Genetic Analysis of the Polyomavirus DnaJ Domain
http://www.100md.com
病菌学杂志 2005年第15期
Department of Biochemistry, Tufts University School of Medicine, 136 Harrison Avenue, Boston, Massachusetts 02111
ABSTRACT
Polyomavirus T antigens share a common N-terminal sequence that comprises a DnaJ domain. DnaJ domains activate DnaK molecular chaperones. The functions of J domains have primarily been tested by mutation of their conserved HPD residues. Here, we report detailed mutagenesis of the polyomavirus J domain in both large T (63 mutants) and middle T (51 mutants) backgrounds. As expected, some J mutants were defective in binding DnaK (Hsc70); other mutants retained the ability to bind Hsc70 but were defective in stimulating its ATPase activity. Moreover, the J domain behaves differently in large T and middle T. A given mutation was twice as likely to render large T unstable as it was to affect middle T stability. This apparently arose from middle T's ability to bind stabilizing proteins such as protein phosphatase 2A (PP2A), since introduction of a second mutation preventing PP2A binding rendered some middle T J-domain mutants unstable. In large T, the HPD residues are critical for Rb-dependent effects on the host cell. Residues Q32, A33, Y34, H49, M52, and N56 within helix 2 and helix 3 of the large T J domain were also found to be required for Rb-dependent transactivation. Cyclin A promoter assays showed that J domain function also contributes to large T transactivation that is independent of Rb. Single point mutations in middle T were generally without effect. However, residue Q37 is critical for middle T's ability to form active signaling complexes. The Q37A middle T mutant was defective in association with pp60c-src and in transformation.
INTRODUCTION
Polyomavirus T antigens function both in replication of the virus and in transformation of the host cell. Large T is central to virus production as the initiator of viral DNA replication (20). Middle T and small T also play important roles in different aspects of polyomavirus infection (21, 23, 55). Defects in viral DNA replication and transcription, as well as defects in viral assembly, have been observed in different mutants of middle T and small T (1, 7, 8, 22, 38). Each of the viral early proteins also contributes to regulation of host cell function. Large T is able to immortalize primary cells (44), to block differentiation (37), and to provoke apoptosis (18, 48). These activities are mediated via association with the retinoblastoma susceptibility (Rb) family of tumor suppressors. Middle T, the major transforming protein, works through activation of cellular signaling pathways that are regulated by src-family tyrosine phosphorylation (15). Small T is able to promote cell cycle progression via association with protein phosphatase 2A (PP2A) (39).
All three T antigens are produced by differential splicing of common primary transcripts (56). As a result, they have the identical N-terminal sequence of 79 amino acids that encompasses a DnaJ domain. DnaJ domains, consisting of approximately 70 amino acids, have a helical structure in which a conserved HPD motif is found between helix 2 and helix 3 (2, 13, 32, 43, 54). DnaJ domains, found in a broad range of proteins, function to stimulate the activity of DnaKs (6, 14). The domains of polyomavirus and simian virus 40 (SV40), like other DnaJs, have been shown to activate the ATPase activity of DnaKs (45, 50, 52). Substitution experiments have shown that SV40 or BK domains can function biologically as DnaJ domains in Escherichia coli (31).
The DnaJ/K cochaperones have diverse cellular activities, including protein folding (19), protein degradation (57), vesicle sorting/uncoating (16, 36), and DNA replication (5, 33). The importance of the J domain to both polyomavirus and SV40 large T function is clear (47, 53). The ability of SV40 large T to drive DNA replication in vivo is highly dependent on an intact DnaJ domain (5). The ability of SV40 large T to transform is dependent on DnaJ function (50, 58). The J domain of SV40 large T has also been implicated in sensitizing cells to apoptosis after genotoxic damage (10). A recent report indicates that the J domain is important for the ability of SV40 large T to release VP1 from complexes with Hsc70 (9). Polyomavirus large T promotes cell cycle progression in a DnaJ-dependent manner (47). Much, but not all (48), of large T's ability to act on the Rb family is dependent on an intact J domain (47, 51, 52, 58). Mutations of the HPD loop that abolish interaction of large T with Hsc70 (5, 48) affect the ability of SV40 large T to alter the turnover of p130 (51). Work done with SV40 by the Pipas group has demonstrated an intact HPD loop is critical for dissociation of Rb/E2F complexes (52), which would account for large T promoter activation at E2F sites. Less is known about the role of the J domain in polyomavirus middle T and small T. Although some mutations in the middle T J domain have been shown to affect transforming ability (11), mutation of the HPD loop was found not to alter the ability of middle T to transform cells (4, 26). This suggests that activation of DnaK, the normal role for DnaJ, is not required for transformation.
There has been little systematic genetic analysis of J domains, even for E. coli DnaJ (17, 24, 25). To map the DnaJ residues necessary for polyomavirus large T function, we carried out extensive mutagenesis of the J domain. This identified residues important for the stability of large T and for the ability of large T to act on the Rb family. These experiments also suggested that the domain is important for large T functions beyond those related to Rb. Further, we compared the effects of mutations in the J domain of large T to the same mutations in middle T. Our results indicated that the two proteins use the J domain differently.
MATERIALS AND METHODS
Cell lines and transfections. NIH 3T3 cells were originally obtained from the American Type Culture Collection. The cells were grown in Dulbecco modified Eagle medium (DMEM; Gibco) and supplemented with 10% calf serum (HyClone). Transfections were performed by the calcium phosphate precipitation method (47). NIH 3T3 cell lines expressing different middle T antigens were obtained by cotransfection of NIH 3T3 cells with a vector for puromycin resistance and the relevant pCMV MT construct at a ratio of 1:10. At 2 days posttransfection the cells were placed in selection medium of DMEM containing 10% calf serum and puromycin at 5 μg/ml to obtain pools of puromycin-resistant colonies.
C33A cells were obtained from Vimla Band and were grown in DMEM supplemented with 10% fetal calf serum (HyClone). C33A cell transfections were performed using HEPES-buffered saline (28). For starvation experiments, cells were washed twice with phosphate-buffered saline (PBS) 18 h after transfection and placed in DMEM containing 0.2% fetal calf serum.
Plasmids and mutagenesis. pCMV large T, pCMV Rb–LT, pCMV P43S large T, HA-Rb, HA-p130 (47), pCMV middle T, pCMV NG59 middle T (12), and INS107AL (5) have all been described previously. Point mutants were constructed on wild-type pCMV large T or pCMV middle T. In general, oligonucleotides of 18 to 20 bases with one or two mismatches to the J domain sequence were used. Standard PCR methods based on PFU Turbo polymerase (Stratagene) were used to generate the mutants. All mutations were confirmed using the dideoxy sequencing method (46). A summary of the mutants constructed is shown in Fig. 1.
The E2F-luciferase construct contains six E2F sites attached to a luciferase construct and was obtained from Amy Yee. The myc-tagged Hsc70 was obtained from Tom Roberts. The cyclin A –89-to-+11 promoter construct has been described previously (48).
Antibodies. The PN116 monoclonal and rabbit anti-T antibody used in Western blots have been described previously (30). Middle T was immunoprecipitated with rabbit polyclonal antibody 45-1 (39). HA11 was obtained from Covance. Myc antibody 9E10 was obtained from James DeCaprio. Polyclonal PP2A-A subunit antibody was obtained from Gernot Walter and Ralf Ruediger. Anti-src antibody (GD11) was obtained from Larry Feig.
Luciferase assays. NIH 3T3 cells plated on 60-mm dishes were transfected at a confluence of 20% with 2 μg of E2F-luciferase or with 1 μg cyclin A-luciferase and 0.5 μg of the relevant LT expression vectors and harvested approximately 40 h posttransfection. For the cyclin A-luciferase assays, cells were placed under serum-starved conditions at 24 h posttransfection. Cells were suspended in buffer (25 mM Tris [pH 7.5], 1 mM EDTA) and subjected to freezing-thawing three times. The lysates were cleared by Eppendorf centrifugation and assayed for luciferase activity.
Extraction and immunoprecipitations. The cells were washed in PBS and extracted in T extraction buffer (TEB; 137 mM NaCl, 10 mM Tris-Cl [pH 8.0 for HA-Rb, pH 7.0 for myc-Hsc70, and pH 9.0 for middle T], 1 mM MgCl2, 1 mM CaCl2 10% [vol/vol] glycerol, and 1% [vol/vol] Nonidet P-40 supplemented with protease inhibitor cocktail, including leupeptin at 1 μg/ml, pepstatin 1 at μg/ml, phenylmethylsulfonyl fluoride at 100 μg/ml, and aprotinin at 2 μg/ml) for 20 minutes at 4°C. The cleared extracts were incubated with the appropriate antibody and protein G- or protein A-Sepharose beads (Pharmacia). The immunoprecipitates were washed with PBS. The immunoprecipitates were boiled in dissociation buffer (62.5 mM Tris-Cl [pH 6.8], 5% [wt/vol] sodium dodecyl sulfate [SDS], 25% [vol/vol] glycerol, 0.0075% [wt/vol] bromophenol blue, 5% [vol/vol] ?-mercaptoethanol).
Electrophoresis. Samples were analyzed on discontinuous buffer SDS gels (34) of 7.5% acrylamide. Radiolabeled gels were exposed and quantified by PhosphorImager (Molecular Dynamics) using ImageQuant software. All of the figures were prepared in Adobe Photoshop.
Large T NT purification. The N-terminal domains (NT; residues 1 to 259) of large T expressed as glutathione S-transferase fusions in pGEX3X were produced in E. coli BL21. Expression was induced with 50 μM IPTG (isopropyl-?-D-thiogalactopyranoside) overnight at 30°C. Cell extracts were incubated with glutathione-beads (Sigma) to harvest the fusion proteins. Washed beads were treated overnight at 4°C with factor Xa (Haematologic Technologies, Inc.) in buffer (20 mM Tris-Cl [pH 8.0], 1 mM calcium chloride, 100 mM NaCl) to release NT. The soluble NT was concentrated by Amicon ultrafiltration.
ATPase assays. The assays were performed in a 20-μl total volume containing 75 mM NaCl, 1.5 mM CaCl2, 40 mM HEPES (pH 7.2), and 2 μCi of [-32P]ATP with 0.4 μg of mouse Hsc70 (Stressgen). In some cases the reactions were supplemented with 0.8 μg of wild-type or mutant purified NT protein. The reaction mixtures were assembled on ice and then incubated at 37°C. The reactions were stopped by plating 2-μl aliquots in triplicate onto polyethyleneimine thin-layer chromatography plates. ATP and ADP were separated by chromatography in 1 M LiCl-1 M formic acid solution for 3 h. The plates were dried and analyzed by using a Molecular Dynamics PhosphorImager.
