E3 Ubiquitin Ligase RLIM Negatively Regulates c-Myc Transcriptional Activity and Restrains Cell Proliferation

RNF12/RLIM is a RING domain-containing E3 ubiquitin ligase whose function has only begun to be elucidated recently. Although RLIM was reported to play important roles in some biological processes such as imprinted X-chromosome inactivation and regulation of TGF-β pathway etc., other functions of RLIM are largely unknown. Here, we identified RLIM as a novel E3 ubiquitin ligase for c-Myc, one of the most frequently deregulated oncoproteins in human cancers. RLIM associates with c-Myc in vivo and in vitro independently of the E3 ligase activity of RLIM. Moreover, RLIM promotes the polyubiquitination of c-Myc protein independently of Ser62 and Thr58 phosphorylation of c-Myc. However, RLIM-mediated ubiquitination does not affect c-Myc stability. Instead, RLIM inhibits the transcriptional activity of c-Myc through which RLIM restrains cell proliferation. Our results suggest that RLIM may function as a tumor suppressor by controlling the activity of c-Myc oncoprotein.


Introduction
RLIM is a RING domain-containing E3 ubiquitin ligase first reported to play an important role in the chicken limb development by controlling CLIM abundance [1]. More recent research revealed its new functions. In Xenopus, Rlim maintains proper stoichiometry of Xlim-1/Ldb1 and confers proper function of the Spemann organizer [2]. It modulates telomere length homeostasis through proteolysis of TRF1 [3]. RLIM and CLIM interact with estrogen receptor α (ERα) and regulate its target gene expression [4]. RLIM was also identified as a component of the TGF-β superfamily signaling pathways [5,6]. It controls embryonic stem cell fate and morphogenesis in Zebrafish embryos by targeting the negative regulator Smad7 for proteasomal degradation [6]. Conditional knockout mouse model revealed that paternal Rnf12/RLIM is a critical survival factor for milk-producing alveolar cells [7]. The most exciting finding was that RLIM initiates imprinted X-chromosome inactivation (iXCI) by targeting REX1 for degradation [8,9]. However, it is dispensable for random form of XCI (rXCI) in mouse embryonic epiblast cells around implantation stage [10]. Our lab recently found that RLIM promotes cell migration in vitro by regulation of TGF-β pathway [11]. Moreover, we identified an interplay between p53 and RLIM: p53 represses the transcription of Rnf12 through interfering with the transcriptional activity of Sp1 [12]. On the other hand, RLIM enhances p53 stability and activity by targeting MDM2 for degradation [13]. However, other functions of RLIM are not well understood. Especially, the substrates for RLIM as an E3 ubiquitin ligase are poorly defined.
In this study, we identified c-Myc as a novel binding partner and substrate for RLIM. RLIM catalyzes non-degradation-associated polyubiquitination of c-Myc independently of c-Myc phosphorylation on Ser62 and Thr58. RLIM-mediated ubiquitination has no effect on c-Myc stability. Instead, it inhibits c-Myc transcriptional activity. Moreover, RLIM restricts cell growth by regulation of c-Myc. Our findings reveal a tumor suppressor role for RLIM which could potentially be exploited for cancer treatment.

Plasmid and siRNA transfection
The cells were seeded one day before transfection. Lipofectamine 2000 reagent (ThermoFisher Scientific) was used for both siRNA and plasmid transfection when the cells reach to about 80-90% confluency for plasmid transfection and 50-70% confluency for siRNA transfection. Medium was changed after 4-6 hours to complete medium with FBS and P/S. Cells were ready for use after 48 hours incubation. The target sequences for siRNAs were as follows. Scr1

Co-immunoprecipitation and Immunoblotting
Cells were lysed with 1 × cell lysis buffer (Cell Signaling Technology) and rotated at 4°C for 30 min. Cell debris was removed by centrifugation and the soluble fraction was collected and precleared with Protein A/G Agarose beads for 2 hours at 4°C. The precleared cell lysate was incubated with indicated antibodies overnight followed by incubation with Protein A/G beads for at least 2 hours at 4°C. Immunoprecipitates were then washed 6 times with cell lysis buffer and boiled in 1 × SDS loading buffer. Samples were resolved on SDS-PAGE and transferred to PVDF membranes and immunoblotting was carried out with antibodies as indicated.

