Discussions

The roles of the proteasome in cancer have been intensively studied. Recently, several studies revealed the critical role of UPS components, the lid subunits (19S) or the 20S subunits, in several types of cancers[27-30,42,43]. In this study, for the first time, we investigated the PSMD3 expression level in BC using BC cell lines and paired human tissue samples from BC patients. In addition, we performed several functional analyses to analyze the PSMD3-HER2 protein-protein interaction. PSMD3 was upregulated in BC cell lines and tumor tissues compared to normal breast cells and tissues. The sixth HER2+ cell lines, HER2+ tumor tissues, and patients with HER2+ from the Oncomine and TCGA databases showed significantly higher levels of PSMD3 compared to normal breast tissue or to other BC subtypes. Interestingly, strong protein-protein interaction between PSMD3-HER2 was observed by IP and FRET assays.

HER2 is located at chr17q12 and in agreement with a previous study on the chr17q copy number (CN) patterns for HER2 and HER2-related genes. PSMD3 locate at ch17q21 and considered as one of the close genes surrounding HER2 that exhibited high (CN) in parallel with HER2[35]. Due to the positive correlation between HER2 and PSMD3 in BC, loss of PSMD3 function analyses in several HER2+ cells was performed. Indeed, silencing of PSMD3 led to destabilize of total HER2 (185 kDa), and HER2 expression was significantly reduced by more than 60% compared to the controls by using two different PSMD3-siRNA.

The main role of the proteasome is to control cell death and apoptosis[44]. Several studies have revealed that inhibiting proteasomal subunits leads to the induction of apoptotic cell death in cancer cells. PSMD4 has been shown to correlate with PARP to induce cell apoptosis[43]. Silencing of PSMD2 regulated cell cycle and apoptosis by modulating p21 and p27 in BC [27]. Our results revealed that downregulation of HER2 by silencing of PSMD3 not only provides a significant inhibition of cell proliferation but also results in activation of cellular apoptosis (measured by Annexin V and activated PARP and caspase-3) in HER2+ cells through inhibiting the main HER2 signaling pathways (ERK/AKT).

In the present study, we showed that transfecting BT-474 and HCC1419 cells with the ubiquitin plasmid had no effect on the ubiquitination process with normal level of HER2. Interestingly, cotransfection of BT474 and HCC1419 with ubiquitin plasmid and PSMD3 Si led to strongly enhanced the ubiquitination process and decreasing the total level of HER2. We treated HER2+ BC cells with MG132, a proteasome inhibitor, with or without silencing PSMD3 in the presence or absence of ubiquitin plasmid. We found that MG132 with co-transfection the cells with ubiquitin and PSMD3 Si plasmids had an effect in enhancing HER2 degradation by non-proteasome process. Collectively, we uncovered the role of PSMD3 in stabilizing HER2.

Cell receptor targeting usually sensitizes the receptors for internalization by a process called receptor-mediated endocytosis (RME) with an association with lysosomal degradation [45]. A number of reports have been published on the degradation of HER2 by a lysosomal-dependent process[46,47]. A previous study demonstrated the use of an anti-HER2 aptamer grafted onto nanostars that mediated HER2 endocytosis and lysosomal degradation[48]. This finding prompted us to investigate whether silencing PSMD3 enhance HER2 degradation by lysosomal pathway. Interestingly, we found that LAMP-1 a lysosomal marker was accumulated at the cell membrane with decrease HER2 immunostaining compared with the preserved form of HER2 at the cell membrane in the control group, suggesting that HER2 is degraded by the lysosomal pathway.

DUBs are considered an important part of the UPS system, and the fundamental role of DUBs is, specifically, to disassemble ubiquitin from target proteins. DUBs contribute to the regulation of various cellular processes, such as preserving proteins from degradations [49], regulating UPS or lysosomal-dependent protein degradation [50] and cell cycle or apoptosis[51]. Among the DUBs that have been identified in the human genome, USP14 has been implicated in several types of cancers, including brain[52], ovarian [53] and liver[54] cancers. A recent study revealed the enhancement of ubiquitination and the degradation of the androgen receptor (AR) in prostate cancer as a result of inhibiting USP14 [55,56]. PSMD3 is not considered a DUB enzyme and does not belong to the ATPase subunits. Interestingly, we found that the protein level of USP14 decreased after knockdown of PSMD3 in HER2+ cells and thus could inhibit trimming of the ubiquitin from HER2. According to Xu et al, phosphorylated and activated USP14 is regulated by the AKT pathway [57]. We demonstrated that inhibiting of AKT pathway did not affect the PSMD3 or USP14 protein level. However, knockdown of PSMD3 led to decrease USP14 protein level.

An overexpression assay was used to overexpress PSMD3 in BC cells. Unexpectedly, the results indicated that PSMD3 has no any oncogenic role in enhancing HER protein level, which indicating that PSMD3 is a proteasome dependent protein.

To the best of our knowledge, we show here for the first time the clinicopathological correlation of PSMD3 with BC using FFPE and TMA analyses and using Oncomine and TCGA databases. Patients with high PSMD3 expression had worse OS, RFS and PFS for BC and OS and DMFS for HER2+BC suggesting the use of PSMD3 as an unfavorable prognostic factor for BC patients. Our results also revealed that patients who were HER2+ had higher PSMD3 levels and shorter OS. Similar results were obtained by univariate and multivariate Cox regression analyses. Unexpectedly, we did not find any significant correlation between low vs high PSMD3 in either the clinical stage, grade, tumor, or node status when we used protein H-score results or between BC molecular subtypes because of the low case numbers that were included in this study. However, according to the TCGA database, high PSMD3 levels have significant differences with tumor size, stage of the disease and, importantly, HER2 status. To our knowledge, this is a novel finding that would be worth validating by using a large series of BC patients based on PSMD3 protein expression and further confirming the correlation between PSMD3 levels and clinicopathological parameters of BC.