Involvement of maternal embryonic leucine zipper kinase (MELK) in mammary carcinogenesis through interaction with Bcl-G, a pro-apoptotic member of the Bcl-2 family
Introduction
Cancer therapies directed at specific molecular targets in signaling pathways of cancer cells, such as tamoxifen, aromatase inhibitors and trastuzumab, have proven useful for treatment of advanced breast cancers.
However, increased risk of endometrial cancer with long-term tamoxifen administration and of bone fracture due to osteoporosis in postmenopausal women undergoing aromatase inhibitor therapy are recognized side effects.
These side effects as well as drug resistance make it necessary to search for novel molecular targets for drugs on the basis of well-characterized mechanisms of action.
Methods
Using accurate genome-wide expression profiles of breast cancers, we found maternal embryonic leucine-zipper kinase (MELK) to be significantly overexpressed in the great majority of breast cancer cells.
To assess whether MELK has a role in mammary carcinogenesis, we knocked down the expression of endogenous MELK in breast cancer cell lines using mammalian vector-based RNA interference.
Furthermore, we identified a long isoform of Bcl-G (Bcl-GL), a pro-apoptotic member of the Bcl-2 family, as a possible substrate for MELK by pull-down assay with recombinant wild-type and kinase-dead MELK.
Finally, we performed TUNEL assays and FACS analysis, measuring proportions of apoptotic cells, to investigate whether MELK is involved in the apoptosis cascade through the Bcl-GL-related pathway.
Results
Northern blot analyses on multiple human tissues and cancer cell lines demonstrated that MELK was overexpressed at a significantly high level in a great majority of breast cancers and cell lines, but was not expressed in normal vital organs (heart, liver, lung and kidney).
Suppression of MELK expression by small interfering RNA significantly inhibited growth of human breast cancer cells.
We also found that MELK physically interacted with Bcl-GL through its amino-terminal region.
Immunocomplex kinase assay showed that Bcl-GL was specifically phosphorylated by MELK in vitro.
TUNEL assays and FACS analysis revealed that overexpression of wild-type MELK suppressed Bcl-GL-induced apoptosis, while that of D150A-MELK did not.
Conclusion
Our findings suggest that the kinase activity of MELK is likely to affect mammary carcinogenesis through inhibition of the pro-apoptotic function of Bcl-GL.
The kinase activity of MELK could be a promising molecular target for development of therapy for patients with breast cancers.
Introduction
Breast cancer is one of the leading causes of cancer death in women worldwide.
According to a 2002 estimate, more than 1,100,000 patients were newly diagnosed with breast cancer, and approximately 410,000 patients died of the disease [1].
Recent improvements in detecting breast cancer at an early stage through mammographic screening have contributed to a decrease in breast cancer-associated mortality.
Mastectomy is among the first options for treatment of localized breast cancer.
Despite surgical removal of primary tumors, however, relapse at local or distant sites occurs in a subset of patients, probably due to undetectable micrometastases at the time of diagnosis [2,3].
Cancer therapies directed at specific molecular targets in signaling pathways of cancer cells, such as tamoxifen, aromatase inhibitors and trastuzumab (Herceptin), have been proven to be useful for treatment of advanced breast cancer [4].
Tamoxifen and aromatase inhibitors suppress the estrogen-related signaling pathway and trastuzumab is the first approved monoclonal antibody for blocking the human epidermal growth factor 2 (HER-2/ErbB-2) signaling pathway [4,5].
Patients with tumors that express estrogen or HER-2 receptors can benefit from either of these therapies and are expected to have a better quality of life and prognosis.
However, increased risk of endometrial cancer with long-term tamoxifen administration and of bone fracture due to osteoporosis in postmenopausal women undergoing aromatase inhibitor therapy are recognized side effects [6,7].
Due to the emergence of these side effects and also because of drug resistance, it is necessary to search for novel molecular targets for drugs on the basis of well-characterized mechanisms of action.
Toward the goal of identifying good molecular targets for drug development, we analyzed the detailed expression profiles of 81 breast tumors, representing 23,040 genes, using a combination of laser-microbeam microdissection and cDNA microarray analysis [8].
After comparing the expression profiles of these breast cancers with those of various normal human tissues [9], we focused on a gene termed maternal embryonic leucine zipper kinase (MELK) that was significantly overexpressed in the great majority of breast cancer cases examined.
In this study, we report evidence indicating that MELK functions as a cancer-specific protein kinase, and that down-regulation of MELK results in growth suppression of breast cancer cells.
In addition, we demonstrate that MELK physically interacts and phosphorylates a long isoform of Bcl-G (Bcl-GL), a pro-apoptotic member of the Bcl-2 family.
Our findings suggest that the kinase activity of MELK is likely to affect mammary carcinogenesis through inhibition of the pro-apoptotic function of Bcl-GL.
We propose that MELK kinase activity could be a promising molecular target for the treatment of breast cancer.
Materials and methods
Cancer cell lines and clinical samples
Human breast cancer cell lines HBL100, HCC1937, MCF-7, MDA-MB-231, MDA-MB-435S, SKBR3, T47D, YMB1, BSY-1 and BT-20, human cervical adenocarcinoma cell line, HeLa, and monkey kidney cells transformed with SV40 T-antigen, COS7 cells, were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA).
Human mammary epithelial cells (HMECs) were purchased from Cambrex Bio Science Walkersville, Inc.
(Walkersville, MD, USA).
HBC4 and HBC5 cells lines were kind gifts from Dr Yamori of the Division of Molecular Pharmacology, Cancer Chemotherapy Center, Japanese Foundation for Cancer Research.
All cells were cultured under their respective depositor's recommendations: RPMI-1640 (Sigma-Aldrich, St Louis, MO, USA) for HBC4, HBC5, HCC1937, SKBR3, T47D, YMB1 and BSY-1 (with 2 mM L-glutamine); Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA, USA) for HBL100, BT-20 and COS7; EMEM (Sigma-Aldrich) with 0.1 mM essential amino acid (Roche, Basel, Switzerland), 1 mM sodium pyruvate (Roche), 0.01 mg/ml insulin (Sigma-Aldrich) for MCF-7; EMEM (Sigma-Aldrich) for HeLa; L-15 (Roche) for MDA-MB-231 and MDA-MB-435S; MEGM (Cambrex Bio Science) for HMECs.
Each medium was supplemented with 10% fetal bovine serum (Cansera International, Ontario, Canada) and 1% antibiotic/antimycotic solution (Sigma-Aldrich).
MDA-MB-231 and MDA-MB-435S cells were maintained at 37°C in humidified air without CO2.
Other cell lines were maintained at 37°C in humidified air with 5% CO2.
Tissue samples from surgically resected breast cancers, and their corresponding clinical information, were obtained from the Department of Breast Surgery, Cancer Institute Hospital, Tokyo, after obtaining written informed consent.
Semi-quantitative RT-PCR analysis
Total RNAs were extracted from each microdissected breast cancer clinical sample and from microdissected normal ductal cells, and polyA(+) RNAs were isolated from normal mammary gland, lung, heart, liver, and kidney (Takara Clontech, Kyoto, Japan).
Subsequently, T7-based amplification and reverse transcription were carried out as described previously [10].
We prepared appropriate dilutions of each single-stranded cDNA for subsequent PCR using the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene as a quantitative control.
The PCR primer sequences were: 5'-GCTGCAAGGTATAATTGATGGA-3' and 5'-CAGTAACATAATGACAGATGGGC-3' for MELK; 5'-CGACCACTTTGTCAAGCTCA-3' and 5'-GGTTGAGCACAGGGTACTTTATT-3' for GAPDH.
Northern blot analysis
Total RNAs were extracted from breast cancer cell lines using the RNeasy kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions.
After treatment with DNase I (Nippon Gene, Osaka, Japan), mRNA was isolated with a mRNA purification kit (GE Healthcare, Buckinghamshire, United Kingdom) following the manufacturer's instructions.
One microgram of each mRNA, along with polyA(+) RNAs isolated from normal mammary gland, lung, heart, liver, kidney and brain (Takara Clontech), were separated on 1% denaturing agarose gels and transferred to nylon membranes (breast cancer Northern blots).
Breast cancer Northern blots and human multiple-tissue Northern blots (Takara Clontech) were hybridized with [α32P]-dCTP-labeled PCR products of MELK cDNA prepared by RT-PCR (see below).
Pre-hybridization, hybridization and washing were performed according to the supplier's recommendations.
The blots were autoradiographed with intensifying screens at -80°C for 12 days.
A probe cDNA (542 base-pairs (bp)) for the sequence in the 3' untranslated region of MELK cDNA (GenBank accession number NM_014791) was prepared by PCR using the primer set 5'-TTATCACTGTGCTCACCAGGAG-3' and 5'-CAGTAACATAATGACAG ATGGGC-3' and were radioactively labeled using the megaprime DNA labeling system (GE Healthcare).
cDNA library screening
We constructed a cDNA library using polyA(+) RNA obtained from T47D breast cancer cells and the Superscript™ plasmid system coupled with Gateway™ technology for cDNA synthesis and a cloning kit (Invitrogen).
We screened 3 × 106 independent clones from this library with a cDNA probe corresponding to nucleotides 1,251 to 2,094 (843 bp) of the MELK cDNA sequence (GenBank accession number NM_014791).
In vitro translation assay
Three variants of MELK (V1, V2 and V3) that were cloned into the pSPORT-1 expression vector were used as templates for transcription/translation experiments in vitro.
The plasmids (1 μg) were transcribed and translated using a TNT Coupled Reticulocyte Lysate system (Promega, Madison, WI, USA) in the presence of ε-labeled biotinylated lysine-tRNA according to the manufacturer's instructions.
Proteins were electrophoresed by SDS-PAGE with a 5% to 20% gradient.
After electroblotting, the biotinylated proteins were visualized by binding streptavidin-horseradish peroxidase followed by chemiluminescent detection (GE Healthcare).
Plasmid construction
To construct MELK expression vectors, the entire coding sequences were amplified by PCR using KOD-Plus DNA polymerase (Toyobo, Osaka, Japan) and cloned into the EcoRI and XhoI sites of the pCAGGSnHC expression vector in frame with the HA (hemagglutinin)-tag at the carboxyl terminus.
The primer set used for PCR reactions with wild-type MELK-V1 was: forward, 5'-CGGAATTCACTATGAAAGATTATG ATGAAC-3' (underlining indicates the EcoRI site); reverse, 5'-AAACTCGAGTACCTTGCAGCTAGATAGGAT-3' (underlining indicates the XhoI site).
Kinase-dead mutant MELK (D150A) was generated with a QuikcChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA) using the following primers; forward, 5'-CATAAATTAAAGCTGATTGCCTTTGGTCTCTGTGCAAAACC-3'; reverse: 5'-GGTTTTGCACAGAGACCAAAGGCAATCAGCTTTAATTTATG-3'.
For construction of the Bcl-GL expression vectors, the corresponding open reading frame was generated by RT-PCR using human testis mRNA as a template and the following set of primers (underlining of forward and reverse primers indicates NotI and XhoI sites, respectively): forward, 5'-ATAAGAATGCGGCCGCGATGTGTAGCACCAGTGG-3' for full-length Bcl-GL (FL); reverse, 5'-CCGCTCGAGTTACTATCAGTCTACTTCTTCATGTG-3' for full-length Bcl-GL (FL); forward 5'-ATAAGAATGCGGCCGCGATCCCCCTAGATGATG-3' for N-1; forward 5'-ATAAGAATGCGGCCGCGTGGCCTTGCAGAAATTC-3' for N-2; forward 5'-ATAAGAATGCGGCCGCGGAATACCAAGATTCGC-3' for N-3; forward 5'-ATAAGAATGCGGCCGCGCAGGCAGGAGGCTTC-3' for N-4; reverse 5'-CCGCTCGAGTTACTATCAGAAGTTCTCTTTCAGGT-3' for C-1; reverse 5'-CCGCTCGAGTTACTATCATCTGGGGTCCACACCCA-3' for C-2; reverse 5'-CCGCTCGAGTTACTATCACAGCTCAACAATTTTGG-3' for C-3.
The reverse primer of the FL construct was commonly used for N-1, N-2, N-3 and N-4 constructs, and the forward primer of it was commonly used for C-1, C-2 and C-3 constructs.
These amplified DNAs were cloned into NotI and XhoI sites of the pCAGGSn3FH vector in frame with Flag-tag at the amino-terminal site.
DNA sequences of all constructs were confirmed by DNA sequencing (ABI3700, PE Applied Biosystems, Foster, CA, USA).
Gene-knockdown of MELK by siRNA
We established a vector-based RNA interference (RNAi) expression system using a psiH1BX3.0 small interfering (si)RNA expression vector as described previously [11].
siRNA expression vectors against MELK (psiH1BX-MELK) and scramble control (SC; the gene coding for 5S and 16S rRNAs in the chloroplast of Euglena gracilis; psiH1BX-SC) were prepared by cloning of double-stranded oligonucleotides into the BbsI site of the psiH1BX3.0 vector.
The target sequence of synthetic oligonucleotides for siRNA were: si-#1, 5'-TCCCACTTGCCTGCCATATCCTTTTCAAGAGAAAGGATATGGCAGGCAAGT-3' and 5'-AAAAACTTGCCTGCCATATCCTTTCTCTTGAAAAGGATATGGCAGGCAAGT-3', si-#2, 5'-TCCCCTATCCTGTTGAGTGGCAATTCAAGAGATTGCCACTCAACAGGATAG-3' and 5'-AAAACTATCCTGTTGAGTGGCAATCTCTTGAATTGCCACTCAACAGGATAG-3', si-#3, 5'-TCCCGACATCCTATCTAGCTGCATTCAAGAGATGCAGCTAGATAGGATGTC-3' and 5'-AAAAGACATCCTATCTAGCTGCATCTCTTGAATGCAGCTAGATAGGATGTC-3', si-#4, 5'-TCCCAGTTCATTGGAACTACCAATTCAAGAGATTGGTAGTTCCAATGAACT-3' and 5'-AAAAAGTTCATTGGAACTACCAATCTCTTGAATTGGTAGTTCCAATGAACT-3'.
SC, 5'-TCCCGCGCGCTTTGTAGGATTCGTTCAAGAGACGAATCCTACAAAGCGCGC-3' and 5'-AAAAGCGCGCTTTGTAGGATTCGTCTCTTGAACGAATCCTACAAAGCGCGC-3'.
Underlining indicates siRNA-targeting sequences designed from MELK mRNA (GenBank accession number NM_014791).
All constructs were also confirmed by DNA sequencing.
Human breast cancer cells lines T47D and MCF-7 were plated in 15 cm dishes (4 × 106 cells/dish), and transfected with 16 μg of psiH1BX-MELK (si-#1, si-#2, si-#3 and si-#4) and psiH1BX-SC (SC; negative control) siRNA plasmids using the FuGENE6 transfection reagent (Roche) according to the supplier's recommendations.
At 24 hours after transfection, cells were re-seeded for the following experiments at the densities given: 1 × 106 cells/10 cm dish for semi-quantitative RT-PCR and western blot analyses; 3 × 106 cells/10 cm dish for colony formation assay and 5 × 105 cells/well for MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay.
We selected T47D or MCF7 cells transfected with MELK-siRNAs in medium containing 0.7 mg/ml or 0.6 g/ml neomycin (geneticine; Invitrogen), respectively.
Four days after neomycin selection, 0.5 μg of total RNA extracted from these cells was reverse-transcribed into cDNA using Superscript II (Invitrogen) to examine the knockdown effect of siRNAs by semi-quantitative RT-PCR using specific primer sets for MELK and β2MG (encoding beta-2-microglobulin).
Cells were also harvested four days after neomycin selection for western blot analysis using anti-MELK antibody (Cell signaling technology, Boston, MA, USA).
The specific primer sets for RT-PCR were: 5'-TTATCACTGTGCTCACCAGGAG-3' and 5'-CAGTAACATAAT GACAGATGGGC-3' for MELK; 5'-TTAGCTGTGCTCGCGCTACT-3' and 5'-TCACATGGTTCACACGGCAG-3' for β2MG as a quantitative control.
Transfectants expressing siRNA were grown for 3 weeks in selective media containing neomycin, then fixed with 4% paraformaldehyde for 15 minutes before staining with Giemsa's solution (Merck, Whitehouse Station, NJ, USA) to assess colony number.
To quantify cell viability, MTT assays were performed at seven days after selection with Cell Counting Kit-8 (Wako, Osaka, Japan) according to the manufacture's protocols.
Absorbance at 570 nm was measured with a Microplate Reader 550 (Bio-Rad, Hercules, CA, USA).
These experiments were performed in triplicate.
Generation and purification of His-tagged recombinant MELK
The entire coding sequences of wild-type MELK (WT-MELK) and kinase-dead MELK (D150A-MELK; asparate changed to alanine at the 150th residue in the kinase subdomain VII, which is considered as the ATP binding site [12]) were subcloned into the pET21a vector (Merck Novagen, Darmstadt, Germany).
The WT-MELK and D150A-MELK recombinant proteins were expressed in Escherichia coli strain BL21 codon-plus (DE3) RIL competent cells (Stratagene), and cultured in TBG-M9 medium (1% tryptone, 0.5% NaCL, 0.5% glucose, 1 mM MgSO4, 0.1% NH4CI, 0.3% KH2PO4, 0.6% Na2HPO4).
After induction with 0.5 μM isopropyl-b-D-thiogalactopyranoside (IPTG) at 25°C for 5 hours, the bacteria pellet was suspended with lysis buffer, and was purified with Ni-NTA superflow (Qiagen) under nondenaturing conditions according to the supplier's instructions.
The entire coding sequence of BCL-GL was also subcloned into the pGEX-6P-1 vector (GE Healthcare).
The glutathione S-transferase (GST)-Bcl-GL recombinant protein was expressed in Escherichia coli strain BL21 codon-plus RIL competent cells (Stratagene).
Purification of the recombinant proteins was performed using Glutathione Sepharose 4B beads (GE Healthcare) under nondenaturing conditions according to the supplier's instructions.
For confirmation of direct binding of BCL-GL and MELK, we removed GST from GST-fused BCL-GL protein using PreScission protease (GE Healthcare) according to the supplier's instructions.
Protein pull down assay
HBC4 cells were lysed with lysis buffer (40 mM Tris-HCL (pH 7.5), 1% Trition X-100, 2.5 mM EDTA, 15 mM dithiothreitol (DTT)) including 0.1% protease inhibitor cocktail III (Calbiochem, San Diego, CA, USA) as previously described [13] with a minor modification.
We incubated 7.5 mg of cell lysates with 0.5 μM of WT-MELK or D150A-MELK recombinant proteins pre-immobilized on Ni-NTA agarose beads in reaction buffer containing 50 mM Tris-HCL (pH 7.5), 50 μM ATP, 10 mM MgCL2, 25 mM NaCL and 1 mM DTT at 30°C for 5 minutes.
The protein-bound beads were washed three times with lysis buffer, and then eluted with SDS sample buffer.
The proteins were separated by SDS-PAGE using a 4% to 12% gradient Bis-Tris gel (Invitrogen), and stained with Silver Stain DAIICHI (Daiichi Pure Chemicals, Tokyo, Japan).
An approximately 30 kDa band, which was seen in pulled-down cell lysates with WT-MELK added but not in those with D150A-MELK added, was extracted.
Its peptide sequence was determined by MALDI-TOF mass spectrometry.
Western blot analysis
To detect the endogenous MELK and Bcl-GL proteins in breast cancer cell lines (BT-20, HBC4, HBC5, HBL100, MCF-7, MDA-MB-231, SKBR3 and T47D) and HMECs, these cells were lysed in lysis buffer (50 mM Tris-HCL pH 8.0, 150 mM NaCL, 0.5% NP-40) including 0.1% protease inhibitor cocktail III (Calbiochem).
After homogenization, the cell lysates were incubated on ice for 30 minutes and centrifuged at 14,000 rpm for 15 minutes to separate only supernatant from cell debris.
The amount of total protein was estimated by protein assay kit (Bio-Rad), and the proteins were then mixed with SDS sample buffer and boiled for 3 minutes before loading into a 10% to 20% gradient SDS-PAGE gel (Bio-Rad).
After electrophoresis, the proteins were transferred onto nitrocellulose membrane (GE Healthcare).
Membranes were blocked by 4% BlockAce (Dainippon Pharmaceutical Co., Ltd, Osaka, Japan), and incubated with anti-MELK polyclonal or anti-Bcl-GL polyclonal antibodies (Abcam, Cambridge, MA, USA).
β-Actin served as a loading control.
Finally, the membranes were incubated with horseradish peroxidase (HRP) conjugated secondary antibody and protein bands were visualized by enhanced chemiluminescence (ECL) detection reagents (GE Healthcare).
Co-immunoprecipitation assay
HeLa cells were transiently co-transfected for 48 hours with 8 μg of plasmid constructs encoding Flag-tagged full-length Bcl-GL or a series of Flag-tagged partial Bcl-GL proteins (FL, N-1, N-2, N-3, N-4, C-1, C-2 and C-3) as well as the same amount of plasmid encoding HA-tagged WT-MELK, using the FuGENE6 transfection reagent (Roche).
Cells were lysed with lysis buffer as described above.
The lysates were pre-cleaned with normal mouse IgG (1.2 μg) and rec-Protein G Sepharose 4B (Zymed, San Francisco, CA, USA) at 4°C for 30 minutes.
Subsequently, the lysate was incubated with anti-Flag agarose M2 gel (Sigma-Aldrich) at 4°C for 12 hours.
After washing three times with lysis buffer, proteins on beads were eluted with SDS sample buffer.
In vitro binding assay
The His-tagged MELK recombinant protein (10 μg) was immunoprecipitated with anti-6XHis-tag monoclonal antibody (2 μg; Takara Clontech) in 500 μl of coupling buffer (50 mM Tris, 75 mM NaCL, 10 mM MgCL2, 1 mM DTT, and complete EDTA-free protease inhibitors; Roche) for 3 hours at 4°C.
Subsequently, 30 μl of protein G Sepharose (Zymed) was added, and the incubation was further continued for 45 minutes.
The immune complexes were washed 3 times with 1 ml of coupling buffer, and resuspended with 500 μl of binding buffer (50 mM Tris, 75 mM NaCL, 10 mM MgCL2, 1 mM DTT, 2% bovine serum albumin, 100 μM ATP and complete EDTA-free protease inhibitors; Roche) containing 5 μg of Bcl-GL recombinant protein.
After incubation for 30 minutes at 4°C for binding, the samples were washed 5 times with washing buffer (50 mM Tris-CL, 75 mM NaCL, 10 mM MgCL2, 1 mM DTT, 0.01% Triton-X100 and complete EDTA-free protease inhibitors; Roche).
The precipitates were then eluted with SDS sample buffer by boiling for 3 minutes and analyzed by western blot with anti-Bcl-GL antibody.
Immunocomplex kinase assay
To use the long isoform Bcl-GL recombinant protein as a substrate for kinase assay, a set of Flag-tagged Bcl-GL (FL, N-1, N-2 N-3 and N-4) expression vectors was transfected into HeLa cells, and the proteins were immunoprecipitated with Flag-conjugated agarose M2 gel (Sigma-Aldrich) at 4°C for 1 hour.
These immunoprecipitates were washed five times with lysis buffer containing 50 mM Tris-HCL (pH 7.5), 150 mM NaCL, 1% NP-40 and 0.1% protease inhibitor cocktail III (Calbiochem).
Aliquots (20 μl) of immunoprecipitates were subjected to immunoblotting using rabbit anti-Flag antibody (Sigma-Aldrich) to check the success of the immunoprecipitation experiments.
His-tagged MELK recombinant proteins (0.5 μM WT-MELK and D150A-MELK) were reacted for 30 minutes at 30°C with aliquots (20 μl) of immunoprecipitates of Flag-tagged Bcl-GL (FL, N-1 N-2, N-3 and N-4) as substrates in 50 μl of kinase buffer containing 30 mM Tris-HCL (pH 7.5), 0.1 mM EGTA, 10 mM DTT, 40 mM NaF, 40 mM sodium β-glycerophosphate (Sigma-Aldrich), 50 μM cold-ATP, 10 Ci of [γ-32P] ATP (GE Healthcare) and 10 mM MgCL2.
The reaction was terminated by addition of SDS sample buffer and boiled for 3 minutes prior to 10% SDS-PAGE.
The gel was then dried and autoradiographed with intensifying screens at room temperature overnight.
TUNEL assay for apoptosis
COS7 cells (1 × 106) were co-transfected with 8 μg each of Flag-tagged pCAGGSn3FH vectors (Flag-Bcl-GL and Flag-Mock) and HA-tagged pCAGGSnHC vectors (HA-WT-MELK, HA-D150A-MELK and HA-Mock) using the FuGENE 6 transfection reagent (Roche) according to the supplier's protocol.
At 24 hours post-transfection, cells were harvested by trypsinization and washed with phosphate buffered saline without CaCL2 and MgCL2 (PBS(-)).
An aliquot of cells (300 μl) was immobilized on the microslide glass (Matsunami Glass, Osaka, Japan) and subjected to a TUNEL assay using an In situ Cell Death Detection Kit, Fluorescein (Roche) according to the supplier's instructions.
The apoptotic cells were observed with a TCS SP2 AOBS microscope (Leica, Tokyo, Japan).
Experiments were carried out in triplicate independently.
Flow cytometry analysis for apoptosis
COS7 cells (1 × 106) were co-transfected with 8 μg each of Flag-tagged pCAGGSn3FH vectors (Flag-Bcl-GL and Flag-Mock) and HA-tagged pCAGGSnHC vectors (HA-WT-MELK, HA-D150A-MELK and HA-Mock) using the FuGENE 6 transfection reagent (Roche).
For FACS analysis, the cells were harvested by trypsinization and fixed with 70% ethanol at room temperature for 30 minutes.
After centrifuging to remove the ethanol, the cells were treated with 0.5 ml of PBS(-) containing 1 mg/ml of RNase I (Sigma-Aldrich) for 30 minutes followed by staining in 1 ml of PBS(-) containing 20 mg/ml of propidium iodide (Sigma-Aldrich) for 30 minutes.
The cells selected from at least 10,000 ungated cells were analyzed for DNA content by flow cytometry (FACS calibur; Becton Dickinson, San Diego, CA, USA).
The data were analyzed using CELLQuest software (FACS calibur; Becton Dickinson).
Assays were done in triplicate independently.
Statistical analysis
Statistical significance was determined by Student's t-test using Statview 5.0 software (SAS Institute, Cary, NC, USA).
A difference of P < 0.05 was considered to be statistically significant.
Results
Up-regulation of MELK in breast cancer
We previously reported the genome-wide expression profile analysis of 81 clinical breast cancers, representing 23,040 genes, by means of cDNA microarray analysis in combination with enrichment of cancer cells with a laser microbeam microdissection system [8].
The MELK gene (GenBank accession number NM_014791) was found to be one of the genes transactivated at a very high level in the great majority of the breast cancers examined.
The subsequent semi-quantitative RT-PCR analysis of the transcript confirmed the elevated expression of MELK in 11 of 12 clinical breast cancer specimens (Figure 1a).
To further examine the expression pattern of this gene in breast cancer cell lines and normal tissues, we performed Northern blot analyses with mRNAs from multiple human tissues and breast cancer cell lines using a cDNA fragment corresponding to the 3' untranslated region of the MELK gene as a probe.
The results showed that an approximately 2.7 kb band was most highly expressed in testis, and weakly expressed in thymus and small intestine (Figure 1b).
Interestingly, Northern blot analyses of breast cancer cell lines using the same probe showed that 2.4 to 2.7 kb bands were significantly over-expressed in breast cancer cells but were not expressed in vital organs (Figure 1c), indicating that some cancer specific transcripts are present over this range.
To characterize the variants specifically transcribed in breast cancer cells, we screened a cDNA library of breast cancer cells.
Subsequent cDNA sequencing analysis identified three different transcriptional variants, designated V1 (2,501 bp; NM_014791), V2 (2,368 bp; AB183427) and V3 (2,251 bp; AB183428).
These three different transcriptional variants consist of 18, 17 and 16 exons, respectively.
The V2 variant lacks exon 3 (86 bp) and the V3 variant lacks exon 3 (86 bp) and exon 4 (117 bp) as a result of alternative splicing (Figure 1d).
The V2 and V3 variants contain early stop codons (TGA) in exon 4 and exon 5, respectively, if they are translated from the same first ATG codon of the V1 variant.
However, if they are translated from a second ATG codon located downstream of the first ATG codon of the V1 variant, the V2 and V3 variant proteins are thought to produce 618 and 581 amino acid peptides, respectively.
Thus, to examine both possibilities, we carried out an in vitro translation assay and found that the V1, V2 and V3 proteins were 75, 71 and 66 kDa, respectively, indicating that the V2 and V3 proteins were translated from the second ATG codon (Figure 1e).
Subsequently, we examined MELK expression in breast cancer cell lines by western blot analysis with anti-MELK polyclonal antibodies.
The results indicate that the V1 protein is dominantly overexpressed in breast cancer cells compared with the V2 and V3 proteins (Figure 1f).
Furthermore, although all of the three variant proteins possess the KA1 domain in their carboxyl terminus, the V2 and V3 proteins lack the amino-terminal portion, which corresponds to the first 48 amino acids of the V1 protein and is considered to be the catalytic kinase domain (Figure 1g).
Although we can not rule out the possibility that the V2 and V3 variants have some functional role in carcinogenesis, for example, dominant-negative functions, in this study we focus on the kinase activity of the V1 transcript due to its predominant expression in cancer cells.
Growth-inhibitory effects of siRNA against MELK
To assess a possible role of MELK in mammary carcinogenesis, we knocked down the expression of endogenous MELK in the breast cancer cell lines T47D and MCF7 (Figure 2a,b), in which MELK was overexpressed at a high level, by means of the mammalian vector-based RNAi technique (see Materials and methods).
We examined expression levels of MELK by semi-quantitative RT-PCR and western blot analyses, and found that the MELK-siRNAs si-#3 and si-#4 significantly reduced its expression at the transcriptional and protein levels compared with scramble-siRNA (SC), si-#1 and si-#2 (Figure 2).
Colony formation and MTT assays revealed that both si-#3 and si-#4 significantlysuppressed growth of T47D and MCF7 cells (MTT assay in T47D, P = 0.0003, P = 0.0013; in MCF-7, P = 0.0001, P = 0.0001; unpaired t-test), in concordance with the results showing the knockdown effect of this gene.
These results suggest that MELK has a critical role in the growth of breast cancer cells.
Identification of Bcl-GL as a MELK-interacting protein
To investigate the biological functions of MELK in breast cancer cells, we searched for substrates of MELK in cancer cells by in vitro protein pull-down assays using wild-type MELK (WT-MELK) and kinase-dead MELK (D150A-MELK) recombinant proteins.
Comparison of silver staining of SDS-PAGE gels containing the pulled-down proteins identified an approximately 30 kDa protein in the lane corresponding to proteins pulled-down with WT-MELK but not in that corresponding to proteins pulled-down with D150A-MELK (Figure 3a).
MALDI-TOF analysis showed this 30 kDa protein to be Bcl-G, a member of the Bcl-2 protein family [14].
According to NCBI database, Bcl-G has three alternative splicing isoforms, termed Bcl-G short isoform (Bcl-GS), median isoform (Bcl-Gm) and long isoform (Bcl-GL), and encoding 252, 276 and 327 amino acid peptides, respectively.
We first examined the expression patterns of these Bcl-G isoforms at the transcriptional level in breast cancer cell lines by semi-quantitative RT-PCR, and found that Bcl-GS and Bcl-GL were expressed in all breast cancer cells examined, whereas the Bcl-Gm transcript was not detected at all (data not shown).
Furthermore, we examined expression of endogenous Bcl-GL in breast cancer cell lines at the protein level by western blot analysis using an anti-Bcl-GL antibody and detected an approximately 30 kDa endogenous Bcl-GL, corresponding to the size identified by protein pull-down experiment, in all the breast cancer cell lines examined (Figure 3b).
Therefore, we focused on Bcl-GL as a candidate interacting protein for MELK.
To validate an interaction between WT-MELK and Bcl-GL, we constructed plasmids designed to express HA-tagged WT-MELK (HA-WT-MELK) and Flag-tagged Bcl-GL (Flag-Bcl-GL).
These plasmids were co-transfected into HeLa cells and the proteins immunoprecipitated with anti-Flag antibody.
Immunoblotting of the precipitates using anti-HA antibodies indicated that Flag-Bcl-GL was co-precipitated with HA-WT-MELK (Figure 3c).
Furthermore, we demonstrated that His-tagged WT-MELK could pull-down with Bcl-GL but His-tagged D150A-MELK could not, indicating that, in vitro, Bcl-GL interacts directly with WT-MELK but not with D150A-MELK (Figure 3d).
To further determine which segment of Bcl-GL can interact with WT-MELK, we performed co-immunoprecipitation analyses using HA-WT-MELK and partial Bcl-GL proteins tagged with Flag (Figure 3e).
After co-transfection of plasmid clones into HeLa cells, we performed immunoprecipitation with an anti-Flag antibody, and then immunoblotting with an anti-HA antibody.
We found that the N-1 construct (amino acids 12 to 327) as well as a full-length Bcl-GL (FL) bound to WT-MELK, whereas N-2 (amino acids 72 to 324), N-3 (amino acids 121 to 327) or N-4 (amino acids 171 to 327) did not (Figure 3f).
Concordantly, all of the partial products that contained the amino-terminal portion of Bcl-GL – C-1 (amino acids 1 to 305), C-2 (amino acids 1 to 295), and C-3 (amino acids 1 to 215) bound to WT-MELK (Figure 3f).
These results suggest that the region corresponding to amino acids 12 to 71 of Bcl-GL is likely to interact with WT-MELK.
MELK phosphorylates Bcl-GL
To examine whether the Bcl-GL protein is a substrate of MELK kinase activity, we performed an immune complex kinase assay.
We first confirmed the exogenous expression of Flag-tagged Bcl-GL (Flag-Bcl-GL) by western blot analysis (Figure 3f).
As shown in Figure 4a, WT-MELK could phosphorylate Bcl-GL, but D150A-MELK could not (single arrowhead).
In addition, autophosphorylation of an approximately 75 kDa protein of MELK was observed (Figure 4a; double arrowheads).
Furthermore, we also confirmed that WT-MELK could phosphorylate GST-Bcl-GL but D150A-MELK could not (Figure 4b), indicating MELK can directly phosphorylate Bcl-GL.
These findings suggest that Bcl-GL is a potential substrate for MELK.
We subsequently examined WT-MELK kinase activity on various partial amino-terminal Bcl-GL products (Figure 3e).
First, we confirmed the amount of immunoprecipitated protein corresponding to partial Bcl-GL (N-1, N-2, N-3 and N-4) as well as to full-length Bcl-GL (FL) by western blot analysis (Figure 3f), and then examined the extent of their phosphorylation (Figure 4c).
The MELK recombinant protein phosphorylated N-1 as well as full-length Bcl-GL, whereas it could not phosphorylate N-2, N-3 or N-4 proteins, corresponding with their inability to interact.
MELK involvement in the apoptotic pathway through Bcl-GL
Because MELK can physically interact with and phosphorylate Bcl-GL (Figures 3d and 4a), we hypothesized that it might be involved in the apoptosis cascade through the Bcl-GL-related pathway.
To investigate this hypothesis, we transiently co-transfected two plasmid clones designed to express HA-tagged MELK (WT or D150A) and Flag-tagged Bcl-GL into COS7 cells, and then performed a TUNEL assay and FACS analysis to measure the proportions of apoptotic cells (see Material and methods).
We first confirmed the exogenous expression of MELK and Bcl-GL in COS7 cells by western blot analysis (Figure 5a).
As indicated in Figure 5b,c, the overexpression of full-length Bcl-GL (HA-Mock+Flag-Bcl-GL) significantly increased the proportion of TUNEL-positive cells compared with the cells transfected with the mock plasmids (HA-Mock+Flag-Mock), indicating that Bcl-GL induces apoptosis, as described previously [14].
In contrast, the co-overexpression of WT-MELK with Bcl-GL (HA-WT-MELK+Flag-Bcl-GL) reduced the proportion of TUNEL-positive cells compared with over-expression of Bcl-GL alone (HA-Mock+Flag-Bcl-GL) (P = 0.0001, unpaired t-test).
However, the co-overexpression of D150A-MELK with Bcl-GL (HA-D150A-MELK+Flag-Bcl-GL) did not affect the proportion of TUNEL-positive cells.
As shown in Figure 5d, FACS analysis of the cells under the same conditions also confirmed that the overexpression of Bcl-GL increased the sub-G1 population of cells compared with the mock-transfected cells.
Similarly to the TUNEL analysis, the overexpression of WT-MELK with Bcl-GL reduced the proportion of sub-G1 cells (P = 0.03, unpaired t-test), while D150A-MELK increased the sub-G1 population.
Our results imply that the kinase activity of MELK may play a critical role in the regulation of the pro-apoptotic function of Bcl-GL.
Discussion
Microarray technologies applied to the study of cancer have contributed to the discovery of genes that play essential roles in carcinogenesis, the discovery of novel molecular target(s) for the development of anti-cancer agents and/or diagnosis, and the identification of genes related to chemo- or radio-sensitivity.
Although such information has been applied in the development of novel therapeutic agents for breast cancer, the range of available treatments for advanced-stage patients is still very limited.
Thus, it is urgent that further molecular targets are discovered to enable development of new anti-cancer drugs that are specific for malignant cells with a minimum risk of adverse reactions.
Using genome-wide expression profiles of breast cancers and normal human tissues [8,9], we identified MELK, one of the transcripts of which was specifically up-regulated in the great majority of clinical breast cancer samples, but expressed in none of 29 normal human tissues examined except testis, thymus and small intestine.
In this study, we characterize the biological function of MELK and indicate that it would be a good candidate as a molecular target for breast cancer therapy, although there would be some concerns over intestinal adverse reactions.
We demonstrated by means of siRNA that knockdown of endogenous MELK expression results in growth suppression of breast cancer cells (Figure 2).
Furthermore, our cDNA microarray data indicate that MELK is also up-regulated relatively frequently in bladder cancers, osteosarcoma and small cell lung cancers (data not shown).
Together, these findings suggest that MELK has an oncogenic role in not only breast cancers but also bladder cancers, bone cancers and small cell lung cancers.
MELK was previously identified as a new member of the snf1/AMPK serine-threonine kinase family that is involved in mammalian embryonic development [15-18].
This gene was shown to play an important role in hematopoiesis, stem cell renewal [19] and cell-cycle progression through an interaction with zinc finger-like protein ZPR9 [20] and splicing factor NIPP1 [12].
To investigate the expression pattern of these proteins in breast cancer cells, we performed semi-quantitative RT-PCR, and observed no significant correlation of gene expression between MELK and these interacting molecules in breast cancer cells (data not shown).
Thus, to investigate the biological significance of MELK in breast cancer cells, we searched for a possible substrate(s) of MELK by means of in vitro pull-down assays with recombinant wild-type MELK (WT-MELK) and kinase-dead MELK (D150A-MELK).
We identified a long isoform of Bcl-G (Bcl-GL), a pro-apoptotic member of the Bcl-2 family, as a potential substrate for MELK.
In addition to physical interaction between these two proteins, in vitro immune complex kinase and in vitro binding assays showed that Bcl-GL was directly bound to MELK and specifically phosphorylated by it in vitro (Figures 3d and 4a,b).
As reported previously [14], we demonstrated by TUNEL assay and FACS analysis that introduction of full-length Bcl-GL into COS7 cells induced apoptosis.
However, under the same conditions, addition of exogenous WT-MELK suppressed induction of apoptosis by Bcl-GL, but addition of D150A-MELK did not (Figure 5b–d).
Thus, we speculate that MELK might promote cell growth by inhibiting the pro-apoptotic function of Bcl-GL through its phosphorylation.
Conclusion
Our findings clearly suggest that MELK is overexpressed in both breast cancer specimens and cancer cell lines, and that its kinase activity possibly plays a significant role in breast cancer cell growth.
Recent anti-cancer drug development is focused on targeting important molecules involved in the oncogenic pathways, represented by imatinib mesylate and trastuzumab.
We found that down-regulation of MELK by treatment with siRNA significantly suppressed the cell growth of breast cancer, indicating its crucial role in the proliferation and tumorgenesis of breast cancer.
In particular, we demonstrate a new biological function for MELK in breast carcinogenesis, the inhibition of apoptosis though its interaction with and phosphorylation of Bcl-GL, a pro-apoptotic member of Bcl-2 family.
Our data should contribute to a better understanding of breast carcinogenesis, and they suggest that MELK is a promising molecular target for breast cancer treatment.
Abbreviations
β2MG = beta-2-microglobulin; DTT = dithiothreitol; GST = glutathione S-transferase; Her-2/ErbB-2 = epidermal growth factor 2; HMEC = human mammary epithelial cell; MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PBS = phosphate-buffered saline; RNAi = RNA interference; SC = scramble control.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
LML performed all experiments and drafted the manuscript.
PJH participated in the discussion and interpretation of data.
NT constructed expression profiles of breast cancer and participated in the screening of candidate genes.
YN was involved in the conception and design of studies, interpretation of data and preparing the final version of the manuscript.
TK guided all molecular aspects of the studies and was involved in the design of studies and interpretation of results.
Breast cancer
A pooled analysis of bone marrow micrometastasis in breast cancer
Pathological validation and significance of micrometastasis in sentinel nodes in primary breast cancer
Pharmacological breast cancer therapy (review)
Trastuzumab (herceptin), a humanized anti-Her2 receptor monoclonal antibody, inhibits basal and activated Her2 ectodomain cleavage in breast cancer cells
Tamoxifen for the prevention of breast cancer: current status of the National Surgical Adjuvant Breast and Bowel Project P-1 study
Incidence and management of side effects associated with aromatase inhibitors in the adjuvant treatment of breast cancer in postmenopausal women
Genome-wide gene-expression profiles of breast-cancer cells purified with laser microbeam microdissection: identification of genes associated with progression and metastasis
Genome-wide profiling of gene expression in 29 normal human tissues with a cDNA microarray
Alterations of gene expression during colorectal carcinogenesis revealed by cDNA microarrays after laser-capture microdissection of tumor tissues and normal epithelia
Involvement of the FGF18 gene in colorectal carcinogenesis, as a novel downstream target of the beta-catenin/T-cell factor complex
Inhibition of spliceosome assembly by the cell cycle-regulated protein kinase MELK and involvement of splicing factor NIPP1
A novel method to identify protein kinase substrates: eEF2 kinase is phosphorylated and inhibited by SAPK4/p38delta
Bcl-G, a novel pro-apoptotic member of the Bcl-2 family
Melk-like kinase plays a role in hematopoiesis in the zebra fish
Expression of Melk, a new protein kinase, during early mouse development
Maternal embryonic leucine zipper kinase/murine protein serine-threonine kinase 38 is a promising therapeutic target for multiple cancers
Cellcycle regulation of pEg3, a new Xenopus protein kinase of the KIN1/PAR-1/MARK family
Maternal embryonic leucine zipper kinase (MELK) regulates multipotent neural progenitor proliferation
Phosphorylation of a novel zinc-finger-like protein, ZPR9, by murine protein serine/threonine kinase 38 (MPK38)
Figures and Tables
Expression and distribution of MELK in human normal tissues and breast cancer cell lines.
(a) Expression of MELK in 12 breast cancer specimens (case number; 42, 102, 247, 252, 302, 473, 478, 502, 552, 646, 769 and 779) by semi-quantitative RT-PCR.
GAPDH served as a quantitative internal control.
(b) Multiple tissue Northern blot analysis demonstrated that an approximately 2.7 kb MELK transcript was detected in the testis, thymus and small intestine.
PBL, peripheral blood leukocytes.
(c) Breast cancer cell line Northern blot analysis revealed that approximately 2.4 to 2.7 kb MELK variants were specifically expressed in breast cancer cell lines, but not in normal vital organs.
(d) Schematic representation of three variant transcripts identified by cDNA library screening (see Materials and methods).
White boxes indicate a coding region and black boxes indicate a non-coding region.
Black and grey triangles indicate initiation codons, and white triangles indicate stop codons.
Exon numbers are shown above each box.
(e) In vitro translation assay of each variant isolated from cDNA library screening.
The number within parentheses represents the predicted molecular weight (kDa) of each variant protein.
(f) Expression of MELK proteins in eight breast cancer cell lines as well as human mammary epithelial cells (HMECs) shown by western blot analysis with an anti-MELK antibody.
β-Actin served as a control.
(g) Schematic representation of the V1, V2 and V3 forms of MELK.
The shaded boxes indicate the catalytic domain (amino acids 11 to 263 of the V1 protein).
The KA1 domain is the kinase-associated domain in the carboxy-terminal region.
Effect of knockdown of MELK by small-interfering RNA (siRNA) on cell viability and proliferation.
Four psiH1 promoter-based siRNA constructs (si-#1, si-#2, si-#3 and si-#4) were introduced into (a) T47D and (b) MCF-7 cell lines.
SC refers to scramble used as a control for siRNA experiments.
Gene silencing was evaluated by semi-quantitative RT-PCR and western blot analyses at four and five days after neomycin selection, respectively.
β2-microglobulin (β2MG) was used as a control for normalization of semi-quantitative RT-PCR, and β-actin was used as a control in western blot analysis.
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assays were performed to evaluate cell viability at 10 days after neomycin selection, and graphed after standardization using the scramble control (SC) as 1.0 (T47D, P = 0.0003, P = 0.0013; MCF-7, P = 0.0001, P = 0.0001; unpaired t-test).
Colony formation assays were carried out three weeks after neomycin selection (see Materials and methods).
Two siRNA constructs (si-#3 and -#4) showed significant knockdown effects against internal MELK expression and inhibited cell growth in both T47D (a) and MCF-7 (b) cell lines.
Values represent the average from triplicate experiments.
Error bars indicate standard deviation.
Identification of Bcl-GL as an interacting protein for MELK.
(a) Silver staining of SDS-PAGE gels that contained the pulled-down cell lysates.
The enlarged area covering Bcl-G shows the differential interaction between it and wild-type MELK (WT-MELK) and kinase-dead MELK (D150A-MELK).
(b) Expression of Bcl-GL in eight breast cancer cell lines as well as human mammary epithelial cells (HMECs) by western blot analysis with an anti-Bcl-GL antibody.
β-Actin was used as a control.
(c) Interaction of MELK with Bcl-GL.
Extracts from HeLa-cells transfected with HA (hemagglutinin)-tagged WT-MELK (HA-WT-MELK) or Flag-tagged Bcl-GL (Flag-Bcl-GL), or a combination of these, were harvested 36 hours after transfection.
The cell lysates were immunoprecipitated with anti-Flag M2 antibody.
Precipitated proteins were separated by SDS-PAGE and western blotting analysis was performed with an anti-HA antibody.
(d) Direct interaction of the MELK and Bcl-GL proteins.
The upper panel indicates the amount of input of WT-MELK and D150A-MELK.
His-tagged WT-MELK bound to Bcl-GL, but His-tagged D150A-MELK did not.
(e) Schematic representation of the amino- and carboxy-terminal deletion constructs of Bcl-GL.
The C-1 and C-2 constructs have the BH2 domain deleted, and the C-3 construct has both the BH2 and BH3 domains deleted.
(f) Determination of the WT-MELK binding regions of Bcl-GL by immunoprecipitation.
The HA-tagged WT-MELK and various peptide sequences of Flag-tagged Bcl-GL (Figure 3e) were pulled down by immunoprecipitation with Flag-M2 antibody and then immunoblotted with rabbit anti-Flag antibody.
The expression of HA-tagged WT-MELK in total cell lysates was confirmed by western blotting analysis.
As a control, immunoprecipitation was performed from cells co-transfected with pCAGGSn3FC (Mock) and HA-tagged WT-MELK (HA-WT-MELK) through all steps.
Arrowheads indicate expression of each Bcl-GL peptide.
MELK phosphorylates Bcl-GL in vitro.
(a) Immunoprecipitates were subjected to immune complex kinase assay with wild-type (WT)-MELK or kinase-dead (D150A)-MELK).
The single arrowhead indicates phosphorylated Bcl-GL, and the double arrowhead points to an autophosphorylated MELK protein.
(b) Phosphorylation of a bacterial glutathione S-transferase (GST) Bcl-GL fusion recombinant protein (GST-Bcl-GL) by His-tagged WT-MELK (WT).
The single arrowhead indicates phosphorylated GST-Bcl-GL protein, and the double arrowhead indicates an autophosphorylated His-tagged MELK protein.
(c) In vitro phosphorylation of various partial amino-terminal constructs of Bcl-GL (N-1, N-2, N-3 and N-4; see Figure 3e) and full-length Bcl-GL (FL) by MELK.
The single arrowheads indicate phosphorylated immunoprecipitated Bcl-GL proteins, and the double arrowhead indicates an autophosphorylated MELK recombinant protein.
MELK involvement in the apoptosis cascade through Bcl-GL.
HA (hemagglutinin)-tagged MELK (HA-wild-type (WT)-MELK or HA-kinase-dead (D150A)-MELK) and Flag-tagged Bcl-GL (Flag-Bcl-GL) expression vectors were co-transfected into COS7 cells for 24 hours.
(a) The expression of MELK and Bcl-GL proteins in the co-transfected cells were examined by western blot analysis.
(b) TUNEL assays after transfection with pCAGGSnHC (HA-Mock), pCAGGSn3FH (Flag-Mock), HA-tagged MELK (WT and D150A), Flag-tagged Bcl-GL expression vectors, and combinations of these.
Apoptotic cells were measured by counting of TUNEL staining (means ± standard deviation, n = 3; P = 0.0001; unpaired t-test).
(c) Representative images of TUNEL assays.
Cells were labeled with DAPI (4',6-diamidino-2-phenylindole) for counting of total cell number.
Apoptotic cells with DNA strand breaks were labeled with green fluorescence.
(d) FACS analysis of cells collected after transfection with pCAGGSnHC (HA-Mock), pCAGGSn3FH (Flag-Mock), HA-tagged MELK (WT and D150A), Flag-tagged Bcl-GL expression vectors, and combinations of these.
Proportions of apoptotic cells are indicated as percentages of sub-G1 populations.
Each value represents the average of three experiments (means ± standard deviation, n = 3).Cdk5 Is Involved in BDNF-Stimulated Dendritic Growth in Hippocampal Neurons
Neurotrophins are key regulators of neuronal survival and differentiation during development.
Activation of their cognate receptors, Trk receptors, a family of receptor tyrosine kinases (RTKs), is pivotal for mediating the downstream functions of neurotrophins.
Recent studies reveal that cyclin-dependent kinase 5 (Cdk5), a serine/threonine kinase, may modulate RTK signaling through phosphorylation of the receptor.
Given the abundant expression of both Cdk5 and Trk receptors in the nervous system, and their mutual involvement in the regulation of neuronal architecture and synaptic functions, it is of interest to investigate if Cdk5 may also modulate Trk signaling.
In the current study, we report the identification of TrkB as a Cdk5 substrate.
Cdk5 phosphorylates TrkB at Ser478 at the intracellular juxtamembrane region of TrkB.
Interestingly, attenuation of Cdk5 activity or overexpression of a TrkB mutant lacking the Cdk5 phosphorylation site essentially abolishes brain-derived neurotrophic factor (BDNF)–triggered dendritic growth in primary hippocampal neurons.
In addition, we found that Cdk5 is involved in BDNF-induced activation of Rho GTPase Cdc42, which is essential for BDNF-triggered dendritic growth.
Our observations therefore reveal an unanticipated role of Cdk5 in TrkB-mediated regulation of dendritic growth through modulation of BDNF-induced Cdc42 activation.
Author Summary

Accurate transmission of information in the nervous system requires the precise formation of contact points between neurons.
Regulation of these contact sites involves fine tuning the number and branching of dendritic processes on neurons.
Throughout development, several secreted factors act to regulate dendrite number and branching.
One important family of these factors is neurotrophins, which are indispensable for the survival and development of neurons.
For example, stimulation of hippocampal neurons with one neurotrophin, brain-derived neurotrophic factor (BDNF), increases the number of dendrites directly extending from the cell body.
Here, we report that BDNF-stimulated dendritic growth requires phosphorylation of the BDNF receptor, TrkB, by a kinase known as cyclin-dependent kinase 5 (Cdk5).
Inhibiting phosphorylation of TrkB by Cdk5 essentially abolishes the induction of dendrites by BDNF.
Our observations reveal that Cdk5 serves as a regulator of neurotrophin function.
Since Cdk5 and neurotrophins both play essential roles in neuronal development, our findings suggest that the interplay between Cdk5 and TrkB may also be implicated in the regulation of other biological processes during development.
Dendritic growth stimulated by brain-derived neurotrophic factor (BDNF) requires phosphorylation of the BDNF receptor, TrkB, by a kinase known as cyclin-dependent kinase 5 (Cdk5).
This study identifies a novel interplay between Cdk5 and TrkB.
Introduction
Neurotrophins are indispensable for multiple aspects of neuronal development, such as the maintenance of neuronal survival, regulation of neuronal architecture, and synaptic plasticity.
Members of the neurotrophins include the prototypic member nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin (NT)–3, and NT-4/5.
Downstream responses of neurotrophins are transduced by a family of receptor tyrosine kinases (RTKs) known as Trks, and also the low-affinity neurotrophin receptor p75.
Although all neurotrophins bind p75, they associate with different Trk receptors with rather remarkable selectivity.
NGF interacts selectively with TrkA, while BDNF and NT-4/5 bind preferentially to TrkB.
NT-3, on the other hand, associates with TrkC with high affinity, although it also binds TrkA and TrkB with low affinity.
Similar to other RTKs, activation of Trks leads to dimerization and autophosphorylation of the receptors, followed by the recruitment and initiation of a myriad of signaling pathways including the Ras/MAPK, PI3K, and PLCγ pathways [1,2].
Interestingly, recent studies have demonstrated that activity of cyclin-dependent kinase 5 (Cdk5), a serine/threonine kinase, is required for the downstream actions of a RTK, ErbB.
Cdk5 was found to phosphorylate ErbB2/3, a phosphorylation that is essential for the activation of the receptors [3,4].
Cdk5 is a member of the cyclin-dependent kinase family, but it is unique in several aspects.
First of all, it is activated by the neural-specific non-cyclin activators p35 and p39.
Secondly, Cdk5 is not involved in the regulation of cell cycle control, but is implicated in neuronal migration, synapse functions/maintenance, and neuronal survival [5,6].
The importance of Cdk5 in neuronal development and migration is underscored by the aberrant phenotypes exhibited by mice lacking Cdk5 and its activators.
Cdk5 knockout mice and p35/p39 double knockout mice both exhibit perinatal death with severe cortical lamination defects [7,8].
Furthermore, swollen soma and nuclear margination is evident in Cdk5-deficient neurons, implicating Cdk5 as an essential regulator of neuronal survival [7].
Interestingly, truncation of the Cdk5 activator p35 into p25 has also been associated with prolonged Cdk5 activation in a number of neurodegenerative diseases [9], thus revealing that precise regulation of Cdk5 activity is essential for maintenance of neuronal survival [10].
Furthermore, an increasing number of studies are pointing to an essential role of Cdk5 at the synapse, where it is not only involved in the formation and maintenance of synapses, but is also indispensable for the regulation of synaptic transmission and synaptic plasticity [5].
While the mechanisms by which Cdk5 regulates such diverse functions remain to be unraveled, the identification of ErbB receptors as Cdk5 substrates suggests that Cdk5 may exert its biological effects by modulating signaling pathways downstream of RTK activation.
This piece of evidence, together with the abundant expression of Cdk5 and Trk receptors in the nervous system and their shared implication in a number of biological functions, prompted us to further examine if Cdk5 also regulates the signaling of Trk receptors.
In the current study, we report the identification of TrkB as a substrate of Cdk5.
More importantly, we found that Cdk5-mediated phosphorylation of TrkB is essential for BDNF-induced dendritic growth through the modulation of Cdc42 activity.
Our findings provide evidence for a crosstalk between the Cdk5 and neurotrophin signaling pathways, and lend further support to the idea that Cdk5 is a modulator of RTK signaling.
Results
TrkB Interacted with p35 and Cdk5
Given the increasing evidence implicating Cdk5 in the modulation of RTK signaling, we sought to examine if Cdk5 may also play a role in Trk signaling.
Literature search revealed that TrkA, TrkB, and TrkC all contain serine- or threonine-directed proline residues at the intracellular juxtamembrane region of the receptors, but only TrkB and TrkC contain Cdk5 consensus sites S/TPXK/H/R (Figure 1A).
To explore the potential interplay between Trk receptors and Cdk5, we first examined if Trk receptors associated with Cdk5 or p35.
TrkA, TrkB, or TrkC was overexpressed together with Cdk5 or p35 in COS7 cells, and immunoprecipitation was performed with Cdk5, p35, or pan-Trk antibody.
Interestingly, all three Trk receptors were observed to associate with Cdk5 (Figure 1B) and p35 (Figure 1C), while no association was observed when immunoprecipitation was performed with IgG control.
Since both TrkB and its ligand BDNF are abundantly expressed in the brain throughout development, we next proceeded to verify the interaction between TrkB and Cdk5/p35 in postnatal brains.
We found that TrkB associated with both p35 and Cdk5 in postnatal day 7 (P7) rat brain lysates (Figure 1D).
Furthermore, Flag-tagged Cdk5 pulled down TrkB from the membrane fraction of adult brain lysates (Figure 1E).
These observations collectively suggest that TrkB interacted with Cdk5/p35 in both postnatal and adult brains.
Since both p35 and Cdk5 are present in brain lysates and likely exist as a complex, the observed interaction between TrkB and Cdk5/p35 did not provide specific information on whether TrkB associated specifically with Cdk5 or p35.
To delineate between these two possibilities, the interaction between TrkB, p35, and Cdk5 was examined in p35+/+ and p35−/− brain lysates (Figure 1F).
Interestingly, we found that in the absence of p35, the association between Cdk5 and TrkB was essentially abolished, indicating that p35 was required for the association between Cdk5 and TrkB in vivo.
Cdk5 Phosphorylated TrkB at Ser478
We next proceeded to examine if Trk receptors, TrkB in particular, served as Cdk5 substrates using in vitro kinase assay.
TrkA, TrkB, and TrkC were overexpressed in COS7 cells and immunoprecipitated by pan-Trk antibody.
Incubation with Cdk5/p25 revealed that TrkB and TrkC, but not TrkA, were phosphorylated by Cdk5/p25 in vitro (Figure 2A).
This is in agreement with the lack of Cdk5 consensus sites in TrkA, and points to the possibility that Cdk5 may phosphorylate TrkB and TrkC at the Cdk5 consensus sites at the juxtamembrane region (Figure 1A).
To examine this possibility, a GST fusion protein containing only the juxtamembrane region of TrkB was prepared.
In vitro kinase assay verified that Cdk5/p35 phosphorylated TrkB at the juxtamembrane region (Figure 2B).
It has previously been proposed that p25 and p35 may confer different substrate specificities.
Results from our in vitro kinase assay suggested that Cdk5 phosphorylated TrkB regardless of whether it was activated by p25 or p35, although further studies will be required to delineate the relative contributions of p25 and p35 to endogenous phosphorylation of TrkB by Cdk5.
We were next interested in identifying the Cdk5 phosphorylation site(s) on TrkB.
Three TrkB-juxtamembrane region mutants were generated: TrkB M1, where Ser478 was mutated to alanine; TrkB M2, where Thr489 was mutated to alanine; and TrkB DM, where both Ser478 and Thr489 were mutated to alanine.
Interestingly, phosphorylation of the TrkB-juxtamembrane region was almost completely abolished when Cdk5/p25 was incubated with TrkB M1 or TrkB DM (Figure 2C), thus revealing that Ser478 was required for Cdk5-mediated phosphorylation of the TrkB-juxtamembrane region.
We further verified the importance of this site for Cdk5-mediated phosphorylation of TrkB by generating a phospho-specific TrkB antibody against Ser478.
Preincubation of the antibody with blocking peptide prevented detection of Ser478-phosphorylated TrkB, indicating that the antibody was sufficiently specific (Figure 2D).
Full-length TrkB mutants lacking the potential Cdk5 phosphorylation sites were overexpressed with or without Cdk5/p35 in HEK293T cells.
Interestingly, Ser478-phosphorylated TrkB was not observed in the absence of Cdk5/p35, indicating that Cdk5 was essential for the phosphorylation of TrkB at Ser478 in HEK293T cells.
More importantly, when TrkB mutants lacking Ser478 were expressed (TrkB M1 and TrkB DM), phosphorylation of TrkB at Ser478 was essentially abolished (Figure 2E).
Taken together, our observations indicate that Cdk5 phosphorylated TrkB at Ser478 at the juxtamembrane region of TrkB.
Ser478 Phosphorylation of TrkB Required Cdk5 Activity In Vivo
To further examine if Cdk5 is essential for phosphorylation of TrkB at Ser478 in vivo, we examined the effect of inhibiting Cdk5 activity on phospho-Ser478 (p-Ser478) TrkB levels in cortical neurons.
We found that at basal level, TrkB was weakly phosphorylated at Ser478.
Interestingly, stimulation with BDNF led to a marked increase in p-Ser478 TrkB levels, indicating that phosphorylation of TrkB at Ser478 was at least in part ligand dependent.
Remarkably, treatment with Cdk5 selective inhibitor roscovitine (Ros) almost abrogated the BDNF-triggered increase in p-Ser478 TrkB levels (Figure 3A), suggesting that Cdk5 was involved in the BDNF-stimulated component of TrkB Ser478 phosphorylation.
To further establish the involvement of Cdk5 in Ser478 phosphorylation of TrkB in vivo, the levels of p-Ser478 TrkB in cdk5+/+ and cdk5−/− brain lysates were examined.
Importantly, we found that Ser478-phosphorylated TrkB was basically undetectable in Cdk5−/− brain lysates (Figure 3B).
Similarly, cortical neurons prepared from Cdk5−/− brains exhibited undetectable levels of p-Ser478 TrkB.
In addition, BDNF stimulation failed to trigger an increase in p-Ser478 TrkB levels (Figure 3C).
These observations strongly suggest that Cdk5 is essential for phosphorylation of TrkB at Ser478 in vivo, and that BDNF-stimulated increase in Ser478 phosphorylation of TrkB requires Cdk5 activity.
BDNF Treatment Enhanced Cdk5 Activity
Since BDNF stimulation was observed to increase Ser478 phosphorylation of TrkB, and Cdk5 was required for phosphorylating TrkB at Ser478, we were interested to examine if BDNF stimulation affects Cdk5 activity.
BDNF has previously been observed to increase Cdk5 activity after 3 d of BDNF stimulation in cortical neurons [11].
In agreement with this observation, we found that BDNF treatment led to an increase in Cdk5 activity within 15 min of BDNF stimulation (Figure 4A).
More importantly, addition of Trk inhibitor K252a essentially abolished BDNF-triggered increase in Cdk5 activity, indicating that the increase in Cdk5 activity was dependent on TrkB activation (Figure 4B).
It has previously been demonstrated that Cdk5 activity is enhanced by phosphorylation at Tyr15 [12].
Given the activation of tyrosine kinase activity of TrkB upon ligand stimulation, we were interested to investigate if BDNF treatment leads to phosphorylation of Cdk5 at Tyr15, thereby enhancing its activity.
We found that BDNF stimulation enhanced association between Cdk5 and TrkB in cortical neurons (Figure 4C).
More importantly, in vitro kinase assay using purified TrkB and Cdk5 revealed that TrkB phosphorylated Cdk5 at Tyr15 (Figure 4D and 4E).
TrkB-mediated phosphorylation of Cdk5 was abolished with the addition of Trk inhibitor K252a, further verifying that Tyr15 phosphorylation of Cdk5 was TrkB dependent (Figure 4E).
These observations collectively indicate that upon BDNF stimulation, Cdk5 was recruited to TrkB and phosphorylated by TrkB at Tyr15, thus leading to enhanced Cdk5 activity to promote phosphorylation of TrkB at Ser478.
Ser478 Phosphorylation of TrkB Was Required for BDNF-Stimulated Dendritic Growth
Given the neural-specific nature of Cdk5, its abundant expression throughout development, and its essential role in the phosphorylation of TrkB at Ser478, we were interested in examining the biological significance of this phosphorylation on the downstream functions of BDNF/TrkB signaling.
As a first step, we examined if Cdk5-mediated phosphorylation of TrkB affects TrkB activation and downstream signaling cascades.
Interestingly, we found that inhibition of Cdk5 activity by Cdk5 selective inhibitor Ros only marginally affected tyrosine phosphorylation of TrkB and initiation of downstream signaling pathways including phosphorylation of Erk1/2, Akt, and CREB (data not shown).
Indeed, BDNF-stimulated increase in TrkB tyrosine phosphorylation was weakly affected in cdk5−/− cortical neurons (Figure 3C).
Furthermore, activation of Akt and Erk1/2 following BDNF stimulation was also comparable in cdk5+/+ and cdk5−/− cortical neurons (data not shown).
Our observations thus revealed that Cdk5-mediated phosphorylation of TrkB did not significantly affect activation of the receptor, nor its initiation and recruitment of downstream signaling pathways.
Although Cdk5-mediated phosphorylation of TrkB had negligible effect on the downstream signaling of TrkB, it cannot be ruled out that Ser478 phosphorylation of TrkB is essential for the downstream functions of BDNF/TrkB signaling.
We thus sought to examine if Cdk5-mediated phosphorylation of TrkB affects its downstream functions.
BDNF has been observed to stimulate dendrite growth and development in hippocampal neurons [13,14].
In accordance with earlier observations, BDNF treatment led to a marked increase in the number of primary dendrites in hippocampal neurons (Figure 5A), although the length and branching of dendrites were not affected (data not shown).
Interestingly, treatment with Cdk5 selective inhibitor Ros almost completely abolished the BDNF-stimulated dendritic growth, without affecting the basal number of dendrites (Figure 5A).
Furthermore, overexpression of dominant negative (DN) Cdk5 (Figure 5B) and transfection with Cdk5 short interfering RNA (siRNA) (Figure 5C) both abrogated BDNF-induced increase in primary dendrites.
More importantly, BDNF similarly failed to induce an increase in primary dendrites in cdk5−/− hippocampal neurons (Figure 5D).
These observations collectively reveal that Cdk5 activity was required for BDNF-induced increase in primary dendrites in hippocampal neurons.
To verify the importance of Ser478 phosphorylation of TrkB in BDNF-triggered dendritic growth, TrkB wild-type (WT) or TrkB M1 was overexpressed in hippocampal neurons.
Remarkably, overexpression of TrkB M1 similarly abolished the BDNF-induced increase in primary dendrites (Figure 5E).
