|Home | About | Journals | Submit | Contact Us | Français|
SUMOylation governs numerous cellular processes and is essential to most eukaryotic life. Despite increasing recognition of the importance of this process, an extremely limited number of SUMO protein ligases (E3s) have been identified. Here we show that at least some members of the functionally diverse tripartite motif (TRIM) superfamily are SUMO E3s. These TRIM proteins bind both the SUMO-conjugating enzyme Ubc9 and substrates and strongly enhance transfer of SUMOs from Ubc9 to these substrates. Among the substrates of TRIM SUMO E3s are the tumor suppressor p53 and its principal antagonist Mdm2. The E3 activity depends on the TRIM motif, suggesting it to be the first widespread SUMO E3 motif. Given the large number of TRIM proteins, our results may greatly expand the identified SUMO E3s. Furthermore, TRIM E3 activity may be an important contributor to SUMOylation specificity and the versatile functions of TRIM proteins.
Covalent conjugation to small ubiquitin-like modifier (SUMO) proteins is a common post-translational modification that alters various properties of the target proteins, including stability, activity, cellular localization, and protein-protein interactions (Geiss-Friedlander and Melchior, 2007; Hay, 2005; Johnson, 2004). Mammal cells expresses three SUMO proteins (SUMO1-3). SUMO2 and SUMO3 are nearly identical in their sequence and function and share about ~50% sequence identity with SUMO1. Similar to the ubiquitination pathway, the SUMO pathway consists of three enzymatic steps performed by the SUMO-activating enzyme E1 (a heterodimer of Aos1/SAE1 and Uba2/SAE2), the SUMO-conjugating enzyme E2 (Ubc9), and one of the SUMO ligases or E3s. SUMO E3s confer SUMOylation specificity and efficiency. However, in contrast to the over six hundred known ubiquitin E3s (Deshaies and Joazeiro, 2009), only a handful of SUMO E3s have been reported. These include protein inhibitor of activated STAT proteins (PIAS proteins) (Hochstrasser, 2001; Johnson and Gupta, 2001; Kahyo et al., 2001), the nuclear pore protein RanBP2 (Pichler et al., 2002), and the polycomb group protein Pc2 (Kagey et al., 2003). With the identification of numerous SUMOylation targets (Golebiowski et al., 2009), it remains unclear whether a large number of SUMO E3s also exist and thus how the specificity and efficiency of SUMOylation are achieved.
Tripartite motif-containing (TRIM) proteins form a superfamily in metazoans with ~20 members in C. elegans and ~65 members in humans and mice (Ozato et al., 2008). These proteins are defined by the TRIM/RBCC motif of a RING domain, one or two B-box domains (which, like the RING domain, are zinc-binding domains), and a predicted coiled-coil region. These domains are invariably arranged in the stated order and present at the N-terminal region of these proteins followed by a more variable C-terminal region. TRIM proteins play important roles in a wide range of processes including cell growth, tumor suppression, DNA damage signaling, senescence, apoptosis, stem cell differentiation, and immune responses against viral, and particularly HIV, infection (Dellaire and Bazett-Jones, 2004; Nisole et al., 2005; Ozato et al., 2008; Salomoni and Pandolfi, 2002; Schwamborn et al., 2009). For example, the promyelocytic leukemia protein (PML), also known as TRIM19, a prototypical TRIM protein, is involved in a chromosomal translocation associated with the vast majority of acute promyelocytic leukemia. PML is the eponymous and main structural component of the PML nuclear bodies (PML NBs), whose mechanism of action still remains elusive (Bernardi and Pandolfi, 2007; Borden, 2002). Likewise, TRIM27 (also known as Ret finger protein or RFP), acquires oncogenic activity when it is fused to the Ret receptor tyrosine kinase (Takahashi et al., 1985). Other examples include the potent anti-HIV infection mediated by rhesus monkey TRIM5α (Stremlau et al., 2004), suppression of stem cell differentiation by TRIM32 (Schwamborn et al., 2009), and the silencing of viral replication in stem cells by TRIM28 (Wolf and Goff, 2007). However, despite an increasing awareness of TRIM proteins’ important cellular functions, the biochemical basis for these functions is poorly understood. Certain TRIM proteins exhibit ubiquitin E3 activity, which is attributed to the RING domain (Meroni and Diez-Roux, 2005). The properties of the other motifs with the characteristic TRIM region remain unknown. In the current study, we show that TRIM proteins are a new class of SUMO E3s. Unlike the other SUMO E3s, the activity of TRIM proteins requires intact RING and B-box domains. Our results provide a biochemical framework for understanding SUMOylation and the function of TRIM proteins.
