In contrast to SQLE HMGCR could be efficiently degraded
In contrast to SQLE, HMGCR could be efficiently degraded in Purvalanol B what lacking UBE2J2. However, this was not the case in cells devoid of UBE2G2, as these cells were unable to support 25-hydroxycholesterol (25-HC)-stimulated degradation of HMGCR (Fig. 1C). This finding is consistent with our recently reported genome-wide haploid mammalian genetic screen, which interrogated sterol-stimulated degradation of endogenous HMGCR and reported on the role of UBXD8 in this process in Hap1 cells . Additionally, this screen allowed us to investigate the E2 ubiquitin conjugating enzyme(s) required for this process (Supplementary Fig. 2A). A total of 34 E2s were identified in the screen, of which only two ERAD E2s were identified as being significantly involved in HMGCR degradation: UBE2G2 and, less prominently, UBE2J2 (Supplementary Fig. 2B). Consistent with UBE2G2 being the strongest hit in the haploid screen, we show here that cells lacking this E2 are unable to efficiently support sterol-stimulated degradation of HMGCR (Fig. 1C). Collectively, our results suggest that the degradation of SQLE and HMGCR depends on different E2 enzymes. While sterol-stimulated degradation of HMGCR in HEK293T cells requires UBE2G2, that of SQLE is predominantly mediated by UBE2J2. As the primary aim of this study was to characterize the ERAD machinery responsible for sterol-mediated SQLE degradation, and to further generalize the role of UBE2J2 herein, we wanted to extend this finding to other mammalian cell types. We therefore generated Hap1-UBE2J2 cells by targeting the endogenous UBE2J2 locus using a CRISPR/Cas9-based approach. The editing of the UBE2J2 locus was confirmed by sequencing, and by the reduction of UBE2J2 mRNA expression (Fig. 2A). Control Hap1 cells showed cholesterol-stimulated breakdown of SQLE, which could be attenuated by blocking the proteasome with the inhibitor MG132. In line with the results obtained in HEK293T cells, absence of UBE2J2 markedly prevented degradation of SQLE, but not that of HMGCR (Fig. 2B). Furthermore, RNAi-mediated silencing of UBE2J2 expression in HepG2 cells, a commonly used human hepatocyte-like cell model, also attenuated sterol-stimulated degradation of SQLE in a proteasome-dependent manner (Fig. 2C and D). Having established the critical necessity of UBE2J2 for degradation of SQLE, we considered the possibility that its expression may be sensitive to the cellular sterol-status, as this may allow coupling of its activity to metabolic demand. However, while expression of the SREBP targets HMGCR and SQLE increased and that of IDOL decreased by sterol-depletion as anticipated, expression of UBE2J2 itself or of MARCH6, UBE2J1 and UBE2G2 remained unchanged in HepG2 cells (Fig. 3A) and HEK293T cells. Moreover, ablation of UBE2J2 did not result in significant changes of UBE2J1 and UBE2G2 in these cells (Fig. 3B), or in Hap1 cells (Fig. 3C). In aggregate, these results suggest that SQLE degradation by MARCH6 and UBE2J2 is a selective post-transcriptional mechanism, which does not rely on sterol-dependent transcriptional regulation. Moreover, our results indicate that ablating UBE2J2 does not change the expression levels of the other ERAD E2s as a compensatory mechanism. To demonstrate that the catalytic activity of UBE2J2 is essential for its function in MARCH6-dependent SQLE degradation, we generated a catalytically inactive UBE2J2 mutant (UBE2J2C94S ) and tested its ability to rescue cholesterol-stimulated degradation of SQLE in UBE2J2 HEK293T cells, which otherwise are unable to support this process. Whereas reintroducing wildtype UBE2J2 largely restored sterol-mediated SQLE degradation in UBE2J2 cells, SQLE remained refractory to degradation in cells in which the mutant was reintroduced (Fig. 4A). In fact, reintroducing wildtype UBE2J2 decreased the basal levels of SQLE, while this was not the case for the catalytically inactive mutant (Fig. 4B). These findings indicate that the catalytic domain of UBE2J2 is essential for its function in MARCH6-dependent SQLE degradation. Collectively, our results support a model in which MARCH6 acts in concert with UBE2J2 in a metabolic feedback network to promote the cholesterol-stimulated and basal degradation of SQLE (Fig. 4C).