Middle T kinase assays. NIH 3T3 cells plated onto 100-mm dishes were transfected at a confluence of 20% with 5 μg of different middle T expression vectors and harvested approximately 40 h posttransfection. After extraction in TEB in the presence of protease inhibitors, the cleared extracts were incubated with rabbit polyclonal antibody to middle T (45-1) and protein A-Sepharose beads for 1 h. Washed immunoprecipitates were incubated in 200 μl of kinase buffer (20 mM Tris [pH 7.5], 5 mM MnCl2) containing 5 to 10 μCi of [-32P]ATP (2,000 Ci/mmol; NEN) for 15 min at room temperature. The reactions were washed with PBS, boiled in dissociation buffer, and subjected to SDS-polyacrylamide gel electrophoresis (PAGE). After electrophoresis, the gel was alkaline treated (0.5 M potassium hydroxide) for 1 h at 55°C to remove background serine/threonine phosphorylation. The gel was then fixed (10% acetic acid, 30% methanol), dried, and analyzed with a PhosphorImager.
Transformation assays. NIH 3T3 cells were grown in 10% calf serum and transfected with 7 μg of carrier DNA and 2 μg of the middle T cDNAs by using the calcium phosphate method. When the confluency reached 100%, the cells were maintained in 5% calf serum-DMEM and allowed to grow for 10 to 14 days. The plates were then washed twice with PBS and fixed for 30 min with 3.7% formaldehyde. The fixed plates were stained with 0.2% crystal violet in 3.7% formaldehyde. Foci were then counted.
RESULTS
Mutant analysis of the polyomavirus J domain demonstrates differences between the middle T and large T proteins. Site-directed mutagenesis was used to screen the J domains of middle and large T antigens. Point mutants were made in vectors expressing either middle T or large T using conservative substitutions such as lysine to arginine or leucine to valine. The mutants are listed in Fig. 1. In all, 63 mutants of large T and 51 mutants of middle T were examined.
The stabilities of mutant large T antigens were compared to wild type after transient expression in NIH 3T3 cells. Cell extracts prepared approximately 40 h after transfection were subjected to SDS-PAGE. Immunoblotting was then used to assess the level of large T protein. Figure 2A shows an example of the results. A majority of the mutants, including P43S, Q36A, and F62Y, retained wild-type levels of expression, whereas W59F large T showed a very low level of expression. In all, nine of the large T mutants showed a substantial decrease in protein levels (Fig. 1). The simplest interpretation of this observation is that the mutations cause unfolding of the domain leading to degradation of the protein. Figure 3A shows the positions in the J structure of mutations that caused loss of large T stability. Four DnaJ mutations also rendered middle T unstable (Fig. 1 and 3C). Two examples, R12K and L13V, are shown in Fig. 2B. Seven mutations (K10R, L17V, M30V, G25A, F27Y, L55V, and W59F) that rendered large T unstable did not appreciably affect the stability of middle T. On the other hand, mutation R12K (Fig. 2B) was unstable in the middle T background but not in large T.
The results just described suggest that mutations of the J domains of large T and middle T have different effects on each protein even though the N-terminal sequences are identical. Possible explanations for the divergence lie within the differences in the C-terminal residues of the two proteins, as well as their association with different cellular proteins. A key difference between middle T and large T is middle T association with PP2A. Furthermore, it is also known that PP2A and Hsc70 compete for binding to middle T (4) suggesting contact may exist between PP2A and Hsc70. If PP2A were to contact the J domain, this interaction might provide additional stability. To test this hypothesis, L55V and G25A mutants of middle T were combined with the NG59 mutation that prevents PP2A binding (49). Figure 2C shows that the two double mutants are significantly less stable than a middle T mutant only in the J domain. This result is also true in the case of the M30V middle T mutant (data not shown). This result suggests that PP2A binding renders middle T more resistant to mutations in the J domain.
Effects of J mutations on large T Rb function. Large T activation of promoters containing E2F sites was shown to be dependent upon an intact Rb binding site, as well as an intact HPD loop in the J domain (52). Earlier experiments suggested a role for some residues outside of the HPD loop in large T for the activation of E2F promoters (47). Each of the mutants in large T was tested for the ability to activate the E2F promoter. Mutants Q32A, A33G, Y34F, H49R, M52V, and N56T were all defective in E2F promoter activity as assayed by cotransfection of the relevant large T constructs and an E2F-luciferase reporter construct (Fig. 4A, upper panel). As expected, the HPD mutant, P43S, was defective in E2F promoter activity (Fig. 4A). The non-HPD mutants defective in E2F activation are found in helix 2 and helix 3 (shown in blue in Fig. 3B). Previously, it has been shown that mutations in the HPD loop that abolish E2F activation have no effect on binding of Rb (47). Coimmunoprecipitation experiments showed that this was also true for the non-HPD mutants (not shown). In our E2F analysis, we also tested C-terminal mutants, including M71V and T76S. Structural analysis (2) suggested that these sequences, while part of the first exon of the T antigens, may not be part of the J domain but rather serve as a linker. These mutants were all wild type for E2F activation, so our data are consistent with this hypothesis.
Polyomavirus large T antigens affect the phosphorylation status of the pRB family member, p130 (52, 58). Mutations within the HPD loop of the J domain of polyomavirus cause a supershift in the p130 molecule seen on SDS-PAGE. The supershift is caused by phosphorylation of the p130 molecule in the C-terminal portion of the molecule. It is has been demonstrated that SV40 large T causes a dephosphorylation of the p130 molecule (57, 58). This result suggests that the large T protein is important for proper cycling of the p130 molecule and that the J mutant is unable to cycle the molecule. Instead, the J mutant leaves the p130 "stuck" in a hyperphosphorylated state. To test whether the non-HPD mutants behaved similarly, experiments were carried out in C33A cells with the various non-HPD mutants and HA-tagged p130 protein. SDS-PAGE analysis of the samples and subsequent immunoblotting with anti-HA was done. Fig. 4B shows that the non-HPD mutants also caused a similar shift in the p130 protein. Therefore, all J mutants defective in the ability to activate E2F sites also affected p130 phosphorylation.
Neither the defect in E2F activation nor the effect on p130 could be explained by differences in protein expression. As shown in Fig. 4A these mutants were expressed at levels close to wild type, suggesting that their defect was caused by loss of J domain activity rather than protein instability.
Interactions of mutant large T's with Hsc70. The HPD loop of large T is known to be critical for Hsc70 binding (5, 47). This binding is necessary for large T transactivation, since reducing large T-Hsc70 interactions by overexpression of exogenous DnaJ domains squelches transactivation of E2F promoters (47). To examine the interaction of the non-HPD mutants with Hsc70, wild-type or mutant large T was cotransfected with myc-tagged Hsc70. Coimmunoprecipitation and Western blotting were used to determine binding of Hsc70 to large T. The experiment in Fig. 5A demonstrates that the P43S HPD and two of six non-HPD (A33G and M52V) mutants were unable to bind the cellular heat shock protein. Figure 5B shows that lack of binding did not result from poor protein expression. A33G retains a low level of activity in the E2F assay despite the lack of hsc70 binding. It is likely that A33G has a lowered affinity for hsc70 that permits modest in vivo activity but does not result in stable complexes that can be immunoprecipitated. However, Q32A, Y34F, H49R, and N56T mutants that were defective in activation of E2F sites were still able to bind the Hsc70 protein in a manner similar to that of the wild type.
Since binding Hsc70 might not be sufficient for stimulation of its ATPase function, activity measurements were also performed. To do this, wild-type, Y34F, and H49R N-terminal domains (residues 1 to 259) were constructed and purified from E. coli using an approach similar to that of Riley et al. (45). The N-terminal domain was used because it can be easily expressed in E. coli as a soluble protein. A time course of ATP hydrolysis was determined for Hsc70 using [-32P]ATP in the presence of wild-type, Y34F, or H49R proteins. In these assays, wild-type NT was able to stimulate Hsc70 ATPase activity, whereas the J domain mutants were defective (Fig. 5C). It seems likely that the defect in E2F activation by Y34F, as well as H49R and probably other mutants that retain binding, is caused by an inability to stimulate DnaK function.
J domain effects on the cyclin A promoter. Whether the J domain participates in large T functions not related to Rb is an open question. Previous work has shown that large T can transactivate the cyclin A promoter (18, 48). In serum-starved cells this activation is completely dependent on Rb binding. However, when the E2F site in the promoter is mutated (–37/–33), large T still transactivates, but that activation is now completely independent of Rb binding (48). Our earlier work showed that HPD mutants in the J domain were also active, although at levels reduced from wild type (48). To determine whether the reduced activity indicated a J domain contribution or whether it simply represented a general loss of protein structure, the new set of J domain mutants were analyzed in cyclin A promoter assays. To avoid E2F effects, a mutant cyclin A (–37/–33) reporter lacking the E2F site was used. Every mutant that was wild type in the E2F assay was also fully competent in activation of the mutant cyclin A promoter (data not shown). The mutants impaired in E2F activation (Q32A, A33G, Y34F, H49R, M52V, and N56T) were similarly impaired in cyclin A activation (Fig. 6). The cyclin A data followed a pattern similar to the E2F data. For example, the mutant that had the most activity on the two promoters was Q32A, whereas Y34F had the least. Since large T activation of the mutant cyclin A promoter is independent of Rb binding, the results suggest a role for the residues within the J domain in large T activities independent of Rb.
J domain mutations and middle T. As for large T, conservative mutations in the J domain were tested for their effect on middle T. The results of Fig. 1 to 3 showed that relatively few conservative mutations affect the stability of middle T. As a screen for middle T activity, immunoprecipitates made from transfected NIH 3T3 cells were tested in an in vitro tyrosine kinase reaction. Almost without exception, single-mutant middle Ts were phosphorylated in complexes with src-family members to an extent similar to that seen with the wild type (Fig. 7A). The middle T versions of non-HPD J-defective large T mutants (A33G, M52V, and N56T) were all active in the in vitro tyrosine kinase reaction and, as expected, these mutants all transformed Rat-1 cells normally (data not shown). All of these results argue against a role for the binding of heat shock proteins in fibroblast transformation assays.
Q37A was an interesting exception. This mutation lies just outside of the HPD loop on the second helix of the J domain. In vitro kinase reactions of transfected Q37A consistently showed very little middle T phosphorylation. To confirm the transfection result, cell pools of NIH 3T3 cells expressing wild type, Q37A, and INS107AL, a mutant defective in both PP2A and src binding, were made by using coselection with puromycin. INS107AL (Fig. 7B, upper panel) was completely inactive in the tyrosine kinase reaction as expected. Q37A also showed little activity in the in vitro kinase reaction even though the protein was present at a level comparable to that seen with the wild type (Fig. 7B, lower panel). To examine the basis for the defect in Q37A-associated kinase activity, middle T immunoprecipitates were made from each of the cell pools. Blotting of these immunoprecipitates for c-src showed that wild type, but not INS107A, was able to associate with src (Fig. 7C). Q37A appeared to have relatively little binding compared to the wild type, although somewhat more than INS107AL. The binding of src to middle T is thought to depend on the association of PP2A (27, 40). To test association with PP2A, cells were harvested, extracted, and immunoprecipitated with anti-middle T antibody. The resulting immunoprecipitates were blotted for the PP2A A subunit as well as the middle T protein. As shown previously, INS107A fails to bind the A subunit of PP2A (Fig. 7D). Q37A bound the PP2A protein at levels close to that of the wild type.
