| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Molecular Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA; and
Second Affiliated Hospital, School of Medicine, Zhejiang University, Zhejiang, China
1Correspondence: Department of Microbiology and Immunology, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St., Baltimore, MD 21205, USA. E-mail: xfyu{at}jhsph.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: proteasomal degradation • substrate receptor • virology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The seven known human Cullin proteins (Cul1, 2, 3, 4a, 4b, 5, and 7) have diverse functions that range from cell cycle regulation to DNA repair and regulation of developmental processes (2)
. Cul1, Cul3, and Cul4 use distinct adaptor/substrate receptors to determine substrate specificity (2
, 3)
. Cul2 and Cul5 are interesting in that they both utilize ElonginB/C adaptor proteins, suggesting that substrate receptors mediate the selection of a particular Cullin. Cellular substrate receptors such as SOCS3 select Cul5 through a Cul5-box (4)
. HIV/SIV Vif molecules specifically assemble with Cul5 through a zinc-binding HCCH motif (5
6
7
8)
.
The adenovirus protein encoded by open reading frame 6 of early region 4 (E4orf6) is another viral protein that has recently been found to interact with Cul5-ElonginB-ElonginC (9
, 10)
. Adenovirus expresses genes from three different regions of the viral genome to modulate p53 function: the gene products of early region 1A (E1A), the 55 kDa product of the E1B region (E1B55K), and the 34 kDa E4orf6 (10
11
12
13
14)
. E4orf6, together with E1B55K, induces the polyubiquitination and degradation of p53 (11
, 12
, 15
16
17
18
19
20
21)
. Although its ability to induce p53 degradation is correlated with its interaction with Cul5-ElonginB-ElonginC (22)
, there has been no direct demonstration of the involvement of Cul5 in E4orf6-mediated p53 degradation. Also, the mechanism by which E4orf6 interacts with Cul5, ElonginB, and ElonginC to form an E3 ligase is poorly understood (9
, 10
, 23)
.
Here we demonstrate that E4orf6-mediated p53 degradation indeed requires Cul5. A dominant-negative mutant of Cul5 and RNA interference (RNAi) mediated Cul5 reduction both inhibited E4orf6-induced p53 degradation. We have also identified an HIV/SIV Vif-like BC-box motif in E4orf6 that is highly conserved among adenoviruses from multiple species. This Vif-like BC-box is essential for the recruitment of Cul5-ElonginB-ElonginC E3 ubiquitin ligase by E4orf6 and is required for E4orf6-mediated p53 degradation. E4orf6 selectively recruited Cul5 despite the lack of a Cul5-box, which is used by cellular substrate receptors to recruit Cul5, or of a newly identified HCCH zinc-binding motif, which is used by primate lentiviral Vif to recruit Cul5. Therefore, the adenovirus E4orf6 proteins represent a novel family of viral BC-box substrate receptors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. pCMVorf6
SLQ, pCMVorf6A136C, pCMVorf6A136S, pCMVorf6A136F, pCMVorf6Q133R, pCMVorf6L132S, pCMVorf6C51S, and pCMVorf6C126S were made by changing residues 131–133 from SLQ to AAA, 136A to C, S, and F, 133Q to R, 132L to S, 51C to S, and 126C to S in pCMV6.9, respectively. pSIVagmTanVif-Myc, Cul5-HA, and ElonginC-HA have been described previously (5
, 6
, 24)
. The human Cul5 mutant Cul5Nedd8 was described previously (24)
.
Transfection, antibodies, and cells
DNA transfection was carried out using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) as described by the manufacturer. The following antibodies were used for this study: anti-c-myc antibody-agarose conjugate (Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-HA mouse monoclonal antibody (mAb; Covance, Princeton, NJ, USA), anti-Cul5 mouse mAb (24)
, anti-c-myc mouse mAb (Sigma, St. Louis, MO, USA), anti-p53 mouse mAb (Oncogene Research Products, Cambridge, MA, USA), anti-ElonginB rabbit polyclonal antibody (Santa Cruz Biotechnology), anti-ElonginC mouse mAb (BD Biosciences, San Jose, CA, USA), and anti-human ribosomal P antigen antibody (Immunovision, Springdale, AR, USA). Anti-E1B55K was provided by Gary Ketner (25)
. 293T cells [AIDS Research Reagents Program, Division of AIDS, National Insititue of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH), Bethesda, MD, USA] were maintained in Dulbecco’s modified Eagle’s medium (Invitrogen) with 10% fetal bovine serum and gentamicin (5 µg/ml; D-10 medium) and passaged on confluence. p53 null SAOS-2 cells (ATCC, Rockville, MD, USA) were maintained in McCoy’s 5A medium (Invitrogen) modified with 1.5 mM L-glutamine, 15% fetal bovine serum, and 5 µg/ml gentamicin.
