| These two authors contributed equally to this work.To whom correspondence should be addressed: Kenji Kohno, Laboratory of Molecular and Cell Genetics, Graduate School of Biological Sciences, Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0192, Japan. Tel: +81–743–72–5640, Fax: +81–743–72–5649 E-mail: kkouno@bs.naist.jp Abbreviations: ER, endoplasmic reticulum; ERAD, ER-associated degradation; CFTR, cystic fibrosis transmembrane conductance regulator; TCRα, T cell-receptor α subunit; tsO45-VSV-G, temperature-sensitive mutant of the vesicular stomatitis virus G protein; NHK, α1-antitrypsin-null Hong Kong. |
The endoplasmic reticulum (ER) is a cellular compartment where newly synthesized secretory and membrane proteins are folded and modified. Proteins that failed to be correctly folded are pulled out or retrotranslocated from the ER to the cytosol, and polyubiquitinated for degradation by the proteasome, which is known as ER-associated degradation (ERAD) (Wiertz et al., 1996; Tsai et al., 2002; Vembar and Brodsky, 2008). It is widely known that Hsp70 family molecular chaperones contribute to ERAD (Meacham et al., 2001; Okuda-Shimizu and Hendershot, 2007; Nakatsukasa et al., 2008). In general, recognition of substrate proteins by Hsp70 family members requires DnaJ-proteins (Hsp40 family), which have the conserved J-domain sequence for interaction with the specific partner Hsp70 family chaperones (Schlenstedt et al., 1995; Nishikawa et al., 2001). An ER-located J-protein ERdj5/JPDI is reported to function together with an ER luminal Hsp70 family chaperone BiP to promote the degradation of misfolded luminal proteins (Ushioda et al., 2008). Cytosolic Hsp70 family chaperones, Hsc70 and/or Hsp70, hereafter called Hsc70/Hsp70, likely contribute to ERAD of membrane proteins for the following reasons. Firstly, Hsc70 associates with mammalian cystic fibrosis transmembrane conductance regulator (CFTR) and promotes the degradation of this protein (Yang et al., 1993). CFTR, a plasma membrane chloride channel whose mutation leads to cystic fibrosis, is widely used as a model substrate to explore folding and degradation of membrane proteins on the ER due to inefficient folding ability. Secondly, CFTR is polyubiquitinated and degraded by the ubiquitin-proteasome system (Ward et al., 1995; Younger et al., 2006; Morito et al., 2008).
As described above, an Hsp70 family member generally requires a specific J-protein for the recognition of a subset of its substrates. In yeast Saccharomyces cerevisiae, Hlj1, an Hsp40 family protein anchored to the cytosolic side on the ER membrane, is involved in the degradation of the CFTR artificially expressed in yeast cells (Youker et al., 2004). However, neither the mammalian ortholog of Hlj1 nor the mammalian ER-localized J-protein recruiting Hsc70/Hsp70 for degradation of mislfolded membrane proteins has been identified. DNAJB12 is a poorly characterized mammalian J-protein. Here we showed that DNAJB12 is an ER-located type-II transmembrane protein. This finding means that, unlike other ER-located transmembrane J-proteins (Brodsky and Schekman, 1993; Dudek et al., 2002), its J-domain faces the cytosol. Moreover, we demonstrated that DNAJB12 associates with Hsc70/Hsp70 via the J-domain of DNAJB12. Further, we showed that DNAJB12 contributes to the degradation of membrane proteins. These findings led us to propose that DNAJB12 is a novel mammalian ER-localized J-protein that cooperates with Hsc70 in the cytosol to promote the ERAD of misfolded membrane proteins.
HEK293T and HeLa cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma) supplemented with 10% fetal bovine serum (FBS) and 4 mM L-glutamine. Proteasome activity was inhibited by treatment with lactacystin (Peptide Institute). Lactacystin was added at a final concentration of 20 μM 3 h prior to pulse-labeling and was present in the medium during the chase period. Plasmids and siRNAs were transfected into cells using effectene (Qiagen) and Lipofectamine RNAiMAX (Invitrogen), respectively. For the transfection, we followed the manufacturer’s instructions.