PML is heavily modified by SUMO, and overexpression of PML in yeast cells enhanced overall SUMOylation (Kamitani et al., 1998; Muller et al., 2000; Quimby et al., 2006). These observations led us to reason that PML may be a SUMO E3. We first examined whether PML stimulates substrate SUMOylation in mammalian cells. The tumor suppressor p53 is a known SUMOylation substrate (Rodriguez et al., 1999). We expressed p53 and the glutathione-S-transferase (GST) fusion of SUMO1 in Pml−/− MEF cells in the presence or absence of PML isoform IV (called PML in this study). This isoform was chosen because it can directly bind to p53 (Rodriguez et al., 1999). In the presence but not absence of PML, a slower migrating form of p53 was detected with a size expected for its conjugation to GST-SUMO1 (Figure 1a). In a similar experiment where Flag-tagged SUMO1 was used, p53 conjugation to Flag-SUMO1 was also strongly increased by the co-expressed PML (Figure 1a). Conjugation occurred mainly at a previously determined p53 SUMOylation site, Lys386 (Rodriguez et al., 1999), as change of this site to Arg abolished the SUMO conjugation (Figure 1a).
To confirm that PML stimulates SUMOylation p53, we expressed p53 and 6xHis-tagged SUMO1 (His-SUMO1), alone or together with PML, in the p53 deficient human lung cancer H1299 cells. The His-SUMO1-conjugated p53 protein was then captured by Ni-NTA beads (nickel-charged affinity resins) under denaturing conditions and analyzed by western blots. PML enhanced conjugation of p53 to His-SUMO1 in a dose-dependent manner (Figure 1b, top panel). The specificity of the Ni-NTA capture was verified by the lack of p53 binding to the beads when His-SUMO1 was replaced by Flag-SUMO1 (Figure 1b). Furthermore, a shorter PML isform (isoform VI, Figure 3a) that cannot interact with p53, failed to stimulate p53 SUMOylation (Figure 1c). In these experiments, PML was also conjugated to SUMO1, as expected (Figure 1a–d).
To examine the generality of PML’s ability to enhance SUMOylation in vivo, we tested two other SUMOylation substrates: the ubiquitin ligase Mdm2 and the transcriptional factor c-Jun (Muller et al., 2000), both of which interact with PML (Bernardi et al., 2004; Salomoni et al., 2005). PML strongly increased conjugation of both SUMO1 and SUMO2 to Mdm2 (Figures 1d and Supplemental Figure 1a). Likewise, PML promoted conjugation of SUMO1 to c-Jun (Supplemental Figure 1b–c). In contrast, PML did not enhance modification of IκBα, a SUMOylation substrate that resides mainly in the cytosol (Supplemental Figure 1c). Therefore, PML enhances SUMOylation of multiple nuclear proteins.
To determine whether PML itself possesses SUMO E3 activity, we performed in vitro SUMOylation assays using purified recombinant proteins. Flag-PML protein was expressed in 293T cells and His-p53 in bacteria, both of which were affinity-purified (Supplemental Figure 2a and b). When incubated in a SUMOylation reaction mix, the addition of Flag-PML strongly enhanced His-p53 modification by both SUMO1 and SUMO2 (Figure 2a). Flag-PML also enhanced Mdm2 modification by SUMO1 (Figure 2b). To rule out the possibility that the SUMO E3 activity was originated from trace amounts of contaminating protein from 293T cells, a GST fusion of PML was expressed in bacteria, which lack the SUMOylation system, and purified with glutathione beads (Supplemental Figure 2c). GST-PML strongly enhanced conjugation of p53 to SUMO1 and SUMO2. This activity depended on both SUMO E1 and Ubc9 (Figure 2c).
To investigate the structural determinant of PML’s SUMO E3 activity, we generated PML deletion mutants that contain either the N-terminal or the C-terminal region (Figure 3a). The N-terminal region, but not the C-terminal region, stimulated Mdm2 SUMOylation in vivo (Figure 3b). To assess the role of individual zinc-binding domains within the characteristic N-terminal TRIM region, we mutated two key zinc-chelating residues (Cys or His) within each domain to either Ser or Ala (Figure 3a). Mutations in each of these domains (Rm, B1m, and B2m), as well as two combined mutations (M4 and M6), reduced PML’s SUMO E3 activity towards p53 in vivo (Figure 3c). The purified mutants also showed reduced SUMO E3 activity in vitro towards both p53 and Daxx, another SUMOylation substrate (Figure 3d). Therefore, the RING and B-box domains are likely required for the E3 activity of PML.