RNAi
The primers for making Cul5 siRNA were designed as described previously (26)
. Cells were collected 72 h after transfection.
Immunoblot analysis
Cells were collected 48 or 72 h after transfection. Cell lysates were prepared as described previously (27)
. Cells (1x105) were lysed in 1x loading buffer (0.08 M Tris, pH 6.8, with 2.0% SDS, 10% glycerol, 0.1 M dithiothreitol, and 0.2% bromphenol blue). The samples were boiled for 10 min, and proteins were separated by SDS-PAGE. Proteins were transferred onto nitrocellulose membranes by semidry transfer (Bio-Rad, Hercules, CA, USA). Membranes were probed with various primary antibodies against proteins of interest. Secondary antibodies were alkaline phosphatase-conjugated anti-human, anti-rabbit, anti-mouse, or anti-goat IgG (Jackson Immunoresearch, West Grove, PA, USA), and staining was carried out with 5-bromo-4-chloro-3indolyl phosphate and nitro blue tetrazolium solutions prepared from chemicals obtained from Sigma.
Immunoprecipitation
For immunoprecipitation experiments, the protein-expressing cells were harvested and washed twice with cold PBS and then lysed with PBS containing 0.5% Triton X-100 and protease inhibitor cocktail (Roche, Basel, Switzerland) at 4°C for 1 h. Cell lysates were clarified by centrifugation at 10,000 g for 30 min at 4°C. Anti-c-myc agarose (Santa Cruz) was mixed with the precleared cell lysates and incubated at 4°C for 3 h. The reaction mixture was then washed three times with cold PBS and eluted with 0.1 M glycine-HCl buffer, pH 2.0. The eluted materials were subsequently analyzed by immunoblotting. For immunoprecipitation in the presence of N,N,N',N'-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN; Sigma), cultural media were replaced with fresh D-10 medium containing 6 µM TPEN. Cells were treated for 16 h, and cell lysates were immunoprecipitated as described above.
In vivo p53 degradation assay
SAOS-2 cells were grown on a 6-well plate and transfected using 10 µl Lipofectamine and one of the following plasmids: 2 µg pCMV6.9 (wild type or mutant), 2 µg pCMV55K, or 1 µg pC53SN3. Total plasmid DNA was kept equal by adding pGFP as needed. All plasmid expression was controlled by the CMV promoter. At 48 h after transfection, cells were harvested by being washed with PBS and lysed in lysis buffer containing protease inhibitor. Lysates were analyzed by SDS-PAGE and immunoblotting. Equal protein loading was verified using anti-ribosomal protein, and equal E4orf6 and E1B55K expression was verified with anti-cmyc and anti-55K. Expression of p53 was assessed with anti-p53.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 10)
and, together with E1B55K, to induce the degradation of the cellular proteins p53 (11
, 12
, 15
16
17
18
19
20
21)
and Mre11 (28)
, the mechanism by which E4orf6 recruits the Cul5-ElonginB-ElonginC complex remains poorly defined (9
, 10)
. By analyzing E4orf6 sequences, we have identified a lentiviral Vif-like BC-box motif (5
, 29)
in human Ad5 E4orf6 (amino acids 131–140). This motif is highly conserved among adenovirus E4orf6 proteins from humans, monkeys, dogs, cattle, sheep, and ducks (Fig. 1
A). Like the divergent BC-box motif present in SIV Vif molecules (5)
, it lacks a downstream Cul5-box (Fig. 1A
). An SLQ to AAA mutation in this putative BC-box inhibited the interaction between mutant E4orf6 and Cul5-ElonginB-ElonginC: Cul5, ElonginB, and ElonginC were not efficiently coimmunoprecipitated with E4orf6SLQ (Fig. 1B
, lane 2) when compared to wild-type E4orf6 (lane 1). Similar results were also obtained in the reverse coimmunoprecipitation experiments: when compared to wild-type E4orf6 (Fig. 1C
, lane 1), a decreased interaction was also observed between mutant E4orf6SLQ (Fig. 1C
, lane 2) and Cul5 in experiments using anti-Cul5-HA.