To construct the plasmid pCAGGS-DNAJB12-Flag, mouse DNAJB12 cDNA was generated from total RNA of NIH3T3 cells by RT-PCR. The primers used were: P1 (5'-CTTGTCATCGTCGTCCTTGTAGTCTCCATGCAGGGAGGCCTGCACC-3'; Flag sequence plus DNAJB12-hybridizing sequence), P2 (5'-TGCTCGGGTGAATTCATGGAATCCAACAAGGATGAAGCCGA-3'; forward; an EcoRI site indicated by underline plus DNAJB12-hybridizing sequence)] and P3 (5'-TTTTTTGAATTCTCACTTGTCATCGTCGTCCTTGTAG-3'; reverse; an EcoRI site indicated by underline plus Flag sequence). Amplification with these primers fuses a Flag tag to the C-terminus of DNAJB12. The resulting RT-PCR product was digested with EcoRI and inserted into the EcoRI site of pCAGGS (Niwa et al., 1991) to obtain pCAGGS-DNAJB12-Flag. The plasmid pCAGGS-HA-DNAJB12-Flag was constructed in the same way as pCAGGS-HA-DNAJB12-Flag except that the primers used are P1, P2, and P4 (5'-AAAAGAATTCATGTACCCATACGATGTTCCAGATTACGCTGAATC-3' (an EcoRI site underlined plus HA-tagging and DNAJB12-hybridizing sequences). Plasmid pCAGGS-DNAJB12H139Q-Flag was constructed by inserting a mutation that alters His139 (CAT) of DNAJB12 to Gln (CAA) into pCAGGS-DNAJB12-Flag by site-directed mutagenesis. The expression plasmids, pEGFP-ΔF508-CFTR (Johnston et al., 1998), and pcDNA3.1-TCRa (Yu et al., 1997), were kindly provided by R.R. Kopito (Department of Biological Sciences, Stanford University, Stanford, CA, USA). pEGFP-WT-CFTR was created by site-directed mutagenesis of pEGFP-ΔF508-CFTR. The plasmid to express vesicular stomatitis virus G tsO45 protein fused with a GFP tag (pcDNA3-VSVG-tsO45-GFP) (Gallione and Rose, 1985) was a gift from J. Lippincott-Schwartz (Cell Biology and Metabolism Branch, National Institute of Child Health and Human Development, National Institutes of Health). To make a plasmid to express NHK, the coding sequence of NHK was excised from pREP9-NHK (a gift from Kazuhiro Nagata, Kyoto Sangyo University, Japan) with HindIII and XhoI and inserted into the corresponding restriction sites of pcDNA3.1 (Invitrogen). The resulting plasmid was named as pcDNA3.1-NHK. Plasmids and siRNAs were transfected into cells using effectene (Qiagen) and Lipofectamine RNAiMAX (Invitrogen), respectively, using protocols provided by the manufacturers.
Anti-mouse DNAJB12 antibody was prepared by immunizing a guinea pig with highly purified protein that has an amino acid sequence corresponding to the N-terminal 100 amino acid residues of mouse DNAJB12. The anti-mouse DNAJB12 antibody recognizes DNAJB12 from both human and mouse. The other antibodies used in this study were mouse monoclonal anti-Flag M2 antibody (Sigma), mouse monoclonal anti-HA antibody (12CA5) (Boehringer Mannheim), rat monoclonal anti-Hsc70 antibody (Stressgen), goat polyclonal anti-GAPDH antibody (ab9484) (Abcam), rabbit polyclonal anti-GFP antibody (MBL), mouse monoclonal anti-GFP antibody (Roche), and rabbit polyclonal anti-A1AT antibody (DakoCytomation). For immunoblotting experiments, appropriate HRP-conjugated secondary antibodies were used. As secondary antibodies for immunofluorescence microscopy, we used Alexa Fluor® 647 Goat Anti-guinea pig IgG (highly cross-adsorbed), and Alexa Fluor® 488 Goat Anti-mouse IgG (highly cross-adsorbed) (Invitrogen).