The inability of Q37A to bind src leads to a defect in binding of important adaptor proteins including those mentioned above, phosphatidylinositol 3-kinase and Shc. Since both molecules are critical signaling molecules involved in middle T transformation, the activity of Q37A in transformation assays was determined. The middle T Q37A mutant, along with wild-type middle T and INS107AL(PP2A-) middle T cDNA, was transiently transfected in triplicate into NIH 3T3 cells. One plate was harvested and lysed 48 h posttransfection and assayed for middle T expression. The mutant proteins were expressed at levels close to that of the wild type (data not shown). The remaining plates were placed under reduced serum conditions and grown for 14 days. The plates were fixed and stained with crystal violet, and foci were counted. Figure 8 shows an example of each middle T transfected plate. The wild-type middle T protein was capable inducing focus formation (211 and 227 foci). Q37A, defective in src binding, was highly defective compared to the wild-type protein (34 and 28 foci). Ins107AL, defective in binding PP2A as well as src, was somewhat more defective compared to Q37A (14 and 17 foci). These results are consistent with those seen for the middle T kinase and src binding assays discussed above.
DISCUSSION
Our primary goal was a thorough genetic analysis of the J domain of polyomavirus T antigens. Previously, the study of the polyomavirus J domain has focused primarily on the importance of the conserved HPD loop (48). Some DnaJ mutants outside the HPD loop have been examined, especially for SV40 (24, 42), but little systematic analysis on DnaJ has been done (25). This work, combined with recent work on SV40 (9, 24), begins to close this gap in our knowledge.
Site-directed mutagenesis revealed several J domain residues necessary for large T and middle T stability. Comparison to the nuclear magnetic resonance (NMR) structure of the polyomavirus J domain (2) shows that, in general, the residues necessary for stability of the protein are located within the hydrophobic core of the domain. Presumably, the mutations cause unfolding that results in rapid degradation of the protein. Short pulse-labeling of the large T mutant L17V, for example, showed significant expression compared to wild type, even though the steady-state concentration measured by Western blotting was extremely low (not shown).
A number of mutations have different effects on the stability of middle T and large T. One possible explanation could be the cellular environment. Degradation of a nuclear large T mutant by cellular proteases might be easier than degradation of a membrane-associated middle T mutant. A second possibility is that J domain contacts with the rest of the protein determine how a given mutation affects stability. Such an interaction could occur within the middle T molecule itself, for example, between the J domain and C-terminal residues. Third, any protein-protein interaction providing additional stability to the domain could counteract the destabilizing effect of any given mutation. We favor the idea that there is increased stability because of an interaction between the J domain and the middle T binding partner PP2A. Unlike other associated proteins, PP2A is stoichiometrically associated with middle T. Further, it is known that PP2A and Hsc70 compete for binding to middle T (5), and our data has demonstrated that a middle T deletion mutant lacking the J domain does not bind PP2A (not shown). Here we have shown that a second mutation that blocked PP2A binding in L55V and G25A made those previously stable mutants unstable.
Genetic analysis of large T revealed residues besides those of the HPD loop that were important for large T effects on the Rb family. Large T proteins mutant at amino acids 32, 33, 34, 49, 52, and 56 were all defective in activating the E2F promoter. Large T HPD mutants cause hyperphosphorylation of the Rb family member p130 (47). The mutants outside the HPD loop also cause p130 hyperphosphorylation. The residues important for affecting Rb function are found in helix 2 (32-34) and helix 3 (49, 52, 56). As can be seen in Fig. 3B, these residues appear to be located on the surface of the domain and are supported by residues important for stability. The importance of sequences in helix 2 and helix 3 can also be seen in SV40 functional assays. Mutations made in residues Y34 (helix 2) and K53 (helix 3) result in an SV40 T antigen defective in J function (17). Interestingly, K53, although conserved in many polyomaviruses, is not present in murine polyomavirus and a Q53A mutant of polyomavirus retained significant activity (not shown). The importance of helix 2 residues 32 to 34 is also highlighted by NMR data on E. coli DnaJ. That study demonstrated that the outer surface of helix 2 is the binding site for DnaK (29). Mutagenesis results have shown that helix 3 residues can be important for E. coli DnaJ function, just as has been shown here for the polyomavirus J domain (25). NMR work suggested conformational changes in helix 3 upon DnaJ/DnaK interactions (35) that might be affected by mutation. The J mutants defective in Rb function can be separated into classes with respect to association with Hsc70. As measured by coimmunoprecipitation, mutants at residues 33 and 52 did not interact with Hsc70. However, other large T mutants at residues 32, 34, 49, and 56 retained near-wild-type levels of binding, even though they are defective in activating the ATPase activity of Hsc70.
Polyomavirus J domains participate in other activities of large T besides regulating the Rb family. In SV40, for example, the J domain is important for replication and can participate in VP1 assembly by promoting its dissociation from hsc70 (5, 9). The results here suggest that the polyomavirus large T J domain can participate in transcriptional activation that does not depend on Rb. Previous examination of the cyclin A promoter showed that, while the H42Q mutant retained activity, it was somewhat (4-fold) reduced in the ability to transactivate (48). Because of the uncertainties in protein stability and because the effect was independent of the E2F site, no conclusion was drawn from that result. Examination of all of the J mutants here showed that a defect in E2F activation was always matched by a decrease in activation of a mutant cyclin A promoter that lacked the E2F site. In particular, the activation of the mutant cyclin A promoter was dependent not only on the HPD residues but also upon residues 32 to 34 and 49, 52, and 56. Detailed work to be presented elsewhere will show that the actual target in the promoter is a Creb/ATF site (36a). Creb/ATF transcription factors function in complexes with other proteins such as CBP, Act, and Oct-1. It seems plausible that, as in the case for Rb/E2F complexes, the chaperone activity of large T targets such complexes.
Polyomavirus middle T does not rely on productive association between DnaK and the J domain for the ability to transform fibroblasts (4, 26). However, it is known that mutations near the N terminus can affect middle T binding to src and PP2A (11, 26), suggesting the possibility that other non-HPD residues could be important for middle T function. Of the 53 mutants examined here, Q37 was the only residue uncovered in the mutant screen that affected middle T transformation without simply destabilizing the protein. Interestingly, the Q37A mutation did not affect large T function in E2F or cyclin A assays (data not shown). This suggests that Q37 is not required for DnaJ activation of DnaK and further supports the idea that the J domains in large T and middle T have different roles. The effect of Q37A mutation on transformation suggests a role for the J domain separate from its action as an effector of DnaK. Since mutations in the N terminus of middle T have been reported to affect transformation (11), our observation that only mutation of Q37 affected middle T function was somewhat surprising. Since we used rather conservative substitutions, this suggests that the structure may tolerate modest changes in amino acids.
Q37A middle T was highly defective in association with c-src (Fig. 7B and C). It is not surprising that the mutant does not transform given the severe defect in tyrosine kinase activity. One possible explanation for the defect in binding is that src is positioned very close to, and is possibly even in direct contact with, the J domain. Q37A could be defective either because loss of direct contact reduces affinity for src or that a conformational change resulting from the mutation causes steric hindrance. This interpretation might seem unexpected since previous mapping of the src binding site has suggested that it is near residue 200 (3). However, there is some suggestion that the src protein may be close to the J domain. Some monoclonal antibodies directed against small T antigen will immunoprecipitate the large T and middle T proteins, so the epitopes must be within the conserved J domain sequence (41). Interestingly, two monoclonal antibodies did not immunoprecipitate middle T-associated kinase activity (41). The simplest interpretation is that the antibodies recognizing the N-terminal J domain sequences are directed against epitopes that are unavailable in the src-middle T complex. In turn, this suggests that src and the J domain are in proximity to one another. There is an alternative explanation for the failure of Q37A to bind src properly that is based on PP2A-middle T interactions. The association of src with middle T is known to be dependent upon recruitment of PP2A (27) in a manner independent of PP2A catalytic activity (40). As noted above, a truncated middle T lacking the J domain fails to bind PP2A. Although Q37A binds PP2A, the mutation could change the orientation so that src could not interact with the middle T-PP2A complex.
In conclusion, the genetics suggest that the J domain in middle T might be important for the formation of the src-middle T complex. For large T, the observation that the J domain participates in non-Rb transactivation suggests that chaperone function may affect other transcription factor protein complexes in the way in which it alters Rb-E2F complexes.
ACKNOWLEDGMENTS
This study was supported by NIH grant CA34722 to B.S.S.
REFERENCES
Berger, H., and E. Wintersberger. 1986. Polyomavirus small T antigen enhances replication of viral genomes in 3T6 mouse fibroblasts. J. Virol. 60:768-770.
Berjanskii, M. V., M. I. Riley, A. Xie, V. Semenchenko, W. R. Folk, and S. R. Van Doren. 2000. NMR structure of the N-terminal J. domain of murine polyomavirus T antigens: implications for DnaJ-like domains and for mutations of T antigens. J. Biol. Chem. 275:36094-36103.
Brewster, C. E., H. R. Glover, and S. M. Dilworth. 1997. pp60c–src binding to polyomavirus middle T-antigen (MT) requires residues 185 to 210 of the MT sequence. J. Virol. 71:5512-5520.
Campbell, K., K. Auger, B. Hemmings, T. Roberts, and P. D. 1995. Identification of regions in polyomavirus middle T and small t antigens important for association with protein phosphatase 2A. J. Virol. 69:3721-3728.
Campbell, K. S., K. P. Mullane, I. A. Aksoy, H. Stubdal, J. Zalvide, J. M. Pipas, P. A. Silver, T. M. Roberts, B. S. Schaffhausen, and J. A. DeCaprio. 1997. DnaJ/hsp40 chaperone domain of SV40 large T antigen promotes efficient viral DNA replication. Genes Dev. 11:1098-1110.
Cheetham, M. E., and A. J. Caplan. 1998. Structure, function and evolution of DnaJ: conservation and adaptation of chaperone function. Cell Stress Chaperones 3:28-36.
Chen, L., and M. M. Fluck. 2001. Role of middle T-small T in the lytic cycle of polyomavirus: control of the early-to-late transcriptional switch and viral DNA replication. J. Virol. 75:8380-8389.
Chen, M. C., D. Redenius, F. Osati-Ashtiani, and M. M. Fluck. 1995. Enhancer-mediated role for polyomavirus middle T/small T in DNA replication. J. Virol. 69:326-333.