To determine whether SUMO E3 activity is a shared feature of TRIM proteins, we screened a total of 14 other TRIM proteins from various subgroups that are distinguished by their C-terminal regions (Ozato et al., 2008). SUMOylation of Mdm2 was noticeably enhanced by half of these TRIM proteins representing all the major subgroups (Figure 4a and b, and Supplemental Figure 3a and b). Of note, a few TRIM family proteins -including TRIM27 and TRIM32 - exhibited much higher activity in stimulating Mdm2 SUMOylation compared to PML.
To verify the SUMO E3 activity of these TRIM proteins, we chose TRIM27 for further analysis since it strongly stimulated Mdm2 SUMOylation in vivo. An E3, in addition to promoting the transfer of ubiquitin or a ubiquitin-like protein from the E2 to the substrate, is characterized by its binding to both the E2 and substrates, either directly or indirectly (Hershko and Ciechanover, 1998; Hochstrasser, 2001). PML fulfills this binding requirement as it interacts with Ubc9 as well as with p53, Mdm2, and c-Jun (Duprez et al., 1999; Guo et al., 2000; Wei et al., 2003). To examine whether TRIM27 interacts with Ubc9, we expressed Flag-TRIM27 and HA-Ubc9 in 293T cells. A co-immunoprecipitation assay showed their specific interaction (Figure 5a). A similar assay also revealed the interaction of TRIM27 with Mdm2 and p53 (Figure 5b). The interaction of TRIM27 with p53, but not with Mdm2, was likely direct as shown by an in vitro pull-down assay with recombinant proteins (Figure 5c). Collectively, these results indicate that TRIM27 binds to both Ubc9 and substrates.
TRIM27 directly interacts with PML and partially resides in the PML nuclear bodies (Cao et al., 1998). To test whether PML influences the ability of TRIM27 to stimulate Mdm2 SUMOylation, we expressed TRIM27 and PML individually or in combination, together with Mdm2. TRIM27 and PML alone enhanced Mdm2 SUMOylation in Pml−/− MEFs, and their combination generated an additive effect (Figure 6a). Thus, TRIM27 acts on Mdm2 independently of PML. We also compared the E3 activities of TRIM27 and PML with that of a well-established SUMO E3, PIASy. The activities of TRIM27 and PML appeared to be weaker than that of PIASy (Figure 6a).
To confirm that TRIM27 functions as a SUMO E3 ligase toward Mdm2, we performed in vitro SUMOylation assays. TRIM27 purified from 293T cells (Supplemental Figure 2b) enhanced conjugation of both SUMO1 and SUMO2 to Mdm2 and p53 in an Ubc9-dependent manner (Figure 6b and c). Furthermore, GST-TRIM27 purified from bacteria, but not GST alone (Supplemental Figure 2c), enhanced conjugation of both SUMO1 and SUMO2 to p53. This activity relied on SUMO E1 and Ubc9 (Figure 6d).
SUMOylation enhances Mdm2 stability by inhibiting its ubiquitination (Lee et al., 2006). When co-expressed with Mdm2 in mammalian cells, the TRIM proteins that increased Mdm2 SUMOylation almost always elevated the steady-state levels of Mdm2 (Figure 4a and b, and Supplemental Figure 3a and b). A cycloheximide (CHX) chase experiment showed that the half-life of Mdm2 was markedly extended by TRIM27 (Figure 6e and f). Mdm2 was also stabilized by expression of exogenous SUMO1 alone and more drastically by the combined expression of SUMO1 and TRIM27 (Figure 6e and f). This effect of TRIM27 was diminished when Ubc9 was knocked down by siRNA (Figure 6g). These results suggest that TRIM27 stabilizes Mdm2 through SUMOylation.
The present study suggests that TRIM proteins are a new class of SUMO E3. This finding may expand the existing number of SUMO E3s many folds and offer a basis for understanding the regulation of SUMOylation. Furthermore, SUMO E3 activity may also explain the diversity of functional roles ascribed to TRIM proteins.