Taken together, our data indicate that Cdk5-mediated phosphorylation of TrkB at Ser478 was required for BDNF-triggered dendritic growth in hippocampal neurons.
Cdk5-Mediated Phosphorylation of TrkB Affected BDNF-Induced Dendritic Growth through Attenuation of Cdc42 Activity
Rho GTPases, including RhoA, Rac1, and Cdc42, are key regulators of actin cytoskeleton dynamics.
Since BDNF stimulation has been observed to activate Rac1 and Cdc42 in neurons [15], we were interested to delineate if Rho GTPases contribute to BDNF-stimulated dendritic growth.
To investigate if Rho GTPases are involved, and to identify the Rho GTPase(s) implicated, hippocampal neurons were transfected with WT or DN Rac1, Cdc42, or RhoA.
We found that while overexpression of WT and DN Rac1 increased the basal number of dendrites in the absence of BDNF treatment, overexpression of both forms of Rac1 abolished BDNF-stimulated dendritic growth.
On the other hand, while overexpression of DN RhoA slightly enhanced primary dendrites irrespective of BDNF stimulation, overexpression of both WT and DN forms of RhoA inhibited BDNF-stimulated dendritic growth.
Remarkably, in contrast to the inhibition of BDNF-stimulated dendritic growth in cells overexpressing WT Rac1 and RhoA, BDNF stimulation of hippocampal neurons overexpressing WT Cdc42 resulted in an increase in primary dendrites, which was nearly abolished by overexpression of DN Cdc42 (Figure 6A).
Our observations therefore suggest that while Rac1 and RhoA may also modulate BDNF-stimulated dendritic growth, it is the activation of Cdc42 following BDNF stimulation that most likely mediates the increase in primary dendrites by BDNF.
To examine if phosphorylation of TrkB by Cdk5 affects dendritic growth through modulating BDNF-triggered activation of Cdc42, we first examined if the BDNF-induced increase in Cdc42 activity was affected by treatment with Cdk5 selective inhibitor Ros.
In agreement with earlier findings, BDNF treatment resulted in an increase in Cdc42 activity.
Interestingly, treatment with Ros significantly reduced BDNF-induced Cdc42 activity in cortical neurons (Figure 6B), suggesting that Cdk5 activity was involved in BDNF-triggered activation of Cdc42.
To investigate if the reduction in Cdc42 activity contributes to the abrogation of BDNF-induced dendritic growth following attenuation of Cdk5 activity, the effect of overexpressing constitutively active (CA) Cdc42 with TrkB M1 on BDNF-induced dendritic growth was examined.
Remarkably, overexpression of CA Cdc42 reversed the abrogation of BDNF-induced dendritic growth by TrkB M1 (Figure 6C).
More importantly, while overexpression of CA Cdc42 had negligible effect on BDNF-stimulated increase in primary dendrites in Cdk5+/+ neurons, overexpression of CA Cdc42 similarly rescued the lack of dendritic growth in cdk5−/− neurons following BDNF stimulation (Figure 6D).
These observations strongly suggest that Cdk5-mediated phosphorylation of TrkB at Ser478 was essential for the BDNF-triggered increase in primary dendrites through modulating BDNF-induced Cdc42 activity.
Discussion
In the current study, we report the identification of TrkB as a novel Cdk5 substrate by providing evidence that Cdk5 phosphorylates TrkB at Ser478, located at the intracellular juxtamembrane region of the receptor.
The near absence of Ser478-phosphorylated TrkB in cdk5−/− brain underscores the importance of Cdk5 in this phosphorylation in vivo.
More importantly, we found that Cdk5-mediated phosphorylation of TrkB is required for BDNF-stimulated increase in primary dendrites.
Furthermore, we demonstrated that Cdk5 activity is involved in BDNF-induced increase in Cdc42 activity, which underlies BDNF-induced dendritic growth in hippocampal neurons.
Overexpression of CA Cdc42 restored BDNF-stimulated increase in primary dendrites in cdk5−/− neurons, lending further support that Cdk5-mediated phosphorylation of TrkB at Ser478 is essential for BDNF-induced Cdc42 activation and increase in primary dendrites.
Our findings therefore reveal an unanticipated role of Cdk5 in mediating downstream functions of Trk signaling.
Activation of Rho GTPases has been implicated in a number of functions downstream of neurotrophin stimulation.
For example, a recent study reported that synaptic maturation involves BDNF-stimulated increase in Cdc42 activity [16].
In addition, activation of Cdc42 is involved in the regulation of retinal growth cone filopodia by BDNF [17].
Activation of Rac1 following neurotrophin stimulation has also been observed to mediate neuronal migration triggered by neurotrophin treatment [18].
Our observation that BDNF-stimulated increase in Cdc42 activity contributes to the increase in primary dendrites corroborates these studies.
It is interesting to note that overexpression of WT and DN Rac1 and RhoA also inhibited BDNF-induced increase in primary dendrites.
While it is rather intriguing to observe similar actions by the WT and DN forms of these two Rho GTPases, our observation nonetheless suggests that Rac1 and RhoA may also play a role in BDNF-stimulated dendritic growth.
Further studies will be required to delineate their involvements in BDNF-dependent regulation of dendritic development.
Although different Rho GTPases have been identified as essential downstream mediators of neurotrophin functions, much less is known about the mechanisms by which neurotrophin treatment results in Rho GTPase activation, and how this process is regulated.
The activity of Rho GTPases is controlled by a number of factors.
Conversion from the GDP-bound, inactive state to the GTP-bound, active state is facilitated by guanine nucleotide exchange factors (GEFs).
The activated Rho GTPases then translocate to the plasma membrane, where they activate other downstream effectors such as PAK1 to modulate actin dynamics [19].
Indeed, neurotrophins have recently been observed to induce Rho GTPase activity through recruitment of a number of GEFs.
TrkA was demonstrated to bind to Kalirin, an association that is essential for NGF-induced Rac1 activation and neurite outgrowth [20].
Furthermore, NGF treatment induces plasma membrane translocation of the GEFs Vav2 and Vav3, an event that is required for activation of Rac1 and Cdc42 and the induction of neurite outgrowth following NGF treatment in PC12 cells [21].
NGF also stimulates activation of the Rac-specific GEF p-Rex1 in PC12 cells [18].
Two recent studies reveal that neurotrophin stimulation in Schwann cells also leads to Rho GTPase activation through activation of GEFs.
TrkC activation results in activation of the Cdc42-specific GEF Dbs [22] and Rac-specific GEF Tiam1 [23], both of which are required for NT-3-stimulated Schwann cell migration.
Finally, TrkB was also recently demonstrated to bind and phosphorylate Tiam1 to mediate a BDNF-triggered change in cell shape [24].
On the other hand, recent studies accentuate the importance of membrane recruitment of Rho GTPase to lipid rafts for the function of these Rho GTPases.
Lipid rafts are microdomains in plasma membrane rich in cholesterol and sphingolipids.
Targeting of activated Rac1 to lipid rafts is required for activation of downstream effector Pak1 [25].
More importantly, neurotrophin-triggered Rac1 activation and morphological changes in hippocampal neurons have also been observed to require localization of Rac1 to lipid rafts [26].
Finally, BDNF has also been observed to increase Cdc42 activity in cerebellar granule neurons through enhancing calcium influx following the activation of PLCγ and PI3K pathways, a series of events that are essential for BDNF-mediated growth cone turning [27].
While a number of mechanisms have been postulated to underlie neurotrophin-mediated activation of Rho GTPases, it appears that the mechanisms implicated may vary with different downstream functions of Trk activation and the GEF involved.
In the current study, we demonstrated that Ser478 phosphorylation of TrkB by Cdk5 is essential for the Cdc42-dependent increase in primary dendrites triggered by BDNF, thus adding a new regulatory component to the mechanisms involved in Rho GTPase activation by neurotrophin.
Although the precise downstream pathways by which this phosphorylation affects Cdc42 activation remains to be determined, our observations provide some interesting insights.
First of all, while inhibition of Cdk5-mediated TrkB phosphorylation at Ser478 essentially abolished BDNF-induced increase in primary dendrites, it was surprising to observe that Cdk5 activity had a negligible effect on TrkB activation and initiation of downstream signaling pathways.
This suggests that Cdk5 activity probably did not affect BDNF-dependent activation of Cdc42 and the induction of primary dendrites through modulating activation of downstream signaling.
This is unexpected because BDNF-stimulated increase in primary dendrites was previously observed to depend on PI3K/Akt pathways in cortical neurons [28].
Nonetheless, accumulating evidence reveals that the location at which Trk receptors are activated may play a pivotal role in determining the precise downstream significance of Trk activation.
For example, BDNF-induced increase in primary dendrites was recently demonstrated to involve TrkB activation in the lipid rafts [13].
In addition, retrograde transport of activated Trk receptors as signaling endosomes is emerging as a key regulator of neuronal survival [29].
Since we examined changes in TrkB downstream signaling cascades only in total lysates, it remains possible that Cdk5 activity may specifically affect TrkB signaling only at certain subcellular/plasma membrane compartments.
Secondly, overexpression of CA Cdc42 restored BDNF-induced dendritic growth in cdk5−/− neurons and in neurons overexpressing TrkB M1 (Figure 6), suggesting that maintenance of Cdc42 activation was sufficient to overcome the lack of BDNF-stimulated dendritic growth when Cdk5-mediated TrkB phosphorylation was absent.
It thus appears that Cdk5 may impair BDNF-induced Cdc42 activation by affecting activation of the Rho GTPase.
On the other hand, it should also be noted that overexpression of both the DN and CA forms of Cdc42 had a negligible effect on the basal number of primary dendrites in both cdk5+/+ and cdk5−/− neurons (Figure 6).
Our observation is in agreement with an earlier study demonstrating that overexpression of DN or CA Cdc42 had no effect on the number of primary dendrites in chick spinal neurons [30].
In addition, it is consistent with the observation that modulation of Cdk5 activity or overexpression of TrkB M1 affected only BDNF-induced dendritic growth, without affecting the basal number of dendrites.
Nonetheless, the inability of CA Cdc42 to mimic BDNF in the induction of primary dendrites suggests that activation of Cdc42 per se was insufficient to trigger dendritic growth in the absence of BDNF, and that additional, BDNF-dependent event(s) are required for the induction of dendritic growth by BDNF.
Although the precise pathways implicated remain to be identified, it is tempting, in light of the emerging importance of lipid rafts in the activation of Rho GTPase, to speculate that BDNF may be required to stimulate translocation of activated Cdc42 to lipid rafts.
In support of this hypothesis, it was observed that activation of Rac1 depends on the translocation of the activated Rho GTPase to lipid rafts [25,26].
In addition, in the absence of cholesterol, CA Rac1 failed to translocate to plasma membrane in fibroblasts [25].
More importantly, depletion of cholesterol similarly abolished BDNF-induced increase in primary dendrites in hippocampal neurons [13].
These observations collectively suggest that the inability of CA Cdc42 to increase dendritic growth in the absence of BDNF treatment may be related to the lack of CA Cdc42 translocation to lipid rafts, which may potentially be induced by BDNF treatment.
A thorough investigation of the importance of lipid rafts in Cdc42 activation and primary dendrite induction by BDNF will shed light on the mechanisms by which BDNF-triggered dendritic growth is regulated.
Given the near absence of Ser478-phosphorylated TrkB in cdk5−/− brain, we believe that Cdk5 functions as the predominant kinase for this phosphorylation in vivo.
Nonetheless, it was interesting to note that prior to BDNF stimulation, a basal level of Ser478-phosphorylated TrkB was detected in cortical neurons that was not inhibited by pretreatment with the Cdk5 inhibitor Ros.
This may suggest that other serine kinases are present to phosphorylate TrkB at Ser478 in the absence of BDNF stimulation.
Nonetheless, given the marked inhibition of BDNF-stimulated increase in TrkB phosphorylation by Ros, we believe that Cdk5 is essential for the BDNF-dependent component of TrkB phosphorylation at Ser478.
Given the abundant expression of Cdk5 and TrkB in neurons throughout development, and their respective concentration at the synapse, it would be interesting to examine if Cdk5 activity is also involved in other downstream functions of TrkB signaling, such as the regulation of neuronal survival and synaptic plasticity.
Preliminary findings from our laboratory reveal that Cdk5 activity is also required for BDNF-stimulated neuronal survival in cortical neurons (unpublished data).
In addition, the juxtamembrane region of Trk receptors has been associated with the regulation of Trk receptor internalization [31] and degradation [32].
Further investigation of whether this phosphorylation also affects the internalization and degradation of the receptor would provide further insights into the biological significance of this phosphorylation.
In addition, since Cdk5 was observed to associate with TrkA without phosphorylating the receptor, further delineation of the consequences of this interaction would be essential for thoroughly understanding the crosstalk between Trk receptors and Cdk5.
A preliminary study revealed that, similar to TrkB, TrkA phosphorylates Cdk5 at Tyr15 (unpublished data).
The differential interaction of TrkA and TrkB with Cdk5, together with the differential localization of TrkA and TrkB in different neuronal populations, may provide a novel mechanism by which Cdk5 can regulate the signaling of different neuronal populations.
In conclusion, our findings have provided evidence for a regulatory role of Cdk5 in Trk-induced dendritic growth, and lend support for an emerging role of Cdk5 as a regulator of RTK signaling.
Given the importance of neurotrophin/Trk signaling in almost all aspects of neuronal development and function, our findings will likely have far-reaching implications for further elucidating the signaling mechanisms involved in the regulation of neuronal survival, synapse formation, and synaptic plasticity.
Materials and Methods
Antibodies, DNA constructs, and siRNAs.
The antibodies against Trk (C-14), Cdk5 (DC-17), p35, and Shc were purchased from Santa Cruz Biotechnology (http://www.scbt.com).
The antibodies against TrkB and SH2B were from BD Biosciences (http://www.bdbiosciences.com).
The polyclonal antibodies recognizing phospho-TrkA (Tyr490), p44/42 mitogen-activated protein kinase (Erk1/2), phospho-p44/42 mitogen-activated protein kinase, AKT, phospho-AKT (Ser473), CREB, and phospho-Ser133 CREB were obtained from Cell Signaling Technology (http://www.cellsignal.com).
Antibodies specific for actin and β-tubulin type III were from Sigma-Aldrich (http://www.sigmaaldrich.com).
Antibody against the p-Ser478 of TrkB was raised by synthetic peptide (CISNDDDSApSPLHHIS; Bio-Synthesis, http://www.biosyn.com) and purified using AminoLink Kit (Pierce, http://www.piercenet.com).
Expression vectors of p35, Cdk5, and DN Cdk5 were prepared as previously described [3].
Flag-tagged and GST-tagged Cdk5 were generated by PCR, and subcloned into the mammalian expression vectors pcDNA3 (Invitrogen, http://www.invitrogen.com) and pGEX-6P-1 (Amersham Biosciences, http://www5.amershambiosciences.com), respectively.
HA-tagged and GST-tagged Rac1, Cdc42, and RhoA constructs were gifts from Yung-Hou Wong (Hong Kong University of Science and Technology, Hong Kong).
The expression vectors of TrkA, TrkB, and TrkC were constructed as described [33].
Three TrkB mutants lacking the potential Cdk5 phosphorylation sites were constructed by mutating Ser478 (TrkB M1), Thr489 (TrkB M2), or both Ser478 and Thr489 (TrkB DM) to alanine using the overlapping PCR technique, followed by subcloning into pcDNA3.
GST-TrkB-Juxta construct was generated by PCR and subcloned into pGEX-6P-1.
Protein purification was performed according to the manufacturer's protocol.
Stealth RNAi molecules for Cdk5 were prepared as previously described [34].
The sequences used were: Cdk5 siRNA, CCUCCGGGAGAUCUGUCUACUCAAA; and control siRNA (Cdk5), CCUAGGGCUAGCUGUUCAUCCCAAA.
Animals, primary cultures, and transfection.
Cdk5 and p35 knockout mice were kindly provided by A.
B.
Kulkarni (National Institutes of Health, Bethesda, Maryland) and T.
Curran (St.
Jude Children's Research Hospital, Memphis, Tennessee), and L.
H.
Tsai (Harvard Medical School, Boston, Massachusetts), respectively.
Mice from different stages were collected and genotyped as described [7,35].
Rat cortical and hippocampal neuron cultures were prepared as previously described [33,34].
Subsequent to digestion with 0.25% trypsin in Hank's Balanced Salt Solution without Ca2+ and Mg2+ at 37 °C for 5 min, the reaction was stopped by 2.5% heat-inactivated horse serum.
The dissociated neurons were seeded in culture dishes coated with 10 μg/ml poly-D-lysine.
Two hours later the medium was replaced by neurobasal medium supplemented with 2 mM L-glutamine and 2% B27 supplement.
Selective Cdk5 inhibitor Ros (Calbiochem, http://www.merckbiosciences.com/html/CBC/home.html) was used to inhibit Cdk5 activity in primary neuron cultures.
Primary cultures at 3 d in vitro (DIV3) were treated with or without BDNF (50 ng/ml) in the presence of Ros (10 or 25 μM) or DMSO for 3 d before harvesting or fixation.
For transfection of primary cultures, cortical and hippocampal neurons were seeded on coverslips in 12-well dishes at a cell density of 2 × 105 per coverslip.
Neurons were transfected using calcium phosphate precipitation at DIV3.
Twenty-four hours after transfection, the cultures were treated with BDNF for 3 d.
Primary hippocampal neuron cultures on coverslips in 12-well dishes were seeded at a cell density of 5 × 104 per coverslip for siRNA transfection.
Cultures were transfected at DIV3 with Lipofectamine 2000 transfection reagent following the manufacturer's protocols (Invitrogen).
The transfected cells were incubated at 37 °C with 5% CO2 for 24 h before treatment, and were then treated with BDNF for 3 d.
Cell cultures and transfection.
COS7 cells and HEK293T cells were obtained from American Type Culture Collection (http://www.atcc.org).
Both cells were maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum, penicillin (50 units/ml), and streptomycin (100 μg/ml) at 37 °C with 5% CO2.
COS7 cells and HEK293T cells were transfected using Lipofectamine Plus transfection reagents following the supplier's instructions (Invitrogen).
The cells were treated and harvested 24 h after transfection.
Protein extraction, immunoprecipitation, in vitro pull-down assay, and Western blot analysis.
Cells were lysed at 4 °C for 30 min in lysis buffer (RIPA: 1× PBS, 1% NP40, 0.1% SDS, and 0.5% sodium deoxycholate) with various protease inhibitors (1 mM phenylmethylsulfonyl fluoride [PMSF], 1 mM sodium orthovanadate [NaOV], 2 μg/ml antipain, 10 μg/ml leupeptin, 30 nM okadaic acid, 5 mM benzamidine, and 10 μg/ml aprotinin).
Brain tissues were homogenized in lysis buffer (0.5% NP-40, 20 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, and 1 mM NaF [pH 7.5]) supplemented with various protease inhibitors (1 mM PMSF, 1 mM NaOV, 2 μg/ml antipain, 10 μg/ml leupeptin, 30 nM okadaic acid, 5 mM benzamidine, and 10 μg/ml aprotinin).
Proteins were resolved by SDS-PAGE and subsequently electro-transferred onto a nitrocellulose membrane.
Immmunoblots were probed with the desired primary antibodies at 4 °C overnight.
After washing with TBS-T, the corresponding HRP-conjugated secondary antibody was added and incubated for 2 h at room temperature.
Proteins were then visualized using enhanced chemiluminescence Western blotting detection reagents with reference to the supplier's instructions (Amersham Biosciences).
For immunoprecipitation, 1–2 mg of protein lysates was incubated with 1 μg of the corresponding antibody at 4 °C overnight with rotation.
Forty microliters of protein G Sepharose (Amersham Biosciences) pre-washed with 1× PBS was added and rotated at 4 °C for 1 h.
After intense washing with the lysis buffer, the immunoprecipitated protein and its associated proteins were analyzed by SDS-PAGE and Western blotting.
Flag-tagged protein was overexpressed in COS7 cells and the cell lysate was obtained as described above.
The cell lysate obtained was incubated with anti-Flag M2 affinity gel (Sigma-Aldrich) at 4 °C overnight with rotation.
The Flag-tagged protein was pulled down by the affinity gel, and the affinity gel was washed twice with lysis buffer (0.5% NP-40, 20 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, and 1 mM NaF [pH 7.5]) with various protease inhibitors (1 mM PMSF, 1 mM NaOV, 2 μg/ml antipain, 10 μg/ml leupeptin, 30 nM okadaic acid, 5 mM benzamidine, and 10 μg/ml aprotinin).
One to two milligrams of proteins prepared from brain tissues was incubated with the affinity gel, and the Flag-tagged protein pulled down by the affinity gel for 1 h.
The affinity gel was washed twice with lysis buffer supplemented with protease inhibitors.
The proteins pulled down by the Flag-tagged protein were subjected to Western blot analysis.
In vitro kinase assay.
Recombinant Cdk5/p35 and Cdk5/p25 were kindly provided by Shin-Ichi Hisanaga (Tokyo Metropolitan University, Tokyo).
TrkA, TrkB, and TrkC were immunoprecipitated from transfected HEK293T cells, and used as substrates for reconstituted Cdk5/p35 or Cdk5/p25 in the in vitro kinase assay.
The kinase assay was performed at 30 °C for 30 min in kinase buffer containing 100 μM [γ-32P] ATP as described [36].
To examine if TrkB phosphorylated Cdk5, recombinant TrkB kinase domain (Upstate Biotechnology, http://www.upstate.com) was incubated with GST-Cdk5 for 30 min at 30 °C, with or without Trk inhibitor K252a pretreatment (100 nM) for 10 min, in the presence of 100 μM [γ-32P] ATP or cold ATP.
To examine if BDNF stimulated Cdk5 activity, primary cortical neurons were treated with BDNF with or without 30 min of K252a pretreatment (100 nM).
The immunoprecipitated Cdk5/p35 complexes from the lysates were washed three times with lysis buffer and twice with kinase buffer.
The in vitro kinase reaction was performed at 30 °C for 30 min with kinase buffer containing 100 μM histone H1 peptide and 100 μM [γ-32P] ATP as described [37].
The phosphorylated proteins were resolved by SDS-PAGE.
After the gel was dried, the phosphorylated proteins were visualized by autoradiography.
For TrkB-mediated Cdk5 phosphorylation, the phosphorylated protein was resolved by SDS-PAGE, and blotted with phospho-Cdk2 (Tyr15; Santa Cruz Biotechnology) or phosphotyrosine antibody (4G10; Upstate Biotechnology).
GTPase activity assay.
GTPase activity was measured as described [38].
Briefly, cultured cortical neurons at DIV7 were pretreated with DMSO or Ros for 30 min, followed by treatment with BDNF for another 5 min.
Cells were lysed at 4 °C and incubated with Pak1-PBD agarose with constant rocking at 4 °C for 1 h.
The proteins bound to the beads were washed three times with lysis buffer at 4 °C, eluted in SDS sample buffer, and analyzed for bound Cdc42 by Western blotting using monoclonal antibody against Cdc42 (Upstate Biotechnology).
GTPase activity was quantified by densitometry analysis of the blots.
Immunohistochemical analysis.
Following fixation in 4% paraformaldehyde and 5% sucrose in PBS with Ca2+ and Mg2+ for 30 min, the cells were washed three times with PBS, and were blocked with 1% bovine serum albumin and 10% goat serum for 20 min.
The cells were then incubated with the corresponding primary antibody (1:150–500) at 4 °C overnight, and were subsequently washed with PBS three times.
Following incubation with FITC or rhodamine conjugated secondary antibody (1:1,000) for 1 h at room temperature, the cells were washed again, stained with DAPI, and mounted with coverslips and MOWIOL (Calbiochem).
Mounted cells were visualized under fluorescent microscope (Leica, http://www.leica.com).
Statistical analysis.
All data were expressed as mean ± standard deviation.
Statistical significance was determined by one-way analysis of variance followed by Bonferroni's post hoc test with 95% confidence.
A p-value of smaller than 0.05 was considered as statistically significant.
Abbreviations
BDNF
brain-derived neurotrophic factor
CA
constitutively active
Cdk5
cyclin-dependent kinase 5
DIV[number]
[number] days in vitro
DN
dominant negative
GEF
guanine nucleotide exchange factor
NaOV
sodium orthovanadate
NGF
nerve growth factor
NT-[number]
neurotrophin-[number]
p-Ser478
phospho-Ser478
P7
postnatal day 7
PMSF
phenylmethylsulfonyl fluoride
Ros
roscovitine
RTK
receptor tyrosine kinase
siRNA
short interferring RNA
WT
wild-type
References
Trk receptors: Roles in neuronal signal transduction
Selectivity in neurotrophin signaling: Theme and variations
Cdk5 is involved in neuregulin-induced AChR expression at the neuromuscular junction
Cyclin-dependent kinase-5 is involved in neuregulin-dependent activation of phosphatidylinositol 3-kinase and Akt activity mediating neuronal survival
Synaptic roles of Cdk5: Implications in higher cognitive functions and neurodegenerative diseases
A decade of CDK5
Targeted disruption of the cyclin-dependent kinase 5 gene results in abnormal corticogenesis, neuronal pathology and perinatal death
p35 and p39 are essential for cyclin-dependent kinase 5 function during neurodevelopment
Cdk5 deregulation in the pathogenesis of Alzheimer's disease
Cdk5: Mediator of neuronal death and survival
Brain-derived neurotrophic factor-induced phosphorylation of neurofilament-H subunit in primary cultures of embryo rat cortical neurons
Fyn and Cdk5 mediate semaphorin-3A signaling, which is involved in regulation of dendrite orientation in cerebral cortex
BDNF-induced recruitment of TrkB receptor into neuronal lipid rafts: Roles in synaptic modulation
Cyclic AMP controls BDNF-induced TrkB phosphorylation and dendritic spine formation in mature hippocampal neurons
Signalling and crosstalk of Rho GTPases in mediating axon guidance
Activity-induced rapid synaptic maturation mediated by presynaptic cdc42 signaling
Cdc42 participates in the regulation of ADF/cofilin and retinal growth cone filopodia by brain derived neurotrophic factor
Involvement of a Rac activator, P-Rex1, in neurotrophin-derived signaling and neuronal migration
GEF means go: Turning on RHO GTPases with guanine nucleotide-exchange factors
Critical role for Kalirin in nerve growth factor signaling through TrkA
Local phosphatidylinositol 3,4,5-trisphosphate accumulation recruits Vav2 and Vav3 to activate Rac1/Cdc42 and initiate neurite outgrowth in nerve growth factor-stimulated PC12 cells
The neurotrophin-3 receptor TrkC directly phosphorylates and activates the nucleotide exchange factor Dbs to enhance Schwann cell migration
Ras activation of a Rac1 exchange factor, Tiam1, mediates neurotrophin-3-induced Schwann cell migration
TrkB binds and tyrosine-phosphorylates Tiam1, leading to activation of Rac1 and induction of changes in cellular morphology
Integrins regulate Rac targeting by internalization of membrane domains
Biological activity of neurotrophins is dependent on recruitment of Rac1 to lipid rafts
Ca2+-dependent regulation of rho GTPases triggers turning of nerve growth cones
BDNF regulates primary dendrite formation in cortical neurons via the PI3-kinase and MAP kinase signaling pathways
Long-distance retrograde neurotrophic signaling
Cdc42 stimulates neurite outgrowth and formation of growth cone filopodia and lamellipodia
Association of the atypical protein kinase C-interacting protein p62/ZIP with nerve growth factor receptor TrkA regulates receptor trafficking and Erk5 signaling
Down-regulation of the neurotrophin receptor TrkB following ligand binding.
Evidence for an involvement of the proteasome and differential regulation of TrkA and TrkB.
SLAM-associated protein as a potential negative regulator in Trk signaling
STAT3 as a downstream mediator of Trk signaling and functions
Mice lacking p35, a neuronal specific activator of Cdk5, display cortical lamination defects, seizures, and adult lethality
Pctaire1 interacts with p35 and is a novel substrate for Cdk5/p35
Cloning of three novel neuronal Cdk5 activator binding proteins
Leukemia inhibitory factor receptor signaling negatively modulates nerve growth factor-induced neurite outgrowth in PC12 cells and sympathetic neurons
Figures and Tables
TrkB Interacted with Cdk5 and p35
(A) TrkA, TrkB, and TrkC all contain proline-directed serine/threonine residues in the juxtamembrane region of the receptors (indicated by arrows).
Nonetheless, only TrkB and TrkC contain Cdk5 consensus sites S/TPXK/H/R.
(B) Cell lysates from HEK293T cells overexpressing Cdk5 and TrkA, TrkB, or TrkC were immunoprecipitated (IP) with Cdk5 antibody and immunoblotted with pan-Trk antibody.
TrkA, TrkB, and TrkC were all observed to associate with Cdk5.
(C) Cell lysates from HEK293T cells overexpressing p35 and TrkA, TrkB, or TrkC were immunoprecipitated with p35 antibody and immunoblotted with pan-Trk antibody.
TrkA, TrkB, and TrkC were all observed to associate with p35.
(D) Brain lysate from P7 rat brain was immunoprecipitated with pan-Trk, p35, or Cdk5 antibody and immunoblotted with p35, Cdk5, and TrkB antibodies.
Rabbit normal IgG was used as a control.
TrkB was observed to associate with both p35 and Cdk5 in P7 rat brain.
(E) The membrane fraction of adult brain lysates was incubated with or without Flag-tagged Cdk5.
Flag-tagged Cdk5 pulled down TrkB from the membrane fraction of adult brain lysates.
(F) Brain lysates from P7 p35+/+ or p35−/− mouse brains were immunoprecipitated with p35 and Cdk5 antibodies and immunoblotted with p35, Cdk5, and TrkB antibodies.
Rabbit normal IgG served as a control.
Association between Cdk5 and TrkB was abolished in p35−/− brain, indicating that p35 was required for the association between Cdk5 and TrkB.
Cdk5 Phosphorylated TrkB at Ser478
(A) Lysates from COS7 cells overexpressing TrkA, TrkB, and TrkC were immunoprecipitated with pan-Trk antibody and incubated with Cdk5/p25 in an in vitro kinase assay.
TrkB and TrkC, but not TrkA, were phosphorylated by Cdk5/p25.
(B) GST-TrkB-juxtamembrane fusion protein was incubated with increasing amount of Cdk5/p35 and subjected to an in vitro kinase assay.
Histone H1 served as control to verify the activity of the Cdk5 kinase.
The TrkB-juxtamembrane region was phosphorylated by Cdk5/p35 in a dose-dependent manner.
(C) Purified WT GST-TrkB-juxtamembrane fusion protein and mutants (M1, M2, and DM) were incubated with Cdk5/p25 in an in vitro kinase assay.
While WT and M2 were strongly phosphorylated by Cdk5/p25, phosphorylation of M1 and DM were markedly attenuated.
Quality of the purified GST and GST-fusion proteins used in the GST pull-down assay was verified by Coomassie blue staining.
(D) Characterization of p-Ser TrkB antibody raised against phosphorylated Ser478 of TrkB.
TrkB was overexpressed with or without p35/Cdk5 in HEK293T cells.
Preincubation of purified p-Ser478 TrkB antibody with blocking peptide completely abolished detection of Ser478 phosphorylation of TrkB.
(E) Full-length TrkB WT, M1, M2, and DM were overexpressed with or without Cdk5/p35 in HEK293T cells.
In the absence of Cdk5/p35, Ser478-phosphorylated TrkB (p-Ser TrkB) was not detected.
Overexpression of Cdk5/p35 resulted in phosphorylation of TrkB WT at Ser478, but phosphorylation at Ser478 was essentially abolished when TrkB M1 and DM were overexpressed.
IP, immunoprecipitation.
Ser478 Phosphorylation of TrkB Required Cdk5 Activity In Vivo
(A) BDNF stimulation resulted in an increase in p-Ser478 TrkB (p-Ser TrkB) levels in cortical neurons.
Treatment with Cdk5 selective inhibitor Ros (25 μM) inhibited the BDNF-induced increase in p-Ser478 TrkB, although Ros treatment also resulted in a slight increase in basal p-Ser478 TrkB.
(B) cdk5+/+ and cdk5−/− brain lysates were immunoblotted against TrkB, phospho-TrkB at Ser478, and β-actin as loading control.
p-Ser478 TrkB was almost completely absent in cdk5−/− brain, indicating the importance of Cdk5 in the phosphorylation of TrkB at Ser478 in vivo.
(C) Cortical neurons isolated from cdk5+/+ and cdk5−/− brain were treated with BDNF for different periods.
Interestingly, while BDNF enhanced TrkB Ser478 phosphorylation in cdk5+/+ cortical neurons, TrkB Ser478 phosphorylation was not detected in cdk5−/− neurons, nor did BDNF stimulation enhance Ser478 phosphorylation, indicating that BDNF-stimulated increase in TrkB Ser478 phosphorylation requires Cdk5 activity.
BDNF Enhanced Cdk5 Activity
(A) Cortical neurons were stimulated with BDNF for different time intervals.
Lysates were immunoprecipitated (IP) with p35 antibody and subjected to in vitro kinase assay using histone H1 as substrate.
BDNF stimulation for 15 min resulted in a marked increase in Cdk5 activity in cortical neurons.
Quantification of the changes in phospho-Histone H1 level following BDNF stimulation was normalized to the value obtained from untreated cultures (time 0) and is shown in the histogram.
*, p < 0.05.
(B) Addition of Trk inhibitor K252a abolished BDNF-induced increase in Cdk5 activity.
Cortical neurons were pretreated with vehicle control (DMSO) or K252a for 30 min before stimulation with BDNF for 15 min.
Lysates were immunoprecipitated with p35 antibody and subjected to in vitro kinase assay using histone H1 as substrate.
We found that K252a pretreatment markedly reduced the increase in Cdk5 activity triggered by BDNF stimulation, indicating that the induction of Cdk5 activity was dependent on TrkB activation.
Quantification of the changes in phospho-Histone H1 level following BDNF stimulation in the presence or absence of K252a treatment was normalized to the value obtained from untreated cultures (time 0) and is shown in the histogram.
*, p < 0.05.
(C) Cortical neurons were treated with BDNF for 20 min.
Lysates were immunoprecipitated with p35 antibody and immunoblotted with TrkB, p35, or Cdk5 antibody.
While association between Cdk5 and p35 was not affected by BDNF stimulation, association between p35 and TrkB increased following 20 min of BDNF stimulation.
(D) Recombinant TrkB was incubated with GST-Cdk5 in an in vitro kinase assay.
TrkB was found to phosphorylate GST-Cdk5 (middle lane).
(E) GST-Cdk5 and recombinant TrkB were pretreated with vehicle (DMSO) or K252a for 10 min, subjected to in vitro kinase assay, and immunoblotted with antibodies against phospho-tyrosine (p-Tyr) and the Tyr15 phosphorylated form of Cdk5 (pTyr15 Cdk5).
Cdk5 was phosphorylated by TrkB at Tyr15.
Addition of K252a abolished phosphorylation of Cdk5 by TrkB, further verifying that Cdk5 was phosphorylated by TrkB.
Attenuation of Cdk5 Activity Abolished BDNF-Induced Increase in Primary Dendrites in Hippocampal Neurons
(A) Hippocampal neurons were stimulated with BDNF for 3 d in the presence or absence of Ros (10 μM).
Interestingly, while BDNF treatment markedly enhanced the number of primary dendrites, treatment with Ros abrogated the increase.
(B) Hippocampal neurons were transfected with Cdk5 or DN Cdk5.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Overexpression of DN Cdk5 abolished the BDNF-induced increase in primary dendrites.
(C) Hippocampal neurons were transfected with Cdk5 siRNA or control siRNA.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Transfection with Cdk5 siRNA attenuated Cdk5 expression in hippocampal neurons.
More importantly, BDNF-induced increase in primary dendrites was abrogated in Cdk5 siRNA–transfected cells.
(D) Hippocampal neurons isolated from cdk5+/+ and cdk5−/− brains were treated with BDNF for 3 d.
BDNF treatment failed to enhance primary dendrites in Cdk5−/− neurons.
(E) Hippocampal neurons were transfected with TrkB WT or TrkB M1.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Overexpression of TrkB M1 markedly reduced the BDNF-induced increase in primary dendrites.
Scale bar = 10 μm.
*, p < 0.05.
Cdk5-Mediated Phosphorylation of TrkB Affected BDNF-Induced Dendritic Growth through Attenuation of Cdc42 Activity
(A) Hippocampal neurons were transfected with the WT or DN form of Rac1, RhoA, or Cdc42.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Overexpression of DN Cdc42 markedly reduced BDNF-induced increase in primary dendrites compared to overexpression of WT Cdc42, indicating that Cdc42 may contribute to the BDNF-dependent induction of dendritic growth in hippocampal neurons.
(B) Cortical neurons were pretreated with Cdk5 selective inhibitor Ros or vehicle (DMSO) for 30 min prior to treatment with BDNF for different time intervals.
Ros pretreatment markedly reduced BDNF-induced increase in Cdc42 activity following 15 and 30 min of BDNF treatment, indicating that Cdk5 activity was involved in BDNF-triggered Cdc42 activation.
Quantification of the changes in Cdc42 activity following BDNF stimulation with or without Ros pretreatment was normalized to the value obtained for the DMSO-treated group at time 0 and is shown in the histogram.
*, p < 0.05.
(C) TrkB WT or TrkB M1 mutant were co-transfected with WT or CA Cdc42 in hippocampal neurons.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Overexpression of CA Cdc42 reversed the abrogation of BDNF-induced increase in primary dendrites following overexpression of TrkB M1.
*, p < 0.05.
(D) Hippocampal neurons isolated from cdk5+/+ and cdk5−/− brains were transfected with WT or CA Cdc42.
Twenty-four hours after transfection, cells were exposed to BDNF for 3 d.
Overexpression of CA Cdc42 rescued the lack of dendritic growth following BDNF treatment in Cdk5−/− hippocampal neurons.
*, p < 0.05.
Competing interests.
The authors have declared that no competing interests exist.
Author contributions.
ZHC, WHC, YC, YPN, and NYI designed the experiments.
ZHC, WHC, YC, and YPN performed the experiments.
ZHC, WHC, YC, YPN, and NYI analyzed the data.
NYI contributed reagents/materials/analysis tools.
ZHC, WHC, YPN, and NYI wrote the paper.
Funding.
The study was supported in part by the Research Grants Council of Hong Kong (HKUST 6130/03M, HKUST 6119/04M, HKUST 3/03C, and HKUST 6431/06M) and the Area of Excellence Scheme of the University Grants Committee (AoE/B-15/01).The interaction of PP1 with BRCA1 and analysis of their expression in breast tumors
Background
The breast cancer susceptibility gene, BRCA1, is implicated in multiple cellular processes including DNA repair, the transactivation of genes, and the ubiquitination of proteins; however its precise functions remain to be fully understood.
Identification and characterization of BRCA1 protein interactions may help to further elucidate the function and regulation of BRCA1.
Additionally, detection of changes in the expression levels of BRCA1 and its interacting proteins in primary human breast tumors may further illuminate their role in the development of breast cancer.
Methods
We performed a yeast two-hybrid study to identify proteins that interact with exon11 of BRCA1 and identified Protein Phosphatase 1β (PP1β), an isoform of the serine threonine phosphatase, PP1.
GST-pull down and co-immunoprecipitation assays were performed to further characterize this interaction.
Additionally, Real-Time PCR was utilized to determine the expression of BRCA1, PP1α, β and γ in primary human breast tumors and normal breast tissue to identify alterations in the expression of these genes in breast cancer.
Results
PP1 and BRCA1 co-immunoprecipitate and the region within BRCA1 as well as the specific PP1 interacting domain mediating this interaction were identified.
Following mRNA expression analysis, we identified low levels of BRCA1 and variable levels of PP1α and β in primary sporadic human breast tumors.
Furthermore, BRCA1, PP1β and PP1γ were significantly higher in normal tissue specimens (BRCA1 p = 0.01, PP1β: p = 0.03, PP1γ, p = 1.9 × 10-6) compared to sporadic breast tumor samples.
Interestingly, we also identified that ER negative tumors are associated with low levels of PP1α expression.
Conclusion
The identification and characterization of the interaction of BRCA1 with PP1 and detection of changes in the expression of PP1 and genes encoding other BRCA1 associated proteins identifies important genetic pathways that may be significant to breast tumorigenesis.
Alterations in the expression of genes, particularly phosphatases that operate in association with BRCA1, could negatively affect the function of BRCA1 or BRCA1 associated proteins, contributing to the development of breast cancer.
Background
The breast cancer susceptibility gene, BRCA1, is involved in the development of a significant proportion of familial breast and ovarian cancers and may also play a role in the development of sporadic breast cancer [1,2].
The BRCA1 protein has an amino terminal RING finger domain, potentially involved in ubiquitination, and two BRCA1 C-terminal domains (BRCT), which interact with a number of proteins involved in the transactivation of genes.
The central portion of BRCA1, mainly encoded by exon11, contains two nuclear localization signals and interacts with proteins involved in DNA repair as well as the centrosomal protein, γ-tubulin [3].
Evidence suggests that BRCA1 functions as a scaffold, coordinating DNA repair [4], the transactivation of genes [5] and centrosome separation or maturation [6].
The phosphorylation of BRCA1 may be important for its function throughout the cell cycle.
BRCA1 is hypophosphorylated during G1/S [7]; however, it becomes phosphorylated from S to G2/M phase.
BRCA1 also interacts with kinases involved in cell cycle control and apoptosis such as the cyclin/cdk complexes [7], and a hypophosphorylated form of BRCA1 has been found at the centrosomes during mitosis [8].
The vertebrate serine/threonine protein phosphatase, PP1, has 3 isoforms: α, β (also known as δ) and γ that are highly conserved across their large catalytic domain, but are divergent at the amino- and carboxy-termini.
Regulatory proteins bind to the unique carboxy termini of the PP1 isoforms to direct their isoform specific activities.
For example, PP1γ knock-out mice have impaired spermiogenesis, indicating that the other PP1 isoforms are not entirely able to compensate for the loss of PP1γ; and the presence of polyploid spermatids in these mice suggests a defect in meiosis [9].
Furthermore, the PP1α isoform has also been found to associate with BRCA1 [10].
We have used a yeast two-hybrid assay to detect proteins that interact with exon11 of BRCA1.
This large exon encodes roughly 60% of the protein, and we wished to identify potentially important interacting proteins outside of the intensely studied RING and C-terminal regions of BRCA1.
The identification of PP1β as a BRCA1 interacting protein and characterization of the interaction of BRCA1 with PP1α, β and γ both in vitro and in vivo, provides new insights into the roles of both BRCA1 and the PP1 isoforms.
These biochemical results led to analysis of the expression of BRCA1 as well as PP1α, β and γ, which has provided unique evidence for their potential deregulation in breast cancer.
Disruption of the interaction of BRCA1 with PP1, or deregulation of its expression, could result in changes in the balance of kinase and phosphatase activities in the cell, leading to tumorigenesis.
Methods
Two-hybrid screening in yeast
Sequences from 900 to 4000 bp of BRCA1, were ligated into the DNA binding domain (DBD) yeast expression vector pAS1 (GAL4 (1–147) DNA-BD, TRP1, ampr).
The yeast strains pJ69-4A (MATa trp1-901, leu2-3, 112 ura3-52, his3-200, gal4Δ, gal80Δ, LYS2::gal1-HIS3, GAL2-ADE2, met2::GAL7-lacZ) and Y187 (MATα, ura3-52, his3-200, ade2-101, trp1-901, leu2-3, 112, gal4Δ, met-, gal80Δ, URA3::GAL1UAS-GAL1TATA-lacZ) were utilized in this study.
Yeast were initially transformed with the DBD-exon11 plasmid, followed by the Human Mammary Gland cDNA library cloned into the pACT2 Activation domain plasmid (Clontech).
A total of 5 × 105 clones were tested.
Co-immunoprecipitation
Coding sequences of PP1α, β or γ were ligated into the pCMVFlag (Sigma) expression vector.
HEK293T cells were transfected with either of the PP1 vectors or the negative control pFLagLaf4 [11].
Cells were lysed using NETN buffer (150 mM NaCl, 1 mM EDTA, 20 mM Tris pH8.0, 0.5% NP40) along with the protease inhibitor cocktail 1 and phosphatase inhibitor cocktail 1 (Sigma); lysates were incubated with 1 μg/ml BRCA1 Ab1 and 3 (Oncogene) or 1 μg/mlM2-Flag (Sigma) rotating overnight at 4°C.
25 μl of protein G-agarose beads (Santa Cruz) were added and samples were incubated at 4°C for 2 hours.
Antibodies used for Westerns were: 1 μg/ml BRCA1 Ab1 plus 1 μg/ml Ab3 (Oncogene) or 1 μg/ml M2 Flag (Sigma).
Construction and expression of bacterial GST fusion proteins
Inserts of BRCA1 were generated using HiFiTaq Polymerase (Invitrogen), using the BRCA1 primers described in Scully et al.
[12].
BR-4 V-A and F-A were generated using in vitro PCR mutagenesis [13] (Primers used were 1forward: GAAAGATCTGTAGAGAGTAGC, 1reverse (V-A): CAAAAGTGGCTTTTG GACTTTG, 1reverse (F-A): CATTCAGCAGTGACTTTTGGAC, 2forward (V-A): CCAAAAGCCACTTTTGAATGTG, 2forward (F-A): GTCACTGCTGAATGTGAAC, 2reverse: CCACTTCATTAGTACTGGAACC).
pLysS bacterial cells, transformed with the BRCA1 constructs and induced for protein expression, were lysed using B-Per bacterial cell lysis reagent (Pierce) plus 1 mM PMSF and DNAse (200U/ml, Invitrogen) according to manufacturer's directions.
Lysates were incubated with 50 μl of a 50% GST slurry overnight at 4°C.
The amount of GST-bound protein was determined by comparison with known amounts of BSA loaded onto the gel, with amounts ranging from 0.1ug to 2ug of BSA.
In vitro GST-interaction studies
1 μg of GST-bound protein was added to HEK293T cell lysate (1 mg of protein per reaction), and incubated rotating at 4°C for 2 hours.
GST pellets were washed 3X with 1 mL of NETN + protease and phosphatase inhibitors (Sigma) and GST-bound proteins were eluted using SDS PAGE loading dye.
Samples were electrophoresed on a 12% SDS-PAGE gel and transferred to a nitrocellulose membrane.
Immunoblots were probed with 1 μg/ml of the PP1 E-9 antibody (Santa Cruz) and bands were visualized using a Fluor-S MultiImager (BioRad).
Binding of PP1 to BR-4 V-A or F-A was compared with binding of PP1 to wild-type BR4 in order to determine the relative degree of interaction.
Tissue samples
Twenty-nine tumors from sporadic breast cancers were obtained as part of a prospective study of molecular alterations in auxiliary node-negative disease [14].
Following frozen section diagnosis of invasive cancer, tumor specimens were sampled by a pathologist and immediately snap frozen and stored in liquid nitrogen.
DNA and RNA have previously been extracted from these tumors by conventional techniques [15].
Normal breast mRNA was obtained, by a pathologist, from adjacent tissue surrounding breast tumors and was isolated as outlined above.
Real-time PCR expression analysis
cDNA was reverse transcribed from 250 ng of RNA using MMLV-RT and random hexamers as directed (Invitrogen).
cDNA was amplified using the Applied Biosystems Taqman Universal PCR Master Mix (no UNG) (cat# 4324018), 8 μl of diluted cDNA (1 μl of cDNA + 7 μl of ddH20), 1 μl of 20 × Assay on Demand gene expression assay mix (HPRT1 control primer/probe mix) and 1 μl of 20 × Assay on Demand gene expression assay mix (test primer/probe mix- catalogue numbers are listed below).
Thermal cycling consisted of a hold at 95°C for 10 minutes followed by 40 cycles of 95°C for 15 seconds and 60°C for 60 seconds, using the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems (ABI)).
Standard curves of serial dilutions ranging from 3 × to 0.005 × of a pool of cDNA were used for quantification, and slopes generated by the standard curves ranged from 3.3 to 3.6.
Test probes were conjugated to the FAM™ fluor (Applied Biosystems), and the HPRT1 endogenous control probe was conjugated to the VIC™ fluor (Applied Biosystems).
The HPRT-1 internal control, which has been determined by our lab and others to have little variation between breast tumors [16,17], was chosen to normalize for variations in mRNA amount and quality between tumors.
We compared the expression levels of the genes to that of the HPRT-1 housekeeping gene for each sample, and all mRNA/HPRT-1 expression levels were approximately one when comparing a pool of cell line cDNA for each experiment to prevent variation between experiments.
For our purposes, mRNA/HPRT-1 ratios of less than 0.5 × the mean level of expression were considered to be under-expressed, while mRNA/HPRT-1 ratios of greater than 2 × the mean level of expression were considered to be over-expressed.
The test and control probes were multiplexed, and analysis was performed on ratios of the test quantity mean values to control quantity mean values for each sample.
ABI SDS2.1 software was used to analyze the results.
The primer/probe pairs used for the experiments were purchased from ABI (cat#: PP1α: Hs00267568-m1, PP1β: Hs00160343-m1, PP1γ: Hs00160351-m1, BRCA1: HS 00173233-m1, HPRT-1: 4326321E_(endogenous control)).
Statistical analysis of tumor characteristics and their association with PP1α and PP1β expression
A descriptive analysis was performed, comparing frequency distributions of tumor characteristics between PP1 groups ('high': greater than 2.2 or 'low': less than or equal to 2.2), using contingency tables.
The median level of expression of PP1α, in all tumors analyzed (familial (not shown) and sporadic), was used to establish the level for high vs.
low PP1 expression.
Association of each characteristic with PP1 expression was investigated by Fisher's exact test [18].
All statistical analyses were performed using SAS statistical software, version 8.2 (SAS Inc., Cary, NC, USA).
p < 0.05 was considered to be statistically significant.
Cell culture conditions and transfections
Transfections were performed using Fugene transfection reagent (Roche) according to manufacturer's directions.
HEK293T cells were acquired from the American Type Tissue Collection and were grown under their recommended conditions.
Results
Association of BRCA1 with PP1 in vitro
We performed a yeast two-hybrid assay to identify proteins that interact with exon11 of BRCA1.
In this study, 9 putative positives were identified including PP1β, an isoform of the serine/threonine phosphatase PP1 (not shown).
These initial yeast two-hybrid studies led to further examination on the interaction of BRCA1 with PP1β.
Overlapping fragments of BRCA1 (Figure 1A) were created to identify the region of BRCA1 required to interact with PP1 and we determined that BR-4 (amino acids 758 to 1064) mediated the interaction with PP1 (Figure 1B).
Sequence analysis of the BR-4 fragment identified a putative PP1 interacting domain, (K/R)/(V/I)/XF [19,20], which is present in several regulatory proteins that interact with PP1.
Similarly to other PP1 interacting domains [19,21], there are 2 positively charged amino acids upstream (893KK894), and a group of negatively charged amino acids downstream (902ECEQKEE908) of the BRCA1 KVTF sequence, further supporting its classification as a putative PP1 interacting domain.
GST-pull down assay to identify the region of BRCA1 interacting with PP1.
(A) Fragments used in GST pull down assays (BR 2 to 5) are diagrammed.
(B) Gel depicting co-precipitation of GST-bound BR-4 with PP1.
Following incubation of GST-BRCA1 proteins with equal amounts of cell lysate, a western blot was performed and probed with an antibody to the catalytic region of PP1.
(C) Analysis of the effect of mutations of the KVTF PP1 interacting domain on the BRCA1- PP1 interaction.
GST-bound-BR4 V-A and GST-bound-BR4 F-A binds PP1 with decreased intensity, compared to WT GST-bound-BR4.
Mutation of the valine to alanine, or phenylalanine to alanine, within the (K/R)/(V/I)/XF PP1 binding domain, has been shown to affect the binding of PP1 with other interacting proteins [22].
When BR-4 containing a V-A or F-A alteration was analyzed for its ability to interact with PP1, a significant decrease was observed in the amount of PP1 that interacted with the BR-4 V-A or F-A GST fusion protein compared to the wild type BR-4 fragment (Figure 1C), indicating that this sequence is required to mediate the interaction between BRCA1 and PP1.
Association of BRCA1 and PP1 in vivo
To establish that the association between BRCA1 and PP1 is a physiological interaction, HEK293T cells were transfected with Flag-epitope tagged PP1 α, β or γ or pFLAG Laf4 [11].
Laf4 was chosen as a negative control since it is expressed in the nucleus, but was not anticipated to interact with BRCA1.
We hypothesized that, although BRCA1 was shown to interact with PP1β in the two-hybrid screen, it might be capable of interacting with all three PP1 isoforms PP1 due to their high degree of sequence similarity.
Protein was immunoprecipitated using antibodies against BRCA1, or an antibody against the Flag epitope to immunoprecipitate Flag epitope tagged PP1α, β, or γ.
BRCA1 coimmunoprecipitated all three PP1 isoforms, and conversely, PP1 α, β and γ coimmunoprecipitated BRCA1 (Figure 2), indicating that the interaction between BRCA1 and PP1 is specific.
Coimmunoprecipitation of BRCA1 and PP1.
HEK293T kidney cells were transfected with vectors encoding untagged BRCA1 under the control of a CMV promoter, and vectors encoding Flag-PP1α, β, γ or Flag-Laf4.
(A) A western blot probed with BRCA1 shows that immunoprecipitation of protein with an antibody against the Flag-PP1α, β or γ proteins, but not Laf4, co-immunoprecipitates BRCA1 (Lanes 1A-1D).
It should be noted that the band observed slightly lower than BRCA1 in lane 1D is a non-specific background band.
Lanes 1E-1H show immunoprecipitation of BRCA1 using antibodies against the amino and carboxy termini of BRCA1.
(B) A western blot probed with an antibody against the Flag epitope.
Lanes 2A-2D indicate immunoprecipitation of the Flag-epitope tagged PP1α, β or γ or Flag-Laf4.
Lanes 2E-2G show co-immunoprecipitation of Flag-PP1α, β or γ with antibodies against BRCA1, and lane 2H shows a lack of coimmunoprecipitation of the negative control Flag-Laf4 by BRCA1.
Analysis of the expression of PP1α, β and γ, and genes encoding other BRCA1 associated proteins in breast tumors
Our investigations into the interaction of BRCA1 with PP1 led us to examine the expression levels of BRCA1, PP1α, β and γ in primary human breast tumors.
We determined that the expression levels of PP1α and β were variable in breast tumors, with several tumors exhibiting very low or high expression, relative to the mean level of expression observed for these genes (Table 1).
In contrast, PP1γ levels were less variable, with most tumors exhibiting expression levels that were not notably outside the mean level of PP1γ expression.
Interestingly, when analyzing levels of PP1α, β and γ in normal breast tissue (Table 2), we observed that levels of PP1β and γ were significantly higher in normal tissue specimens (PP1β: p = 0.03, PP1γ, p = 1.9 × 10-6) compared to sporadic breast tumor samples.
A similar trend for decreased PP1α expression was also observed, although significance was not reached for this isoform.
Distribution of gene expression in Primary Sporadic Human Breast Tumors
   Gene Name Number of Tumors (N) Mean expression level (M) N < 0.5 × M N > 2 × M   PP1α  24 3.1 4 (17%) 2 (8.4%)  PP1β  26 2.8 5 (19 %) 5 (19 %)  PP1γ  23 1.3 1 (4.3%) 1 (4.3%)  BRCA1 25 0.6 7 (28%) 2 (8%)
1test gene/HPRT-1 expression ratios as measured by quantitative real-time PCR
Gene Expression Levels in Normal Breast Tissue Compared to Sporadic Breast Tumors
   Name of Gene Tissue Type Mean Expression Level1 Significance (p)   BRCA1 Normal 0.9 0.01   Sporadic 0.6   PP1α  Normal 3.6 0.29   Sporadic 3.1   PP1β  Normal 3.9 0.03   Sporadic 2.8   PP1γ  Normal 3.2 1.9 × 10-6   Sporadic 1.3 
1.
mRNA/HPRT-1 expression ratios as measured by quantitative real-time PCR, relative to the pool of cDNA.
BRCA1 has previously been reported to have a decreased level of expression in tumors, relative to normal breast tissue [2].
We also observed that 28% of the tumors we analyzed had low levels of BRCA1, and 8% of those tumors had extremely low to non-detectable levels of BRCA1 expression.
Not surprisingly, BRCA1 expression levels in normal breast tissue were significantly higher those seen in sporadic breast tumors (p = 0.01).
Tumor characteristics and association with PP1α expression
To identify an association with the characteristics of the breast cancer cases used in this study and their level of PP1α and β expression, contingency-table Fisher's exact tests were performed.
The group of breast tumors with a 'low' level of PP1α mRNA was compared to those with 'high' PP1α expression.
Although no correlations were observed in age, menopausal status, tumor size, histologic grade, PgR status and lymphatic invasion, we did detect an association between low PP1α expression and ER status (p = 0.02).
Tumors with 'low' PP1α levels were more likely to be negative for ER receptor than those with 'high' PP1α levels (87% vs.
25%).
No correlations were observed between 'high' or 'low' PP1β expression and any of the tumor characteristics outlined above.
Discussion
BRCA1 interacts with a large number of proteins, and may function as a scaffold protein to bring cell cycle or DNA repair processes together, but the mechanisms by which BRCA1 functions in these processes have yet to be fully elucidated.
We performed a yeast two-hybrid study to identify proteins that interact with exon11 of BRCA1.
The region of BRCA1 encoded by exon 11 is known to interact with a number of proteins involved in DNA repair [23], as well as γ-tubulin [3] and several kinases including Aurora-A kinase [24] and ChkII [25].
Identification of additional interacting partners, particularly ones that could modify the activity of a BRCA1 through changes in phosphorylation, may further aid in clarifying its function and regulation.
In this yeast two-hybrid study, we identified the serine/threonine phosphatase PP1β as a BRCA1 interacting protein, which could have important consequences on both the activity of BRCA1 and the regulation of PP1β activity.
PP1 has 3 isoforms encoded by different genes that are 97% conserved across their catalytic domains and distinct roles for each isoform have yet to be determined.
When we coimmunoprecipitated Flag-epitope tagged PP1α, β or γ with BRCA1, we observed that all 3 isoforms interacted with BRCA1.
Additionally, we have identified the functional PP1 interacting domain within BRCA1.
This domain is found in other PP1 regulatory proteins, suggesting that BRCA1 may regulate the activity of PP1 and could act as a scaffold protein to promote the dephosphorylation of BRCA1 associated proteins by PP1.
The expression levels of BRCA1 and the PP1 isoforms were analyzed in primary human breast tumors.
Low levels of BRCA1 mRNA were identified, consistent with decreased expression rather than mutation as a method for BRCA1 inactivation in these tumors.
Additionally, we observed variable levels of expression for PP1α and β, but not PP1γ.
Interestingly, decreased levels of PP1β and γ were identified when comparing their gene expression levels from normal breast tissue with expression levels from sporadic breast tumors (Table 2).
This decreased expression may lead to perturbations of PP1 protein levels, altering the balance of kinase and phosphatase activities acting on specific substrates and potentially disrupting important cellular functions.
The use of primary human breast tumors allowed us to examine correlations between tumor and patient characteristics with PP1α or β expression.
Interestingly, we observed a statistically significant association between the level of PP1α expression and estrogen receptor (ER) status using our small sample of 24 sporadic tumors, which will need to be investigated further in a larger sample set.
ER turnover is affected by a basal level of phosphorylation that is maintained through a balance of kinase and phosphatase activities [26].
Activation of protein phosphatase PP2A has recently been shown to increase the level of ER mRNA stability [27].
Our results indicate that tumors with low expression of PP1α are more likely to be ER negative than tumors with high expression of PP1α (87% vs.
25%) and it is possible that PP1α has a role in ER mRNA stability, similarly to that of PP2A.
Conclusion
We have characterized the interaction of PP1 with BRCA1 and have identified a PP1 binding domain within BRCA1 that is necessary for this interaction to occur.
Furthermore, expression of the PP1 isoforms as well as several genes encoding BRCA1 interacting proteins was analyzed in primary human invasive breast tumors.
We detected low levels of BRCA1 expression in 28% of tumors, respectively, and variable levels of PP1α and β.
Moreover, significant decreases in expression were observed for BRCA1, PP1β and PP1γ when comparing normal breast tissue with invasive breast cancers.
Additionally, a significant association of PP1α levels with ER status in breast tumors was identified that could lead to additional studies into the effect of PP1α on ER mRNA stability and its role in the development of breast cancer.
The studies presented here provide evidence for an important role for PP1 in the development of breast cancer, possibly through its association with BRCA1, and suggest that deregulation of the balance of kinase and phosphatase activity in the cell may be an important component of breast tumorigenesis.
Abbreviations
BRCA1: Breast cancer susceptibility gene 1, BRCT (BRCA1 Carboxy-terminal domain), AD: Activating Domain, DBD: DNA binding domain, MEF: Mouse embryonic fibroblast, PP1: Protein Phosphatase 1
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
SLW performed the biochemical analysis for the interaction of BRCA1 and PP1 and designed the expression studies, interpreted results and drafted the manuscript.
LB-C performed the Real-Time expression analysis of gene expression, DP performed statistical analysis on the results and ILW supervised all work and aided in the drafting of the manuscript.
All authors have read and approved the final manuscript.
Pre-publication history
The pre-publication history for this paper can be accessed here:
A role for BRCA1 in sporadic breast cancer
Preferential allelic expression can lead to reduced expression of BRCA1 in sporadic breast cancers
Identification of a gamma-tubulin-binding domain in BRCA1
Association of BRCA1 with the hRad50-hMre11-p95 complex and the DNA damage response
BRCA1 interacts with components of the histone deacetylase complex
Roles of BRCA1 in centrosome duplication
BRCA1 is a 220-kDa nuclear phosphoprotein that is expressed and phosphorylated in a cell cycle-dependent manner
BRCA1 is associated with the centrosome during mitosis
Spermiogenesis is impaired in mice bearing a targeted mutation in the protein phosphatase 1cgamma gene
Regulation of BRCA1 phosphorylation by interaction with protein phosphatase 1alpha
LAF-4 is aberrantly expressed in human breast cancer
Association of BRCA1 with Rad51 in mitotic and meiotic cells

neu/erbB-2 amplification identifies a poor-prognosis group of women with node-negative breast cancer.
Toronto Breast Cancer Study Group.

Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes
Statistical modeling for selecting housekeeper genes

Structural basis for the recognition of regulatory subunits by the catalytic subunit of protein phosphatase 1
Rapid identification of protein phosphatase 1-binding proteins by mixed peptide sequencing and data base searching.
Characterization of a novel holoenzymic form of protein phosphatase 1.
Structural basis of protein phosphatase 1 regulation
Identification of binding sites on protein targeting to glycogen for enzymes of glycogen metabolism
BASC, a super complex of BRCA1-associated proteins involved in the recognition and repair of aberrant DNA structures
BRCA1 phosphorylation by Aurora-A in the regulation of G2 to M transition
hCds1-mediated phosphorylation of BRCA1 regulates the DNA damage response
Various phosphorylation pathways, depending on agonist and antagonist binding to endogenous estrogen receptor alpha (ERalpha), differentially affect ERalpha extractability, proteasome-mediated stability, and transcriptional activity in human breast cancer cells
Protein phosphatase 2A regulates estrogen receptor alpha (ER) expression through modulation of ER mRNA stabilityStructure of the APPL1 BAR-PH domain and characterization of its interaction with Rab5
APPL1 is an effector of the small GTPase Rab5.
Together, they mediate a signal transduction pathway initiated by ligand binding to cell surface receptors.
Interaction with Rab5 is confined to the amino (N)-terminal region of APPL1.
We report the crystal structures of human APPL1 N-terminal BAR-PH domain motif.
The BAR and PH domains, together with a novel linker helix, form an integrated, crescent-shaped, symmetrical dimer.
This BAR–PH interaction is likely conserved in the class of BAR-PH containing proteins.
Biochemical analyses indicate two independent Rab-binding sites located at the opposite ends of the dimer, where the PH domain directly interacts with Rab5 and Rab21.
Besides structurally supporting the PH domain, the BAR domain also contributes to Rab binding through a small surface region in the vicinity of the PH domain.
In stark contrast to the helix-dominated, Rab-binding domains previously reported, APPL1 PH domain employs β-strands to interact with Rab5.
On the Rab5 side, both switch regions are involved in the interaction.
Thus we identified a new binding mode between PH domains and small GTPases.
Introduction
Endocytosis induced by ligand−receptor interaction has been directly linked to signal transduction mediated by Rab5 and its effector APPL1 (Adaptor protein containing PH domain PTB domain and Leucine zipper motif; Miaczynska et al, 2004; Mao et al, 2006).
The small GTPase Rab5 is a generally acknowledged prominent regulator of vesicle trafficking enroute from the plasma membrane to early endosomes (Li, 1996), whereas APPL1 (also called DIP13α) is identified with signaling pathways of adiponectin, insulin, EGF, follicle stimulating hormone receptor, neurotrophin receptor (TrkA), oxidative stress, and DCC-mediated apoptosis (Liu et al, 2002; Miaczynska et al, 2004; Lin et al, 2006; Mao et al, 2006; Varsano et al, 2006; Nechamen et al, 2007).
Within this milieu, APPL1 specifically binds to the GTP-bound, active form of Rab5.
In response to extracellular stimuli, Rab5 hydrolyzes its bound GTP, releasing APPL1 from an endocytic structure, and allowing APPL1 to further interact with components of nucleosome remodeling and histone deacetylase complexes.
The interaction with Rab5 is essential for APPL1 localization to the endosomes and is indispensable for the functional cycle of APPL1 (Miaczynska et al, 2004).
Human APPL1, a multidomain protein 709 amino-acid (aa) residues in length contains an amino (N)-terminal BAR (Bin1/Amphiphysin/RVS167) domain and a PH (pleckstrin homology) domain followed by a carboxy (C)-terminal PTB (phosphotyrosine binding) domain (Sakamuro et al, 1996; Liu et al, 2002; Miaczynska et al, 2004).
The Rab5-binding site is located in the N-terminal BAR-PH region (Miaczynska et al, 2004), while the C-terminal region is found to interact with a host of other proteins, including the adiponectin receptor (Mao et al, 2006), Akt2/PKBβ kinase (Mitsuuchi et al, 1999), tumor suppressor DCC (Liu et al, 2002), TrkA, and TrkA interacting protein GIPC1 (Lin et al, 2006).
Based on aa sequence analysis, BAR domains have been identified in many proteins involved in intracellular trafficking, but sequence homology is low in general among known BAR domains (Farsad et al, 2001; Habermann, 2004).
