(E) Western blot assays were performed using Myc or HA antibodies in HEK293T cells transfected with HA-AR-V7 plasmids with or without Myc-SIAH2 plasmids for 48 h. 0.05 was considered statistically significant. Results Rutaecarpine selectively promotes the K48-linked ubiquitination accumulation and degradation of AR-V7 We first determined the expression of AR-FL and AR-V7 in various human PC cell lines and a prostatic stromal myofibroblast cell line WPMY-1. Our western blot results showed that AR-FL and AR-V7 were expressed in 22Rv1, LNCaP, and C4-2, but not in WPMY-1, PC3, and DU145 cells. Compared to LNCaP and C4-2 cell lines, the 22Rv1 cell line had the highest level of AR-V7 (Figure ?(Figure1A).1A). To identify a potential AR-V7 inhibitor, a natural product library containing 113 kinds of nature products (e.g., flavonoids, alkaloids, phenols, anthraquinones, quinones, and terpenes) was used to screen out the inhibitor of AR-V7 in 22Rv1 cells (Figure ?(Figure1B).1B). Among them, Rut, which is extracted from the dried fruit of 0.05. (H) Co-IP assay was performed using AR-V7 antibody or control IgG beads and immunoblotted for K48-Ub and AR-V7 in 22Rv1 treated with Rut for 12 h, and exposed to MG132(10 M) for 6 h before harvest. (I) Quantitative data of (H) are shown. (J) Immunoblot analysis of AR-V7 in 22Rv1 cells exposed to Rut with or without Bortezomib (BTZ) for 12 h. (K) Quantitative data of (J) are shown. Mean SD (n = 3). * 0.05, **P 0.01, #P 0.05 versus BTZ (-), BML-190 Rut (-), ###P 0.001 versus BTZ (-), Rut (-). In contrast, Rut failed to decrease the protein level of AR-FL, indicating a special role of Rut in the regulation of AR-V7 expression. The time-chasing experiments also confirmed that Rut suppressed AR-V7 protein expression after 12 h (Figure ?(Figure1D).1D). The immunofluorescence results further demonstrated that Rut reduced the overall expression of AR-V7 in the cell (Figure S1A-B). These results suggest that Rut selectively down-regulates AR-V7. We next determined whether Rut decreases the transcription of AR-V7 or promotes its degradation. Our Q-PCR results showed that Rut did not decrease the mRNA level of AR-V7 from 2.5 to 10 mol/L (Figure ?(Figure1E),1E), while the cycloheximide (CHX)-chasing experiments showed that Rut shortened the half-life of AR-V7 protein (Figure ?(Figure1F-G),1F-G), suggesting that Rut promotes AR-V7 degradation. Our Co-IP assay further confirmed that Rut increased the Lys(K)48-linked ubiquitination of AR-V7 (Figure ?(Figure1H-I).1H-I). Moreover, the 20S proteasome inhibitor, bortezomib, notably reversed the Rut-induced AR-V7 protein down-regulation (Figure ?(Figure1J-K),1J-K), suggesting that Rut induced a proteasome-mediated degradation of AR-V7. These results suggest that Rut selectively promotes the K48-linked ubiquitination and proteasome-mediated AR-V7 degradation. Previous studies have shown that Rut is an inhibitor of cyclooxygenase-2 (COX-2) 24, 25. To determine whether the Rut-induced K48-linked ubiquitination of AR-V7 was associated with its COX-2 inhibitory activity, we used parecoxib, another COX-2 inhibitor. Unlike Rut, treatment with parecoxib decreased the protein levels of both AR-V7 and AR-FL in 22Rv1, LNCaP, and C4-2 cells (Figure S2A). Additionally, parecoxib exhibited Rabbit Polyclonal to KCY a similar inhibitory effect on the cell proliferation among 22Rv1, LNCaP, and C4-2 cell lines, which had a notable difference in the protein level of AR-V7 (Figure S2B-C). Unlike Rut, parecoxib at lower concentrations failed to affect the expression of AR-V7 and AR-FL (Figure S2D). More importantly, the knockdown of COX-2 using siRNA did not affect the protein level of AR-V7 and cell viability of PC cell lines (Figure S2E-F). Together, these findings demonstrate that COX-2 inhibition is not required for Rut-induced K48-linked ubiquitination of AR-V7. Rutaecarpine BML-190 enhances the interaction between GRP78 and AR-V7 To explore the underlying molecular mechanism of Rut-induced BML-190 AR-V7 degradation, Co-IP combined with biomass spectrum assay was performed to identify the AR-V7 interacting proteins. The purified proteins from Co-IP using anti-AR-V7 antibodies were separated by SDS-PAGE, followed by silver staining (Figure ?(Figure2A).2A). Biomass spectrum analysis showed that AR-V7 interacted with several chaperones, including HSP7C (heat shock cognate 71 kDa protein), GRP78, HS90B (Hsp90-beta), HS90A (Hsp90-alpha) (Figure ?(Figure2B-C).2B-C). Indeed, molecular chaperones, such as HSP40, HSP70, and HSP90, are critical to the ubiquitin-mediated degradation of AR in certain PC cells and are proposed as anti-PC targets 26-28. We therefore wonder whether the chaperone machinery similarly controls the protein quality of AR-V7. Co-IP assay demonstrated that GRP78, but not HSP7C, strongly interacted with AR-V7 (Figure ?(Figure2D),2D), indicating that GRP78 may be the molecular chaperone.