GST pull down assay
The human RLIM gene was subcloned into pET28a (6×His)expression plasmid and the human c-MYC gene was subcloned into pGEX-4T-2 expression plasmid with GST tag (pGEX-4T-2 empty vector as a negative control). Recombinant proteins were induced to be expressed in Escherichia coli strain BL21 (DE3) by 0.1mM IPTG for 1-2 hours at 28°C (GST and His-tagged RLIM) or 25°C (GST tagged c-Myc). The bacteria were lysed by lysis buffer (1mM DTT, 1mM PMSF, 1% Triton in 1x PBS). GST or GST-c-Myc protein-containing cell lysate were incubated with equal amount of GST-Sepharose (GE Healthcare Life Sciences) for 4 hours at 4°C. After washing the beads 6 times with GST-pull down buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM DTT, and 0.01% Triton X-100), the supernatant was removed by centrifugation. Then equal amount of 6 × His-RLIM protein-containing E. coli cell lysate was added into the beads and incubate at 4°C for at least 1 hour. After 6 times washing with GST-pull down buffer, proteins bound to the beads were analyzed by SDS-PAGE followed by immunoblotting with anti-His antibody.

Construction of stable cell lines
U2OS cells were transfected with pcDNA3.1 or pcDNA3.1-HA-RLIM plasmid. Stably transfected cells were selected by G418 resistance (800μg/mL) for 2 weeks. Mixed clones were used for experiments.

Cell Proliferation Assay
The cells were plated in 6 replicates at a density of 1500-3000 cells per well in a 96 well plate. Cell proliferation was monitored by the Cell Counting Kit-8 (CCK-8) as instructed. The amount of formazan produced which was directly proportional to the number of living cells was measured by absorbance at 490 nm in a microplate reader. The measurement was taken every 24 hours for 3 or 4 consecutive days.

Dual-Luciferase Reporter Assay
293T cells were seeded in 4 replicates at a density of 6-8 × 10 4 cells per well in 24 well plates. 24 hours later, cells were transfected with myc-c-Myc plasmid and/or HA-RLIM plasmid together with an E-box reporter plasmid as well as a control reporter vector pRL-TK (Promega). After 30 hours incubation, the cells were harvested and lysed with 100ul 1 × PLB (passive lysis buffer, Promega) for 15min at room temperature with gentle shaking. Luciferase assays were performed by Luminometer LB9507. Firefly and Renilla luciferase activity was performed sequentially with a 10 second interval. Luciferase activity was analyzed based on the ratio of Firefly/ Renilla (co-transfecting pRL-TK vector as an internal control to normalize cell number and transfection efficiency) and the 4 replicates were averaged.

Statistical Analysis
Experiments were performed at least three times and representative results were shown. Statistical analysis was performed with a two-tailed unpaired t test or one-way ANOVA. p< 0.05 was considered statistically significant.

RLIM interacts with c-Myc in cells and in vitro
To study the interaction between RLIM and c-Myc, we transfected 293T cell with HA-RLIM and myc-c-Myc expression plasmids alone or together and performed a set of reciprocal co-IP-IB assays with myc and HA antibodies respectively. HA-RLIM and myc-c-Myc were specifically coimmunoprecipitated by myc and HA antibodies respectively only in cells with coexpression of both, but not in cells that express either one alone (Fig 1A). Consistent with these results, endogenous RLIM also bound to endogenous c-Myc (Fig 1B and 1C). These results indicate RLIM indeed associates with c-Myc. To access whether the two proteins interact with each other directly, we performed GST pull-down assay using bacteria-expressed His-RLIM and GST-c-Myc proteins. GST-c-Myc but not GST alone specifically pulled down His-RLIM protein (Fig 1D), which demonstrated the direct association of the two proteins. In addition, we constructed a catalytic dead RLIM expression plasmid by mutating an essential amino acid in the catalytic domain (C596A), and performed the co-IP assay. We found that RLIM C596A still interacted with c-Myc, suggesting that the E3 ligase activity of RLIM is dispensable for its interaction with c-Myc (Fig 1E). Together, these results demonstrate that RLIM is a bona fide binding partner of c-Myc.

RLIM promotes c-Myc ubiquitination
Since RLIM is an E3 ligase, we hypothesized that RLIM might serve to ubiquitinate c-Myc in vivo. We transfected 293T cells with Flag-ubiquitin (Ub), myc-c-Myc, HA-RLIM and HA-R-LIM C596A plasmids in different combinations, then immunoprecipitated myc-c-Myc and examined its ubiquitination status. We found that ubiquitination of c-Myc was significantly increased only by co-expression of WT RLIM but not catalytic-dead RLIM (C596A) (Fig 2A), suggesting that RLIM promotes ubiquitination of c-Myc. To further confirm this, we transfected 293T cells as above except using 6 × His-Ub construct instead of Flag-Ub. We affinitypurified His-tagged proteins under denaturing conditions and blotted with myc antibody. We found there was a significant increase of slow-moving bands only when WT RLIM was coexpressed ( Fig 2B). There was some increase of slow-moving bands in RLIM C596A co-expression group (Fig 2B), suggesting the E3 ligase activity of RLIM was not completely abolished by the point mutation. It was reported that phosphorylation of c-Myc on Ser62 and Thr58 regulates its ubiquitination by FBW7. This prompted us to examine whether phosphorylation of c-Myc on Ser62 and Thr58 also affects its ubiquitination by RLIM. We constructed a double point mutation of c-Myc (S62A and T58A) to abolish its phosphorylation on these two sites. Co-expression of WT or double mutant (DM) c-Myc (c-Myc DM ) with RLIM followed by ubiquitination assay showed that both WT c-Myc and c-Myc DM were ubiquitinated to similar extent by RLIM (Fig 2C), indicating that the ubiquitination of c-Myc by RLIM was not dependent on Ser62 and Thr58 phosphorylation. Taken together, these findings suggest that RLIM is a new E3 ligase for c-Myc.