The BAR domain typically contains three long kinked α-helices (α1, α2, and α3) that form a well-packed, crescent-shaped, symmetrical, six-helix bundle, side-by-side antiparallel homodimer; a structure proposed to exert its function as a convex membrane-curvature sensor or stabilizer.
The concave surface of the BAR dimer is proposed to bind preferentially to a negatively charged, curved membrane largely through electrostatic interactions.
Furthermore, some BAR domains have been found to bind to small GTPases, a class of intracellular molecular switches (Tarricone et al, 2001; Habermann, 2004); thus, their membrane association is directly linked to regulation of signal transduction and trafficking.
However, currently available structural information suggests that bindings of the BAR domain to GTPases and to membrane lipids are incompatible, because both interactions appear to compete for the same concave surface region of the BAR dimer (Tarricone et al, 2001).
The BAR domain of APPL1 is required for Rab5 binding and membrane recruitment (Miaczynska et al, 2004), although the mechanisms remain to be elucidated.
The PH domain is approximately 100-residue long, and has been identified in over 100 different eukaryotic proteins such as kinases, isoforms of phospholipase C (PLC), GTPases, and their regulators; most of which participate in cell signaling and cytoskeletal regulation (Rebecchi and Scarlata, 1998).
Despite their minimal sequence homology, the three-dimensional (3D) structures of PH domains are remarkably conserved.
They possess a common core consisting of seven β-strands and a C-terminal α-helix (Rebecchi and Scarlata, 1998).
Some PH domains specifically bind to phosphatidylinositol phosphates, suggesting that one possible function of this family is to anchor the host proteins to membranes.
PH domains are also suggested to bind to the Gβγ complex of the heterotrimeric G protein, protein kinase C, and small GTPases.
Nevertheless, none of these functions is absolutely conserved.
For instance, the PH domain of APPL1 alone is insufficient for binding to the membrane (Miaczynska et al, 2004).
The PH domain immediately follows the C-terminus of the BAR domain; such a BAR-PH motif is essential for Rab5 binding.
The same motif has also been found in a homolog Rab5 effector APPL2, centaurin-β family members, GRAF2, and oligophrenin (Habermann, 2004), but the 3D structure organization of BAR-PH motif and its functional implication remained elusive until now.
In order to address the functional roles of the BAR-PH motif in APPL1 and related proteins, we have carried out structure-function studies on human APPL1 and determined the crystal structures of the Rab5-binding region of APPL1 as well as the BAR domain alone.
The results show that two BAR-PH molecules form an integrated, symmetric homodimer, and the PH domain has extensive intermolecular interactions with the BAR domain.
The BAR dimer of APPL1 has a stronger curvature than other reported BAR structures.
Further mutagenesis analyses allowed us to identify the binding sites on both APPL1 and Rab5.
In sharp contrast to the presumed conflict between concurrent membrane association and GTPase binding by the BAR dimer (Habermann, 2004), the novel binding mode of the BAR-PH dimer should permit simultaneous interactions with both.
Results
Protein expression and crystallography
Recombinant proteins of human APPL1 N-terminal fragments including the BAR (residues 5−265) and BAR-PH domains (residues 5−385) were expressed in Escherichia coli, then purified using His tag affinity chromatography.
The samples were crystallized after removing the tag with thrombin, which generated a four-residue (Gly–Ser–His–Met) peptide N-terminal to the native Asp5 residue.
The BAR domain crystal diffracted up to 1.8-Å resolution on a beamline at the Argonne Advanced Photon Source (APS) synchrotron facility.
The crystal belongs to P21212 space group.
Phases of the structural factors were determined using the Se-Met-based single-wavelength anomalous dispersion (SAD) method (Supplementary data).
There is one APPL1 BAR molecule per asymmetric unit, with ∼41% solvent content.
Regions of the N-terminus (up to Thr12), Leu75−Asp79, and C-terminus (i.e., Pro260−Asp265) were missing from the final refined model because of lack of interpretable electron density.
The BAR-PH crystal diffracted to 2.05-Å resolution at the synchrotron facility.
The crystal also belongs to P21212 space group.
There is one APPL1 BAR-PH molecule per asymmetric unit, with ∼45% solvent content.
Phases of this crystal form were calculated using a combination of molecular replacement and SAD methods, and further improved with density modification.
Regions of N-terminal non-native tripeptide (i.e., Gly–Ser–His), Gly76−Asp78, Asn288−Ser295, and C-terminus (i.e., Ser380−Glu385) lacked interpretable electron density and were omitted from the final refined model.
Data collection and refinement statistics are summarized in Table I.
BAR domain structure and dimerization
From the two crystal forms of APPL1 peptides, we obtained two crystallographically independent BAR domain models.
In both cases, the APPL1 BAR domain has three long helices, namely α1, α2, and α3.
In addition, the APPL1 BAR domain contains an extra nine-turn α-helix, α4 (Figure 1; Supplementary Figure 1).
The two models could be superimposed onto each other with a moderate, 1.3-Å, Cα-atom root mean square deviation (r.m.s.d.) if flexible terminal and loop regions (i.e., residues 5−18, 75−79, 151−153, and 255−265) were omitted.
Thus, the overall structure of the BAR domain remains the same either alone or in the context of BAR-PH motif.
In each of the two crystal forms, two BAR molecules form a tightly packed dimer, which assumes a crescent-like shape, a hallmark of the BAR dimer structure (Figure 1).
In the BAR dimer, the helix α1 forms an antiparallel helix bundle with its symmetry counterpart, giving shape to the concave surface of the crescent-like dimer.
Helix α4 packs against α3 of the symmetry mate on the convex side of the dimer, and its C-terminus points to the tip of the crescent.
Over 4400 Å2 solvent accessible surface (SAS) from each protomer is buried in the dimer interface.
The addition of each α4 helix to the canonical BAR motif results in approximate 1900 Å2 buried SAS on the two protomers, corresponding to over 40% of the total buried SAS.
Although the overall folding of APPL1 BAR domain is similar to previously reported BAR domain 3D structures (i.e., arfaptin2, PDB file 1I4T; amphiphysin, 1URU; and endophilin, 1ZWW), those structures are in general more similar to each other than to the APPL1 BAR domain.
For instance, using 150 Cα atoms of the common helical regions, the r.m.s.d.
values between the dimer of APPL1 BAR and 1I4T, 1URU, and 1ZWW were 3.7, 3.8, and 4.4 Å, respectively, while those among 1I4T, 1URU, and 1ZWW range between 2.4 and 2.6 Å.
In addition, the APPL1 α1 and α2 helices lack extensive patches of positively charged aa residues on the concave surface (Figure 2); such patches are thought to be essential for some BAR containing proteins to induce in vitro tubule formation (Carlton et al, 2004).
The curvature of the concave face of the BAR dimer is thought to play an important role in membrane bending and/or curvature sensing (Habermann, 2004).
We implemented a computing algorithm to calculate the curvature radius (rC) and found that the APPL1 BAR dimer has an rC about 55 Å (Supplementary data; Supplementary Figure 2), significantly smaller than the rC values of other BAR dimers (Peter et al, 2004).
Thus, the APPL1 BAR dimer has the strongest curvature among known BAR dimer structures.
Structure of the APPL1 PH domain
The APPL1 PH domain encompasses residues Asn276−Leu379 and has a typical PH folding (Supplementary Figure 3).
The core structure of PH domain consists of a pair of nearly orthogonal β-sheets of four and three antiparallel β-strands (β1–β2–β3–β4 and β5–β6–β7; Supplementary Figure 1).
The C-terminal α-helix, αC, packs against both β-sheets and contributes to the core of the domain.
In the PH domain, connecting loops are named after the preceding β-strands (e.g., the loop between β1 and β2 is called L1, etc).
The canonical ligand-binding site is composed of β1, L1, β2, L3, and L6 (exemplified in the crystal structure of PLC-δ1, PDB file 1MAI) and, roughly speaking, is confined to a triangular area with L1, L3, and L6 as the three vertices.
Some positively charged or polar residues that have been previously identified as critical for lipid binding in this ligand-binding triangle are not conserved in APPL1 (Supplementary Figures 1 and 3), consistent with the fact that APPL1 alone lacks membrane binding ability.
Packing of the BAR and PH domains
In our crystal structure of the APPL1 BAR-PH dimer, the two PH domains are located at the opposite ends of the crescent-shaped dimer, and each has fairly extensive contact with the BAR domain of its symmetry mate (Figure 1).
The addition of the PH domain expands the BAR dimer in the longest dimension from 140 to 170 Å, but hardly changes the height of the dimer (i.e., the dimension along the two-fold axis direction) and its curvature.
Using its β1, β2, L3, and L7 regions, the PH domain contacts the BAR domain of its symmetry mate in two places (Figures 1B, C and 3; Supplementary Figure 3).
First, the motif of D15SPxxR20 (where x stands for any aa residue) at the N-terminal of BAR domain contacts β1, β2, and L3 regions of PH domain.
For instance, the hydroxyl group of Tyr283 in β1 forms a 2.6-Å hydrogen bond with the backbone carbonyl oxygen of Asp15 (Figure 3A).
Second, the conserved D334xxDRRYCF342 motif in the loop L7 of the PH domain is directly in contact with the loop connecting α2 and α3 in the BAR domain (Figure 3B).
The buried SAS from each BAR-PH molecule in the dimer is about 6600 Å2.
Thus, in the presence of PH domain, the buried SAS is 50% larger than that of the dimer formed by BAR domain alone (4400 Å2).
The canonical ‘ligand'-binding triangle of the PH domain is oriented about 60° from the concave side of the BAR dimmer, so that both the PH triangle and BAR concave surface could be brought within the vicinity of a curved membrane simultaneously.
Nevertheless, as discussed earlier, key residues for lipid interaction are not conserved in APPL1.
The C-terminus of PH is exposed to solvent in the dimer, consistent with the fact that it connects to the C-terminal region including the PTB domain.
Because of the interaction between D15SPxxR20 motif and PH domain, the rest N-terminus peptide (residues 5−12) clearly became ordered in the BAR-PH crystal structure, in comparison to the BAR domain-alone crystal structure, where residues N-terminal to Leu13 were invisible in the electron density map.
The fixed N-terminal peptide in the BAR-PH structure has an extended backbone conformation between Met4 (remnant from the His-tag cleavage) and Pro8, followed by a one-turn 310 helix (Figure 3C).
This region has several important intramolecular contacts mainly with helices α1 and α3.
For instance, the Leu7 side chain inserts between the aromatic rings of Phe26 in α1 and Phe182 in α3.
Meanwhile, the side chain of Asn186 forms two hydrogen bonds with the backbone amide and carbonyl groups of Leu7, respectively.
All these hydrophobic and hydrogen bond interactions appear conserved among BAR-PH containing proteins (Supplementary Figure 1).
Furthermore, the N-terminus is surrounded by a number of regions from the dimer partner, including the helix α4 and flexible loop connecting α1 and α2 (where residues 76−78 were mobile in the crystal structure).
For instance, Lys6' (where the prime stands for the dimer partner) forms a salt bridge with Asp243 in α4 between the protomers.
In addition, Ile9' forms hydrophobic interactions with Met247, Ile251, and Leu254 in α4 (Figure 3C).
To investigate roles of the BAR−PH interaction in solution, a double point mutation, S16E/P17E, at the BAR−PH interface was made.
These residues are located in the region N-terminal to the BAR domain and form close contacts with β2 and L3 of the PH domain (Figure 3A).
The recombinant protein of S16E/P17E double mutant in the context of BAR-PH was expressed predominantly in the insoluble fraction of cell lysate; however, the same mutations behaved normal in the BAR-only construct (data not shown).
Moreover, expression of the APPL1 PH domain alone in E.
coli did not produce soluble recombinant protein.
The data suggest that the dimer interaction between PH and BAR domains is critical for the solubility and stability of the APPL1 PH domain.
Consistent with this, our analytical ultracentrifugation (AUC) data showed that BAR-PH protein has a higher dimerization affinity in solution (kd=0.34 nM) than BAR domain alone (kd=0.13 μM; Supplementary data).
APPL1−Rab5 interaction in solution
To study APPL1−Rab5 interaction in solution, we performed glutathione S-transferase (GST)-mediated pull-down assays.
The APPL1 BAR-PH domain (residues 5−385) and a longer fragment with a 40-residue extension downstream of the PH domain, APPL1 (5−419), were each effectively pulled down by GTP-bound GST−Rab5 fusion protein (Figure 4).
The APPL1 protein was pulled down by either WT Rab5 preloaded with non-hydrolysable GTP analog (GppNHp) or Rab5-Q79L defective in GTP hydrolysis (with or without preloaded GTP analog), but could not be effectively pulled down by either the WT Rab5 preloaded with GDP or Rab5-S34N defective in GTP binding (Figure 4 and data not shown).
In contrast to the BAR-PH domain, we confirmed that APPL1 BAR domain alone (residues 5−265) cannot directly interact with Rab5 (data not shown) (Miaczynska et al, 2004).
Furthermore, using different Rab5 truncation variants, we demonstrated that the N-terminus (residues 1−15) and C-terminus (residues 185−215) of Rab5 are dispensable for interaction with APPL1 (Figure 4).
Binding affinity between full-length Rab5-Q79L and APPL1 (5−419) was quantitatively determined in a surface plasmon resonance (SPR) experiment.
Rab5 was coupled to the SPR biosensor chip in random orientations, and APPL1 (5−419) was applied as the analyte at concentrations of 0.15−12 μM (Supplementary Figure 4).
The dissociation constant, kd, for the Rab5−APPL1 interaction measured from this experiment was 0.9 (±0.7) μM, with kon and koff of 1.3 (±0.6) × 103 M−1 s−1, and 1.2 (±0.4) × 10−3 s−1, respectively.
This kd value is typical for an interaction between a Rab and its effectors (Eathiraj et al, 2005).
A major Rab5-binding site in the PH domain
To identify Rab5-binding site(s) in APPL1, GST−Rab5-Q79L (full length) was used to pull down APPL1 variants having surface point mutations.
The WT APPL1 (5−419) fragment was used as the parental construct for the mutagenesis, because this fragment is easily distinguishable from GST−Rab5 by size on SDS–PAGE gels without the need for Western blot analysis.
A total of 31 point mutations were made at 27 distinct, solvent-exposed positions (Figure 5; Supplementary Figures 1 and 3), based on the structural information of BAR-PH motif.
Most of these point mutations were located in the PH domain or near the BAR−PH interface, which are the surface regions most conserved between APPL1 and APPL2 (Supplementary Figure 5).
Substitution mutants were designed to maximize potential mutational effects on Rab5 binding (e.g., by flipping charges or switching between hydrophobic and hydrophilic residues) without disrupting the overall structure.
In addition, the flexible L1 loop (residues 289−294) was truncated and replaced with one Gly residue.
All of these APPL1 variants, as well as the WT construct, were expressed in E.
coli, with comparable yields from the soluble fractions (data not shown), in contrast to the mutations at BAR−PH interface mentioned earlier.
This suggested that the thirty or so surface mutations had little effect on the stability of BAR-PH dimer.
Among them, seven mutants, including V25D, N308D, M310K, A318D, G319R, L321D, and D324A, either abolished or significantly reduced (i.e., retaining <30%) Rab5 binding compared with the WT APPL1 (Figure 5A).
In the 3D structure, most of these residues cluster in an elongated surface area formed by β3, L3, and β4 of the PH domain, defining a major Rab5-binding site (Figure 5B; Supplementary Figure 3).
In addition, the effect of the V25D mutant suggests that the BAR domain also contributes to Rab binding either directly or indirectly.
On the other hand, L1 loop seems not to be required for Rab5 binding; significance of the apparent, positive effect of the truncation mutant (Figure 5A) remains to be studied.
We further extended these binding studies and confirmed the above Rab5-binding site in vivo in the cell, by monitoring Rab5-mediated APPL1 recruitment to early endosomes in the cell via confocal microscopy.
In this case, the RFP (DsRed-monomer)−Rab5-Q79L fusion protein was expressed in PC12 cells, targeted to the early endosomes, and recruited effectively the coexpressed GFP (green fluorescence protein)−APPL1 to these early endosomes (Figure 5C).
Importantly, APPL1 (5−385), that is, the BAR-PH domain, was sufficient to target to Rab5-Q79L containing early endosomes (Figure 5C).
In contrast, one of the Rab5-binding defective mutants (A318D) failed to target the early endosomes and exhibited a diffused pattern throughout the cytoplasm in the cell (Figure 5C).
Interaction between small GTPase and BAR domain has been exemplified in a complex crystal structure of Rac and arfaptin2 before (Tarricone et al, 2001).
Based on the following observations, however, we excluded the possibility of a Rac−arfaptin2-like binding mode for the Rab5−APPL1 interaction.
First, the linear dimension of Rab5 is less than 50 Å, which is significantly smaller than the distance (∼60 Å) between the putative Rab5-binding site in the PH domain and the central region of the BAR dimmer, where Rac binds with arfaptin2.
Second, the isolated APPL1 BAR domain did not bind to Rab5 in our pull-down assay.
Third, we mutated APPL1 Asn52, which is at the position equivalent to Rac-binding site in arfaptin2, to either a smaller (Ala) or larger (Arg) side-chain residue, and the mutations showed no effect on the binding to Rab5.
APPL1 as a Rab21 effector
Rab5 subfamily contains several members, including Rab5, Rab21, and Rab22.
Among them, Rab5 and Rab22 share a higher overall sequence identity with each other than with Rab21 (Pereira-Leal and Seabra, 2000).
This difference was used to explain the ability of Rab5 and Rab22, but not Rab21, to share some common effectors such as EEA1 and rabenosyn5 (Kauppi et al, 2002; Eathiraj et al, 2005).
Therefore, we tested APPL1 binding specificity towards other members in the Rab5 subfamily, using GST−Rab21 (full length) and GST−Rab22 (2−192) to pull down APPL1 (5−419).
Interestingly, APPL1 would bind to Rab21 in a GTP-dependent manner (Figure 4), indicating that APPL1 is an effector for both Rab5 and Rab21.
On the other hand, we were unable to detect any binding between APPL1 and Rab22 in the pull-down assay (Figure 4).
We could not rule out possible in vivo interaction between them because our recombinant Rab22 might not have folded correctly in E.
coli based on the following observations: (1) the expression level of Rab22 was 10- to 20-fold lower than Rab5 and Rab21, and (2) the GTP loading rate of Rab22 was lower too (data not shown).
Therefore, we focused our study on Rab5 and Rab21 for their interactions with APPL1.
We demonstrated that Rab21 and Rab5 have similar but not identical binding profiles towards APPL1 variants (Figure 5), which may be explained by their sequence divergence.
This differential binding to Rab5 and Rab21 by APPL1 may allow in vivo analysis of the functional roles of each Rab−APPL1 interaction, for example, by specifically abolishing one interaction while retaining others.
A novel Rab effector binding mode between Rab5 and APPL1
Next we investigated the APPL1-binding regions in Rab5.
Since residues responsible for APPL1 binding are likely to be located in the switch I, switch II, and interswitch regions, whose conformations change between different nucleotide binding states, these regions became the main objects of our investigation.
In addition to relevant Rab5 mutations that we made in previous studies, several point mutations in the Rab5 switch I region (i.e., residues 40−53) were tested for APPL1 binding.
We found that point mutations in the 42−48 region significantly reduced APPL1 binding, while L38R, Q49A, E50A, and I53N showed little or no detectable effect (Figure 6A).
Consistent with our previous structural studies on Rab5–rabaptin5 interaction (Zhu et al, 2004), all mutations within the 38−50 region did not interfere with Rab5−rabaptin5 binding (Figure 6A).
Furthermore, with the knowledge of crystal structures of Rab5–rabaptin5 and Rab22–rabenosyn5 complexes, it is clear that both Rab5 effectors rabaptin5 and rabenosyn5 bind to the so called invariant hydrophobic triad of Rab5 (i.e., Phe57, Trp74, and Tyr89) (Merithew et al, 2001).
Mutation of any of these residues usually strongly inhibits the Rab-effector binding (Zhu et al, 2004; Eathiraj et al, 2005).
Interestingly, in our mutagenesis analysis, the APPL1-binding was affected by W74R and Y89R, but not by F57R point mutation in Rab5 (Figure 6).
Taken together, our results indicate that APPL1 binds to Rab5 regions including the 40−48 loop and switch II, ∼30 Å across.
In addition, we showed that the two effectors, APPL1 and rapaptin5, could compete for Rab5-binding (data not shown), confirming that the binding sites of APPL1 and rabaptin5 on the Rab5 surface overlap with each other.
To further define the Rab5−APPL1 binding mode, we performed extensive pull-down analyses between variants of Rab5 and APPL1, looking for reversal mutants that could rescue the lost binding ability of others.
We identified one such pair; APPL1-N308D abolished the binding to Rab5, while Rab5-L38R had no effect on APPL1 binding.
However, Rab5-L38R was found to bind with APPL1-N308D, but not with the other tested APPL1 variants of similar hydrophobic-to-charged mutations, including V25D, A318D, and L321D (Supplementary Figure 6).
This result suggests that Rab5-L38R restores binding for APPL1-N308D through complementary, electrostatic, yet specific interactions.
It further implies that the position 308 in the β3 strand of APPL1 PH domain is in the vicinity of position 38 in the α1 helix of Rab5 in their complex.
Discussion
BAR-stabilized PH domain is essential for Rab5 binding
Since both APPL1 and APPL2 bind to Rab5, their Rab-binding sites are likely located in a surface region that is conserved between the two APPL proteins.
There are no deletion/insertion differences in the BAR-PH region between them (Supplementary Figure 1), and an inspection of the APPL1 BAR-PH dimer surface indicates that the most conserved surface region is located on the PH domain surface and the BAR-PH junction (Supplementary Figure 5).
Furthermore, neither PH domain (Miaczynska et al, 2004) nor BAR domain alone (data not shown) can directly bind Rab5, suggesting that the dimer interface between PH and BAR domains plays a critical role in Rab5 binding directly or indirectly.
This binding mode between Rab5 and APPL1 is apparently distinct from that between Rac and arfaptin, which only requires BAR dimerization (Tarricone et al, 2001).
To investigate further the structural basis of APPL1 and Rab5 interaction, we have performed extensive mutagenesis analyses.
A BAR dimer breaking mutant (F210D/F211D) and the BAR−PH interface mutant (S16E/P17E) are both insoluble when expressed in E.
coli (data not shown), supporting the notion that the functional form of APPL1 BAR-PH domain is a dimer.
Importantly a series of surface point mutants are soluble, allowing us to analyze the in vitro binding properties between these APPL1 mutants and Rab5 (Figure 5).
The results indicate that Rab5 specifically binds to the PH domain of APPL1 in the context of BAR-PH dimer, and this binding may marginally extend to the neighboring BAR domain.
Our structure-functional analyses are consistent with existing biochemical data.
For example, a previously reported triple mutation of APPL1 within the PH domain, K280E/Y283C/G319R, disrupts Rab5 binding (Miaczynska et al, 2004).
This effect can be fully explained based on the importance of the BAR−PH and PH−Rab5 interfaces.
Combined results from our mutagenesis pull-down experiments (Figures 4, 5 and 6), crystal structures of the BAR-PH domain of APPL1 (Figure 1), and structures of GTPase domain of human Rab5 in different nucleotide binding modes (Zhu et al, 2003, 2004) clearly explain the requirement of GTP-bound Rab5 for APPL1 binding.
Based on available information, we have modeled the interaction between the two proteins.
With the assumption that both proteins remain rigid bodies, our complex model satisfies constraints imposed by the mutagenesis pull-down results (Figure 7).
Over 1200 Å2 SAS area combined from both the APPL1 dimer and Rab5 would be buried in their interface.
In this putative Rab5−APPL1 binding mode, APPL1 interface includes L2, β3, L3, and β4 regions.
Note that the L3-β4 region showed weak electron density in the crystal structure, indicating its higher mobility and possible adaptability in forming a complex with Rab5.
On the Rab5 side, two regions that harbor binding-defective mutations are involved in the complex formation: the loop 42−48 and switch II (Figure 6).
Furthermore, the reversal mutation pair, Rab5-L38R and APPL1-N308D (Supplementary Figure 6) would directly interact with each other inside the interface of our complex model.
The bound Rab5 molecules would extend the concave surface of the APPL1 dimer, with both the N- and C-termini of Rab5 exposed to solvent.
Considering that there are about 30 residues C-terminal to our Rab5 model, which are necessary for membrane association but excluded from the crystallography study, our model would allow Rab5 molecules to anchor to the membrane through the added C-terminal tails and to interact with APPL1 at both ends of the BAR-PH dimer (Figure 7).
The Rab5 C-terminal tail is likely flexible, supporting that recruitment of APPL1 to the endocytic vesicle may not require its direct contact with the membrane.
In the complex, the Rab5 molecule does not block the C-terminus of the PH domain, allowing peptide extension of the APPL1 molecule from the BAR-PH domain.
Rab5–APPL1 interaction represents a novel Rab effector binding mode
In contrast to the α-helix dominant Rab-binding motifs of all other effectors of known 3D structures (Ostermeier and Brunger, 1999; Zhu et al, 2004; Eathiraj et al, 2005; Wu et al, 2005; Wei et al, 2006), the Rab5-binding motif of APPL1 is mainly composed of two β-strands, β3 and β4, and their connecting loop L3.
Although the exact binding position on the Rab protein and orientation of the effector helices may differ among available complex structures, all these Rab-binding domains interact with the invariant hydrophobic triad.
However, we have identified a Rab5 mutation in the hydrophobic triad, F57R, that does not interfere with APPL1 binding, but abolishes the binding to another Rab5 effector, rabaptin5 (Figure 6A; Zhu et al, 2004).
In contrast, several point mutations in the switch I region of Rab5 affect the binding of APPL1 but not rabaptin5.
The 42−48 region in Rab5 has not been previously reported to be involved in effector binding.
GTPase binding has emerged as a major function of PH domains in addition to lipid binding (Lemmon, 2004).
For example, PH domains in some guanine nucleotide-exchange factors (GEF) have been shown to bind directly to their cognate small GTPases (Rossman et al, 2002, 2003; Lu et al, 2004), and our data now show direct interaction between the APPL1 PH domain and Rab5.
So far, only two crystal structures of small GTPase−PH domain complexes are available.
One is Ran−RanBD1 (PDB file 1RRP).
The interactions between the Ran GTPase domain and RanBD1 PH core domain is fairly minor, occurring between the switch I region of the GTPase (equivalent to the 40's in Rab5) and strand β2 of the PH domain.
This interaction alone is unlikely to be sufficient to form a stable complex.
Indeed, Ran has a long C-terminal peptide beside the GTPase domain, while the PH domain of RanBD1 has an extra N-terminal peptide.
These two terminal peptides wrap around the partner proteins forming the major interaction between Ran and RanBD1.
Such an interaction seems not to be required for Rab5 and APPL1, because the GTPase domain of Rab5 and BAR-PH domain of APPL1 are sufficient to mediate their interaction.
The second published small GTPase−PH complex is that of Ral−Exo84 (PDB file 1ZC3).
In this complex, the PH domain of Exo84 uses L1, β5, and L6 to interact with the interswitch and switch II regions of Ral forming an intermolecular β-sheet extension mediated by the PH β5 strand and GTPase β2 strand (Jin et al, 2005).
Our mutagenesis analysis points to a different surface region (β3, L3, and β4) of the PH domain for Rab5 binding.
Therefore, the Rab5−APPL1 interaction represents a new GTPase−PH binding mode.
APPL1 is a representative of BAR-PH containing proteins
Both APPL1 and APPL2 are identified as Rab5 effectors, and their overall aa sequences are highly homologous.
In particular, residues on the APPL1 BAR dimer interface, BAR−PH interface, and the presence of the α4 helix seem well conserved in APPL2 (Supplementary Figure 5).
Therefore, APPL2 BAR-PH domain most likely forms a homodimer very similar to that of APPL1.
Furthermore, these conserved structural features may also extend to other BAR-PH containing proteins (Supplementary Figure 1; Habermann, 2004).
For instance, no helix breaking aa sequence appears in the middle of their predicted α4 regions.
Based on the APPL1 BAR-PH crystal structure, we find that, in general, the PH domain is more conserved than the BAR domain, and most of the highly conserved positions are located closer to the BAR−PH interface rather than the central region of the symmetric dimer.
For example, the two major contact regions between PH and BAR domains (i.e., D15SPxxR20 and D334xxDRRYCF342) are conserved at the aa sequence level among BAR-PH containing proteins.
In addition, correlated mutations are present between these proteins at the BAR−PH interface.
Thus, we propose that all BAR-PH containing proteins share similar 3D structures in the corresponding regions and that the BAR-PH motif may function as a general structural unit to interact with membrane-bound proteins and other molecular moieties.
In some BAR containing proteins, it is proposed that there exists an amphipathic helix N-terminal to the α1 helix of BAR domain, and they are called an N-BAR motif (Peter et al, 2004; Gallop et al, 2006).
It is suggested that this extra N-terminal region facilitates membrane binding (or bending).
A similar N-BAR structure was predicted for APPL1 and APPL2 (Habermann, 2004), but our current APPL1 crystal structure does not show such a structural motif.
Instead, the N-terminal region assumes an extended conformation and packs in the groove formed between helices α1 and α3 on the convex side of the crescent-shaped dimer (Figure 3C).
Since the N-terminal regions of the other BAR-PH containing proteins (Supplementary Figure 1) share similar sequences, we suggest that none of these proteins contains an N-BAR motif in their 3D structure.
Lacking both the amphipathic helix N-terminal to the BAR domain and the lipid-binding motif in the PH domain (Supplementary Figure 1) may explain the Rab5-dependent membrane association of APPL1.
In contrast, the PH domains of centaurin-β1/2 contain the key, basic residues for phosphoinositide binding (Dinitto and Lambright, 2006; Supplementary Figure 1).
If their PH domains are oriented similarly to that in the APPL1 dimer, the canonical (i.e., L1–L3–L6), ligand-binding triangle in their PH domains likely contributes to direct membrane association of these proteins.
Functional implications of the BAR-PH structure
While it is suggested that Rab5−APPL1 interaction mediates a signal transduction pathway between the plasma membrane and the nucleus, the mechanism by which Rab5 binding stimulates APPL1 translocation to the nucleus remains elusive.
The current BAR-PH structure may help to clarify the mechanism.
Interestingly, the sequence of ‘PKKKENE' was identified in the BAR domain of APPL2 as a potential nuclear localization signal (Miaczynska et al, 2004).
The corresponding region in APPL1 is the solvent exposed loop connecting α2 and α3 at the tip of the dimer (Figure 1B) and has a fairly conserved sequence (Supplementary Figure 1).
In addition, our preliminary data suggest that there is no detectable binding between the BAR-PH domain and the C-terminal region of APPL1 (data not shown), which makes it unlikely that Rab5 may regulate APPL1 through interference with the intramolecular interaction of the latter.
It seems more probable that the Rab5−APPL1 complex recruits downstream effectors to propagate the signal transduction process.
Unlike other Rab effectors, APPL1/2 proteins function in the signaling pathway from the so-called signaling endosome to nucleus.
Our data show that APPL1 interacts with the Rab5 protein using a novel binding mode; it remains to be proven whether such a binding mode is essential for APPL1 function.
Whereas it has been shown that APPL1 does not bind other Rab proteins miscellaneously (Miaczynska et al, 2004), we demonstrate that APPL1 is also an effector of Rab21, indicating that APPL1 adopts a binding mode shared by both Rab5 and Rab21.
It raises the possibility that, besides Rab5, other members of this Rab subfamily may also be involved in the APPL1 signaling pathway.
Materials and methods
Protein expression and purification
Constructs of human APPL1 (GenBank ID: NP_036228) (5−265) (i.e., the BAR domain) and APPL1 (5−385) (i.e., the BAR-PH domain) were inserted into the vectors pET28a and pET15b (Novagen), respectively, between NdeI and BamHI restriction sites.
The N-terminal few residues in the native sequence are hydrophobic and were deleted in an attempt to improve the solubility.
Point mutations were introduced into the pET15b-APPL1 (5−419) parental construct using QuickChangeTM site-directed mutagenesis kit (Stratagene).
His-tagged proteins of APPL1 (5−265) and APPL1 (5−385) were expressed as soluble recombinant proteins in BL21 StarTM (DE3) strain of E.
coli (Invitrogen), and cells were harvested after induction with 0.1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) for 8 h at 25°C.
The cells were lysed with lysozyme, and the lysate supernatant was purified with His-SelectTM affinity beads (Sigma).
In both cases, the His tag was removed with thrombin.
After further purification with Resource-QTM anion-exchange chromatography (GE Healthcare), both protein samples were concentrated to ∼30 mg ml−1 in (20 mM Tris–HCl (pH 8.0) and 0.1% (v/v) β-mercaptoethanol (βME)) and stored at −85°C until needed.
APPL1 (5−419) mutants were expressed similarly.
Se-Met-substituted proteins were expressed in E.
coli B834 (DE3) pLysS cells (Novagen) in minimal media supplemented with 40 mg l−1 Se-Met (Sigma) and purified using the same procedure as the native protein.
Recombinant proteins of human Rab5a variants (GenBank ID: NM_004162), human Rab21 (BC021901), and human Rab22a (NM_020673) fused with an N-terminal GST were expressed in BL21 E.
coli and purified with GST-affinity chromatography.
The sample was concentrated to ∼20 mg ml−1 and stored in 1 × phosphate-buffered saline (PBS) with 0.1% (v/v) βME at −80°C.
Recombinant protein of human rabaptin5 (551–862) (GenBank ID: CAA62580) was expressed and purified as described previously (Zhai et al, 2003); two additional point mutations, C719S and C734S, were introduced to reduce aggregation.
Protein crystallization and data collection
Crystals of APPL1 (5–265) were grown at 20°C with the hanging drop vapor diffusion method.
The Se-Met incorporated protein sample diluted to 10–20 mg ml−1 was mixed 1:1 (v/v) with the reservoir solution of 0.1 M magnesium formate and 0.1% (v/v) βME.
Crystals were transferred to a cryo-protectant solution of (88% saturated Li2SO4, 14 mM magnesium formate, 20 mM Tris–HCl (pH 8.0), and 0.1% (v/v) βME) by gradually changing the drop solution in 20 min, followed by cooling in liquid nitrogen.
A data set was collected at selenium edge at sector 22 BM of the Argonne APS facility.
Crystals of APPL1 (5–385) were also grown in hanging drops at 20°C.
The reservoir contained 6% (w/v) polyethylene glycol 6000 (PEG6K), 0.6 M NaCl, and 0.1% (v/v) βME.
The crystals were quickly soaked in a solution of (7.5% (w/v) PEG6K, 0.25 M NaCl, and 10% (v/v) glycerol) and flash-cooled under liquid nitrogen.
Both native and SAD data sets were collected at the APS facility.
Analyzing the crystal content by SDS–PAGE confirmed the integrity of the protein sample.
All data were processed with the program suite HKL2000.
Pull-down assay for analyzing protein–protein interactions
In the Rab5−APPL1 pull-down experiment, 30 μg GST−Rab fusion protein (52 kDa) was incubated with 60 μl of 30% slurry of GSH–Sepharose 4B (GE Healthcare) at 22°C for 30 min.
Nucleotide loading reaction was performed on the GSH beads in an exchange buffer of (1 × PBS, 2 mM DTT, 1 mM MgCl2, 4 mM EDTA, and 400 μM GppNHp or GDP) at 22°C for 30 min.
Increasing the magnesium ion concentration to 20 mM terminated the loading reaction.
Soluble fractions of cell lysate containing all His-tagged APPL1 (5–419) variants were analyzed by SDS–PAGE to confirm their comparable expression level and solubility.
The GSH resin carrying nucleotide-loaded GST−Rab fusion protein was incubated with ∼50 μl cell lysate (∼200 μg APPL1 variant, 50 kDa) at 22°C for 30 min, then washed three times with 200 μl of (1 × PBS, 2 mM DTT, and 4 mM MgCl2) and resuspended in 20 μl of 2 × reducing SDS sample buffer.
The sample was subjected to SDS–PAGE analysis, visualized with Coomassie blue stain.
The same samples were analyzed with chemiluminescence Western blot (GE Healthcare) and His-tag antibody then detected on films which were semi-quantified using the computer software ImageJ (http://rsb.info.nih.gov/ij/) including its default calibration.
The relative band intensity of each mutant versus WT from multiple experiments is shown in Figure 5.
Coordinate deposit
Coordinates and the structural factors of the APPL1 crystal structures have been deposited to PDB under codes 2Q12 (BAR domain structure) and 2Q13 (BAR-PH domain structure).
Supplementary Material
Sorting nexin-1 mediates tubular endosome-to-TGN transport through coincidence sensing of high- curvature membranes and 3-phosphoinositides
Membrane and juxtamembrane targeting by PH and PTB domains
Structural basis of family-wide Rab GTPase recognition by rabenosyn-5
Generation of high curvature membranes mediated by direct endophilin bilayer interactions
Mechanism of endophilin N-BAR domain-mediated membrane curvature
The BAR-domain family of proteins: a case of bending and binding?
Exo84 and Sec5 are competitive regulatory Sec6/8 effectors to the RalA GTPase
The small GTPase Rab22 interacts with EEA1 and controls endosomal membrane trafficking
Pleckstrin homology domains: not just for phosphoinositides
Rab5 GTPase and endocytosis
APPL1 associates with TrkA and GIPC1 and is required for nerve growth factor-mediated signal transduction
Mediation of the DCC apoptotic signal by DIP13 alpha
PH domain of ELMO functions in trans to regulate Rac activation via Dock180
APPL1 binds to adiponectin receptors and mediates adiponectin signalling and function
Structural plasticity of an invariant hydrophobic triad in the switch regions of Rab GTPases is a determinant of effector recognition
APPL proteins link Rab5 to nuclear signal transduction via an endosomal compartment
Identification of a chromosome 3p14.3–21.1 gene, APPL, encoding an adaptor molecule that interacts with the oncoprotein-serine/threonine kinase AKT2
APPL1, APPL2, Akt2 and FOXO1a interact with FSHR in a potential signaling complex
Structural basis of Rab effector specificity: crystal structure of the small G protein Rab3A complexed with the effector domain of rabphilin-3A
The mammalian Rab family of small GTPases: definition of family and subfamily sequence motifs suggests a mechanism for functional specificity in the Ras superfamily
BAR domains as sensors of membrane curvature: the amphiphysin BAR structure
Pleckstrin homology domains: a common fold with diverse functions
Multifunctional roles for the PH domain of Dbs in regulating Rho GTPase activation
A crystallographic view of interactions between Dbs and Cdc42: PH domain-assisted guanine nucleotide exchange
BIN1 is a novel MYC-interacting protein with features of a tumour suppressor
The structural basis of arfaptin-mediated cross-talk between Rac and Arf signalling pathways
GIPC is recruited by APPL to peripheral TrkA endosomes and regulates TrkA trafficking and signaling
Molecular dissection of Rab11 binding from coiled-coil formation in the Rab11-FIP2 C-terminal domain
Structural basis for recruitment of RILP by small GTPase Rab7
The interaction of the human GGA1 GAT domain with Rabaptin-5 is mediated by residues on its three-helix bundle
High resolution crystal structures of human Rab5a and five mutants with substitutions in the catalytically important phosphate-binding loop
Structural basis of Rab5–Rabaptin5 interaction in endocytosis
Crystal structure of the APPL1 BAR-PH dimer.
(A) The overall structure of the dimer.
The left view is along the direction of the dyad symmetry and on the concave surface, and the right view differs by 90°.
One protomer is shown in ribbon diagram, and the other is shown in molecular surface model.
Helix α1 is colored yellow, α2 magenta, α3 green, α4 blue, and PH domain red.
(B, C) A close view of the BAR–PH interface.
These images were generated using the program PyMol.
Electrostatic potential distribution of APPL1.
Electrostatic potentials of APPL1 BAR-PH protomer (left) and dimer (middle and right) are mapped on their molecular surfaces.
Negatively charged regions (⩽−0.5 V) are colored red, positively charged regions (⩾+0.5 V) blue, and neutral regions gray.
The right view is looking down along the dyad axis of the dimer at the concave surface, and the side view differs by 90°.
Clusters of acidic residues which are potentially important in Rab5 binding are circled.
This figure was generated with the program CCP4mg.
APPL1 domain packing.
(A, B), Stereoviews of two major interacting regions between the PH and BAR domains.
The PH domain from one APPL1 protomer is colored gray, and the BAR domain from the dimer-mate is colored yellow.
Helix backbones are shown in ribbon representation, otherwise in ropes.
Side-chain and/or main-chain atoms of selected residues are shown in stick models.
Nitrogen atoms are colored blue, oxygen red, and sulfur green.
Hydrogen bonds (<3.0 Å) are shown as dash-lines.
Note that most residues displayed are highly conserved among the known BAR-PH containing proteins (Supplementary Figure 1).
(C) Stereoview of interactions of the APPL1 N-terminal region.
The two protomers are colored gray and yellow, respectively.
This figure was generated with the programs Molscript and Raster3D.
Pull-down analysis of APPL1-Rab interaction.
GST fusion proteins of Rab5, Rab21, and Rab22 were used to pull down His-tagged APPL1 (5−385) and (5−419) fragments in the presence of GDP or GTP analog (GppNHp).
The results were analyzed with SDS–PAGE and anti-His Western blot.
Mutational analyses of APPL1.
(A) Quantifying the Rab binding ability of APPL1 variants.
The Rab5- and Rab21 binding abilities of APPL1 variants relative to that of WT APPL1 (5−419) were estimated based on chemiluminescence-labeled Western blot (see the Materials and methods) and shown as the white and gray bars, respectively.
Standard deviations calculated from multiple experiments were represented by the thin lines.
(B) Mutational effects on binding to Rab5 and Rab21.
Distribution of point mutations is shown on the molecular surface of the BAR-PH dimer where the two protomers are colored gray and light green, respectively.
APPL1 mutations that affect binding strongly (i.e., <30% binding comparing to WT) are colored red; otherwise, the mutants are colored blue.
The position showing reversed binding property is colored yellow.
Mutation positions are selectively labeled; note that the PH domain contains residues 276−379.
(C) Mutational effects on APPL1 targeting to Rab5-positive early endosomes in the cell.
RFP−Rab5-Q79L was coexpressed with GFP−APPL1 (full length, FL; BAR-PH domain; or BAR-PH mutant) in PC12 cells as indicated, followed by confocal fluorescence microscopy.
Shown are typical confocal microscopic images indicating the RFP−Rab5-Q79L labeled early endosomes (red) and the colocalization of GFP−APPL1 or mutants (green) in the same cells.
Scale bar, 16 μm.
Mutational analyses of Rab5.
(A) Pull-down assay on Rab5 variants.
Point mutations in switch regions were introduced in the full-length Rab5-Q79L background of the GST fusion construct.
Each mutant was expressed in E.
coli, purified, and equal amounts of each Rab5 sample was used to pull down recombinant proteins of His-tagged WT APPL1 (5−419) (top panel) and His-tagged rabaptin5 (551−862) (bottom panel).
The results were visualized by Coomassie blue stain.
(B) Molecular surface model of Rab5 GTPase domain.
Its N-terminus, switch I (SW1) and switch II (SW2) regions are labeled.
Rab5 mutations that have effects and have no effect on binding are labeled red and blue, respectively.
The position showing reversed binding property with the APPL1 N308D mutant is colored yellow.
Putative complex model of the APPL1 BAR-PH dimer and Rab5.
(A) APPL1–Rab5 interaction.
The two BAR-PH protomers are shown in molecular surface models and colored gray and cyan, respectively.
Positions of APPL1 mutations are colored similarly to Figure 5B.
Overlaying APPL1, Rab5 is shown in a green backbone trace and positioned according to the mutagenesis data; the N-and C-termini of its GTPase domain are labeled.
Rab5 mutants that affect binding are marked with red spheres, and those having no effect are marked with blue spheres.
Positions of the reversal mutation pair are labeled.
(B) Membrane recruitment of APPL1 mediated by Rab5.
The APPL1 BAR-PH protomers are colored gray and yellow, and APPL1 C-terminal peptides are represented by ovals.
Rab5 molecules are shown in green molecular surface models, and their membrane anchored C-terminal tails are represented by red curves.
Crystallography data collection and refinement statistics
   (a) Data statistics BAR BAR−PH   Wavelength (Å) 0.9793 1.0000   Space group P21212 P21212   Unit cell        a (Å) 53.0 103.7    b 129.2 105.7    c 36.9 36.4   Resolution (Å) 50 (1.86)a−1.80 50 (2.12)−2.05   Rmerge (%) 8.1 (44.0) 6.7 (43.2)   Number of reflections 23 548 (1897) 23 286 (1989)   Completeness (%) 96.1 (78.5) 89.9 (78.0)   Redundancy 3.3 (3.1) 4.5 (3.2)   I/σ(I) 10.2 (2.2) 16.4 (2.7)         (b) Refinement statistics       Rworking (%)/# of reflectionsb 21.2/21 703 20.5/20 882   Rfree (%)/# of reflectionsb 25.5/1200 26.8/1200   Number of non-hydrogen atoms        Protein 1990 2943    Solvent 147 157   R.m.s.d.
from ideal values        Bond length (Å)/angle (deg) 0.016/1.51 0.013/1.30   Ramachandran plot (%)c 98.3/1.7/0/0 94.7/5.3/0/0   Average B-factor (Å2)        Protein 34.1 (25.8)d 47.9 (38.3)d    Solvent 41.4 47.7  aNumbers in parentheses are the corresponding numbers for the highest-resolution shell, unless otherwise mentioned.
     bReflections of ∣Fobs∣>0.0.
     cCalculated using PROCHECK.
Numbers reflect the percentage of residues in the core, allowed, generously allowed and disallowed regions, respectively.
     dWilson B-factors calculated using a 4 Å cutoff.
   .Structural Basis of PP2A Inhibition by Small t Antigen
The SV40 small t antigen (ST) is a potent oncoprotein that perturbs the function of protein phosphatase 2A (PP2A).
ST directly interacts with the PP2A scaffolding A subunit and alters PP2A activity by displacing regulatory B subunits from the A subunit.
We have determined the crystal structure of full-length ST in complex with PP2A A subunit at 3.1 Å resolution.
ST consists of an N-terminal J domain and a C-terminal unique domain that contains two zinc-binding motifs.
Both the J domain and second zinc-binding motif interact with the intra-HEAT-repeat loops of HEAT repeats 3–7 of the A subunit, which overlaps with the binding site of the PP2A B56 subunit.
Intriguingly, the first zinc-binding motif is in a position that may allow it to directly interact with and inhibit the phosphatase activity of the PP2A catalytic C subunit.
These observations provide a structural basis for understanding the oncogenic functions of ST.
Author Summary

The study of how DNA tumor viruses induce malignant transformation has led to the identification of key pathways that also play a role in spontaneously arising cancers.
One such virus, simian virus 40 (SV40), produces two proteins, the large T and small t antigens, that bind and inactivate tumor suppressor genes important for cell transformation.
Specifically, SV40 small t antigen (ST) binds to and perturbs the function of the abundant protein phosphatase 2A (PP2A).
PP2A is a family of heterotrimeric enzymes, composed of a structural A subunit, a catalytic C subunit, and one of several regulatory B subunits.
Here we have determined the structure of SV40 ST in complex with the PP2A structural subunit Aα.
SV40 ST consists of an N-terminal J domain and a C-terminal unique domain that contains two separate zinc-binding motifs.
SV40 ST binds to the same region of PP2A as the regulatory subunit B56, which provides a structural explanation for the displacement of regulatory B subunits by SV40 ST.
Taken together, these observations provide a structural basis for understanding the oncogenic functions of ST.
The crystal structure of full-length SV40 small t antigen (ST) in complex with the A subunit of its target, protein phosphatase 2A, contributes to our understanding of the oncogenic functions of ST.
Introduction
Simian virus 40 (SV40) is a DNA tumor virus in the polyomavirus family.
SV40 may play a role in a subset of human cancers, and the study of transformation induced by SV40 has led to many insights into the pathways involved in spontaneously arising cancers [1].
The Early Region of SV40 is essential for transformation and encodes two oncoproteins, the large T antigen (LT) and small t antigen (ST), through alternative splicing.
LT binds to a number of host proteins including the retinoblastoma and p53 tumor suppressors.
ST, which shares its N terminus with LT but has a unique C-terminal end, is also a potent oncoprotein that plays a critical role in the transformation of several human cell types [2,3].
For example, the cointroduction of LT, ST, the telomerase catalytic subunit hTERT (human telomerase reverse transcriptase), and an oncogenic allele of H-RAS imparts a tumorigenic phenotype to a wide range of primary human cells [4–6].
The tumorigenic activity of ST is strictly dependent on its interaction with protein phosphatase 2A (PP2A), since mutant versions of ST that are unable to bind PP2A fail to induce tumorigenic activity [5].
PP2A is a large family of heterotrimeric enzymes that accounts for the majority of total Ser/Thr phosphatase activity in most tissues and cells [7–10].
Although a dimer comprised of a ~65 kDa scaffolding A subunit and a ~36 kDa catalytic C subunit constitutes the core enzymatic activity of PP2A, the binding of a third regulatory B subunit to the AC core enzyme regulates PP2A activity, cellular localization, and substrate specificity.
PP2A can exist in cells as either the AC core complex or an ABC heterotrimeric holoenzyme that is the dominant form in most cells.
The scaffold A subunit is composed of 15 HEAT (huntingtin, elongation factor 3, A subunit of protein phosphatase 2A, and target of rapamycin) repeats [11], while the C subunit contains a catalytic domain that shares sequence homology with other Ser/Thr phosphatases such as protein phosphatase 1 and protein phosphatase 2B (calcineurin).
In humans there are at least 18 B subunits that can be classified into B (B55), B′ (B56), B′′, and B′′′ families based on sequence similarity.
PP2A regulates a wide array of cellular processes.
In particular, several lines of evidence implicate PP2A as a tumor suppressor gene.
Specifically, PP2A inhibitors such as microcystin and okadaic acid are potent tumor promoters and the ST antigen acts as a potent oncogene [12–16].
Consistent with its role as a tumor suppressor, low frequency mutations of the Aα and Aβ subunits are found in breast, lung, and colorectal cancers [12,17–20].
Mutations in PP2A Aα result in functional haploinsufficiency that depletes complexes containing B56γ, thereby leading to cell transformation [16,21].
In contrast, mutations in Aβ disrupt the interaction of Aβ with the small GTPase RalA, leading to constitutive RalA phosphorylation and activity [22].
ST interacts with the PP2A AC core dimer through the direct interaction between ST and the PP2A scaffolding A subunit [23,24].
Prior work has suggested that ST associates with the PP2A C subunit through interactions with the A subunit.
While the C subunit interacts with HEAT repeats 11–15 of A subunit, ST interacts with HEAT repeats 3–7, which is also the binding site for PP2A B subunits [25,26].
The ability of ST to displace multiple B subunits from the A subunit has been demonstrated both in vitro [27,28] and in vivo [16,29].
The displacement of B subunits by ST inhibits PP2A activity towards multiple substrates, but increases phosphatase activity towards histone H1 [22,30].
Therefore, ST can be considered as a viral B subunit that changes the biochemical properties of PP2A.
The 174 amino acid ST shares its N-terminal 78 residues with the N-terminal sequence of LT, including a region that exhibits J domain (also termed the DnaJ domain) function [3,31,32].
The unique C-terminal 96 residues of ST antigen interact with the PP2A A subunit.
The C-terminal unique domain is rich in Cys residues, six of which are organized into two CxCxxC clusters and bind two zinc ions, providing structural stability to ST [33,34].
The N-terminal J domain is not required for the interaction of ST with the A subunit of PP2A but may contribute to the high affinity binding, since the removal of the first 51 residues from ST caused a 140-fold reduction in inhibition towards the phosphatase activity of PP2A AC core complex [35].
The N-terminal J domain shares sequence homology with the DnaJ family of molecular co-chaperones, which promote the ATPase and chaperone activities of heat shock protein 70 (Hsp70), an important chaperone in the cell [36,37].
Hsp70 binding region has been mapped to the surface formed by J domain helices 2–3 [38,39].
The DnaJ-like domain in ST and the DnaJ domain in E.
coli are interchangable without a loss in co-chaperone activity [40].
While the structure of the J domain can be predicted from prior structural determinations of LT, the overall structure of ST remains to be unraveled and it is unclear how ST may interact with PP2A and regulate PP2A activities.
We have determined the crystal structure of full-length SV40 ST in complex with the full-length Aα subunit of PP2A.
This structure reveals two novel zinc-binding motifs formed by the unique C-terminal domain, the structural linkage of the J and unique domain of ST, and the interaction site of ST with the structural A subunit.
Together with our biochemical data, we provide a structural basis for understanding the tumorigenic activity of ST protein.
Results
Overall Structure
The protein complex containing full-length SV40 ST and full length murine PP2A Aα subunit (A-ST complex) were co-expressed in E.
coli and purified to homogeneity.
Crystal structure of the complex was determined by a combination of molecular replacement, using the PP2A A subunit structure as the searching model, and single-wavelength anomalous dispersion of intrinsic zinc atoms in ST, and was refined at 3.1 Å resolution (Table 1).
Four complexes were found in each asymmetric unit.
In each complex, the scaffolding A subunit contains 15 HEAT repeats that forms a horseshoe shape.
The four A-ST complexes in the asymmetric unit have essentially the same structure, except HEAT repeats 11–15 that show substantial conformational variation (see below).
ST contains an N-terminal J domain and a C-terminal unique domain.
These two domains sit on the concave and convex sides of the ridge of the A subunit horseshoe structure, respectively, by interacting with intra-repeat loops of the Aα subunit HEAT repeats 3–7 (Figure 1), which is also the binding site for B56γ1 in the A-B56γ1-C trimeric PP2A holoenzyme structure [25,26].
Summary of Crystallographic Analysis of the PP2A Aα-SV40 Small t Antigen Complex
Overall Structure of SV40 ST in Complex with the A Subunit of PP2A
A cartoon illustration of the “front” and “top” views of the PP2A A subunit–SV40 ST complex.
The scaffold Aα subunit of PP2A, the J domain, and the unique domain of SV40 ST are colored blue, green, and magenta, respectively.
In addition, two zinc ions in the ST unique domain are yellow.
ST interacts with the intrarepeat loops of HEAT repeats 3–7 of PP2A A subunit.
Structure of SV40 ST
SV40 ST exhibits an all α-helix structure with two zinc-binding sites in the unique domain (Figure 2A).
The J domain contains three helices and has a structure similar to the previously solved crystal structure of the J domain of SV40 LT and a NMR structure of polyomavirus DnaJ-like domain [41,42], with Cα root mean square deviations (RMSDs) of 1.84 and 1.83 Å, respectively.
The DnaJ domain in the E.
coli DnaJ protein forms a complex with Hsp70 in the flexible L3 region [38,39].
Interestingly, the expected Hsp70 binding region of the J domain is not involved in the interaction with PP2A A subunit and appears widely accessible to Hsp70.
In contrast, the Hsp70 binding region of J domain in LT is involved in the binding of retinoblastoma protein [41].
The linker between helices 2 and 3, which has been implicated in Hsp70 interaction, as well as the linker between J and unique domains (residues 72–86), is not visible in any of the four complexes in the asymmetric unit, suggesting structural flexibility of these two linkers in the A-ST complex.
Structure of SV40 Small t Antigen
(A) Structural organization of SV40 ST and the zinc coordination of ST unique domain.
The peptide chain is color-coded from blue to red, going through the rainbow colors, from the N terminus (blue) to the C terminus (red).
(B) Interface of the J and unique domains.
The J and unique domains are in green and pink, respectively.
Key residues in the interface between the J domain and the unique domain—Trp24, Gly25, Pro28, His70, Trp135, Tyr139, Trp147, and Ile163—are labeled.
(C) Sequence alignment of ST proteins from SV40 (strain VA45-54-2), baboon polyomavirus (PyV), BK polyomavirus (BKV), and JC polyomavirus (JCV).
Sequence conservation is indicated below the aligned sequences.
Secondary structures in the determined crystal structure are indicated above the aligned sequences.
Solid and empty stars indicate residues interacting with PP2A A subunit using side-chain and main-chain, respectively.
Solid and empty squares represent residues involved in interactions between the J domain and the unique domain with side-chain and main-chain atoms, respectively.
The ST unique domain is composed of four helices.
Three of these helices are directly involved in the interaction with two zinc ions.
The first zinc-binding site is positioned between two roughly parallel helices (α4 and α5) and the L5 loop, coordinated by four conserved cysteine residues (Cys103, Cys111, Cys113, and Cys116).
Cys103 stems from the C-terminal end of α4, while Cys113 and Cys116 are positioned in the N-terminal end of α5 (Figure 2A).
The second zinc is located between two roughly perpendicular helices (α5 and α6) and coordinated by three conserved cysteines in the second cysteine cluster (Cys138, Cys140, and Cys143), with Cys140 and Cys143 from the N-terminal end of α6 and the conserved His122 from the middle of α5 (Figure 2A).
Although these two zinc ions are located in discrete positions, both of them coordinate with α5.
These two zinc-binding motifs are completely distinct from the previously proposed GAL4-type zinc cluster for ST and are also different from classical zinc finger structures [33,34,43].
The J and unique domains have an interface that is mostly hydrophobic (Figure 2B).
These two domains of SV40 ST interact with each other by several hydrogen-bonding and hydrophobic interactions.
The indole rings of Trp135 and Trp147 make hydrogen bonds with the backbone carbonyl group of Gly25 and Trp24, respectively.
The hydroxyl group of Tyr139 forms a hydrogen bond with the imidazole ring of His70.
Pro28 and Ile163 make a hydrophobic interaction between two domains (Figure 2B).
It should be noted that both the N and C termini of ST are located at the interface of the J and unique domains .
In particular, the N terminus is located in the joint point of J domain and the A subunit, and directly interacts with the A subunit.
The interface and relative orientation of these two domains are essentially identical in all four A-ST complexes in the asymmetric unit.
The Cα RMSDs of ST among these four complexes are between 0.95 Å and 1.03 Å.
Therefore, the J and unique domains of ST are structurally coupled and ST consists of one globular fold.
Since residues in the A-ST interface are highly conserved among ST proteins (Figure 2C), this domain organization should be conserved among ST proteins in the polyomavirus family.
The Interface between ST and PP2A A Subunit
The interactions between the Aα subunit and SV40 ST are formed through the intra-loop regions of repeats 3 to 7 in the Aα subunit (Figures 1 and 3).
Detailed interactions are summarized in Figure 3.
ST uses both its J domain and the second zinc-binding motif (in particular the L6 loop) for interacting with PP2A Aα subunit, while the first zinc finger is not directly involved in interactions with PP2A Aα subunit, but may be involved in interactions with PP2A catalytic C subunit (see below).
One notable feature of this interaction is that five of eight residues in the ST unique domain (Figures 2C and 3) in the A-ST interface interact with the Aα subunit through the backbone of those residues.
The backbone conformations of those ST residues (Leu133, Met146, Trp147, Phe148, and Gly149) are largely determined by the second zinc-binding site, demonstrating the importance of the second zinc-binding site in the Aα subunit interaction.
In contrast, all of the Aα subunit residues interact with ST through their side chains, and some of those residues were identified by previous A subunit mutagenesis studies [44].
The Interface between SV40 ST and PP2A A Subunit
Color assignment for different subunit or domains is the same as in Figure 1.
Hydrogen bonds between ST and the PP2A A subunit are indicated by dashed lines.
Structural Comparison of A-ST Complex with A-B56-C PP2A Holoenzyme
When the A-ST complex structure is superimposed with the previously reported A-B56-C PP2A holoenzyme structure [25,26], it is obvious that ST and the B56 subunit interact with the same region of PP2A A subunit and have very similar “footprints” on the scaffold A subunit (Figure 4A).
The majority of Aα residues involved in the A-ST interaction—including Glu100, Trp140, Phe141, Asp177, Pro179, and Met180—are also involved in the A-B56 interaction (Figure 4B).
The side chain conformations of those Aα residues are quite similar in both structures, except Trp140.
The position of the indole ring of Trp140 is flipped between these two complex structures, mostly due to the intercalation of Pro132 of ST between Trp140 and Phe141 by forming a hydrophobic core (Figures 3 and 4B).
The largely overlapping Aα binding sites of ST and B56 explains how ST competes with B56 for the binding of the Aα subunit [25,26].
Structural Comparison of the A-ST Complex and the A-B56-C PP2A Holoenzyme
(A) Structural superposition of these two complexes.
These two complexes are superimposed using A subunit HEAT repeats 2–10.
The J and unique domain of ST are colored green and pink, respectively.
The Cα trace of B56γ1 are shown in light orange.
It is clear that ST and B56γ1 bind to the same sites on PP2A A subunit.
(B) The PP2A A subunit residues involved in both ST and B56γ1 interactions.
ST and B56γ1 share a common footprint on the ridge of A subunit.
In the A-B56-C PP2A holoenzyme structure, both B56 and C subunits sit on the same side (intra-repeat loop side) of the horseshoe shape formed by the scaffolding A subunit [25,26].
When HEAT repeats 2–10 in the A-ST complex are superimposed with corresponding regions of the A-B56-C complex, the first zinc-binding motif, in particular helix α4 and its flanking region, is in close proximity to the C subunit active site (Figure 5).
The closest atoms between ST and C subunit in this superposition are within van der Waals distances.
It should be noted that, when the A-ST complex structure is superimposed with the PP2A AC core complex, the distance between the ST first zinc-binding motif and C subunit appears further apart than in the A-B56-C holoenzyme.
However, since the PP2A A subunit has substantial structural flexibility (see below), it is plausible that ST may interact directly with the C subunit near its catalytic active site, just like the B56 subunit that interacts with both A and C subunits.
The First Zinc-Binding Motif May Directly Interact with and Inhibit the Catalytic C Subunit of PP2A
The structures of the A-ST and A-B56-C complexes are superimposed.
The PP2A catalytic subunit is shown in the surface model.
The active site of the PP2A C subunit is indicated.
The potential binding site of Hsp70 is represented by a gray sphere.
Structural Comparison of the A Subunit Conformations in Different Complexes
The A-ST interface and the structure of HEAT repeats 2–10 of the A subunit are quite rigid, since these structures can be well superimposed in all four A-ST complexes in the asymmetric unit.
In contrast, HEAT repeats 11–15 apparently have highly variable structures, and the first HEAT repeat is not well folded in some cases (Figure 6A and 6B).
While the RMSDs of all 15 HEAT repeats among four A-ST complexes in the asymmetric unit range from 1.29 Å to 3.02 Å, the RMSD of HEAT repeats 2–10 are on average 1.09 Å.
The conformational variation of the Aα subunit results mostly from conformational flexibility in HEAT repeats 10–13, since HEAT repeats 13–15 among the four complexes are easily superimposed, with an average RMSD of 1.15 Å (Figure 6C).
Structural comparison of the A-ST crystal structure with previously reported crystal structures of A, AC complex, and A-B56-C complexes [25,26,45,46] also support the conclusion that HEAT repeats 2–10 and HEAT repeats 13–15 form two relatively rigid blocks.
However, there is substantial structural flexibility between these two structural blocks, due to the result of accumulative conformational changes in HEAT repeats 10–13 (Figures S1 and S2).
Structural Comparison of the PP2A A Subunit and the Structural Flexibility
(A) The structural alignment of PP2A A subunits.
PP2A A subunit structures were aligned based on their structures of HEAT repeats 2–10.
PP2A A subunit structures that used for the alignment are from the four A-ST complexes in the asymmetric unit, the A subunit structure alone (PDB code: 1B3U), the AC dimer structure (PDB code: 2IE3), and the A-B56-C trimeric structure (PDB code: 2IAE).
The HEAT repeats 2–10 may form a rigid structural block since no significant structural variations were observed for this region among all A subunit structures.
(B) The amplitude of conformational variations of the PP2A A subunit HEAT repeats 11–15.
(C) HEAT repeats 13–15 may form the other relatively rigid structural block in the PP2A A subunit.
There is no major conformational variation in HEAT repeats 13–15 among all A subunit structures.
Therefore, the structural variations observed in (A) and (B) are mostly due to the conformational flexibility of HEAT repeats 10–13.
(D) Superposition of PP2A A subunit structures from Aα alone, A-ST complex, and A-B56-C trimeric complex.
The structure of the PP2A A subunit in the A-ST complex can have a conformation very similar with that of the A subunit in the A-B56-C trimeric complex.
Mutagenesis of the A-ST Interface
To determine whether the amino acid residues observed in our structure to be interaction points between ST and Aα, we generated a set of ST (R7A, R21A, P132A, and W147A) and Aα (D177A, R183A, E216A, Q217A, and R258A) mutants and performed in vitro binding assays.
We found that substitution of alanine for residues Arg7, Arg21, or Pro132 of ST abrogated interaction between ST and wild-type Aα (Figure 7).
In addition, we found that the W147A ST mutant showed reduced binding to Aα compared to wild-type ST.
Analysis of Aα mutants indicated that single alanine substitutions at position Glu216 disrupted PP2A A-ST interaction (Figure 7).
These observations provide strong support to the A-ST interface observed in our crystal structure.
Mutational Analysis of the A-ST Interface
(A) Mutation of ST residues predicted to interact with PP2A Aα.
Top: Expression of ST in whole cell lysates (WCL).
Middle: Isolation of PP2A Aα complexes and immunoblotting with anti-PP2A Aα antibodies.
Bottom: Isolation of PP2A Aα complexes and immunoblotting for ST.
(B) Mutation of PP2A Aα residues predicted to interact with ST.
Top: Isolation of PP2A Aα complexes and immunoblotting with anti-PP2A Aα.
Bottom: Isolation of PP2A Aα complexes and immunoblotting for ST.
Discussion
Structure of Small t Antigen and Its Interaction with PP2A A Subunit
Here we report the crystal structure of SV40 ST in complex with the murine PP2A A subunit.
It is striking that all four A-ST complexes in the asymmetric unit of our crystal lattice have essentially the same structure, except the flexible HEAT repeats 11–15 of the A subunit that are not involved in the A-ST interaction.
This observation argues strongly that the ST structure as well as the A-ST interface observed in our crystal structure are independent of crystal packing and should be physiologically relevant.
Since the human and murine Aα subunits are identical except for one residue that is distant from the ST binding site (Ser324 in human, Thr324 in mouse), this structure is likely representative of the interaction of ST with the human PP2A.
Moreover, since most ST residues in the structural core and the A-ST interface are conserved among ST and middle t (MT) proteins in the polyomavirus family, this structure provides a structural basis for understanding the oncogenic activities of each of the ST and MT proteins in the polyomavirus family.
We found that the N-terminal J and the C-terminal unique domains of ST are structurally coupled.
The conserved hydrophobic interface between the J and unique domains may be important for the structural integrity of ST.