For immunofluorescent staining, cells were fixed with 1% acetic acid in ethanol, permeabilized with digitonin in phosphate-buffered saline (PBS), and then blocked with PBS containing 3% bovine serum albumin (BSA). Primary and secondary antibodies were diluted in the blocking solution, and washes were performed with 0.1% Tween 20 PBS. Coverslips were mounted on 20 μl of Prolong Gold Antifade Reagent (Invitrogen) placed on a glass holder. Images were analyzed using a confocal fluorescence microscope FV1000 (Olympus).
To study the association of DNAJB12 with CFTR and/or Hsc70 in vivo, HeLa cells expressing both DNAJB12-Flag and GFP-WT-CFTR were lysed in ice-cold lysis buffer [10 mM Tris-HCl (pH 7.4 at 4°C), 130 mM NaCl, 10 mM EDTA, 1% Nonidet P-40 (NP-40), 1% BSA] supplemented with protease inhibitors (10 μg/ml leupeptin, 10 μg/ml pepstatin A, and 1 mM benzamidine). The lysate was centrifuged at 15,000 rpm for 10 min to remove the cell debris. The resulting supernatant was incubated with anti-Flag antibody-conjugated agarose beads (Sigma) at 4°C for overnight. The immunocomplexes were collected by centrifugation, washed with lysis buffer three times, re-suspended in 20 μl of 2×SDS-PAGE gel loading buffer, incubated at 65°C for 10 min, separated by SDS-PAGE and then subjected to immunoblot analysis.
For the knockdown of DNAJB12 in HEK293T cells, we used Lipofectamine RNAiMAX (Invitrogen). We followed the manufacture’s instructions except that 20 nM siRNAs were incubated with 3.5×105 cells in collagen coated 6-well plates. The siRNAs used are sijb12 #1 (5'-UAUCUGUGUCGCCAAAGUAGCGUGC-3' and 5'-GCACGCUACUUUGGCGACACAGAUA-3'), and sijb12 #3 (5'-AUAGGCAUCAGCUGCACAAACACCC-3' and 5'-GGGUCAAGCAAUGUAAAGAUUACUA-3') (both were synthesized by Invitrogen).
Growing cultures of HEK293T cells were preincubated in methionine- and cysteine-free DMEM supplemented with 10% dialyzed FBS at 37°C for 30 min. Following this preincubation, the cells were radio-labeled with [35S]-EXPRESS protein labeling mixture (PerkinElmer) and chased in normal DMEM/10% FBS. Cellular proteins were normally extracted from the labeled cells with 1% NP40 lysis buffer. Exceptions were: (1) 0.3 % SDS RIPA lysis buffer [50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.3% SDS, 0.5% deoxycholic acid and protease inhibitors] for the extraction of CFTR-GFP; and (2) 0.5% NP40 lysis buffer [50 mM Tris-HCl (pH 7.5 at 4°C), 150 mM NaCl, 0.5% NP-40 and 0.5% deoxycholic acid and protease inhibitors] for VSV-G. The extracted proteins were incubated with appropriate antibodies, and then immunoprecipitated using protein A-coupled Sepharose beads (GE Healthcare). Immunoprecipitates were washed twice in lysis buffer and then eluted by incubating in 2×SDS-PAGE gel load buffer at 65°C for 10 min. The proteins were separated by 7 or 10% SDS-PAGE and the radioactive signals were detected with BAS2500 (Fujifilm).