RLIM-mediated ubiquitination of c-Myc is not degradative
One consequence of ubiquitination is the proteasome mediated protein degradation. To investigate whether RLIM regulates c-Myc protein level, we transfected 293T cells with myc-c-Myc and increasing amount of WT RLIM or RLIM C596A expressing plasmids. As expected, RLIM C596A did not change myc-c-Myc protein level (Fig 3A, lanes 4-8). Surprisingly, overexpression of WT RLIM did not significantly change myc-c-Myc level either (Fig 3A, lanes 1-4). Similar results were also observed in H1299 cells (Fig 3B). To examine whether endogenous RLIM regulates endogenous c-Myc protein level, we designed 4 different siRNAs to knockdown endogenous RLIM. Upon transfection into 293T cells, the four siRNAs effectively reduced RLIM protein level by more than 85% (Fig 3C). However, we did not observe significant change of endogenous c-Myc protein level (Fig 3C). Same experiment in H1299 cells confirmed the observation (Fig 3D).
We then hypothesized that RLIM may regulate the turnover of c-Myc protein.
To this end, we transfect 293T cells with RLIM expression plasmid or RLIM siRNAs, blocked new protein synthesis by cycloheximide (CHX) and harvested cells at different time points for WB. We found that RLIM was relatively stable with a half-life of about 60-120 minutes (Fig 3E and 3F). However, we did not observe significant changes of c-Myc turnover rate in either RLIM overexpression or knockdown conditions (Fig 3E and 3F), indicating that RLIM may not regulate c-Myc on the protein level.

RLIM negatively regulates c-Myc function
Regulation of protein activity is another function of ubiquitination. To examine whether RLIM regulates c-Myc function, we firstly employed a luciferase reporter system in which the expression of Firefly luciferase gene is controlled by c-Myc binding element (E-box). We transfected 293T cells with the reporter and a control construct (Renilla luciferase) in combination with HA-RLIM and myc-c-Myc plasmids alone or together. c-Myc overexpression increased Firefly luciferase activity by three folds (Fig 4A). Interestingly, co-expression of RLIM completely abolished c-Myc effects on Firefly luciferase activity (Fig 4A). This suggests that RLIM negatively regulates the transcriptional activity of c-Myc. We further examined whether RLIM regulates c-Myc target gene expression. We transfected U2OS cells with c-Myc expression vector alone or together with RLIM expression construct and examined the expression of two typical c-Myc target genes by real-time PCR. Consistent with the luciferase reporter assay results, c-Myc overexpression increased E2F2 and Nucleolin mRNA levels by two folds, while co-expression of RLIM completely abolished c-Myc effects (Fig 4B). To investigate the effects of RLIM on c-Myc activity in physiological conditions, we utilized two different siRNAs (as used in Fig  3C and 3D) to knockdown endogenous RLIM. Real-time PCR revealed a 1.5 to 2.5 fold increase of E2F2 and Nucleolin mRNA levels upon knockdown of RLIM (Fig 4C). Collectively, these data suggest that RLIM is a negative regulator for c-Myc.
As c-Myc is known to regulate cell proliferation, we hypothesized that RLIM may control cell proliferation by regulation of c-Myc. To this end, we constructed U2OS stable cell lines that overexpress either RLIM or empty vector and monitored their proliferation. We found that overexpression of RLIM significantly reduced cell growth rate when compared with control cell line (Fig 4D, upper panel. Control+si-Scr vs. RLIM+si-Scr). Importantly, concomitant knockdown of endogenous c-Myc almost completely abolished RLIM effects on cell growth (Fig 4D, upper panel. Control+si-c-Myc vs. RLIM+si-c-Myc). WB confirmed RLIM overexpression and knockdown efficiency of c-Myc (Fig 4D, lower panel). To further confirm this observation, we knocked down c-Myc and/or RLIM by siRNAs in U2OS cells and monitored their effects on cell growth. Indeed, depletion of RLIM significantly enhanced cell growth as compared to cells transfected with scramble control siRNA (Fig 4E, upper panel. si-Scr vs. si-RLIM). More importantly, concomitant ablation of c-Myc completely eliminated the effects of RLIM depletion on cell growth (Fig 4E, upper panel. si-c-Myc vs. si-double). WB confirmed the knockdown efficiency of RLIM and c-Myc (Fig 4E, lower panel). Previously, our lab found that RLIM regulates HCT116 cell colony formation in a p53-dependent manner [13]. To determine whether p53 was involved in the present experimental settings, we utilized the p53-defecient cell line, H1299, to repeat the proliferation assay. We depleted RLIM and/or c-Myc by siRNAs and monitored cell growth. Interestingly, we got similar results as in U2OS cells ( Fig  4F, upper panel). Knockdown efficiency is confirmed by WB (Fig 4F, lower panel).These results confirmed our hypothesis that RLIM restrains cell growth by negatively controlling c-Myc activity, at least in some cell lines.