Previous studies have been shown that each ST molecule contains two zinc ions [33,34].
Since ST contains two cysteine cluster motifs (CXCXXC) that are absolutely conserved, it was proposed that ST may resemble GAL4, which contains a Zn(II)2Cys6 binuclear cluster [33,34].
In our ST crystal structure, instead of forming a binuclear cluster, these two zinc ions are located in two separate positions and form two novel zinc-binding motifs.
The first and second cysteine clusters (residues 111–116, 138–143, respectively) form two zinc-binding motifs together with the conserved Cys103 and His122, respectively.
Both zinc ions interact with helix α5 and stabilize the structure of the C-terminal unique domain.
The Aα subunit of PP2A consists of 15 HEAT repeats, with each repeat containing two antiparallel helices.
Overall the PP2A Aα subunit forms a horseshoe shape structure.
SV40 ST “rides” on the structural ridge formed by intra-HEAT-repeat loops of PP2A Aα subunit (Figure 1).
Previous work has demonstrated that the unique domain of SV40 ST is the essential binding domain for PP2A AC core complex [5].
In our structure the second zinc-binding motif, in particular the L6 loop, forms extensive interactions with the PP2A Aα subunit.
In support of our structural observations, in vitro structure-directed mutagenesis studies demonstrate that the interface between the second zinc-binding motif, in particular Pro132 that intercalates between Trp140 and Phe141 of PP2A A subunit, is essential for A-ST interactions (Figure 7).
Point mutations of two residues in the ST J domain (R7A and R21A), or mutation of an A subunit residue that forms a hydrogen bond with ST Arg21 (E216A, Figure 3), all abolish the A-ST interaction (Figure 7), confirming the J domain is directly involved in A subunit interaction.
In addition, this structure is completely consistent with previous mutagenesis studies of PP2A A subunit [44], since essentially all A subunit mutations that have weaker ST binding activities map to the A-ST interface in our structure.
Potential Interaction between ST and PP2A C Subunit
Previous work has indicated that a region of the unique domain encompassing the first cysteine cluster is necessary for the binding of ST to PP2A, and its N-terminal flanking region (residues 97–103) are also important for PP2A interactions [35,47].
However, the first zinc-binding motif does not interact with the A subunit of PP2A in the crystal structure.
Instead, the crystal structure suggests that the first zinc-binding motif may directly interact with the C subunit near its active site, since the first zinc-binding motif is spatially close to the active site of the PP2A C subunit in the structural superposition of PP2A and A-ST complexes (Figure 5).
This hypothesis is supported by structural flexibility of the scaffold A subunit.
Our structural comparison indicates that HEAT repeats 10–13 of PP2A Aα subunit have substantial structural flexibility (Figure 6).
This flexibility of the Aα subunit may allow for the accommodation of different types of regulatory B subunits or B-like proteins, such as ST, into the AC core enzyme that interact with both A and C subunit.
Consistent with prior reports [23,24], we failed to observe stable direct interaction between purified ST and PP2A C subunit using a glutathione S-transferase (GST) pull-down assay (unpublished data).
It is possible that ST directly interacts with the C subunit via the formation of a stable complex between ST and the A subunit.
Therefore, ST may form trivalent interactions with the PP2A AC complex—two of them (via the J domain and the second zinc-binding motif) interacting with the A subunit, and the third one (via the first zinc-binding motif) binding to the C subunit.
This hypothesis explains why the first zinc-binding motif does not directly interact with PP2A A subunit, yet is required for the interaction with and the inhibition of phosphatase activity of PP2A AC core complex [35].
This model is also supported by the observation that ST fragments containing both J domain and the first but not the second zinc-binding motif interact with PP2A AC complex and inhibit PP2A AC dimer phosphatase activity [29,35].
ST may interfere with substrate binding via its interaction near the active site of the C subunit.
Therefore, in addition to competing with the PP2A B subunit for PP2A A subunit binding, ST may directly modulate the phosphatase activity of the AC core complex, which accounts for substantial proportion of PP2A enzyme in the cell.
Future work will be needed to understand if and how ST may directly interact with the C subunit.
The Role of the ST J Domain in the Interaction with PP2A
Prior studies have demonstrated that the unique domain but not the J domain was sufficient for interaction with PP2A AC complex.
Although not essential for A-ST interaction, the deletion of the J domain significantly decreased the inhibitory activity of ST on the PP2A AC core dimer [35], suggesting that the J domain enhances the binding of ST to the PP2A A subunit.
Consistent with this view, mutation of either Arg7 or Arg21, two residues on the J domain surface involved in PP2A A interaction, disrupts the interaction between ST and A subunit.
Alternatively, the J domain may play a role in stabilizing the spatial position of the first zinc-binding motif by allowing its efficient interaction with the C subunit.
This interaction may be particularly important in the AC-ST complex formation, because the structural flexibility of HEAT repeats 10–13 of the A subunit may not permit the C subunit to stay in a fixed position and interact with ST efficiently.
Indeed, although the unique domain of ST binds to PP2A A, this binding fails to inhibit PP2A AC phopshatase activity [35].
In this regard, we note that the second zinc-binding motif binds to the A subunit primarily through loop–loop interactions and on the concave side of the A subunit structure only, while the J domain interacts with the convex side of A subunit.
While the second zinc-binding motif may be the primary docking site, the J domain may fix the relative orientation between ST unique domain and the A subunit, with the N-terminal J and C-terminal unique domains sitting on the convex and concave side of the horseshoe shape, respectively.
This orientation stabilization may be important for the first zinc-binding motif to effectively inhibit the phosphatase activity of the PP2A C subunit.
In addition to the inhibition of the phosphatase activity of the PP2A AC dimer, the J domain may also play a role in the oncogenic activity of ST by providing an additional binding site for Hsp70, even when in complex with the PP2A AC complex, as suggested by our crystal structure.
The potential simultaneous interaction with PP2A and Hsp70 may couple these two functions of ST.
For example, ST may bring PP2A and Hsp70 together to allow for the dephosphorylation of protein(s) bound to Hsp70.
In summary, our structural and biochemical studies reveal the structure of the ST family and define the interaction between ST and the A subunit of PP2A.
In addition, our work suggests that ST may directly interact and regulate the activity or substrate specificity of the PP2A catalytic C subunit, and Hsp70 may bind to PP2A-bound ST and thus define PP2A activity and/or substrate specificity.
Taken together, our work provides a structural basis for the oncogenic activity of ST and MT antigens in the polyomavirus family.
Since ST binds to PP2A Aα in a manner similar to that used by the regulatory B subunits, these findings provide not only new insights into the regulation of PP2A but may also provide a foundation for the development of small molecules that alter the function of PP2A.
Materials and Methods
Expression and purification of SV40 ST in complex with PP2A A subunit.
Full-length SV40 ST (strain VA45‐54‐2) and full-length mouse PP2A Aα subunit with tobacco etch virus protease (TEV) cleavage sites were cloned into pGEX4T1 vector (Amersham Biotech, http://www.gelifesciences.com) and pET28a vector (Novagen, http://www.emdbiosciences.com/html/NVG/home.html), respectively, and co-transformed into E.
coli strain BL21 (star) (Invitrogen, http://www.invitrogen.com).
Mouse and human Aα subunits are identical in protein sequence, except for one residue that is distant from the ST binding site (Ser324 in human, Thr324 in mouse).
Coexpression of Aα subunit and SV40 ST was induced by the addition of 0.1 mM IPTG at OD600 = 0.6 upon shifting the temperature from 37 °C to 18 °C, and cells were grown for an additional 18 h.
Cells were then collected by centrifugation and resuspended with lysis buffer (30 mM Tris-HCl [pH 8.0], 50 mM NaCl, 5 mM β-mercaptoethanol) including protease inhibitors (PMSF, leupeptin, and benzamidine).
Resuspended cells were lysed by sonication, and cell debris removed by centrifugation at 26,000 g for 1 h.
Soluble fractions were filtered with 0.8 μm syringe filters and applied into a Ni-NTA affinity column pre-equilibrated with 30 mM Tris-HCl (pH 8.0), 50 mM NaCl, 5 mM β-mercaptoethanol.
Target protein complexes (the Aα subunit with GST-tag and SV40 ST with His-tag) were eluted with elution buffer (30 mM Tris-HCl [pH 8.0], 50 mM NaCl, 300 mM imidazole, 5 mM β-mercaptoethanol) and dialyzed overnight at 4 °C in 30 mM Tris-HCl (pH 8.0), 50 mM NaCl, 5 mM DTT.
Dialyzed protein was applied to a GST affinity column to remove free SV40 ST, and on-column cleavage with TEV protease was performed at 4 °C overnight.
The flow-through fraction of the GST column was reapplied into the Ni-NTA column to remove cleaved His-tag and TEV protease.
The flow-through fraction of the Ni-NTA column was concentrated and applied into a Superdex 200 size-exclusion column (Amersham Biotech) pre-equilibrated with 30 mM Tris-HCl (pH 8.0), 50 mM NaCl, 5 mM DTT.
Fractions of the heterodimeric complex of full-length Aα subunit and full-length SV40 ST were pooled and concentrated up to 10 mg/ml and used for the crystallization trial.
Crystals of the A-ST complex were obtained at room temperature with the hanging drop vapor diffusion method.
After optimizing initial conditions and extensively searching for additives, combined with microseeding, the optimal crystallization condition was 16% PEG 3350, 0.2 M ammonium formate, 30 mM spermine, 6% 6-aminocaproic acid, 10 mM DTT.
Crystals were cryoprotected by gradually increasing the PEG 3350 concentration up to 26% and frozen with liquid nitrogen.
Data collection and structure determination of the PP2A Aα subunit and SV40 ST complex.
Data collection was performed at Advanced Light Source beamline 5.0.1.
The best dataset has a resolution of 3.1 Å, with high redundancy.
The space group was P1, with the unit cell dimensions a = 94.69 Å, b = 105.50 Å, c = 111.98 Å, α = 115.64°, β = 109.57°, γ = 94.10°.
These datasets were integrated and scaled using HKL2000 and SCALEPACK [48].
There were four heterodimers in one asymmetric unit with 54% solvent content.
Initial molecular replacement was performed with Phaser [49] by using the Aα subunit structure in the trimeric PP2A complex [25] as the search model.
The program Phaser found all four Aα subunits with pseudo D2 symmetry.
However, both Rfree and Rwork were over 55% after overall rigid-body refinement, suggesting substantial conformational changes in the Aα subunit.
Therefore, rigid body refinement was performed in Refmac5, with every single HEAT repeat as a separate group [50].
After this rigid body refinement, Rfree dropped to about 40%.
Since SV40 ST was demonstrated to contain two zinc-binding sites [34] that gave significant anomalous signal at the experimental wavelength (λ = 1.000 Å), the same dataset was scaled as an anomalous dataset using HKL2000.
The anomalous difference map, calculated using phases derived from the refined molecular replacement solution, gave the positions of eight zinc sites (for four ST molecules in the asymmetric unit) without any ambiguity.
Zinc sites were further refined with Sharp [51], and phases were calculated together with the partial model of the Aα subunit.
The calculated map showed clear density of SV40 ST after the density modification with DM.
Since the conformation of each four Aα subunit was different, noncrystallographic symmetry averaging was helpful only to build SV40 ST.
Further model building was done with Xtalview [52] and COOT [53].
TLS refinement was performed using Refmac5 in the CCP4 suite [50].
Because of the conformational variations in the A subunit, noncrystallographic symmetry averaging was not applied during refinement.
The final model has an Rfree of 30.5%, and an Rwork of 24.4%.
Zinc sites are confirmed by anomalous difference map.
None of the nonglycine residues are in the disallowed region of the Ramanchandran plot.
The final structural models of ST contain residues 1–44, 48–78, 85–172 (chain e); 1–40, 49–74, 87–172 (chain f); 1–41, 51–70, 92–172 (chain g); and 1–41, 51–73, 94–172 (chain h).
In vitro binding assay.
GST-tagged ST and Aα mutants were generated using the QuickChange Site-Directed Mutagenesis Kit (Stratagene) and expressed in E.
coli strain BL21 (star).
Wild-type Aα and Aα mutants were isolated with glutathione-sepharose (Amersham Biosciences) GST-tags were removed from wild-type ST and ST mutants and equal amounts of wild-type ST or ST mutants were added to Aα-glutathione-sepharose precipitates.
Binding assays were performed in 30 mM Tris (pH 8.0), 50 mM NaCl, 5 mM DTT, 0.2% NP-40 buffer for 4 h.
The beads were washed five times, and the proteins were eluted with reduced glutathione, followed by SDS-PAGE and immunoblotting.
For immunoblotting, we used affinity-purified polyclonal antibodies against SV40 ST [16] and monoclonal antibodies (clone 6F9) against Aα (Abcam, http://www.abcam.com).
Supporting Information
Accession Numbers
Coordinates and structural factors have been deposited in the Protein Data Bank (PDB, http://www.rcsb.org/pdb) with the accession code 2PF4.
Abbreviations
GST
glutathione S-transferase
HEAT
huntingtin, elongation factor 3, A subunit of protein phosphatase 2A, and TOR (target of rapamycin)
Hsp70
heat shock protein 70
LT
large T antigen
MT
middle T antigen
PDB
Protein Data Bank
PP2A
protein phosphatase 2A
RMSD
root mean square deviation
ST
small t antigen
SV40
simian virus 40
TEV
tobacco etch virus protease
References
Simian virus 40 infection in humans and association with human diseases: Results and hypotheses
Cellular targets of the SV40 small-t antigen in human cell transformation
The role of the SV40 ST antigen in cell growth promotion and transformation
Critical role for SV40 small-t antigen in human cell transformation
Enumeration of the simian virus 40 early region elements necessary for human cell transformation
Species- and cell type-specific requirements for cellular transformation
Protein phosphatase 2A: The Trojan horse of cellular signaling
Protein phosphatase 2A: A highly regulated family of serine/threonine phosphatases implicated in cell growth and signalling
Protein phosphatase 2A: Who shall regulate the regulator?
Protein phosphatase 2A: A panoply of enzymes
Molecular model of the A subunit of protein phosphatase 2A: Interaction with other subunits and tumor antigens
Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the A alpha subunit gene
A truncated isoform of the PP2A B56 subunit promotes cell motility through paxillin phosphorylation
Inhibitors of protein kinases and phosphatases
SV40 early region oncoproteins and human cell transformation
Identification of specific PP2A complexes involved in human cell transformation
Alterations of the PPP2R1B gene in human lung and colon cancer
Alterations in protein phosphatase 2A subunit interaction in human carcinomas of the lung and colon with mutations in the A beta subunit gene
Alterations of the PPP2R1B gene located at 11q23 in human colorectal cancers
PPP2R1B gene alterations inhibit interaction of PP2A-Abeta and PP2A-C proteins in colorectal cancers
Cancer-associated PP2A alpha subunits induce functional haploinsufficiency and tumorigenicity
The tumor suppressor PP2A A beta regulates the RalA GTPase
Control of protein phosphatase 2A by simian virus 40 small-t antigen
Identification of binding sites on the regulatory A subunit of protein phosphatase 2A for the catalytic C subunit and for tumor antigens of simian virus 40 and polyomavirus
Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme
Structure of the protein phosphatase 2A holoenzyme
Polyoma small and middle T antigens and SV40 small t antigen form stable complexes with protein phosphatase 2A
Comparison of heterotrimeric protein phosphatase 2A containing different B subunits
The interaction of SV40 small tumor antigen with protein phosphatase 2A stimulates the map kinase pathway and induces cell proliferation
Dephosphorylation of simian virus 40 large-T antigen and p53 protein by protein phosphatase 2A: Inhibition by small-t antigen
SV40 17KT antigen complements dnaj mutations in large T antigen to restore transformation of primary human fibroblasts
Involvement of PP2A in viral and cellular transformation
Effect of zinc ions on the biochemical behavior of simian virus 40 small-t antigen expressed in bacteria
Simian virus 40 small-t antigen binds two zinc ions
Identification of structural elements involved in the interaction of simian virus 40 small tumor antigen with protein phosphatase 2A
The J-protein family: Modulating protein assembly, disassembly and translocation
Structure, function and evolution of DnaJ: Conservation and adaptation of chaperone function
Role of the J-domain in the cooperation of Hsp40 with Hsp70
Interaction of the Hsp70 molecular chaperone, DnaK, with its cochaperone DnaJ
Genetic analysis of the polyomavirus DnaJ domain
Structural basis for the inactivation of retinoblastoma tumor suppressor by SV40 large T antigen
NMR structure of the N-terminal J domain of murine polyomavirus T antigens.
Implications for DnaJ-like domains and for mutations of T antigens.
Zinc fingers, zinc clusters, and zinc twists in DNA-binding protein domains
Binding specificity of protein phosphatase 2A core enzyme for regulatory B subunits and T antigens
The structure of the protein phosphatase 2A PR65/A subunit reveals the conformation of its 15 tandemly repeated HEAT motifs
Structure of protein phosphatase 2A core enzyme bound to tumor-inducing toxins
Mutations which affect the inhibition of protein phosphatase 2A by simian virus 40 small-t antigen in vitro decrease viral transformation
Processing of X-ray diffraction data collected in oscillation mode
Pushing the boundaries of molecular replacement with maximum likelihood
Efficient anisotropic refinement of macromolecular structures using FFT
Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methods
XtalView/Xfit—A versatile program for manipulating atomic coordinates and electron density
Coot: Model-building tools for molecular graphics
Author contributions.
USC, WCH, and WX conceived and designed the experiments.
USC, SM, AAS, JDA, and WX performed the experiments.
All authors analyzed the data.
USC, SM, AAS, and JDA contributed reagents/materials/analysis tools.
USC, WCH, and WX wrote the paper.
Funding.
This work was supported in part by an Investigator's Award from the Burroughs Welcome Fund (WX), the Keck Center for Pathogenesis at the University of Washington (WX), US National Cancer Institute P01 CA50661 (WCH), and a Howard Hughes Medical Institute predoctoral fellowship (JDA).
Competing interests.
The authors have declared that no competing interests exist.UXT is a novel and essential cofactor in the NF-κB transcriptional enhanceosome
As a latent transcription factor, nuclear factor κB (NF-κB) translocates from the cytoplasm into the nucleus upon stimulation and mediates the expression of genes that are important in immunity, inflammation, and development.
However, little is known about how it is regulated inside the nucleus.
By a two-hybrid approach, we identify a prefoldin-like protein, ubiquitously expressed transcript (UXT), that is expressed predominantly and interacts specifically with NF-κB inside the nucleus.
RNA interference knockdown of UXT leads to impaired NF-κB activity and dramatically attenuates the expression of NF-κB–dependent genes.
This interference also sensitizes cells to apoptosis by tumor necrosis factor-α.
Furthermore, UXT forms a dynamic complex with NF-κB and is recruited to the NF-κB enhanceosome upon stimulation.
Interestingly, the UXT protein level correlates with constitutive NF-κB activity in human prostate cancer cell lines.
The presence of NF-κB within the nucleus of stimulated or constitutively active cells is considerably diminished with decreased endogenous UXT levels.
Our results reveal that UXT is an integral component of the NF-κB enhanceosome and is essential for its nuclear function, which uncovers a new mechanism of NF-κB regulation.
Introduction
Nuclear factor κB (NF-κB) is a widely expressed primary transcription factor composed of a heterodimeric complex (p65–p50).
A myriad of unrelated exogenous and endogenous stimuli are capable of inducing NF-κB activity.
In turn, NF-κB regulates the expression of an equally diverse array of cellular genes that are important in immunity, inflammation, and development (Ghosh et al., 1998; Li and Verma, 2002).
Aberrance of its function has been linked to such pathological processes as cancer and abnormal development (for review see Rayet and Gelinas, 1999).
Determining molecular mechanisms that regulate the activation of NF-κB is crucial to understand how multiple intracellular signaling pathways converge to activate a single transcription factor.
NF-κB is normally localized in the cytoplasm as an inactive complex through physically associating with its inhibitory molecule IκBα.
Extensive studies have been performed to address how various stimuli trigger its translocation from the cytoplasm into the nucleus (Hayden and Ghosh, 2004).
Seminal works from several laboratories have determined a sequence of biochemical events that result in the ubiquitin-dependent degradation of IκB proteins (Chen et al., 1996; Hatakeyama et al., 1999; Spencer et al., 1999).
Consequently, this releases NF-κB to move into the nucleus and switch on the expression of target genes (Ghosh and Karin, 2002; Li and Verma, 2002).
NF-κB belongs to the Rel homology domain (RHD) family of transcription factors that exploit similar strategies to achieve initial activation.
Recently, an alternative pathway was identified to regulate another member (p100) of this family (Senftleben et al., 2001).
Compared with the fruitful know-how about the molecular events of NF-κB activation in cytoplasm, much less is understood concerning its active regulation and functional interaction with other proteins inside the nucleus.
Recent progress has shed light on the importance of nuclear events in shaping the strength and duration of the NF-κB transcriptional response, which is achieved partly by posttranslational modification of the NF-κB transcription factor complex or the histones that surround various NF-κB target genes (Chen and Greene, 2004).
For example, I κ-B kinase α (IKKα) was demonstrated to accelerate both the turnover of NF-κB and its removal from proinflammatory gene promoters (Lawrence et al., 2005).
This kinase could also phosphorylate histone H3 and was critical for NF-κB–responsive gene expression (Yamamoto et al., 2003).
Additionally, the acetyltransferase activity of p300/CREB-binding protein (CBP) was required for the activation of NF-κB–dependent transcriptions.
p300/CBP proteins were also found to directly associate with NF-κB, forming a bridge to the basal transcriptional machinery (Gerritsen et al., 1997; Chan and La Thangue, 2001).
Although a couple of cofactors were recently shown to reside in the NF-κB enhanceosome, much remains to be done to understand their specific functions and regulatory mechanisms.
This indicates that NF-κB assembles a much higher order transcription complex than once expected and that there are additional important layers of regulation for the NF-κB transactivation process.
Ubiquitously expressed transcript (UXT) is ∼18 kD and predominantly localizes in the nucleus (Markus et al., 2002).
It was demonstrated to be widely expressed in human and mouse tissues, and its expression was markedly elevated in some human tumors (Schroer et al., 1999; Zhao et al., 2005).
Computer modeling predicted that UXT was an α-class prefoldin (PFD) family protein (Gstaiger et al., 2003).
Most members of this family are small molecular mass proteins (14–23 kD) and are composed of coiled-coil structures.
Yeast and human PFDs 1–6 were previously found to assemble into a hexameric complex that functioned as a new type of molecular chaperone (Vainberg et al., 1998; Siegert et al., 2000).
Until now, the functional characterization of UXT was scarce.
One recent study indicated that UXT bound to the N terminus of the androgen receptor and regulated androgen receptor–responsive genes that are important in prostate growth suppression and differentiation (Markus et al., 2002; Taneja et al., 2004).
Another investigation suggested UXT to be a component of the centrosome (Zhao et al., 2005).
However, much remains to be done as to the in vivo function of UXT and its regulatory roles in cellular processes.
During a systematic screening for proteins that interacted with components of the NF-κB enhanceosome, we identified UXT as a novel p65-interacting protein.
This interaction is confirmed both in vitro and in vivo.
In this study, we show that RNAi knockdown of UXT leads to impaired NF-κB activity and dramatically attenuates the expression of NF-κB– dependent genes.
This interference also sensitizes cells to apoptosis by TNF-α.
Furthermore, UXT forms a signal- dependent complex with NF-κB and is recruited to the NF-κB enhanceosome upon stimulation.
Interestingly, the UXT protein level correlates with constitutive NF-κB activity in human prostate cancer cell lines.
The presence of NF-κB within the nucleus of stimulated or constitutively active cells is considerably diminished with decreased endogenous UXT protein levels.
Collectively, our investigation reveals that UXT is an integral component of the NF-κB enhanceosome and is essential for its nuclear function, which uncovers a new mechanism of NF-κB regulation.
Results
Identification of UXT as a p65-interacting protein
To identify new components of the NF-κB enhanceosome, we performed a systematic yeast two-hybrid screening in which the cDNA fragment harboring the RHD of p65 (amino acids 1–312) was used as bait.
Several positive clones were identified to encode full-length UXT (Fig.
1 A).
In addition, previously confirmed p65-interacting proteins (e.g., IκBα and PIAS3) were screened out.
UXT interacts with p65 in vitro and in vivo.
(A) Interaction between p65 and UXT in a yeast two-hybrid assay.
(B) Subcellular localization of endogenous and exogenous UXT.
293T cells were transfected with (top) or without (bottom) FLAG-UXT.
Immunofluorescentmicroscopy was performed with the indicated primary antibodies.
(C) Full-length HA-p65 and FLAG-UXT proteins were labeled with [35S]methionine by in vitro translation.
The products were mixed and immunoprecipitated with the indicated antibodies coupled onto protein A/G beads.
The immunoprecipitates were resolved by SDS-PAGE and visualized by autoradiography.
10% of the input proteins for pull down are shown at the left.
(D) 293T cells were transfected with HA-UXT.
24 h after transfection, cells were treated with 10 ng/ml TNF-α for the indicated times and fractionated to cytoplasmic and nuclear fractions, which were immunoprecipitated and immunoblotted with the indicated antibodies, respectively.
(E) 293T cells were treated with 10 ng/ml TNF-α for the indicated times.
Whole cell lysates were immunoprecipitated and immunoblotted with the indicated antibodies.
Bar, 10 μm.
UXT was previously reported to be expressed almost exclusively inside the nucleus of most cells (Markus et al., 2002).
This was confirmed in our investigation for either endogenous or overexpressed UXT (Fig.
1 B).
To further substantiate its interaction with p65, an in vitro coimmunoprecipitation assay was applied in which full-length HA-p65 and FLAG-UXT proteins were generated and labeled, respectively, with [35S]methionine by in vitro translation.
The products were mixed and immunoprecipitated with either control IgG or anti-HA antibody.
As shown in Fig.
1 C, UXT could be coprecipitated by antibody against the HA epitope but not by control IgG, which suggests that UXT indeed interacts directly with full-length p65.
To address the physiological relevance of this interaction in mammalian cells, we expressed HA-UXT in 293T cells and then stimulated cells with or without TNF-α for the indicated times.
The fractionated cytoplasmic or nuclear extracts were immunoprecipitated with either anti-p65 antibody or IgG as a control, respectively.
There was no detectable UXT that interacted with cytoplasmic p65 in the presence or absence of TNF-α (Fig.
1 D), which was consistent with the unique subcellular location of UXT.
In addition, there was only a marginal amount of endogenous p65 in the nucleus devoid of TNF-α treatment.
Consequently, no UXT was coimmunoprecipitated from this nuclear extract even though there existed a large amount of UXT.
In contrast, there exhibited a strong interaction between nuclear p65 and UXT upon TNF-α stimulation.
Furthermore, we tested whether endogenous UXT and p65 could interact in response to TNF-α.
As shown in Fig.
1 E, endogenous UXT was coimmunoprecipitated by p65 antibody from cells treated with TNF-α.
In contrast, UXT was barely detected in the immunoprecipitates without TNF-α treatment.
One possible explanation for this phenomenon is that only after p65 translocation into the nucleus could UXT have access to p65.
However, we could not formally rule out the possibility that posttranslational modifications of either protein were prerequisites for this interaction in vivo.
Collectively, these results indicate that UXT interacts in vivo with p65 upon TNF-α stimulation.
RHD of p65 mediates its interaction with UXT
The p65 subunit of NF-κB harbors an N-terminal conserved region (∼300 amino acid residues) known as RHD and a C-terminal transactivation domain.
To explore the UXT-binding region within p65, we constructed a series of p65 deletion mutants (Fig.
2 A).
It was found that the loss of amino acids 1–285 at the N terminus of p65 resulted in its complete inability to interact with UXT (Fig.
2 B, top).
In contrast, p65 fragments spanning amino acids 1–286, 1–312, or 1–372 fully retained their binding capability and interacted with UXT as well as the wild type (Fig.
2 B, middle).
Because the RHD of p65 consisted of two Ig-like domains (Chen et al., 1998), we made two additional deletion mutants of p65 (amino acids 1–190 and 191–551), each of which contained only one Ig-like domain.
Immunoprecipitation assays revealed that neither of them was able to interact with UXT (Fig.
2 B, bottom).
In addition, we generated several point mutations of UXT and did not observe a considerable change on UXT and p65 interaction.
We also had attempted in vain to express truncation mutants of UXT in mammalian cells, which prevented us from further dissecting UXT.
Collectively, these data suggest that the intact RHD of p65 is both essential and sufficient to mediate interaction with UXT.
RHD of NF-κB mediates its interaction with UXT.
(A) Schematic illustration of p65 and its mutants.
RHD, Rel homology domain; NLS, nuclear localization signal; TAD, transactivation domain.
(B) Tagged full-length UXT was transfected into 293T cells along with p65 and its deletion mutants as indicated.
Whole cell lysates were immunoprecipitated and immunoblotted with the indicated antibodies.
(C) 293T cells were transfected with FLAG-UXT together with HA-p50, myc-cRel, and HA–lymphoid enhancer binding factor 1.
Cell lysates were immunoprecipitated and immunoblotted with the indicated antibodies.
The RHD structure defined a highly conserved family of transcription factors (Rel family) that are important in immunity and inflammation (Ghosh et al., 1998).
This led us to wonder whether this interaction was also applicable to other proteins of this family.
To explore this possibility, we transfected UXT into 293T cells along with p50 or c-Rel.
Interestingly, UXT was also capable of binding to p50 or c-Rel specifically and strongly (Fig.
2 C).
In contrast, lymphoid enhancer binding factor 1, a transcription factor unrelated to the Rel family, did not have any affinity to UXT.
This phenomenon suggested that UXT recognized a consensus structure fold and that there was probably a unified theme for its interaction with p65 and other members of the Rel family.
Overexpression of UXT does not markedly affect NF-κB activation
Because UXT interacted specifically inside the nucleus with p65 in a signal-dependent manner, we went on to address whether this interaction was important for regulating NF-κB activity and genes responsive to it.
We transfected 293T cells with UXT or other control plasmids, prepared nuclear extracts, and performed electrophoretic mobility shift assay (EMSA) as indicated.
Consistently, there was no NF-κB binding to its cognate probe without TNF-α treatment.
Notably, the overexpression of UXT alone did not induce any detectable basal NF-κB binding activity.
In contrast, robust NF-κB binding activity was induced upon TNF-α stimulation.
Expectedly, this activity was severely impaired by A20, a potent inhibitor of NF-κB signaling (Wertz et al., 2004).
However, in response to TNF-α, we observed neither inhibitory nor synergically stimulatory effects on NF-κB binding affinity (Fig.
3 A).
Overexpression of UXT does not markedly affect NF-κB activation induced by TNF-α.
(A) 293T cells were transfected with an equal amount of UXT, A20, or control vector.
24 h after transfection, cells were stimulated with 10 ng/ml TNF-α for the indicated times.
Equal amounts (10 μg) of nuclear extracts were subjected to EMSA.
For competition analysis, 100-fold excess of unlabeled wild-type or mutant κB probes were added to the reaction mixtures.
For supershift assays, nuclear extracts were incubated with antibody as indicated.
(B) 293T cells were transfected with equal amounts of UXT, IκBαSR, or empty vector.
24 h after transfection, cells were stimulated with 10 ng/ml TNF-α for 30 min or left untreated.
Relative mRNA levels of A20 and IL-8 were analyzed by real-time RT-PCR.
Data represent means ± SD (error bars) of at least three independent experiments.
Alternatively, we explored whether overexpressing UXT had any measurable effects on inducible genes such as A20 or interleukin-8 (IL-8) that were regulated by NF-κB.
On transfection of UXT alone, we did not observe any change on the basal expression of these two genes (unpublished data).
Thus, we transfected 293T cells with UXT or IκBα super repressor (SR) and analyzed the amount of A20 or IL-8 mRNA induced by TNF-α, respectively, via quantitative real-time RT-PCR.
IκBαSR was a well-established potent inhibitor of NF-κB activation (Diao et al., 2005).
Consistently, A20 or IL-8 inductions by TNF-α were severely attenuated in the presence of IκBαSR.
Interestingly, we observed marginal synergic inductions of both A20 and IL-8 by TNF-α in the presence of UXT (Fig.
3 B), which suggested that UXT might modulate NF-κB function.
Knockdown of UXT sharply attenuates NF-κB activation by multiple stimuli
Because we failed via overexpressing UXT to convincingly demonstrate the involvement of UXT in NF-κB regulation, we turned to investigate the effect if endogenous UXT expression was reduced via RNAi.
Two UXT siRNAs were fished out, which are designated as 242 and 428 hereafter.
We confirmed the effectiveness of them against UXT by monitoring the mRNA level (Fig.
4 A), protein level (Fig.
4 B), and cellular immunofluorescence of UXT (Fig.
4 C).
428 was reproducibly better than 242 and was used more frequently in later experiments.
In addition, two negative control siRNAs were used.
One was a nonspecific siRNA and was named as control; the other is a mutant form of siRNA 428 (428m) that could partially bind to UXT mRNA but lost the interfering ability.
Knockdown of UXT attenuates NF-κB activation induced by specific stimuli.
(A) 293T cells were transfected with the indicated siRNA.
48 h after transfection, the endogenous expression of UXT mRNA was monitored by RT-PCR (top) or by real-time RT-PCR (bottom).
(B) Cell lysates from A were subjected to Western blotting for determining endogenous UXT protein levels after siRNA transfection.
(C) FLAG-UXT was cotransfected with the indicated siRNA.
24 h after transfection, UXT protein levels were monitored by immunofluorescence.
(D) 293T cells were cotransfected with 3×κB-Luc and siRNA as indicated.
48 h after transfection, cells were stimulated with 10 ng/ml TNF-α or 20 ng/ml IL-1 for 7 h before luciferase assays were performed.
(E) The indicated siRNA and 3×κB-Luc were transfected into 293T cells along with MyD88 or TRAF6.
48 h after transfection, cells were assayed as in D.
(F) p65 was transfected into 293T cells along with the indicated siRNA and 3×κB-Luc.
Cells were treated and assayed as in E.
(G) c-Rel was transfected into 293T cells along with the indicated siRNA and mp40-Luc.
Cells were treated and assayed as in E.
(H) RAW264.7 cells were cotransfected with 3×κB-Luc and the indicated siRNA.
48 h after transfection, cells were stimulated with 500 ng/ml lipopolysaccharide for 7 h before luciferase assays were performed.
(I) 293T cells were transfected with the indicated siRNA and stimulated by 10 ng/ml TNF-α for the indicated times.
Endogenous mRNA expressions of IL-8, A20, and IκBα were measured by real-time RT-PCR.
Data represent means ± SD (error bars) of at least three independent experiments.
Bar, 200 μm.
Initially, we used a κB-Luc reporter gene to evaluate the effect of UXT knockdown on NF-κB activation status.
Excitedly, luciferase assay revealed that the decrease of endogenous UXT considerably inhibited NF-κB transcriptional activity induced by TNF-α, IL-1β (Fig.
4 D), or lipopolysaccharide (Fig.
4 H).
Notably, this phenomenon also held true for basal luciferase expression.
Likewise, similar effects were observed in cells with reduced UXT expression when the cells were stimulated by overexpressing MyD88 or TRAF6, which are well known to induce NF-κB activity (Fig.
4 E; Muzio et al., 1997; Deng et al., 2000).
In addition, gene activation induced by p65 alone was also markedly attenuated in cells with a decreased expression of endogenous UXT (Fig.
4 F).
Our data have shown that UXT could interact with other Rel family proteins like c-Rel (Fig.
2 C).
c-Rel was previously found to regulate the expression of the IL-12 p40 subunit by specifically interacting with its promoter (Gri et al., 1998).
Thus, a reporter gene was used that harbored firefly luciferase under the control of the mouse p40 promoter.
Consistently, c-Rel activation was severely impaired in cells expressing reduced amounts of UXT.
This attenuation was in direct proportion to the intensity of RNAi, as evidenced in oligonucleotides 428 versus 242 (Fig.
4 G).
To make it more physiologically relevant, we also investigated how the induction of NF-κB–dependent genes (IL-8, A20, and IκBα) was influenced by knocking down endogenous UXT expression.
For cells transfected with control siRNA or UXT siRNA, quantitative real-time RT-PCR was used to measure endogenous mRNA levels of IL-8, A20, and IκBα induced by TNF-α during a time course.
Consistently, these inductions were sharply attenuated when endogenous UXT expression was suppressed (Fig.
4 I).
Collectively, our data strongly indicate that UXT is required for inducing genes tightly regulated by NF-κB and that it plays an essential role in NF-κB function.
Knockdown of UXT attenuates NF-κB binding to its cognate promoter
As was stated in the Introduction, NF-κB activation involved a series of molecular events both in the cytoplasm and in the nucleus.
To rule out the possibility that UXT might act on processes other than directly on NF-κB itself, we examined the effects of UXT knockdown on the phosphorylation and degradation of IκBα.
We did not find any differences in this regard between UXT-deficient and normal cells during TNF-α stimulation (Fig.
5 A).
This was further supported by the observation that IKK kinase activity was not affected at all in terms of UXT knockdown (unpublished data).
Given that UXT was expressed almost exclusively inside the nucleus and interacted directly with NF-κB, these data strongly indicate that UXT performed its function via targeting NF-κB itself.
Knockdown of UXT attenuates the activity and amount of nuclear NF-κB.
(A) 293T cells were transfected with the indicated siRNAs.
After 48 h, cells were induced by 10 ng/ml TNF-α for the indicated times.
Western blotting was performed on the cell extracts to check the phosphorylation and degradation of IκBα.
(B) 293T cells were treated as in A.
EMSA was performed to test endogenous NF-κB or sp1 binding to their cognate probes.
(C) 293T cells were treated as in A.
(D) 293T cells were transfected with the indicated siRNAs.
After 48 h, cells were induced by 10 ng/ml TNF-α for 30 min.
Cytoplasmic and nuclear fractions were prepared and immunoblotted with the indicated antibodies, respectively.
(E) Cells were treated as in D and stained with anti-p65 primary antibody and FITC-conjugated secondary antibody.
The nucleus was counterstained with DAPI.
Quantification was performed to 100–200 cells in the same ranges of microscopy field for the presence of an appreciable nuclear signal of p65.
Only those showing typical focused nuclear p65 were counted.
The data are presented as percentages of cells with nuclear p65 versus total cells.
(F) 293T cells were transfected with siRNA 428.
After 48 h, cells were induced by 10 ng/ml TNF-α or 10 ng/ml TNF-α plus 20 ng/ml LMB for the indicated times and stained with anti-p65 primary antibody and FITC-conjugated secondary antibody.
The nucleus was counterstained with DAPI.
(G) 293T cells were cotransfected with the indicated siRNAs, 3×κB-Luc, and chimeric p65 (1–312)-VP16.
Luciferase assays were performed 48 h after transfection.
(H) 293T cells were cotransfected with the indicated siRNAs, Gal4-Luc, and chimeric Gal4 BD-p65 (285–551).
Luciferase assays were performed 48 h after transfection.
Data represent means ± SD (error bars) of at least three independent experiments.
Bars, 20 μm.
Via EMSA, we then analyzed endogenous NF-κB DNA binding activity in cells with reduced UXT expression.
Expectedly, TNF-α alone induced endogenous NF-κB to bind to its cognate probe strongly and specifically.
Considerably, this interaction was markedly diminished in nuclear extracts from UXT-specific knockdown cells.
In addition, this reduction was correlated with the effectiveness of the siRNA administered (Fig.
5 B).
In addition, this was also true for other components of the NF-κB transcriptional enhanceosome such as p50 and RNA polymerase II.
Understandably, deficiency of the endogenous UXT level had no effect on the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcriptional complex (Fig.
5 C).
Collectively, these finding indicate that UXT plays an important role in the NF-κB enhanceosome.
UXT maintains the presence of NF-κB inside the nucleus
A previous study has predicted UXT as an α-class PFD family protein via computer modeling (Gstaiger et al., 2003).
Some members of this family were previously found to assemble into a hexameric complex that functioned as a molecular chaperone in protein folding and stability (Vainberg et al., 1998).
Thus, we hypothesized that UXT performed its function by serving as a specific molecular chaperone for NF-κB inside the nucleus.
To explore this hypothesis, we investigated whether the nuclear presence of p65 was influenced by UXT.
Therefore, 293T cells were transfected with or without the indicated siRNAs and stimulated with or without TNF-α as shown.
Interestingly, the amount of nuclear p65 was markedly diminished in cells that were treated with siRNA against UXT even though these cells were stimulated with TNF-α.
Notably, siRNA 428 was better than siRNA 242 in causing the loss of nuclear p65.
In contrast, control siRNA displayed no such effect on the nuclear presence of p65.
The transcription factor sp1 as a negative control was expressed constitutively inside the nucleus and remained intact in response to UXT knockdown.
The cytoplasmic reservoir of p65 did not seem to be affected by siRNA against UXT (Fig.
5 D).
This suggested that UXT positively regulated p65 inside the nucleus, and the loss of it affected the sustained action of p65 in the nucleus, which led to the attenuated EMSA results as observed in Fig.
5 B.
Alternatively, we performed immunofluorescence analysis to support this speculation.
Interestingly, when transfecting cells with siRNA against UXT and stimulating them with TNF-α, there were an apparently decreased percentage of cells that displayed focused nuclear p65.
Approximately 90% of control cells displayed p65 inside the nucleus upon stimulation, whereas only 35–50% displayed p65 in the case of the UXT knockdown specimen (Fig.
5 E).
Collectively, these data indicate that UXT is a stabilizing factor for the NF-κB transcriptional enhanceosome.
One possible mechanism for the aforementioned phenomenon is that the loss of UXT releases NF-κB from the enhanceosome, which consequently causes NF-κB to be exported out into the cytoplasm.
To address this hypothesis, leptomycin B (LMB), a specific inhibitor of nuclear export (Ullman et al., 1997; Ghosh and Karin, 2002), was used.
Cells were transfected with siRNA against UXT and stimulated with TNF-α in the presence or absence of LMB.
Immunofluorescence analysis indicated that LMB considerably increased the nuclear accumulation of p65 even though the endogenous UXT was knocked down, which suggests that UXT contributed to the sustained presence of p65 inside the nucleus.
We have also tested other possibilities, but no obvious correlation could be drawn at present (see Discussion).
Our data formerly indicated that RHD of p65 was essential and sufficient to mediate the interaction between p65 and UXT (Fig.
2 B).
We wondered whether this stabilizing effect of UXT was applicable to any proteins containing this domain.
To address this speculation, we constructed two fusing proteins, p65 (1–312)-VP16 and Gal4 BD-p65 (285–551): the former one harbors the RHD, whereas the latter one spans all of p65 except RHD.
Luciferase assays were performed to monitor the effect of UXT knockdown on activations of indicated reporters induced by these chimeric proteins.
Interestingly, the knockdown of UXT inhibited the activity induced by p65 (1–312)-VP16 (Fig.
5 G) but not that induced by Gal4 BD-p65 (285–551) (Fig.
5 H).
Collectively, these data indicate that UXT directly modulates the nuclear function of NF-κB.
The siRNA-resistant form of UXT rescues the phenotype of its endogenous knockdown
To rule out possible off-target effects of the siRNA against UXT, a siRNA-resistant UXT (UXTr) was generated in which silent mutations were introduced into the sequence targeted by siRNA 428 and without changing the amino acid sequence of the proteins expressed.
The usefulness of UXTr was confirmed so that siRNA 428 no longer had any effects on UXTr but siRNA 242 could still interfere with UXTr expression (Fig.
6 A).
The exogenous wild-type UXT or UXTr was transfected together with siRNA 428 and stimulated with TNF-α as indicated, respectively.
Nuclear and cytoplasmic extracts were probed with indicated antibodies.
As shown in Fig.
6 B, the expression of UXTr rescued the loss of nuclear p65 caused by endogenous UXT impairment, whereas the expression of wild-type UXT failed to do so.
This rescuing effect was also confirmed with EMSA assays (Fig.
6 C).
Collectively, these data indicate that the loss of nuclear p65 was directly caused by impairment of UXT function and that there was a direct functional connection between p65 and UXT.
Rescue of the UXT knockdown effects with a siRNA-resistant form.
(A) The HA-tagged wild-type UXT and siRNA-resistant UXT (UXTr) were cotransfected with the indicated siRNAs.
After 24 h, cell lysates were subjected to Western blotting for determining exogenous UXT protein levels.
(B) 293T cells were transfected with the indicated exogenous UXT and siRNAs.
After 48 h, cells were induced by 10 ng/ml TNF-α for the indicated times.
Cytoplasmic and nuclear fractions were prepared and immunoblotted with the indicated antibodies, respectively.
(C) 293T cells were treated as in B.
EMSA was performed to test endogenous NF-κB or sp1 binding to their cognate probes.
UXT is dynamically recruited to the NF-κB enhanceosome upon stimulation in vivo
UXT forms a dynamic complex with NF-κB and is recruited to the NF-κB enhanceosome upon stimulation.
(A and B) 293T cells transfected with (A) or without (B) UXT.
(C) 293T cells were induced by 10 ng/ml TNF-α for 30 min, and nuclear extracts were prepared and incubated with the indicated antibodies to perform EMSA supershift assays.
(D) 293T cells were transfected with CARM1 siRNA.
48 h after transfection, the endogenous CARM1 mRNA was shown by RT-PCR.
(E) 293T cells were transfected with CARM1 or UXT siRNA and treated with 10 ng/ml TNF-α for 30 min or were left untreated.
Cover slides were subjected to immunofluoresence assay with anti-p65 antibody.
(F) 293T cells were transfected and treated as in E.
Nuclear lysates were used for EMSA to check endogenous NF-κB or sp1 DNA binding activities.
Bar, 20 μm.
Alternatively, we performed supershift assay to further substantiate this observation.
Stimulation of 293T cells with TNF-α led to a strong NF-κB DNA-binding band composed of κB probe, NF-κB, and its cofactors in EMSA.
Interestingly, this EMSA band was markedly diminished when antibody against UXT was introduced into the reaction mixture (Fig.
7 C), which strongly suggests that UXT is an integral component in this band and that it is of importance to foster an NF-κB conformation amenable to its binding.
Collectively, these data indicate that UXT forms a dynamic complex with p65 in vivo and is recruited to the NF-κB enhanceosome after stimulation.
The NF-κB enhanceosome consisted of a growing list of transcriptional cofactors.
Our current finding that UXT was one of them and that it was able to maintain the stability of the NF-κB enhanceosome led us to ask whether other factors could also accomplish this function.
Coactivator-associated arginine methyltransferase 1 (CARM1) is known to be a transcriptional cofactor in the NF-κB enhanceosome.
Cells with a reduced expression of CARM1 showed an impaired expression of NF-κB–dependent genes upon stimulation (Covic et al., 2005; Teferedegne et al., 2006).
However, immunofluorescence and EMSA experiments indicated that the knockdown of CARM1 did not affect the nuclear presence of p65 (Fig.
7, E and F), which suggested that UXT played a unique role toward NF-κB function.
Knockdown of UXT sensitizes 293T cells to apoptosis induced by TNF-α
It was well established that NF-κB promotes the survival of most cells through the transcriptional induction of antiapoptotic genes (Barkett and Gilmore, 1999; Zong et al., 1999).
Normally, 293T cells would not display apoptotic phenomenon in the presence of TNF-α (<3%).
However, loss of NF-κB function would make cells prone to apoptosis.
This was confirmed by knocking down UXT upon TNF-α stimulation (∼15%; unpublished data).
Cycloheximide (CHX) is a potent protein synthesis inhibitor.
However, at the concentration of ≤10 μg/ml, CHX only reduces but does not completely block de novo protein synthesis (Tang et al., 2001).
Interestingly, a low concentration of CHX could dramatically augment the apoptotic effects (Yeh et al., 1997; Kelliher et al., 1998).
Taking advantage of this cell model, we explored whether UXT was important for cell survival in response to TNF-α treatment.
As was shown in Fig.
8, TNF plus 5 μg/ml CHX resulted in ∼10% cell death after 18 h of treatment.
More importantly, when siRNA of UXT, which blocked NF-κB activity, was used in combination with TNF and CHX, we detected drastically morphological changes under light microscopy and ∼50% of cells undergoing apoptosis.
Alternatively, we used annexin V and TUNEL methods to quantitatively measure the percentages of cells undergoing apoptosis.
Consistently, a considerable increase of apoptotic cells was observed in samples with reduced UXT (Fig.
8, B and C).
These indicated that the knockdown of UXT sensitizes 293T cells to apoptosis induced by TNF-α.
Knockdown of UXT sensitizes 293T cells to TNF-α–induced apoptosis.
(A) 293T cells were transfected with the indicated siRNAs.
48 h after transfection, cells were treated with 50 ng/ml TNF-α plus 5 μg/ml CHX or were left untreated for 18 h.
Representative microscopic images are shown.
(B and C) Cells were prepared as in A.
After that, cells were analyzed using the annexin V (B) or TUNEL assay (C) to monitor cell apoptosis.
The percentages indicate the fractions of positive annexin V cells in total cells.
Data represent means ± SD (error bars) of at least three independent experiments.
Bar, 10 μm.
Endogenous UXT expression correlates with NF-κB activation in human prostate cancer cell lines
Prostate cancer began as an androgen-dependent tumor and progressed into an androgen-independent tumor.
In this progress, NF-κB activity started to behave out of control (Chen and Sawyers, 2002; Zerbini et al., 2003).
In androgen-independent PC-3 cells, more NF-κB localized inside the nucleus and constitutively displayed DNA binding activity, whereas it exhibited barely detectable activity in androgen-sensitive LNCaP cells (Palayoor et al., 1999).
We confirmed these observations as shown in Fig.
9 B, which indicate that these two cells are ideal models for studying NF-κB regulation within the nucleus.
UXT protein level correlates with the constitutive NF-κB activity in human prostate cancer cell lines.
(A) Equal amounts of whole cell lysates from LNcaP or PC-3 prostate cancer cells were immunoblotted with the indicated antibodies.
(B) 10 μg of nuclear extracts from LNCaP or PC-3 were subjected to EMSA with a radiolabeled κB or sp1 probe.
NS, nonspecific band.
(C) PC-3 cells were transfected with the indicated siRNAs.
48 h after transfection, cytoplasmic and nuclear fractions were prepared and immunoblotted with the indicated antibodies, respectively.
(D) PC-3 cells were cotransfected with the indicated siRNAs and 3×κB-Luc.
Luciferase assays were performed 48 h after transfection.
Data represent means ± SD (error) of at least three independent experiments.
(E) PC-3 cells were transfected with the indicated siRNAs.
48 h after transfection, nuclear extracts were prepared.
Endogenous NF-κB or sp1 DNA binding activities were examined by EMSA.
In addition, a former study indicated that the UXT mRNA level was considerably elevated in PC-3 as compared with LNCaP cells (Markus et al., 2002).
We confirmed this observation in terms of UXT mRNA (not depicted) and also found that this was true for the protein level of endogenous UXT (Fig.
9 A).
Given that UXT was recruited onto the NF-κB enhanceosome upon stimulation and that it is essential to maintain the presence of NF-κB inside the nucleus, these results suggest a possible correlation between UXT levels and NF-κB activation.
To further explore this possibility, we determined whether the inhibition of endogenous UXT expression would affect the constitutive amount of nuclear NF-κB in PC-3 cells.
As shown in Fig.
9 C, this interference indeed reduced the amount of nuclear NF-κB in this cell, which is consistent with that observed in 293T cells stimulated by TNF-α.
In addition, luciferase assay revealed that the decrease of UXT expression suppressed the constitutive transcriptional activity of NF-κB in PC-3 cells (Fig.
9 D).
Furthermore, EMSA confirmed that the loss of UXT also suppressed the constitutive DNA binding activity of NF-κB but not the control sp1 (Fig.
9 E).
These results indicate that the elevated expression of UXT strongly correlates with constitutive NF-κB activity in prostate cancer cell lines and again substantiates the notion that UXT is essential for NF-κB function in the nucleus.
Discussion
The NF-κB family of transcription factors is crucial for many key cellular processes.
Recently, the regulation of NF-κB activity has been cast in the limelight (Dixit and Mak, 2002; Ghosh and Karin, 2002).
A handful of transcriptional cofactors were implicated in this process.
For example, p300/CBP played a major role in the acetylation of p65 in vivo (Chen et al., 2002).
Conversely, acetylated p65 was subjected to deacetylation by HDAC3 (Chen et al., 2001).
Recent work also revealed that SIRT1 physically interacted with p65 and promoted p65 deacetylation (Yeung et al., 2004).
In addition, NF-κB was reported to be phosphorylated at multiple sites during activation.
For example, IKKα was demonstrated to accelerate both the turnover of NF-κB and its removal from proinflammatory gene promoters (Lawrence et al., 2005).
An emerging theme is that there are additional important layers of regulation for nuclear NF-κB.
Regulating the duration of the nuclear presence of p65 is another potential mechanism to modulate the NF-κB transcriptional response.
In this study, we characterized UXT as a novel and essential cofactor for NF-κB function in its enhanceosome.
Several lines of findings support this argument.
(1) UXT was shown to interact directly with p65 both in vitro and in vivo.
Importantly, this interaction was dependent on external stimuli.
Because UXT was almost exclusively present inside the nucleus, we reasoned that only after NF-κB translocation into the nucleus could this interaction take place, which was also substantiated by the observation that NF-κB was constitutively inside the nucleus in PC-3 cells and regulated by UXT as a result.
(2) ChIP and EMSA assays demonstrated that UXT localized inside the NF-κB enhanceosome in vivo and was recruited to it upon stimulation.
(3) The knockdown of UXT did not influence the molecular events of NF-κB activation outside the nucleus.
Instead, it decreased the amount of nuclear p65 and severely impaired NF-κB activation.
This decrease could be rescued by a siRNA-resistant exogenous UXT.
(4) Overexpression of UXT caused marginal synergic inductions of both A20 and IL-8 in response to TNF-α.
We reasoned that there was a sufficient amount of endogenous UXT within the nucleus in 293T cells so that it would be difficult to demonstrate the synergic effect more dramatically.
Convincingly, RNAi of UXT resulted in apparent attenuations of NF-κB–responsive reporter expression by various stimuli and, ultimately, the inducible expression of genes tightly regulated by NF-κB.
(5) The reduction of endogenous UXT tamed cells prone to apoptosis that was induced by TNF-α, which is a known index for the impairment of NF-κB function.
(6) PC-3 cells displayed constitutive NF-κB activity inside the nucleus.
This was nicely correlated with the elevated presence of both NF-κB and UXT within the nucleus, which was also consistently substantiated by the loss of NF-κB and its activity when endogenous UXT was diminished in PC-3 cells.
Thus, UXT might function to extend the duration of NF-κB or its enhanceosome inside the nucleus.
This function was probably achieved by fostering a favorable conformation for NF-κB in its enhanceosome.
UXT was recently predicted as a new member of the α-class PFD family protein.
Yeast and human PFDs 1–6 assembled into a hexameric complex, which functioned as a molecular chaperone in protein folding.
Our preliminary data also suggested that UXT forms oligomers in vivo (unpublished data).
Some members of the PFD family protein were implicated to participate in transcriptional regulation, such as PFDN5 (MM1) in c-myc transcription (Satou et al., 2001) and URI in the rapamycin-sensitive transcription response (Gstaiger et al., 2003).
However, the specific mechanisms of their action remain unknown.
We have generated several point mutations of UXT (C75A, L32P, L50P, L59P, and L32P/L50P/L59P) and did not reveal any substantial correlation between UXT and p65 interaction.
Our speculation is that a three-dimensional juxtaposed motif may be involved in this interaction.
Structural analysis of UXT and NF-κB interaction is under way in our laboratory, and hopefully this will shed light on how UXT performs this regulatory role.
Recently, studies suggested that the ubiquitin–proteasomal degradation pathway was also involved in stringent control of the promoter-bound p65 (Ryo et al., 2003; Saccani et al., 2004).
An NF-κB coactivator, Pin1, was found to bind p65 and prevent SOCS-1–mediated ubiquitination (Ryo et al., 2003).
We have tried to address whether the ubiquitin–proteasome pathway is critical for p65 stability in the context of UXT, but no conclusion can be made as of now.
The problem lies in the fact that most proteins are instantly degraded after ubiquitination, which is also the case for p65.
To prevent this obstacle, proteasome inhibitors were used to stabilize ubiquitinated proteins.
Unfortunately, a critical step during NF-κB activation involved proteasome function (i.e., IκBα degradation), which makes it unpractical to probe endogenous p65 degradation inside the nucleus.
The few papers that reported nuclear p65 ubiquitination often used the overexpression of p65 to answer this question, which was a controversial approach.
After this method, we did observe traces of p65 ubiquitination when UXT was knocked down (unpublished data), which was comparable with published reports (Ryo et al., 2003).
However, we believe it is too early to correlate the loss of UXT with p65 ubiquitination.
Substantially, we found that LMB, a specific inhibitor of nuclear export, increased the nuclear accumulation of p65 even though the endogenous UXT was knocked down.
More likely, UXT influenced p65 nucleocytoplasmic shuttling.
Given the findings from this study, we favor the notion that UXT is a nuclear chaperone that promotes formation of the NF-κB enhanceosome.
Nuclear chaperones usually mediate nucleosome assembly and remodeling (Philpott et al., 2000; Loyola and Almouzni, 2004).
Furthermore, they were implicated to directly regulate transcription factors.
For example, a nuclear chaperone termed FACT (facilitates chromatin transcription) could facilitate transcript elongation through nucleosomes (Orphanides et al., 1998).
Another human nuclear chaperone (bZIP-enhancing factor) promoted the transcriptional activity of bZIP (basic region–leucine zipper DNA-binding domain) proteins (Virbasius et al., 1999).
JDP2 was recently demonstrated to be a chaperone for AP1 (Jin et al., 2006).
In addition, nuclear chaperones can also negatively regulate transcription.
For example, nuclear chaperones p23 and Hsp90 bound to glucocorticoid-induced regulatory complexes and consequently disassembled these complexes (Freeman and Yamamoto, 2002).
In conclusion, nuclear chaperones have regulatory functions targeted to their specific client proteins other than the classic functions involved in protein synthesis and maturation.
More work needs to be done before the function and mechanism of UXT action are fully understood.
Numerous reports have documented considerable correlations between NF-κB activation and specific types of cancer (for review see Rayet and Gelinas, 1999).
By promoting proliferation and inhibiting apoptosis, NF-κB could tip the balance between proliferation and apoptosis toward malignant behavior in tumor cells (for review see Rayet and Gelinas, 1999).
Prostate cancer begins as an androgen-dependent tumor and progresses into an androgen-independent tumor.
In this progress, NF-κB activity is up-regulated (Chen and Sawyers, 2002; Zerbini et al., 2003).
In this study, we provide a new thread of explanation for this correlation and demonstrate that UXT is essential for the constitutive activity of NF-κB in prostate cancer cells.
Collectively, our study reveals that UXT is an integral component of the NF-κB enhanceosome and is essential for its function in the nucleus.
UXT may function as a specific molecular chaperone for NF-κB in this process.
Future investigations will focus on how UXT interacts with NF-κB and regulates the dynamic processes of NF-κB within the nucleus.
Materials and methods
Reagents
Monoclonal UXT antibodies 6D3 and 105.128 were provided by M.I.
Greene (University of Pennsylvania, Philadelphia, PA) and W.
Krek (Eidgenössische Technische Hochschule Honggerberg, Zurich, Switzerland), respectively.
UXT siRNA duplexes were chemically synthesized by GenePharma.
The UXT siRNA sequences were as follows: 242, AGCACUCGGAGUUAUAUAUdTdT; 428, CCAAGGACUCCAUGAAUAUdTdT; and 428M1, CCAACCCCUCCAUGAAUAUdTdT.
The control siRNA sequence was UUCUCCGAACGUGUCACGUdTdT, and the CARM1 siRNA sequence was GCAGUCCUUCAUCAUCACCdTdT.
Other commercially available reagents and antibodies used were as follows: HA, p65, p65, p50, IκBα, and RNA polymerase II antibodies were purchased from Santa Cruz Biotechnology, Inc.
Flag, sp1, and β-actin antibodies were obtained from Sigma-Aldrich.
Anti-myc was purchased from Wolwobiotech.
rhTNFα and RhIL-1β was purchased from R&D Systems, and lipopolysaccharide was obtained from Sigma-Aldrich.
Plasmids
UXT and its deletion mutants were constructed by PCR from the human thymus library and subsequently cloned into mammalian expression vectors as indicated.
The UXT siRNA-resistant form was generated by introducing four silent mutations (373 ACAAAAGATAGC 384) in the siRNA 428 target sequence.
Flag-IκBαSR, Flag-TRAF6, and the reporter genes (3×κB-luc and pRL-SV40) have been described previously (Diao et al., 2005).
pcDNA3-HA-p65 was a gift from G.
Pei (Shanghai Institute of Biochemistry and Cell Biology [SIBCB], Shanghai, China), and its deletion mutants were constructed by PCR.
MyD88 was amplified from the human thymus library and subcloned into pcDNA3.1-Flag vector.
HA–lymphoid enhancer binding factor 1 was provided by L.
Li (SIBCB, Shanghai, China), HA-p50 and mp40-luc were provided by B.
Sun (SIBCB, Shanghai, China), myc-cRel was provided by B.
Huang (Northeast Normal University, Changchun, China), and A20 was provided by L.
Guo (SIBCB, Shanghai, China).
Yeast two-hybrid screening
A cDNA fragment encoding residues 1–312 of human p65 was inserted in frame into the Gal4 DNA-binding domain vector pGBKT7.
A human thymus cDNA library (CLONTECH Laboratories, Inc.) was screened according to protocols recommended by the manufacturer.
Cell culture and transfection
293T and RAW264.7 cells were cultured in DME (Invitrogen) supplemented with 10% FBS (Hyclone).
PC-3 cells were cultured in RPMI 1640 medium (Invitrogen) supplemented with 10% FBS, and LNCaP cells were cultured in Ham's F12 medium (Invitrogen) supplemented with 10% FBS.
All transfections were performed using LipofectAMINE 2000 (Invitrogen) according to the manufacturer's instructions.
Reporter assays
Cells were seeded in 24-well plates and transfected with 40 pmol siRNA combined with reporters and other constructs as indicated.
The total amount of DNA was kept constant by supplementing with pcDNA3.
pRL-SV40 (Promega) was cotransfected to normalize transfection efficiency.
48 h after transfection, cells were treated with the indicated reagents or left untreated.
Luciferase activity was analyzed with the Dual Luciferase Reporter Assay System (Promega).
In vitro translation
UXT and p65 were in vitro translated and labeled with [35S]methionine using the TNT Quick Coupled Transcription/Translation System (Promega) according to the manufacturer's instructions.
Real-time RT-PCR
Total RNA was isolated with TRIzol (Invitrogen) according to the manufacturer's instructions.
Reverse transcription of purified RNA was performed using oligonucleotide dT primer.
The quantification of gene transcripts was performed by real-time PCR using SYBR green I dye (Invitrogen).
Expression values were normalized with control β-actin.
The primers used are listed as follows: IL-8, sense (AGGTGCAGTTTTGCCAAGGA) and antisense (TTTCTGTGTTGGCGCAGTGT); IκBα, sense (CTGAGCTCCGAGACTTTCGAGG) and antisense (CACGTGTGGCCATTGTAGTTGG); A20, sense (GCGTTCAGGACACAGACTTG) and antisense (GCAAAGCCCCGTTTCAACAA); UXT, sense (TTTGGGCTGTAACTTCTTCGT) and antisense (ATATTCATGGAGTCCTTGGTG); CARM1, sense (TGCCGACCGCCTATGACT) and antisense (CCCGTGTTGGCTAAAGGAA); β-actin, sense (AAAGACCTGTACGCCAACAC) and antisense (GTCATACTCCTGCTTGCTGAT).
EMSA
EMSAs were performed as described previously (Yang et al., 2006).
The probes used are as follows: wild-type κB probe (AGTTGAGGGGACTTTCCCAGGC), mutant κB probe (AGTTGAGGCGACTTTCCCAGGC), and sp1 probe (ATTCGATCGGGGCGGGGCGAGC).
ChIP assay
The ChIP assay kit (Upstate Biotechnology) was used according to the manufacturer's instructions with some variations.
Formaldehyde cross-linking was performed at room temperature for 10 min before glycine was added to a final concentration of 125 mM for 5 min.
The cells were rapidly collected and lysed in SDS lysis buffer.
Suspended chromatin was sheared by sonication to a mean size of 200–1,000 bp, centrifuged to pellet debris, and diluted 10 times with dilution buffer.
Extracts were precleared for 2 h with salmon sperm DNA and BSA-saturated protein A/G beads.
Immunoprecipitations were performed at 4°C overnight using antibodies as indicated with IgG as a negative control.
Immune complexes were collected and washed sequentially with Tris-SDS-EDTA buffer I, II, and III followed by two washes with Tris-EDTA buffer.
Immune complexes were then extracted with elution buffer and DNA: protein complexes were disrupted by heating at 65°C overnight.
After proteinase K digestion for 1 h, DNA was extracted with phenolchloroform and precipitated in ethanol.
About one twentieth of precipitated DNA was used as template in each PCR reaction.
The following promoter-specific primers were used: human A20, sense (CAGCCCGACCCAGAGAGTCAC) and antisense (CGGGCTCCAAGCTCGCTT); human GAPDH, sense (AGCTCAGGCCTCAAGACCTT) and antisense (AAGAAGATGCGGCTGACTGT); and human IκBα, sense (TAGTGGCTCATCGCAGGGAG) and antisense (TCAGGCTCGGGGAATTTCC).
Immunofluorescence and microscopy
Cells grown on coverslips were fixed with 4% PFA, permeabilized in 0.1% Triton X-100, blocked by 1% BSA, and stained with the indicated primary antibodies followed by FITC-conjugated anti–mouse IgG (Jackson ImmunoResearch Laboratories).
Nuclei were counterstained with DAPI (Sigma-Aldrich).
Slides were mounted by Aqua-Poly/Mount (Polysciences).
Images were captured at room temperature using a confocal microscope (TCS SP2 ACBS; Leica) with a 63× NA 1.4 oil objective (Leica) except those in Fig.
2 C, which were captured using a camera (DP70; Olympus) on a microscope (BX51; Olympus) with a 10× NA 0.3 objective.
The acquiring software was TCS (Leica) or DPcontrol (Olympus).
Apoptosis assay
293T cells were transfected with the indicated siRNA.
48 h after transfection, cells were treated with 50 ng/ml TNF-α and 5 μg/ml CHX or left untreated for 18 h.
Floating and adherent cells were collected and analyzed using the annexin V–phycoerythrin Apoptosis Detection kit I (BD Biosciences) and In Situ Cell Death Detection kit (TUNEL; Roche) according to the manufacturer's instructions.
