Previous reports have demonstrated that SB and SB-derived polyphenols reduce the growth of several cancer cell lines in vitro
], but their anti-proliferative effect in PaCa cells is less explored. The results of this study showed that SB extract and all SB-derived polyphenols reduced the proliferation of PaCa cell lines in a dose-dependent manner. While baicalin (10% of raw material, ref. 4
) had only a minimal effect on PaCa growth, its aglycone baicalein (5% of raw material) had the most potent growth inhibitory effect, consistent with a previous report in multiple myeloma cell lines [16
]. In general, the growth-inhibitory effect was more clearly seen in the BrdU assay, which detects changes in DNA synthesis, compared to the MTT assay, which measures cell viability. Although baicalein was most effective in Kras wildtype BxPC-3 cells, it is important to note that it also inhibited growth of Kras mutated pancreatic cancer cells at concentrations that can be achieved in vivo in mice fed a diet supplemented with 1% SB extract for 8 weeks (manuscript submitted). Furthermore, Bonham et al. reported that most of the activity seen with the SB extract could be attributed to baicalein in prostate cancer cells [14
]. Based on these reports and our study, we decided to investigate the mechanisms of baicalein-induced growth inhibition in PaCa cells.
Anti-apoptotic Bcl-2 family proteins play an important role in response to a variety of death stimuli. Mcl-1 is one of the anti-apoptotic Bcl-2 family proteins, and its over-expression has been reported in a variety of hematopoietic, lymphoid, and some solid tumors including PaCa [6
]. In these cancers, Mcl-1 plays a key role in the resistance to conventional chemotherapy. Importantly, some papers showed that Bcl-2 or Bcl-xL knock-down by siRNA is not enough to promote apoptosis, but Mcl-1 knock-down alone can clearly induce apoptosis in melanoma and chronic lymphocytic leukemia cells [30
]. From these reports, Mcl-1 is considered to be a promising therapeutic target in several cancers, but the importance of Mcl-1 in PaCa is unknown. In this context, we first confirmed the expression of anti-apoptotic Bcl-2 family proteins in six PaCa cell lines ranging from undifferentiated to well differentiated cell lines. Mcl-1 was expressed in five PaCa cell lines, but Bcl-2 was detected in only MIA PaCa-2 and Panc-1 cells. Miyamoto et al. reported that Bcl-2 protein was strongly expressed in MIA PaCa-2 and Panc-1 cells [6
]. Others showed that Bcl-2 protein was strongly expressed in BxPC-3 cells, but weakly expressed in Panc-1 cells [32
]. The exact reasons for this discrepancy are unclear, but different culture conditions, cell passages, and different antibodies may be potential causes. We carefully explored which anti-apoptotic Bcl-2 family proteins play a critical role on PaCa apoptosis by siRNA-mediated knock-down of Bcl-2, Bcl-xL, and Mcl-1. We confirmed a marked induction of apoptosis following knock-down of Mcl-1, whereas Bcl-2 and Bcl-xL knock-down resulted only in minimal induction of apoptosis. However, double knock-down of Mcl-1 and Bcl-xL in MIA PaCa-2 cells had a significantly more robust effect on apoptosis than Mcl-1 knock-down alone. Among the anti-apoptotic Bcl-2 family proteins, Mcl-1 seems to play the most critical pro-survival role in pancreatic cancer cells (as evident by the single knock-down of each protein). However, once Mcl-1 is knocked-down, additional reduction of other anti-apoptotic Bcl-2 family members, such as Bcl-xL, greatly elevates the induction of cell death. To our knowledge, this is the first report comparing the effect of three anti-apoptotic Bcl-2 family proteins in PaCa.
In this study, we demonstrated that baicalein treatment significantly induced apoptosis through a caspase-dependent mechanism. We observed a close correlation between baicalein-induced apoptosis and decrease in Mcl-1 expression, and over-expression of Mcl-1 confers resistance against baicalein treatment. It is of note that baicalein did not change expression of Bcl-xL proteins in any pancreatic cancer cell line. In addition, genetic knock-down of Mcl-1 by siRNA significantly abrogated the apoptosis-inducing effect of baicalein, suggesting that the pro-apoptotic effect of baicalein was mediated through reduction of Mcl-1. However, baicalein still had an additional (albeit significantly less robust) effect on apoptosis in cells, in which Mcl-1 was knocked-down. The reasons for this phenomenon may include the lack of complete knock-down of Mcl-1 by siRNA and consequently an additional effect of baicalein on the remaining Mcl-1 expression, or alternative pathways (independent of Mcl-1) of apoptosis induction by baicalein.
BH3-only proteins are classified in activator BH3-only protein and sensitizer BH3-only protein [3
]. Activator BH3-only proteins can directly activate pro-apoptotic multi-domain molecules Bak/Bax, but the effect of sensitizer BH3-only protein is indirect; sensitizer BH3-only proteins lower the threshold of apoptosis by occupying the binding pocket of anti-apoptotic proteins, such as Bcl-xL or Bcl-2 [3
]. In this report, baicalein reduced Bim and PUMA, and induced sensitizer BH3-only protein Bad. However Bad knock-down by siRNA did not reverse baicalein-induced apoptosis, indicating that Bad is not involved in baicalein-induced apoptosis. In addition to the increased expression of Bad, baicalein also increased phosphorylation of Bad on Ser112, which is important for translocation of Bad from the mitochondrial membrane to the cytosol, thereby inactivating the pro-apoptotic function of Bad. Again, this finding suggests that the pro-apoptotic effects of baicalein in pancreatic cancer cells do not involve activation of Bad. We observed a strong binding between Mcl-1 and Bak in PaCa cells. Bak has been reported to play a critical role on UV- or chemotherapeutic agent-induced apoptosis. To clarify the involvement of Bak in this setting, further studies are necessary to delineate the role of Bak in baicalein-induced apoptosis in PaCa cells. Our results suggest the following possible scenario: In untreated pancreatic cancer cells, Mcl-1 exerts a strong survival function by binding to pro-apoptotic Bcl-2 molecules, i.e. Bak. Baicalein leads to reduced expression on Mcl-1, thereby possibly freeing pro-apoptotic Bcl-2 proteins, i.e. Bak, from their suppression by Mcl-1.
Mcl-1 is considered as a promising therapeutic target because of its short half-life and multiple mechanisms regulating its expression. So far, some therapeutic approaches have been reported to abrogate the anti-apoptotic function of Mcl-1. It is reported that some BH3 mimetic, such as obatoclax, can bind to and disable Bcl-2 family proteins, including Mcl-1 [33
]. Recently, the multi-kinase inhibitor sorafenib has been reported to down-regulate Mcl-1 protein expression through a translational mechanism [34
]. Furthermore, certain natural products, such as piceatannol, have been shown to reduce Mcl-1 protein expression [35
]. Based on our data, we propose that baicalein may be a therapeutic tool targeting Mcl-1. Importantly, SB and SB-derived polyphenols are known to have almost no or very minimal toxicity in vitro and in vivo [11
], rendering them intriguing drug candidates.
In summary, we demonstrated that the pro-apoptotic effect of baicalein in PaCa cells is mediated through reducing the expression of the pro-survival protein Mcl-1, at least partially via a transcriptional mechanism. Mcl-1 is highly expressed in PaCa cell lines and down-regulation of Mcl-1 by siRNA showed a strong pro-apoptotic effect. These data suggest that baicalein may represent a promising therapeutic agent in PaCa and may have a beneficial value as a sensitizing agent for standard chemotherapeutic drugs by targeting a major survival molecule.