DNAJB12 is a poorly characterized protein carrying a J-domain and a hydrophobic segment that is deduced to act as a transmembrane domain (Fig. 1A and Fig. 2). This protein is highly conserved from Schizosaccharomyces pombe to human (Fig. 2), suggesting its importance in cells. To examine the cellular localization of DNAJB12, endogenous DNAJB12 was visualized in HeLa cells by immunostaining with anti-DNAJB12 antibody. As shown in Fig. 1B, DNAJB12 (left panel) was detected in a pattern that overlapped with calnexin, an ER marker (middle and right panels) indicating that DNAJB12 is located at the ER. As described above, DNAJB12 is likely to be a membrane-spanning protein carrying one transmembrane domain. To determine the membrane topology of DNAJB12, we fused DNAJB12 with an HA tag at the N-terminus and a Flag tag at the C-terminus and expressed it in HeLa cells. The microsomes, prepared from the cells, were then treated with proteinase K, and analyzed by immunoblotting with anti-Flag and anti-HA antibodies. As shown in Fig. 1C, both anti-Flag and anti-HA antibodies detected the full-length HA-DNAJB12-Flag (lanes 1 and 4). In contrast, a proteolytic fragment (approximately 20 kDa) was detected only with anti-Flag antibody (lanes 2 and 5). This fragment and the full-length protein were almost undetectable when the proteolysis was performed in the presence of detergent Triton X-100 (lanes 3 and 6). The right panel of Fig. 1C shows a control experiment demonstrating that a luminal protein, protein disulfide isomerase (PDI), was efficiently degraded only in the presence of Triton X-100 (compare lane 9 with lanes 7 and 8). These immunofluorescent staining and protease protection assays indicate that it is highly likely that DNAJB12 is an ER-located type-II transmembrane protein with a J-domain facing the cytosol (Fig. 1D).
 View Details | Fig. 1. Cellular localization and membrane topology of DNAJB12. (A) Architecture of DNAJB12. The putative J-domain (magenta), glycine/phenylalanine-rich domain (orange), and transmembrane domain (blue) are indicated. (B) Endogenous DNAJB12 and calnexin (an ER marker) were immunostained in HeLa cells with antibodies specific to these proteins, and visualized by secondary antibodies conjugated with Alexa 647 and Alexa 488, respectively. Bars indicate 10 μm (C) Microsomes prepared from HeLa cells expressing HA-DNAJB12-Flag were treated with proteinase K. Where indicated, 1% Triton X-100 was added into the proteolysis reaction mixture. The resulting proteolytic products were analyzed by immunoblotting using anti-HA, anti-Flag and anti-PDI antibodies. Full-length HA-DNAJB12-Flag and its partial proteolytic fragment are indicated by black and white arrows, respectively. (D) Comparison of the topology of DNAJB12 with those of the other well-characterized mammalian ER-localized J-proteins. Transmembrane domains are indicated by blue boxes. Note that DNAJB12 is unique in that its J-domain faces the cytosol. |
 View Details | Fig. 2. Sequence alignments of DNAJB12 homologues. The source organisms were: Mouse, Mus musculus; Human, Homo sapiens; Drosophila, Drosophila melanogaster; C. elegans, Caenorhabditis elegans; and S. pombe, Schizosaccharomyces pombe. The putative J-domains are boxed with red line. The positions of putative HPD motif (asterisks), glycine/phenylalanine-rich domain (orange bar), and transmembrane domain (blue bar) in the mouse DNAJB12 sequence were indicated. Black highlighting completely conserved amino acid residue; gray, amino acids resides identical in 60% or 80% of the sequences compared. |
These observations led us to examine whether DNAJB12 associates with cytosolic HSP70 family chaperones in vivo (Fig. 3A). To test this, HeLa cells expressing C-terminally Flag-tagged DNAJB12 (DNAJB12-Flag) were subjected to immnoprecipitation with anti-Flag antibody. This resulted in the co-immunoprecipitation of Hsc70 (lane 2, middle panel). Hsc70 was not detected in the immunoprecipitate of the cells that does not express DNAJB12-Flag (lane 1, middle panel). Since the highly conserved HPD motif in J-domains is generally required for their interaction with the partner Hsp70 family chaperones (Kampinga and Craig, 2010), we constructed a DNAJB12 mutant carrying a substitution mutation that alters this HPD motif to QPD (H139Q). When the H139Q mutant version of DNAJB12-Flag was expressed in HeLa cells, its expression level was almost comparable to that of wild-type DNAJB12-Flag (compare lane 3 with lane 2 in the bottom panel). Nevertheless, Hsc70 was not co-immunoprecipitated with the mutant (lane 3, middle panel). This result indicates that DNAJB12 interacts with Hsc70 in a manner that depends on its J-domain. This finding also supports our model on the membrane topology of DNAJB12 that the J-domain of DNAJB12 faces the cytosol.