The flow cytometer used was a FACSCalibur (BD Biosciences).



















































Arabidopsis COP1 shapes the temporal pattern of CO accumulation conferring a photoperiodic flowering response
The transcriptional regulator CONSTANS (CO) promotes flowering of Arabidopsis under long summer days (LDs) but not under short winter days (SDs).
Post-translational regulation of CO is crucial for this response by stabilizing the protein at the end of a LD, whereas promoting its degradation throughout the night under LD and SD.
We show that mutations in CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1), a component of a ubiquitin ligase, cause extreme early flowering under SDs, and that this is largely dependent on CO activity.
Furthermore, transcription of the CO target gene FT is increased in cop1 mutants and decreased in plants overexpressing COP1 in phloem companion cells.
COP1 and CO interact in vivo and in vitro through the C-terminal region of CO.
COP1 promotes CO degradation mainly in the dark, so that in cop1 mutants CO protein but not CO mRNA abundance is dramatically increased during the night.
However, in the morning CO degradation occurs independently of COP1 by a phytochrome B-dependent mechanism.
Thus, COP1 contributes to day length perception by reducing the abundance of CO during the night and thereby delaying flowering under SDs.
Introduction
Exposure to light influences many aspects of the plant life cycle, a process referred to as photomorphogenesis.
Light promotes seed germination and seedling growth, thereby ensuring that young plants are exposed to an optimal environment for photosynthesis.
Photomorphogenesis also has important functions in the development of adult plants (Neff et al, 2000).
The mechanisms controlling adult photomorphogenic traits such as control of flowering in response to day length are less well understood than those that occur in the seedling.
However, a genetic pathway that promotes flowering of Arabidopsis in response to long days (LDs) has been defined (Searle and Coupland, 2004; Imaizumi and Kay, 2006).
Within this pathway, the transcriptional regulator CONSTANS (CO) has an important function by promoting flowering specifically under LDs.
Here, we demonstrate that the ubiquitin ligase CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1), a major regulator of seedling photomorphogenesis (Deng et al, 1992), negatively regulates CO protein abundance in the vascular tissue of adult plants as part of the mechanism by which Arabidopsis discriminates between LD and SD during flowering-time control.
CO is a major regulator of photoperiodic flowering.
Mutations in CO delay flowering specifically under LD, whereas its overexpression from a viral promoter causes extreme early flowering under LD and SD.
CO contains two B-box-type zinc-finger motifs near its N terminus and a CCT (CONSTANS, CONSTANS-LIKE, TOC1) domain at its C terminus (Putterill et al, 1995).
The latter domain is plant specific, but shows similarity to the DNA-binding domain of the HAP2 subunit of the CCAAT box-binding complex, suggesting that CO might bind to DNA directly (Wenkel et al, 2006).
The closely related genes FLOWERING LOCUS T (FT) and TWIN SISTER OF FT (TSF) are highly and rapidly increased in expression in response to CO expression (Samach et al, 2000; Wigge et al, 2005; Yamaguchi et al, 2005).
These genes encode RAF kinase inhibitor-like proteins that exert an effect as potent inducers of flowering (Kardailsky et al, 1999; Kobayashi et al, 1999).
CO activates FT in the companion cells of the phloem within the vascular tissue, and FT protein is then proposed to move through the phloem sieve elements to the shoot apical meristem (An et al, 2004; Corbesier et al, 2007; Jaeger and Wigge, 2007; Mathieu et al, 2007), where it changes gene expression patterns and induces flowering (Abe et al, 2005; Wigge et al, 2005; Searle et al, 2006).
The mechanism by which CO activity is controlled by day length involves both transcriptional and post-translational regulation.
CO transcription is regulated by the circadian clock so that its expression rises around 12 h after dawn and stays high until the following dawn (Suarez-Lopez et al, 2001).
Exposure to light between 10 and 14 h after dawn further promotes CO transcription through the activity of the photoreceptor FLAVIN-BINDING, KELCH REPEAT, F-BOX 1 (FKF1) and its interacting partner GIGANTEA (GI) (Suarez-Lopez et al, 2001; Imaizumi et al, 2003; Sawa et al, 2007).
At the post-translational level, CO protein is stabilized when plants are exposed to light, whereas in darkness CO protein is rapidly degraded through ubiquitination and the activity of the proteasome.
These mechanisms combine to ensure that a peak in CO protein abundance occurs under LDs when plants are exposed to light between 10 and 16 h after dawn, whereas under SD, when plants are exposed to darkness during this interval, CO protein does not accumulate (Valverde et al, 2004).
The importance of ubiquitination and degradation of CO protein by the proteasome in these processes was demonstrated by use of proteasome inhibitors.
These regulatory steps ensure that transcription of FT and TSF occurs under LDs but not under SDs.
The photoreceptors required for post-translational regulation of CO have been characterized.
Mutations in the genes encoding the photoreceptors phytochrome A (phyA) and cryptochrome 2 (cry2) delay flowering, and these mutations also reduce the accumulation of CO protein (Valverde et al, 2004).
Similarly, far-red light or blue light promotes flowering and stabilizes CO protein, and these regions of the spectrum activate phyA and cry2, respectively.
In contrast, red light delays flowering and reduces the accumulation of CO protein.
This response appears to be mainly controlled by phytochrome B (phyB), because phyB mutations cause early flowering and allow increased accumulation of CO protein.
COP1 is a major negative regulator of photomorphogenic responses, so that cop1 mutants undergo photomorphogenesis in darkness in the absence of photoreceptor activation (Deng et al, 1991).
COP1 encodes a RING finger protein with a coiled-coil motif and WD40 repeats (Deng et al, 1992), and exerts an effect as a ubiquitin ligase that promotes the degradation of transcription factors implicated in seedling photomorphogenesis (Osterlund et al, 2000).
In mammalian and plant cells, COP1 seems to exert an effect as part of a complex that also includes DEETIOLATED 1 (DET1), DAMAGED DNA-BINDING PROTEIN 1 (DDB1), cullin 4A and RING BOX 1 (RBX1) (Chory et al, 1989; Schroeder et al, 2002; Wertz et al, 2004; Hoecker, 2005; Chen et al, 2006).
However, the SUPPRESSOR OF PHYTOCHROME A-105 1 (SPA) family of proteins is plant specific, related in sequence to COP1 and modulates the ubiquitin ligase activity of COP1.
SPA proteins contain a coiled-coil domain and WD40 repeats related to those of COP1 as well as a kinase-like domain not present in COP1 (Hoecker et al, 1999).
Quadruple mutants in which the four SPA genes are mutated exhibit a phenotype similar to that of cop1 mutants (Laubinger et al, 2004).
Furthermore, SPA1 and COP1 physically interact and SPA1 modulates COP1 activity in vitro (Hoecker and Quail, 2001; Saijo et al, 2003; Seo et al, 2003).
Several protein targets for COP1 are transcription factors that regulate seedling photomorphogenesis (Osterlund et al, 2000; Seo et al, 2003; Duek et al, 2004; Jang et al, 2005; Yang et al, 2005).
Each of these proteins was shown based on mutagenesis studies to have a function in the regulation of seedling growth in response to light.
COP1 targets each of these proteins for degradation in the dark, but in the light COP1 activity is suppressed allowing these transcription factors to accumulate and promote seedling photomorphogenesis.
In addition to these roles in seedling development, COP1 also influences photomorphogenesis of adult plants.
Although null mutant alleles of COP1 cause seedling lethality, plants homozygous for weaker cop1 alleles are viable.
These plants are early flowering, particularly under SDs, indicating that COP1 is required for the suppression of flowering (McNellis et al, 1994).
Furthermore cop1 mutants, but not wild-type (WT) plants, flower in darkness if provided with sugar (Nakagawa and Komeda, 2004).
In addition, spa1 mutants flower early and SPA proteins modulate CO abundance so that in spa1 spa3 spa4 triple mutants 16 h after dawn under LDs increased levels of CO protein were detected (Ishikawa et al, 2006; Laubinger et al, 2006).
Here, we analysed the role of COP1 in the light regulation of flowering time by genetic and molecular studies.
We show that COP1 represses CO activity in the vascular tissue, and reduces CO protein levels particularly under SDs and in the dark, thereby facilitating a flowering response to day length.
Results
Genetic and spatial interactions between COP1 and CO in the regulation of flowering time
Previously cop1 mutants were shown to flower earlier than WT plants under short days (SDs) and at a similar time to WT plants under LDs (Mcnellis et al, 1994).
Under our conditions, cop1–4 mutants flowered dramatically earlier than WT plants under SDs, as shown previously, but in addition flowered earlier than WT plants under LDs.
The cop1–4 mutant produced around 53 leaves fewer than WT plants before flowering under SDs, whereas under LDs the difference between mutant and WT was around 5 leaves (Figure 1A–C; Supplementary Table 1).
Therefore, the photoperiod response of cop1–4 mutants was severely reduced so that they flowered after forming only 3 leaves more under SDs than LDs, whereas WT plants formed around 45 leaves more under SDs.
In WT plants, CO promotes early flowering under LDs but not SDs.
To test whether the early flowering of cop1–4 mutants under SDs was caused by activation of CO under these conditions, the cop1–4 co-10 double mutant was constructed and its flowering time was measured.
The double mutant flowered after forming around 30 leaves more than cop1–4 mutants under SDs, demonstrating that CO has an important role in the early flowering of cop1–4 mutants under SDs (Figure 1B and C).
Nevertheless, cop1–4 co-10 plants formed 20 leaves fewer than co-10 plants under these conditions, indicating that part of the early flowering of the cop1–4 mutant occurs independently of CO.
Under LDs, cop1–4 co-10 plants also flowered at a time intermediate between co-10 and cop1–4 (Figure 1A and C).
These genetic results suggest that COP1 exerts an effect as a negative regulator of CO under SDs, so that CO promotes flowering of cop1–4 mutants but not WT plants under SDs.
CO is expressed only in the vascular tissue and exerts an effect in the phloem companion cells to activate the transcription of the flowering-time gene FT (Takada and Goto, 2003; An et al, 2004).
To test whether COP1 also regulates flowering from the phloem, COP1 or HA:COP1 was expressed from the SUC2 promoter, which is active specifically in the phloem companion cells (Imlau et al, 1999).
The SUC2:COP1 transgene was introduced into WT Columbia plants and into cop1–4 mutants, whereas SUC2:HA:COP1 was introduced into SUC2:CO plants.
SUC2:COP1 delayed flowering of cop1–4 mutants under LDs and SDs, and of WT plants under LDs (Figure 1D–G).
Therefore, COP1 exerts an effect in the companion cells, where CO is expressed, to delay flowering.
However, SUC2:COP1 cop1–4 plants still flower earlier than WT plants under SDs, suggesting that COP1 expression in companion cells is not sufficient to completely rescue the early-flowering phenotype of cop1–4 mutants, and therefore that COP1 probably also exerts an effect in additional cell types to delay flowering.
The observation that SUC2:HA:COP1 delays flowering of SUC2:CO plants under LDs and SDs supports the idea that the delay of flowering associated with SUC2:COP1 is at least in part caused by reduction of CO activity (Figure 1G).
Taken together, the flowering-time phenotypes of plants misexpressing COP1 in the phloem are consistent with the idea that COP1 exerts an effect in the phloem companion cells to repress the promotion of flowering by CO.
COP1 reduces FT mRNA levels
FT transcription is activated by CO and likely represents a direct target of CO protein (Samach et al, 2000; Wigge et al, 2005).
Therefore, if COP1 exerts an effect to repress CO activity this should be reflected in reduced FT mRNA levels.
In WT plants grown under SDs, CO mRNA is present during the night but FT mRNA is not expressed, because CO protein is rapidly degraded in the dark (Suarez-Lopez et al, 2001; Valverde et al, 2004).
The effects of COP1 on CO transcription and CO activity were tested by analysing CO and FT mRNA levels at 4-h intervals for 24 h in SD-grown plants of different genotypes (Figure 2A and B).
In WT Columbia, CO mRNA was detected during the night under SDs, but was absent in co-10 and cop1–4 co-10 plants as expected due to the T-DNA insertion present in the CO gene in the co-10 allele (Materials and methods).
In cop1–4 mutants, the CO mRNA pattern is similar to that observed in WT plants but rises earlier, appearing weakly 8 h after dawn, whereas in WT plants CO mRNA was first detected 12 h after dawn.
In contrast, FT mRNA was detected in cop1–4 mutants but not WT plants, consistent with the early flowering of these mutants under SDs.
Similarly, under LDs, FT mRNA levels were much higher in cop1–4 plants than in WT plants consistent with the earlier flowering of the mutants under these conditions (Supplementary Figure 1).
However, under LDs, CO mRNA was consistently detected at lower levels in cop1–4 mutants than in WT plants (Supplementary Figure 1).
The expression of FT mRNA in cop1–4 mutants requires CO activity, as demonstrated by the absence of FT mRNA in cop1–4 co-10 plants (Figure 2A).
These results are consistent with the idea that COP1 delays flowering of WT plants under SDs, and to a lesser extent under LDs, by repressing CO activity and thereby preventing FT expression.
The abundance of FT mRNA was also tested in transgenic plants expressing COP1 or HA:COP1 mRNAs at high levels in the phloem companion cells from the SUC2 promoter (Figure 2C–E).
CO and FT mRNA levels were compared through a LD time course in SUC2:COP1 and WT plants.
CO mRNA levels were very similar in both genotypes at all times, whereas FT mRNA levels were severely reduced in SUC2:COP1 plants (Figure 2E), consistent with the overexpression of COP1 in phloem companion cells leading to a reduction in CO activity at the post-transcriptional level.
Similarly, 16 h after dawn under SDs, when FT mRNA reaches peak levels in cop1–4 mutants (Figure 2A), SUC2:COP1 cop1–4 plants displayed severely reduced levels of FT mRNA (Figure 2C).
Finally, in SUC2:HA:COP1 SUC2:CO plants the level of FT mRNA was lower than in the SUC2:CO progenitor plants, but the level of CO mRNA was unaffected (Figure 2D).
Therefore, analysis of FT mRNA in these transgenic plants supports the conclusion that COP1 delays flowering by repressing at the post-transcriptional level the capacity of CO to promote FT transcription in the phloem companion cells.
COP1 and CO physically interact in vitro and in vivo
The observation that the ubiquitin ligase COP1 represses CO-mediated activation of FT suggested that CO might be a substrate for COP1.
To test this hypothesis, we first investigated whether COP1 was able to physically interact with CO.
In the yeast two-hybrid system, we detected no interaction between CO and COP1, although an interaction between COP1 and the CO-related protein CO-LIKE3 (COL3) was previously detected by this method (Datta et al, 2006), and we were able to confirm this interaction.
Therefore, whether COP1 and CO interact in vitro was tested using a co-immunoprecipitation assay (Figure 3).
COP1 attached to the GAL4 activation domain (GAD:COP1) and CO were made in an in vitro transcription/translation system and combined.
GAD:COP1 was precipitated with anti-GAD antibody and CO was co-precipitated with GAD:COP1 (Figure 3).
In contrast, CO was not co-immunoprecipitated with GAD alone.
These experiments suggest that CO interacts with COP1 in the GAD:COP1 fusion protein.
Fragments of CO were also combined with GAD:COP1 to determine which regions of CO are required for the interaction with COP1.
Two segments of CO were tested: one containing the region between amino acids 107 and 373, which was called COΔB-box because it did not contain the zinc-finger B-boxes found at the N terminus of CO, and a second containing the region between amino acids 1 and 331, which was named COΔCCT, because the CCT domain near the C terminus of CO was removed.
In vitro precipitation experiments demonstrated that COΔB-box was co-immunoprecipitated with GAD:COP1, whereas COΔCCT was not.
Therefore, the N-terminal region containing the B-boxes is not required for interaction with COP1, suggesting that the interaction with COP1 is mediated by the C-terminal region of CO that contains the CCT domain.
Whether the interaction between CO and COP1 also occurred in vivo in plant cells was tested using fluorescent resonance energy transfer (FRET).
Microprojectile bombardment was used to co-express cyan fluorescent protein (CFP):COP1 and yellow fluorescent protein (YFP):CO in leaf epidermal cells of Arabidopsis.
CFP:COP1 and YFP:CO colocalized to the nucleus and also colocalized in speckles within the nucleus (Figure 4A and B).
Physical interaction of CFP:COP1 and YFP:CO was tested by measuring FRET using photoacceptor bleaching, as previously described (Wenkel et al, 2006) (Figure 4C and D).
Quantification of FRET signals demonstrated that FRET occurred between YFP:CO and CFP:COP1 both in the nucleus and specifically in nuclear speckles (Figure 4C and D).
In control experiments using YFP and CFP, YFP:CO and CFP or YFP and CFP:COP1 FRET was detected at significantly lower levels (Figure 4C).
These experiments demonstrate that YFP:CO and CFP:COP1 colocalize and physically interact in the nuclei of plant cells.
COP1 and phyB have complementary roles in repressing CO protein levels under LDs and SDs
The genetic and molecular experiments described earlier supported the hypothesis that COP1 negatively regulates CO activity at the post-transcriptional level.
Therefore, we tested the effect of COP1 on CO protein levels.
First, CO protein abundance was examined in nuclei of WT Columbia, co-10, transgenic 35S:CO and cop1–4 plants harvested 16 h after dawn under LDs, when CO protein is expected to be at highest abundance (Valverde et al, 2004) (Figure 5A).
As shown previously, CO was clearly detectable in 35S:CO transgenic plants that overexpress the protein, but was below the level of detection in nuclei of WT plants.
However, in cop1–4 mutants, CO was clearly detected, suggesting that in WT plants COP1 has a major function in reducing CO protein levels at this time.
Strong support that the protein detected in cop1–4 mutants was indeed CO protein came from the analysis of cop1–4 co-10 double mutants, in which the protein detected in cop1–4 mutants was no longer present (Figure 5A).
CO mRNA shows a diurnal rhythm in abundance in WT plants and in cop1–4 mutants, therefore the diurnal pattern of CO protein abundance was tested under LDs and SDs in cop1–4 mutants (Figure 5B and C).
Under SDs of 8 h light, cop1–4 mutants flower dramatically earlier than WT plants (Figure 1) and CO protein was present for most but not all of the diurnal cycle (Figure 5C).
CO was strongly detected soon after dawn, was absent or present at much lower abundance 4 and 8 h after dawn, and then was present strongly for the remainder of the night from 10 to 24 h after dawn.
The appearance of CO protein from 10 h after dawn is likely due to an increase in CO mRNA levels, as the abundance of CO mRNA increased steeply between 4 and 14 h after dawn in the same plants used for the protein analysis (Supplementary Figure 2).
In contrast, CO mRNA abundance fell between 14 and 24 h after dawn, whereas CO protein levels were high throughout this time.
This comparison suggests that impairing COP1 function causes CO protein to be relatively stable in the dark.
However, the steep decline in CO protein abundance between 0.5 and 4 h after dawn suggests that a second post-translational mechanism, independent of COP1, might negatively regulate CO protein levels in the morning.
Under LDs of 16-h photoperiods, CO protein was detected from dawn until 4 h into the photoperiod, was undetectable 6 h after dawn and then was present for the remainder of the photoperiod and throughout the night (Figure 5B).
This pattern was similar to that detected under SDs, but the protein was detectable for longer and was only absent at one time point, 6 h after dawn.
The broader peak in CO protein under LDs is likely due to CO mRNA being expressed for longer under LDs, as previously described (Suarez-Lopez et al, 2001; Imaizumi et al, 2003).
The photoreceptor phyB was previously shown to promote the degradation of CO protein early in the day in 35S:CO plants, and this was proposed to contribute to the inhibitory effect of phyB on flowering time (Valverde et al, 2004).
To test whether phyB is responsible for the reduction in CO protein levels early in the day in cop1 mutants, the phyB-9 cop1–6 double mutant was tested for flowering time and CO protein levels.
Under SDs, phyB-9 cop1–6 plants flowered at a very similar time to cop1–6 mutants, demonstrating that the early flowering of cop1–6 mutants is not enhanced by loss of function of phyB (Supplementary Figure 3).
Under LDs, the double mutant flowered significantly later than either single mutant, which indicates a complexity in the interaction between COP1 and phyB under these conditions that cannot be simply explained by regulation of CO protein levels (see Discussion).
To test whether phyB is responsible for the reduction in CO protein early in the day in cop1 mutants, protein was extracted from phyB-9 cop1–6 and cop1–6 plants 6 and 16 h after dawn under SDs.
In the cop1–6 plants, CO protein was undetectable 6 h after dawn, as observed for cop1–4 mutants, but in phyB-9 cop1–6 plants CO protein accumulated strongly 6 h after dawn (Figure 5D).
In contrast, CO mRNA was present at similar levels in cop1–6 and phyB-9 cop1–6 plants 6 h after dawn (Supplementary Figure 3).
These results indicate that phyB is required for post-transcriptional regulation of CO expression early in the day and independently of COP1.
However, in the samples harvested 16 h after dawn CO protein was present at similar levels in cop1–6 and phyB-9 cop1–6 plants, indicating that phyB does not influence CO protein levels at that time of day (Supplementary Figure 3).
SPA1 interacts with COP1 and is implicated in the degradation of some COP1 substrates.
Recently, CO protein was shown to be more abundant in spa1 spa3 spa4 triple mutants 16 h after dawn under LDs (Laubinger et al, 2006).
The diurnal pattern of CO protein abundance in spa1–7 mutant plants was tested to compare with that described for cop1–4 mutants (Figure 5C and E).
A similar pattern of CO protein accumulation was observed in spa1–7 and cop1–4 mutants between 14 h after dawn and the following morning, but the rise in the abundance of the protein was delayed in the spa1–7 mutant, so that it could not be detected until 14 h after dawn.
In contrast, in the cop1–4 mutant CO protein was strongly detected 10 h after dawn.
These results suggest a functional relationship between COP1 and SPA1 proteins in the degradation of CO, and that of the four SPA proteins SPA1 has the major role in regulating CO levels.
The delayed increase in CO abundance in spa1–7 compared with cop1–4 mutants might be due to the activity of other SPA proteins.
Degradation of CO protein in red light is not impaired by the cop1–4 mutation
Arabidopsis plants flower later under red light and previously this was proposed to be at least partly due to degradation of CO protein under these conditions (Valverde et al, 2004).
To test whether red light-mediated degradation of CO protein is also impaired in the cop1–4 mutants, CO protein abundance was compared in 35S:CO and cop1–4 plants grown under white and red light (Figure 6).
Similar levels of CO protein were detected in both lines grown under 16 h of white light (Figure 6).
Furthermore, when both genotypes were exposed to 16 h of red light, CO protein abundance fell sharply in both 35S:CO plants and cop1–4 mutants (Figure 6).
The reduced levels of CO protein observed in cop1–4 mutants under red light compared with white light are not due to lower levels of CO mRNA, which were identical under both conditions (Figure 6).
This result demonstrates that a COP1-independent mechanism is required for CO protein degradation under red light.
Discussion
We demonstrated that COP1 ubiquitin ligase is required to shape the diurnal pattern of CO protein accumulation as part of the flowering response of Arabidopsis to photoperiod.
COP1 delays flowering of WT plants under SDs by preventing CO protein accumulation during the night and thereby ensuring that FT transcriptional activation does not occur.
Under LDs WT plants flower early, but even under these conditions COP1 modulates CO protein levels, lowering the abundance of the protein towards the end of the day and during the night.
These effects on CO are consistent with the extreme early flowering of cop1–4 mutants under LDs and SDs, the largely day length-insensitive phenotype of cop1–4 mutants and suppression of these phenotypes to a large extent by co mutations.
However, in cop1–4 mutants CO protein abundance is still reduced in the early morning and in red light, indicating that a second mechanism independent of COP1 regulates CO protein abundance under these conditions.
The mechanism that exerts an effect in the early morning is shown to depend on the phyB photoreceptor.
Our observations place COP1 within the regulatory network for photoperiod perception and regulation of flowering in Arabidopsis (Figure 7), and extend the characterized functions of COP1 beyond those previously described in seedling photomorphogenesis.
COP1 reduces CO protein abundance to confer a photoperiodic flowering response
COP1 represses seedling photomorphogenesis by catalysing the ubiquitination and therefore degradation of proteins that promote seedling photomorphogenesis (Osterlund et al, 2000; Seo et al, 2003; Duek et al, 2004; Jang et al, 2005; Yang et al, 2005).
In addition to their effects on seedling development, cop1 mutations severely impair the development of adult plants, although the mechanisms by which this occurs are less well understood.
Altered adult traits include photoperiodic flowering so that cop1 mutants flower at similar times under LDs and SDs (Mcnellis et al, 1994).
In addition, cop1 mutants flower in constant darkness in the presence of sucrose, whereas WT seedlings do not (Nakagawa and Komeda, 2004).
Under these conditions, cop1 mutants exhibit higher expression of FT and SOC1 mRNAs than WT plants.
We showed that the early flowering of cop1–4 mutants under LDs or SDs largely depends on CO function and that in these mutants CO protein persists in the dark under SDs and LDs.
These results suggest that post-translational regulation of CO is impaired in cop1–4 mutants.
Under LDs, CO protein levels are high in cop1–4 mutants throughout most of the day and rise earlier after dawn than under SDs, as shown for CO mRNA (Suarez-Lopez et al, 2001; Imaizumi et al, 2003).
This effect on CO mRNA levels is at least in part due to FKF1- and GI-mediated degradation of the transcriptional repressor CYCLING DOF FACTOR 1 (CDF1), which is triggered by light, allowing CO mRNA abundance to rise earlier during the day under LDs (Imaizumi et al, 2005; Sawa et al, 2007).
Perhaps surprisingly, our results indicate that under LDs, when CO promotes flowering of WT plants, COP1 also has a strong negative effect on CO protein levels.
At the end of the day and during the night, CO protein accumulates to much higher levels in cop1–4 mutants than WT plants, although CO mRNA abundance is actually lower than in WT plants.
These results indicate that under LDs COP1 has an unexplained role in increasing CO mRNA abundance, and a major function in lowering CO activity by reducing CO protein abundance.
Under SDs, CO mRNA is expressed and rises during the night.
Our demonstration of COP1-mediated degradation of CO protein in the dark under SDs provides a molecular explanation for why CO does not promote flowering under these conditions, and is consistent with previous demonstrations that application of proteasome inhibitors led to stabilization of the CO protein.
The importance of this process in conferring a photoperiodic response is illustrated by the extreme early flowering of cop1–4 mutants under SDs, which is responsible for almost abolishing the response to photoperiod.
COP1 therefore have an important function in turning over CO protein in the light and dark under these conditions.
COP1 activity is higher in the dark than light.
One mechanism by which this light regulation occurs is through exclusion of COP1 from the nucleus in the light (Von Arnim and Deng, 1994), whereas in addition COP1 activity is repressed by direct interaction with activated cryptochromes (Wang et al, 2001; Yang et al, 2001).
Our observation that CO protein levels are very high in the dark under LD or SD in cop1–4 mutants is consistent with COP1 activity being high in the dark under both day lengths and rapidly turning over CO protein.
The increase in CO protein at the end of the day under LD indicates that even in the light COP1 contributes to keeping CO protein levels low.
COP1 activity might only be reduced and not fully suppressed at the light intensities used in these experiments.
If so, then higher light intensities might promote flowering at least partly by more effectively repressing COP1 activity and allowing CO protein levels to rise higher.
This would suggest a role for COP1 and CO in the regulation of flowering by light intensity.
Degradation of CO protein in the morning or in red light does not require COP1
Previously, two distinct post-transcriptional mechanisms were postulated to shape the diurnal pattern of CO protein accumulation.
One of these occurred early in the day and involved a phyB-mediated signal and another occurred in the dark during the night (Valverde et al, 2004).
Degradation of CO protein in red light was proposed to involve the same phyB pathway that caused rapid turnover of the protein early in the day.
We observed that in cop1–4 mutants the CO protein was still effectively degraded in red light and that there was a strong diurnal trough in CO protein levels early in the day.
Degradation of CO early in the day was shown to require phyB but not COP1.
The degradation of CO in red light likely occurs by the same phyB-dependent mechanism acting early in the day, as was shown by Valverde et al (2004), and this could be tested by comparing cop1–6 phyB-9 and cop1–6 plants under red light.
Also, we cannot exclude that other phytochromes related to phyB may also have a function in CO regulation.
In particular, phyC and phyE were demonstrated to influence flowering time (Halliday and Whitelam, 2003; Monte et al, 2003; Balasubramanian et al, 2006).
Nevertheless, our data suggest that a second ubiquitin ligase may be responsible for phyB-mediated turnover of CO early in the day and in continuous red light (Figure 7).
Interestingly, the bHLH transcription factor PHYTOCHROME INTERACTING FACTOR 3, which is involved in seedling photomorphogenesis and phytochrome signalling, was also recently shown to be degraded by a red light-activated ubiquitin-mediated process soon after dawn (Al-Sady et al, 2006).
There might be a common mechanism promoting the degradation at dawn of several transcription factors involved in light signalling.
Alternatively, a set of ubiquitin ligases might exist that specifically promote the degradation of individual transcription factors at this time.
Further genetic and biochemical approaches are required to understand the mechanisms underlying CO protein degradation at dawn.
Spatial regulation of photoperiodic response by COP1
CO and FT are expressed in the vascular tissue and their expression in the phloem companion cells is sufficient to promote flowering (Takada and Goto, 2003; An et al, 2004).
Furthermore, reducing FT expression specifically in the phloem companion cells delays flowering (Mathieu et al, 2007).
Thus, the perception of photoperiod that is mediated through transcriptional and post-transcriptional regulation of CO likely takes place in the companion cells.
Similarly, Cry2, which positively regulates CO accumulation, exerts an effect in the companion cells to regulate flowering (Endo et al, 2007).
In contrast, phyB, a photoreceptor that delays flowering at least in part by reducing CO abundance, appears to exert an effect non-cell autonomously from the mesophyll cells (Endo et al, 2005).
This result suggests that a signalling step downstream of phyB exerts an effect non-cell autonomously to trigger degradation of CO protein, although definitive conclusions on the site of action of phytochromes influencing flowering will require a better understanding of the spatial requirement for other phytochromes, such as phyC and phyE.
We showed that expression of COP1 in the vascular tissue from the SUC2 promoter complemented the flowering-time phenotype of cop1–4 mutants under LDs and reduced FT mRNA levels.
Under SDs, cop1–4 SUC2:COP1 plants still flowered earlier than WT, but this was probably due to a CO-independent process causing early flowering in the cop1–4 mutant, because cop1–4 co-10 plants also flowered earlier than WT under SDs.
In WT plants, SUC2:COP1 also delayed flowering under LDs but not under SDs.
The day length specificity of this effect suggests that the overexpression in companion cells affects flowering through CO, and indicates that COP1 levels are a limiting factor on CO degradation under these conditions.
Taken together, these results suggest that COP1 exerts an effect in companion cells to regulate FT expression.
This observation is consistent with our suggestion that COP1 exerts an effect to degrade CO at the end of the day and during the night, but not as part of the phyB pathway, which exerts an effect mainly in the morning or in red light.
The temporal patterns of COP1 activity, therefore, support our understanding of the spatial pattern of activity of the pathways responsible for post-translational regulation of CO (Figure 7).
COP1 and CO interact in vitro and in nuclear speckles in vivo
COP1 directly interacts with target proteins and directs them for degradation (Hoecker, 2005; Jiao et al, 2007).
CO is composed of three domains, zinc-finger B-boxes, a central domain and the C-terminal CCT domain (Wenkel et al, 2006).
CO and COP1 interact directly in vitro as demonstrated by immunoprecipitation experiments.
This interaction was almost abolished when the C-terminal part of CO was removed, suggesting that COP1 interacts with the C-terminal region of CO, as was previously observed for interactions between COP1 and COL3 or between CO and SPA1 (Datta et al, 2006; Laubinger et al, 2006).
The interaction between COP1 and HY5 occurs through a defined domain that includes adjacent valine and proline residues that are essential for the interaction (Holm et al, 2001).
Conserved pairs of valine-proline residues in the CCT domain of COL3 are also important for the interaction with COP1 (Datta et al, 2006).
The region of CO that interacts with COP1 contains three VP motifs, but changing all of these to AA did not impair the interaction with COP1 in vitro (data not shown).
A similar result was previously observed for the interaction between CO and SPA1 (Laubinger et al, 2006).
Therefore, the interaction between CO and COP1 likely involves a different motif than observed for the interactions between COP1 and COL3 or HY5.
Direct interaction between COP1 and CO was further supported by transient expression of COP1 and CO fused to fluorescent proteins in Arabidopsis leaf cells.
The proteins colocalized in the nucleus and both occurred in speckles.
Previously, COP1 was shown to colocalize with its target proteins HY5, HYH and LAF1 in nuclear speckles in onion epidermal cells (Osterlund et al, 2000; Holm et al, 2002; Seo et al, 2003).
COP1 speckles were proposed to represent nuclear sites for proteasome-mediated protein degradation (Al-Sady et al, 2006).
The presence of CO and COP1 in nuclear speckles similar to those observed for other targets of COP1-mediated degradation supports the idea that direct interaction between COP1 and CO is required for CO degradation in the nucleus.
This degradation presumably requires the SPA proteins, perhaps acting directly in a larger order complex with COP1, as SPA proteins also regulate CO abundance at least at the end of a LD and interact directly with COP1 (Laubinger et al, 2006).
The mechanism by which CO is degraded by the SPA–COP1 complex has therefore strong parallels with that of HY5.
However, the precise relationship between SPA1 and COP1 activity and whether the proteins exert an effect in a larger order complex that interacts directly with substrates is still not clear (Saijo et al, 2003; Seo et al, 2003; Hoecker, 2005).
COP1 and the external coincidence model controlling flowering of Arabidopsis in response to photoperiod
CO promotes flowering and FT transcription under LDs but not SDs.
CO activity is proposed to be restricted to LDs by an external coincidence model in which circadian clock control and light signalling combine to trigger CO activity (Searle and Coupland, 2004; Imaizumi and Kay, 2006).
The cop1–4 mutant causes CO mRNA to accumulate earlier under SDs, so that cop1–4 mutations may in part accelerate flowering under SDs by causing CO mRNA to be expressed in the light, as previously shown for toc1-1 mutants (Yanovsky and Kay, 2002).
However, the major time of expression of FT under SDs in cop1–4 is later during the night, suggesting that the earlier phase of CO expression in cop1–4 has a small part in the acceleration of flowering under these conditions.
Rather the major role of COP1 in this system appears to be to degrade CO protein in the dark, and thereby ensure that no FT transcription occurs under SDs (Figure 7).
The importance of this activity is demonstrated by the high abundance of CO protein in the cop1–4 mutant under SDs and the extreme early flowering of cop1–4 mutants under these conditions.
During the final revision of this paper, another study described the role of cryptochrome signalling in suppressing COP1-mediated degradation of CO in the dark (Liu et al, 2008).
Our data extend the model of photoperiodic flowering in Arabidopsis by providing a molecular explanation for why CO mRNA expression during the night in SDs does not lead to FT transcription and promotion of flowering.
The day length-insensitive early-flowering phenotype of cop1–4 mutants and the strong suppression of this phenotype caused by co null alleles demonstrate that degradation of CO under SDs is essential in conferring a photoperiodic flowering response.
Materials and methods
Plant material
WT Arabidopsis thaliana plants and all mutants used in this study were Col-0.
The cop1–4 allele was previously characterized (McNellis et al, 1994).
The co-10 allele was previously used (Laubinger et al, 2006) and was confirmed to have a T-DNA insertion 342 bp after the ATG.
Homozygous cop1–4 co mutant plants were found using PCR-based markers.
The cop1–6 phyB-9 and cop1–6 seeds were kindly provided by Dr Jorge Casal (Boccalandro et al, 2004).
Analysis of flowering time
For analysis of flowering time and gene expression, plants were grown on soil in controlled environment rooms under LDs (16 h light–8 h dark) or SDs (8 h light–16 h dark).
Flowering time was measured by scoring the number of rosette and cauline leaves on the main stem of at least eight individuals.
Data are expressed as average±s.d.
mRNA expression analysis
Arabidopsis RNA was isolated with the Plant RNeasy kit (Qiagen) according to the manufacturer's instructions.
RNA was analysed by RT–PCR.
Detailed protocols and the origins of the primer sequences are presented in Supplementary data.
Plant transformation
The COP1 full-length cDNA was isolated by RT–PCR and produced as entry clone through BP reaction of Gateway system from Invitrogen.
Then, the entry clone was utilized for the construction of destination vectors for plant transformation, FRET experiments and in vitro-binding assay.
All plasmids for plant transformation were introduced into Agrobacterium strain GV3101 (pMP90RK) and transformed into WT Columbia, cop1–4 or SUC2:CO (An et al, 2004) plants by the floral dip method (Clough and Bent, 1998).
In vitro-binding assay
For the in vitro expression of GAD:COP1, we produced the vector pJIC39 containing pT7:GAD:GATEWAY cassette and T7 terminator, so that by LR reaction with the COP1 entry clone, the construct expressing GAD∷COP1 is produced.
Vector (Wenkel et al, 2006; Turck et al, 2007) pJIC26 is similar but contains only the GAD domain and was used for expressing full open reading frame of CO or parts of the ORF.
COΔB-box and COΔCCT (Laubinger et al, 2006) were also tested for the binding with COP1.
The detailed method used for the in vitro precipitation experiments is presented in Supplementary data.
Confocal microscopy, CO:COP1 colocalization and FRET analysis
To express CFP:COP1 and YFP:CO in plants, the CO and COP1 genes were cloned into the GATEWAY vectors pENSG:CFP or pENSG:YFP by recombination reaction.
In these vectors, CFP:COP1 and YFP:CO are expressed under the control of the constitutive 35S promoter (Laubinger et al, 2006).
The method used to analyse FRET is described in detail in Supplementary data.
Immunological techniques
WT Columbia, cop1–4 and spa1–7 were grown in temperature-controlled light cabinets either under LDs (16 h light and 8 h dark) or SDs (8 h light and 16 h dark).
Plants were grown on solid germination medium for 2 weeks, harvested at specified zeitgeber time (ZT), frozen in liquid nitrogen and kept at −80°C until further use.
For the red light experiments, 35S:CO and cop1–4 plants were grown in LD (16 h light–8 h dark) for 12 days, moved to red light conditions at ZT 0 and maintained for 16 h under red light.
Nuclear extracts were prepared from the plants at different ZT times as described previously (Valverde et al, 2004).
Nuclear proteins (17 μg) were separated employing 10% bis-Tris NuPAGE gels (Invitrogen), transferred to nitrocellulose membranes and probed with an anti-CO antibody followed by a horseradish peroxidase-conjugated secondary antibody.
Immunoreactive proteins were visualized by Pico chemiluminescence substrate system (Pierce).
The membrane was subsequently reprobed with an antibody against histone H3a (Abcam) as a loading control.
Supplementary Material
FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex
Photoactivated phytochrome induces rapid PIF3 phosphorylation prior to proteasome-mediated degradation
CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis
The PHYTOCHROME C photoreceptor gene mediates natural variation in flowering and growth responses of Arabidopsis thaliana
Promotion of photomorphogenesis by COP1
Arabidopsis CULLIN4 forms an E3 ubiquitin ligase with RBX1 and the CDD complex in mediating light control of development
Arabidopsis thaliana mutant that develops as a light-grown plant in the absence of light
Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana
FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis
Arabidopsis CONSTANS-LIKE3 is a positive regulator of red light signaling and root growth
cop1—a regulatory locus involved in light-controlled development and gene-expression in Arabidopsis
COP1, an Arabidopsis regulatory gene, encodes a protein with both a zinc-binding motif and a G-beta homologous domain
The degradation of HFR1, a putative bHLH class transcription factor involved in light signaling, is regulated by phosphorylation and requires COP1
CRYPTOCHROME2 in vascular bundles regulates flowering in Arabidopsis
Phytochrome B in the mesophyll delays flowering by suppressing FLOWERING LOCUS T expression in Arabidopsis vascular bundles
Changes in photoperiod or temperature alter the functional relationships between phytochromes and reveal roles for phyD and phyE
Regulated proteolysis in light signaling
The phytochrome A-specific signaling intermediate SPA1 interacts directly with COP1, a constitutive repressor of light signaling in Arabidopsis
SPA1, a WD-repeat protein specific to phytochrome A signal transduction
Identification of a structural motif that confers specific interaction with the WD40 repeat domain of Arabidopsis COP1
Two interacting bZIP proteins are direct targets of COP1-mediated control of light-dependent gene expression in Arabidopsis
Photoperiodic control of flowering: not only by coincidence
FKF1 F-box protein mediates cyclic degradation of a repressor of CONSTANS in Arabidopsis
FKF1 is essential for photoperiodic-specific light signalling in Arabidopsis
Cell-to-cell and long-distance trafficking of the green fluorescent protein in the phloem and symplastic unloading of the protein into sink tissues
The Arabidopsis SPA1 gene is required for circadian clock function and photoperiodic flowering
FT protein acts as a long-range signal in Arabidopsis
HFR1 is targeted by COP1 E3 ligase for post-translational proteolysis during phytochrome A signaling
Light-regulated transcriptional networks in higher plants
Activation tagging of the floral inducer FT
A pair of related genes with antagonistic roles in mediating flowering signals
The SPA quartet: a family of WD-repeat proteins with a central role in suppression of photomorphogenesis in Arabidopsis
Arabidopsis SPA proteins regulate photoperiodic flowering and interact with the floral inducer CONSTANS to regulate its stability
COP1-mediated ubiquitination of CONSTANS is implicated in cryptochrome regulation of flowering in Arabidopsis
Export of FT protein from phloem companion cells is sufficient for floral induction in Arabidopsis
Genetic and molecular analysis of an allelic series of cop1 mutants suggests functional roles for the multiple protein domains
Isolation and characterization of phyC mutants in Arabidopsis reveals complex crosstalk between phytochrome signaling pathways
Flowering of Arabidopsis cop1 mutants in darkness
Light: an indicator of time and place
Targeted destabilization of HY5 during light-regulated development of Arabidopsis
The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors
The COP1–SPA1 interaction defines a critical step in phytochrome A-mediated regulation of HY5 activity
Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis
FKF1 and GIGANTEA complex formation is required for day-length measurement in Arabidopsis
De-etiolated 1 and damaged DNA binding protein 1 interact to regulate Arabidopsis photomorphogenesis
Induction of flowering by seasonal changes in photoperiod
The transcription factor FLC confers a flowering response to vernalization by repressing meristem competence and systemic signaling in Arabidopsis
LAF1 ubiquitination by COP1 controls photomorphogenesis and is stimulated by SPA1
CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis
TERMINAL FLOWER2, an Arabidopsis homolog of HETEROCHROMATIN PROTEIN1, counteracts the activation of FLOWERING LOCUS T by CONSTANS in the vascular tissues of leaves to regulate flowering time
Arabidopsis TFL2/LHP1 specifically associates with genes marked by trimethylation of histone H3 lysine 27
Photoreceptor regulation of CONSTANS protein in photoperiodic flowering
Light inactivation of Arabidopsis photomorphogenic repressor Cop1 involves a cell-specific regulation of its nucleocytoplasmic partitioning
Direct interaction of Arabidopsis cryptochromes with COP1 in light control development
CONSTANS and the CCAAT box binding complex share a functionally important domain and interact to regulate flowering of Arabidopsis
Human De-etiolated-1 regulates c-Jun by assembling a CUL4A ubiquitin ligase
Integration of spatial and temporal information during floral induction in Arabidopsis
TWIN SISTER OF FT (TSF) acts as a floral pathway integrator redundantly with FT
The signaling mechanism of Arabidopsis CRY1 involves direct interaction with COP1
Light regulates COP1-mediated degradation of HFR1, a transcription factor essential for light signaling in Arabidopsis
Molecular basis of seasonal time measurement in Arabidopsis
Genetic characterization of the interaction between CO and COP1.
(A, B) cop1–4 mutants flowered earlier than wild-type Columbia plants irrespective of photoperiod, and the co-10 mutation suppresses the extreme effect of the cop1–4 mutation on flowering time under 16 h LD (A) and 8 h SD (B).
(C) Flowering times in LD and SD of genotypes shown in (A, B).
Flowering time is expressed as total leaf number (TLN) at flowering.
(D) COP1 expression under the phloem-specific promoter SUC2 largely rescued the early-flowering cop1–4 mutant phenotype.
The plants were grown under SD.
(E) Simultaneous expression of CO and COP1 in the phloem tissue.
SUC2:CO SUC2:HA:COP1 transgenic plants flowered later than SUC2:CO transgenic plants.
(F) SUC2:COP1 caused late flowering of wild-type Columbia plants under LD.
(G) Flowering times expressed as TLN at flowering under LD and SD of genotypes shown in (D–F).
Effect of COP1 on CO and FT mRNA levels.
(A) CO and FT mRNA analysis in wild-type (WT) Columbia, cop1–4 mutants, co-10 mutants and cop1–4 co-10 double-mutant plants under 8 h SDs.
(B) Quantification of the mRNA levels shown in (A).
Expressed as a ratio between UBQ10 mRNA level and FT or CO mRNA level.
(C) COP1 and FT mRNAs in cop1–4 mutants and two SUC2:COP1 cop1–4 transformants.
All plants were grown under SD and harvested 16 h after dawn.
(D) FT, COP1 and CO mRNAs in SUC2:CO and three SUC2:CO SUC2:HA:COP1 transformants.
All plants were grown under SD and harvested 8 h after dawn.
(E) CO and FT mRNAs in WT Columbia plants and in a SUC2:COP1 Columbia transformant.
All plants were grown under LD and harvested at 4-h intervals.
All genotypes are in the accession Columbia, and in (C, D) the numbers represent independent transgenic plants.
In (A, E) 2-week-old seedlings were sampled, whereas in (C, D) rosette leaves of 3-week-old plants were harvested.
In vitro interaction between CO and COP1 detected by co-immunoprecipitation.
35S-methionine-labeled CO, COΔB-box or COΔCCT was incubated with 35S-methionine-labeled GAD:COP1 or GAD and co-immunoprecipitated with anti-GAD antibodies.
Supernatant fractions and pellet fractions were resolved by SDS–PAGE and visualized by autoradiography using a phosphorimager.
Quantification of the fractions of prey proteins that were co-immunoprecipitated by the indicated bait proteins GAD:COP1 or GAD.
Error bars denote the standard error of the mean of two replicate experiments.
CO protein physically interacts with COP1 in plant cells.
(A) Transient co-expression of 35S:YFP:CO and 35S:CFP:COP1 constructs.
A 35S:dsRED construct was cotransformed to highlight the transformed cell.
The arrows represent the nucleus in which CO and COP1 are colocalized.
(B) Enlargement of the nucleus shown in each of the panels represented in (A).
(C) Quantification of FRET in vivo between CFP:CO and YFP:COP1.
YFP:CO detected as an increase in CFP fluorescence after photobleaching of YFP.
Quantification of FRET efficiencies after acceptor photobleaching measured in nuclei and nuclear speckles.
Data are mean±s.d.
of 10–20 cells from three separate experiments.
(D) Visualization of increase in CFP fluorescence after YFP photobleaching.
Left-hand panel, cells expressing CFP:COP1 and YFP, which exerts an effect as a negative control.
Right-hand panel, cells expressing CFP:COP1 and YFP:CO.
Scale bar: 6 μm in (A) and 8 μm in (D).
Detection of CO protein in cop1–4, cop1–6 phyB-9 and spa1–7 plants.
(A) CO protein was detected in 35S:CO transgenic plants and cop1–4 mutants, but not in WT Columbia, co-10 or cop1–4 co-10 mutants.
Plants were grown under 16 h LDs and harvested 16 h after dawn.
(B, C) CO protein in cop1–4 mutants under 16 h LD or 8 h SD.
Numbers above each lane represent hours after dawn that the sample was harvested.
Light bar represents day; dark bar represents night.
(D) CO protein detection in cop1–6 and cop1–6 phyB-9 plants grown under SDs.
Samples were harvested 6 and 16 h after dawn.
The reduction in CO protein at 6 h in cop1–6 plants (see also (C)) does not occur in cop1–6 phyB-9 plants.
(E) CO protein detection in spa1–7 mutants under 8 h SD.
Numbers and bars as described for (B, C).
In WT plants, CO protein could not be detected and therefore is not included as control ((A); Valverde et al, 2004).
For all panels, histone 3a was used as a loading control.
Comparison of CO protein and mRNA in plants exposed to white or red light.
35S:CO or cop1–4 seedlings (12-day old) grown in LD were divided into two groups and exposed to 16 h red or white light, respectively.
Samples were harvested for RNA and protein analysis at the end of the 16 h light period under both conditions.
(A) CO and ubiquitin mRNA levels in 35S:CO or cop1–4 plants exposed to white (W) or red (R) light.
Numbers in parentheses represent the numbers of cycles used to amplify the cDNA prior to separation on a gel.
(B) CO and histone protein levels in the same plants used for (A).
In both genotypes, CO is detected in white light (WL)-grown plants but not in red light (RL)-grown plants.
Model for regulation of CO stability during photoperiodic flowering control in wild-type plants and cop1–4 mutants.
Photoperiodic flowering in Arabidopsis involves two mechanisms of CO protein degradation: a phyB-dependent mechanism occurs early in the day or in response to red light and a second mechanism involving COP1 occurs late in the day and during the night.
(top) In wild-type plants under LDs, CO accumulates in the evening due to an increase in CO mRNA and photoreceptor-mediated repression of COP1.
CO can promote FT expression at this time and thereby flowering.
During the night, COP1 is active and causes rapid degradation of CO protein by ubiquitination and activity of the proteasome.
(second from top) Under SD, CO mRNA is expressed during the night and the protein is degraded through COP1 activity.
CO protein does not accumulate and FT mRNA is absent, resulting in late flowering.
(second from bottom) In cop1–4 mutants under LDs, CO is not degraded in the dark and accumulates to high levels.
CO also accumulates to high levels at the end of the day, consistent with COP1 targeting CO for degradation at that time.
However, CO protein still disappears early in the day, suggesting a COP1-independent mechanism of degradation at that time.
(bottom) Under SDs in cop1–4 mutants, CO accumulates to a high level during the night and promotes FT expression at higher level than in wild-type plants.
Enhanced CO activity at these times is responsible for the early flowering of cop1–4 mutants under SDs.
In the morning, CO protein is degraded by a COP1-independent mechanism.
The symbols represent CO protein abundance (red circles), COP1 (blue spheres), ubiquitin (small yellow circles on CO) and an unknown red light-activated degradation mechanism that is also active in the morning (dark orange spheres).
A full-colour version of this figure is available at The EMBO Journal Online.The DEAD-box helicase DDX3X is a critical component of the TANK-binding kinase 1-dependent innate immune response
TANK-binding kinase 1 (TBK1) is of central importance for the induction of type-I interferon (IFN) in response to pathogens.
We identified the DEAD-box helicase DDX3X as an interaction partner of TBK1.
TBK1 and DDX3X acted synergistically in their ability to stimulate the IFN promoter, whereas RNAi-mediated reduction of DDX3X expression led to an impairment of IFN production.
Introduction
Innate immune responses are stimulated when microbial molecules referred to as pathogen-associated molecular patterns (PAMPs) bind to a cognate pattern recognition receptor (PRR) (Medzhitov and Janeway, 1997).
PRRs can be expressed at the cell surface, the endosomal compartment or in the cytoplasm of mammalian cells (Akira, 2006).
Three major families of PRRs have been identified: the Toll-like receptors (TLRs), the RIG-I-like RNA helicases and the NOD-like receptors (NLRs).
A total of 11 different TLRs interact with lipid, protein or nucleic acid PAMPs either at the cell surface or the endosome (Akira, 2006).
By contrast, the RIG-I-like helicases exclusively detect cytosolic RNA of viral origin (Yoneyama and Fujita, 2007).
The NLRs bind to a large variety of PAMPs or danger-associated molecular patterns that include peptides, small molecules and nucleic acids (Kanneganti et al, 2007; Petrilli et al, 2007).
Cytosolic DNA, a PAMP resulting from infection with DNA-containing intracellular pathogens, is recognized by a yet unknown number of DNA receptors.
One component of the DNA recognition machinery is DAI (DLM-1/ZBP1), a protein that does not belong to either of the three major PRR families (Ishii et al, 2006, 2008; Takaoka et al, 2007; Wang et al, 2008).
PRR engagement triggers several intracellular signalling cascades that culminate in the production and secretion of pro-inflammatory cytokines, chemokines and type-I interferons (IFNs) (predominantly IFN-α and -β collectively termed IFN-I (Pestka et al, 2004; Kawai and Akira, 2007a, 2007b)).
These secreted mediators, in turn, activate a gene expression programme and orchestrate a protective immune response against the invading pathogen.
Interferons are the most important antiviral cytokines and loss of IFN-I signalling leads to a severe immunodeficiency towards viral infection (Muller et al, 1994).
As a primary response to infection, IFN-β production can be triggered by several PRRs, most notably TLR3 and TLR4, RIG-I and MDA5 and the cytoplasmic DNA receptors.
TLR-induced signalling is initiated by the recruitment of an adaptor called TRIF and proceeds through the ubiquitin ligase TRAF3 and the kinases TANK-binding kinase 1 (TBK1)/IKK-i to phosphorylate and activate the transcription factors IFN-regulatory factors (IRF) 3 and 7 (Akira, 2006; Honda and Taniguchi, 2006; He et al, 2007).
RIG-I and MDA5 signalling uses a different adaptor called MAVS (IPS-1, VISA, Cardif; Kawai et al, 2005; Meylan et al, 2005; Xu et al, 2005; Sun et al, 2006), but also involves TRAF3, TBK1/IKK-i and IRF3/IRF7 (Saha et al, 2006; Yoneyama and Fujita, 2007).
DNA-mediated signalling initiated by DAI (DLM-1/ZBP1) and/or additional cytoplasmic receptors is poorly understood, but similar to the other PRRs it funnels into the TBK1/IKK-i–IRF3 pathway to activate the IFN-β gene (Ishii et al, 2006; Takaoka et al, 2007).
A plethora of studies thus converge to emphasize the role of TBK1 as an activator of IRF3, hence as a central regulator of IFN-I genes in response to many different PAMPs.
TBK1-mediated phosphorylation of IRF3 is believed to occur in the C-terminal domain of IRF3 and triggers a conformational change that allows for the dimerization and subsequent nuclear translocation of IRF3 (Honda and Taniguchi, 2006; Hiscott, 2007).
In some cells, TBK1 function is shared by the highly related IKK-i (or IKKɛ) kinase, whereas in others TBK1 has a non-redundant function as an IRF3 kinase (Hemmi et al, 2004; McWhirter et al, 2004; Matsui et al, 2006).
The Gram-positive bacterium Listeria monocytogenes has adapted to an intracellular lifestyle by virtue of its ability to escape from endosomal or phagosomal compartments to the cytoplasm of the host cell (Portnoy et al, 2002; Hamon et al, 2006).
Disruption of the endosomal/phagosomal membrane results from the activity of the major Listeria virulence protein, a haemolysin called Listeriolysin O (Schnupf and Portnoy, 2007).
Intracellular infection with L.
monocytogenes stimulates robust expression of IFN-I genes and this was shown to require TBK1 as well as IRF3 (O'Connell et al, 2004, 2005; Stockinger et al, 2004; Decker et al, 2005).
The nature of the PAMP delivered to the cytoplasm by L.
monocytogenes, although speculatively DNA (Stetson and Medzhitov, 2006), requires final clarification.
None of the known PRR adapters are essential to connect the Listeria ligand and TBK1 (Stockinger et al, 2004; Soulat et al, 2006; Sun et al, 2006).
The L.
monocytogenes ligand and cytoplasmic DNA pathways thus share the lack of known players to address and activate TBK1.
To start bridging the gap between the L.
monocytogenes ligand and the IRF3 pathway, we sought to identify TBK1-interacting proteins from macrophages.
We report here the surprising finding that a DEAD-box helicase not closely related to, and with properties clearly distinct from, the RIG-I family is phosphorylated by TBK1 to enhance expression of IFN-I genes.
Results
Identification of DDX3X as a novel TBK1 target
To identify novel interactors of TBK1, we generated stable RAW264.7 cell lines that express a GS-TAP-tagged version of TBK1 and purified the TBK1 protein complex by tandem affinity purification (Figure 1A and B) (Burckstummer et al, 2006).
Mass spectrometry analysis identified TBK1 along with the core complex components TRAF family member-associated NF-kappa-B activator (TANK; 17 peptides), TBK-binding protein 1 (TBKBP1, also referred to as SINTBAD; 10 peptides) and TBKBP2 (also referred to as NAP1 or AZI2; 15 peptides), indicating that this native purification was efficient and in agreement with previously published data on the TBK1 core complex (Bouwmeester et al, 2004).
In addition, we identified the DEAD-box helicase DDX3X (RefSeq NP_034158) with five peptides.
Immunoprecipitation experiments using tagged TBK1 suggested that its interaction with DDX3X and the transcription factor IRF3 are significantly weaker than the interaction between TBK1 and TANK and therefore not detected by coimmunoprecipitation under stringent conditions (Supplementary Figure 1A).
Likewise, we could not detect any association of DDX3X with IRF3 (Supplementary Figure 1B).
Requirement of DDX3X for IFN-β production
We addressed the functional relevance of the TBK1/DDX3X interaction by reducing DDX3X expression and measuring the impact on type-I IFN production.
siRNA transfection led to a decrease in DDX3X or TBK1 expression by at least 50% (Figure 2A and B).
The cells were infected with L.
monocytogenes and IFN-β mRNA production was evaluated by quantitative RT–PCR.
As expected, TBK1 was required for IFN-β mRNA transcription (Figure 2C).
Strikingly, we observed the same level of reduction for the cells in which DDX3X expression was reduced by siRNA, suggesting that DDX3X is necessary for IFN-β induction.
Similar results were obtained for Listeria-mediated induction of RANTES (Figure 2C), another IRF3-dependent gene.
To verify that the DDX3X knockdown does not have a general effect on the response to L.
monocytogenes, we measured the induction of TNF-α after infection.
TNF-α is strongly induced upon L.
monocytogenes infection (data not shown), but it does not rely on the TBK1/IRF3 axis (Doyle et al, 2002).
In agreement with this, we did not observe any defect in the TNF-α induction neither when TBK1 nor when DDX3X expression was reduced by RNA interference (Figure 2C).
Next, we wanted to address whether DDX3X has a specific role during L.
monocytogenes infection or a more general role in IFN-β induction.
To this end, we stimulated the siRNA-treated cells with different PAMPs such as LPS, transfected poly(I:C) or transfected poly(dA:dT) targeting, respectively, TLR4, MDA5 or the cytosolic DNA receptor.
In all of these experimental conditions, reduction of DDX3X expression caused a decrease in IFN-β induction (Figure 2D), confirming that DDX3X is necessary for type-I IFN induction in general.
TBK1/IRF3-mediated production of IFN-β triggers a positive feedback loop to fortify IFN production (Honda et al, 2005).
Newly produced IFN-β engages the type-I IFN receptor (IFNAR) which, in turn, stimulates the production of IRF7.
Subsequently, IRF7 cooperates with IRF3 to cause a second wave of type-I IFN production.
To address whether DDX3X is also involved in this second loop of signalling initiated by IFNAR engagement, we decided to analyse the impact of reduced DDX3X expression on IRF7 transcription.
We observed a diminished IRF7 induction upon TBK1 or DDX3X knockdown in cells infected with L.
monocytogenes (Figure 2E) and this reduction was similar to the one observed for induced expression of the IFN-β gene (Figure 2C).
By contrast, there was no effect of the TBK1 or DDX3X knockdown on the IFN-β-stimulated IRF7 expression (Figure 2E).
Similar results were obtained for Mx2, another downstream target of the IFNAR (Figure 2E), indicating that DDX3X does not have a function in IFN-induced gene expression.
Overall, these experiments suggest that DDX3X is required for TBK1/IRF3-mediated IFN-β production.
Impact of TBK1 on DDX3X function
The function of DDX3X that may be best documented so far is its role in the nuclear export of human immunodeficiency virus-1 (HIV-1) mRNA.
To test whether this function is affected by TBK1 activity, we used an established experimental set-up involving a Rev reporter plasmid (Yedavalli et al, 2004).
This assay monitors the DDX3X-dependent nuclear export of a Rev response element-containing RNA encoding the CAT gene (Figure 3A).
We found that DDX3X by itself has only a moderate impact on Rev-mediated nuclear export (Figure 3B).
Coexpression of TBK1 strongly increased the activity of DDX3X.
This effect was neither seen with an inactive mutant of TBK1 (K38M) (Fitzgerald et al, 2003) nor with an ATPase-deficient mutant of DDX3X (K230A) (Yedavalli et al, 2004), implying that both the kinase activity of TBK1 and the helicase activity of DDX3X are required for this function of DDX3X.
Synergistic activation of the IFN-β promoter by TBK1 and DDX3X
Is DDX3X sufficient to activate the IFN promoter? We decided to address this question using an IFN-β reporter plasmid in HEK293 cells.
MAVS and, to a lesser extent, TBK1 induced the IFN-β promoter, whereas DDX3X by itself did not (Figure 4A).
Nevertheless, coexpression of TBK1 with DDX3X led to a synergistic promoter activation that was also observed for the ATPase-deficient mutant of DDX3X (Figure 4A), indicating that the ATPase and the helicase functions of DDX3X are dispensable for its synergy with TBK1.
DDX3X did not synergize with TBK1 with regard to NF-κB promoter induction (Figure 4B), sustaining the specific role of DDX3X in the IFN pathway.
Next, we performed epistasis analyses to position DDX3X in the IRF3 pathway.
To this end, we used mouse embryonic fibroblasts (MEFs) from TBK1- or IRF3-deficient mice along with wild-type controls.
Again, MAVS was a strong activator of IFN-β and MAVS-mediated activation depended more on the presence of IRF3 than on TBK1 (Figure 4C).
This result can be attributed to a certain degree of redundancy between TBK1 and IKK-i (Hemmi et al, 2004).
When TBK1 was expressed at low levels (compare Supplementary Figure 2), TBK1 by itself was only a weak inducer of IFN-β.
In contrast, DDX3X and TBK1 together gave a strong synergistic activation of the IFN-β promoter (Figure 4C).
Importantly, this effect was also seen in TBK1−/− MEFs, but not in IRF3−/− MEFs, confirming that the DDX3X effect on IFN requires TBK1 and does not bypass the requirement for IRF3.
Noteworthy, this synergy depends on kinase activity of TBK1 as inactive TBK1 S172A cannot synergize with DDX3X (Figure 4C).
Overall, these data support that DDX3X has a positive role in the IFN pathway and is therefore required for efficient IFN production.
How does DDX3X affect IFN-β gene expression? Our data are consistent with the possibility that DDX3X impinges on IFN-β promoter activity directly.
We therefore asked whether DDX3X can be recruited to the IFN promoter.
GS-TAP-tagged DDX3X, isolated by affinity purification using rabbit IgG agarose, was also recruited to the enhanceosome region upon L.
monocytogenes infection (Figure 5C).
This recruitment was specific because a control region (Figure 5A) could not be amplified by PCR under these conditions.
This suggests that DDX3X exerts a direct effect on the IFN-β promoter.
To address whether DDX3X recruitment to the IFN-β promoter depends on IRF3, we synthesized a biotinylated double-stranded DNA fragment corresponding to the IFN-β enhancer (Panne et al, 2007), along with two variants in which the IRF3-binding site was either mutated or deleted (Supplementary Figure 3A).
We used these oligos as affinity reagents to monitor IRF3 or DDX3X engagement using lysates from untransfected RAW264.7 cells.
We found that the apparent majority of IRF3 strongly associated with the wild type, but only weakly with the mutated enhancer sequence (Supplementary Figure 3B).
On the other hand, a minor fraction of DDX3X associated with both the wild-type and the mutated enhancer sequence, suggesting that its binding is independent of IRF3.
DDX3X as a TBK1 kinase substrate
Having established the importance of DDX3X in the TBK1-dependent IRF3 pathway, we further investigated the molecular nature of the relationship between TBK1 and DDX3X.
Given the weak affinity of DDX3X for TBK1, we considered the possibility of a kinase–substrate interaction.
We observed that coexpression of TBK1 with DDX3X is associated with a slight shift in DDX3X migration on an SDS–PAGE (Supplementary Figure 1).
To investigate whether this shift in migration was dependent on TBK1 kinase activity, we generated different inactive mutants of TBK1, mutating either a putative phosphorylation site in the activation loop (S172A), the ATP-binding site (K38M) or the catalytic base (D135N).
Activity of the TBK1 mutants was monitored by an in vitro kinase assay or by immunoblotting using an antibody that was raised against phosphoserine 172 in the activation loop of TBK1.
Whereas TBK1 wt was active in the kinase assay (Supplementary Figure 4) and phosphorylated on S172 (Figure 6A, middle panel), all mutants were inactive and unphosphorylated on S172, suggesting that the antibody is a valuable tool to monitor the activation status of TBK1 in vivo.
How does loss of TBK1 kinase activity affect DDX3X migration? The aforementioned shift in migration was strictly dependent on TBK1 kinase activity, as it was not seen when different inactive mutants of TBK1 (TBK1 S172A, K38M or D135N) were coexpressed with DDX3X (Figure 6A).
To unequivocally prove that the shift in migration reflects a phosphorylation event, we added the non-selective kinase inhibitor staurosporine and showed that the TBK1-induced shift was strongly diminished in the presence of the drug.
Likewise, treatment of the cell extract with calf intestinal phosphatase led to a decreased shift in DDX3X migration, providing strong evidence that coexpression of TBK1 causes DDX3X phosphorylation.
In parallel, we compared DDX3X to IRF3 in this assay (Figure 6A).
Strikingly, the phosphorylation pattern observed for IRF3 mirrored the one obtained for DDX3X.
These results indicate that DDX3X and IRF3 are very similar in terms of their TBK1 substrate quality.
To demonstrate whether the shift in DDX3X migration was specific for TBK1, we coexpressed DDX3X with a number of kinases including IKK-i, the closest relative of TBK1.
Interestingly, IKK-i could also phosphorylate DDX3X, but DDX3X phosphorylation was not observed when the kinases IKKα, IRAK1, Abl or FAK2 were coexpressed (Figure 6B).
This suggests that DDX3X is rather specifically phosphorylated by TBK1 and IKK-i.
So far, the data presented supported the possibility that DDX3X is a novel TBK1 substrate, although we could not formally exclude the presence of an intermediate kinase activated by TBK1.
