br Our data revealed that mechanistically HDAC inhibition
Our data revealed that mechanistically, HDAC6 inhibition reduces TNBC migration by increasing the acetylation of the molecular cha-perone Hsp90, which is crucial to the stability and function of nu-merous substrate proteins required to maintain cellular homeostasis and cell survival . The pharmacological inactivation or knockdown of HDAC6 was found to promote Hsp90 hyperacetylation and the loss of Hsp90 chaperone activity, thus accelerating the degradation of client proteins . The Hsp90 substrates aurora-A and cortactin are also known to play important roles in cell motility. An elegant study per-formed by Wang et al.  demonstrated that the overexpression of aurora-A increased the level of the active form of cofilin by enhancing SSH1 expression. Another research group found that HDAC inhibitor treatment downregulated aurora-A expression in cancer cells . Cofilin activation generates barbed ends on free Latrunculin A filaments, in-itiates actin polymerization, induces localized cell membrane protru-sion, and directs cell migration, which in turn promote breast cancer cell metastasis . Cofilin activity is tightly regulated by phosphor-ylation. The serine/threonine kinases LIMK and TESK render cofilin inactive by phosphorylating the Ser3 residue , whereas SSH1 de-phosphorylates and thus activates cofilin, which in turn stimulates the severing and depolymerization of actin [25,35]. Consistent with these results, we found that MPT0G211 treatment led to significant Hsp90 acetylation, resulting in the dissociation of the Hsp90/aurora-A com-plex and the proteosomal degradation of aurora-A (Fig. 3), as well as the downregulation of SSH1, phosphorylation of cofilin (Fig. 4), and the subsequent inhibition of actin polymerization (Fig. 5).
Regarding cortactin, a cytoplasmic protein involved in the MPT0G211-mediated inhibition of breast cancer motility, studies of primary metastatic breast carcinomas have observed amplification of the CTTN gene, which encodes cortactin, and the consequent
overexpression of cortactin . Other studies have linked the over-expression of cortactin with invasiveness in various cancers, including breast cancer, and suggested that cortactin may be an important bio-marker of invasiveness [17,27,37]. Cortactin can be detected at the leading edges of migrating cells in areas of dynamic actin assembly, where it acts as a nucleation promoting factor (NPF) that binds and activates the Arp2/3 complex responsible for the polymerization and branching of F-actin [27,37]. Zhang et al. identified cortactin as a substrate of HDAC6 and specified that HDAC6 modulates the actin binding ability of cortactin via multiple acetylated lysine residues in the repeat region of the latter . Therefore, the inhibition of HDAC6 leads to the hyperacetylation of cortactin, which is thus prevented from translocating to the membrane ruﬄe in the cell periphery and asso-ciating with F-actin, thus impairing cell motility. Our results, which were consistent with those of previous studies, clearly show that MPT0G211 treatment significantly increases cortactin acetylation, leading to reduced cortactin/F-actin binding (Fig. 5C) and inhibited breast cancer cell migration.
Paclitaxel has been used as frontline chemotherapeutic agent for breast cancer; however, recent studies have pointed that paclitaxel exacerbate breast cancer metastasize to lung, although the underlying mechanism are still unclear [38,39]. In our study, we combined MPT0G211 with paclitaxel in cell migration assay, it showed that the combination therapy significantly inhibited breast cancer migration (Fig. 2A). Furthermore, in mouse metastasis model, the combination of MPT0G211 with paclitaxel caused more significant reductions in the number of tumor nodules (Fig. 6A).
Taken together, this study demonstrates that MPT0G211, a novel and potent HDAC6 inhibitor, suppresses TNBC metastasis by inhibiting HDAC6 activity and not only increasing Hsp90 acetylation to promote aurora-A degradation and inhibit the cofilin–F-actin pathway, but also by inhibiting cortactin/F-actin binding and thus impairing cell motility. These results suggest that MPT0G211 may be a useful option for cancer treatment.
Conflict of interest statement
The authors declare that they have no conflicts of interest.
The Transparency document associated this article can be found, in online version.
Appendix A. Supplementary data
 C.L. Arteaga, M.X. Sliwkowski, C.K. Osborne, E.A. Perez, F. Puglisi, L. Gianni, Treatment of HER2-positive breast cancer: current status and future perspectives,