|
The presence of Ala or Ser instead of a conserved Cys at position 6 in the E4orf6 BC-box represents a major deviation from the cellular BC-boxes (Fig. 2
A) and suggests that these three residues should be exchangeable at this position. Indeed, wild-type E4orf6, E4orf6A136C, and E4orf6A136S all interacted efficiently with Cul5-ElonginB-ElonginC (Fig. 2B
, lanes 5–7). Similar observations have been reported for the HIV-1 Vif BC-box mutants (6)
. However, a more bulky amino acid at this position (E4orf6A136F) disrupted the interaction (lane 3). Cys-to-Phe mutations have also been shown to disrupt cellular BC-box protein interactions with ElonginC (30
31
32)
as well as HIV-1 Vif interactions with ElonginC (6)
.
|
The highly conserved Leu (L132) at position 2 of the BC-box (33)
is critical for the interaction of SOCS1 (31)
, ElonginA (30)
, VHL (32)
, and HIV-1 Vif (24)
with ElonginC. Substitution of L132 with Ser also abolished the E4orf6L32S-ElonginC interaction (Fig. 2B
, lane 4). Residue 133 is often Q; however, a number of adenovirus isolates have an R at this position. To determine whether this conservation of this residue was required for E4orf6 function, we generated E4orf6Q133R. Immunoprecipitation of wild-type and E4orf6Q133R resulted in comparable interactions with Cul5, ElonginB, and ElonginC, suggesting that position 133 can tolerate a Q or R without affecting E3 ligase assembly (Fig. 2C
).
ElonginC interacts with the BC-box through a hydrophobic binding pocket (34)
. To confirm that the interaction between ElonginC and E4orf6 occurs through a BC-box motif, we constructed two ElonginC mutants, EloC2-HA and EloC4-HA, in which A100 and L103 or A100, L101, L103, and L104 were replaced with hydrophilic Ser. Although both the wild-type and mutant ElonginC constructs retained the ability to bind ElonginB (Fig. 2D
, lanes 2–4), only wild-type ElonginC interacted efficiently with E4orf6 (lane 2).
A functional BC-box in adenovirus E4orf6 is required for p53 degradation
To investigate whether the BC-box motif is required for E4orf6 function, p53 degradation induced by E4orf6 and its mutants was examined in p53-null SAOS-2 cells (Fig. 3
). SAOS-2 cells were transfected with expression vectors encoding p53, E1B55K, and E4orf6 or its mutants, and p53 stability was examined by immunoblotting. There was little p53 degradation in the absence of E1B55K (lane 8) and/or E4orf6 (Fig. 3A
, lane 7; Fig. 3B
, lanes 4 and 5). Efficient degradation of p53 was observed when both E4orf6 and E1B55K were present (lane 1). Degradation of p53 was inhibited by the BC-box mutations in E4orf6 that also disrupted the Cul5-ElonginB-ElonginC interaction, including E4orf6SLQ (lane 2), E4orf6A136F (lane 3), and E4orf6L132S (lane 4). Degradation of p53 was maintained by E4orf6A136C (lane 5), E4orf6A136S (lane 6), and E4orf6Q133R (Fig. 3B
, lane 2), which were competent for Cul5-ElonginB-ElonginC interaction. These results demonstrate the importance of a functional Vif-like BC-box and highlight the importance of the association of E4orf6 with Cul5-ElonginB-ElonginC for efficient p53 degradation.
|
The E4orf6-Cul5 complex does not require zinc
Like primate lentiviral Vif, E4orf6 lacks both a Cul2- and a Cul5-box, which have been demonstrated to be important for the recruitment of Cul2 and Cul5, respectively, by cellular substrate receptors (Fig. 4
C, D). We have recently demonstrated that a novel zinc-binding H-x5-C-x-17–18-C-x-3–5-H motif in primate lentiviral Vif molecules is critical for Cul5 selection (Fig. 4E
; refs 5
6
7
). Interestingly, E4orf6 has also been shown to be a zinc-binding protein (15)
. To determine whether zinc is also required for the interaction of E4orf6 with Cul5, we used the membrane-permeable zinc chelator TPEN (Fig. 4A
). HIV Vif, which contains a functional H-x5-C-x-17–18-C-x-3–5-H, zinc-binding motif, and the cellular substrate receptor SOCS-3 were used as controls (Fig. 4A
, lanes 4–9). 293T cells were transfected with E4orf6, HIV Vif, or SOCS-3 expression vectors and treated with increasing amounts of TPEN or control DMSO. TPEN treatment disrupted the association of HIV Vif with Cul5 but not ElonginB or ElonginC in a dose-dependent manner (Fig. 4A
, lanes 5 and 6). However, TPEN treatment did not affect the interaction of either E4orf6 or SOCS-3 with Cul5 (Fig. 4A
, lanes 2, 3, 8 and 9). Taken together, these data suggest that zinc chelation by 7 µM TPEN does not disrupt the E4orf6-Cul5 interaction (Fig. 4F
).