To rule out any indirect effect, we expressed His-tagged IRF3, DDX3X and Grb2 in Escherichia coli and purified them by HisTrap chromatography.
Equal amounts of purified proteins were then used for an in vitro kinase assay using TBK1 purified from HEK293 cells.
Noteworthy, TBK1 alone (lane 1) or the substrates alone (lanes 2, 6 and 10) did not produce any signal.
IRF3 was phosphorylated by TBK1 whereas Grb2 was not, suggesting that the assay was performed under conditions that warrant substrate specificity (Figure 6C).
The signal obtained for DDX3X was equivalent to the one obtained for IRF3, implying that DDX3X is also a TBK1 substrate in vitro.
DDX3X phosphorylation site mapping
To map the phosphorylation site systematically, we decided for a peptide array that displays 73 peptides containing all serine and threonine residues of DDX3X (Supplementary Figure 5).
As a control, we included a peptide derived from the TBK1 activation loop that contains putative autophosphorylation site S172 (Figure 6A).
As another control, we included a peptide derived from the C-terminus of IRF3 that contains S396, one of the TBK1 target sites (Servant et al, 2003).
As the control peptides contained several serines and threonines, we sequentially exchanged these for alanines to map which serine or threonine within these peptides was actually phosphorylated.
Incubation of the peptide array with purified TBK1 produced a number of phosphorylation signals (Figure 7A).
The TBK1 control peptide was phosphorylated by TBK1, suggesting that the activation loop is indeed an autophosphorylation site.
Mutagenesis indicates that the phosphorylation occurs predominantly at S172, highlighting the relevance of the antibody that recognizes phospho-S172 in TBK1 (Figure 6A).
The peptide derived from IRF3 was efficiently phosphorylated as well, but unexpectedly, the phosphorylation occurred exclusively at S402.
Analysing the remaining hits in the peptide array, we identified 11 potential phosphorylation sites in 9 DDX3X-derived peptides (Figure 7A).
Four of these target sites (S181, S183, S240 and S269) were found in the DEAD domain that contains the ATPase activity of DDX3X.
The seven remaining sites were scattered throughout the helicase domain (S429, T438, S442, S456, S520, T542 and S543).
We used the information gathered in the peptide array to derive a TBK1 consensus phosphorylation site (Figure 7B).
Strikingly, TBK1 showed a strong preference for serine over threonine.
In fact, none of the 12 peptides displayed on the array that contain exclusively threonines was phosphorylated.
Furthermore, we identified several preferences of TBK1 in terms of amino acids adjacent to the actual site of phosphorylation, the most striking being a preference for a hydrophobic amino acid (L or F) in the +1 position as also seen in the two control peptides.
It is interesting to note that S386, which has been identified as critical for IRF3 activation (Mori et al, 2004), also contains a Leu in the +1 position.
To test which of those putative phosphorylation sites are surface-exposed and therefore accessible to TBK1 in the context of the folded protein, we mapped them onto the structure of a truncated version of DDX3X that was recently published (Hogbom et al, 2007).
All sites except T542/S543 mapped to the surface of DDX3X (Figure 7C), implying that the vast majority of these sites are bona fide TBK1 phosphorylation sites in vivo.
Phosphorylation of DDX3X required for synergy with TBK1
To obtain phosphorylation-deficient mutants for functional experiments, we created three DDX3X mutants.
We mutated either all potential phosphorylation sites collectively (Pan-M) or the phosphorylation sites in the DEAD domain (Dead-M) or those in the helicase domain separately (Helic-M) (Figure 8A).
Mutagenesis did not include T542 or S543 as those residues were predicted to disrupt the secondary structure of DDX3X.
Again, we purified those DDX3X mutants from E.
coli and analysed TBK1-mediated phosphorylation in vitro.
DDX3X wt was strongly phosphorylated (Figure 8B).
In contrast, phosphorylation of Pan-M was strongly diminished as was the case for the Dead-M mutant.
The Helic-M mutant was also reduced in terms of TBK1 phosphorylation, but the effect was less pronounced.
On the basis of these results, we conclude that both the DEAD domain and the helicase domain contain at least one TBK1 phosphorylation site and that the predominant phosphorylation event occurs in the DEAD domain.
We tested these DDX3X mutants for their ability to synergize with TBK1 in the IFN-β reporter assay.
Although DDX3X wt showed a strong synergy with TBK1, all the phosphosite mutants failed to enhance activity of the IFN-β promoter (Figure 8C).
This establishes a causal link between TBK1-mediated phosphorylation of DDX3X and the ability of the two proteins to stimulate IFN-β expression.
Discussion
As far as current knowledge goes, TBK1 is arguably the most important kinase in the IFN pathway, yet we know of only two substrates, IRF3 and IRF7 (Pichlmair and Reis, 2007).
In this study, we show that the DEAD-box helicase DDX3X is required for IFN production.
DDX3X is found to be a TBK1 substrate in vitro and in vivo and phosphorylation by TBK1 is necessary for DDX3X to stimulate the IFN-β promoter.
During these studies, we were also able to advance the biochemical characterization of TBK1: first, we identified and characterized a set of inactive mutants (Figure 6).
Second, we developed a phospho-specific TBK1 antibody that can be used to monitor TBK1 activity in vivo (Figure 6).
Third, we investigated the properties of TBK1 in vitro, using both recombinant proteins and peptides as substrates.
On the basis of these analyses, we are providing what we believe may be the first consensus sequence for a TBK1 substrate, which could be helpful for the future identification of TBK1 target sites.
DEAD-box helicases constitute a large family of proteins that comprises at least 38 members in humans (Linder, 2006).
These enzymes are characterized by the presence of nine conserved motifs organized in two highly conserved domains: a DEAD domain (named after the characteristic amino-acid sequence D-E-A-D) and a helicase domain.
Their nucleic acid unwinding activity requires the hydrolysis of ATP through Walker motifs present in the DEAD domain.
Although DEAD-box helicases are usually classified as RNA helicases, it has been suggested that some of them, including DDX3X, can also bind to DNA (Franca et al, 2007).
Genetic and biochemical experiments mainly realized in yeast suggested that DEAD-box helicases could be involved at many steps of RNA metabolism, including the initiation of the transcription, RNA splicing, nuclear export, control of mRNA translation and the regulation of RNA stability.
DDX3X is an X chromosome-encoded DEAD-box helicase that has been implicated in nuclear export of HIV-1 RNA through Rev/Crm1 (Yedavalli et al, 2004) and transcriptional regulation of the p21waf1/cip1 promoter (Chao et al, 2006).
Both of these activities require access to the nuclear compartment, which is achieved by the means of a nuclear localization signal and a nuclear export signal.
Incubation of cells with nuclear export inhibitor leptomycin B traps DDX3X in the nucleus (Yedavalli et al, 2004), confirming that it is shuttling between nucleus and cytoplasm.
DDX3X belongs to the same family of helicases as RIG-I and MDA5 and yet, its mode of action in innate immunity seems to be distinct: unlike RIG-I and MDA5, DDX3X does not exert an effect as a PRR, but is situated rather downstream in the signalling cascade that controls IFN-β production.
Furthermore, unlike RIG-I and MDA5, the DDX3X function in the IRF pathway does not rely on its ATPase activity, implying that it may rather exert an effect as a scaffold than as a helicase.
A similar scenario has been described for other transcriptional co-activators that exert an effect as bridging factors to build the transcriptional activator complex (Valineva et al, 2005).
Our data suggest that DDX3X is required for the function of the TBK1/IRF3 module and that DDX3X exerts at least part of this function at the promoter level.
We would like to propose that activated TBK1 phosphorylates DDX3X which, in the nucleus, gets recruited to the IFN promoter where it stimulates transcription of the IFN-β gene.
How this effect is brought about is currently unclear and will be the objective of future studies.
DDX3X has been reported to promote transcription by direct interaction with transcription factors (Chao et al, 2006).
In our case, however, we fail to observe a direct interaction between IRF3 and DDX3X and the association of DDX3X with the IFN-β enhancer appeared to be independent of the IRF3-binding site.
This is reminiscent of the direct association of DDX3X with the E-Cadherin promoter (Botlagunta et al, 2008).
On the basis of available literature, however, it is likely that the bulk of the DDX3X molecular functions involve its RNA binding and helicase function.
So far, our attempts to find evidence for a post-transcriptional role of DDX3X in the IFN pathway have failed.
We cannot rule out the possibility that DDX3X may affect IFN-β production at other stages of gene regulation as well.
For instances, DDX3X has been reported to bind to the cap-binding protein complex (Shih et al, 2008).
As capping is a cotranscriptional event required for quality control of nascent transcripts (Orphanides and Reinberg, 2002), one could speculate that DDX3X modulates transcriptional elongation of the IFN-β gene.
Additionally, there is the possibility of DDX3X having an effect on nuclear export or translation of the IFN-β mRNA.
In fact, DDX3X has been implied in the nuclear export of HIV-1 RNA (Yedavalli et al, 2004) on top of cap-dependent mRNA translation (Shih et al, 2008), overall suggesting that DDX3X may indeed interfere with IFN induction at different stages (transcription, nuclear export and mRNA translation) to allow for a fine-tuning or fail saving of the process.
This may also explain why TBK1 seems to rely on at least two substrates (IRF3 and DDX3X) to regulate IFN-β production.
The fact that DDX3X is necessary for IFN production implies that it has an antiviral function.
In apparent contradiction to this notion, hepatitis C virus (HCV) and HIV-1 seem to exploit DDX3X and use it to their advantage (Yedavalli et al, 2004; Ariumi et al, 2007): DDX3X is needed to sustain the viral life cycle of HIV-1 by facilitating the nuclear export of HIV-1 RNA (Yedavalli et al, 2004).
Likewise, DDX3X binds to the HCV core protein and supports viral replication (Ariumi et al, 2007).
This obvious contradiction, however, may be resolved by interpreting it in light of the constant evolutionary struggle between host and pathogen: DDX3X may have evolved as an antiviral protein, but viruses then, in turn, developed strategies to hijack DDX3X function and use it for their own benefit, thereby counteracting the antiviral effect of DDX3X.
On the basis of this, we postulate that additional viral DDX3X antagonists may exist that inactivate DDX3X either by marking it for proteasome-mediated degradation, by direct proteolytic cleavage or by altering its phosphorylation pattern.
Materials and methods
Antibodies
The TBK1-specific antiserum was obtained from Cell Signaling.
The IRF3-specific antiserum was obtained from Invitrogen.
The DDX3X-specific antiserum was generated by immunizing rabbits using a recombinant purified domain of DDX3X.
Immunizations were carried out by Eurogentec (Belgium).
Plasmids
All ESTs were obtained from RZPD (Berlin, Germany).
Genes were based on the following RefSeq IDs: Human DDX3X (NM_001356), human IRF3 (NM_001571).
HIV-1 Rev and the reporter plasmid pDM128 were kind gifts of Dr Thomas Hope (University of California, San Diego) and Dr Bryan Cullen (Duke University).
Mutagenesis of DDX3 was performed by custom gene synthesis (Genscript Corporation, Piscataway, NJ, USA; www.genscript.com).
Tandem affinity purification
Tandem affinity purification using the GS-TAP cassette was performed as described previously (Burckstummer et al, 2006).
Elution was performed by boiling the sample in SDS sample buffer.
siRNA treatment of RAW264.7 cells
siRNAs for TBK1 and DDX3X were purchased from Dharmacon.
RAW264.7 cells (2 × 104) were transfected with 25 nM siRNA complexed with HiPerFect (Qiagen) according to the manufacturer's instructions.
Transfection was repeated once after 48 h.
At 72 h after the first transfection, 5 × 105 cells were seeded onto 6 cm dish to be stimulated 24 h later.
RAW264.7 stimulation
siRNA-treated RAW264.7 cells were infected with L.
monocytogenes LO28 at a multiplicity of infection of 10 for 4 h.
In parallel, cells were transfected with 30 μg/ml poly(I:C) (Amersham Biosciences) complexed with HiPerFect (Qiagen) for 3 h or 10 μg/ml poly(dA:dT) (Sigma) complexed with PolyFect (Qiagen) for 4 h according to manufacturer's instructions.
Additionally, cells were treated with 100 ng/ml LPS derived from Salmonella minnesota (Alexsis) or with 500 U/ml IFN-β (Calbiochem, La Jolla, CA) for 4 h.
RNA isolation, cDNA synthesis and quantitative PCR
After stimulation, total RNA was extracted from RAW264.7 cells using Nucleospin RNA II kit (Macherey & Nagel).
The RNA was then reverse transcribed with oligo(dT)18 primer and RevertAid M-MuLV Reverse Transcriptase (MBI Frementas) according to the manufacturer's instructions.
The cDNA was then analysed by quantitative PCR using the iCycler IQ machine (Bio-Rad).
The target genes (IFN-β, RANTES, TNFα, IRF7 and Mx2) were quantified using the standard curve method and normalized to the endogenous control GAPDH.
The PCR was performed by using SybrGreen (Molecular Probes) and the Taq polymerase (MBI Fermentas) and the following primers: GAPDH for, 5′-CATGGCCTTCCGTGTTCCTA-3′; GAPDH rev, 5′-GCGGCACGTCAGATCCA-3′; IFN-β for, 5′-TCAGAATGAGTGGTGGTTGC-3′; IFN-β rev, 5′-GACCTTTCAAATGCAGTAGATTCA-3′; TNFα for, 5′-CAAAATTCGAGTGACAAGCCTG-3′; TNFα rev,5′-GAGATCCATGCCGTTGCC-3′; IRF7 for, 5′-CTGGAGCCATGGGTATGCA-3′; IRF7 rev, 5′-AAGCAGAAGCCCAGACTGCT-3′; Mx2 for, 5′-CCAGTTCCTCTCAGTCCCAAGATT-3′; Mx2 rev, 5′-TACTGGATGATCAAGGGAACGTGG-3′; RANTES for, 5′-CTCACCATCATCCTCACTGC-3′; RANTES rev, 5′-ACTTGGCGGTTCCTTCG-3′.
Rev reporter assay
HEK293 cells were seeded at 5 × 105 cells per well in six-well plates.
Cells were transfected with 500 ng of pDM128 (Rev reporter plasmid), 30 ng of Rev expression plasmid, 300 ng of DDX3X expression plasmid and 500 ng of TBK1 expression plasmid.
All samples were normalized for DNA content using an empty plasmid.
Cells were harvested 24 h post-transfection and CAT levels were analysed by ELISA (Roche) according to the manufacturer's instructions.
Generation of point mutant of TBK1 and DDX3X
QuikChange® Multi Site-Directed Mutagenesis Kit (Stratagene) was used to generate point mutants with appropriate primers regarding the sequence to target.
Luciferase assay
HEK293 and MEF cells were transfected for 24 h with the IFN-β reporter (firefly luciferase) plasmid (pTA-Luc-IFN-β) or the NF-κB firefly luciferase plasmid (pTA-Luc-NF-κB) together with the Renilla luciferase control plasmid (phRG-TK) in complex with ExGen500 (MBI Fermentas) according to the manufacturer's instructions.
As target vectors, N-HA-MAVS or N-HA-DDX3X plasmid (or derived mutants) and/or myc-TBK1 (or derived mutants) plasmid were used as indicated.
Empty HA vector was used to equalize the DNA content of each transfection.
At 24 h after the transfection, the cells were lysed and the extracts were analysed by using the Dual-Luciferase Reporter Assay System (Promega) and the luminometer Lumat (LB9501; Berthold).
For quantification, the firefly luciferase activity was then normalized to the Renilla luciferase activity.
ChIP
A total of 3 × 107 RAW264.7 wt or NTAP(GS)-DDX3X cells were used per time point.
ChIP assays were performed as recently described (Zupkovitz et al, 2006) using 3 μg of IRF3-specific antibody (Zymed, Invitrogen) followed by a protein A pulldown or using rabbit IgG agarose for NTAP(GS)-DDX3X.
For the analysis of the enhanceosome-binding site of ifnb1 promoter, the following primers were used: forward, 5-caggatgagcagctactctgc-3, and reverse, 5-ctgctacctgcaagatgagg-3.
For the 1 kb upstream control region, the following primers were used: forward, 5-gccaatgatgtggttcagc-3, and reverse, 5-gagacctgtctgctgttagtgg-3.
Generation of the antiserum directed against phosphoserine 172 in TBK1
The phospho-specific antiserum was custom-synthesized by Eurogentec using a phosphorylated peptide derived from the activation loop of TBK1 (DDEQFVpSLYGTEE) (Eurogentec).
Transient transfection of HEK293 and preparation of cell extracts
HEK293 cells were transiently transfected using PolyFect (Qiagen) according to the manufacturer's instructions.
Cells were harvested 24 or 48 h post-transfection and lysed in Frackelton buffer (10 mM Tris–HCl pH 7.5, 50 mM NaCl, 30 mM sodium pyrophosphate, 1% Triton X-100, 1 mM DTT, 100 μM sodium orthovanadate, 50 μM NaF and protease inhibitors).
Kinase assays
For peptide-based assays, Myc-TBK1 was isolated from HEK293 cells by immunoprecipitation using anti-Myc agarose (Sigma).
Precipitated TBK1 was incubated with a biotinylated peptide derived from IRF3 (DLHISNSHPLSLC) in the presence of 50 μM ATP, 5 μCi [γ-32P]ATP and kinase buffer (40 mM Tris–HCl pH 7.5, 10 mM MgCl2 and 1 mM DTT) for 30 min at 30°C.
The reaction was terminated by addition of guanidinium chloride (2.5 M final concentration) and the reaction was spotted onto a SAM2 Biotin Capture membrane (Promega) and further treated according to the manufacturer's instructions.
Kinase activity was then measured by scintillation counting.
For those kinase assays where recombinant proteins were used as substrates, we purified those substrates from E.
coli using HisTrap chromatography (Amersham Biosciences).
Recombinant TBK1 was expressed in HEK293 cells and purified by affinity chromatography (data not shown).
Purified TBK1 was incubated with three dilutions of each purified substrate in the presence of 50 μM ATP and 5 μCi [γ-32P]ATP using kinase buffer (40 mM Tris–HCl pH 7.5, 10 mM MgCl2 and 1 mM DTT) at 30°C.
In some instances, the ATP concentration was lowered to 10 μM to increase substrate specificity.
The reaction was terminated by boiling in SDS sample buffer after 5–15 min.
Samples were separated by SDS–PAGE, the gels were dried and analysed by autoradiography.
Peptide array
The peptide array was custom synthesized by Jerini Peptide Technologies (Berlin, Germany).
The peptide sequences that were spotted on the array in triplicate are depicted in Supplementary Figure 5.
Recombinant TBK1 was expressed in HEK293 cells and purified by affinity chromatography (data not shown).
The peptide array was incubated with TBK1 in the presence of 50 μM ATP and 100 μCi [γ-32P]ATP using kinase buffer (40 mM Tris–HCl pH 7.5, 10 mM MgCl2 and 1 mM DTT) at 30°C.
After 30 min, the reaction was terminated by addition of 0.1 M phosphoric acid.
The slides were then washed with 0.1 M phosphoric acid for five times and with deionized water for five times.
After a final wash with methanol, the slides were dried and analysed by autoradiography.
Structural mapping of phosphorylation sites and building of the TBK1 phosphorylation consensus
The structure of DDX3 was downloaded from the Protein Database (PDB) (accession number 2i4i).
Phosphorylation sites were highlighted using PyMOL (www.pymol.org).
The TBK1 phosphorylation consensus was built by manual inspection of the TBK1 target peptides taking into consideration only those amino acids that were seen more than three times (out of 13 positive peptides).
Amino-acid sizes reflect the relative abundance of a given amino acid in that particular position (e.g.
Leu in position +1 was found in 5 out of 13 peptides and has therefore been represented as 5/13 of the size of the S/T peak).
Statistical analysis
Sets of data were analysed using the paired Student's t-test (two-tailed, equal variance).
Statistical significance was assessed based on the P-value: *P<0.05, **P<0.01 and ***P<0.001.
For the siRNA experiments (Figure 2), we calculated the standard deviation and the statistical significance (t-test) of the percentage of reduction based on a minimum of three independent experiments.
For reporter gene assays (Figures 3, 4 and 8), we chose to show the absolute level of induction (versus relative).
The difference in transfection efficiency across different experiments precludes the calculation of standard deviations based on absolute levels.
We therefore decided to show one representative experiment and hereby state that we have replicated each result at least twice, with a statistically significant difference in all three cases.
Supplementary Material
TLR signaling
DDX3 DEAD-box RNA helicase is required for hepatitis C virus RNA replication
Oncogenic role of DDX3 in breast cancer biogenesis
A physical and functional map of the human TNF-alpha/NF-kappa B signal transduction pathway
An efficient tandem affinity purification procedure for interaction proteomics in mammalian cells
DDX3, a DEAD box RNA helicase with tumor growth-suppressive property and transcriptional regulation activity of the p21waf1/cip1 promoter, is a candidate tumor suppressor
The yin and yang of type I interferon activity in bacterial infection
IRF3 mediates a TLR3/TLR4-specific antiviral gene program
IKKepsilon and TBK1 are essential components of the IRF3 signaling pathway
Human DEAD-box ATPase DDX3 shows a relaxed nucleoside substrate specificity
Listeria monocytogenes: a multifaceted model
TRAF3 and its biological function
The roles of two IkappaB kinase-related kinases in lipopolysaccharide and double stranded RNA signaling and viral infection
Triggering the innate antiviral response through IRF-3 activation
Crystal structure of conserved domains 1 and 2 of the human DEAD-box helicase DDX3X in complex with the mononucleotide AMP
IRFs: master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors
Regulation of the type I IFN induction: a current view
A Toll-like receptor-independent antiviral response induced by double-stranded B-form DNA
TANK-binding kinase-1 delineates innate and adaptive immune responses to DNA vaccines
Intracellular NOD-like receptors in host defense and disease
Antiviral signaling through pattern recognition receptors
Signaling to NF-kappaB by Toll-like receptors
IPS-1, an adaptor triggering RIG-I- and Mda5-mediated type I interferon induction
Dead-box proteins: a family affair—active and passive players in RNP-remodeling
Cutting edge: role of TANK-binding kinase 1 and inducible IkappaB kinase in IFN responses against viruses in innate immune cells
IFN-regulatory factor 3-dependent gene expression is defective in Tbk1-deficient mouse embryonic fibroblasts
Innate immunity: the virtues of a nonclonal system of recognition
Cardif is an adaptor protein in the RIG-I antiviral pathway and is targeted by hepatitis C virus
Identification of Ser-386 of interferon regulatory factor 3 as critical target for inducible phosphorylation that determines activation
Functional role of type I and type II interferons in antiviral defense
Type I interferon production enhances susceptibility to Listeria monocytogenes infection
Immune activation of type I IFNs by Listeria monocytogenes occurs independently of TLR4, TLR2, and receptor interacting protein 2 but involves TANK-binding kinase 1
A unified theory of gene expression
An atomic model of the interferon-beta enhanceosome
Interferons, interferon-like cytokines, and their receptors
The inflammasome: a danger sensing complex triggering innate immunity
Innate recognition of viruses
The cell biology of Listeria monocytogenes infection: the intersection of bacterial pathogenesis and cell-mediated immunity
Regulation of antiviral responses by a direct and specific interaction between TRAF3 and Cardif
Listeriolysin O: a phagosome-specific lysin
Identification of the minimal phosphoacceptor site required for in vivo activation of interferon regulatory factor 3 in response to virus and double-stranded RNA
Candidate tumor suppressor DDX3 RNA helicase specifically represses cap-dependent translation by acting as an eIF4E inhibitory protein
Cytoplasmic Listeria monocytogenes stimulates IFN-beta synthesis without requiring the adapter protein MAVS
Recognition of cytosolic DNA activates an IRF3-dependent innate immune response
IFN regulatory factor 3-dependent induction of type I IFNs by intracellular bacteria is mediated by a TLR- and Nod2-independent mechanism
The specific and essential role of MAVS in antiviral innate immune responses
DAI (DLM-1/ZBP1) is a cytosolic DNA sensor and an activator of innate immune response
The transcriptional co-activator protein p100 recruits histone acetyltransferase activity to STAT6 and mediates interaction between the CREB-binding protein and STAT6
Regulation of innate immune responses by DAI (DLM-1/ZBP1) and other DNA-sensing molecules
VISA is an adapter protein required for virus-triggered IFN-beta signaling
Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function
Function of RIG-I-like receptors in antiviral innate immunity
Negative and positive regulation of gene expression by mouse histone deacetylase 1
The DEAD-box helicase DDX3X is a target of TBK1.
(A) Schematic representation of the tandem affinity purification protocol: GS-TAP-tagged TBK1 expressed in RAW264.7 cells was purified using rabbit immunoglobulin G (IgG) agarose and eluted by tobacco etch virus (TEV) protease cleavage.
Next, the remaining complex was purified using streptavidin agarose and eluted by boiling in SDS sample buffer.
(B) The final TAP eluate was separated on an SDS–PAGE and stained by silver staining.
Protein complex composition was analysed by LC-MSMS.
The position of several complex components is depicted next to the region where it was identified.
DDX3X is required for IFN-β induction.
Expression of TBK1 or DDX3X was targeted in RAW264.7 macrophages using specific siRNAs.
Expression levels of TBK1 (A) or DDX3X (B) were monitored by immunoblotting.
Pan-ERK was monitored as a loading control.
(C) siRNA-treated RAW264.7 cells were infected with L.
monocytogenes for 4 h.
Induction of IFN-β mRNA (left panel), RANTES mRNA (middle panel) and TNF-α mRNA (right panel) were measured by quantitative RT–PCR.
(D) Similar cells were transfected with poly(I:C) (left panel), treated with LPS (middle panel) or transfected with poly(dA:dT) (right panel).
Induction of IFN-β mRNA was measured by quantitative RT–PCR.
(E) siRNA-treated RAW264.7 cells were infected with L.
monocytogenes or treated with IFN-β for 4 h.
Induction of IRF7 and Mx2 was measured by quantitative RT–PCR.
Sets of data were analysed using the paired Student's t-test (two-tailed, equal variance).
Statistical significance was assessed based on the P-value: *P<0.05, **P<0.01 and ***P<0.001.
TBK1 has an impact on Rev/DDX3X-dependent nuclear RNA export.
(A) Schematic representation of the Rev reporter construct pDM128.
In the absence of Rev-mediated nuclear mRNA export, the CAT gene is spliced and therefore not translated.
If Rev is present, the mRNA is exported out of the nucleus, splicing is prevented and the CAT mRNA is translated.
(B) HEK293 cells were transfected with the reporter plasmid pDM128 and constant amounts of Rev DDX3X wt or K230A and TBK1 wt or K38M were added as indicated.
CAT expression was measured 24 h post-transfection by ELISA (Roche).
DDX3X and TBK1 led to synergistic activation of the IFN-β promoter.
Cells were transfected with a firefly luciferase reporter plasmid and a Renilla luciferase plasmid.
Firefly luciferase activity was measured after 24 h and normalized to Renilla luciferase activity.
(A) HEK293 cells were transfected with the IFN-β reporter plasmid and 1 μg of plasmids encoding MAVS, DDX3X or DDX3X-K230A and 500 ng of a TBK1 expression plasmid as indicated.
(B) HEK293 cells were transfected with the NF-κB reporter plasmid and 1 μg of plasmids encoding HA-tagged MAVS, DDX3X or DDX3X-K230A and 500 ng of a TBK1 expression plasmid as indicated.
(C) MEFs from wt mice (black bars), TBK1-deficient mice (grey bars) or IRF3-deficient mice (white bars) were transfected with the IFN-β reporter plasmid and 1 μg of plasmids encoding HA-tagged MAVS, DDX3X, TBK1 and TBK1-S172A as indicated.
DDX3X is recruited to the enhanceosome-binding site on the IFN-β promoter.
(A) Schematic representation of the IFN-β promoter.
Enhanceosome-binding site (white box) is flanked by specific primers, whereas a 1 kb upstream region is flanked by control primers.
(B) RAW264.7 macrophages were infected with L.
monocytogenes for 0–3 h.
Quantitative PCR was realized on ChIP samples treated with either an IRF3-specific antiserum (black bar) or an unrelated serum (white bar).
(C) RAW264.7 NTAP(GS)-DDX3X was infected with L.
monocytogenes for 0–3 h.
Quantitative PCR was realized on samples immunoprecipitated with IgG beads with primers specific for either the enhanceosome-binding site (black bar) or the control region (white bar).
Sets of data were analysed using the paired Student's t-test (two-tailed, equal variance).
Statistical significance was assessed based on the P-value: *P<0.05 and **P<0.01.
DDX3X is phosphorylated by TBK1.
(A, B) HEK293 cells were transiently transfected with Myc- or HA-tagged constructs as indicated.
Cells were treated with 10 μM staurosporine 1 h before lysis as indicated.
Cells were lysed 48 h post transfection.
Cell extracts were treated with 5 U CIP for 1 h as indicated.
Cell extracts were analysed by immunoblotting using either anti-Myc (Rockland), anti-HA (Covance Research) or an antiserum against phosphoserine 172 in TBK1.
(C) GST-tagged IRF3, DDX3X and Grb2 were expressed in E.
coli and purified by HisTrap chromatography.
Purity was assessed by Coomassie staining of three different dilutions of each protein (right panel).
TBK1-mediated phosphorylation was assessed by incubating the three potential substrates (IRF3, DDX3X and Grb2) with TBK1 in the presence of [γ-32P]ATP.
Lane 1 contains TBK1 without substrate, whereas lanes 2, 6 and 10 contain the substrates without TBK1 (left panel).
Mapping of the TBK1 phosphorylation site in DDX3X.
(A) Peptides derived from DDX3X or control peptides derived from TBK1 or IRF3 along with alanine mutants were incubated with TBK1 in the presence of [γ-32P]ATP.
Each array contained a total of 82 peptides spotted in triplicate.
(B) Phosphorylation sites obtained from the peptide array (see also text) were used to build a TBK1 phosphorylation consensus sequence.
(C) Putative DDX3X phosphorylation sites were mapped onto the structure of a truncated version of DDX3X.
Phosphorylation-deficient DDX3X mutants fail to synergize with TBK1.
(A) Phosphorylation sites obtained from the peptide array were mutated as follows: Dead-M (S181A, S183A, S240A and S269A), Helic-M (S429A, T438A, S442A, S456A and S520A) or Pan-M (S181A, S183A, S240A, S269A, S429A, T438A, S442A, S456A and S520A).
(B) Mutated DDX3X proteins were purified from E.
coli and analysed in the in vitro kinase assay as described in Figure 6B.
(C) HEK293 cells were transiently transfected with the IFN-β reporter plasmid and MAVS, DDX3X wt, different DDX3X mutants (Dead-M, Helic-M and Pan-M) and TBK1 as indicated.
Reporter activity was quantified as described in Figure 4.Phospho-dependent interactions between NBS1 and MDC1 mediate chromatin retention of the MRN complex at sites of DNA damage
Mammalian cells respond to DNA double-strand breaks (DSBs) by recruiting DNA repair and cell-cycle checkpoint proteins to such sites.
Central to these DNA damage response (DDR) events is the DNA damage mediator protein MDC1.
MDC1 interacts with several DDR proteins, including the MRE11–RAD50–NBS1 (MRN) complex.
Here, we show that MDC1 is phosphorylated on a cluster of conserved repeat motifs by casein kinase 2 (CK2).
Moreover, we establish that this phosphorylation of MDC1 promotes direct, phosphorylation-dependent interactions with NBS1 in a manner that requires the closely apposed FHA and twin BRCT domains in the amino terminus of NBS1.
Finally, we show that these CK2-targeted motifs in MDC1 are required to mediate NBS1 association with chromatin-flanking sites of unrepaired DSBs.
These findings provide a molecular explanation for the MDC1–MRN interaction and yield insights into how MDC1 coordinates the focal assembly and activation of several DDR factors in response to DNA damage.
Introduction
To counter the threat to the genome from DNA-damaging agents, cells use the DNA damage response (DDR): a set of events involving activation of DNA repair mechanisms and cell-cycle ‘checkpoint' signalling (Kastan & Bartek, 2004).
DNA double-strand breaks (DSBs)—the most cytotoxic DNA lesions—activate the protein kinase ataxia telangiectasia mutated (ATM) to phosphorylate the carboxy-terminal tail of histone H2AX in the vicinity of the break (Stucki & Jackson, 2006).
This chromatin modification is crucial for the relocalization of several proteins to sites flanking DSBs, generating ionizing radiation-induced foci (IRIF) that promote efficient repair and sustained DNA damage signalling (Fernandez-Capetillo et al, 2004; Stucki & Jackson, 2006).
A core IRIF component is the mediator of the DNA-damage checkpoint protein 1 (MDC1; Goldberg et al, 2003; Lou et al, 2003; Stewart et al, 2003; Stucki et al, 2005).
MDC1 colocalizes with phospho-H2AX (γH2AX) owing to direct interactions between the C-terminal twin BRCT (BRCT2) domains of MDC1 and the γH2AX phospho-epitope (Stucki et al, 2005).
MDC1 then mediates IRIF formation by other DDR factors such as p53 binding protein 1 (53BP1), breast cancer protein 1 (BRCA1) and the MRE11–RAD50–NBS1 (MRN) complex (Goldberg et al, 2003; Stewart et al, 2003).
We and others have recently shown that a cluster of repeated consensus ATM phosphorylation sites in MDC1 is crucial for the recruitment of BRCA1 and 53BP1 to IRIF, but not for NBS1 focus formation (Huen et al, 2007; Kolas et al, 2007; Mailand et al, 2007).
On phosphorylation, these motifs are bound by the FHA domain of the ubiquitin E3 ligase ring-finger protein 8 (RNF8), which then generates ubiquitin conjugates at sites of DSBs that mediate BRCA1 and 53BP1 recruitment.
Although MDC1 was shown to directly bind to MRN some years ago (Goldberg et al, 2003), the molecular basis of how MDC1 recruits MRN to IRIF has remained elusive.
Here, we show that a cluster of repeated motifs in MDC1, which are phosphorylated by casein kinase 2 (CK2), are bound by MRN through phospho-dependent interactions with NBS1 that require its amino-terminal FHA and BRCT2 domains.
Furthermore, we show that this MDC1–NBS1 interaction is crucial for the targeting and retention of NBS1 on chromatin-flanking DNA DSBs.
Results
Phosphorylated SDTD repeats of MDC1 bind to MRN
A search of published databases for the post-translational modifications of MDC1 identified many phosphorylation sites; however, two studies (Beausoleil et al, 2004; Olsen et al, 2006) identified eight sites within a cluster of short-repeat sequences that shared the consensus motif Ser-Asp-Thr-Asp (SDTD) in which both the serine and threonine are phosphorylated.
This SDTD motif is repeated six times between Ser 218 and Asp 455 in human MDC1, and similar evolutionarily conserved motifs exist in MDC1 orthologues in other vertebrates (Fig 1A).
To investigate whether the SDTD motifs might interact with other DDR components, we generated a synthetic peptide that corresponded to MDC1 residues 325–340 bearing dual phosphorylation on Ser 329 and Thr 331.
This peptide and its unphosphorylated equivalent were coupled to beads and used to retrieve interacting proteins from HeLa cell nuclear extracts.
This approach identified four prominent protein bands on a silver-stained gel that bound in a phosphorylation-dependent manner (Fig 1B).
Mass spectrometric analysis identified three of these as MRE11, RAD50 and NBS1.
We noted that control binding assays with the γH2AX phospho-peptide retrieved MRN and MDC1 from nuclear extracts, whereas the phosphorylated SDTD peptide bound MRN only (Fig 1C).
This indicates that the phosphorylated MDC1 SDTD peptide does not interact with MRN indirectly by bridging contacts with MDC1.
CK2 phosphorylation of MDC1 mediates MRN binding
To study phosphorylation of the MDC1 SDTD region and to investigate its regulation, we raised a rabbit polyclonal antiserum against a synthetic peptide corresponding to one of the MDC1 SDTD repeats phosphorylated on Ser 329 and Thr 331 (supplementary Fig S1A online).
This detected MDC1 isolated from HeLa nuclear extracts, indicating that MDC1 S329/T331 phosphorylation is present in proliferating human cells (Fig 2A, left panels).
The signal from this antibody was lost when MDC1 precipitated with γH2AX peptide-coupled beads was treated with phosphatase, confirming its phospho-specificity (Fig 2A, left panels; detection of RAD50 indicates equal protein loading).
Interestingly, although γ-irradiation of osteosarcoma (U2OS) cells caused little change in the amount of MDC1 S329/T331 phosphorylation, it reduced the overall levels of MDC1 (Fig 2A, right panels).
This suggests that, after DNA damage, S329/T331-phosphorylated MDC1 is more stable than the unmodified protein.
Consistent with these data, S329/T331-phosphorylated MDC1 was detected by immunofluorescence in both control and γ-irradiated cells (Fig 2B; note the absence of staining in cells treated with MDC1 small interfering RNA (siRNA)), confirming that SDTD-phosphorylated MDC1 is present in the absence of damage and also forms IRIF.
Notably, we did not detect any cell-cycle-dependent alteration of MDC1 S329/T331 phosphorylation relative to total MDC1 protein content by using flow cytometry, which suggests that SDTD phosphorylation occurs throughout interphase (supplementary Fig S1B online).
The MDC1 SDTD motif does not conform to consensus sequences targeted by ATM or other DDR kinases.
Instead, both the serine and threonine residues in the SDTD sequence match the CK2 consensus motif of phosphorylated serine/threonine followed by an acidic residue at the +3 position (p[S/T]xxD/E).
Consistent with this, a bacterially expressed protein (glutathione S-transferase (GST)-SDTD2) encompassing two MDC1 SDTD motifs (contained within amino-acid residues 296–340 of MDC1) was efficiently phosphorylated by recombinant CK2 in vitro, as shown by a marked decrease in electrophoretic mobility and detection by anti-MDC1 pS329/pT331 antiserum (Fig 2C, upper and lower panels, respectively).
By contrast, the equivalent protein bearing alanine substitutions of Thr 301 and Thr 331 (GST-SDAD2), while being phosphorylated by CK2 to some degree (presumably on CK2 consensus serine residues retained in the SDAD motifs present in this fusion protein), was detected only weakly with the anti-MDC1 pS329/pT331 antiserum (Fig 2C).
Moreover, although CK2-mediated phosphorylation of the GST-SDTD2 protein allowed it to retrieve MRN from HeLa nuclear extracts, the CK2-treated GST-SDAD2 mutant protein was unable to bind to MRN (Fig 2D).
To investigate whether CK2 phosphorylation of the MDC1 SDTD region promotes MRN binding in vivo, we transiently expressed a haemagglutinin (HA)-tagged MDC1 SDTD fragment coupled to a nuclear localization sequence in U20S cells (HA-SDTD6; supplementary Fig S1C online).
Immunoblotting showed that treating these cells with the CK2 inhibitor 4,5,6,7-tetrabromo-benzimidazole increased the electrophoretic mobility of the MDC1 SDTD fragment and reduced the ability of the fragment to co-immunoprecipitate MRN (Fig 2E).
NBS1 directly binds to CK2-phosphorylated MDC1
Next, we used the recombinant, purified MRN complex in peptide pull-down experiments.
These showed that MRN bound to the phosphorylated but not the unphosphorylated form of the SDTD peptide—confirming the direct nature of the interaction—yet did not bind to either version of the H2AX C-terminal peptide (Fig 3A).
This indicates the specificity of MRN for the phosphorylated MDC1 SDTD motif and also shows that if, as previously reported, NBS1 binds to γH2AX directly (Kobayashi et al, 2002), this interaction must be less stable than MRN binding to the phosphorylated SDTD motif.
Within MRN, only NBS1 contains phosphorylation-specific interaction domains: an N-terminal FHA domain followed directly by a BRCT2 domain (Durocher et al, 2000; Manke et al, 2003; Yu et al, 2003; Becker et al, 2006).
To investigate whether these domains bind to the phosphorylated SDTD motif of MDC1, we expressed an N-terminal fragment of NBS1 containing these motifs (residues 1–343; HA-fNBS1) in rabbit reticulocyte lysates (despite many attempts using a wide range of expression constructs, we were unable to express soluble recombinant versions of the FHA or FHA/BRCT2 domains of NBS1 in bacteria).
We also generated versions of this NBS1 region in which either the FHA domain (fNBS1R28A and fNBS1H45A) or the BRCT2 region (fNBS1K160M) was mutated to abolish phospho-dependent interactions.
These proteins were then tested for binding to an MDC1 fragment encompassing the SDTD repeats (GST-SDTD6) that had or had not been treated with CK2 (Fig 3B).
Significantly, the wild-type NBS1 fragment bound specifically to CK2-phosphorylated—but not unphosphorylated—GST-SDTD6, whereas this binding was abolished by mutation of the FHA or BRCT2 domain of NBS1 (Fig 3B).
Furthermore, CK2-phosphorylated GST-SDTD2 retrieved NBS1 and RAD50 from both MRC5- and NBS1-complemented Nijmegen breakage syndrome (NBS) cell extracts, but did not mediate such interactions when the FHA domain mutant of NBS1 was expressed in NBS cells (Fig 3C; see supplementary Fig S2 online for expression levels).
Taken together, these results indicate that the integrity of both the FHA and BRCT2 domains of NBS1 is needed for binding of NBS1 to the phosphorylated SDTD region of MDC1.
MDC1 SDTD motifs retain MRN at DNA DSB sites
An analysis of the recruitment dynamics of fluorescently tagged proteins to laser-induced DNA damage showed similar recruitment kinetics for NBS1 and MDC1, and highlighted the importance of the FHA domain of NBS in this recruitment (Lukas et al, 2004).
Therefore, we thought that the MDC1–MRN interaction, which depends on the FHA/BRCT2 domains of NBS1, might promote retention of the MRN complex in IRIF.
To test this, we used siRNA to deplete endogenous MDC1 in U2OS cells stably expressing siRNA-resistant wild-type MDC1 (MDC1WT) or an MDC1 mutant without the SDTD region (MDC1SDTDΔ; Fig 4A; supplementary Fig S3A online).
The wild-type MDC1 derivative efficiently co-immunoprecipitated MRN at physiological salt concentrations, whereas only low levels of MRN were recovered in immunoprecipitates of the MDC1SDTDΔ mutant (Fig 4B), consistent with the SDTD region being the principal MRN interaction interface.
This residual interaction with MRN shown by the MDC1SDTDΔ mutant raises the possibility of other MDC1–MRN interactions, and would be consistent with previous findings (Goldberg et al, 2003); however, it is noteworthy that this weaker interaction was undetectable under higher salt conditions (supplementary Fig S3B online).
Moreover, although both MDC1WT and MDC1SDTDΔ efficiently formed IRIF, MDC1WT, but not MDC1SDTDΔ, supported IRIF formation by NBS1 (Fig 4C; see supplementary Fig S3C online for quantification).
By contrast, IRIF formation by 53BP1 and BRCA1 was normal after expression of either form of MDC1 (Fig 4D; supplementary Fig S4A online).
This is consistent with 53BP1 and BRCA1 requiring the RNF8-binding motifs of MDC1 that are still intact in the MDC1SDTDΔ mutant (Fig 4A), and shows that the SDTD region of MDC1 specifically mediates IRIF formation by NBS1 by promoting efficient MDC1–MRN interactions.
To examine further the NBS1 recruitment defect of cells expressing MDC1SDTDΔ, we used the ‘laser scissors' technique to generate DNA DSBs along defined sub-micrometre tracks in live cells (Limoli & Ward, 1993).
In MDC1WT-expressing cells, the extent of NBS1 recruitment closely resembled that of MDC1 (Fig 4Ea).
By contrast, although MDC1SDTDΔ recruitment was indistinguishable from that of the MDC1WT protein, the NBS1 recruitment pattern was altered.
Thus, in MDC1SDTDΔ cells, although NBS1 was recruited to the laser path, its pattern no longer resembled the broad distribution of MDC1; instead, NBS1 was restricted to a discrete micro-focal pattern lying within the broader track of MDC1 staining (Fig 4Eb).
This indicates that SDTD phosphorylation of MDC1 is required to retain NBS1 on γH2AX-coupled chromatin, but that NBS1 might recognize other structures induced by laser microirradiation independently of MDC1.
Consistent with these results, and in line with our finding that the FHA domain of NBS1 is required for MDC1 SDTD binding, similar micro-focal patterns of NBS1 recruitment were observed in MDC1-depleted cells and in cells in which the FHA domain of NBS1 was mutated (Lukas et al, 2004).
Discussion
MDC1 binds to γH2AX flanking unrepaired DNA DSBs, and then acts as a ‘mediator' to recruit other DDR factors to such sites (Stucki & Jackson, 2006).
For example, the ATM-phosphorylated TQXF motifs on MDC1 are recognized by the ubiquitin E3 ligase enzyme RNF8 (Huen et al, 2007; Kolas et al, 2007; Mailand et al, 2007).
RNF8 then generates ubiquitin adducts on histone H2A—and possibly other chromatin components—which lead to IRIF formation by 53BP1 and BRCA1 (Huen et al, 2007; Mailand et al, 2007).
This paper describes how MDC1 facilitates the accumulation and retention of the MRN complex in IRIF, indicating a new mechanism for MDC1 mediator function.
Specifically—and in contrast to the situation for 53BP1 and BRCA1—our work highlights how phospho-dependent interactions between MDC1 and MRN might precede the detection and signalling of DNA damage.
This characteristic of MDC1–MRN binding might therefore allow this complex to be efficiently and rapidly recruited to sites of DNA damage, thus ensuring that the DDR is launched without delay.
By contrast, the ATM- and RNF8-dependent nature of 53BP1 and BRCA1 recruitment means that it is likely to happen more slowly, possibly to ensure that the chromatin alterations and other processes triggered by these proteins are invoked only at the right time and in the correct chromatin context.
This model—in which various factors are recruited with different kinetics and factor dependencies—suggests a more complex, hierarchical and highly regulated DDR than has hitherto been apparent.
Together with data obtained from proteomics screens (Beausoleil et al, 2004; Olsen et al, 2006), our findings indicate that CK2 phosphorylates several sites on MDC1, often at dual phosphorylation sites within the SDTD motif.
These findings are in agreement with previous work showing that CK2 has an impact on various DDR events.
For example, it was previously shown that CK2-dependent phosphorylation of the adaptor/mediator proteins XRCC1 and XRCC4 promotes the recruitment of polynucleotide kinase (PNK), Aprataxin, and Aprataxin- and PNK-like factor (APLF) to sites of chromosomal damage to facilitate the repair of DSBs and DNA single-strand breaks, respectively (Clements et al, 2004; Loizou et al, 2004; Iles et al, 2007).
Interestingly, in these studies, CK2 was found to mediate factor recruitment through direct interactions between the CK2 phosphorylation sites on XRCC1/XRCC4 and the FHA domains of PNK, Aprataxin and APLF.
It is noteworthy that the NBS1 recruitment defects we observed in MDC1SDTDΔ mutant cells are similar to those reported for cells containing NBS1 with mutations in its FHA domain (Lukas et al, 2004), suggesting that the FHA domain of NBS1 binds to MDC1.
Although our data support this idea, the fact that mutations in either the FHA or BRCT2 domain of NBS1 impair the NBS1–MDC1 interaction suggests that both domains are intimately involved in making such contacts.
It might be that these domains are orientated such that they are able to simultaneously bind to both the phosphorylated serine and the phosphorylated threonine of a single SDTD motif, or to two adjacent SDTD motifs in the MDC1 protein.
This dual-interaction mode might provide a greater degree of binding selectivity and discrimination than would be possible by phospho-dependent interactions involving either domain alone.
Alternatively, it is possible that the proximity of the FHA and BRCT2 domains of NBS1 makes them conformationally inter-dependent, so that mutation of one affects the structure and function of the other.
The tightly interconnected FHA/BRCT2 domain architecture is conserved in virtually all known NBS1 counterparts (Becker et al, 2006), pointing to this relationship being of crucial functional importance.
Resolution of the above issues will await structural determinations of the N terminus of NBS1, both alone and in a complex with the phosphorylated MDC1 SDTD motif.
After DNA damage, we have consistently observed stabilization of the CK2-phosphorylated MDC1 relative to the total MDC1 pool, which is downregulated.
This is consistent with a recent study showing 26S proteasome-dependent turnover of the mediator proteins topoisomerase II binding protein 1 (TOPBP1), Claspin, 53BP1 and MDC1 following DNA damage (Zhang et al, 2006).
Whether CK2 phosphorylation itself, or MRN binding to the phosphorylated SDTD motifs, protects a sub-pool of MDC1 from proteolytic degradation remains to be determined.
It is also noteworthy that CK2 phosphorylations on XRCC1 mediate binding to PNK, Aprataxin and APLF (Clements et al, 2004; Loizou et al, 2004; Iles et al, 2007) and, conversely, that FHA domain-containing proteins such as PNK, Aprataxin and APLF are able to bind to CK2 phosphorylation sites on both XRCC1 and XRCC4.
This raises the possibility that the phosphorylated SDTD regions of MDC1 might interact with proteins in addition to NBS1, and that the FHA/BRCT2 region of NBS1 might have more than one physiological target.
Finally, a recent study of MDC1-deficient mouse cells complemented with various MDC1 domain-deletion constructs showed the central repeat region of MDC1 and FHA and BRCT2 domains are important for DSB repair by sister-chromatid recombination (Xie et al, 2007).
This work also showed that MDC1-dependent recruitment of BRCA1 and 53BP1 to IRIF was not required for this repair, or for 53BP1-dependent DSB repair by non-homologous end-joining.
So far, a functional role for the marked redistribution of DDR factors to regions surrounding DSBs has not been defined.
Perhaps future studies using the MDC1 separation-of-function mutants created in this study will help to explain the different and possibly overlapping functional contributions made by recruiting BRCA1, 53BP1 and the MRN complex to γH2AX-associated chromatin.
Methods
See the supplementary information online for details.
Supplementary information is available at EMBO reports online (http://www.emboreports.org).
Note added in proof.
While this work was under revision, two related papers have been published: Melander F, Bekker-Jensen S, Falck J, Bartek J, Mailand N, Lukas J (2008) Phosphorylation of SDT repeats in the MDC1 N terminus triggers retention of NBS1 at the DNA damage-modified chromatin.
J Cell Biol 181: 213–226; Spycher C, Miller ES, Townsend K, Pavic L, Morrice NA, Janscak P, Stewart GS, Stucki M (2008) Constitutive phosphorylation of MDC1 physically links the MRE11–RAD50–NBS1 complex to damaged chromatin.
J Cell Biol 181: 227–240.
Supplementary Material
The authors declare that they have no conflict of interest.
Large-scale characterization of HeLa cell nuclear phosphoproteins
Detection of a tandem BRCT in Nbs1 and Xrs2 with functional implications in the DNA damage response
The ataxia-oculomotor apraxia 1 gene product has a role distinct from ATM and interacts with the DNA strand break repair proteins XRCC1 and XRCC4
The molecular basis of FHA domain:phosphopeptide binding specificity and implications for phospho-dependent signaling mechanisms
H2AX: the histone guardian of the genome
MDC1 is required for the intra-S-phase DNA damage checkpoint
RNF8 transduces the DNA-damage signal via histone ubiquitylation and checkpoint protein assembly
APLF (C2orf13) is a novel human protein involved in the cellular response to chromosomal DNA strand breaks
Cell-cycle checkpoints and cancer
NBS1 localizes to γ-H2AX foci through interaction with the FHA/BRCT domain
Orchestration of the DNA-damage response by the RNF8 ubiquitin ligase
A new method for introducing double-strand breaks into cellular DNA
The protein kinase CK2 facilitates repair of chromosomal DNA single-strand breaks
MDC1 is coupled to activated CHK2 in mammalian DNA damage response pathways
Mdc1 couples DNA double-strand break recognition by Nbs1 with its H2AX-dependent chromatin retention
RNF8 ubiquitylates histones at DNA double-strand breaks and promotes assembly of repair proteins
BRCT repeats as phosphopeptide-binding modules involved in protein targeting
Global, in vivo, and site-specific phosphorylation dynamics in signaling networks
MDC1 is a mediator of the mammalian DNA damage checkpoint
γH2AX and MDC1: anchoring the DNA- damage-response machinery to broken chromosomes
MDC1 directly binds phosphorylated histone H2AX to regulate cellular responses to DNA double-strand breaks
Distinct roles of chromatin-associated proteins MDC1 and 53BP1 in mammalian double-strand break repair
The BRCT domain is a phospho-protein binding domain
A role for the deubiquitinating enzyme USP28 in control of the DNA-damage response
Phosphorylated MDC1 SDTD motifs bind to MRN.
(A) Domain architecture of MDC1 and alignment of the SDTD region from MDC1 proteins in human (hMDC1), chimpanzee (ptMDC1), pig (ssNFBD1), dog (cfMDC1) and mouse (mmMDC1).
RBM indicates RNF8-binding motifs.
Identical and similar residues are boxed in black and grey, respectively.
CK2 consensus sites are underlined, and residues identified as phosphorylated (Beausoleil et al, 2004; Olsen et al, 2006) are numbered (′phosphorylated residues identified in both studies).
(B) Silver-stained SDS–polyacrylamide gel of an SDTD peptide pull-down.
PP, S329 T331 doubly phosphorylated peptide and −, its non-phosphorylated equivalent; B, bead-interacting proteins removed from extracts in a pre-clearing step; M, molecular weight markers.
(C) Immunoblot of SDTD and histone H2AX phosphopeptide-interacting proteins.
INP, input (5% of total); γ, phosphorylated H2AX peptide.
CK2, casein kinase 2; MRN, MRE11–RAD50–NBS1; SDTD, Ser–Asp–Thr–Asp motif.
Casein kinase 2-dependent phosphorylation of MDC1 SDTD motifs induces interaction with MRN.
(A) Left panels: SDTD peptide interacting proteins, isolated from HeLa nuclear extracts (as in Fig 1C), were subjected to phosphatase or mock treatment, and then analysed with the indicated antibodies.
Right panels: whole-cell extracts (WCE) from osteosarcoma (U2OS) cells 1 h after 5 Gy irradiation or control cells were immunoblotted with the indicated antibodies.
(B) At 72 h after control (CNTL) or MDC1-targeting siRNA, control cells or cells incubated for 2 h following 5 Gy of X-rays were processed for immunofluorescence with MDC1 and MDC1 pS329pT331 antibodies.
(C) Immunoblot analysis of purified GST-SDTD2 and GST-SDAD2 after in vitro phosphorylation by recombinant CK2, or mock treatment.
(D) A 250 ng portion of purified GST-SDTD2 or GST-SDAD2 was mock treated or phosphorylated by CK2 in vitro and then incubated with HeLa nuclear extracts.
GST fusions and interacting proteins were captured on glutathione-Sepharose beads and immunoblotted.
(E) Expression plasmids encoding HA-SDTD6 were transfected into U2OS cells.
After 48 h, cells were treated with 4,5,6,7-tetrabromo-benzimidazole (TBB; 75 μM; Sigma-Aldrich (Poole, Dorset, UK), T0826) or dimethyl sulphoxide (DMSO) for 6 h.
Extracts were immunoprecipitated (IP) with monoclonal antibodies against HA or GFP (CNTL) and immunoblotted.
CK2, casein kinase 2; DAPI, 4′,6–diamidino–2–phenylindole; GFP, green fluorescent protein; GST, glutathione S-transferase; HA, haemagglutinin; INP, input; MRN, MRE11–RAD50–NBS1; SDAD, Ser–Asp–Ala–Ser motif; SDTD, Ser–Asp–Thr–Asp motif; siRNA, small interfering RNA.
Binding of MRN to MDC1 phospho-SDTD motifs requires the NBS FHA and BRCT2 domains.
(A) MRN interacts directly with phosphorylated SDTD peptides.
SDTD and H2AX peptide-coupled beads were incubated with 250 ng of recombinant human MRN, and interacting proteins were analysed by western blotting.
INP, input (20%); PP, MDC1 S329/T331 double phosphorylated peptide; γ, phosphorylated H2AX peptide; −, non-phosphorylated equivalent peptides.
(B) A 200 ng portion of CK2-phosphorylated or mock-phosphorylated GST-SDTD6 was incubated in pull-down reactions with 25 μl of in vitro-translated (IVT) HA fusions corresponding to amino-acid residues 1–348 of NBS1 (HA-fNBS1), or analogous proteins bearing point mutations predicted to abolish either FHA (R28A/H45A) or BRCT2 (K160M) phosphorylation-dependent interactions.
INP, input (20%).
(C) A 2.0 mg portion of cell extracts prepared from MRC5, NBS (ILB1) or NBS (ILB1) fibroblasts stably expressing NBS1 or NBS1R28A was incubated with GST-SDAD2 treated as in (B).
GST fusions were retrieved (as in Fig 2C) and interacting proteins were immunoblotted.
INP, input (2.5%).
CK2, casein kinase 2; GST, glutathione S-transferase; HA, haemagglutinin; MRN, MRE11–RAD50–NBS1; SDTD, Ser–Asp–Thr–Asp motif.
The MDC1 SDTD–MRN interaction recruits NBS1 to damaged chromatin.
(A) SDTD deletion relative to the MDC1 domain architecture.
(B) Indicated expression constructs were transfected into human embryonic kidney 293 cells.
After 48 h, extracts were prepared, immunoprecipitated (IP) with GFP antibodies and immunoblotted.
Immunoprecipitations and washes were performed at 150 mM salt; INP, input (5%).
(C,D) Indicated osteosarcoma (U2OS) cell lines were treated with two rounds of control (CNTL) or MDC1-targeting siRNA for 72 h.
Cells were then treated with 5 Gy of X-rays and processed for immunofluorescence 4 h later with MDC1, (C) NBS1 or (D) 53BP1 antibodies (non-irradiated cells are shown in supplementary Fig S4B,C online).
(E) U2OS cells stably expressing siRNA-resistant MDC1wt (a) or MDC1SDTDΔ (b) were treated with siRNA against MDC1 as above, and then the nuclei were subjected to laser micro-irradiation.
After 30 min, cells were pre-extracted with detergent, fixed and immunostained with antibodies against MDC1 and NBS1.
Asterisks indicate resistance to MDC1 siRNA.
GFP, green fluorescent protein; MRN, MRE11–RAD50–NBS1; SDTD, Ser–Asp–Thr–Asp motif; siRNA, small interfering RNA.The GET Complex Mediates Insertion of Tail-Anchored Proteins into the ER Membrane
Summary
Tail-anchored (TA) proteins, defined by the presence of a single C-terminal transmembrane domain (TMD), play critical roles throughout the secretory pathway and in mitochondria, yet the machinery responsible for their proper membrane insertion remains poorly characterized.
Here we show that Get3, the yeast homolog of the TA-interacting factor Asna1/Trc40, specifically recognizes TMDs of TA proteins destined for the secretory pathway.
Get3 recognition represents a key decision step, whose loss can lead to misinsertion of TA proteins into mitochondria.
Get3-TA protein complexes are recruited for endoplasmic reticulum (ER) membrane insertion by the Get1/Get2 receptor.
In vivo, the absence of Get1/Get2 leads to cytosolic aggregation of Get3-TA complexes and broad defects in TA protein biogenesis.
In vitro reconstitution demonstrates that the Get proteins directly mediate insertion of newly synthesized TA proteins into ER membranes.
Thus, the GET complex represents a critical mechanism for ensuring efficient and accurate targeting of TA proteins.
Introduction
The biogenesis of transmembrane proteins presents the cell with several compounding challenges.
Prior to membrane insertion, hydrophobic transmembrane domains (TMDs) are prone to aggregation, and the spontaneous insertion of TMDs across lipid bilayers, even when thermodynamically favored, can be slow.
Moreover, proteins containing TMDs must find their correct target membrane for insertion among the different membrane-surrounded compartments present in eukaryotic cells.
To face these challenges, cells have evolved diverse mechanisms for chaperoning membrane proteins, often from the earliest stages of their biosynthesis on the ribosome to their proper destinations.
Such pathways have been the subject of intense investigations and include the signal recognition particle (SRP)/Sec61 translocon system that imports secretory pathway proteins into the endoplasmic reticulum (ER) (Egea et al., 2005; Rapoport et al., 1999; Wickner and Schekman, 2005) and the transport inner membrane/transport outer membrane (Tim/Tom) translocases that mediate insertion of transmembrane proteins into both mitochondrial membranes (Neupert, 1997; Pfanner and Meijer, 1997).
Far less is known about the machinery responsible for the insertion of an important class of proteins that are anchored to the lipid bilayer by a single TMD located near their C termini.
This topological arrangement allows tail-anchored (TA) proteins to be tethered to internal membranes while presenting their functional N-terminal domains to the cytosol (Borgese et al., 2007; Wattenberg and Lithgow, 2001).
TA proteins are found throughout the secretory pathway, in the nuclear envelope, peroxisomes, and mitochondria.
Within the secretory pathway, TA proteins play diverse roles, such as enabling vesicular traffic (e.g., many of the SNAREs, which mediate fusion of secretory vesicles, are TA proteins [Beilharz et al., 2003]), aiding in protein translocation, and promoting folding or degradation of membrane proteins (Borgese et al., 2007; Wattenberg and Lithgow, 2001).
Secretory pathway TA proteins are first inserted into the ER membrane, and are then sorted to their ultimate destination (Bulbarelli et al., 2002).
In contrast, mitochondrial TA proteins are inserted directly into the mitochondrial membrane, where they facilitate mitochondrial fission, provide key components of the translocation machinery, and act in apoptosis (Borgese et al., 2007; Wattenberg and Lithgow, 2001).
The membrane specificity of TA proteins is largely encoded in their TMDs and flanking regions (Egan et al., 1999).
These signals, however, are not absolute, as some TA proteins, such as the mammalian oncoprotein Bcl2 (Krajewski et al., 1993; Lithgow et al., 1994), are found in both the mitochondria and the ER.
Moreover, it is not well understood how targeting determinants in the TMDs are decoded by cellular machinery (Borgese et al., 2007).
Because of its position near the C terminus, the TMD of TA proteins is occluded by the ribosome until translation is completed.
Thus, TA proteins cannot exploit the classic cotranslational SRP/Sec61 translocation mechanism used by most secretory pathway proteins (Yabal et al., 2003).
Early studies with cell extracts indicated that some TA proteins, such as CytB5, could integrate into membranes without the assistance of specialized machinery (Brambillasca et al., 2006; Rachubinski et al., 1980).
However, most TA proteins, such as the mammalian Sec61β and synaptobrevin, have more hydrophobic TMDs, rendering them reliant on an incompletely characterized, ATP-dependent mechanism (Abell et al., 2007; High and Abell, 2004; Stefanovic and Hegde, 2007; Favaloro et al., 2008).
Recently, biochemical studies identified the mammalian soluble ATPase, Asna1/TRC40, as part of a cytosolic complex that interacts with the newly synthesized TA protein, Sec61β, in vitro (Stefanovic and Hegde, 2007; Favaloro et al., 2008).
This complex can then deliver Sec61β to the surface of ER-derived vesicles (microsomes), where upon it can undergo ATP-dependent membrane insertion.
While these studies have provided critical molecular insights into the ATP-dependent biogenesis of TA proteins, they leave several important questions unaddressed.
First, it is unclear how broad a role the Asna1/TRC40 system plays in vivo.
Indeed, a recent report established that the cytosolic chaperone pair Hsc70/Hsp40 is sufficient to mediate efficient ATP-dependent insertion of Sec61β in vitro (Abell et al., 2007).
Second, the identity of the proteins necessary for recruiting Asna1/TRC40 to the ER is unknown.
Finally, it is unknown how cells ensure proper partitioning of TA proteins between the ER and mitochondria.
Based on a large-scale genetic interaction map of the secretory pathway, we previously suggested that three otherwise unassociated yeast proteins (Mdm39/Get1, Rmd7/Get2, and Arr4/Get3, the yeast homolog of Asna1/TRC40) cooperate to carry out a common function that strongly impacts on trafficking and, accordingly, named them Get1–3 (Golgi ER trafficking 1–3) (Schuldiner et al., 2005).
In agreement with this idea, we and others have found that all three Get proteins physically associate (Auld et al., 2006; Ho et al., 2002; Schuldiner et al., 2005), and that loss of any of the GET genes leads to a pronounced Kar2 secretion phenotype, suggestive of a defect in retrograde Golgi to ER trafficking (Schuldiner et al., 2005).
However, the full range of phenotypes that have now been reported for the respective get deletions are difficult to reconcile with an isolated defect in trafficking.
These include mitochondrial dismorphogenesis (Dimmer et al., 2002) for Δget1 (Δmdm39); defects in DNA replication or damage response (Zewail et al., 2003) and V-type ATPase dysfunction (Sambade et al., 2005) for Δget2 (Δhur2/Δrmd7); sensitivity to toxic metal ions (Shen et al., 2003) and effects on protein degradation machinery (Auld et al., 2006) for Δget3 (Δarr4); and defects in meiotic spore formation (Auld et al., 2006; Enyenihi and Saunders, 2003) for all deletions in GET genes.
Thus, the underlying molecular function(s) of the Get proteins, and the extent to which they are working together to perform a single molecular role, remained unresolved.
Here we show, both in vivo and in vitro, that the GET complex is the machinery responsible for insertion of secretory pathway TA proteins into the ER membrane, and that the reduction in inserted TA proteins can, in turn, explain the wide array of phenotypes observed for deletions in the GET genes.
Results
Get1 and Get2 Form a Membrane Receptor for Get3 on the Face of the ER
We began our functional analysis of the GET complex by exploring how Get1 and Get2 determine the subcellular localization of Get3 (for analysis of the physical and functional relationship between the Get proteins see Figures S1 and S2 available online).
Earlier studies established that Get3, which, unlike Get1 and Get2, is not predicted to have TMDs, is found on the surface of the ER as well as in the cytosol.
Moreover, in the absence of Get1 and/or Get2, Get3 loses its ER localization, and is found both in the cytosol as well as in poorly characterized punctate structures (Auld et al., 2006; Schuldiner et al., 2005).
Here we reveal that, rather than being membrane vesicles, these punctate structures are in fact cytosolic detergent-insoluble aggregates (Figure S3).
We further show, through in vitro experiments with microsomes and proteoliposomes containing Get1 and Get2, that the Get1/Get2 complex is directly responsible for recruiting Get3 to the ER membrane in an ATP-independent manner (Figure 1).
This appears to be the primary role of Get1/2 complex, as, in the absence of Get3, there is no apparent additional cost to deleting Get1/2 (Auld et al., 2006; Schuldiner et al., 2005) (Figure S4).
The fact that Get3 shuttles between the cytosol and the ER suggests that it may deliver substrates to the membrane.
In the context of this model, the formation of aggregates and the exacerbated phenotype found in Δget1/Δget2 cells (Auld et al., 2006; Schuldiner et al., 2005) (Figure S4) would be explained by disruption of the Get3 cycle, leading to sequestration of potential substrates.
