RanBPM is positively correlated with p21, and correlated with the survival of NSCLC patients. To explore the roles of RanBPM in the development of NSCLC, we performed immunohistochemical (IHC) staining using antibodies against RanBPM on a tumor tissue microarray consisting of 52 NSCLC patient specimens. NSCLC tumor specimens express much lower RanBPM levels than the normal tissues adjacent to the tumor (Fig. 1A-B). It was reported that RanBPM knockdown induced cell cycle arrest at S-phase 9. However, the detailed mechanism remains unclear. We analyzed the correlation of RanBPM (RANBP9) with cyclin-dependent kinases inhibitors (CKIs) by using Gene Expression Profiling Interactive Analysis (GEPIA) web server (Fig. 1D and Figure S1A). The expression levels of RanBPM is positively correlated with p21(CDKN1A) in NSCLC tumor samples (Fig. 1C). RanBPM knockdown specifically decreased p21 level (Figure S1B). We further validated the correlation of RanBPM with p21 protein levels in NSCLC tumor specimens by IHC analysis (Fig. 1D-E). These data indicated that the expression of RanBPM is positively correlated with p21 in the tumor specimens of NSCLC patients.
RanBPM protein interacts with p21. RanBPM, as a scaffolding protein, is a crucial component of multiple-protein complex that mediate diverse cellular functions by modulating and/or assembling with various kinds of proteins19. Given RanBPM is positively correlated with p21, we further tested whether RanBPM protein physically interacts with p21. RanBPM or p21 was separately immunoprecipitated from the lysates of A549 cells, and the protein of p21 or RanBPM was detected by western blotting. As shown in Fig. 2A and B, both RanBPM and p21 were detected in their individual immunoprecipitated complexes, but not in the isotype-matched negative control IgG complexes. We also detected the colocalization of RanBPM and p21 proteins in the nucleus (Fig. 2C). Furthermore, the plasmids encoding Flag-RanBPM or Myc-p21 were transfected into HEK293T cells, and RanBPM or p21 protein was co-immunoprecipitated (co-IP) with an anti-Flag or anti-Myc antibody. The exogenously expressed RanBPM was pulled down by the ectopically-overexpressed p21, while the exogenously expressed p21 was pulled down by the ectopically-overexpressed RanBPM as well (Fig. 2D-E). To further validate whether RanBPM physically interacts with p21 protein, we performed GST-pull down assay by using the purified recombinant His-RanBPM and GST-p21 proteins. The GST-p21, but not the GST control, was able to pull down the His-RanBPM protein under cell-free conditions (Fig. 2F), demonstrating a direct protein interaction between RanBPM and p21. Collectively, these results suggested that RanBPM physically interacts with p21 protein in vivo and in vitro.
RanBPM stabilizes and deubiquitinates p21 protein. Since RanBPM protein physically interacts with p21, we next investigated whether RanBPM affects the steady-state levels of p21 protein. As shown in Fig. 3A, knockdown of RanBPM with two independent RanBPM specific short interfering RNAs (siRNAs) significantly decreased p21 level in A549 and H1299 cells. Conversely, RanBPM overexpression led to the accumulation of endogenous p21 protein regardless of the p53 status (Figs. 3B). To determine whether RanBPM regulates p21 at the level of gene transcription, we measured the mRNA levels of p21 gene by qRT-PCR after RanBPM downregulation or overexpression. RanBPM did not significantly change the levels of p21 mRNA (Fig. 3C-D). The data suggested that RanBPM regulates the post-translational modifications, rather than gene transcription of p21 gene.
To further elucidate the mechanism by which RanBPM sustains the protein stability of p21, we monitored the protein degradation of p21 when the cells were treated with proteasome inhibitor MG132. In the absence of MG132, RanBPM overexpression or downregulation elevated or declined p21 protein levels in A549 or H1299 cells, while dysregulation of p21 caused by RanBPM could be blocked by the proteasome inhibitor MG132 (Fig. 3E-F). The data demonstrated that RanBPM regulates p21 protein by ubiquitin-proteasome pathway. Furthermore, when the protein biosynthesis was inhibited with cycloheximide (CHX), the knockdown of endogenous RanBPM decreased the half-life of the p21 protein (Fig. 3G and H), while RanBPM overexpression profoundly extended the half-life of the p21 protein (Fig. 3I). To further validate the underlying mechanism by which RanBPM regulates the stability of p21, we measured the levels of polyubiquitination of p21 protein by co-transfecting the plasmids encoding Myc-p21 and HA-Ubiquitin to HEK293T cells. RanBPM overexpression reduced the levels of polyubiquitylated p21 (Fig. 3J), whereas RanBPM knockdown significantly increased the levels of p21 polyubiquitylation (Fig. 3K).
RanBPM facilitates p21 deubiquitination in a USP11-dependent manner. It has well known that USP11 plays a key role in the maintenance of p21 protein stability17. Interestingly, USP11 recently was identified as a potential binding partner for RanBPM protein 20. We validated the protein interaction with sequential immunoprecipitation assays. The plasmids encoding Flag-RanBPM, HA-USP11 and Myc-p21 were co-transfected to HEK293T cells, and RanBPM protein was first immunoprecipitated with anti-Flag M2 agarose beads and eluted with Flag peptides. p21 proteins were secondly immunoprecipitated with anti-Myc antibody. The data showed that the USP11 was detected in both immunoprecipitates obtained with anti-Flag or anti-Myc antibodies (Fig. 4A). These data indicated that RanBPM, p21 and USP11 were present in the same protein complex.
