USP16 is positively correlated with the c-Myc gene signature
To explore potential DUBs that may regulate c-Myc signalling, we performed differential expression analysis of control and c-Myc-overexpressing PCa cells (GSE51384), revealing 310 c-Myc mediated up-regulated genes (Fig. 1a and Additional file 2. Table S2). This set of genes was defined as a c-Myc gene signature and then imported into GSEA software for gene set enrichment analysis. Using this dataset, we screened for ubiquitin-specific proteases (USP) that were positively correlated with the c-Myc signature in four publicly available human PCa datasets (GSE62872, GSE79021, GSE134501, and GSE134160). Merging of these four independent analyses revealed a set of five USPs (USP16, USP22, USP28, USP38, and USP40) that consistently identified across all dataset analyses (Fig. 1b).
Next, we transfected shRNAs of the five USPs into PC3 cells, and the knockdown efficiency of shRNAs was measured by qRT-PCR (Fig. 1c). Analysis of c-Myc protein levels by Western blot revealed that knockdown of USP16 significantly decreased the abundance of c-Myc (Fig. 1d). Taken together, these data suggest that USP16 is strongly associated with the c-Myc signalling pathway and may play an important role in PCa. The positive correlation between USP16 and the c-Myc gene signature in PCa datasets was shown in Fig. 1e.
Targeting USP16 inhibits CRPC cells proliferation in vitro
To better understand the role of USP16 in PCa cells in vitro, we silenced USP16 in two CRPC cell lines: PC3 and DU145 cells, which characterized by androgen-independent growth. The knockdown efficiency of USP16 was confirmed by Western blot (Fig. 2a). Cells proliferation were then analysed using a CCK-8 assay, and the results revealed that knockdown of USP16 markedly reduced cell growth in PCa cells (Fig. 2a and b). Colony formation assays yielded the similar results that USP16 deletion results in significant reductions in cell colony numbers relative to controls (Fig. 2d and e). Concordantly, we found the ectopic expression of USP16 restored the proliferation rate of USP16 knockdown cells (Fig. 2f and g). Collectively, these results indicate that USP16 is necessary for the growth of CRPC cells.
USP16 knockdown suppresses growth of PCa tumour xenografts
To further elucidate the role of USP16 in the growth of PCa cells in vivo, PC3 cells stably expressing shRNA targeting USP16 (shUSP16) or vector control (shCON) were subcutaneously injected into 6-week-old male nude mice. After 8 weeks, the mice were sacrificed and the xenografts were extracted for further investigation. The control group xenografts(shCON) were larger and weighed significantly heavier than those in the USP16 knockdown group(shUSP16) (Fig. 3a and b). In addition, the inhibition of USP16 led to a delayed tumour onset in nude mice (Fig. 3c). IHC staining analysis of the xenograft tissues revealed that inhibiting USP16 reduced Ki67 expression, indicating USP16 knockdown impaired the proliferation of PCa cells (Fig. 3d–f). These results demonstrate that inhibiting USP16 significantly suppressed PCa cell growth in vivo.
USP16 stabilizes c-Myc in a deubiquitination activity-dependent manner
In previous assays, we observed the malignant effects of USP16 in PCa cells, we would like to further uncover the underlying mechanisms of how USP16 exert such effects. We found that knockdown of USP16 dramatically reduced c-Myc protein abundance but did not affect its mRNA levels (Fig. 4a and b), suggesting that the regulation of c-Myc by USP16 occurs at the post-transcriptional level. Moreover, treatment with proteasome inhibitor MG132 significantly attenuated the effect of USP16 knockdown on c-Myc protein level (Fig. 4c and d). Next, we examined whether USP16 could regulate the stability of c-Myc using a cycloheximide (CHX) chase assay. PC3 cells were treated with 50 μg/L CHX and c-Myc protein levels were measured at a series of indicated time points. We found that ectopic expression of USP16 enhanced the stability of c-Myc protein, while USP16 knockdown reduced the half-life of c-Myc protein (Fig. 4e and f). These data indicate that USP16 specifically sustains c-Myc stability through the ubiquitination-proteasome pathway.