The SUMO E3 activity of TRIM proteins relies on the RING domain, extending the well-described ubiquitin E3 role of this domain. It also requires intact B1- and B2-boxes. These two zinc-binding domains fold into similar ternary structures to the RING domain, suggesting that all three domains share an evolutionarily ancestral motif. B-boxes almost always exist as part of the TRIM motif, and their biochemical activity had remained elusive. To our knowledge, the present study provides the first activity linked to these domains. The SUMO E3 motif of TRIM proteins is thus distinct from the other known SUMO E3 motifs including those present in RanBP2 and Pc2, while the E3 activity of the five mammalian PIAS proteins and their yeast homologues, Siz1 and Siz2, is associated with a different RING-like domain, SP-RING. However, like the other known E3s, TRIM proteins stimulate the conjugation of both SUMO1 and SUMO2/3 to targets. Selective attachment of different SUMO variants to substrates in vivo by TRIM proteins may be in part controlled by isopeptidases as suggested by a recent study for the SUMOylation of RanGAP1 (Zhu et al., 2009).
The SUMO E3 activity appears to be prevalent among TRIM proteins. A screen of 14 TRIM proteins from various subgroups on a single substrate (Mdm2) suggested that more than half of them have SUMO E3 activity. This is likely an underestimate of the actual number of SUMO E3s within the TRIM family as other TRIM proteins may have different substrates. A SUMO E3 is expected to bind to both Ubc9 and its substrates. PML binds to Ubc9 via its RING domain and to its substrates p53 and Mdm2 via its C-terminal region. Other TRIM proteins may act similarly. The consequence for SUMOylation varies among different target proteins. The effect of SUMOylation on p53 function is still a matter of debate (Hoeller et al., 2006). While Mdm2 SUMOylation has been proposed to enhance Mdm2 stability (Lee et al., 2006), it is also stimulated by the tumor suppressor Arf (Tago et al., 2005), which inhibits the function of Mdm2 (Sherr, 2006). The SUMO E3 for Mdm2 has not been identified, but it could be a TRIM protein. Identification of endogenous SUMO E3 for Mdm2 should help define the functional consequence of Mdm2 SUMOylation.
The TRIM family is notable not only because of its large size but also due to their roles in a wide range of processes, including those that protect against cancer, viral infection, and neurological disorders. We propose that the SUMO E3 activity is a unifying biochemical mechanism for these functions. For example, TRIM proteins are often identified as part of subcellular proteinaceous bodies (Reymond et al., 2001). The best-studied bodies, the PML NBs, contain a large number of stable and transient components, and the function of these transient components is often altered when they flux through the bodies. The function of these proteins may be altered either transiently or permanently by PML-mediated SUMOylation. To date, the function of the majority of TRIM proteins still remains elusive, but the identification of TRIM proteins as SUMO E3s should be instrumental in understanding these proteins.
Some TRIM proteins are also ubiquitin E3 ligases (Gillot et al., 2009; Meroni and Diez-Roux, 2005). Consistent with a recent report on the ubiquitin E3 activity of TRIM27 (Gillot et al., 2009; Meroni and Diez-Roux, 2005), we found that TRIM27 can also promote p53 ubiquitination (Supplemental Figure 3c). Thus, a single TRIM protein can have dual E3 activities. A continuum may exist among TRIM proteins in the relative strength of the ubiquitin and SUMO E3 activity; some TRIM proteins may modulate their target protein primarily through ubiquitination while others through SUMOylation. A TRIM protein may also attach ubiquitin and SUMO onto the same target simultaneously or sequentially to achieve versatile outcomes. It is now of great interest to determine the regulation of the TRIM proteins’ SUMO E3 activity, its physiological targets, and its dynamic interplay with the ubiquitin E3 activity.
Antibodies against PML (PG-M3, sc-966, H-238, and sc-5621), p53 (DO-1, sc-126), Ubc9 (sc-10759) and Mdm2 (SMP14, sc-965), and horse radish peroxidase (HRP)-conjugated anti-rabbit, anti-mouse, and anti-goat IgGs were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against p53 (Ab-6) and Mdm2 (Ab-1) were purchased from EMD Chemicals (Gibbstown, NJ), TRIM27 antibody (18791) from Immuno-Biological Lab (Minneapolis, MN), and Flag peptides and anti-Flag antibody (M2) from Sigma (St. Louis, MO). Glutathione Sepharose 4B (17-0756-01) (GE healthcare); SUMO E1, Ubc9, SUMO1, SUMO2 and ATP (Boston Biochem, Boston, MA); and MG132 (BML-PI102) (Enzo life science, Plymouth Meeting, PA) were purchased from the indicated sources.