We further presumed that RanBPM might promote interaction between USP11 and p21, and regulate USP11-dependent deubiquitination of p21. We tested the impact of RanBPM on the protein interaction of USP11 with p21. The ectopic overexpression of RanBPM promoted the protein interaction between USP11 and p21, whereas RanBPM knockdown with siRNAs significantly reduced the USP11-p21 interactions (Fig. 4B-C). Furthermore, sub-cellular localization of p21 and USP11 were further detected by immunofluorescent staining after RanBPM was knocked down in A549 cells. The knockdown of RanBPM did not seem to change the sub-cellular localization of p21 and USP11 (Figures S2).
Because RanBPM affected the interaction of USP11 with p21, we hypothesized that RanBPM is involved in USP11-mediated regulation of p21. To address this, Flag-USP11 was transfected into RanBPM-knockdown cells. As expected, the effect of USP11 on p21 markedly decreased after RanBPM knockdown (Fig. 4D). Ectopical overexpression of RanBPM dramatically promoted, whereas RanBPM knockdown decreased USP11-mediated p21 deubiquitination (Fig. 4E-F). These results suggest that RanBPM plays a vital role in USP11-mediated regulation of p21.
As well known, RanBPM is a scaffold protein without any enzymatic activity19, we deduced that RanBPM sustains p21 protein stability depending on the deubiquitinase activity of USP11. USP11 knockdown significantly reduced the protein levels of p21, which has been reported in a previous study17. We further clarified whether RanBPM facilitates p21 deubiquitination in a USP11-dependent manner. As shown in Fig. 4G and 4H, USP11 knockdown impaired RanBPM-sustained p21 protein stability and the deubiquitination level. These results indicated that RanBPM stabilizes and deubiquitinates p21 protein by promoting the binding of USP11 to p21.
RanBPM regulates DNA damage response in a p21-dependent manner. A549 and H1299 cells with RanBPM knockdown showed a lower ATM activation and defective homology-directed repair (HDR), and DNA damage induced more cell apoptosis 9,21. To demonstrate the crucial roles of RanBPM in DDR, we measured the cell proliferation and clonogenic formation in response to genotoxic stress using CCK-8 assays. The data indicated that RanBPM knockdown with siRNAs increased the sensitivity of A549 to genotoxic stress (Fig. 5A-C). The rescue of exogenous p21 in the RanBPM-depleted cells fully reversed the effect of RanBPM ablation (Fig. 5D-G). These data suggested that the RanBPM-mediated DDR is dependent on p21.
DNA damage promotes the translocation of RanBPM into the nucleus and regulates p21 protein stability. It was reported that p21 regulates DNA damage response (DDR) by p53-dependent or independent pathways22,23, Since RanBPM stabilizes and deubiquitinates p21 protein, we next investigated whether DNA damage elevates p21 protein levels through RanBPM-mediated pathways. In agreement with previous reports24, etoposide elevated p21 protein levels in A549 and H1299 cells (Fig. 6A and B), while etoposide-elevated p21 protein levels significantly decreased in the RanBPM-depleted cells (Fig. 6A). Furthermore, RanBPM knockdown significantly decreased the doxorubicin-triggered p21 elevation (Fig. 6B), but had no impact on the mRNA levels of p21 gene (Fig. 6C and D). These results suggested that RanBPM regulates DNA damage-elevated p21 protein.
Since p21 acts as a tumor suppressor in the nucleus, we hypothesized RanBPM might translocate into the nucleus to participate in DDR. We performed cell fractionation assays or immunofluorescence to confirm the hypothesis. As shown in Fig. 6E, DNA damage significantly elevated the amounts of RanBPM in the nucleus. In addition, DNA damage promoted the translocalization of RanBPM proteins to the nucleus (Fig. 6F).
Furthermore, we analyzed the physical protein interaction of RanBPM with p21. A549 cells were treated with etoposide or doxorubicin, and cell lysates were subjected to co-immunoprecipitation with anti-RanBPM or anti-p21 antibody. DNA damage significantly promoted the protein interaction between RanBPM and p21 (Fig. 6G and H). These data suggested that RanBPM is indispensable for the p21 protein stability in physiological conditions, or elevated protein levels in response to DNA damage (Fig. 6I).
DNA damage promotes the translocation of RanBPM and regulates p21 protein stability through ATM-mediated pathways. DDR induced posttranslational modifications of proteins such as phosphorylation, which are crucial for controlling protein stability, localization and activity25. The DDR signaling pathway orchestrated by the ATM and ATR kinases is the central regulator of this network in response to DNA damage26. ATM has been reported as a binding partner of RanBPM21. We speculated DNA damage induced p21 protein accumulation through ATM-mediated translocation of RanBPM. Firstly, etoposide or doxorubicin up-regulated p21 protein levels in A549 and H1299 cells. However, the elevated p21 protein was remarkably decreased by ATM inhibitor Ku55933 (Fig. 7A-B). We performed cell fractionation assays to monitor the translocation of RanBPM protein. As shown in Fig. 7C, DNA damage significantly increased the amounts of RanBPM proteins in the nucleus. Intriguingly, ATM inhibition by Ku55933 reversed DNA damage-induced nuclear translocation of RanBPM. The data were validated by immunofluorescence (Fig. 7D-E). These results suggested that DNA damage significantly promoted the nuclear translocation of RanBPM protein through ATM-dependent pathways.