 View Details | Fig. 3. DNAJB12 binds Hsc70 and accelerates degradation of CFTR. (A) HeLa cells were mock-transfected or transfected with a plasmid to express DNAJB12-Flag or DNAJB12 H139Q-Flag. After 36 h, the cells were lysed and the resulting lysate was subjected to immunoprecipitation with anti-Flag agarose beads. Ten μg of the lysate proteins (Input) and the immunoprecipitates from 1 mg lysate proteins (IP) were separated by SDS-PAGE and analyzed by immunoblotting (IB) with the indicated antibodies. (B) Lysates from HeLa cells expressing both GFP-WT-CFTR and DNAJB12 variants were subjected to immunoprecipitation with anti-Flag agarose beads. GFP-WT-CFTR that was present in the lysate (Input) and immunoprecipitates (IP) were analyzed by immunoblotting (IB) with anti-GFP antibody. (C) HEK293T cells were transfected with the GFP-WT-CFTR expression plasmid alone or together with the DNAJB12-Flag expression plasmid. After 30 h of culture, cells were pulse-labeled with 35S-methionine/cysteine for 1 h and chased for the indicated time. The cells were then lysed and GFP-WT-CFTR was collected from the lysate by immunoprecipitation with anti-GFP antibody. Then, the fate of newly synthesized GFP-WT-CFTR was analyzed by SDS-PAGE and autoradiography. Where indicated, lactacystin (10 μM) was added to the medium during the pulse-labeling and chase period. (D) The GFP-WT-CFTR remaining (the sum of B and C forms) at each time point was calculated and plotted. The intensity observed at 0 h was taken as 100%. Error bars indicate standard deviations from the average of three independent experiments. |
Our finding that DNAJB12 is an ER protein that can interact with Hsc70 led us to explore its involvement in the quality control of membrane proteins including CFTR. Wild-type CFTR (WT-CFTR) was expressed as a GFP-fused protein in HeLa cells and detected by immunoblotting with anti-GFP antibody (Fig. 3B) (Johnston et al., 1998). Since GFP-WT-CFTR is converted from the B form (ER form) to the mature C form (Golgi-plasma membrane form) during its biosynthesis, two protein bands were detected (Fig. 3B, upper panel, lane 1). When DNAJB12-Flag was co-expressed with GFP-WT-CFTR, both the B and C forms were hardly detectable (lane 2). On the other hand, the co-expression of the H139Q mutant version of DNAJB12-Flag gave a more modest effect, as only the level of the C form was decreased significantly (lane 3) (see Discussion section for the interpretation of this result). As shown in the lower panel of Fig. 3B, GFP-WT-CFTR was co-immunoprecipitated efficiently with the wild-type DNAJB12 but not with the H139Q mutant, indicating that the HPD motif in the J-domain of DNAJB12 is important for the efficient binding of this protein with WT-CFTR.
To study the involvement of DNAJB12 in the quality control of GFP-WT-CFTR, we performed pulse-chase experiments. Cells expressing GFP-WT-CFTR were incubated with 35S-methionine/cysteine for 30 min to label newly synthesized proteins and chased with excess unlabeled methionine/cysteine for different time periods. As expected, the radioactive signal pulled down by anti-GFP antibody indicated the conversion of GFP-WT-CFTR from the B form to the C form (Fig. 3C, top panel). When DNAJB12-Flag was co-expressed with GFP-WT-CFTR, the C form was hardly detected during the chase period (middle panel), and the level of the total GFP-WT-CFTR signal (the sum of the B and C forms) was rapidly decreased (Fig. 3D). This finding indicates that DNAJB12 accelerates the degradation of WT-CFTR. The DNAJB12-dependent degradation of WT-CFTR is mainly due to ERAD, because this process was clearly compromised by the proteasome inhibitor, lactacystin (Fig. 3C, bottom panel and Fig. 3D).