Get3 Binds the TA Protein Sed5 and Is Necessary for Its Membrane Targeting
To help identify factors that might be shuttled from the cytosol to the ER by the GET system, we performed a yeast two-hybrid (Y2H) screen for polypeptides that can interact with Get3.
Y2H analysis, which reports on weak interactions occurring within the nucleus of assayed strains, is well suited for identifying Get3 binding proteins, as it can detect transient interactions that are independent of the presence of Get1 and Get2.
We used yeast expressing Get3 as bait to screen a genomic library encoding prey proteins (James et al., 1996).
Physical interactions caused activation of the Gal4-driven HIS3 reporter gene, allowing growth on plates lacking histidine.
The strongest hit from the screen was a fragment of Sed5 (amino acid 197 to the C terminus) (Figure 2A), a TA protein that acts as a SNARE in vesicular traffic within the Golgi and between the Golgi and the ER (Hardwick and Pelham, 1992).
The Get3-Sed5 interaction was dependent on the presence of the C-terminal TMD (Figure 2A).
We next examined whether Get3, as part of the GET complex, plays a role in recruiting newly synthesized Sed5 in the cytosol and inserting it into membranes.
We visualized the subcellular localization of Sed5 with an N-terminal fusion protein with GFP (GFP-Sed5) (Weinberger et al., 2005).
N-terminal GFP fusion was compatible with the correct targeting of Sed5 to the Golgi in control cells (Banfield et al., 1994; Weinberger et al., 2005) (Figure 2B).
Deletion of Get3 led to a large pool of soluble protein and a corresponding decrease in Golgi-like puncta containing Sed5 (Figure 2B).
In a Δget1/Δget2 background, this defect was more pronounced; there was only modest Golgi staining and, instead, we observed cytosolic fluorescence and a few large punctate structures that were distinct from the Golgi, as visualized by Anp1-GFP staining (Figure 2B).
Red fluorescent protein (RFP) fused to Get3 (Get3-tdRFP) and GFP-Sed5 colocalized in these punctate structures (Figure 2C).
Thus, in the absence of the Get proteins, a substantial fraction of Sed5 remains in the cytosol.
Consistent with this, subcellular fractionation experiments indicate that deletion of the GET genes leads to reduced levels of endogenous untagged Sed5 in membranes, while not interfering with membrane association of the Golgi protein Emp47 or the ER protein Sec61 (Figure 2D).
Decreased Sed5 SNARE activity in vesicles traveling between the Golgi and ER could slow down retrograde traffic and reduce the efficiency of cellular retrieval mechanisms of ER resident proteins (Hardwick and Pelham, 1992; Yamaguchi et al., 2002).
We therefore tested whether reduced Sed5 function could explain the Kar2 secretion phenotype observed in the get mutants.
Consistent with this hypothesis, lowering protein levels of the essential Sed5 protein, by using a repressible tetO7 promoter (Mnaimneh et al., 2004), caused Kar2 secretion at levels that were similar to those observed in deletions of GET complex members (Figure 2E).
Moreover, overexpression of Sed5, presumably by allowing sufficient Sed5 to insert into membranes by alternate, potentially spontaneous TA-insertion pathways (see Discussion), suppressed the Kar2 secretion defect in a triple-deletion strain that has no GET complex members (Figure 2F).
We therefore conclude that the GET complex plays a major role in the biogenesis of the TA protein Sed5.
In addition, the Kar2 secretion phenotype of these cells could be explained by the reduced levels of Sed5 in the membranes of get deletion mutants.
The GET Complex Plays a Broad Role in Insertion of TA Proteins into Membranes
The role of the GET complex as a specific chaperone system for the TA protein Sed5 cannot explain the diversity of the phenotypes displayed by deletions of GET genes.
The recent finding that the mammalian Get3 homolog Asna1 was involved in the insertion of in vitro synthesized Sec61β (Stefanovic and Hegde, 2007; Favaloro et al., 2008) suggests that the GET complex has a broader role in TA protein biogenesis.
Consistent with this idea, by a directed Y2H approach, we detected physical interactions between Get3 and several additional secretory pathway TA proteins, including the SNAREs Tlg2 and Sec22 and the peroxisomal TA protein Pex15.
These interactions, as observed for Sed5, were dependent on the presence of the C-terminal TMD (Figure 3A).
Although these data suggest that Get3 can specifically recognize a range of C-terminally located TMDs, it appears to have some selectivity, as we could not detect physical interactions for the mitochondrial TA protein Fis1 (Figure 3A), which, like other mitochondrial TA proteins, has a shorter, more hydrophilic TMD than secretory pathway TA proteins (Beilharz et al., 2003; Borgese et al., 2001, 2007).
To test whether the observed physical interactions reflect an in vivo role for the GET complex in the biogenesis of secretory pathway TA proteins, we looked at the effect of loss of the Get1/Get2 receptor on the subcellular localization of a functionally diverse range of TA proteins.
We focused predominantly on ER-localized TA proteins, as interpretation of effects on their localization is not complicated by trafficking defects seen in get mutant strains.
Accordingly, we expressed N-terminal fluorescent protein fusions to Sbh1, Sbh2—the yeast homolog of Sec61β shown to interact with Asna1 (Stefanovic and Hegde, 2007)—Scs2, and Ysy6.
As observed for Sed5, localization of these TA proteins was normal in control cells, but was altered in a Δget1/Δget2 background (Figure 3B).
During logarithmic growth, we could observe both the presence of large puncta (that colocalize with Get3) and also proper ER localization, (Figure S5).
Following the diauxic shift, which occurs as cells exit log phase, we observed a more pronounced defect.
In most cells, the majority of the protein was either cytosolic or in one or two large puncta that also contained Get3 (Figure S6).
These defects are specific for TA proteins entering the secretory pathway, as the two mitochondrial TA proteins examined (Fis1 and Tom22) properly localized in both the control and a Δget1/Δget2 background (Figure 3C).
Loss of the GET Complex Leads to Mislocalization of a Subset of TA Proteins
The finding that Get3 is able to distinguish between TA proteins destined for the secretory pathway and those destined to mitochondria suggests that, in addition to increasing the efficiency of TA protein membrane insertion, the GET complex helps ensure that TA proteins accurately find their destination membrane.
Consistent with this idea, we observed that, when overexpressed, a subset of secretory pathway TA proteins mislocalize to mitochondria in Δget1/Δget2 strains (an example of the TA protein Ubc6 is given in Figure 4A).
This effect was particularly pronounced for Pex15.
In wild-type (WT) cells, this protein is thought to be first inserted into the ER, and then transported to the peroxisome via Pex19 (Elgersma et al., 1997; Hoepfner et al., 2005; Tam et al., 2005) (supported by data in Figure S7).
However, in Δget1/Δget2 cells, Pex15 initially formed cytosolic aggregates (Figure S8), but, after extended overexpression, began to appear in mitochondrial membranes (Figure 4B).
This suggests that, once Pex15 saturated Get3, it could insert inappropriately into mitochondrial membranes.
Indeed, we observed that, in the absence of Get3, this lag phase is shortened dramatically, and Pex15 is found in the mitochondria at much earlier time points (Figure 4B).
Hence, in addition to increasing the efficiency of insertion of TA proteins, recognition by Get3 represents a key decision step in defining the membrane destination of a TA protein, thus overcoming the intrinsic potential for TA proteins to spontaneously insert into a wide range of membranes.
Loss of TA Proteins Recapitulates the Pleiotropic GET Phenotypes
The diversity of TA protein functions suggests that the pleiotropic effects associated with loss of the GET complex might be a secondary consequence of TA protein mislocalization.
To test this idea, we assembled a library of strains carrying mutant alleles for each of the predicted 55 yeast TA proteins (Beilharz et al., 2003) (including the six TA proteins localized to mitochondria), and plated these strains on various conditions for which we observed sensitivity in the get deletion strains.
The library consisted of 43 deletion strains (Giaever et al., 2002) for nonessential TA proteins and 12 hypomorphic alleles of the essential ones.
Hypomorphic alleles were made by using the decreased abundance by mRNA perturbation (DAmP) method (Schuldiner et al., 2005), which typically results in ∼5- to 10-fold decrease in levels of the endogenous protein.
We found that loss of a large number of TA proteins (Δpep12, Δtlg2, Δsec22, Δvam3, Δscs2, Δsso2, Δgos1, and bos1-DAmP) caused pronounced copper sensitivity (see Figure 5 for the most sensitive strains).
This is consistent with an important role of the late secretory pathway in cellular copper homeostasis (Labbe and Thiele, 1999).
More generally, individual TA protein mutant strains were sensitive to only a subset of the conditions.
However, for every condition tested, we found a subset of TA protein deletions/depletions that fully recapitulate the sensitivities found in the Δget1 and Δget2 strain (Figure 5 shows the strains with strongest sensitivities).
Thus, a broad defect in secretory pathway TA protein insertion could fully account for the diverse phenotypes observed upon loss of the Get1/Get2 receptor.
In Vitro Evidence that the GET Complex Directly Mediates Insertion of Newly Synthesized TA Proteins into the ER membrane
The above studies establish that the Get proteins play a critical role in the biosynthesis and proper localization of a wide range of TA proteins.
Given the pleiotropy of phenotypes displayed by the get deletants, we wished to assess whether the Get proteins are directly required for insertion of TA proteins into the ER membrane.
We therefore developed an in vitro system for studying this process, which takes advantage of our ability to prepare cytosol extracts and microsomes from get mutant yeast strains.
To monitor membrane insertion, we engineered a glycosylation site after the TA sequence of each substrate examined.
Following translocation, this site is expected to gain access to the glycosylation machinery in the lumen of the ER (Borgese et al.
[2001] and Figure S9) and, as such, serve as a proxy for translocation.
Indeed, when we combined cytosol and microsomes from WT cells, we observed efficient translocation of preproalpha factor (a canonical Sec61 substrate) (Figure 6A) and of the three secretory pathway TA proteins that we tested: Sed5 (Figure 6B), Sec22 (Figures 6C and 6D), and Ysy6 (Figure S9).
In contrast, we did not detect any translocation of the mitochondrial TA protein Fis1 (Figure 6B), indicating that our in vitro system faithfully recapitulates the target membrane specificity of TA protein insertion.
To evaluate the role of the Get proteins in TA protein insertion, we prepared extracts from a Δget3 strain and microsomes from a Δget1/2 strain.
Strikingly, these mutant extracts and microsomes were defective for insertion of TA proteins (Figures 6B–6D and S9), while being fully proficient in supporting the translocation of preproalpha factor (Figure 6A and data not shown).
Addition of recombinant Get3 to Δget3 extracts during (Figure 6C), but not after (data not shown), translation allowed for robust insertion, thus demonstrating that this defect is a proximal consequence of not having Get3 in the in vitro system, and not due to altered cellular physiology in Δget3 strains.
Get3 also appears to be limiting in our WT extracts, as we saw enhanced translocation when recombinant Get3 was added (Figures 6C and S9).
Critically, this Get3-mediated insertion is completely dependent on the presence of the Get1/2 complex in the microsomes (Figures 6D and S9), providing further evidence that these three proteins cooperate to carry out insertion of TA proteins.
Taken together, these data establish that the GET system is directly responsible for mediating insertion of newly synthesized TA proteins into the ER membrane.
Discussion
The present study defines a pathway by which cells ensure the efficient and accurate biogenesis of TA proteins destined for the secretory pathway.
The soluble cytosolic ATPase, Get3, specifically queries newly synthesized proteins for the presence of C-terminally localized hydrophobic domains.
Get1 and Get2 then serve as an ER membrane receptor, which recruits the Get3-TA complex, thereby promoting the proper insertion of TA proteins into the ER (Figure 7A).
Once inserted, TA proteins can then be routed to their ultimate destination within the secretory pathway.
In the absence of the heteromeric Get1/Get2 receptor, TA proteins bound to Get3 fail to reach ER membranes, and are instead trapped in large cytosolic aggregates (Figure 7B).
This leads to a broad depletion of TA proteins, which in turn can account for the otherwise confusing array of phenotypes associated with loss of Get proteins.
Binding to Get3 is also a decisive step in the insertion pathway, as in its absence, secretory pathway TA proteins may insert into mitochondrial membranes (Figure 7C).
The finding that the GET pathway is not essential for yeast viability provides in vivo support to in vitro studies that had suggested additional mechanisms by which TA proteins can find their destination membranes (Rabu and High, 2007).
Nonetheless, several considerations suggest that the GET pathway is the major route used to target a broad range of TA proteins to the secretory pathway.
First, our Y2H analysis indicates that Get3 can bind multiple secretory pathway TA proteins in a TMD-dependent manner.
Second, for all secretory pathway TA proteins examined, the interaction with Get3 caused sequestration of the TA proteins into cellular aggregates in the absence of Get1 and Get2.
This suggests that, when Get3 is present, most of the natural flux of TA proteins flows through the GET pathway.
Indeed, yeast fail to grow when Get3 is overexpressed in the absence of Get1 and Get2 (Figure S4).
Third, deletion of Get3, which would eliminate the GET pathway without actively preventing TA proteins from utilizing alternate pathways by trapping them in nonproductive Get3 complexes, still leads to diverse cellular defects.
Finally, in vitro reconstitution experiments directly establish that Get3 cooperates with the Get1/2 complex in mediating the insertion of newly synthesized TA proteins.
Thus the ability of cells to survive in the complete absence of the Get proteins may be analogous to the viability of yeast missing the SRP, which is made possible by the existence of alternate pathways for insertion of the numerous secreted and membrane-bound proteins that normally utilize this machinery (Ogg et al., 1992).
Possible alternate routes for TA protein biogenesis that have been suggested by in vitro studies include spontaneous insertion, which occurs efficiently for some TA proteins, such as CytB5 (Brambillasca et al., 2006).
In addition, purified Hsc70/Hsp40 can promote the ATP-dependent (Abell et al., 2007) and SRP the GTP-dependent insertion of other TA proteins, such as Sec61β, (Abell et al., 2004).
Such back-up systems, however, would lack the strong membrane specificity conferred by the ER localization of the Get1/2 complex, as well as the preferential binding of Get3 to TA proteins destined to the secretory pathway.
The potential importance of such specificity is illustrated by the observation that some TA proteins, including Pex15 and Ubc6, mislocalize to the mitochondria when the GET system is impaired.
This argues that, shortly after synthesis, Get3 competes with other factors (possibly Hsc70 and/or components that play an analogous role to Get3 in the targeting of mitochondrial TA proteins) for TMD binding, and that Get3 recognition commits the TA proteins to their subsequent insertion into ER membrane.
It remains to be determined whether a dedicated protein machinery exists that ensures the accurate targeting of mitochondrial TA proteins, or whether the shorter, more hydrophilic nature of their TMDs prevents Get3 binding, thereby allowing for efficient, spontaneous insertion into the mitochondria.
The interaction between Get3 and a TA protein substrate may thus represent a critical and potentially regulated decision step for establishing the destination target of TA proteins.
Regulation could globally alter Get3 function or specifically affect the interaction between Get3 and target TA proteins.
Along these lines, we have recently found that the function of Get3 is modulated by its redox state (our unpublished data and Metz et al.
[2006]).
In addition, Get3 is transcriptionally upregulated under both cytosolic (Auld et al., 2006) and ER (Travers et al., 2000) stress conditions.
It has also been found that many TA proteins are palmitoylated (Roth et al., 2006) or phosphorylated (such as for Sed5 [Weinberger et al., 2005]) on residues that are immediately adjacent to the TMD.
Such modifications could modulate Get3 recognition by creating negatively charged flanking regions or by altering the hydrophobicity of the TMD, thereby enabling the coordinated regulation of subclasses of TA proteins and altering the physiology of the cell.
While the present studies focused on TA biogenesis in yeast, recent observations suggest that the GET pathway plays an essential role in TA biogenesis in higher eukaryotes.
Biochemical studies revealed that the mammalian Get3 homolog, Asna1/TRC40, binds the TA protein, Sec61β, and facilitates its posttranslational insertion into ER membranes (Stefanovic and Hegde, 2007; Favaloro et al., 2008).
An in vivo role of Asna1 in TA biosynthesis in metazoans is suggested by the impaired capacity for insulin secretion in Caenorhabditis elegans mutants of asna1 (Kao et al., 2007).
In light of our findings, an attractive hypothesis is that impaired insulin secretion results from compromised biogenesis of one or more of the SNARE TA proteins.
The broader importance of the GET pathway is underscored by the finding that complete loss of ASNA1 causes early embryonic lethality in mice (Mukhopadhyay et al., 2006) and arrested growth at the L1 stage in C.
elegans (Kao et al., 2007).
The molecular identity of the Get3 ER receptor in metazoans remains to be established.
However, we find that Ysy6 translated in rabbit reticulocyte extracts inserts into yeast microsomes in a Get1/2-dependent manner, suggesting that the GET pathway is highly conserved (data not shown).
Consistent with this, PSI-BLAST analysis identifies the WRB protein as an excellent and ubiquitously expressed candidate for a Get1 ortholog.
In summary, the GET complex in yeast and likely metazoans constitutes the major machinery necessary for membrane selective, and ATP-dependent insertion of TA proteins.
This finding should now enable mechanistic studies to explore central questions, including how the GET system selects substrate and exploits ATP hydrolysis to overcome the energetic barriers to insertion of transmembrane proteins into lipid bilayers.
Experimental Procedures
Strains and Media
Due to a high rate of reversion, all deletions in GET genes were constructed by sporulating from a heterozygous diploid carrying deletions in all three genes (his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 LYS2/LYS2 MET15/met15Δ0 ura3Δ0/ura3Δ0 can1Δ::STE2pr-spHIS5/CAN1 lyp1Δ::STE3pr-LEU2/LYP1 cyh2/CYH2 GET1/Δget1::cgURA3 GET2/Δget2::NATr GET3/Δ::Kanr).
Following sporulation, single-, double-, and triple-deletion strains of the correct genotype were chosen.
For Figure 5, all strains were chosen to be Δmet15 to be isogenic with deletion strains taken from the Yeast Consortium Deletion Library (Giaever et al., 2002) or made by the DAmP method (Schuldiner et al., 2005).
Deletion constructs used were pFA6-NAT and pFA6-Kan (Longtine et al., 1998) or pCG-URA (Kitada et al., 1995).
Galactose (GAL) inducible strains were made on the same background as the deletions, only pFA6-Kanr-GALp or pFA6- Kanr-pGAL-GFP cassette was used (Longtine et al., 1998).
C-terminally tagged Get3-GFP::His, Anp1-RFP::Kanr, and Pex3-RFP::Kanr were taken from the whole genome GFP tag library (Huh et al., 2003).
The tetO7-SED5 strain was picked from the essential gene promoter shut-off collection (Mnaimneh et al., 2004).
All N-terminal-tagged proteins with mCherry were created with a pFA6-based vector (Kind gift from David Breslow, University of California, San Francisco), carrying a URA3-TEF2 promoter-mCherry, and were integrated into the gene by one-step PCR-based homologous recombination, with appropriate primers that also introduced an N-terminal linker (GDGAGL) between the mCherry and the proteins.
pRS315-GFPSed5 was a kind gift of Anne Spang (Weinberger et al., 2005).
p416MET25-Get3-tdRFP was constructed by fusing the tdRFP (Campbell et al., 2002) open reading frame to the 3′ end of the GET3 open reading frame via an engineered NotI site coding for three alanines.
For colocalization purposes, a mitochondrial targeting sequence containing RFP was used as a mitochondrial marker (kind gift of Jodi Nunnari, University of California, Davis).
For the strong overexpression plasmid employed in Figure 1H, SED5 was cloned into the 2 μm plasmid BFGIII under the control of its own promoter.
Plates used for drug sensitivity assays were: SD + 100 mM HU (Sigma), SD + 1μg/ml tunicamycin (Sigma), SD + 100 μg/ml hygromycin (Sigma), and SD + 1mM CuSO4.
YPD plates were used for heat sensitivity assays at 39°C.
For induction of the GAL, promoter cells were grown in YP + 2%Galactose.
Microsome Binding Experiments
Microsomes were isolated from Δget1/2/3 and Δget3 yeast strains as previously described (Wuestehube and Schekman, 1992), then resuspended in reaction buffer (20 mM HEPES/KOH, pH 6.8, 5 mM MgAc2, 150 mM KAc, 250 mM sorbitol).
A volume of 10 μl of microsomes were mixed with 1 μl of 2 μM Get3 purified from Escherichia coli (Metz et al., 2006), 0.5 μl 100 mM glutathione (Sigma), 1.25 μl 100 mM batho cuproine disulfonic acid (BCS; SERVA), 1 μl 100 mM ATP (Sigma), and 14 μl of 2× ATPase buffer (200 mM HEPES/KOH, 20 mM MgCl2, 40% glycerol, pH 7.0), and were incubated at 30°C for 1 hr.
After incubation, samples were immediately mixed with 490 μl of 50% Optiprep (PROGEN Biotechnik GmbH) solution in the reaction buffer, placed in 2 ml ultracentrifugation tubes, and overlayed with 1160 μl 40% Optiprep solution in reaction buffer, and, finally, with 450 μl of the reaction buffer.
Samples were centrifuged at 166,000 × g for 3 hr at 4°C.
After centrifugation, four fractions were collected: (1) 630 μl; (2) 430 μl; (3) 430 μl; (4) 640 μl.
All fractions were precipitated with 50% TCA.
Pellets were washed twice with 500 μl cold acetone and dried at 37°C for 1–5 min.
Final pellets were resuspended in 1× SDS-PAGE sample buffer.
Purification of GET Components
Get3 was purified as previously described (Metz et al., 2006).
Epitope-tagged versions of Get1 and Get2 were copurified from yeast (see Experimental Procedures in the Supplemental Data).
Liposome Binding Experiments
Proteoliposomes were prepared as described previously (Denic and Weissman, 2007) and incubated with recombinant Get3.
as described above in Microsome Binding Experiments.
Y2H System
The Y2H system was performed as previously described (Metz et al., 2006).
For more details see Experimental Procedures in the Supplemental Data.
Fluorescence Microscopy
For Figure 2C, microscopy was performed with a Leica DM IRE2 microscope (Leica Microsystems, Wetzlar, Germany).
For Figure 2B, a DeltaVison restoration microscope was employed.
Raw images were deconvolved with the additive algorithm of Softworx software.
For live cell imaging, yeast were incubated in synthetic complete medium at room temperature.
Fixed yeast cells were mounted in ProLong Gold antifade reagent with DAPI (Invitrogen).
For Figures 3 and 4, microscopy was performed in the UCSF Nikon Imaging Center with a Yokogawa CSU-22 spinning disc confocal on a Nikon TE2000 microscope.
For more detailed information see Experimental Procedures in the Supplemental Data.
Crude Fractionation
OD600 units of 25–50 were harvested from log-phase cells growing in YPAD medium, washed once in water, and resuspended in 1 ml buffer (20 mM HEPES/KOH, pH 7.3, 100 mM KCl, 1 mM glutathione, complete protease inhibitors, phosphatase inhibitors [Roche], 1 mM EDTA, 1 mM EGTA, 3 mM BCS, 1 mM PMSF).
Cells were broken by bead beating with 800 μl glass beads for 10 min.
Homogenates were cleared at 2000 rpm in a microcentrifuge and the supernatant (input) was subjected to two sequential centrifugation steps (13,000 rpm in a microcentrifuge [Heavy membranes] and 40,000 rpm in a TLA45 rotor in a tabletop ultracentrifuge [Light membranes]).
Pellets from both steps were resuspended in 250 μl (Heavy) or 50 μl (Light) of the same buffer as above.
Equal protein concentrations of the collected fractions and the remaining supernatant (Other) were loaded, resolved by SDS-PAGE, transferred to a nitrocellulose membrane, and analyzed by immunoblotting with antisera against Sed5, Sec61, or Emp47.
Kar2 Secretion Assays
Kar2 secretion assays were performed as previously described (Schuldiner et al., 2005).
For more details see Experimental Procedures in the Supplemental Data.
In Vitro Transcription
mRNAs were prepared with the mMessage mMachine kit (with cap analog, either SP6- or T7-driven, as appropriate) from Ambion.
Alpha factor mRNA was transcribed from pDJ100 (Garcia et al., 1991).
For other messages, template DNA was derived from PCR products amplified with a 5′ primer containing the T7 promoter/alpha factor 5′ UTR/kozak sequence/start codon and 5′ region of homology, and a 3′ primer containing, in antiparallel order, the 3′ region of homology preceding the stop codon/opsin tag/alpha factor 3′ UTR/polyA tail.
Primer sequences are available upon request from the corresponding authors.
In Vitro Translation and Translocation
Yeast translational extracts were prepared from cells grown to OD600 1–2 in YPD.
Cells were washed and resuspended in 1 ml of lysis buffer (100 mM KOAc, 2 mM Mg(OAc)2, 2 mM DTT, 20 mM HEPES-KOH, pH 7.4, complete protease inhibitors from Roche) for every 6 g of dry cell pellet.
The cell slurry was frozen in liquid nitrogen and lysed by bead beating.
The thawed lysates were spun in an SS34 rotor at 10,000 × g for 10 min.
The low-speed supernatant was then spun in a TLA110 rotor at 49,000 rpm for 30 min.
The high-speed supernatant was collected (avoiding the very top and bottom layers) and passed over a 5 × 5 ml (attached in series) HiTrap desalting column (GE Healthcare) equilibrated in lysis buffer with 14% glycerol.
OD280 fractions >40 were collected, pooled, and stored as frozen aliquots at −80°C.
Prior to use, extracts were treated with micrococcal nuclease (Amersham or NEB) to remove any endogenous mRNAs, as described previously (Garcia et al., 1991).
Translation reactions contained 9.5 μl of nuclease-treated extract, 2.5 μl of 6× mix (132 mM HEPES-KOH, pH 7.4, 720 mM KOAc, 9 mM Mg(OAc)2, 4.5 mM ATP, 0.6 mM GTP, 150 mM creatine phosphate [Roche], 0.24 mM of each amino acid except methionine [Promega], 10.2 mM DTT, 0.5 μl of creatine phosphokinase [10 mg/ml in 50% glycerol; Roche], 0.5 μl RNasin [Promega], 1 μl S35-labeled methionine [ARC; >1000 Ci/mmol]).
Unless indicated otherwise, reactions were programmed with 1 μl of mRNA (0.1–1.0 μg) and incubated at room temperature for 1 hr.
Further translation was stopped by addition of cycloheximide (1 mM).
Microsomes (0.06 OD280) were then added and translocation allowed to proceed for an additional 30 min at room temperature.
RNAs were digested with an RNase cocktail (Ambion).
Finally, loading buffer was added and the translation products were analyzed by SDS-PAGE followed by phosphorimager analysis.
Preperation of Translocation-Competent, ER-Derived Microsomes
Preperation of translocation-competent, ER-derived microsomes was performed as previously described (Brodsky, 2005).
For more details see Experimental Procedures in the Supplemental Data.
Immunoprecipitation and EndoH for In Vitro Reconstitution Experiments
Immunoprecipitation and EndoH for in vitro reconstitution experiments were performed as previously described (Stefanovic and Hegde, 2007).
For more details see Experimental Procedures in the Supplemental Data.
References
Signal recognition particle mediates post-translational targeting in eukaryotes
Post-translational integration of tail-anchored proteins is facilitated by defined molecular chaperones
The conserved ATPase Get3/Arr4 modulates the activity of membrane-associated proteins in Saccharomyces cerevisiae
Localization of Sed5, a putative vesicle targeting molecule, to the cis-Golgi network involves both its transmembrane and cytoplasmic domains
Bipartite signals mediate subcellular targeting of tail-anchored membrane proteins in Saccharomyces cerevisiae
Targeting of a tail-anchored protein to endoplasmic reticulum and mitochondrial outer membrane by independent but competing pathways
How tails guide tail-anchored proteins to their destinations
Unassisted translocation of large polypeptide domains across phospholipid bilayers
An in vitro assay for the selective endoplasmic reticulum associated degradation of an unglycosylated secreted protein
Trafficking of tail-anchored proteins: transport from the endoplasmic reticulum to the plasma membrane and sorting between surface domains in polarised epithelial cells
A monomeric red fluorescent protein
A molecular caliper mechanism for determining very long-chain fatty acid length
Genetic basis of mitochondrial function and morphology in Saccharomyces cerevisiae
Targeting of tail-anchored proteins to yeast mitochondria in vivo
Targeting proteins to membranes: structure of the signal recognition particle
Overexpression of Pex15p, a phosphorylated peroxisomal integral membrane protein required for peroxisome assembly in S.
cerevisiae, causes proliferation of the endoplasmic reticulum membrane.
Large-scale functional genomic analysis of sporulation and meiosis in Saccharomyces cerevisiae
Distinct targeting pathways for the membrane insertion of tail-anchored (TA) proteins
In vitro protein translocation across microsomal membranes of Saccharomyces cerevisiae
Functional profiling of the Saccharomyces cerevisiae genome
SED5 encodes a 39-kD integral membrane protein required for vesicular transport between the ER and the Golgi complex
Tail-anchored protein biosynthesis at the endoplasmic reticulum: the same but different
Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry
Contribution of the endoplasmic reticulum to peroxisome formation
Global analysis of protein localization in budding yeast
Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast
ASNA-1 positively regulates insulin secretion in C.
elegans and mammalian cells.
Cloning of the Candida glabrata TRP1 and HIS3 genes, and construction of their disruptant strains by sequential integrative transformation
Investigation of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes
Pipes and wiring: the regulation of copper uptake and distribution in yeast
The protein product of the oncogene bcl-2 is a component of the nuclear envelope, the endoplasmic reticulum, and the outer mitochondrial membrane
Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae
The yeast Arr4p ATPase binds the chloride transporter Gef1p when copper is available in the cytosol
Exploration of essential gene functions via titratable promoter alleles
Targeted disruption of the mouse Asna1 gene results in embryonic lethality
Protein import into mitochondria
Signal recognition particle receptor is important for cell growth and protein secretion in Saccharomyces cerevisiae
The Tom and Tim machine
Membrane protein chaperones: a new twist in the tail?
Synthesis of rat liver microsomal cytochrome b5 by free ribosomes
Posttranslational protein translocation across the membrane of the endoplasmic reticulum
Global analysis of protein palmitoylation in yeast
A genomic screen for yeast vacuolar membrane ATPase mutants
Exploration of the function and organization of the yeast early secretory pathway through an epistatic miniarray profile
The Saccharomyces cerevisiae Arr4p is involved in metal and heat tolerance
Identification of a targeting factor for posttranslational membrane protein insertion into the ER
Pex3p initiates the formation of a preperoxisomal compartment from a subdomain of the endoplasmic reticulum in Saccharomyces cerevisiae
Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation
Targeting of C-terminal (tail)-anchored proteins: understanding how cytoplasmic activities are anchored to intracellular membranes
Control of Golgi morphology and function by Sed5 t-SNARE phosphorylation
Protein translocation across biological membranes
Reconstitution of transport from endoplasmic reticulum to Golgi complex using endoplasmic reticulum-enriched membrane fraction from yeast
Translocation of the C terminus of a tail-anchored protein across the endoplasmic reticulum membrane in yeast mutants defective in signal peptide-driven translocation
Sly1 binds to Golgi and ER syntaxins via a conserved N-terminal peptide motif
Novel functions of the phosphatidylinositol metabolic pathway discovered by a chemical genomics screen with wortmannin
Supplemental Data
Get1 and Get2 Act as a Membrane Receptor for the Soluble Get3
(A) Western blots with αGet3 showing binding of recombinant Get3 ATPase to microsomes prepared from Δget3 or Δget1/2/3 strains in the presence or absence of ATP.
Shown are Optiprep gradient fractions, which separate microsomes from unbound protein.
(B) Western blots with αGet3 or αPhs1 showing binding of recombinant Get3 to proteoliposomes reconstituted with either Phs1 as a control protein (−Get1/−Get2+PHS1) or purified Get1-PC and Get2-HA (+Get1/+Get2).
Shown are optiprep gradient fractions as above.
Get3 Binds to Sed5 and Is Important for Its Biogenesis In Vivo
(A) Yeast two-hybrid assay with Get3 as bait and Sed5197–340 (the strongest hit from the Y2H screen) as prey (in the presence or absence of its TMD).
The growth on medium lacking histidine (−HIS) is indicative of a physical interaction.
(B) Fluorescence microscopy demonstrating a shift in the subcellular localization of GFP-Sed5 from Golgi in control (WT) strains to a partially cytosolic localization in a Δget3 strain, and both cytosolic and few large puncta in Δget1/2 strains.
The GFP-Sed5 puncta in get mutants do not colocalize with the Golgi marker Anp1-RFP.
(C) Fluorescence microscopy demonstrating colocalization of GFP-Sed5 and Get3-tdRFP in cytosolic aggregates that form in a Δget1/2 background.
(D) Western blots of cell fractionation experiments to determine levels of Sed5 in membrane fractions.
Control (WT), Δget1/2 or Δget3 strains were divided into three fractions (Heavy Mem, Lighter Mem, and remainder of cellular proteins [Other]) and compared to input protein (Input) with Western blots immunostained against either Sed5 or the control Golgi transmembrane protein, Emp47, and ER transmembrane protein Sec61.
(E) Western blots of secreted proteins with αKar2.
Assay for Kar2 secretion was performed on a control strain (WT), mutants of the GET complex (Δget1, Δget2, Δget3), and on a yeast strain harboring a repressible allele of the essential TA protein Sed5 (tet-SED5), either in the presence (+Dox) or absence (−Dox) of the corepressor doxycycline.
(F) Western blots of secreted proteins with αKar2.
Assay for Kar2 secretion was performed on the triple mutant (Δget1/2/3) either alone or overexpressing SED5 from a high copy plasmid (+ OE SED5), and compared to a control strain (WT).
The GET Complex Affects the Biogenesis of a Wide Variety of TA Proteins
(A) Y2H assay showing Get3 as bait and various TA proteins (in the presence or absence of their TMDs) as prey.
The growth on medium lacking histidine (−HIS) is indicative of a physical interaction.
(B) Fluorescence microscopy of control (WT) and Δget1/2 strains expressing a broad variety of TA proteins.
GFP-Scs2, GFP-Sbh1, and GFP-Ysy6 under a galactose-inducible (GAL) promoter.
Cherry-Sbh2 was expressed from a plasmid under the constitutive TEF2 promoter.
(C) Fluorescence microscopy of control (WT) and Δget1/2 strains expressing two mitochondrial TA proteins, Cherry-Fis1 and Cherry-Tom22, expressed from a plasmid under the constitutive TEF2 promoter.
Role of GET Proteins in Creating Membrane Specificity
(A) Fluorescence microscopy showing the localization of GFP-Ubc6 and mitochondrially targeted dsRED (MTS-RFP) in a control (WT) or Δget1/2 strain.
(B) Fluorescence microscopy of a time course monitoring the subcellular localizations of the peroxisomal TA protein GFP-Pex15 as well as dsRED targeted to the mitochondria (MTS-RFP) following induction of Pex15 from a galactose inducible promoter in a control (WT), get1/2, or get3 strain.
Reduced Levels of TA Proteins Can Explain the Diverse Array of GET Complex Phenotypes
Serial dilutions in different conditions: SD + CuSO4 (Cu), SD + hydroxyurea (HU), SD + tunicamycin (Tunic.), SD + hygromycin (Hygro.), and YPD incubated at 39°C (39°C).
Strains shown are: control cells (WT), get mutants, five TA protein deletion strains, and a strain carrying a hypomorphic allele (DAmP) of an essential TA protein.
Copper sensitivity in the Δget3 strain is more pronounced in methionine prototrophic than auxotrophic cells.
We used Δmet15 cells for this panel, resulting in a less sensitive phenotype compared with MET+ cells depicted in Figure S4.
In Vitro Reconstitution of GET-Dependent Insertion of TA Proteins
(A) Autoradiograph of in vitro-translated, 35S methionine-labeled, α-factor (αfac) following incubation in the presence of microsomes derived from WT or Δget1/2 strains.
The position of untranslocated prepro-αfac and glycosylated, translocated pro-αfac (gαfac) are indicated.
(B) Autoradiograph of in vitro translated, 35S methionine-labeled Sed5 and Fis1 following incubation with microsomes derived from WT or Δget1/2 strains.
Prior to SDS-PAGE analysis, samples were immunoprecipitated with an anti-opsin antibody and then treated with EndoH, as indicated.
The position of untranslocated Sed5 and Fis1 as well as glycosylated translocated Sed5 (gSed5) are indicated.
(C) Graph representing the dose dependence of Sec22 translocation on addition of recombinant Get3 to either WT- or Δget3-derived translation extracts.
WT microsomes were added following translation, and the amount of glycosylated Sec22 relative to total Sec22 was calculated.
Results from three independent experiments are shown; data are presented as mean ± SD.
(D) Autoradiograph of in vitro-translated, 35S methionine labeled, Sec22 following translation in WT cytosol supplemented with optimal levels of Get3.
Translocation was terminated at the indicated times following addition of microsomes derived from either WT or Δget1/2 strains.
The position of untranslocated Sec22 as well as glycosylated translocated Sec22 (gSec22) are indicated.
Schematic Model for GET Complex Function
(Top) WT cells.
Get3 recognizes newly synthesized, ER-destined TA proteins.
The Get3-TA complexes dock onto the Get1/Get2 receptor.
This allows insertion of TA proteins.
(Middle) Cells lacking the receptor (Δget1/2).
Get3-TA complexes fail to reach the ER and, instead, are sequestered in cytosolic aggregates.
(Bottom) Cells lacking Get3 (Δget3).
Newly synthesized TA proteins intended for the ER are no longer shuttled into the GET pathway.
To varying degrees, depending on the TA proteins, they may use alternate ATP/GTP-dependant pathways or spontaneous routes for membrane insertion.
This could lead to misinsertion into the mitochondria, inefficient insertion into the ER, or aggregation in the cytosol.African swine fever virus protein p30 interaction with heterogeneous nuclear ribonucleoprotein K (hnRNP-K) during infection
Heterogeneous nuclear ribonucleoprotein K (hnRNP-K) was identified as interacting cellular protein with the abundant immediate early protein p30 from African swine fever virus (ASFV) in a macrophage cDNA library screening.
The interacting regions of hnRNP-K with p30 were established within residues 35–197, which represent KH1 and KH2 domains responsible for RNA binding.
Colocalization of hnRNP-K and p30 was observed mainly in the nucleus, but not in the cytoplasm of infected cells and infection modified hnRNP-K subcellular distribution and decreased the incorporation of 5-fluorouridine into nascent RNA.
Since similar effects were observed in cells transiently expressing p30, this interaction provides new insights into p30 function and could represent a possible additional mechanism by which ASFV downregulates host cell mRNA translation.
Structured summary
MINT-6742660:
hnRNP-K (uniprotkb:P61978) physically interacts (MI:0218)
with p30 (uniprotkb:Q8V1E7) by pull down (MI:0096)
MINT-6742673, MINT-6742696, MINT-6742729:
hnRNP-K (uniprotkb:P61978) physically interacts (MI:0218)
with p30 (uniprotkb:Q8V1E7) by two hybrid (MI:0018)
MINT-6742711:
p30 (uniprotkb:Q8V1E7) and hnRNP-K (uniprotkb:P61978) colocalize (MI:0403) by fluorescence microscopy (MI:0416)

Introduction
African swine fever virus (ASFV), the only member of the Asfarviridae family, is a large, cytoplasmic, double-stranded DNA virus that is responsible of a haemorrhagic and frequently fatal disease of swine [1].
The viral genome comprises more than 150 open reading frames and expression is regulated in a temporal fashion.
So, viral genes are classified as early or late depending on their requirement for viral DNA synthesis [2].
Little is known about the regulatory functions of ASFV proteins during infection and early virus proteins are important candidates to play critical roles in the modification of cellular metabolism to take advantage of host cell functions.
ASFV open reading frame CP204L encodes a 30 kDa protein named p30 or p32 [3,4] which represents the most abundantly expressed viral protein early in infection.
p30 exhibits a predominantly cytoplasmic location within infected cells and results phosphorylated in serine residues at the N-terminal [4] before its final incorporation to the viral particle.
p30 is also one of the most antigenic ASFV proteins [5], eliciting virus neutralizing antibodies in infected animals [6–8].
Previous studies have demonstrated a role for p30 in first stages of infection, since antibodies against p30 are able to inhibit virus internalization into the host cell [6].
Nevertheless, the regulatory function of p30 upon infection remains largely unknown.
To explore the potential targets of p30 during infection, we have used the yeast two-hybrid system to screen a porcine macrophage (the natural viral host cell) cDNA library for cellular proteins that may interact with p30.
We have identified heterogeneous nuclear ribonucleoprotein K (hnRNP-K) as the first cellular ligand of p30.
hnRNP-K is a multifunctional protein since it has been described to interact with diverse molecules of cellular [9,10], and viral origin [11–14], being involved in a variety of cellular functions such as regulation of transcription and translation [15], RNA splicing, mRNA stability and transport of pre-mRNA out of nucleus to cytoplasm.
It has been demonstrated to interact with different molecules involved in signal transduction such as Src, Fyn and Lyn [16].
Recently, a number of new hnRNP-K partners have been identified by using a proteomic approach [17] and new roles have been suggested for this protein.
ASFV p30 was found to interact directly with host hnRNP-K during ASFV infection and we have mapped the interacting regions of both proteins.
p30 modifies hnRNP-K subcellular distribution and could contribute to modulate hnRNP-K functions related to processing and export of mRNAs during ASFV infection.
Materials and methods
Plasmids
For the yeast two-hybrid assay, plasmids pGBT9 and pACT2 (BD Sciences) were used as sources of the GAL4 DNA-binding domain (BD) and transcriptional activation domain (AD), respectively.
To generate pGBT9-p30, the complete p30 coding sequence was directly amplified from ASFV genome of E70 isolate and inserted in EcoRI and BamHI sites in pGBT9.
Unrelated ASFV protein p54 was used as negative control and cloned in pGBT9, as previously published [18].
A cDNA library from porcine macrophage, cloned in XhoI site in pACT2, was kindly provided by Linda Dixon [19].
p30 and hnRNP-K different mutant truncations used for mapping the regions involved in the interaction were generated by PCR from pGBT9-p30 and pGEX-RNP-K, respectively.
The fragments generated were inserted in frame with AD or BD in pGBT9 or pACT2.
pACT2-K, containing complete hnRNP-K coding sequence fused to AD, was derived from pGEX-RNP-K.
pCMV-p30 was also derived from pGBT9-p30 and used in transfection experiments.
For the glutathione S-transferase (GST) pull-down experiments, plasmids pGEX-RNP-K (kindly provided by Dr.
Levens) and pGEX-4T (GE Healthcare) were used to express GST-hnRNP-K fusion protein or GST alone in Escherichia coli.
Yeast two-hybrid
pGBT9-p30 and unrelated control protein pGBT9-p54 were independently used as baits to screen a pACT2 cDNA library from pig macrophages in Saccharomyces cerevisiae reporter strain Y190 as previously published [18,20,21].
Yeast were sequentially transformed with bait plasmid and pACT2 library by the lithium acetate method.
After auxotrophic and colony size selection, resulting clones were analyzed for expression of GAL4-dependent β-galactosidase.
Plasmid DNA from those clones exhibiting β-galactosidase activity was isolated and retransformed into yeast strain Y190 with pGBT9-p30 to eliminate false positives.
The sequence of inserts was determined by sequencing using specific primers and compared with the data base of the NCBI using the BLAST program.
pGBT9-p30, pGBT9-p54 and pACT2-K were individually transformed in yeast and tested for β-galactosidase activity to exclude activation of gene reporter by itselves.
The different mutant truncations of p30 were individually transformed with pACT2-K in Y190 and resulting clones tested for expression of GAL4-dependent β-galactosidase.
Similarly, different mutant truncations of hnRNP-K were transformed with pGBT9-p30 and tested.
GST pull-down experiments
GST-hnRNP-K and GST proteins were produced in E.
coli BL21 cells, previously transformed with vector pGEX-RNP-K or pGEX-4T.
Cells were induced with 0.1 mM IPTG for 2 h at 37 °C.
Bacteria were harvested and suspended in lysis buffer (PBS, 1% Triton X-100, 1 mM PMSF, 5 mM DTT, and anti-proteases), and sonicated on ice.
GST-hnRNP-K and GST alone were purified from cleared lysates by mixing with glutathione-sepharose 4B beads (GE HealthCare), 5 ml of cleared lysate/400 μl of beads, for 1 h at 4 °C.
After extensive washing, GST-hnRNP-K or GST beads were incubated in binding buffer (50 mM HEPES, ph 7.5, 50 mM NaCl, 0.1% Nonidet P-40 with protease inhibitor mixture (Roche Molecular Biochemical)) at 4 °C for 1 h with insect cell extracts containing either p30 or p54 ASFV proteins overexpressed in a baculovirus system [8].
Equal amounts of GST, GST-hnRNP-K and ASFV proteins p30 and p54 were used as judged by Coomasie Blue staining.
Alternatively, monolayers of Vero cells infected with BA71V and lysed in 50 mM HEPES, ph 7.5, 50 mM NaCl, 0.1% Nonidet P-40 containing protease inhibitor, were similarly mixed with GST-hnRNP-K or GST beads.
In all cases, after extensive washing, bound proteins were eluted and analyzed by Western blotting with anti-p30 monoclonal antibody (diluted 1:500) or anti-p54 serum (diluted 1:60).
Virus infections and immunofluorescence analyses
Vero cells were grown in DMEM 5% onto cover slips, at 50–60% confluence, allowed to attached and then infected or mock infected with 1 pfu/cell of ASFV strain BA71V.
Cells were fixed from 4 to 24 hpi in PBS-paraformaldehyde 3.8% and permeabilized in PBS-TX-100 0.1% for immunofluorescence analyses.
In transient expression analysis pCMV-p30 was transfected in Vero cells using Fugene6 (Roche Molecular Biochemical) according to manufacturer’s indications.
Anti-hnRNP-K rabbit serum and mouse monoclonal anti-p30 were used at 1:400 and 1:100, respectively.
Secondary antibodies used were Alexa 488-conjugated goat anti-rabbit IgG (Molecular Probes) and Rhodamine Red conjugated goat anti-mouse IgG (Molecular Probes), both at dilution 1:300.
Incorporation of 5-fluorouridine (FU; Sigma) into nascent RNA was determined as previously [22].
Briefly, transfected or BA71V infected cells were pulse labelled with 5 mM FU at 37 °C for 20 min.
After washing with PBS to remove residual FU, cells were fixed and permeabilized as described.
Monoclonal antibody against halogenated UTP (anti-BrU; Sigma) was used diluted 1:500 to label nascent RNA.
Specific fluorescence was measured using Lasersharp Processing 3.2 program (BioRad).
Specificity of labelling and absence of signal crossover was excluded in all cases by examination of single labelled control samples, including labelling controls with omission of primary antibody.
When indicated, viral factories and cell nuclei were identified by direct staining of cells with Hoechst 33342 1 μg/ml (Sigma).
Finally, cells were mounted onto slides using Fluorsave reagent (Calbiochem).
Conventional microscopy was carried out in a Leica photomicroscope with a digital camera and digitized images were obtained with Qwin program (Leica).
Confocal microscopy was carried out on an MRC1024 system (BioRad) mounted on a Nikon Eclipse 300 microscope equipped with 63 and 100X objectives.
Statistical analysis of colocalization was performed using Lasersharp Processing 3.2 program (BioRad).
Preparation of nuclear and cytoplasmic extracts
Preconfluent monolayers of Vero cells were mock infected or infected with ASFV strain BA71V.
Cells were washed and harvested from 0 to 24 hpi in cold PBS.
Isolation of nuclear and cytoplasmic fractions was performed as described previously [23].
Briefly, cell pellets were suspended in three volumes of hypotonic buffer and homogenization was carried out with a glass Dounce homogenizator.
After centrifugation at 3300 × g for 15 min, supernatant was considered as the cytoplasmic fraction while pellets containing nuclei were suspended in half volume of low salt buffer.
Extraction of nuclei was performed by adding 1 volume of high salt buffer for 30 min followed by centrifugation at 25000 × g.
Resulting supernatant was considered as the nuclear fraction.
After dialysis, 20 μg from each cytosolic sample and 13 μg from each nuclear sample were analyzed by Western blot with specific antibodies (anti-p30, anti-hnRNP-K and anti-β-tubulin diluted 1:500, 1:10.000 and 1:500, respectively).
Quality of fractionation process was assessed by detection of β-tubulin on both nuclear and cytosolic samples.
Bands obtained corresponding to hnRNP-K were densitometrically quantified using an image analyzer with TINA software package (Raytest).
Results
Identification of hnRNP-K as cellular interacting protein with ASFV p30
To identify cellular proteins interacting with ASFV early protein p30, yeast two-hybrid system was used to screen a porcine macrophage cDNA library.
After selection from a total of 5 × 106 transformants screened, two potential positive clones were obtained in the reporter gene assay.
DNA sequence analysis showed that cDNA contained in these clones, identical in size and composition, matched the cDNA sequence encoding hnRNP-K.
cDNA sequences from positive clones represent nucleotides from 169 to 870 of the hnRNP-K cDNA sequence (GeneBank™ accession number 241477), with 98% nucleotide identity, corresponding to amino acid residues 13–246 of hnRNP-K protein.
To confirm this result, the plasmid contained in these two clones was isolated and retransformed together with pGBT9-p30 detecting β-galactosidase activity.
Conversely, clones obtained screening the library with pGBT9-p54 never encoded a similar sequence.
Moreover, no reporter gene activity was detected in clones transformed with pACT2-K alone, excluding activation of gene reporter by itself.
p30 binds directly to hnRNP-K in vitro
The interaction was further confirmed by in vitro binding assays using a GST-hnRNP-K fusion protein bound to glutathione-sepharose 4B beads.
GST pull-down experiments were carried out followed by Western blot with specific antibodies.
First, p30 and another unrelated ASFV protein (in this case, p54 as negative control) were used to bind GST fusion protein or GST alone.
Protein p30, and not p54, was retained in the presence of GST-hnRNP K, showing a band of appropriate size (30 kDa) for p30 in Western blotting.
This band did not appear in the presence of GST alone, indicating specific interaction of p30 with hnRNP K and not with GST (Fig.
1A).
Identical results were obtained in subsequent experiments using BA71V infected or mock-infected cells extracts instead of baculovirus infected cell extracts in the pull-down assay (Fig.
1B).
Mapping the interacting domains of p30 and hnRNP-K
To examine the regions of both proteins involved in this interaction, different truncations of p30 fused to GAL4-BD domain and different truncations of hnRNP-K fused to GAL4-AD domain were tested for interaction (Fig.
2).
As p30 sequence does not contain any previously characterized functional domain, we analyzed independently p30 amino terminus, central region and carboxy terminus fragments (1–70, 1–140 and 140–204 aa residues) for hnRNP-K interaction.
In addition, diverse truncations of hnRNP-K were performed attending to previously well characterized functional domains [10] and tested for interaction using the yeast two-hybrid system.
The results showed that none of the three different p30 truncations interacted with hnRNP-K (Fig.
2A).
On the other hand, we could determine that the hnRNP-K fragment from amino acid residue 35–197 contained the interacting region with p30 (Fig.
2B).
This region harbours KH1 and KH2 domains, identical each other, which have been previously identified as functional elements responsible for RNA binding.
Partial colocalization of protein p30 and hnRNP-K in ASFV infected cells
To ascertain the time in infection and the subcellular compartment at which the interaction occurs, ASFV infected cells were examined by confocal laser scanning microscopy at different times post infection (at least 30 infected cells were analyzed for each time point at 0, 6, 8 and 12 hpi).
Newly synthesized viral protein p30 could not be detected in the nucleus or cytoplasm of examined cells before 8 hpi.
Since that time point, p30 and hnRNP-K were found to colocalize in discrete areas within the nucleus of infected cells.
Statistical analysis of colocalization showed that the totality of protein p30 present in the nucleus was detected colocalizing with hnRNP-K since 12 hpi (Fig.
3B).
In contrast, most hnRNP-K was detected also in other parts of the nucleus exhibiting no colocalization with p30, indicating a partial colocalization of hnRNP-K with p30.
Interestingly, in the cytoplasmic region of those infected cells hnRNP-K was barely detected and colocalization percentages were not significant (data not shown).
ASFV infection modifies hnRNP-K distribution
Subcellular distribution of p30 and hnRNP-K were analyzed by immunofluorescence microscopy in ASFV or mock infected cells.
ASFV infection resulted in an intensification of nuclear, but not cytoplasmic, staining of hnRNP-K when compared to uninfected cells.
When nuclei of infected cells were examined in detail, nuclear hnRNP-K was detected in characteristic granular structures coincident with areas completely devoid of nucleic acid staining.
Interestingly, these kind of hnRNP-K accumulation was also observed in cells transiently expressing p30 (pCMV-p30), but not in mock-infected cells, suggesting the involvement of p30 on these changes (Fig.
4) during ASFV infection.
To confirm the retention of hnRNP-K in the nucleus of infected cells observed by immunofluorescence, the levels of hnRNP-K were assessed by Western blot at different times post infection using total, cytoplasmic and nuclear protein cell extracts.
Total levels of hnRNP-K (Fig.
5A) were determined as infection proceeded (monitored by detection of p30) in order to discard that these changes could be to due changes in newly synthesized hnRNP-K or protein degradation.
While no changes in total hnRNP-K levels were observed, quantitative analysis revealed a reduction of hnRNP-K levels in cytoplasm protein extracts from 12 hpi until the end of the infectious cycle (Fig.
5B).
This reduction was in agreement with the observed increase of nuclear hnRNP-K levels, as judged by Western blot analysis.
These results are consistent with the immunofluorescence analysis and indicate nuclear accumulation of hnRNP-K during ASFV infection.
These changes in hnRNP-K distribution might disrupt the normal behaviour of this ribonucleoprotein which is involved in diverse biological processes including regulation of transcription and translation.
To examine the transcriptional activity in ASFV infected cells we performed immunodetection of nascent RNA after FU pulse labelling (Fig.
5C).
When compared with mock-infected cells, the incorporation of FU into nascent RNA was almost completely abolished in infected cells from 12 hpi, coincident with the time point at which hnRNP-K redistribution was observed.
Discussion
Infection of eukaryotic cells with large DNA viruses often results in extensive interactions of viral gene products with macromolecular pathways of the host cell.
By using the yeast two-hybrid system, we identified cellular hnRNP-K as an interacting protein with ASFV early protein p30.
This interaction was further confirmed by an in vitro GST-fusion pull-down assay, using either p30 obtained from baculovirus system or ASFV infected cell extracts.
hnRNP-K is closely coupled to gene expression [15,24] and signal transduction pathways [16,25].
One feature which distinguishes hnRNP-K from other hnRNPs is the presence of three repeated K homology motives (KH) which are responsible for RNA-binding [26–28].
Interestingly, the p30 interacting region within hnRNP-K contains KH1 and KH2 motives, suggesting that the interaction described might modify hnRNP-K functions such as processing and export of the cellular mRNAs than on signal transduction pathways.
hnRNP-K has been previously described to interact with several proteins from diverse viruses.
Core protein from Hepatitis C virus (HCV) and Dengue virus (DEN), are proposed to relieve the repressive effect of hnRNP-K on transcription regulation of different human genes in vivo, disrupting the multiple functions of hnRNP-K and contributing to virus pathogenesis [11,12].
Studies with phosphoprotein IE63 (also named ICP27) from Herpes simplex virus (HSV) suggest that an IE63-mediated phosphorylation of hnRNP-K by casein kinase 2, results in inhibition of binding to RNA, affecting transport of cellular mRNAs and altering the subcellular location of hnRNP-K [13].
Nevertheless, no homology sequence in nucleotides or amino acids has been found when comparing these virus proteins interacting with hnRNP-K, including p30, indicating that this common interaction could occur in multiple and diverse ways.
During ASFV infection, hnRNP-K normal function could be affected since changes related to hnRNP-K subcellular distribution are observed.
First, accumulation of hnRNP-K within the nucleus of ASFV infected cells was evident and subsequently the levels of cytoplasmic hnRNP-K decreased while nuclear levels were increased, as infection proceeded, supporting the idea that accumulation of hnRNP-K within nucleus is most likely due to relocation from cytoplasm.
Changes in hnRNP-K subcellular distribution have been reported previously in poliovirus infection where virus infection affects the normal nucleo-cytoplasmic trafficking of the host cell [29].
Second, this hnRNP-K accumulation is coincident with the appearance of granular structures in the nucleus where the ribonucleoprotein is detected.
Since no nucleic acids can be detected in these areas, these structures could represent inactive sites within the nucleus where hnRNP-K remains sequestered as consequence of infection, as postulated for HSV1 [13].
Similar structures have been previously reported during HCV infection.
In this case, HCV core protein and hnRNP-K also colocalized in granules in the nucleus [12].
The presence of p30 in nucleus colocalizing with hnRNP-K suggests an involvement of this interaction in the appearance of these characteristic structures.
Moreover, the identification of these structures also in cells transiently expressing p30 supports this idea.
Interestingly, neither conventional nuclear localization nor classical nuclear import signal have been previously found within ASFV protein p30 sequence, since it exhibits a predominantly cytoplasmic distribution.
In this report we have demonstrated that small amounts of p30 are also present in the nucleus of the infected cell, so it can not be excluded that interaction with shuttling protein in cytoplasm could facilitate p30 traffic to the cell nucleus.
Nevertheless, colocalization of both proteins in the cytoplasm during infection was not detectable by immunofluorescence, probably because of the low relative abundance of hnRNP-K in cytoplasm and its shuttling properties which difficult its visualization by immunofluorescence microscopy.
In conclusion, protein p30 could have further regulatory actions associated to its specific binding to hnRNP-K protein in the nucleus of infected cells.
It is well known that viruses have evolved different mechanisms to alter host cell transcription and translation to promote transcription of its own DNA [30].
Here, we confirmed that all changes observed in hnRNP-K distribution during ASFV infection take place simultaneously with an extraordinary reduction in 5-fluouridine (FU) incorporation into nascent RNA, which represents an alteration in the cellular transcriptional activity.
One way to inhibit cellular transcription is the recruitment of host cell RNA polymerase II.
Nevertheless this is not a requirement for ASFV, since α-amanitin does not inhibit the production of infectious virus [31].
However, in cell systems previously studied (macrophages and Vero cells), ASFV infection produces a general shut-off of protein synthesis that affects up to 65% of the cellular proteins [2,32,33].
This inhibition of protein synthesis is detectable very early after ASFV infection, so it will be interesting to determine whether the interaction here described is contributing to the general host cell shut-off or to a concrete cellular pathway.
References

Proteins specified by African swine fever virus: V.
Identification of immediate early, early and late proteins.
Characterization of p30, a highly antigenic membrane and secreted protein of African swine fever virus
Sequence and characterization of the major early phosphoprotein p32 of African swine fever virus
Comparison of a radioimmunoprecipitation assay to immunoblotting and ELISA for detection of antibody to African swine fever virus
The African swine fever virus proteins p54 and p30 are involved in two distinct steps of virus attachment and both contribute to the antibody-mediated protective immune response
Neutralizing antibodies to different proteins of African swine fever virus inhibit both virus attachment and internalization
High level expression of the major antigenic African swine fever virus proteins p54 and p30 in baculovirus and their potential use as diagnostic reagents
hnRNP K: one protein multiple processes
Diverse molecular interactions of the hnRNP-K protein
The heterogeneous nuclear ribonucleoprotein K (hnRNP-K) interacts with dengue virus core protein
Hepatitis C virus core protein interacts with heterogeneous nuclear ribonucleoprotein K
The multifunctional herpes simplex virus IE63 protein interacts with heterogeneous ribonucleoprotein K and with casein kinase 2
Heterogeneous nuclear ribonuclear protein K interacts with Sindbis virus nonstructural proteins and viral subgenomic mRNA
Heterogeneous nuclear ribonucleoprotein K is a transcription factor
The K protein domain that recruits the interleukin 1-responsive K protein kinase lies adjacent to a cluster of c-Src and Vav SH3-binding sites.
Implications that K protein acts as a docking platform.
Landscape of the hnRNP K protein-protein interactome
African swine fever virus protein p54 interacts with the microtubular motor complex through direct binding to light-chain dynein
A viral mechanism for inhibition of the cellular phosphatase calcineurin
A179L, a viral Bcl-2 homologue, targets the core Bcl-2 apoptotic machinery and its upstream BH3 activators with selective binding restrictions for Bid and Noxa
The MyD116 African swine fever virus homologue interacts with the catalytic subunit of protein phosphatase 1 and activates its phosphatase activity
Promyelocytic leukemia (PML) nuclear bodies are protein structures that do not accumulate RNA

The hnRNP proteins
Insulin alters heterogeneous nuclear ribonucleoprotein K protein binding to DNA and RNA
Characterisation of the nucleic-acid-binding activity of KH domains.
Different properties of different domains.
KH domain: one motif, twofolds
Essential role for KH domains in RNA binding: impaired RNA binding by a mutation in the KH domain of FMR1 that causes fragile X syndrome
Effects of poliovirus infection on nucleo-cytoplasmic trafficking and nuclear pore complex composition
Early shutoff of host protein synthesis in cells infected with herpes simplex viruses
Effect of inhibitors of the host cell RNA polymerase II on African swine fever virus multiplication
Identification of cellular proteins modified in response to African swine fever virus infection by proteomics
African swine fever virus-induced polypeptides in porcine alveolar macrophages and in Vero cells: two-dimensional gel analysis
Interaction of viral protein p30 with hnRNP-K in the GST fusion protein binding assay.
(A) Viral protein p30, produced in the baculovirus system, was incubated with GST or GST-hnRNP-K beads (GST-K).
Unrelated ASFV protein p54 was similarly used as negative control.
After extensive washing and elution from beads, resulting protein complexes were electrophoresed and analyzed by WB with monoclonal anti-p30 or anti-p54 serum (see Section 2).
(B) As described before, ASFV infected (BA71V) or mock infected (MI) protein cell extracts were incubated with GST-hnRNP-K (GST-K) beads and analyzed by WB with anti-p30 monoclonal antibody.
In all cases C+ indicates p30 or p54 protein extracts used as positive Western blot controls.
Regions of p30 and hnRNP-K involved in the interaction.
(A) Schematic representation of the diverse p30 truncations tested for interaction with full length hnRNP-K in the yeast two-hybrid assay.
(B) Schematic representation of the diverse hnRNP-K truncations tested for interaction with complete p30 in the yeast two-hybrid assay.
Structure with details of the different functional regions is shown [10].
NLS, nuclear localization signal; KH, K homology domains; KNS, bidirectional nuclear export signal.
In all cases correspondent clones were analyzed for β-galactosidase activity indicating interaction (+) or no interaction (−) with respective partner tested.
Colocalization of p30 and hnRNP-K in the nucleus of infected cells.
Cells were infected with BA71V strain and analyzed at different times post infection by confocal immunofluorescence microscopy acquiring 0.1 μm optical sections from the Z-axis.
A representative image of an infected cell at 8 hpi is shown (A).
p30 was detected with a monoclonal anti-p30 followed by Alexa 647-conjugated goat anti-mouse antibody (red) and hnRNP-K with an anti-hnRNP-K specific serum followed by Alexa 488-conjugated goat anti-rabbit antibody (green).
Discrete spots of colocalization of both proteins (orange) can be discerned in the cell nucleus.
Bar, 10 μm.
(B) Quantification of colocalization during infection.
As described above, the nuclei of 30 infected cells were analyzed each time for hnRNP-K (black bars), p30 (white bars) and colocalization (grey bars) fluorescence emission.
Data are expressed as percentages and normalized to total fluorescence due to hnRNP-K exclusively within the cell nucleus.
ASFV infection induces changes in subcellular distribution of hnRNP-K.
(A) Vero cells were either infected with BA71V (1 pfu/cell) or transfected with pCMV-p30 and analyzed by immunofluorescence microscopy at 8 hpi or 24 h post transfection, respectively.
ASFV protein p30 was detected with a mouse anti-p30 monoclonal antibody and rhodamine red conjugated corresponding secondary antibody (red).
hnRNP-K distribution in the cell nucleus was detected with a rabbit antiserum anti-hnRNP-K and Alexa Fluor 488 conjugated corresponding secondary antibody (green).
Nuclei of cells were stained with Hoechst 3332.
Accumulation of hnRNP-K in infected cells as spots which are coincident with the absence of nucleic acids staining can be appreciated in BA71V infected and pCMV-p30 transfected cells.
Details of hnRNP-K accumulation are amplified from selected areas.
(B) Comparison between uninfected and 12 hpi BA71V infected cell nuclei is shown in more detail.
Arrows indicate the position of some of these structures in detail within the nucleus of an infected cell (bottom), but not in a non infected cell (top).
VF indicates the perinuclear viral factory, identified by Hoechst staining in an infected cell.
Bar, 16 μm.
Analysis of hnRNP-K levels during ASFV infection by Western blot.
Vero cells were infected with 5 pfu/cell of BA71V strain and analyzed by WB with specific antibodies at different times after infection.
(A) p30 and hnRNP-K were detected in total protein cell extracts with specific antibodies (see Section 2).
β-Tubulin was detected to ensure that same amount of total protein was loaded.
(B) Levels of hnRNP-K were analyzed independently from nuclear and cytoplasmic protein infected cell extracts.
One of three representative experiments is shown including the relative intensities of quantified hnRNP-K (rel.
int.).
β-Tubulin detection was included as quality control of the cellular fractionation.
(C) Vero cells were infected with BA71V (0.5 pfu/cell) and analyzed at 12 hpi for incorporation of FU into nascent RNA after FU pulse (see Section 2) to evaluate transcriptional activity.
Arrows indicate the position of infected cells identified by p30 detection with specific antibody.
Note the increase of nuclear hnRNP-K staining of infected cells when compared to the non infected cells.
Bar, 30 μm.SIRT3 interacts with the daf-16 homolog FOXO3a in the Mitochondria, as well as increases FOXO3a Dependent Gene expression
Cellular longevity is a complex process relevant to age-related diseases including but not limited to chronic illness such as diabetes and metabolic syndromes.
Two gene families have been shown to play a role in the genetic regulation of longevity; the Sirtuin and FOXO families.
It is also established that nuclear Sirtuins interact with and under specific cellular conditions regulate the activity of FOXO gene family proteins.
Thus, we hypothesize that a mitochondrial Sirtuin (SIRT3) might also interact with and regulate the activity of the FOXO proteins.
To address this we used HCT116 cells overexpressing either wild-type or a catalytically inactive dominant negative SIRT3.
For the first time we establish that FOXO3a is also a mitochondrial protein and forms a physical interaction with SIRT3 in mitochondria.