Discussion
In this study, we revealed a novel function of RLIM as an E3 ubiquitin ligase for c-Myc. We found RLIM interacts with c-Myc in cells by reciprocal immunoprecipitation. The interaction was confirmed in vitro by GST-pulldown assay using purified proteins indicating a direct association of the two. We further showed that RLIM can promote polyubiquitination of c-Myc in cells. Compared to previously identified E3 ligases for c-Myc, RLIM showed some novel properties. It catalyzes polyubiquitination of c-Myc, but does not lead to c-Myc degradation. Instead, RLIM attenuates c-Myc transcriptional activity towards at least some of its target genes (E2F2 and Nucleolin). Moreover, RLIM restrains cell proliferation through regulation of c-Myc, as depletion of c-Myc abolished the effects of RLIM on cell growth. MYC is documented to be deregulated in up to 70% of human cancers [15]. Thus, our results provide evidence that RLIM could function as a tumor suppressor by controlling the activity of one of the most important oncoprotein. It will be of great importance to investigate whether RLIM is mutated or dysregulated in various type of cancers, especially in those without c-Myc dysregulations. Indeed, we found that RLIM expression is downregulated in certain percent of hepatocellular carcinoma patient samples (Pingzhao Zhang, unpublished data). This suggests that RLIM may serve as a potential drug target in treatment of certain type of cancers.
We found that expression of WT RLIM, but not catalytic dead mutant RLIM, increased c-Myc polyubiquitination in cells. However, we cannot exclude the possibility that RLIM promotes c-Myc polyubiquitination through regulation of other proteins. An in vitro ubiquitination assay using purified proteins will definitely help to clarify this. c-Myc exerts its transactivation function by heterodimerization with Max [34]. The heterodimer then GFP was used to monitor transfection efficiency. c-Myc and GFP levels were quantified using ImageJ software and the ratios of c-Myc to GFP are shown. (C) and (D) 293T (C) and H1299 (D) cells were transfected with scramble siRNAs or siRNAs against RLIM. Endogenous c-Myc protein was detected by anti-c-Myc antibody. c-Myc and actin levels were quantified using ImageJ software and the ratios of c-Myc to actin are shown. (E) and (F) 293T cells were transfected with HA-RLIM (E) plasmid or siRNA against endogenous RLIM (F). Cells were harvested at different time points after cycloheximide treatment and subjected to WB. Quantification of c-Myc protein level relative to actin are summarized from 3 independent experiments and shown in the right panels.  recognizes E-box elements within target gene promoters or enhancers and recruits various transcriptional cofactors to promote gene expression [35]. We found RLIM-mediated polyubiquitination of c-Myc attenuates its transactivation activity. It will be interesting to investigate whether ubiquitination of c-Myc blocks its recruitment to target gene promoters or which step in the transactivation process is affected. RLIM was proposed to mediate ubiquitination-dependent cofactor exchange on LIM homeodomain transcription factors [36]. Probably the same mechanism is also underlying its regulation of c-Myc activity. However, further experiments need to be carried out to prove this model. c-Myc regulates various biological processes including DNA replication, RNA transcription and elongation, cell growth and proliferation, metabolism, ribosomal biogenesis, genome stability etc. [14,15,34,[37][38][39][40][41][42][43][44][45][46][47]. Here, we provided evidence that RLIM participates in cell proliferation control through regulation of c-Myc. It will be intriguing to examine whether RLIM also participates in other biological processes regulated by c-Myc.
Recently, our lab revealed an interplay between RLIM and p53. It is very interesting that RLIM interacts with two of the most important cancer-relevant pathways. Further investigation needs to be done to fully elucidate the role that RLIM plays in coordinating the most important tumor suppressor, p53 and oncoprotein, c-Myc.