Plasmids for expressing PML (isoform IV), TRIM27, TRIM32, TRIM5δ, Mdm2, c-Jun, IκBα, SUMO1, and SUMO2 in mammalian cells and/or in the in vitro translation system were constructed in pRK5 with an N-terminally fused Flag, HA, or 6xHis (His) tag, or glutathione-S-transferase (GST) as indicated. Other TRIM plasmids were kindly provided by Drs. V. Yu, A.-M. Herr, F. Rauscher, III, and T. C. Cox. TRIM25 was purchased from Addgene (Cambridge, MA). For expressing proteins in Escherichia coli, full-length PML and TRIM27 were fused to GST in pGEX-1ZT, a derivative of pGEX-1λT. p53 was fused to the 6xHis tag at the C-terminus and the Flag tag at the N-terminus in pET28a. Point mutations were made by overlap PCR. All plasmids generated for this study were confirmed by sequencing.
In vivo SUMOylation assays were performed as described previously with minor changes (Chen and Chen, 2003). Cells cultured in 6-cm plates were transfected with His-SUMO1 or His-SUMO2 expression plasmids, MDM2, GFP, and TRIM protein expression plasmids. 24 h after transfection, cells were treated with MG132 (20 nM) for 4 h. Cells from each plate were collected into two aliquots. One aliquot was lysed in lysis buffer and analyzed by western blot to examine the expression of transfected proteins. The second aliquot was lysed in buffer A (6 M guanidinium-HCl, 0.1 M Na2HPO4/NaH2PO4, 10 mM Tris-Cl pH 8.0, 5 mM imidazole, and 10 mM β-mercaptoethanol), and incubated with Ni2+-NTA beads (Qiagen) for 4 h at room temperature or overnight at 4 °C. The beads were washed sequentially with buffers A, B (8 M urea, 0.1 M Na2PO4/NaH2PO4, 10 mM Tris-HCl pH 8.0, 10 mM β-mercaptoethanol), and C (same as B except pH=6.3). Beads with bound proteins were then boiled in SDS sample buffer, the proteins were fractioned by SDS–PAGE and were analyzed by western blot for the presence of conjugated Mdm2, p53 or PML.
Flag-tagged PML, its mutants, and TRIM27 were expressed in Pml−/− MEFs or 293T cells through transient transfection and purified by anti-Flag M2 beads as described (Tang et al., 2006; Tang et al., 2004). Whole cell lysates were made in IP-lysis buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1% Triton X-100, 0.5% NP-40, 1 mM PMSF, 1 mM DTT, plus EDTA-free “complete” protease inhibitors). The lysates were immunoprecipitated with the anti-Flag M2 beads for 4 h to overnight. After extensive wash, the proteins were eluted from beads with 100 μg/ml 3xFlag peptide in elution buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 5 mM DTT, 25 μM ZnCl2, and 10% glycerol).
GST-fusion proteins and His-p53/pET28a were induced and purified in Escherichia coli BL21(DE3). Escherichia coli BL21(DE3) expressing the appropriate GST-fusion proteins was cultured in Terrific Broth (TB) medium with 50 μM ZnCl2. The proteins were induced for 4 h with 0.2 mM isopropyl-β-D-thiogalactoside (IPTG), followed by centrifugation, and re-suspended in lysis buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM PMSF, 5 mM DTT, 25 μM ZnCl2 plus EDTA- free “complete” protease inhibitors). Cellswere sonicated and then centrifuged for 10 min at 4 °C. Extracts were incubatedwith 30 μl glutathione Sepharose 4B beads (50% slurry) for 4 h at 4 °C. After extensive washing, the beads with the GST-fusion proteins were used for in vitro SUMOylation assay. BL21 (DE3) containing His-p53/pET28a was cultured in TB medium without ZnCl2. His-p53 was captured with TALON His-Tag purification resins (Clontech, Mountain View, CA). After extensive washing, His-p53 was eluted from the TALON beads by 1x PBS buffer containing 300 mM imidazole. His-p53 was then passed through a PD-10 desalting column (GE Healthcare) pre-equilibrated with a buffer containing 20 mM Tris-HCl, pH8.0, 150 mM NaCl, 0.1 mM EDTA, and 5% glycerol.
In vitro-translated proteins were made using TNT® Coupled Transcription/Translation System with S6 RNA polymerase (Promega).