To examine whether DNAJB12 is involved in the degradation of misfolded membrane proteins, we employed other model ERAD substrates and performed pulse-chase experiments. One of the substrates, CFTR mutant ΔF508, exhibits severe defects in protein folding, hence this protein is rapidly degraded before maturation into the C form (Ward et al., 1995). Another substrate, T cell-receptor α subunit (TCRα), is misfolded when produced in the absence of its partner subunits (Yu et al., 1997). Overexpression of DNAJB12 accelerated the degradation of these two proteins moderately but clearly (Fig. 4A, B). We also looked at the effect of DNAJB12 overexpression on the degradation of a temperature-sensitive mutant of the vesicular stomatitis virus G protein (tsO45-VSV-G) whose folding is compromised at 39°C (Gallione and Rose, 1985). Remarkably, overexpression of DNAJB12 accelerated the degradation of this protein drastically (Fig. 4C). These findings indicate that DNAJB12 can accelerate the degradation of membrane proteins.
 View Details | Fig. 4. Degradation of misfolded ER-membrane proteins is accelerated by DNAJB12. HEK293T cells were cotransfected with plasmids to express both DNAJB12-Flag and one of the model substrates [GFP-ΔF508-CFTR (A), a TCRα-HA (B), a tsO45-VSV-G-GFP (C) or NHK (D)]. Pulse-chase experiments were then performed basically as described in the legend to Fig. 3 except that, in panel C, cells were cultured at 39°C during the pulse-labeling and chase period. Pulse-chased samples were then subjected to immunoprecipitation using following antibodies: anti-GFP antibody (A), anti-HA antibody (B), anti-GFP antibody (C), and anti-A1AT antibody (D). The fate of the newly synthesized model substrates were then analyzed as described in the legend to Fig. 3C, D. |
We next examined whether DNAJB12 participates in the degradation of an ER luminal soluble protein. α1-antitrypsin is a plasma serine protease inhibitor. The A1AT null Hong Kong variant of α1-antitrypsin (NHK) fails to fold properly in the ER and, thus, is degraded by the ubiquitin-proteasome system (Sifers et al., 1988). This protein has been widely used as a substrate of ERAD of ER luminal soluble proteins. Importantly, the overexpression of DNAJB12 did not affect the stability of NHK (Fig. 4D). Thus, DNAJB12 can specifically enhance the ERAD of membrane proteins.
Cellular function of DNAJB12 was further explored by knocking down the expression of endogenous DNAJB12 using siRNAs. HEK293T cells were transfected with one of the two siRNA duplexes, here called sijb12 #1 and sijb12 #3, that were designed to decrease the expression of DNAJB12. Fig. 5A shows that both sijb12 #1 and sijb12 #3 siRNAs successfully suppressed the expression of DNAJB12. In the cells that were transfected with DNAJB12 siRNAs, GFP-Δ508-CFTR, a mutant CFTR having significantly less folding ability than WT-CFTR, was more slowly degraded than that of the cells transfected with negative control siRNA (Fig. 5B). On the other hand, the same knockdown of DNAJB12 did not affect the stability of NHK (Fig. 5C). Collectively, these findings clearly indicate that DNAJB12 is involved in the ERAD of membrane proteins.