Chromy, L. R., J. M. Pipas, and R. L. Garcea. 2003. Chaperone-mediated in vitro assembly of Polyomavirus capsids. Proc. Natl. Acad. Sci. USA 100:10477-10482.
Cole, S. L., and M. J. Tevethia. 2002. Simian virus 40 large T antigen and two independent T-antigen segments sensitize cells to apoptosis following genotoxic damage. J. Virol. 76:8420-8432.
Cook, D. N., and J. A. Hassell. 1990. The amino terminus of polyomavirus middle T antigen is required for transformation. J. Virol. 64:1879-1887.
Culleré, X., P. Rose, U. Thathamangalam, A. Chatterjee, K. Mullane, D. Pallas, T. Benjamin, T. Roberts, and B. Schaffhausen. 1998. Serine 257 phosphorylation regulates association of polyoma middle T antigen with 14-3-3 proteins. J. Virol. 72:558-565.
Cupp-Vickery, J. R., and L. E. Vickery. 2000. Crystal structure of Hsc20, a J.-type Co-chaperone from Escherichia coli. J. Mol. Biol. 304:835-845.
Cyr, D. M., T. Langer, and M. G. Douglas. 1994. DnaJ-like proteins: molecular chaperones and specific regulators of Hsp70. Trends Biochem. Sci. 19:176-181.
Dilworth, S. M. 2002. Polyoma virus middle T antigen and its role in identifying cancer-related molecules. Nat. Rev. Cancer 2:951-956.
Evans, P. R., and D. J. Owen. 2002. Endocytosis and vesicle trafficking. Curr. Opin. Struct. Biol. 12:814-821.
Fewell, S. W., J. M. Pipas, and J. L. Brodsky. 2002. Mutagenesis of a functional chimeric gene in yeast identifies mutations in the simian virus 40 large T antigen J domain. Proc. Natl. Acad. Sci. USA 99:2002-2007.
Fimia, G. M., V. Gottifredi, B. Bellei, M. R. Ricciardi, A. Tafuri, P. Amati, and R. Maione. 1998. The activity of differentiation factors induces apoptosis in polyomavirus large T-expressing myoblasts. Mol. Biol. Cell 9:1449-1463.
Fink, A. L. 1999. Chaperone-mediated protein folding. Physiol. Rev. 79:425-449.
Francke, B., and W. Eckhart. 1973. Polyoma gene function required for viral DNA synthesis. Virology 55:127-135.
Freund, R., A. Sotnikov, R. T. Bronson, and T. L. Benjamin. 1992. Polyoma virus middle T is essential for virus replication and persistence as well as for tumor induction in mice. Virology 191:716-723.
Garcea, R., and T. Benjamin. 1983. Host range transforming gene of polyomavirus plays a role in virus assembly. Proc. Natl. Acad. Sci. USA 80:3413-3417.
Garcea, R. L., D. A. Talmage, A. Harmatz, R. Freund, and T. L. Benjamin. 1989. Separation of host range from transformation functions of the hr-t gene of polyomavirus. Virology 168:312-319.
Genevaux, P., F. Lang, F. Schwager, J. V. Vartikar, K. Rundell, J. M. Pipas, C. Georgopoulos, and W. L. Kelley. 2003. Simian virus 40 T antigens and J domains: analysis of Hsp40 cochaperone functions in Escherichia coli. J. Virol. 77:10706-10713.
Genevaux, P., F. Schwager, C. Georgopoulos, and W. L. Kelley. 2002. Scanning mutagenesis identifies amino acid residues essential for the in vivo activity of the Escherichia coli DnaJ (Hsp40) J-domain. Genetics 162:1045-1053.
Glenn, G. M., and W. Eckhart. 1995. Amino-terminal regions of polyomavirus middle T antigen are required for interactions with protein phosphatase 2A. J. Virol. 69:3729-3736.
Glover, H. R., C. E. Brewster, and S. M. Dilworth. 1999. Association between src-kinases and the polyoma virus oncogene middle T-antigen requires PP2A and a specific sequence motif. Oncogene 18:4364-4370.
Graham, F. L., and A. J. van der Eb. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-467.
Greene, M., K. Maskos, and S. Landry. 1998. Role of the J-domain in the cooperation of Hsp40 with Hsp70. Proc. Natl. Acad. Sci. USA 95:6108-6113.
Holman, P., O. Gjoerup, T. Davin, and B. Schaffhausen. 1994. Characterization of an immortalizing N-terminal domain of polyomavirus large T antigen. J. Virol. 68:668-673.
Kelley, W. L., and C. Georgopoulos. 1997. The T/t common exon of simian virus 40, JC, and BK polyomavirus T antigens can functionally replace the J.-domain of the Escherichia coli DnaJ molecular chaperone. Proc. Natl. Acad. Sci. USA 94:3679-3684.
Kim, H. Y., B. Y. Ahn, and Y. Cho. 2001. Structural basis for the inactivation of retinoblastoma tumor suppressor by SV40 large T antigen. EMBO J. 20:295-304.
Konieczny, I., and M. Zylicz. 1999. Role of bacterial chaperones in DNA replication. Genet. Eng. 21:95-111.
Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
Landry, S. J. 2003. Structure and energetics of an allele-specific genetic interaction between dnaJ and dnaK: correlation of nuclear magnetic resonance chemical shift perturbations in the J-domain of Hsp40/DnaJ with binding affinity for the ATPase domain of Hsp70/DnaK. Biochemistry 42:4926-4936.
Lemmon, S. K. 2001. Clathrin uncoating: auxilin comes to life. Curr. Biol. 11:R49-R52.
Love, T. M., R. de Jesus, J. A. Kean, Q. Sheng, A. Ledger, and B. Schaffhausen. 2005. Activation of CREB/ATF sites by polyoma large T antigen. J. Virol. 79:4180-4190.
Maione, R., G. M. Fimia, P. Holman, B. Schaffhausen, and P. Amati. 1994. Retinoblastoma antioncogene is involved in the inhibition of myogenesis by polyomavirus large T antigen. Cell Growth Differ. 5:231-237.
Martens, I., S. A. Nilsson, S. Linder, and G. Magnusson. 1989. Mutational analysis of polyomavirus small-T-antigen functions in productive infection and in transformation. J. Virol. 63:2126-2133.
Mullane, K. P., M. Ratnofsky, X. Cullere, and B. Schaffhausen. 1998. Signaling from polyomavirus middle T and small T defines different roles for protein phosphatase 2A. Mol. Cell. Biol. 18:7556-7565.
Ogris, E., I. Mudrak, E. Mak, D. Gibson, and D. C. Pallas. 1999. Catalytically inactive protein phosphatase 2A can bind to polyomavirus middle tumor antigen and support complex formation with pp60c–src. J. Virol. 73:7390-7398.
Pallas, D. C., C. Schley, M. Mahoney, E. Harlow, B. S. Schaffhausen, and T. M. Roberts. 1986. Polyomavirus small t antigen: overproduction in bacteria, purification, and utilization for monoclonal and polyclonal antibody production. J. Virol. 60:1075-1084.
Peden, K. W., and J. M. Pipas. 1992. Simian virus 40 mutants with amino-acid substitutions near the amino terminus of large T antigen. Virus Genes 6:107-118.
Qian, Y. Q., D. Patel, F. U. Hartl, and D. J. x. McColl. 1996. Nuclear magnetic resonance solution structure of the human Hsp40 (HDJ-1)J-domain. J. Mol. Biol. 260:224-235.
Rassoulzadegan, M., Z. Naghasfar, A. Cowie, A. Carr, M. Grisoni, R. Kamen, and F. Cuzin. 1983. Expression of the large T protein of polyoma virus promotes the establishment in culture of "normal" rodent fibroblast cell lines. Proc. Natl. Acad. Sci. USA 80:4354-4358.
Riley, M. I., W. Yoo, N. Y. Mda, and W. R. Folk. 1997. Tiny T antigen: an autonomous polyomavirus T antigen amino-terminal domain. J. Virol. 71:6068-6074.
Sanger, F., S. Nicklen, and A. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5468.
Sheng, Q., D. Denis, M. Ratnofsky, T. Roberts, J. DeCaprio, and B. Schaffhausen. 1997. The DnaJ domain of polyoma large T is required to regulate Rb family tumor suppressor function. J. Virol. 71:9410-9416.
Sheng, Q., T. M. Love, and B. Schaffhausen. 2000. J. domain-independent regulation of the Rb family by polyomavirus large T antigen. J. Virol. 74:5280-5290.
Silver, J., B. Schaffhausen, and T. Benjamin. 1978. Tumor antigens induced by nontransforming mutants of polyoma virus. Cell 15:485-496.
Srinivasan, A., A. J. McClellan, J. Vartikar, I. Marks, P. Cantalupo, Y. Li, P. Whyte, K. Rundell, J. L. Brodsky, and J. M. Pipas. 1997. The amino-terminal transforming region of simian virus 40 large T and small t antigens functions as a J domain. Mol. Cell. Biol. 17:4761-4773.
Stubdal, H., J. Zalvide, K. S. Campbell, C. Schweitzer, T. M. Roberts, and J. A. DeCaprio. 1997. Inactivation of pRB-related proteins p130 and p107 mediated by the J. domain of simian virus 40 large T antigen. Mol. Cell. Biol. 17:4979-4990.
Sullivan, C. S., P. Cantalupo, and J. M. Pipas. 2000. The molecular chaperone activity of simian virus 40 large T antigen is required to disrupt Rb-E2F family complexes by an ATP-dependent mechanism. Mol. Cell. Biol. 20:6233-6243.
Sullivan, C. S., and J. M. Pipas. 2002. T antigens of simian virus 40: molecular chaperones for viral replication and tumorigenesis. Microbiol. Mol. Biol. Rev. 66:179-202.
Szyperski, T., M. Pellecchia, D. Wall, C. Georgopoulos, and K. Wuthrich. 1994. NMR structure determination of the Escherichia coli DnaJ molecular chaperone: secondary structure and backbone fold of the N-terminal region (residues 2-108) containing the highly conserved J domain. Proc. Natl. Acad. Sci. USA 91:11343-11347.
Templeton, D., S. Simon, and W. Eckhart. 1986. Truncated forms of the polyomavirus middle T antigen can substitute for the small T antigen in lytic infection. J. Virol. 57:367-370.
Treisman, R., A. Cowie, J. Favaloro, P. Jat, and R. Kamen. 1981. The structures of spliced mRNAs encoding polyoma virus early region proteins. J. Mol. Appl. Gen. 1:83-92.
Yaglom, J., A. L. Goldberg, D. Finley, and M. Sherman. 1996. The molecular chaperone Ydj1 is required for p34cdc28-dependent phosphorylation of Cln 3 that signals its degradation. Mol. Cell. Biol. 16:3679-3684.