USP16 deubiquitinates c-Myc
To explore the functional links between USP16 and c-Myc, we transfected Flag-tagged and V5-tagged plasmids into HEK293T cells, followed by immunoprecipitation with an anti-Flag antibody. Ectopically expressed USP16 was found to significantly interact with c-Myc and vice versa (Fig. 5a and b). Furthermore, USP16 protein was detected when Flag-c-Myc was immunoprecipitated by Flag antibody, and inversely c-Myc was detected when Flag-USP16 was immunoprecipitated in PC3 cells (Fig. 5c and d). The interaction between endogenous c-Myc and USP16 was also demonstrated in PC3 cells (Fig. 5e). Next, we confirmed the co-localization of USP16 and c-Myc in PC3 and DU145 cells using immunofluorescent staining (Fig. 5f). Together, these data indicate that USP16 both interacts and co-localizes with c-Myc.
To identify whether USP16 serves as a DUB of c-Myc, HEK293T cells were transfected with plasmids encoding HA-ubiquitin and Flag-c-Myc with wild-type USP16 or its catalytically inactive mutant USP16-C205S and treated with MG132. As shown in Fig. 5g, the wild-type USP16, but not USP16-C205S, markedly reduced the ubiquitination of c-Myc. Besides, the knockdown of USP16 significantly enhanced the polyubiquitination of c-Myc (Fig. 5h).
Next, we identified which polyubiquitin modification of c-Myc protein was regulated by USP16. HEK293T cells were transfected with v5-USP16 and Flag-c-Myc, along with one each of the different HA-ubiquitins (WT, K11, K48, or K63). Cell lysates were then immunoprecipitated with an anti-Flag antibody and subjected to immunoblotting analysis using an anti-HA antibody. The results revealed that the K48-linked ubiquitination of c-Myc was substantially reduced by USP16 (Fig. 5i).
USP16 regulates PCa cell growth through c-Myc
c-Myc is an oncoprotein involved in cell proliferation, and overexpression of c-Myc is known to enhance the viability of several cancer cells [24]. Given that USP16 regulates the stability of c-Myc, we examined whether USP16 regulates cell growth through c-Myc. We either disrupted or overexpressed c-Myc under conditions of USP16 knockdown. The protein levels of USP16 and c-Myc were measured using Western blot (Fig. 6a). Colony formation assays results suggest that c-Myc knockdown could abolish the effect of USP16 knockdown in terms of both cell proliferation and growth (Fig. 6b). Moreover, c-Myc overexpression restored the proliferation and colony formation abilities of USP16 silenced cells (Fig. 6c). These findings were further confirmed by CCK-8 assays (Fig. 6d and e).
We then performed RNA‐Seq analysis of PC3 cells with or without USP16 knockdown and the gene expression profiles were analysed by GSEA (Fig. 6f). Consistent with the previous screening results of published datasets, the Myc targets and Myc gene signature gene set were significantly enriched in the control group (Fig. 6g and h). Together, these data indicate that USP16 regulates PCa cell proliferation mainly throughs stabilizing c-Myc.
USP16 expression is elevated in prostate cancer
Next, we sought to confirm our results in clinical samples. We characterized USP16 expression in PCa (n = 70) and adjacent normal tissues (n = 70) via IHC staining. The staining scores of normal prostate tissues were markedly weaker than PCa tissues (Fig. 7a). Furthermore, we noticed that the USP16 staining scores were significantly correlated with the Gleason scores (χ2 test; p < 0.05; Fig. 7b and c), which indicates the essential role played by USP16 in PCa development.
To further assess the association between USP16 and c-Myc in PCa, we detected the expression of USP16 and c-Myc using tissue microarrays containing 82 human PCa tissues. Consequently, a positive correlation was found between the staining scores of USP16 and c-Myc (Fig. 7d-f). Thus, these results revealed the clinical relevance of USP16-mediated regulation of c-Myc in PCa development.