|
Cul5 is required for E4orf6 function
All Cullin family members are known to be modified by the ubiquitin-like small molecule Nedd8, which is critical for Cullin function (35
36
37
38)
. To determine whether the Cul5-containing complex is required for E4orf6 function, we first studied the effect of a Cul5 dominant-negative mutant (Cul5Nedd8) on E4orf6-induced degradation of p53 in 293T cells in which E1B55K is constitutively expressed. In these cells, p53 degradation occurred in the presence (Fig. 5
A, lane 2) but not the absence (Fig. 5A
, lane 1) of E4orf6. This E4orf6-induced degradation of p53 was significantly decreased in the presence of the Cul5 dominant-negative mutant Cul5Nedd8 (Fig. 5A
, lane 3). To further address whether Cul5-containing complexes are required for E4orf6 function, we silenced Cul5 expression through RNA interference. Cul5 expression was significantly reduced by siRNA oligos specific for Cul5 (Fig. 5B
, lane 1) but not by control siRNA (Fig. 5B
, lane 2). Knockdown of Cul5 blocked E4orf6-induced p53 degradation (Fig. 5B
, lane 1) when compared to the control (Fig. 5B
, lane 2). Collectively, these data demonstrate that Cul5 is required for E4orf6-induced degradation of p53.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 23
, 28
, 39)
. However, until very recently a direct role for Cul5 in E4orf6/E1B55K-mediated p53 degradation had not been demonstrated. In the present study, we have used two independent methods involving a Cul5 dominant-negative mutant and Cul5 siRNA to demonstrate that Cul5 is required for E4orf6/E1B55K-mediated degradation of p53. In the presence of E4orf6 and E1B55K, p53 was efficiently degraded; however, introduction of either a Cul5 mutant or siRNA resulted in the stabilization of p53, directly implicating Cul5 in adenovirus-mediated degradation of p53. While the present study was in review, another group also demonstrated that a Cul5 mutant blocked E4orf6/E1B55K-mediated p53 degradation (39)
.
E4orf6 selectively assembles with the Cul5-E3 ligase through a mechanism that is apparently distinct from cellular substrate receptors. Cellular substrate receptors such as SOCS3 use a Cul5-box downstream from the BC-box to select Cul5 (4)
. Even a subtle alteration in the Cul5-box, such as occurs in the case of SOCS1, has been shown to diminish Cul5 selection (4)
. Careful examination of the E4orf6 sequence did not reveal a Cul5-box like motif, suggesting that E4orf6 may recruit Cul5 via another mechanism. Primate lentiviruses use a highly conserved H-x5-C-x-17–18-C-x-3–5-H zinc-binding motif in Vif to recruit Cul5 (5
6
7
8)
. However, although E4orf6 is also a zinc-binding protein, a Vif-like H-x5-C-x-17–18-C-x-3–5-H motif was not found in E4orf6. We further observed that HIV/SIV Vif molecules could not efficiently assemble with Cul5 in the presence of the membrane-permeable zinc chelator TPEN, but the assembly of the Cul5-E3 ligase with E4orf6 and the cellular substrate receptor SOCS-3 was resistant to 7 µM TPEN. These data may be interpreted in two ways: First, E4orf6 does not interact with Cul5 via a zinc-binding motif such as a Vif H-x5-C-x-17–18-C-x-3–5-H motif. Second, E4orf6 may interact with Cul5 via a zinc-dependent mechanism that is not sensitive to the low concentration of TPEN used in this assay. Future experiments need to be conducted to determine which conclusion is most likely.
Although the mechanism by which E4orf6 selectively interacts with Cul5 is still unclear, we have shown that a functional BC-box is present in E4orf6. E4orf6 has a motif that closely resembles the viral BC-box identified in HIV/SIV Vif molecules. We have demonstrated that a functional interaction between E4orf6 and the Cul5 E3 ligase proceeds via this viral Vif-like BC-box, with the sequence SLQxxAxxxL spanning residues 130–140. Studies examining cellular BC-box proteins, HIV/SIV Vif BC-box proteins, and the E4orf6 Vif-like BC-box have identified a BC-box consensus sequence (A,P,S,T)1L2xxx(C,A,S)6xxx(A,I,L,V)10, in which position 1 is frequently A, P, S, or T; position 2 is invariably L; position 6 is a small amino acid, C, A, or S; and position 10 is a hydrophobic amino acid A, I, L, or V (4
, 6
, 26
, 29
, 31)
.