Overexpression of a wild-type SIRT3 gene increase FOXO3a DNA-binding activity as well as FOXO3a dependent gene expression.
Biochemical analysis of HCT116 cells over expressing the deacetylation mutant, as compared to wild-type SIRT3 gene, demonstrated an overall oxidized intracellular environment, as monitored by increase in intracellular superoxide and oxidized glutathione levels.
As such, we propose that SIRT3 and FOXO3a comprise a potential mitochondrial signaling cascade response pathway.
Introduction
One theme emerging over the last several years is that aging is a complex, genetically regulated cellular process 1.
Genes critical to longevity have been isolated and characterized in several model systems including yeast and C.
elegans 1.
Sirtuin family genes first came to attention because their Saccharomyces cerevisiae homolog (Sir2) was shown to regulate both replicative and overall lifespan by caloric restriction (CR) 2.
In humans and mice there are seven Sirtuins, two that are located in the nucleus (SIRT1, 6, and 7), one in the cytoplasm and nucleus (SIRT2), and 3 are localized to the mitochondria (SIRT3, 4, and 5) 3.
While yeast Sir2 is a NAD+ dependent deacetylase, mammalian Sirtuins have more varied targets, including either deacetylase or ADP-ribosyltransferase activity 4.
It is clear that sirtuins regulate a series of essential intracellular processes that defend the cell against multiple types of cellular damage, including oxidative damage 5.
These observations imply that Sirtuin genes may be metabolic guardians that could play a central role in longevity by regulating cellular or organelle renewal.
That is, as cells and critical intracellular organelles slowly degrade from age related oxidative stress, Sirtuin genes may respond to this by activating signaling pathways to repair damage.
When outlining a course of investigation for the potential role of Sirtuins as organelle surveillance genes, we chose to examine the mitochondrial inner membrane protein SIRT3 for several reasons.
First, it has long been suggested that the mitochondria play a role in aging and oxidative stress 6.
Second, mitochondrial electron transport chains reduce of O2 to form superoxide (O2•-), leading to oxidative stress which plays a role in genomic instability and aging 7,8.
Third, SIRT3 localization to the inner mitochondrial membrane is consistent with a role in monitoring and responding to conditions harmful to electron transport, activating processes to minimize mitochondrial damage.
The FOXO family of transcription factors: FOXO1, FOXO3a, FOXO4 and FOXO6 are the human homologs of the daf-16 C.
elegans gene.
Daf-16 is the master regulator of dauer formation 9 and contributes to the regulation of lifespan in nematodes 10.
FOXO transcription factors appear to regulate a wide range of intracellular processes including metabolism and cellular resistance to various forms of oxidative stress 11,12.
It is also established that sirtuins interact with, and under specific cellular conditions, regulate that activity of FOXO gene family proteins 13,14.
Thus, we chose to investigate if SIRT3 might also interact with and regulate the activity of any of the FOXO proteins.
MATERIALS AND METHODS
Cell lines and Plasmids
HCT-116 (human colon carcinoma) and Cos-7 cells were cultured in McCoy's 5A media, containing 10% heat-inactivated (56°C, 30min) fetal bovine serum (FBS) and.
supplemented with penicillin (100 Units/mL) and streptomycin (100μg/mL).
The expression vector pcDNA4 was obtained from Promega, Inc.
pCMV-SIRT3-wt and pCMV-SIRT3-mt were a kind gift from Toren Finkel (National Institutes of Health).
These expression plasmids express either the wild-type gene or an acetylation null gene where amino acid 248 has been changed from a histidine to tyrosine 15.
Permanent cell lines expressing wild-type SIRT3 (wt-SIRT3) or the deacetylase mutant (mt-SIRT3) (supplemental Fig.
S1) were constructed by transfecting pcDNA4, pCMV-SIRT3-wt, or pCMV-SIRT3 in HCT116 cells using FuGene 6 and selected with zeosin (supplemental Fig.
S2).
For this work cell lines expressing the various SIRT3 genes are referred to as wt-SIRT3 or mt-SIRT3 while plasmids expressing these genes are denoted as pCMV-SIRT3-wt or pCMV-SIRT3-mt (supplemental Fig.
S1).
The FOXO3a expression plasmids (pCMV-wt-FOXO3a, pCMV-m-FOXO3a, and pCMV-N-FOXO3a) were obtained from Addgene (Cambridge, Ma).
IP and Mitochondrial Fractionation
Cell extracts were prepared using a modification of the previously described method 16,17 and protein concentrations were determined using the Bradford assay (BioRad Labs., Hercules, CA) on a Beckman DU-640 spectrophotometer.
Mitochondria extracts were prepared using kit provided by Pierce (Rockford, Il).
Briefly cells were lysed by dounce homogenization in lysis buffer followed by centrifugation at 700 x g to remove the nuclear fraction along with cell debris.
The supernatant was then spun at 3,000 x g for 15 minuets to isolate the mitochondria fraction.
Trypan blue exclusion analysis was performed to check for lysis efficiency following dounce homogenization.
Western samples were loaded into denaturing SDS-polyacrylamide gels, membranes were probed with anti-SIRT3 (See supplement) or anti-FOXO3a antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) and incubated with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA).
Co-transfections, Luciferase Assays, and Chromatin Immunoprecipitation
For transient assays, cells were plated at 2 ×106 cells per 100-mm plate, co-transfected, and harvested after 36 hrs.
Cells were co-transfected with 1 µg of pCMV-β-gal and analysis for luciferase and β-galactosidase activity.
Luciferase activity was normalized to β-galactosidase activity and the results are presented as luciferase/β-galactosidase relative units.
The results are presented as relative-fold induction over the sham treated control.
Chromatin Immunoprecipitation
Intracellular superoxide measurements and thiol analysis
Steady state levels of superoxide were estimated using the fluorescent dye, dihydroethidium (DHE) as described in Slane et al 8.
Glutathione (GSH) and glutathione disulfide (GSSG) contents were determined by the recycling method 18 and a yellow color change was detected spectrophotometrically at 412 nm as DTNB is converted to TNB by the conversion of 2 GSH to GSSG.”.
RESULTS
The daf-16 homolog, FOXO3a, forms a complex with SIRT3
The FOXO family of transcription factors regulate metabolism and confer stress resistance 11,12.
It has also been shown that the well studied SIRT1 interacts with, and under specific cellular conditions, regulates the activity of FOXO gene family proteins 14,19.
We sought to establish whether the mitochondrial localized SIRT3 may also form a physical interaction with the FOXO family protein, FOXO3a, using co-immunoprecipitation (Co-IP) techniques.
Carboxy-terminally myc tagged wild-type (p-myc-hSIRT3-wt), and mutant (p-myc-hSIRT3-mt) SIRT3 expression vectors were transfected into Cos-7 cells followed by Co-IP with an anti-myc antibody.
Western analysis with an anti-FOXO3a antibody subsequently showed a physical interaction between both wild-type and mutant SIRT3 with FOXO3a (Fig.
1A).
To ensure that this interaction was not an artifact of the COS-7 cell line, or the transient transfection conditions, HCT116 cells, which contain undetectable levels of endogenous SIRT3 by western analysis, were stably transfected with p-myc-hSIRT3-wt, and p-myc-hSIRT3-mt SIRT3 expression vectors (Fig S2).
These cells were used to determine if the FOXO3a – SIRT3 interaction occurs in the mitochondria or the cytosol.
Co-IP experiments were repeated using the HCT116 SIRT3 expressing stable cell lines.
FOXO3a was found to interact with SIRT3 in both whole cell (Fig.
1B) and mitochondrial extracts (Fig.
1C).
All fractions and samples in the Co-IP experiments were checked for the presence of SIRT3, using a SIRT3 specific antibody (Biomol, Plymouth Meeting , PA), as well as tubulin (Santa Cruz Biotechnology, Santa Cruz, CA), and Cytochrome C (Mitosciences, Eugene, OR) to ensure fraction purity (data not shown).
These experiments demonstrate that both wild type and mutant SIRT3 form a physical interaction with FOXO3a in mitochondrial extracts.
FOX3a is a mitochondrial protein
FOXO3a interacts with SIRT3 in mitochondrial extracts in vitro thus, we chose to investigate if any of the FOXO3a proteins might be localized to the mitochondrial tissue lysates as well.
Western analysis of mitochondrial samples from human heart (Fig.
2A) and liver (Fig.
2B) (Microscience, Inc) lysates clearly shows that FOXO3a is present in the mitochondrial cellular fractions.
Anti-tubulin and anti-cytochrome C antibodies were used as controls to determine mitochondrial and non-mitochondrial protein fractions.
As expected, similar results were seen in HCT116 and HCT116 cells genetically altered to express either the wild-type or deacetylation null SIRT3 genes (Fig.
2C).
SIRT3 activates FOXO3a dependent gene expression
To determine if FOXO3a activates promoter activity in our system, three FOXO3a expression vectors were employed that: (1) express the wild-type gene that is still inhibited by Akt (pCMV-FOXO3a-wt); (2) express a mutant gene that appears to be a dominant negative (pCMV-mt-FOXO3a); and (3) express a dominant positive (constitutively active) gene that cannot be phosphorylated and inhibited by Akt (pCMV-N-FOXO3a) 20.
Co-transfection of these three plasmids with p3400-MnSOD-Luc, which contains the MnSOD promoter and its two consensus FOXO binding sites (supplemental Fig.
S3) cloned upstream of luciferase, demonstrated that pCMV-N-FOXO3a increased p3400-MnSOD-Luc luciferase activity (Fig.
3A, bar 3).
In contrast, co-transfection with pCMV-mt-FOXO3a decreased MnSOD expression of the reporter (bar 2) as compared to the control, wild-type vector, pCMV-wt-FOXO3a (bar 1).
These results clearly demonstrate that FOXO3a can activate the expression of the reporter gene.
A FOXO3a luciferase reporter construct was made that contains three FOXO3a binding sites fused to the luciferase gene (p3x-FOXO-tk-Luc) and this reporter vector was used to determine FOXO3a dependent gene expression as measured by luciferase levels.
p3x-FOXO-tk-Luc was subsequently co-transfected into HCT116 cells with either pCMV-SIRT3-wt or pCMV-SIRT3-mt expression vectors.
Expression of the wild-type SIRT3 increases luciferase activity (Fig.
3B, bar 1 versus bar 2).
These results show that SIRT3 activates FOXO3a dependent gene expression.
To determine if FOXO3a is downstream of SIRT3 in the activation of the p3x-FOXO-Luc the reporter plasmid was co-transfected with either the wild-type or mutant SIRT3 expression vector (as done in Fig.
3B), and either pCMV-wt-FOXO3a or pCMV-mt-FOXO3a.
These experiments will determine if a mutant FOXO3a protein will change the ability of SIRT3 to activate p3x-FOXO-tk-Luc gene expression.
Transfection of p3x-FOXO-tk-Luc or pCMV-SIRT3-wt (bar 2) is a positive control and is identical to Fig.
2B, bar 1.
In contrast, co-transfection of pCMV-SIRT3-wt with a mutant FOXO3a expression vector (pCMV-mt-FOXO3a) decreased p3x-FOXO-tk-Luc driven luciferase activity (Fig.
3C, bar 2 versus 3.
These results suggest that in transient assays FOXO3a dependent gene expression that is induced by SIRT3 is prevent by expression of a mutant FOXO3a protein.
Overexpression of SIRT3 induced MnSOD
p3400-MnSOD-Luc was transfected into HCT116 cells genetically altered to over express either wild-type (wt-SIRT3) or a catalytically inactive deacetylation mutant (mt-SIRT3) SIRT3 gene (Supplemental Fig.
S1).
HCT116 cells were used since they have nearly undetectable SIRT3 protein (data not shown).
These results showed an increase in luciferase activity in cells overexpressing the wild-type SIRT3 gene, as compared to the mutant gene (Fig.
4A).
p3400-MnSOD-Luc was also co-transfected into HCT116 cells with either pCMV-SIRT3-wt or pCMV-SIRT3-mt SIRT3 expression vector.
These transient transfection experiments showed that that SIRT3, but not the deacetylation mutant gene, induces MnSOD promoter activity (Fig.
4B).
SIRT3 alters intracellular superoxide levels
SIRT3 is localized in the inner mitochondrial membrane 21 along with the electron transport chain.
The mitochondria are a major source of superoxide formation; and the accumulation of superoxide, from decreased election transport, is thought be a source of oxidative damage associated with degenerative diseases and aging 22.
Thus, it seemed logical to propose that one function of SIRT3 may be to mediate intracellular superoxide levels.
HCT116 based wt-SIRT3 or mt-SIRT3 cells were isolated and intracellular superoxide levels were estimated by DHE oxidation 8.
These experiments showed that SIRT3 deacetylation mutant cells (mt-SIRT3) have significantly increased intracellular superoxide (O2●-) levels 8, relative to either vector control or wt-SIRT3 cells (Fig.
5A).
Intracellular reduced (GSH) and oxidized glutathione (GSSG) concentrations were also altered in mutant mt-SIRT3 cells, relative to either vector control or wt-SIRT3 cells, consistent with a relatively oxidizing intracellular environment in the Sirt3 mutant expressing cells (Fig.
5B).
DISCUSSION
In this study, we suggest that FOXO3a is a mitochondrial protein and its function may be regulated by a physical interaction with SIRT3 in the mitochondria.
This idea is based on the observations that: (1) FOXO3a is present in the mitochondria and physically interacts with SIRT3; (2) SIRT3 increases the DNA-binding of FOXO3a at two promoters resulting in an increase in gene expression; and (3) altered SIRT3 function results in altered and increased intracellular O2●-.
The results of these experiments all support the hypothesis that SIRT3 may be a mitochondrial surveillance factor that detects metabolic imbalances involving O2•- production and induces specific pathways to respond to these conditions.
While it is tempting to suggest that SIRT3 is initiating a signaling pathway in the mitochondria and sending it to the nucleus to activate gene expression via FOXO3a our results do not confirm this specific idea and the experimental assays to validate this idea do not exist.
However, one possibility is that SIRT3 is both a mitochondrial and a cytoplasmic protein.
Several manuscripts have been published suggest that SIRT3 is a mitochondrial protein 21,22 while others imply a more complex cellular localization pattern 24,25.
In our two model systems using HCT116 and SIRT3 knockout derived MEFs, SIRT3 appears present only in the mitochondria.
However, this does not rule out a small amount of the protein might be in either the cytoplasm or nucleus.
Several previous publications suggest that SIRT3 may play a role in the regulation of mitochondrial metabolism 26 as well as acetyl-CoA synthetase acetylation 4.
It also appears that SIRT3 is the primary mammalian Sir2 homolog that regulates global mitochondrial lysine acetylation 1.
However, the mechanism and physiological function of SIRT3 has not been determined.
The results presented here, when coupled with others, suggest that at least one function of SIRT3 serves as a sensor of small reactive oxygen molecules and initiates specific cellular signaling pathways.
In this model the induction of SCO2 and MnSOD would be an appropriate response to detoxification O2•-.
The mitochondria are key organelles that maintain multiple cellular processes and decreases in function are critical to age-related diseases such as diabetes, neurodegeneration, and carcinogenesis 27.
SIRT3 is located near the electron transport chain 28 and is at an ideal location to function as a guardian gene against the accumulation of small reactive molecules that might induce cellular damages.
Mitochondria are a major source of superoxide formed by the reduction of oxygen during electron transport.
The oxidative damage caused by the reactive oxygen species derived from superoxide is believed to be a major cause of degenerative diseases associated with aging 22.
However, it has also been suggested that that physiological levels of O2•- and H2O2 production are essential for normal biological signaling processes leading to the attainment of a normal lifespan 26.
Thus, these results suggest a potential mechanism where the normal production of reactive oxygen species, such as O2●-, induces specific pathways that act to repair and, perhaps more importantly, to renew organelle functions.
On the other hand, the induction of proteins such as SCO2, should decrease O2•- levels indirectly by increasing the efficiency of the four electron reduction of O2.
SCO2 is a mitochondrial membrane-bound protein involved in copper supply for the assembly of cytochrome c oxidase (CcO) 29.
CcO catalyzes the four-electron reduction of molecular oxygen to H2O and couples this reduction with proton translocation across the inner membrane to improve mitochondrial respiration efficiency 30.
CcO increases the efficiency of four electron transfer to O2 to 2H2O in Complex IV and as such, reduces the residence time of electrons at upstream sites capable of mediating one electron reductions of O2.
Thus, the induction of SCO2 could theoretically decrease O2•- production at sites in the electron transport chains upstream of Complex IV (Complexes I, II, and III).
This work also suggests that there may be mitochondrial surveillance genes that sense organelle distress and pass signals to the nucleus to initiate the expression of nuclear genes to prevent or repair any functional damage to mitochondria.
While no direct mechanism is presented for how this signaling is transduced out of the mitochondria, it does seem logical that such a reparative pathway would exist and has been proposed by others 27.
Finally, since SIRT3 is the primary deacetylase in the mitochondria 1 we would propose that deacetylation, perhaps induced by nutrient deprivation or oxidative stress, may activate families of evolutionary conserved reparative genes.
Supplementary Material
Mammalian Sir2 Homolog SIRT3 Regulates Global Mitochondrial Lysine Acetylation
AGEID: a database of aging genes and interventions
Calorie restriction and SIR2 genes--towards a mechanism
Reversible lysine acetylation controls the activity of the mitochondrial enzyme acetyl-CoA synthetase 2
Resveratrol protects dopaminergic neurons in midbrain slice culture from multiple insults
Aging lowers steady-state antioxidant enzyme and stress protein expression in primary hepatocytes
A mutation in succinate dehydrogenase cytochrome b causes oxidative stress and ageing in nematodes
Mutation of succinate dehydrogenase subunit C results in increased O2•-, oxidative stress, and genomic instability
Regulation of the Caenorhabditis elegans longevity protein DAF-16 by insulin/IGF-1 and germline signaling
Tissue-specific activities of C.
elegans DAF-16 in the regulation of lifespan.
Cell cycle and death control: long live Forkheads
FoxOs at the crossroads of cellular metabolism, differentiation, and transformation
SIRT2 deacetylates FOXO3a in response to oxidative stress and caloric restriction
Mitochondria--a nexus for aging, calorie restriction, and sirtuins?
SIRT3, a mitochondrial Sirtuin deacetylase, regulates mitochondrial function and thermogenesis in brown adipocytes
Thioredoxin reductase regulates AP-1 activity as well as thioredoxin nuclear localization via active cysteines in response to ionizing radiation
Distinct effects on gene expression of chemical and genetic manipulation of the cancer epigenome revealed by a multimodality approach
Glucose deprivation-induced oxidative stress in human tumor cells.
A fundamental defect in metabolism?.
DAF-16/FOXO targets genes that regulate tumor growth in Caenorhabditis elegans
Forkhead transcription factor FOXO3a protects quiescent cells from oxidative stress
SIRT3, a human SIR2 homologue, is an NAD-dependent deacetylase localized to mitochondria
Mitochondrial superoxide and aging: uncoupling-protein activity and superoxide production
Where in the cell is SIRT3?--functional localization of an NAD+-dependent protein deacetylase
SirT3 is a nuclear NAD+-dependent histone deacetylase that translocates to the mitochondria upon cellular stress
Localization of mouse mitochondrial SIRT proteins: shift of SIRT3 to nucleus by co-expression with SIRT5
Mechanism of superoxide-mediated damage relevance to mitochondrial aging
Mitochondria damage checkpoint, aging, and cancer
Loss of imprinting in colorectal cancer linked to hypomethylation of H19 and IGF2
The human cytochrome c oxidase assembly factors SCO1 and SCO2 have regulatory roles in the maintenance of cellular copper homeostasis
p53 regulates mitochondrial respiration
The daf-16 homolog, FOX3a forms a complex with SIRT3.
(A) FOXO3a binds to SIRT3 in vitro.
Cos-7 cells were transfected with either SIRT3 wild-type (p-myc-hSIRT3-wt) or deacetylation mutant (p-myc-hSIRT3-mt) vectors and cell lysates were immunoprecipitated (IPd) with an anti-Myc antibody followed by Western analysis with an anti-FOXO3a antibody.
(B) HCT116 cell lysates were IPd with either an anti-FOXO3a or anti-SIRT3 antibody, resolved by SDS-PAGE, and immunoblotted with anti-FOXO3a antibody.
(C) Mitochondrial factions from HCT116 cells were IPd with either an anti-FOXO3a or anti-SIRT3 antibody and immunoblotted with anti-FOXO3a antibody.
The daf-16 homolog, FOX3a is a mitochondrial protein.
Mitochondrial protein fraction from heart (A) and liver (B) (DNA Technologies, ReadyWestern™) were analyzed by western analysis for the presence of the FOXO3a protein.
The PVDF membrane was also probed with both and a-tubulin and a-cytochrome c antibodies that localize to the cytoplasm and mitochondria, respectively.
(C) Mitochondrial protein fractions from of pcDNA, wt-SIRT3, or mt-SIRT3 cell lines were isolated and immunoreactive FOXO3a protein levels were determined.
SIRT3 activates FOXO3a dependent gene expression.
(A) pMnSOD-Luc was transiently co-transfected with CMV based FOXO3a expression vectors encoding the wild-type (pCMV-wt-FOXO3a), dominant negative (pCMV-m-FOXO3a), or dominant positive (pCMV-N-FOXO3a) FOXO3a genes.
Results for all the experiments presented in this figure were normalized to β-galactosidase activity and the results are presented as luciferase/β-galactosidase relative units.
(B) SIRT3 activates FOXO3a dependent gene expression.
A reporter plasmid containing three FOXO DNA-binding sites (p3x-FOXO-Luc) was transfected with either pCMV-SIRT3-wt or pCMV-SIRT3-mt and luciferase activity was determined as described above.
(C) The p3x-FOXO-Luc reporter plasmid was transfected with wild type or mutant FOXO3a expression vectors (pCMV-wt-FOXO3a or pCMV-m-FOXO3a respectively) along with wild type or mutant SIRT3 expression vectors (pCMV-SIRT3-wt or pCMV-SIRT3-mt) and luciferase activity was determined and normalized to β-galactosidase as described above.
All experiments are mean of at least three separate experiments and error bars for all data shown represent one standard deviation and statistical significance was established by Student's t-test.
* Indicates P < 0.05 by t-test.
Cells were fixed with 1% formaldehyde to crosslink protein-DNA interactions, sonicated, and fixed cells were immunoprecipitated with either an anti-FOXO3a antibody.
DNA was eluted and purified before analysis specific primers and visualized by ethidium bromide staining.
Overexpression of SIRT3 induced MnSOD expression.
(A) A MnSOD promoter luciferase reporter plasmid (pMnSOD-Luc) was transfected into wt-SIRT3, mt-SIRT, and vector control cell lines.
Luciferase activity was normalized to β-galactosidase activity and presented as luciferase/β-galactosidase relative units.
All results are the mean of at least three separate experiments.
Error bars around data points represent one standard deviation about the arithmetic mean.
(B) pMnSOD-Luc was co-transfected into HCT116 cells with either pCMV-SIRT3-wt or pCMV-SIRT3-mt expression vectors or luciferase assays were done as described above.
* Indicates P < 0.05 by t-test.
SIRT3 expression alters intracellular superoxide levels.
(A) Altered steady-state levels of superoxide as shown by increased oxidation of DHE in HCT116 SIRT3 overexpressing cells.
Control (pcDNA vector), wild-type SIRT3 (wt-SIRT3), or a deacetylation SIRT3 mutant (mut-SIRT3) cell lines were analyzed by flow cytometry for the amount of hydrolyzed DHE per 10,000 cells represented as Mean Florescent Intensity (MFI).
Each bar represents the average of three separate experiments.
Error bars indicate standard error.
* Indicates P < 0.05 by t-test.
(B) SIRT3 overexpressing cells contain increased GSH/GSSG redox ratios.
Total levels of glutathione (GSH) was determined and normalized to total protein (left panel).
GSSG/GSH ratios were determined and represented as % GSSG (right panel).
Error bars represent of standard deviation of three independent experiments.
* Indicates P < 0.05 by t-test.Ethylene- and pathogen-inducible Arabidopsis acyl-CoA-binding protein 4 interacts with an ethylene-responsive element binding protein
Six genes encode proteins with acyl-CoA-binding domains in Arabidopsis thaliana.
They are the small 10-kDa cytosolic acyl-CoA-binding protein (ACBP), membrane-associated ACBP1 and ACBP2, extracellularly-targeted ACBP3, and kelch-motif containing ACBP4 and ACBP5.
Here, the interaction of ACBP4 with an A.
thaliana ethylene-responsive element binding protein (AtEBP), identified in a yeast two-hybrid screen, was confirmed by co-immunoprecipitation.
The subcellular localization of ACBP4 and AtEBP, was addressed using an ACBP4:DsRed red fluorescent protein fusion and a green fluorescent protein (GFP):AtEBP fusion.
Transient expression of these autofluoresence-tagged proteins in agroinfiltrated tobacco leaves, followed by confocal laser scanning microscopy, indicated their co-localization predominantly at the cytosol which was confirmed by FRET analysis.
Immuno-electron microscopy on Arabidopsis sections not only localized ACBP4 to the cytosol but also to the periphery of the nucleus upon closer examination, perhaps as a result of its interaction with AtEBP.
Furthermore, the expression of ACBP4 and AtEBP in Northern blot analyses was induced by the ethylene precursor 1-aminocyclopropane-1-carboxylic acid, methyl jasmonate treatments, and Botrytis cinerea infection, suggesting that the interaction of ACBP4 and AtEBP may be related to AtEBP-mediated defence possibly via ethylene and/or jasmonate signalling.
Introduction
In Arabidopsis, six genes encode proteins that contain a conserved acyl-CoA-binding domain (Leung et al., 2004).
They are the 10-kDa ACBP6 (Engeseth et al., 1996; Xiao et al., 2008), of which homologues are prevalent in eukaryotes, and larger ACBPs (ACBP1 to ACBP5), of which homologues have not been well-investigated in other organisms.
Some of the larger ACBPs contain ankyrin repeats (ACBP1 and ACBP2), and kelch motifs (ACBP4 and ACBP5) that can potentially interact with protein partners (Li and Chye, 2003, 2004; Leung et al., 2004).
ACBP2 has been previously shown to interact with AtEBP (Li and Chye, 2004).
Our observations that various members of the Arabidopsis ACBP gene family consist of additional structural domains other than the conserved acyl-CoA-binding domain, plus their varying affinities for acyl-CoA esters, imply that they do not have redundant roles in plant lipid metabolism (Chye, 1998; Chye et al., 2000; Leung et al., 2004, 2006).
The acyl-CoA-binding domain in each ACBP has been shown to bind acyl-CoA esters, hence these ACBPs may mediate the subcellular transfer of acyl-CoA esters in plant lipid metabolism (Chye et al., 2000; Leung et al., 2004, 2006).
Membrane-associated ACBP1 and ACBP2 could possibly maintain an acyl-CoA pool at the plasma membrane and participate in membrane biogenesis (Chye, 1998; Chye et al., 1999, 2000; Li and Chye, 2003; Xiao et al., 2008).
We have also shown that transgenic Arabidopsis overexpressing ACBP1 showed enhanced tolerance to Pb(II)-induced stress, implying that ACBP1 could be involved in lipid bilayer membrane repair at the plasma membrane in response to Pb(II) stress (Xiao et al., 2008).
ACBP3 has been demonstrated to be extracellularly-targeted (Leung et al., 2006) while ACBP4 and ACBP5 are predicted to be localized to the cytosol (Leung et al., 2004, 2006).
The preference of ACBP4 and ACBP5 in oleoyl-CoA binding suggests that they could participate in the transfer of oleoyl-CoA esters to the endoplasmic reticulum (ER) from the chloroplasts, in which de novo fatty acid biosynthesis occurs (Leung et al., 2004).
To elucidate the function of ACBP4, the significance of its acyl-CoA-binding domain has been addressed by using site-directed mutagenesis (Leung et al., 2004).
The role of its kelch motifs in mediating protein–protein interactions is investigated here because the identification of its interactors will provide a better understanding of ACBP4 function in planta.
Kelch motifs, structural repeats first observed in the Drosophila actin cross-linking protein kelch, allow protein folding into a cylindrical ‘β-propeller structure’ (Adams et al., 2000) forming a potential protein–protein interaction domain (Andrade et al., 2001).
A bait-containing sequence encoding ACBP4 was constructed for yeast two-hybrid screens using a cDNA library derived from A.
thaliana to identify proteins that interact directly with ACBP4.
Co-immunoprecipitation assays were used to confirm the protein–protein interactions.
Subsequently, localization of ACBP4 and its interacting protein, AtEBP, was confirmed using transient expression of GFP- and DsRed-tagged fusion proteins in Nicotiana tabacum.
When the spatial and temporal expression of ACBP4 and AtEBP was examined by Northern blot analyses, their similar induced expression by the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC), methyl jasmonate (MeJA) treatments, and Botrytis cinerea implicate the feasibility of their potential roles in plant defence.
Materials and methods
Yeast strain
The two-hybrid library screens were performed in the Saccharomyces cerevisiae strain YPB2 [MATa ara3 his3 ade2 lys2 trp1 leu2, 112 canr gal4 gal80 LYS2::GAL1-HIS3, URA3::(GAL1UAS17mers)-lacZ] (Kohalmi et al., 1998).
Cotransformants were plated on synthetic dextrose agar plates lacking leucine, tryptophan, and histidine [SD-leu-trp-his] supplemented with 10 mM 3-AT (Kohalmi et al., 1998).
Construction of a bait vector of GAL4(DB)-ACBP4 fusion
The bait plasmid pAT188 was prepared by inserting a 2 kb XhoI-NotI fragment encoding ACBP4 from pAT181 (Leung et al., 2004) into the SalI-NotI sites of pBI-880 (a variant of pPC62 as described by Chevray and Nathans, 1992; Kohalmi et al., 1998).
All constructs were confirmed by restriction digestion and nucleotide sequence analysis.
Yeast two-hybrid screening
S.
cerevisiae strain YPB2 was transformed with bait plasmid pAT188 and transformants were plated on synthetic dextrose agar plates lacking leucine [SD-leu].
An aliquot of transformants was also tested on [SD-leu-his] medium supplemented with 10 mM 3-amino-1, 2, 4-triazole (3-AT) because an absence of growth on this medium would confirm that the DB-‘bait’ fusion protein is unable to initiate transcription of HIS3.
Subsequently, the bait-carrying strain was tested negative for β-galactosidase activity using the X-Gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) colony filter assay.
This further showed that the bait was not able to activate transcription of the lacZ reporter gene.
The prey vector pBI-771, a variant of pPC86 (Chevray and Nathans, 1992; Kohalmi et al., 1998), was introduced into this strain and its inability to grow on [SD-leu-trp-his] medium supplemented with 10 mM 3-AT and its lack of β-galactosidase activity were confirmed before the bait was further used in cDNA library screening.
To ensure sufficient coverage in the identification of potential proteins interacting with ACBP4, yeast two-hybrid screenings were also performed at the Molecular Interaction Facility, University of Wisconsin–Madison using yeast strains and vectors as previously described by James et al.
(1996).
For bait preparation, ACBP4 (amino acids 1–669) was cloned in-frame with the GAL4 DNA-binding domain of bait vector pBUTE (a kanamycin-resistant version of GAL4 bait vector pGBDUC1).
The resulting vector was subject to DNA sequence analysis to confirm the presence of an in-frame fusion, before use in transformation of S.
cerevisiae mating type strain PJ69-4A, followed by testing for autoactivation of the β-galactosidase reporter gene.
Library screenings were conducted using the Molecular Interaction Facility Arabidopsis library collection representing cDNAs from flowering Arabidopsis plants.
Approximately 50 million clones were screened.
Of these, positive yeast clones were tested for interaction by selection on histidine drop-out and β-galactosidase assays.
Plasmids were rescued and analysed by restriction endonuclease analysis.
Positive prey plasmids were retransformed into the mating type of PJ69-4A and validated in mating and selection assays with the ACBP4 bait, the empty bait vector, and unrelated control baits.
Positive clones were subsequently identified by nucleotide sequence analysis using the GAL(TA)-specific forward primer BC304 (5′-CTATTCGATGATGAAGATACC-3′) and the ADH1-terminator reverse primer, JN069 (5′-TTGATTGGAGACTTGACC-3′) (Kohalmi et al., 1998).
Co-immunoprecipitation
To corroborate the interaction from yeast two-hybrid analysis, co-immunoprecipitation studies were performed according to Mongiat et al.
(2003).
All constructs used in these interaction assays were derivatives of vector pBluescriptII KS(–) (pKS).
The HindIII-SacI fragment from pBI-771 carrying GAL4(TA) (amino acids 768–881) was cloned into corresponding restriction sites on pKS.
The GAL4(TA)-ACBP4 fusion construct was prepared by inserting ACBP4 cDNA from pAT181, on a 2 kb EcoRI-BamHI fragment, into the EcoRI-BglII sites of pKS-TA with the 5′ of TA-ACBP4 adjacent to the T3 promoter.
Two putative interactors, ADF3 (identified at the Molecular Interaction Facility, University of Wisconsin–Madison) and AtEBP (from a yeast two-hybrid screen in our laboratory) were selected for further studies.
Their full-length cDNAs were generated by the Reverse-Transcriptase-Polymerase Chain Reaction (RT-PCR) using the Superscript™ First-strand synthesis system (Invitrogen, Carlsbad, CA, USA).
The cDNA fragments were subsequently cloned into pGEM-T Easy (Promega, Madison, WI, USA).
Potential ‘ATG’ start codons in the multiple cloning sites of pGEM-T Easy vector upstream of the ADF3 or AtEBP cDNA were eliminated by restriction endonuclease digestion followed by filling-in with Klenow and re-ligation.
The cDNAs of both ADF3 and AtEBP were verified by nucleotide sequence analysis.
Subsequently, GAL4(TA)-ACBP4 and each candidate were in vitro transcribed and translated by a TNT quick coupled wheat germ transcription-translation system (Promega, Madison, WI, USA) in the presence of [35S]methionine (ICN Pharmaceuticals Inc., Costa Mesa, CA, USA), according to the manufacturer's instructions.
The proteins were analysed by 12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS–PAGE) and autoradiography.
Co-immunoprecipitation with monoclonal anti-GAL4(TA) antibody (Clontech, USA) was performed following Mongiat et al.
(2003).
Construction of plasmids used in subcellular localization
All binary vectors used in this study were derivatives of plasmids pGDG and pGDR which contain genes encoding the autofluorescent proteins GFP and DsRed, respectively (Goodin et al., 2002).
The 2 kb XhoI-BamHI fragment encoding the complete ACBP4 peptide was generated by PCR using primers ML350 and ML682 with pAT181 as template, and cloned into pGEM-T Easy vector to generate plasmid pAT280.
The 2 kb XhoI-BamHI ACBP4 fragment derived from plasmid pAT280 was cloned into the XhoI and BamHI sites of pGD-DsRed to obtain pAT282 in which ACBP4 is fused to 5′ of DsRed.
The plasmid pAT225 in which AtEBP is fused to 3′ of GFP has been previously described (Li and Chye, 2004).
The cloning junctions in all constructs were confirmed by nucleotide sequence analysis.
Transient expression by agroinfiltration
Tobacco (Nicotiana tabacum var.
Xanthi) plants were grown in a greenhouse at 22 °C for 6 weeks.
Two days before agroinfiltration, they were maintained in a growth chamber at 22 °C under 16/8 h light/dark as specified by Goodin et al.
(2002).
Derivatives of Agrobacterium tumefaciens strain LBA4404 containing autofluorescent protein fusion constructs were cultured on LB solid medium supplemented with kanamycin (50 μg ml−1) and streptomycin (25 μg ml−1) at 28 °C for 2 d.
For agroinfiltration, Agrobacterium was grown at 28 °C overnight, in LB medium supplemented with kanamycin (50 μg ml−1) and streptomycin (25 μg ml−1).
Preparation of Agrobacterium suspension and agroinfiltration of tobacco leaves in planta were carried out following the procedures of Yang et al.
(2000).
Confocal laser-scanning microscopy
Tobacco leaf epidermal cells from agroinfiltration were examined under a Zeiss LSM 510 inverted confocal laser-scanning microscope (Zeiss, Jena, Germany) following the settings described by Goodin et al.
(2002) with minor modifications.
Single optical sections were scanned as resulting images for each transient expression.
For each plasmid construct, 10–15 cells were imaged with similar results.
GFP fluorescence was excited at 488 nm, filtered through a primary dichroic (UV/488/543), a secondary dichroic of 545 nm, and subsequently through BP505–530 nm emission filters to the photomultiplier tube (PMT) detector.
DsRed fluorescence was excited at 543 nm, the emission was passed through similar primary and secondary dichroic mirrors and finally through a BP560–615 nm emission filter to the PMT detector.
Fluorescence resonance energy transfer (FRET) pairs GFP/DsRed were analysed using a confocal laser-scanning microscope (Zeiss LSM510 META).
FRET measurements of DsRed emission with zero contribution from GFP, was accomplished as described by Erickson et al.
(2003) using the following settings: excitation at 488 nm and emission filters, BP 505–530 nm for GFP and BP 600–637 nm for DsRed.
Western blot analysis
Protein extracts were prepared by homogenizing Arabidopsis protein from 3-week-old wild-type (Col-0) Arabidopsis rosettes according to Chye et al.
(1999).
Total proteins were separated on SDS–PAGE and transferred onto Hybond-C membranes (Amersham).
The blots were blocked in TTBS (TBS plus 0.05% Tween 20) containing 5% non-fat milk for 2 h and incubated for an additional 2 h with anti-ACBP4 primary antibodies.
The blots were washed three times with TTBS and then incubated with secondary antibody for 1 h.
Either the Amplified Alkaline Phosphatase Goat Anti-rabbit Immuno-blot Assay Kit (BioRad) or the ECL Western Blotting Detection Kit (Amersham) was used following the manufacturer's instructions to detect cross-reacting bands.
ACBP4-specific antibodies were generated by rabbit immunization using a synthetic peptide RMQTLQLRQELGEAE (corresponding to amino acids 566 to 580 of ACBP4).
Immuno-electron microscopy
Arabidopsis leaves were fixed in a solution of 4% (v/v) paraformaldehyde and 0.5% (v/v) glutaradehyde in 0.1 M phosphate buffer (pH 7.2) for 20 min under vacuum and then a further 3 h at room temperature.
The specimens were then dehydrated in a graded ethanol series, infiltrated in stepwise increments of LR white resin (London Resin, Theale, Berkshire, UK) and polymerized at 45 °C for 24 h.
Materials for immuno-gold labelling were prepared according to the procedure of Varagona and Raikhel (1994) with the modification as described.
Specimens (90 nm) were sectioned using a Leica Reichert Ultracut S microtome and mounted on formvar-coated slotted grids.
Grids were incubated in a blocking solution of TTBS containing 1% (w/v) fish skin gelatin and 1% (w/v) BSA for 30 min.
Anti-ACBP4 antibodies diluted 1:50 in blocking solution were added and incubated at room temperature for 2 h.
The grids were then rinsed three times, each for 5 min, in TTBS and then incubated with 10 nm gold-conjugated goat anti-rabbit IgG secondary antibody (Sigma), diluted 1:20 with blocking solution.
Grids were rinsed three times, each for 5 min in TTBS, following by three 5-min rinses in distilled water.
After being stained in 2% (w/v) uranyl acetate for 6 min followed by 2% (w/v) lead citrate for 6 min, the sections were visualized and photographed using Philips EM208s electron microscope operating at 80 kV.
Controls were performed excluding the primary antibody.
Plant materials, growth conditions and treatment
Tobacco (N.
tabacum var.
Xanthi) plants were grown in a greenhouse at 22 °C for 6 weeks.
Two days before agroinfiltration, they were maintained in a growth chamber at 22 °C under 16/8 h dark/light as specified by Goodin et al.
(2002).
Arabidopsis thaliana ecotype Columbia (Col-0) was grown under cycles of 8 h dark at 21 °C and 16 h light at 23 °C.
For Arabidopsis treatments in northern blot experiments, seedlings were grown on Murashige and Skoog (1962) medium with 2% sucrose in continuous light for 2–3 weeks and then treated with 1 mM 1-aminocyclopropane-1-carboxylic acid (ACC, Sigma-Aldrich, St Louis), 100 μM methyl jasmonate (MeJA, Sigma-Aldrich, St Louis) or water (control).
Plant samples were collected at 0, 4, 8, 12, and 24 h post-treatment.
Pathogen infection
Three-week-old wild-type Arabidopsis plants were inoculated with Botrytis cinerea by spraying with a spore suspension (2×105 spores ml−1) in a solution containing 1% glucose or with water containing 1% glucose as a control.
After inoculation, the plants were placed in a growth chamber with high humidity (100%) at 22 °C under a 16/8 h light/dark photoperiod as described by Xiao et al.
(2004).
Plant samples were collected at 0, 24, 48, and 72 h post-inoculation.
Northern blot analysis
Total RNA was isolated from plant tissues following the procedure of Nagy et al.
(1988).
Northern blot analysis was performed as described previously (Xiao et al., 2004).
Briefly, 30 μg of total RNA were separated on a 1.5% agarose gel containing 6% formaldehyde and transferred to Hybond N membranes (Amersham).
To generate probes for use in Northern blot analyses, specific primers were designed for PCR-amplification: ACBP4 (ML350, 5′-CCTCGAGAATGGCTATGCCTAGGGC-3′ and ML682, 5′-GGATCCACAAGGCGAATCATCATCT-3′), AtEBP (ML826, 5′-ACAGAGAAAATGTGTGGCGG-3′ and ML827, 5′-CAAGCATCCACATAT CCACC-3′) and PDF1.2 (ML741, 5′-TAAGTTTGCTTCCATCATCACCC-3′ and ML742, 5′-TTAACATGGGACGTAACAGATACA-3′).
Templates used in PCR were plasmid pAT282 (consisting of the ACBP4 cDNA) and the first-strand wild-type pool of cDNAs (for AtEBP and PDF1.2).
The fragments were labelled with the PCR Digoxigenin Probe Synthesis Kit according to the manufacturer's instructions (Roche, Germany).
Hybridization and detection were performed according to the standard procedures as advised by the manufacturer (Roche).
The blots were washed under conditions of high stringency (2× SSC, 0.1% SDS for 2×15 min at room temperature; 0.5× SSC, 0.1% for 2×15 min at 68 °C; 0.1× SSC, 0.1% SDS for 2×15 min at 68 °C).
Results
Yeast two-hybrid screening
The yeast YPB2 transformed with the bait GAL4(DB)-ACBP4 could not grow on [SD-leu-his] and was tested negative on X-Gal colony filter assays (data not shown), suggesting that the pAT188 bait alone could not activate the transcription of reporter genes HIS3 and lacZ and was deemed appropriate for two-hybrid screens.
A GAL4(TA) tagged A.
thaliana cDNA library was introduced into the yeast YPB2 harbouring plasmid pAT188.
The number of independent transformants was determined to be 2×106 following transformation and plating of an aliquot of the yeast transformation mixture on [SD-leu-trp].
A total of 100 putative positives were selected on [SD-leu-trp-his] supplemented with 10 mM 3-AT medium.
When these putative positives were further screened for β-galactosidase activity using the X-Gal colony filter assay, nine yeast clones that appeared blue, at varying intensities, were identified as putative clones encoding interactors.
Putative library plasmids were retrieved and their nucleotide sequences were searched against the BLAST server http://www.ncbi.nlm.nih.gov/cgi-bin/BLAST.
Only one clone was in-frame to GAL4(TA), encoding a full-length ethylene-responsive element binding factor (ERF) protein AtEBP (Arabidopsis genome locus: AT3G16770).
An AP2/EREBP (ethylene-responsive element binding protein) domain is present in AtEBP at amino acids 76–143 (Okamuro et al., 1997).
In another independent yeast two-hybrid screen using the Molecular Interaction Facility (University of Wisconsin–Madison), six putative positives were identified following selection on histidine drop-out and β-galactosidase assays.
Subsequently, they were used to retransform yeast mating type strain PJ69-4A, and were validated in mating and selection assays using the ACBP4 bait, the empty bait vector, and unrelated baits.
Five clones were tested positive and further identified by nucleotide sequence analysis.
Results from analysis using the BLAST revealed that only one clone was in-frame and it encoded a full-length actin-depolymerizing factor 3 (ADF3, At5g59880) protein.
However, this was not investigated further in this study because it did not bind ACBP4 in subsequent co-immunoprecipitation, possibly due to the absence of some as yet unidentified essential cofactor(s) for binding in an in vitro co-immunoprecipitation reaction.
Results of X-Gal filter assays are shown in Fig.
1A.
Positive protein–protein interaction results in activation of the reporter gene β-galactosidase in yeast cells, which turns yeast colonies blue in filter assays using X-Gal.
Without interaction, the yeast colonies remain ‘colourless’.
As shown in Fig.
1Aa, the GAL4(DB)-ACBP4 fusion interacted with GAL4(TA)-AtEBP, as indicated by the blue colour arising from the production of significant levels of β-galactosidase.
No interactions were observed in control yeast cells harbouring GAL4(DB)-ACBP4+GAL4(TA) (Fig.
1Ab) and GAL4(DB)+GAL(TA)-AtEBP (Fig.
1Ac).
Therefore, from yeast two-hybrid analysis, AtEBP was identified as a putative protein that interacts with ACBP4.
(A) Colony filter β-galactosidase assays of candidate proteins AtEBP from yeast two-hybrid screens.
(a) YPB2/GAL4(DB)-ACBP4+GAL4(TA)-AtEBP; (b) YPB2/GAL4(DB)-ACBP4+GAL4(TA); (c) YPB2/GAL4(DB)+GAL(TA)-AtEBP.
(B) Co-immunoprecipitation of ACBP4 and AtEBP using the anti-GAL4(TA) monoclonal antibody.
Autoradiograph of a 12% SDS-PAGE (left panel) showing the in vitro transcribed and translated ADF3, AtEBP, and GAL4(TA)-ACBP4, respectively, as indicated.
The right panel shows the co-immunoprecipitation of equimolar amounts of GAL4(TA)-ACBP4 and ADF3 or AtEBP using the anti-GAL4(TA) antibody.
Arrows indicate the positions of these proteins.
Corroboration of ACBP4-interacting proteins by co-immunoprecipitation
Proteins, generated from plasmid derivatives of pBluescriptII KS using in vitro transcription/translation, were analysed by 12% SDS-PAGE.
An autoradiograph of the gel showed that the estimated molecular masses of the in vitro translation products of GAL4(TA)-ACBP4, AtEBP and ADF3, were 84 kDa, 28 kDa, and 16 kDa, respectively, according to their calculated molecular masses (Fig.
1B).
Co-immunoprecipitation of in vitro transcription/translation products to the GAL4(TA)-ACBP4 fusion protein, immobilized to protein A/agarose beads, using monoclonal antibody against GAL4(TA), showed that the GAL4(TA)-ACBP4 fusion protein significantly binds AtEBP (Fig.
1B).
However, no binding of GAL4(TA)-ACBP4 to ADF3 was observed (Fig.
1B), perhaps due to the lack of cofactors which must be present for their in vitro interaction.
Co-localization of ACBP4:DsRed and GFP:AtEBP
To verify the subcellular localization of ACBP4 and AtEBP in vivo, ACBP4 was tagged to the N-terminus of DsRed, and the fusion from the CaMV 35S promoter expressed while AtEBP was tagged to the C-terminus of GFP.
Following agroinfiltration of tobacco leaves with both ACBP4:DsRed and GFP:AtEBP, observations were carried out by confocal microscopy using a green filter to investigate the fluorescence pattern of GFP and a red filter to visualize the fluorescence of DsRed (Fig.
2).
GFP:AtEBP was located mainly in the nucleus, with some signals at the cytosol and the plasma membrane (Fig.
2A).
ACBP4:DsRed was localized predominantly to the cytosol, inclusive of signals detected in the cytosol surrounding the nucleus (Fig.
2B).
Signals of both fusion proteins were common to the cytosol.
Confocal images indicating co-localization of ACBP4:DsRed and GFP:AtEBP fusion proteins transiently-expressed in tobacco leaves.
Representative tobacco leaf epidermal cells are shown by laser-scanning confocal microscopy following agroinfiltration of plasmid pAT282 or pAT225.
(A) GFP:AtEBP expressed from pAT225 in a tobacco leaf.
(B) ACBP4:DsRed expressed from pAT282 in a tobacco leaf.
(C–F) FRET detection in tobacco leaf epidermal cells co-expressing GFP:AtEBP and ACBP4:DsRed; (C) differential interference contrast image of D-F; (D) green channel shows GFP:AtEBP; (E) red channel shows FRET signal of ACBP4:DsRed; (F) co-localization of two signals is indicated by a yellow colour in merged images of (D) and (E).
Arrowheads indicate the position of nuclei.
Bar=20 μm.
In FRET analysis, in cells co-expressing GFP:AtEBP and ACBP4:DsRed, not only GFP:AtEBP green fluorescence (Fig.
2D) but also ACBP4:DsRed red fluorescence (Fig.
2E), which overlapped with the GFP signals (Fig.
2F), were detected, indicating that FRET occurred between GFP:AtEBP and ACBP4:DsRed.
Detection of ACBP4 protein in Arabidopsis
Results from Western blot analysis using total protein from 3-week-old Arabidopsis revealed that anti-ACBP4 antibodies cross-reacted with a band of apparent molecular mass of 73.1 kDa (Fig.
3A, lane 3), as previously predicted for ACBP4 (Leung et al., 2004).
Localization of ACBP4 in Arabidopsis leaves.
(A) Western blot analysis using affinity-purified anti-peptide antibodies against ACBP4.
Lanes 1 and 2: gel identically loaded as lanes 3 and 4, respectively, stained with Coomassie blue to show the amount of protein blotted in western blot analysis.
Lanes 3 and 4: western blot analysis of total protein using anti-ACBP4 antibodies (lane 3) and preimmune serum (lane 4).
Lane M, molecular marker in kDa.
(B, C, D) Immuno-gold labelling of ACBP4 in an Arabidopsis leaf cell using transmission electron microscopy.
Transverse sections were stained with affinity-purified ACBP4-specific antibodies.
(B) Transverse sections of leaves stained with ACBP4-specific antibodies.
(C) Magnification of the boxed area in (B).
(D) Control labelling of a leaf cell using secondary antibodies alone.
Arrowheads, gold particles.
V, vacuole; C, cytosol; Ch, chloroplast; N, nucleus; Cw, cell wall; Bars in (B) represent 2 μm, and in (C, D), 0.2 μm.
Immuno-electron microscopy was carried out using transverse sections of leaves of 2-week-old Arabidopsis germinated and grown in MS medium under a 16/8 h light/dark regime.
Although immuno-gold labelling with the anti-ACBP4 antibodies was mostly evident in the cytosol, some signals were detected at the periphery of the nucleus, (Fig.
3B, C).
In the control, when the primary antibody was replaced by blocking solution, no significant immuno-gold labelling was observed (Fig.
3D).
The immunolocalization of signals at the periphery of the nucleus may have culminated from the interaction of ACBP4 with AtEBP.
ACBP4 and AtEBP show overlapping expression patterns
To address the coexpression of ACBP4 and its interactor, AtEBP, further, their spatial expression patterns were examined.
Northern blot analyses were carried out using ACBP4 and AtEBP full-length cDNA probes, generated in PCR using gene-specific primers.
Both ACBP4 and AtEBP accumulated in leaves and stems (Fig.
4, lanes L and S) of Arabidopsis young seedlings, with lower expression in the flowers and siliques (Fig.
4, lanes F and Si).
ACBP4, but not AtEBP, showed higher expression in roots (Fig.
4, lane R).
Taken together, ACBP4 and AtEBP appear to have some overlapping expression patterns in leaves and stems, which may represent the potential organs for their interaction in vivo.
Expression patterns of ACBP4 and AtEBP in Arabidopsis on Northern blot analyses.
Total RNAs were extracted from wild-type Arabidopsis leaves (L), stems (S), roots (R), flowers (F), and siliques (Si).
A gel blot containing about 30 μg total RNA for each lane was hybridized with an ACBP4-specific cDNA probe.
The membrane was stripped and reprobed with an AtEBP-specific full-length cDNA probe.
Ethidium bromide staining of rRNAs shown at the bottom indicates the relative amounts of total RNA loaded in each lane.
The blots were washed under conditions of high stringency.
Expressions of ACBP4 and AtEBP are induced by ACC and MeJA treatments and by Botrytis infection
It has been reported that in young Arabidopsis seedlings, ethephon induces the expression of AtEBP after 12 h, indicating that AtEBP is involved in ethylene signalling (Büttner and Singh, 1997).
As ACBP4 was shown to interact with AtEBP in vitro and both displayed some similarity in spatial expression, it was investigated to find whether ACBP4 is regulated by ethylene and/or jasmonates.
To this end, 2-week-old Arabidopsis seedlings were treated with 1 mM 1-aminocyclopropane-1-carboxylic acid (ACC, the direct precursor of ethylene) and 100 μM methyl jasmonate.
AtEBP mRNA and ACBP4 mRNA were induced in seedlings at 4, 8, 12, and 24 h (for ACBP4) or 8, 12, and 24 h (for AtEBP) following treatment with ACC and MeJA (Fig.
5A).
An ACC-inducible and MeJA-inducible gene encoding plant defensin PDF1.2 (Penninckx et al., 1998) was used as a positive control in these experiments.
Induction of AtEBP (At3g16770) and ACBP4 (At3g05420) after ACC and MeJA treatments detected in Northern blot analysis was compared with information available from microarray data analysis on AtEBP expression (www.weigelworld.org/resources/microarray).
The expression of both AtEBP and ACBP4 were not inducible in microarrays at 1 h and 3 h after ACC and MeJA treatments and no data were available for a period exceeding 4 h.
Northern blot analyses of ACBP4 and AtEBP expression following 1-aminocyclopropane-1-carboxylic acid and methyl jasmonate treatments and B.
cinerea infection.
(A) Accumulation of ACBP4 and AtEBP transcripts in wild-type Arabidopsis plants grown in MS media under continuous light following treatment with 1 mM ACC and 100 μM MeJA.
Thirty micrograms total RNA per lane were hybridized to the respective probes as indicated on the right of the figure.
The ACC- and MeJA-inducible PDF1.2 transcript was used as a positive control.
Ethidium bromide-stained rRNA is shown below the blots to indicate the relative amounts of total RNA loaded per lane; hpt, hours post-treatment.
(B) Inducible expression of ACBP4 and AtEBP transcripts in wild-type Arabidopsis plants infected with B.
cinerea.
Thirty micrograms total RNA per lane was hybridized to the respective probes as indicated on the right of the figure.
Ethidium bromide-stained rRNA is shown to indicate the relative amounts of total RNA loaded per lane; hpi, hours post-inoculation.
The blots were washed under conditions of high stringency.
In Arabidopsis, both ethylene and jasmonate have been reported to be essential for the induction of a functional defence response towards the necrotrophic fungal pathogen Botrytis cinerea (Thomma et al., 2001; Diaz et al., 2002).
Expression of the ethylene downstream regulator ERF1 is up-regulated upon infection by B.
cinerea (Berrocal-Lobo et al., 2002).
Since ACBP4 and AtEBP mRNAs accumulated in response to ethylene and jasmonate treatments, Arabidopsis plants were subsequently infected with B.
cinerea and tested for ACBP4 and AtEBP expression.
Both ACBP4 and AtEBP mRNAs accumulated in the infected plants at 48 h and 72 h post-inoculation, while the control plant remained uninduced at these corresponding time points (Fig.
5B).
Our findings again suggest that both ACBP4 and AtEBP are probably associated with the ethylene- and jasmonate-mediated plant defence responses.
Discussion
Kelch-motif containing ACBP4 was used as bait in yeast two-hybrid screens from which an interactor (AtEBP) was retrieved.
The interaction of AtEBP and ACBP4 was further substantiated by co-immunoprecipitation and by using autofluorescent protein fusions in the transient expression of tobacco leaf epidermal cells.
ACBP4 and AtEBP showed overlapping expression patterns in leaves and stems and both were inducible by ACC, MeJA treatment, and infection with the fungal pathogen, Botrytis cinerea.
Co-localization of ACBP4 and AtEBP
AtEBP was predicted to be targeted to the nucleus using the PSORT server for the prediction of the subcellular localization of proteins (http://psort.nibb.ac.jp).
However, another server LOCtree (http://cubic.bioc.columbia.edu/services/loctree/; Nair and Rost, 2005) scored the Reliability Index (RI) value of AtEBP nuclear localization to be merely 1, in a range of RI values from 1–10, with 10 denoting the most confident prediction.
LOCtree is a novel system of support vector machines that predict subcellular localization by the incorporation of a hierarchical ontology of localization classes modelled onto biological processing pathways (Nair and Rost, 2005).
It is significantly more accurate than other traditional networks at predicting subcellular localization (Nair and Rost, 2005).
In this study, GFP:AtEBP was not confined to the nucleus but was also detected in the cytosol where it could interact with ACBP4.
ACBP4:DsRed, transiently-expressed in tobacco leaves, was predominantly targeted to the cytosol but immuno-electron microscopy indicated localization of ACBP4 in the cytosol with signals detected at the periphery of the nucleus, perhaps as a consequence of its interaction with AtEBP.
Many protein factors, such as the photoreceptor phytochrome B, COP1, and some bZIP transcription factors demonstrate light-regulated movement between the cytoplasm and the nucleus (Yamamoto and Deng, 1999).
Also, some transcription factors such as the Arabidopsis floral identity protein LEAFY (LFY), do move between cells (Wu et al., 2003).
Therefore, interactions between ACBP4 and AtEBP at the cytosol, a location common to both may permit their similar translocation across subcellular compartments or between cells.
Interaction of ACBPs and transcription factors
When Arabidopsis cDNA libraries were screened for interacting proteins of Arabidopsis ACBPs by the yeast two-hybrid system in our laboratory, as well as at the Molecular Interaction Facility, University of Wisconsin–Madison, while we could not identify protein interactors for ACBP5, a common interacting protein (AtEBP) was identified for both ACBP2 (Li and Chye, 2004) and ACBP4.
AtEBP, containing one AP2/EREBP domain, belongs to the ERF subfamily of AP2/EREBP family of plant transcription factors involved in plant growth and developmental regulation (Riechmann and Meyerowitz, 1998).
The conserved AP2/EREBP domain unique to plants, has been reported to be involved in DNA binding (Ohme-Tagaki and Shinshi, 1995) and in mediating protein–protein interactions (Okamuro et al., 1997).
Proteins of the ERF subfamily have been demonstrated to be mainly expressed in response to biological or physical stress, such as pathogen attack, ethylene or abscisic acid (ABA) treatment, drought, and cold treatment (Zhang et al., 2004; Zhang et al., 2005).
Recently, Ogawa et al.
(2005) demonstrated that AtEBP conferred resistance to Bax and abiotic stress-induced plant cell death in plant cells overexpressing AtEBP.
Furthermore, the function of AtEBP as a transcriptional activator may be related to ethylene signalling based on the analysis of gene expression levels in ethylene-related mutants (Ogawa et al., 2005).
Several ERF proteins have been reported to interact with other proteins including a transcriptional factor, a nitrilase-like protein, and an ubiquitin-conjugated enzyme (Büttner and Singh, 1997; Xu et al., 1998; Koyama et al., 2003).
AtEBP was reported to interact in particular with an ocs element binding protein (Büttner and Singh, 1997) and ACBP2 (Li and Chye, 2004).
A highly conserved motif RAYD element within the AP2/EREBP domain contains a conserved core region that is predicted to form an amphipathic α-helix.
This α-helical structure has been implicated in a role in DNA binding or in mediating protein–protein interactions important for RAP2.3 (AtEBP) function (Okamuro et al., 1997).
The interactions between ACBPs and AtEBP imply that certain ACBPs could be involved in the regulation of plant development or defence through interactions with the transcription factor AtEBP.
Long-chain acyl-CoAs have been demonstrated to regulate gene expression in bacteria, yeast, and mammals (Black et al., 2000).
Petrescu et al.
(2003) has reported that recombinant mouse ACBP in rat hepatoma cells and in transfected COS-7 cells interacts with the hepatocyte nuclear factor-4α (HNF-4α), a nuclear binding protein that regulates the transcription of genes involved in lipid and glucose metabolism.
HNF-4α also catalyses the hydrolysis of bound long-chain fatty acyl-CoAs, and subsequently binds the fatty acid product, thus allowing cross-talk between acyl-CoA-binding sites and free fatty acid binding sites in HNF-4α (Hertz et al., 2005).
ACBP4 may play a role in plant defence and related signalling pathways
ACBP4 has been reported to bind oleoyl-CoA esters in vitro (Leung et al., 2004).
Enzymes that use acyl-CoA esters but do not contain any acyl-CoA-binding domain could possibly dock to acyl-CoA-binding proteins, via protein–protein interactions at the ankyrin repeats of ACBP1 and ACBP2 or via such interactions at the kelch motifs of ACBP4 and ACBP5, to retrieve acyl-CoA substrates.
If ACBP4 were a transporter and pool former of acyl-CoA esters, it would donate acyl-CoA esters to regulatory factors reminiscent of yeast ACBP in the gene regulation of OLE1, in which saturated fatty acids induce OLE1 transcription while unsaturated fatty acids repress its expression (Choi et al., 1996).
In plants, fatty acid-derived signals have been implicated in the regulation of plant defence and development (Farmer et al., 1998).
Calcium-independent phospholipase A2β, a multifunctional signalling enzyme that catalyses the hydrolysis of saturated fatty acyl-CoAs at physiologically relevant concentrations, is selectively autoacylated by oleoyl-CoA, is protected from autoacylation by Ca2+-activated calmodulin, and is rescued from calmodulin-mediated inhibition of phospholipase A2 activity by oleoyl-CoA (Jenkins et al., 2006).
The present study demonstrates that ACBP4, like its identified protein partner AtEBP, is induced by the defence signals ethylene, and jasmonate, and the fungal pathogen B.
cinerea.
The clear roles of ethylene and jasmonate in plant defence signalling, development, and in environmental stress mitigation is relatively well-established.
Our results now suggest a possible role for ACBP4, working in conjunction with AtEBP, in mediating plant defence- and ethylene-related signalling pathways.
While the precise roles for ACBP4 and AtEBP need to be addressed further, it appears that the roles of the new family of six Arabidopsis ACBPs (Leung et al., 2004) are not restricted to binding acyl-CoAs in various subcellular compartments in plant lipid metabolism, but may possibly be extended to the transfer of acyl-CoAs in relation to plant defence- and ethylene-related signalling.
The kelch repeat superfamily of proteins: propellers of cell function
A combination of the F-box motif and kelch repeats defines a large Arabidopsis family of F-box proteins
Constitutive expression of ETHYLENE-RESPONSE-FACTOR1 in Arabidopsis confers resistance to several necrotrophic fungi
Long-chain acyl-CoA-dependent regulation of gene expression in bacteria, yeast and mammals
Arabidopsis thaliana ethylene-responsive element binding protein (AtEBP), an ethylene-inducible, GCC box DNA-binding protein interacts with an ocs element binding protein
Protein interaction cloning in yeast: identification of mammalian proteins that react with the leucine zipper of Jun
Regulatory elements that control transcription activation and unsaturated fatty acid-mediated repression of the Saccharomyces cerevisiae OLE1 gene
Arabidopsis cDNA encoding a membrane-associated protein with an acyl-CoA binding domain
Isolation of a gene encoding Arabidopsis membrane-associated acyl-CoA-binding protein and immunolocalization of its gene product
Single amino acid substitutions at the acyl-CoA-binding domain interrupt 14[C]palmitoyl-CoA binding of ACBP2, an Arabidopsis acyl-CoA-binding protein with ankyrin repeats
The role of ethylene and wound signaling in resistance of tomato to Botrytis cinerea
Characterization of an acyl-CoA-binding protein from Arabidopsis thaliana
DsRed as a potential FRET partner with CFP and GFP
Fatty acid signaling in Arabidopsis
pGD vectors: versatile tools for the expression of green and red fluorescent protein fusions in agroinfiltrated plant leaves
Thioesterase activity and acyl-CoA/fatty acid cross-talk of hepatocyte nuclear factor-4α
Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast
Highly selective hydrolysis of fatty acyl-CoAs by calcium-independent phospholipase A2β: enzyme autoacylation and acyl-CoA-mediated reversal of calmodulin inhibition of phospholipase A2 activity
Identification and characterization of protein interactions using the yeast 2-hybrid system
Isolation of tobacco ubiquitin-conjugating enzyme cDNA in a yeast two-hybrid system with tobacco ERF3 as bait and its characterization of specific interaction
ACBP4 and ACBP5, novel Arabidopsis acyl-CoA-binding proteins with kelch motifs that bind oleoyl-CoA
Arabidopsis ACBP3 is an extracellularly targeted acyl-CoA-binding protein
Membrane localization of Arabidopsis acyl-CoA binding protein ACBP2
Arabidopsis Acyl-CoA-binding protein ACBP2 interacts with an ethylene-responsive element-binding protein, AtEBP, via its ankyrin repeats
Perlecan protein core interacts with extracellular matrix protein 1 (ECM1), a glycoprotein involved in bone formation and angiogenesis
Analysis of gene expression in transgenic plants
Mimicking cellular sorting improves prediction of subcellular localization
Functional analysis of Arabidopsis ethylene-responsive element binding protein conferring resistance to Bax and abiotic stress-induced plant cell death
Ethylene-inducible DNA binding proteins that interact with an ethylene-responsive element
The AP2 domain of APETALA2 defines a large new family of DNA binding proteins in Arabidopsis
Concomitant activation of jasmonate and ethylene response pathways is required for induction of a plant defensin gene in Arabidopsis.
Physical and functional interaction of acyl-CoA binding protein (ACBP) with hepatocyte nuclear factor-4α (HNF-4α)
The AP2/EREBP family of plant transcription factors
The complexity of disease signaling in Arabidopsis
Immunocytochemistry for light and electron microscopy
Modes of intercellular transcription factor movement in the Arabidopsis apex
COS1: an Arabidopsis coronatine insensitive1 suppressor essential for regulation of jasmonate-mediated plant senescence and defence
Overexpression of membrane-associated acyl-CoA-binding protein ACBP1 enhances lead tolerance in Arabidopsis
A nitrilase-like protein interacts with GCC box DNA-binding proteins involved in ethylene and defence responses
Protein nucleocytoplasmic transport and its light regulation in plants
In vivo analysis of plant promoters and transcription factors by agroinfiltration of tobacco leaves
Tomato stress-responsive factor TSRF1 interacts with ethylene responsive element GCC box and regulates pathogen resistance to Ralstonia solanacearum
Expressing TERF1 in tobacco enhances drought tolerance and abscisic acid sensitivity during seedling development