Zalvide, J., H. Stubdal, and J. DeCaprio. 1998. The J domain of SV40 large T antigen is required to functionally inactivate Rb family proteins. Mol. Cell. Biol. 18:1408-1415.(Kerry A. Whalen, Rowena d)
ABSTRACT
Polyomavirus T antigens share a common N-terminal sequence that comprises a DnaJ domain. DnaJ domains activate DnaK molecular chaperones. The functions of J domains have primarily been tested by mutation of their conserved HPD residues. Here, we report detailed mutagenesis of the polyomavirus J domain in both large T (63 mutants) and middle T (51 mutants) backgrounds. As expected, some J mutants were defective in binding DnaK (Hsc70); other mutants retained the ability to bind Hsc70 but were defective in stimulating its ATPase activity. Moreover, the J domain behaves differently in large T and middle T. A given mutation was twice as likely to render large T unstable as it was to affect middle T stability. This apparently arose from middle T's ability to bind stabilizing proteins such as protein phosphatase 2A (PP2A), since introduction of a second mutation preventing PP2A binding rendered some middle T J-domain mutants unstable. In large T, the HPD residues are critical for Rb-dependent effects on the host cell. Residues Q32, A33, Y34, H49, M52, and N56 within helix 2 and helix 3 of the large T J domain were also found to be required for Rb-dependent transactivation. Cyclin A promoter assays showed that J domain function also contributes to large T transactivation that is independent of Rb. Single point mutations in middle T were generally without effect. However, residue Q37 is critical for middle T's ability to form active signaling complexes. The Q37A middle T mutant was defective in association with pp60c-src and in transformation.
INTRODUCTION
Polyomavirus T antigens function both in replication of the virus and in transformation of the host cell. Large T is central to virus production as the initiator of viral DNA replication (20). Middle T and small T also play important roles in different aspects of polyomavirus infection (21, 23, 55). Defects in viral DNA replication and transcription, as well as defects in viral assembly, have been observed in different mutants of middle T and small T (1, 7, 8, 22, 38). Each of the viral early proteins also contributes to regulation of host cell function. Large T is able to immortalize primary cells (44), to block differentiation (37), and to provoke apoptosis (18, 48). These activities are mediated via association with the retinoblastoma susceptibility (Rb) family of tumor suppressors. Middle T, the major transforming protein, works through activation of cellular signaling pathways that are regulated by src-family tyrosine phosphorylation (15). Small T is able to promote cell cycle progression via association with protein phosphatase 2A (PP2A) (39).
All three T antigens are produced by differential splicing of common primary transcripts (56). As a result, they have the identical N-terminal sequence of 79 amino acids that encompasses a DnaJ domain. DnaJ domains, consisting of approximately 70 amino acids, have a helical structure in which a conserved HPD motif is found between helix 2 and helix 3 (2, 13, 32, 43, 54). DnaJ domains, found in a broad range of proteins, function to stimulate the activity of DnaKs (6, 14). The domains of polyomavirus and simian virus 40 (SV40), like other DnaJs, have been shown to activate the ATPase activity of DnaKs (45, 50, 52). Substitution experiments have shown that SV40 or BK domains can function biologically as DnaJ domains in Escherichia coli (31).
The DnaJ/K cochaperones have diverse cellular activities, including protein folding (19), protein degradation (57), vesicle sorting/uncoating (16, 36), and DNA replication (5, 33). The importance of the J domain to both polyomavirus and SV40 large T function is clear (47, 53). The ability of SV40 large T to drive DNA replication in vivo is highly dependent on an intact DnaJ domain (5). The ability of SV40 large T to transform is dependent on DnaJ function (50, 58). The J domain of SV40 large T has also been implicated in sensitizing cells to apoptosis after genotoxic damage (10). A recent report indicates that the J domain is important for the ability of SV40 large T to release VP1 from complexes with Hsc70 (9). Polyomavirus large T promotes cell cycle progression in a DnaJ-dependent manner (47). Much, but not all (48), of large T's ability to act on the Rb family is dependent on an intact J domain (47, 51, 52, 58). Mutations of the HPD loop that abolish interaction of large T with Hsc70 (5, 48) affect the ability of SV40 large T to alter the turnover of p130 (51). Work done with SV40 by the Pipas group has demonstrated an intact HPD loop is critical for dissociation of Rb/E2F complexes (52), which would account for large T promoter activation at E2F sites. Less is known about the role of the J domain in polyomavirus middle T and small T. Although some mutations in the middle T J domain have been shown to affect transforming ability (11), mutation of the HPD loop was found not to alter the ability of middle T to transform cells (4, 26). This suggests that activation of DnaK, the normal role for DnaJ, is not required for transformation.
There has been little systematic genetic analysis of J domains, even for E. coli DnaJ (17, 24, 25). To map the DnaJ residues necessary for polyomavirus large T function, we carried out extensive mutagenesis of the J domain. This identified residues important for the stability of large T and for the ability of large T to act on the Rb family. These experiments also suggested that the domain is important for large T functions beyond those related to Rb. Further, we compared the effects of mutations in the J domain of large T to the same mutations in middle T. Our results indicated that the two proteins use the J domain differently.
MATERIALS AND METHODS
Cell lines and transfections. NIH 3T3 cells were originally obtained from the American Type Culture Collection. The cells were grown in Dulbecco modified Eagle medium (DMEM; Gibco) and supplemented with 10% calf serum (HyClone). Transfections were performed by the calcium phosphate precipitation method (47). NIH 3T3 cell lines expressing different middle T antigens were obtained by cotransfection of NIH 3T3 cells with a vector for puromycin resistance and the relevant pCMV MT construct at a ratio of 1:10. At 2 days posttransfection the cells were placed in selection medium of DMEM containing 10% calf serum and puromycin at 5 μg/ml to obtain pools of puromycin-resistant colonies.
C33A cells were obtained from Vimla Band and were grown in DMEM supplemented with 10% fetal calf serum (HyClone). C33A cell transfections were performed using HEPES-buffered saline (28). For starvation experiments, cells were washed twice with phosphate-buffered saline (PBS) 18 h after transfection and placed in DMEM containing 0.2% fetal calf serum.
Plasmids and mutagenesis. pCMV large T, pCMV Rb–LT, pCMV P43S large T, HA-Rb, HA-p130 (47), pCMV middle T, pCMV NG59 middle T (12), and INS107AL (5) have all been described previously. Point mutants were constructed on wild-type pCMV large T or pCMV middle T. In general, oligonucleotides of 18 to 20 bases with one or two mismatches to the J domain sequence were used. Standard PCR methods based on PFU Turbo polymerase (Stratagene) were used to generate the mutants. All mutations were confirmed using the dideoxy sequencing method (46). A summary of the mutants constructed is shown in Fig. 1.
The E2F-luciferase construct contains six E2F sites attached to a luciferase construct and was obtained from Amy Yee. The myc-tagged Hsc70 was obtained from Tom Roberts. The cyclin A –89-to-+11 promoter construct has been described previously (48).
Antibodies. The PN116 monoclonal and rabbit anti-T antibody used in Western blots have been described previously (30). Middle T was immunoprecipitated with rabbit polyclonal antibody 45-1 (39). HA11 was obtained from Covance. Myc antibody 9E10 was obtained from James DeCaprio. Polyclonal PP2A-A subunit antibody was obtained from Gernot Walter and Ralf Ruediger. Anti-src antibody (GD11) was obtained from Larry Feig.
Luciferase assays. NIH 3T3 cells plated on 60-mm dishes were transfected at a confluence of 20% with 2 μg of E2F-luciferase or with 1 μg cyclin A-luciferase and 0.5 μg of the relevant LT expression vectors and harvested approximately 40 h posttransfection. For the cyclin A-luciferase assays, cells were placed under serum-starved conditions at 24 h posttransfection. Cells were suspended in buffer (25 mM Tris [pH 7.5], 1 mM EDTA) and subjected to freezing-thawing three times. The lysates were cleared by Eppendorf centrifugation and assayed for luciferase activity.
Extraction and immunoprecipitations. The cells were washed in PBS and extracted in T extraction buffer (TEB; 137 mM NaCl, 10 mM Tris-Cl [pH 8.0 for HA-Rb, pH 7.0 for myc-Hsc70, and pH 9.0 for middle T], 1 mM MgCl2, 1 mM CaCl2 10% [vol/vol] glycerol, and 1% [vol/vol] Nonidet P-40 supplemented with protease inhibitor cocktail, including leupeptin at 1 μg/ml, pepstatin 1 at μg/ml, phenylmethylsulfonyl fluoride at 100 μg/ml, and aprotinin at 2 μg/ml) for 20 minutes at 4°C. The cleared extracts were incubated with the appropriate antibody and protein G- or protein A-Sepharose beads (Pharmacia). The immunoprecipitates were washed with PBS. The immunoprecipitates were boiled in dissociation buffer (62.5 mM Tris-Cl [pH 6.8], 5% [wt/vol] sodium dodecyl sulfate [SDS], 25% [vol/vol] glycerol, 0.0075% [wt/vol] bromophenol blue, 5% [vol/vol] ?-mercaptoethanol).
Electrophoresis. Samples were analyzed on discontinuous buffer SDS gels (34) of 7.5% acrylamide. Radiolabeled gels were exposed and quantified by PhosphorImager (Molecular Dynamics) using ImageQuant software. All of the figures were prepared in Adobe Photoshop.
Large T NT purification. The N-terminal domains (NT; residues 1 to 259) of large T expressed as glutathione S-transferase fusions in pGEX3X were produced in E. coli BL21. Expression was induced with 50 μM IPTG (isopropyl-?-D-thiogalactopyranoside) overnight at 30°C. Cell extracts were incubated with glutathione-beads (Sigma) to harvest the fusion proteins. Washed beads were treated overnight at 4°C with factor Xa (Haematologic Technologies, Inc.) in buffer (20 mM Tris-Cl [pH 8.0], 1 mM calcium chloride, 100 mM NaCl) to release NT. The soluble NT was concentrated by Amicon ultrafiltration.
ATPase assays. The assays were performed in a 20-μl total volume containing 75 mM NaCl, 1.5 mM CaCl2, 40 mM HEPES (pH 7.2), and 2 μCi of [-32P]ATP with 0.4 μg of mouse Hsc70 (Stressgen). In some cases the reactions were supplemented with 0.8 μg of wild-type or mutant purified NT protein. The reaction mixtures were assembled on ice and then incubated at 37°C. The reactions were stopped by plating 2-μl aliquots in triplicate onto polyethyleneimine thin-layer chromatography plates. ATP and ADP were separated by chromatography in 1 M LiCl-1 M formic acid solution for 3 h. The plates were dried and analyzed by using a Molecular Dynamics PhosphorImager.