Although the Vif-like BC-box motifs in E4orf6 from diverse adenoviruses comply with these rules, the two BC-box motifs reported previously (23)
are significantly divergent from the rules. The invariable L at position 2 is not conserved in either of the two BC-boxes (Fig. 6
A). This L residue has been shown to be critical for the interaction of the BC-box-containing protein with ElonginC in the case of SOCS1, VHL, and ElonginA. Mutation of L145 to A in HIV/SIV Vif abolished interaction with ElonginC and prevented A3G degradation. Similarly, we have demonstrated that mutation of L132 to S in E4orf6 Vif-like BC-box also abolishes the interaction with Cul5, ElonginB, and ElonginC and inhibits p53 degradation. If this L is critical, it should be conserved. Our data demonstrate that it is conserved in the Vif-like BC-box of E4orf6 molecules; however, it is not conserved in either BC-box 1 or BC-box 2.
|
Position 1 of BC-box 1 or BC-box 2 was frequently not A, P, S, or T, and bulky amino acids H and N were found in position 6 of BC-box 1 (Fig. 6A
). Furthermore, position 6 of either BC-box 1 or BC-box 2 could not tolerate a change from C to S (Fig. 6B
), which can be tolerated by other well-characterized BC-box motifs (6
, 33)
. We have shown in both the HIV/SIV Vif and the E4orf6 Vif-like BC boxes that amino acids in the sixth position of the BC-box consensus have to be small, presumably for structural reasons. Ala, Cys, or Ser are all tolerated in Vif with regard to its function in interacting with Cul5-ElonginB-ElonginC (6)
. Vif function was disrupted when a large bulky amino acid such as F was introduced at this position (6)
. A similar situation has also been reported in individuals with VHL syndrome, in whom this position in the BC-box is commonly mutated from C to F, T, or W, all of which are large, bulky amino acids. Our results for the E4orf6 Vif-like BC-box are consistent with this trend (Fig. 3)
. On the other hand, the C residue in the same position in the previously reported BC-boxes in E4orf6 cannot tolerate a change to S (Fig. 6B, C
). This situation is consistent with the fact that these C residues are probably involved in maintaining the global structure of E4orf6.
Taken together, the data presented here further define the mechanism by which the adenovirus proteins E4orf6 and E1B55K induce the degradation of cellular p53. E4orf6 interacts with ElonginC via a conserved Vif-like viral BC-box to recruit Cul5 and degrade cellular p53. E4orf6 lacks both a Cul5 box and an HCCH zinc-binding motif, yet it specifically selects Cul5 (10)
, suggesting its use of an additional, possibly zinc-dependent, interface for Cul5 selection. Future studies to define this E4orf6 and/or E1B55K-Cul5 interface would aid in the identification of additional cellular substrate receptors and functions for Cul5.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 1, 2006. Accepted for publication January 18, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Thomas, R. S. Broughton, F. D. Goodrum, and D. A. Ornelles E4orf1 Limits the Oncolytic Potential of the E1B-55K Deletion Mutant Adenovirus J. Virol., March 15, 2009; 83(6): 2406 - 2416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Schwartz, S. S. Lakdawala, H. D. Eshleman, M. R. Russell, C. T. Carson, and M. D. Weitzman Distinct Requirements of Adenovirus E1b55K Protein for Degradation of Cellular Substrates J. Virol., September 15, 2008; 82(18): 9043 - 9055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Stanley, E. S. Ehrlich, L. Short, Y. Yu, Z. Xiao, X.-F. Yu, and Y. Xiong Structural Insight into the Human Immunodeficiency Virus Vif SOCS Box and Its Role in Human E3 Ubiquitin Ligase Assembly J. Virol., September 1, 2008; 82(17): 8656 - 8663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Mahrour, W. B. Redwine, L. Florens, S. K. Swanson, S. Martin-Brown, W. D. Bradford, K. Staehling-Hampton, M. P. Washburn, R. C. Conaway, and J. W. Conaway Characterization of Cullin-box Sequences That Direct Recruitment of Cul2-Rbx1 and Cul5-Rbx2 Modules to Elongin BC-based Ubiquitin Ligases J. Biol. Chem., March 21, 2008; 283(12): 8005 - 8013. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