 View Details | Fig. 5. DNAJB12 is involved in the degradation of a membrane protein. (A) HEK293T cells were transfected with DNAJB12 siRNAs using Lipofectamine RNAiMAX. After 48 h, the cellular proteins from the culture were separated by SDS-PAGE and the levels of DNAJB12 (upper) and GAPDH (lower; negative control) in the cells were examined by immunoblotting (IB) using antibodies to DNAJB12 and GAPDH, respectively. Note that the membrane after the protein transfer was split into two halves at the position of 37 kDa. The upper and lower parts were then used to detect DNAJB12 and GAPDH, respectively. siRNAs employed for the knockdown experiments are indicated at the top of each lane. (B–C) HEK293T cells were initially transfected with the DNAJB12 siRNA, sijb12#1, or the control siRNA, siLuc, using Lipofectamine RNAiMAX. After 24 h, the cells were further transfected with a plasmid that expresses either GFP-ΔF508-CFTR (B) or NHK (C). Following an additional 24 h of incubation, pulse-chase experiments were performed. The pulse-chased samples were then subjected to immunoprecipitation using anti-GFP antibody (B), or anti-A1AT antibody (C). The fate of the newly synthesized model substrates was then analyzed as described in the legend to Fig. 3C, D. |
Protein quality control systems play a vital role in the maintenance of homeostasis by degrading misfolded proteins that arise in the cells (Ding and Yin, 2008). Previous studies have shown that misfolded proteins in the mammalian ER membrane are degraded by the proteasome in the cytosol in a process called ERAD (Gallione and Rose, 1985; Ward et al., 1995; Yu et al., 1997). The ERAD of misfolded proteins in the mammalian ER membrane requires the cytosolic Hsp70/Hsc70 chaperone (Younger et al., 2006). However, the J-protein required for this process remained elusive. Here, we have presented several lines of evidence that DNAJB12 is the J-protein that contributes to the ERAD of membrane proteins (Fig. 6). Firstly, the overexpression of DNAJB12 enhanced the degradation of several membrane proteins including WT-CFTR. Secondly, the DNAJB12-dependent degradation of WT-CFTR was diminished by lactacystin, a proteasome inhibitor. Thirdly, DNAJB12 interacted with both WT-CFTR and Hsc70 in a manner that requires its J-domain. Fourthly, knockdown of the endogenous DNAJB12 stabilized Δ508-CFTR. Finally, neither the overexpression nor knockdown of DNAJB12 affected the degradation of a model ER luminal substrate of ERAD, indicating that DNAJB12-dependent degradation is specific to membrane proteins.
 View Details | Fig. 6. Model for the role of DNAJB12 in degradation of ER membrane proteins. See main text for explanation. |
Our data indicated that DNAJB12 is an ER-localized single membrane spanning protein having its J-domain on the cytosolic side. DNAJB12’s membrane topology is unique as that of a mammalian ER-localized J-protein (Fig. 1D). In contrast to the other characterized mammalian ER-localized J-proteins all of which have a J-domain on the luminal side (Chevaliar et al., 2000; Kurisu et al., 2003; Hosoda et al., 2003; Dong et al., 2008; Jin et al., 2009; Zahedi et al., 2009), DNAJB12 has its J-domain in the cytosol (Fig. 1D). Thus, topologically, DNAJB12 is well suited for its function because DNAJB12 needs to interact with not only misfolded proteins on the ER membrane but also with Hsc70 in the cytosol.
We showed that DNAJB12 stably associates with Hsc70 and its substrate protein WT-CFTR (Fig. 3A, B). This association was abolished by a mutation (H139Q) in the J-domain of DNAJB12 (Fig. 3A, B). However, this mutant still retained some of its ability, although at reduced levels, to promote the degradation of WT-CFTR (Fig. 3B, lane 3). Thus, it may be possible that DNAJB12 has another function that is independent of its interaction with Hsc70. Alternatively, the DNAJB12 mutant may still be able to weakly associate with its partner proteins to promote the degradation of membrane proteins.
We are grateful to Dr. Douglas Cyr (University of North Carolina-Chapel Hill, USA) for sharing information about DNAJB12 before publication. We thank Drs. Ron Kopito (Stanford University), Jennifer Lippincott-Schwartz (NICHD), and Kazuhiro Nagata (Kyoto Sangyo University) for plasmids, and Junko Iida-Hashimoto and Hisayo Masuda for technical assistance. We also thank Drs. Jeffrey Brodsky (University of Pittsburgh), Daisuke Morito (Kyoto Sangyo University) and Akira Kitamura (Hokkaido University) for critical discussions and advice. This work was supported by Grants-in-Aids for Scientific Research on Priority Areas (Nos. 14037240 and 19058010 to K.K.; No. 20058023 to Y.K.), and Scientific Research B (20380062 to K.K.) of KAKENHI from MEXT of Japan. Y.Y. was supported by a fellowship from the Japan Society for the Promotion of Science. H.K. was supported by an international research fellowship from the Global COE program in NAIST from MEXT of Japan.
|
Index