Middle T kinase assays. NIH 3T3 cells plated onto 100-mm dishes were transfected at a confluence of 20% with 5 μg of different middle T expression vectors and harvested approximately 40 h posttransfection. After extraction in TEB in the presence of protease inhibitors, the cleared extracts were incubated with rabbit polyclonal antibody to middle T (45-1) and protein A-Sepharose beads for 1 h. Washed immunoprecipitates were incubated in 200 μl of kinase buffer (20 mM Tris [pH 7.5], 5 mM MnCl2) containing 5 to 10 μCi of [-32P]ATP (2,000 Ci/mmol; NEN) for 15 min at room temperature. The reactions were washed with PBS, boiled in dissociation buffer, and subjected to SDS-polyacrylamide gel electrophoresis (PAGE). After electrophoresis, the gel was alkaline treated (0.5 M potassium hydroxide) for 1 h at 55°C to remove background serine/threonine phosphorylation. The gel was then fixed (10% acetic acid, 30% methanol), dried, and analyzed with a PhosphorImager.
Transformation assays. NIH 3T3 cells were grown in 10% calf serum and transfected with 7 μg of carrier DNA and 2 μg of the middle T cDNAs by using the calcium phosphate method. When the confluency reached 100%, the cells were maintained in 5% calf serum-DMEM and allowed to grow for 10 to 14 days. The plates were then washed twice with PBS and fixed for 30 min with 3.7% formaldehyde. The fixed plates were stained with 0.2% crystal violet in 3.7% formaldehyde. Foci were then counted.
RESULTS
Mutant analysis of the polyomavirus J domain demonstrates differences between the middle T and large T proteins. Site-directed mutagenesis was used to screen the J domains of middle and large T antigens. Point mutants were made in vectors expressing either middle T or large T using conservative substitutions such as lysine to arginine or leucine to valine. The mutants are listed in Fig. 1. In all, 63 mutants of large T and 51 mutants of middle T were examined.
The stabilities of mutant large T antigens were compared to wild type after transient expression in NIH 3T3 cells. Cell extracts prepared approximately 40 h after transfection were subjected to SDS-PAGE. Immunoblotting was then used to assess the level of large T protein. Figure 2A shows an example of the results. A majority of the mutants, including P43S, Q36A, and F62Y, retained wild-type levels of expression, whereas W59F large T showed a very low level of expression. In all, nine of the large T mutants showed a substantial decrease in protein levels (Fig. 1). The simplest interpretation of this observation is that the mutations cause unfolding of the domain leading to degradation of the protein. Figure 3A shows the positions in the J structure of mutations that caused loss of large T stability. Four DnaJ mutations also rendered middle T unstable (Fig. 1 and 3C). Two examples, R12K and L13V, are shown in Fig. 2B. Seven mutations (K10R, L17V, M30V, G25A, F27Y, L55V, and W59F) that rendered large T unstable did not appreciably affect the stability of middle T. On the other hand, mutation R12K (Fig. 2B) was unstable in the middle T background but not in large T.
The results just described suggest that mutations of the J domains of large T and middle T have different effects on each protein even though the N-terminal sequences are identical. Possible explanations for the divergence lie within the differences in the C-terminal residues of the two proteins, as well as their association with different cellular proteins. A key difference between middle T and large T is middle T association with PP2A. Furthermore, it is also known that PP2A and Hsc70 compete for binding to middle T (4) suggesting contact may exist between PP2A and Hsc70. If PP2A were to contact the J domain, this interaction might provide additional stability. To test this hypothesis, L55V and G25A mutants of middle T were combined with the NG59 mutation that prevents PP2A binding (49). Figure 2C shows that the two double mutants are significantly less stable than a middle T mutant only in the J domain. This result is also true in the case of the M30V middle T mutant (data not shown). This result suggests that PP2A binding renders middle T more resistant to mutations in the J domain.
Effects of J mutations on large T Rb function. Large T activation of promoters containing E2F sites was shown to be dependent upon an intact Rb binding site, as well as an intact HPD loop in the J domain (52). Earlier experiments suggested a role for some residues outside of the HPD loop in large T for the activation of E2F promoters (47). Each of the mutants in large T was tested for the ability to activate the E2F promoter. Mutants Q32A, A33G, Y34F, H49R, M52V, and N56T were all defective in E2F promoter activity as assayed by cotransfection of the relevant large T constructs and an E2F-luciferase reporter construct (Fig. 4A, upper panel). As expected, the HPD mutant, P43S, was defective in E2F promoter activity (Fig. 4A). The non-HPD mutants defective in E2F activation are found in helix 2 and helix 3 (shown in blue in Fig. 3B). Previously, it has been shown that mutations in the HPD loop that abolish E2F activation have no effect on binding of Rb (47). Coimmunoprecipitation experiments showed that this was also true for the non-HPD mutants (not shown). In our E2F analysis, we also tested C-terminal mutants, including M71V and T76S. Structural analysis (2) suggested that these sequences, while part of the first exon of the T antigens, may not be part of the J domain but rather serve as a linker. These mutants were all wild type for E2F activation, so our data are consistent with this hypothesis.
Polyomavirus large T antigens affect the phosphorylation status of the pRB family member, p130 (52, 58). Mutations within the HPD loop of the J domain of polyomavirus cause a supershift in the p130 molecule seen on SDS-PAGE. The supershift is caused by phosphorylation of the p130 molecule in the C-terminal portion of the molecule. It is has been demonstrated that SV40 large T causes a dephosphorylation of the p130 molecule (57, 58). This result suggests that the large T protein is important for proper cycling of the p130 molecule and that the J mutant is unable to cycle the molecule. Instead, the J mutant leaves the p130 "stuck" in a hyperphosphorylated state. To test whether the non-HPD mutants behaved similarly, experiments were carried out in C33A cells with the various non-HPD mutants and HA-tagged p130 protein. SDS-PAGE analysis of the samples and subsequent immunoblotting with anti-HA was done. Fig. 4B shows that the non-HPD mutants also caused a similar shift in the p130 protein. Therefore, all J mutants defective in the ability to activate E2F sites also affected p130 phosphorylation.
Neither the defect in E2F activation nor the effect on p130 could be explained by differences in protein expression. As shown in Fig. 4A these mutants were expressed at levels close to wild type, suggesting that their defect was caused by loss of J domain activity rather than protein instability.
Interactions of mutant large T's with Hsc70. The HPD loop of large T is known to be critical for Hsc70 binding (5, 47). This binding is necessary for large T transactivation, since reducing large T-Hsc70 interactions by overexpression of exogenous DnaJ domains squelches transactivation of E2F promoters (47). To examine the interaction of the non-HPD mutants with Hsc70, wild-type or mutant large T was cotransfected with myc-tagged Hsc70. Coimmunoprecipitation and Western blotting were used to determine binding of Hsc70 to large T. The experiment in Fig. 5A demonstrates that the P43S HPD and two of six non-HPD (A33G and M52V) mutants were unable to bind the cellular heat shock protein. Figure 5B shows that lack of binding did not result from poor protein expression. A33G retains a low level of activity in the E2F assay despite the lack of hsc70 binding. It is likely that A33G has a lowered affinity for hsc70 that permits modest in vivo activity but does not result in stable complexes that can be immunoprecipitated. However, Q32A, Y34F, H49R, and N56T mutants that were defective in activation of E2F sites were still able to bind the Hsc70 protein in a manner similar to that of the wild type.
Since binding Hsc70 might not be sufficient for stimulation of its ATPase function, activity measurements were also performed. To do this, wild-type, Y34F, and H49R N-terminal domains (residues 1 to 259) were constructed and purified from E. coli using an approach similar to that of Riley et al. (45). The N-terminal domain was used because it can be easily expressed in E. coli as a soluble protein. A time course of ATP hydrolysis was determined for Hsc70 using [-32P]ATP in the presence of wild-type, Y34F, or H49R proteins. In these assays, wild-type NT was able to stimulate Hsc70 ATPase activity, whereas the J domain mutants were defective (Fig. 5C). It seems likely that the defect in E2F activation by Y34F, as well as H49R and probably other mutants that retain binding, is caused by an inability to stimulate DnaK function.
J domain effects on the cyclin A promoter. Whether the J domain participates in large T functions not related to Rb is an open question. Previous work has shown that large T can transactivate the cyclin A promoter (18, 48). In serum-starved cells this activation is completely dependent on Rb binding. However, when the E2F site in the promoter is mutated (–37/–33), large T still transactivates, but that activation is now completely independent of Rb binding (48). Our earlier work showed that HPD mutants in the J domain were also active, although at levels reduced from wild type (48). To determine whether the reduced activity indicated a J domain contribution or whether it simply represented a general loss of protein structure, the new set of J domain mutants were analyzed in cyclin A promoter assays. To avoid E2F effects, a mutant cyclin A (–37/–33) reporter lacking the E2F site was used. Every mutant that was wild type in the E2F assay was also fully competent in activation of the mutant cyclin A promoter (data not shown). The mutants impaired in E2F activation (Q32A, A33G, Y34F, H49R, M52V, and N56T) were similarly impaired in cyclin A activation (Fig. 6). The cyclin A data followed a pattern similar to the E2F data. For example, the mutant that had the most activity on the two promoters was Q32A, whereas Y34F had the least. Since large T activation of the mutant cyclin A promoter is independent of Rb binding, the results suggest a role for the residues within the J domain in large T activities independent of Rb.
J domain mutations and middle T. As for large T, conservative mutations in the J domain were tested for their effect on middle T. The results of Fig. 1 to 3 showed that relatively few conservative mutations affect the stability of middle T. As a screen for middle T activity, immunoprecipitates made from transfected NIH 3T3 cells were tested in an in vitro tyrosine kinase reaction. Almost without exception, single-mutant middle Ts were phosphorylated in complexes with src-family members to an extent similar to that seen with the wild type (Fig. 7A). The middle T versions of non-HPD J-defective large T mutants (A33G, M52V, and N56T) were all active in the in vitro tyrosine kinase reaction and, as expected, these mutants all transformed Rat-1 cells normally (data not shown). All of these results argue against a role for the binding of heat shock proteins in fibroblast transformation assays.
Q37A was an interesting exception. This mutation lies just outside of the HPD loop on the second helix of the J domain. In vitro kinase reactions of transfected Q37A consistently showed very little middle T phosphorylation. To confirm the transfection result, cell pools of NIH 3T3 cells expressing wild type, Q37A, and INS107AL, a mutant defective in both PP2A and src binding, were made by using coselection with puromycin. INS107AL (Fig. 7B, upper panel) was completely inactive in the tyrosine kinase reaction as expected. Q37A also showed little activity in the in vitro kinase reaction even though the protein was present at a level comparable to that seen with the wild type (Fig. 7B, lower panel). To examine the basis for the defect in Q37A-associated kinase activity, middle T immunoprecipitates were made from each of the cell pools. Blotting of these immunoprecipitates for c-src showed that wild type, but not INS107A, was able to associate with src (Fig. 7C). Q37A appeared to have relatively little binding compared to the wild type, although somewhat more than INS107AL. The binding of src to middle T is thought to depend on the association of PP2A (27, 40). To test association with PP2A, cells were harvested, extracted, and immunoprecipitated with anti-middle T antibody. The resulting immunoprecipitates were blotted for the PP2A A subunit as well as the middle T protein. As shown previously, INS107A fails to bind the A subunit of PP2A (Fig. 7D). Q37A bound the PP2A protein at levels close to that of the wild type.