In vitro SUMOylation reactions were performed at 37 °C for 1 h in 20 μl volume containing purified His-p53 (~100 ng) or in vitro translated HA-Mdm2 (4 μl), mammalian or bacterial cell-expressed PML or TRIM27 protein (5–10 ng), SAE1/SAE2 (125 nM), Ubc9 (50 nM), and His-SUMO1 or His-SUMO2 (32 μM). The reaction buffer contained 50 mM Tris-HCl pH 7.5, 2.5 mM Mg2+-ATP, and 2.5 mM DTT. Mdm2-containing reaction mixes were incubated with Ni2+-NTA beads. After wash in Urea buffer, the conjugated Mdm2 was detected by anti-Mdm2 antibody (SMP14, Santa Cruz). p53-containing mix was directly fractionated by SDS-PAGE (8%) and analyzed by western blot using anti-p53 antibody (DO-1).
Cells cultured on coverslips were fixed with 4% paraformaldehyde for 15 min at room temperature, permeabilized with 0.2% Triton X-100, blocked with 1% BSA and incubated with anti-TRIM27, anti-p53, anti-Mdm2, and anti-PML antibodies as indicated, followed by Texas-red conjugated anti-Rabbit IgG (Vector Laboratories, Burlingame, CA) and FITC-conjugated anti-mouse IgG antibody (Zymed Laboratories). The cells were mounted with DAPI-containing medium (Vector Laboratories, Burlingame, CA) and the images were acquired with a confocal microscope.
Whole cell lysates were made in lysis buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 0.5% Triton X-100, 0.5% NP-40, 1 mM PMSF, 1 mM EDTA, 1 mM DTT, plus EDTA-free “complete” protease inhibitors (Roche: 04693132001). The lysates were then immunoprecipitated with the anti-Flag M2 beads for 4 h to overnight. Beads were washed multiple times and boiled in SDS-containing loading buffer. Protein samples were resolved by SDS-PAGE and transferred onto PVDF membrane and probed with the indicated antibodies.
GST-pull down was performed as described previously (Townson et al., 2006) with some modifications. GST only or GST-fusion proteins were induced in Escherichia coli BL21. Cellswere lysed in lysis buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1% Triton X-100, 1mM EDTA, 0.1% SDS, 1 mM PMSF, plus complete protease cocktail) and sonicated. 400 μg of GST or 1 mg of GST-fusion protein bacterial crude extracts was incubated with 30 μl glutathione Sepharose 4B beads (50% slurry). Flag-TRIM27 was purified from 293T cells as described above. For the binding assay, the beads were incubated in lysis buffer and Flag-TRIM27 elute was added and incubated at 4 °C. The beads were washed three times with lysis buffer. Bound proteins were eluted in SDS sample buffer, resolved by SDS-PAGE, and analyzed by western blot.
HeLa cells transfected with appropriate vectors were cultured for 18 h. Cells were divided into four fractions and further cultured on 60-mm dishes overnight. The cells were treated with 50 μg/ml cycloheximide (CHX) and collected at different time points. Cells were lysed in RIPA buffer (50 mM Tris-HCl at pH 7.5, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, 0.1% SDS, 0.1% sodium deoxycholate, 1 mM PMSF, plus complete protease cocktail). Proteins were separated by SDS-PAGE and analyzed by Western blots.
Cells were seeded in 6-cm plates and transfected with 80 pmol siRNA per well using Lipofetamine 2000 according to the manufacturer’s instruction. Ubc9 siRNA was ordered from Thermo Scientific Dharmacon (Cat# M-004910-00). Control siRNA was ordered from Qiagen (Cat# 1027281).
In vitro ubiquitination reactions were performed at 37 °C for 2 h in 20 μl volume containing purified His-p53 (~100 ng) or bacterial-expressed TRIM27 protein (5–10 ng), E1 (156 nM), Ubc5a (50 nM), and Ubiquitin (147μM). The reaction buffer contained 50 mM Tris-HCl pH 7.5, 2.5 mM Mg2+-ATP, and 2.5 mM DTT. p53 ubiquitination was analyzed by Western blot using anti-p53 antibody (DO-1).
We thank Dr. P. P. Pandolfi for providing Pml−/− MEF cells and Drs. V. Yu, A.-M. Herr, F. Rauscher, III, and T. C. Cox for TRIM-expressing plasmids. We also thank E. Fischer and S. Slattery for technical assistance and A. Stonestrom for help with manuscript preparation. Supported by NIH (CA088868 and GM060911) to X.Y.
Conflict of interest
The authors declare no conflict of interest.