The inability of Q37A to bind src leads to a defect in binding of important adaptor proteins including those mentioned above, phosphatidylinositol 3-kinase and Shc. Since both molecules are critical signaling molecules involved in middle T transformation, the activity of Q37A in transformation assays was determined. The middle T Q37A mutant, along with wild-type middle T and INS107AL(PP2A-) middle T cDNA, was transiently transfected in triplicate into NIH 3T3 cells. One plate was harvested and lysed 48 h posttransfection and assayed for middle T expression. The mutant proteins were expressed at levels close to that of the wild type (data not shown). The remaining plates were placed under reduced serum conditions and grown for 14 days. The plates were fixed and stained with crystal violet, and foci were counted. Figure 8 shows an example of each middle T transfected plate. The wild-type middle T protein was capable inducing focus formation (211 and 227 foci). Q37A, defective in src binding, was highly defective compared to the wild-type protein (34 and 28 foci). Ins107AL, defective in binding PP2A as well as src, was somewhat more defective compared to Q37A (14 and 17 foci). These results are consistent with those seen for the middle T kinase and src binding assays discussed above.
DISCUSSION
Our primary goal was a thorough genetic analysis of the J domain of polyomavirus T antigens. Previously, the study of the polyomavirus J domain has focused primarily on the importance of the conserved HPD loop (48). Some DnaJ mutants outside the HPD loop have been examined, especially for SV40 (24, 42), but little systematic analysis on DnaJ has been done (25). This work, combined with recent work on SV40 (9, 24), begins to close this gap in our knowledge.
Site-directed mutagenesis revealed several J domain residues necessary for large T and middle T stability. Comparison to the nuclear magnetic resonance (NMR) structure of the polyomavirus J domain (2) shows that, in general, the residues necessary for stability of the protein are located within the hydrophobic core of the domain. Presumably, the mutations cause unfolding that results in rapid degradation of the protein. Short pulse-labeling of the large T mutant L17V, for example, showed significant expression compared to wild type, even though the steady-state concentration measured by Western blotting was extremely low (not shown).
A number of mutations have different effects on the stability of middle T and large T. One possible explanation could be the cellular environment. Degradation of a nuclear large T mutant by cellular proteases might be easier than degradation of a membrane-associated middle T mutant. A second possibility is that J domain contacts with the rest of the protein determine how a given mutation affects stability. Such an interaction could occur within the middle T molecule itself, for example, between the J domain and C-terminal residues. Third, any protein-protein interaction providing additional stability to the domain could counteract the destabilizing effect of any given mutation. We favor the idea that there is increased stability because of an interaction between the J domain and the middle T binding partner PP2A. Unlike other associated proteins, PP2A is stoichiometrically associated with middle T. Further, it is known that PP2A and Hsc70 compete for binding to middle T (5), and our data has demonstrated that a middle T deletion mutant lacking the J domain does not bind PP2A (not shown). Here we have shown that a second mutation that blocked PP2A binding in L55V and G25A made those previously stable mutants unstable.
Genetic analysis of large T revealed residues besides those of the HPD loop that were important for large T effects on the Rb family. Large T proteins mutant at amino acids 32, 33, 34, 49, 52, and 56 were all defective in activating the E2F promoter. Large T HPD mutants cause hyperphosphorylation of the Rb family member p130 (47). The mutants outside the HPD loop also cause p130 hyperphosphorylation. The residues important for affecting Rb function are found in helix 2 (32-34) and helix 3 (49, 52, 56). As can be seen in Fig. 3B, these residues appear to be located on the surface of the domain and are supported by residues important for stability. The importance of sequences in helix 2 and helix 3 can also be seen in SV40 functional assays. Mutations made in residues Y34 (helix 2) and K53 (helix 3) result in an SV40 T antigen defective in J function (17). Interestingly, K53, although conserved in many polyomaviruses, is not present in murine polyomavirus and a Q53A mutant of polyomavirus retained significant activity (not shown). The importance of helix 2 residues 32 to 34 is also highlighted by NMR data on E. coli DnaJ. That study demonstrated that the outer surface of helix 2 is the binding site for DnaK (29). Mutagenesis results have shown that helix 3 residues can be important for E. coli DnaJ function, just as has been shown here for the polyomavirus J domain (25). NMR work suggested conformational changes in helix 3 upon DnaJ/DnaK interactions (35) that might be affected by mutation. The J mutants defective in Rb function can be separated into classes with respect to association with Hsc70. As measured by coimmunoprecipitation, mutants at residues 33 and 52 did not interact with Hsc70. However, other large T mutants at residues 32, 34, 49, and 56 retained near-wild-type levels of binding, even though they are defective in activating the ATPase activity of Hsc70.
Polyomavirus J domains participate in other activities of large T besides regulating the Rb family. In SV40, for example, the J domain is important for replication and can participate in VP1 assembly by promoting its dissociation from hsc70 (5, 9). The results here suggest that the polyomavirus large T J domain can participate in transcriptional activation that does not depend on Rb. Previous examination of the cyclin A promoter showed that, while the H42Q mutant retained activity, it was somewhat (4-fold) reduced in the ability to transactivate (48). Because of the uncertainties in protein stability and because the effect was independent of the E2F site, no conclusion was drawn from that result. Examination of all of the J mutants here showed that a defect in E2F activation was always matched by a decrease in activation of a mutant cyclin A promoter that lacked the E2F site. In particular, the activation of the mutant cyclin A promoter was dependent not only on the HPD residues but also upon residues 32 to 34 and 49, 52, and 56. Detailed work to be presented elsewhere will show that the actual target in the promoter is a Creb/ATF site (36a). Creb/ATF transcription factors function in complexes with other proteins such as CBP, Act, and Oct-1. It seems plausible that, as in the case for Rb/E2F complexes, the chaperone activity of large T targets such complexes.
Polyomavirus middle T does not rely on productive association between DnaK and the J domain for the ability to transform fibroblasts (4, 26). However, it is known that mutations near the N terminus can affect middle T binding to src and PP2A (11, 26), suggesting the possibility that other non-HPD residues could be important for middle T function. Of the 53 mutants examined here, Q37 was the only residue uncovered in the mutant screen that affected middle T transformation without simply destabilizing the protein. Interestingly, the Q37A mutation did not affect large T function in E2F or cyclin A assays (data not shown). This suggests that Q37 is not required for DnaJ activation of DnaK and further supports the idea that the J domains in large T and middle T have different roles. The effect of Q37A mutation on transformation suggests a role for the J domain separate from its action as an effector of DnaK. Since mutations in the N terminus of middle T have been reported to affect transformation (11), our observation that only mutation of Q37 affected middle T function was somewhat surprising. Since we used rather conservative substitutions, this suggests that the structure may tolerate modest changes in amino acids.
Q37A middle T was highly defective in association with c-src (Fig. 7B and C). It is not surprising that the mutant does not transform given the severe defect in tyrosine kinase activity. One possible explanation for the defect in binding is that src is positioned very close to, and is possibly even in direct contact with, the J domain. Q37A could be defective either because loss of direct contact reduces affinity for src or that a conformational change resulting from the mutation causes steric hindrance. This interpretation might seem unexpected since previous mapping of the src binding site has suggested that it is near residue 200 (3). However, there is some suggestion that the src protein may be close to the J domain. Some monoclonal antibodies directed against small T antigen will immunoprecipitate the large T and middle T proteins, so the epitopes must be within the conserved J domain sequence (41). Interestingly, two monoclonal antibodies did not immunoprecipitate middle T-associated kinase activity (41). The simplest interpretation is that the antibodies recognizing the N-terminal J domain sequences are directed against epitopes that are unavailable in the src-middle T complex. In turn, this suggests that src and the J domain are in proximity to one another. There is an alternative explanation for the failure of Q37A to bind src properly that is based on PP2A-middle T interactions. The association of src with middle T is known to be dependent upon recruitment of PP2A (27) in a manner independent of PP2A catalytic activity (40). As noted above, a truncated middle T lacking the J domain fails to bind PP2A. Although Q37A binds PP2A, the mutation could change the orientation so that src could not interact with the middle T-PP2A complex.
In conclusion, the genetics suggest that the J domain in middle T might be important for the formation of the src-middle T complex. For large T, the observation that the J domain participates in non-Rb transactivation suggests that chaperone function may affect other transcription factor protein complexes in the way in which it alters Rb-E2F complexes.
ACKNOWLEDGMENTS
This study was supported by NIH grant CA34722 to B.S.S.
REFERENCES
Berger, H., and E. Wintersberger. 1986. Polyomavirus small T antigen enhances replication of viral genomes in 3T6 mouse fibroblasts. J. Virol. 60:768-770.
Berjanskii, M. V., M. I. Riley, A. Xie, V. Semenchenko, W. R. Folk, and S. R. Van Doren. 2000. NMR structure of the N-terminal J. domain of murine polyomavirus T antigens: implications for DnaJ-like domains and for mutations of T antigens. J. Biol. Chem. 275:36094-36103.
Brewster, C. E., H. R. Glover, and S. M. Dilworth. 1997. pp60c–src binding to polyomavirus middle T-antigen (MT) requires residues 185 to 210 of the MT sequence. J. Virol. 71:5512-5520.
Campbell, K., K. Auger, B. Hemmings, T. Roberts, and P. D. 1995. Identification of regions in polyomavirus middle T and small t antigens important for association with protein phosphatase 2A. J. Virol. 69:3721-3728.
Campbell, K. S., K. P. Mullane, I. A. Aksoy, H. Stubdal, J. Zalvide, J. M. Pipas, P. A. Silver, T. M. Roberts, B. S. Schaffhausen, and J. A. DeCaprio. 1997. DnaJ/hsp40 chaperone domain of SV40 large T antigen promotes efficient viral DNA replication. Genes Dev. 11:1098-1110.
Cheetham, M. E., and A. J. Caplan. 1998. Structure, function and evolution of DnaJ: conservation and adaptation of chaperone function. Cell Stress Chaperones 3:28-36.
Chen, L., and M. M. Fluck. 2001. Role of middle T-small T in the lytic cycle of polyomavirus: control of the early-to-late transcriptional switch and viral DNA replication. J. Virol. 75:8380-8389.
Chen, M. C., D. Redenius, F. Osati-Ashtiani, and M. M. Fluck. 1995. Enhancer-mediated role for polyomavirus middle T/small T in DNA replication. J. Virol. 69:326-333.
Chromy, L. R., J. M. Pipas, and R. L. Garcea. 2003. Chaperone-mediated in vitro assembly of Polyomavirus capsids. Proc. Natl. Acad. Sci. USA 100:10477-10482.
Cole, S. L., and M. J. Tevethia. 2002. Simian virus 40 large T antigen and two independent T-antigen segments sensitize cells to apoptosis following genotoxic damage. J. Virol. 76:8420-8432.
Cook, D. N., and J. A. Hassell. 1990. The amino terminus of polyomavirus middle T antigen is required for transformation. J. Virol. 64:1879-1887.
Culleré, X., P. Rose, U. Thathamangalam, A. Chatterjee, K. Mullane, D. Pallas, T. Benjamin, T. Roberts, and B. Schaffhausen. 1998. Serine 257 phosphorylation regulates association of polyoma middle T antigen with 14-3-3 proteins. J. Virol. 72:558-565.
Cupp-Vickery, J. R., and L. E. Vickery. 2000. Crystal structure of Hsc20, a J.-type Co-chaperone from Escherichia coli. J. Mol. Biol. 304:835-845.
Cyr, D. M., T. Langer, and M. G. Douglas. 1994. DnaJ-like proteins: molecular chaperones and specific regulators of Hsp70. Trends Biochem. Sci. 19:176-181.
Dilworth, S. M. 2002. Polyoma virus middle T antigen and its role in identifying cancer-related molecules. Nat. Rev. Cancer 2:951-956.
Evans, P. R., and D. J. Owen. 2002. Endocytosis and vesicle trafficking. Curr. Opin. Struct. Biol. 12:814-821.
Fewell, S. W., J. M. Pipas, and J. L. Brodsky. 2002. Mutagenesis of a functional chimeric gene in yeast identifies mutations in the simian virus 40 large T antigen J domain. Proc. Natl. Acad. Sci. USA 99:2002-2007.
Fimia, G. M., V. Gottifredi, B. Bellei, M. R. Ricciardi, A. Tafuri, P. Amati, and R. Maione. 1998. The activity of differentiation factors induces apoptosis in polyomavirus large T-expressing myoblasts. Mol. Biol. Cell 9:1449-1463.
Fink, A. L. 1999. Chaperone-mediated protein folding. Physiol. Rev. 79:425-449.
Francke, B., and W. Eckhart. 1973. Polyoma gene function required for viral DNA synthesis. Virology 55:127-135.
Freund, R., A. Sotnikov, R. T. Bronson, and T. L. Benjamin. 1992. Polyoma virus middle T is essential for virus replication and persistence as well as for tumor induction in mice. Virology 191:716-723.
Garcea, R., and T. Benjamin. 1983. Host range transforming gene of polyomavirus plays a role in virus assembly. Proc. Natl. Acad. Sci. USA 80:3413-3417.
Garcea, R. L., D. A. Talmage, A. Harmatz, R. Freund, and T. L. Benjamin. 1989. Separation of host range from transformation functions of the hr-t gene of polyomavirus. Virology 168:312-319.
Genevaux, P., F. Lang, F. Schwager, J. V. Vartikar, K. Rundell, J. M. Pipas, C. Georgopoulos, and W. L. Kelley. 2003. Simian virus 40 T antigens and J domains: analysis of Hsp40 cochaperone functions in Escherichia coli. J. Virol. 77:10706-10713.
Genevaux, P., F. Schwager, C. Georgopoulos, and W. L. Kelley. 2002. Scanning mutagenesis identifies amino acid residues essential for the in vivo activity of the Escherichia coli DnaJ (Hsp40) J-domain. Genetics 162:1045-1053.
Glenn, G. M., and W. Eckhart. 1995. Amino-terminal regions of polyomavirus middle T antigen are required for interactions with protein phosphatase 2A. J. Virol. 69:3729-3736.
Glover, H. R., C. E. Brewster, and S. M. Dilworth. 1999. Association between src-kinases and the polyoma virus oncogene middle T-antigen requires PP2A and a specific sequence motif. Oncogene 18:4364-4370.
Graham, F. L., and A. J. van der Eb. 1973. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52:456-467.
Greene, M., K. Maskos, and S. Landry. 1998. Role of the J-domain in the cooperation of Hsp40 with Hsp70. Proc. Natl. Acad. Sci. USA 95:6108-6113.
Holman, P., O. Gjoerup, T. Davin, and B. Schaffhausen. 1994. Characterization of an immortalizing N-terminal domain of polyomavirus large T antigen. J. Virol. 68:668-673.
Kelley, W. L., and C. Georgopoulos. 1997. The T/t common exon of simian virus 40, JC, and BK polyomavirus T antigens can functionally replace the J.-domain of the Escherichia coli DnaJ molecular chaperone. Proc. Natl. Acad. Sci. USA 94:3679-3684.
Kim, H. Y., B. Y. Ahn, and Y. Cho. 2001. Structural basis for the inactivation of retinoblastoma tumor suppressor by SV40 large T antigen. EMBO J. 20:295-304.
Konieczny, I., and M. Zylicz. 1999. Role of bacterial chaperones in DNA replication. Genet. Eng. 21:95-111.
Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
Landry, S. J. 2003. Structure and energetics of an allele-specific genetic interaction between dnaJ and dnaK: correlation of nuclear magnetic resonance chemical shift perturbations in the J-domain of Hsp40/DnaJ with binding affinity for the ATPase domain of Hsp70/DnaK. Biochemistry 42:4926-4936.
Lemmon, S. K. 2001. Clathrin uncoating: auxilin comes to life. Curr. Biol. 11:R49-R52.
Love, T. M., R. de Jesus, J. A. Kean, Q. Sheng, A. Ledger, and B. Schaffhausen. 2005. Activation of CREB/ATF sites by polyoma large T antigen. J. Virol. 79:4180-4190.
Maione, R., G. M. Fimia, P. Holman, B. Schaffhausen, and P. Amati. 1994. Retinoblastoma antioncogene is involved in the inhibition of myogenesis by polyomavirus large T antigen. Cell Growth Differ. 5:231-237.
Martens, I., S. A. Nilsson, S. Linder, and G. Magnusson. 1989. Mutational analysis of polyomavirus small-T-antigen functions in productive infection and in transformation. J. Virol. 63:2126-2133.
Mullane, K. P., M. Ratnofsky, X. Cullere, and B. Schaffhausen. 1998. Signaling from polyomavirus middle T and small T defines different roles for protein phosphatase 2A. Mol. Cell. Biol. 18:7556-7565.
Ogris, E., I. Mudrak, E. Mak, D. Gibson, and D. C. Pallas. 1999. Catalytically inactive protein phosphatase 2A can bind to polyomavirus middle tumor antigen and support complex formation with pp60c–src. J. Virol. 73:7390-7398.
Pallas, D. C., C. Schley, M. Mahoney, E. Harlow, B. S. Schaffhausen, and T. M. Roberts. 1986. Polyomavirus small t antigen: overproduction in bacteria, purification, and utilization for monoclonal and polyclonal antibody production. J. Virol. 60:1075-1084.
Peden, K. W., and J. M. Pipas. 1992. Simian virus 40 mutants with amino-acid substitutions near the amino terminus of large T antigen. Virus Genes 6:107-118.
Qian, Y. Q., D. Patel, F. U. Hartl, and D. J. x. McColl. 1996. Nuclear magnetic resonance solution structure of the human Hsp40 (HDJ-1)J-domain. J. Mol. Biol. 260:224-235.
Rassoulzadegan, M., Z. Naghasfar, A. Cowie, A. Carr, M. Grisoni, R. Kamen, and F. Cuzin. 1983. Expression of the large T protein of polyoma virus promotes the establishment in culture of "normal" rodent fibroblast cell lines. Proc. Natl. Acad. Sci. USA 80:4354-4358.
Riley, M. I., W. Yoo, N. Y. Mda, and W. R. Folk. 1997. Tiny T antigen: an autonomous polyomavirus T antigen amino-terminal domain. J. Virol. 71:6068-6074.
Sanger, F., S. Nicklen, and A. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5468.
Sheng, Q., D. Denis, M. Ratnofsky, T. Roberts, J. DeCaprio, and B. Schaffhausen. 1997. The DnaJ domain of polyoma large T is required to regulate Rb family tumor suppressor function. J. Virol. 71:9410-9416.
Sheng, Q., T. M. Love, and B. Schaffhausen. 2000. J. domain-independent regulation of the Rb family by polyomavirus large T antigen. J. Virol. 74:5280-5290.
Silver, J., B. Schaffhausen, and T. Benjamin. 1978. Tumor antigens induced by nontransforming mutants of polyoma virus. Cell 15:485-496.
Srinivasan, A., A. J. McClellan, J. Vartikar, I. Marks, P. Cantalupo, Y. Li, P. Whyte, K. Rundell, J. L. Brodsky, and J. M. Pipas. 1997. The amino-terminal transforming region of simian virus 40 large T and small t antigens functions as a J domain. Mol. Cell. Biol. 17:4761-4773.
Stubdal, H., J. Zalvide, K. S. Campbell, C. Schweitzer, T. M. Roberts, and J. A. DeCaprio. 1997. Inactivation of pRB-related proteins p130 and p107 mediated by the J. domain of simian virus 40 large T antigen. Mol. Cell. Biol. 17:4979-4990.
Sullivan, C. S., P. Cantalupo, and J. M. Pipas. 2000. The molecular chaperone activity of simian virus 40 large T antigen is required to disrupt Rb-E2F family complexes by an ATP-dependent mechanism. Mol. Cell. Biol. 20:6233-6243.
Sullivan, C. S., and J. M. Pipas. 2002. T antigens of simian virus 40: molecular chaperones for viral replication and tumorigenesis. Microbiol. Mol. Biol. Rev. 66:179-202.
Szyperski, T., M. Pellecchia, D. Wall, C. Georgopoulos, and K. Wuthrich. 1994. NMR structure determination of the Escherichia coli DnaJ molecular chaperone: secondary structure and backbone fold of the N-terminal region (residues 2-108) containing the highly conserved J domain. Proc. Natl. Acad. Sci. USA 91:11343-11347.
Templeton, D., S. Simon, and W. Eckhart. 1986. Truncated forms of the polyomavirus middle T antigen can substitute for the small T antigen in lytic infection. J. Virol. 57:367-370.
Treisman, R., A. Cowie, J. Favaloro, P. Jat, and R. Kamen. 1981. The structures of spliced mRNAs encoding polyoma virus early region proteins. J. Mol. Appl. Gen. 1:83-92.
Yaglom, J., A. L. Goldberg, D. Finley, and M. Sherman. 1996. The molecular chaperone Ydj1 is required for p34cdc28-dependent phosphorylation of Cln 3 that signals its degradation. Mol. Cell. Biol. 16:3679-3684.
Zalvide, J., H. Stubdal, and J. DeCaprio. 1998. The J domain of SV40 large T antigen is required to functionally inactivate Rb family proteins. Mol. Cell. Biol. 18:1408-1415.(Kerry A. Whalen, Rowena d)