Discussion

GBA is a bioactive sesquiterpene coumarin compound isolated fromFerula species (Bedniak, 1962; Iranshahi et al., 2018).Ferula assafoetida and Ferula Szowitsiana , as rich sources of GBA, are herbaceous plants of the genus Ferula belonging to the Umbelliferae family and are distributed throughout central Asia, eastern Iran, and Afghanistan (Abd El-Razek et al., 2001; Iranshahi et al., 2007; Yaqoob and Nawchoo, 2016). GBA is an herbal compound with multiple biological activities including anticancer, cancer chemopreventive, hepatoprotective, and antiviral activities (Bedniak et al., 1967; Syrov et al., 1990; Iranshahi et al., 2008; Iranshahy and Iranshahi, 2011; Kim et al., 2011; Kasaian et al., 2014). GBA exerts its anticancer activity in association with apoptosis induction, P-glycoprotein (P-gp) inhibition, and anti-proliferative actions (Hanafi-Bojd et al., 2011; Kim et al., 2011). The anti-tumor activities of GBA inspired us to investigate the effects of GBA in MT-2 cell line to find if GBA can improve the therapeutic efficacy of ATO, as a potential agent in combination with antivirals and IFN-α in the treatment of ATL (Kchour et al., 2009), and determine whether the GBA+ATO combination could be an effective treatment for patients with ATL.
Results of the present study indicate that GBA is cytotoxic to MT-2 cells in a dose-dependent manner with an IC50 value of 80 µM at 48 h. The growth inhibitory effects of GBA have been reported in several studies. Eskandani et al. reported that GBA increases the percentage of early/late apoptotic cells and attenuates the growth rates of OVCAR-3 human ovarian carcinoma cells in a dose- and time-dependent manner with an IC50 of 37, 12.1, and 10 µM at 24, 48, and 72 h, respectively (Eskandani et al., 2015a). In another study, Eskandani et al. showed that intact GBA and GBA-loaded solid lipid nanoparticles (GBA-SLNs) inhibited the growth of A549 human lung epithelial carcinoma cells by upregulation of proapoptotic protein caspase 9 and downregulation of antiapoptotic protein Bcl-xL (Eskandani et al., 2015b). Kim et al. in 2011 reported that GBA plays its anticancer effects against human umbilical vein epithelial cells (HUVECs) and Lewig lung cancer (LLC) cells through anti-angiogenic and anti-proliferative activities (Kim et al., 2011). Zhang et al. in 2012 demonstrated that GBA preferentially inhibits the growth of androgen receptor (AR)+prostate cancer cells (with an IC50 of approximately 80 µM at 72 h) compared to AR- prostate cancer cells by downregulating the AR levels and consequent inhibition of AR signaling pathway which is a critical regulator of the G1/S transition and important for development and progression of prostate cancer. Additionally, combination of GBA and bicalutamide, an AR antagonist, showed a greater than additive cytotoxicity in LNCaP prostate cancer cells (Zhang et al., 2012). In the current study, we showed that the combination of GBA and ATO was also cytotoxic against MT-2 cells in a dose-dependent manner and induced the most cytotoxicity  on MT-2 cells after 48 hours. Indeed, the combination of 20 µM GBA and 4 µM ATO at 48h significantly decreased the proliferative activity of MT-2 cells compared to each agent alone. In accordance with our findings, Kim et al. in 2019 reported a significant dose-dependent cytotoxicity for the combination of GBA and TNF related apoptosis inducing ligand (TRAIL) in resistant H460/R non-small cell lung cancer cells (NSCLCs) (Kim et al., 2019). It was also shown that GBA in combination with nanomicellar curcumin significantly inhibits the growth of murine C26 and human Caco-2 colon carcinoma cells in a dose-dependent manner (Jafari et al., 2019). Results obtained in our study indicate that GBA is able to inhibit MT-2 cells growth specifically when combined with ATO. Cell cycle analysis also showed that GBA or ATO alone induced apoptotic cell death and sub-G1 phase arrest in a low percentage of treated cells (2.30% and 4.30%, respectively), whereas the combination of GBA and ATO at the equal concentrations significantly increased the sub-G1 apoptotic population (55.56%). In line with the findings of our study, several reports have indicated that GBA increases sub-G1 apoptotic population. Kim et al. demonstrated that the combination of GBA (25 or 50 μM) and TRAIL (25 ng/ml) in resistant H460/R NSCLCs increased distribution of apoptotic cells in sub-G1 phase to 24.75% and 31.09%, respectively, compared with 25 μM GBA (6.50%), 50 μM GBA (12.26%), and 25 ng/ml TRAIL (11.51%) (Kim et al., 2019). Oh et al. have also shown that GBA at the concentrations of 25 and 50 µM significantly increases sub-G1 phase cells population in H460 NSCLCs to 7.71% and 7.92%, respectively, compared with the untreated control (0.17%) (Oh et al., 2015). By contrast, Kim et al. have detected no apparent changes in the frequency of GBA-treated LLC cells in the sub-G1 phase, indicating that GBA could not induce apoptotic cell death in LLC cells. Nevertheless, cell cycle analysis showed that GBA inhibited the LLC cells growth by causing an evident G2/M phase arrest (Kim et al., 2011). Contrary to the findings of Kim et al. study, it is clear that the GBA+ATO combination treatment synergistically induced sub-G1 apoptotic population in MT-2 cells.
Pumping therapeutic agents out of cell, mediated by multiple drug efflux transporters of the ATP-binding cassette (ABC) family, is a serious impediment to the effective chemotherapy. Overexpression and enhanced efflux activity of MDR1/P-gp or ABC sub-family B member 1 (ABCB1) have been reported in patients with ATL. It has been also reported that HTLV-1 Tax protein is a potent transcriptional activator of MDR1/P-gp gene promoter. MDR1/P-gp overexpression endows ATL cells with a drug resistance phenotype (Lau et al., 1998). Therapeutic agents with the ability to inhibit the P-gp mediated efflux activity could be used in combination with existing medicines to increase their potency against ATL cells. Several studies have investigated the effects of GBA on MDR-1/P-gp. Studying the effects of GBA on P-gp efflux activity via rhodamine 123 efflux assay in doxorubicin-resistant MCF7/Adr breast cancer cells revealed that GBA is a potent P-gp inhibitor that is more effective in inhibiting the function of ABCB1/P-gp compared to verapamil, a typical P-gp inhibitor (Hanafi-Bojd et al., 2011). Using GBA against six Multi-drug resistance clinical isolates ofStaphylococcus aureus showed that GBA has efflux inhibitory effects on P-gp and its mechanism of action is comparable to verapamil (Bazzaz et al., 2010). In another study, Kim et al. showed that GBA in combination with TRAIL inhibits MDR1 efflux activity by repressing the MDR1 expression in resistant H460/R NSCLCs, thereby leading to enhanced TRAIL-induced apoptosis induction in NSCLCs (Kim et al., 2019). The present study confirms previous findings about GBA inhibitory effect on the functionality of the P-gp efflux pump. Results obtained from efflux assay showed that the rate of MTX accumulation in MT-2 cells increased in the presence of GBA. The enhanced accumulation of MTX in GBA-treated MT-2 cells may be attributed to efflux inhibitory properties of GBA. Therefore, GBA increased intracellular accumulation of ATO in MT-2 cells through inhibition of P-gp efflux activity and therefore could overcome the P-gp mediated Multi-drug resistance and induced apoptotic cell death in MT-2 cells.
The expression levels of genes involved in cell proliferation and apoptosis is altered in ATL, resulting in an increase in cell proliferation and survival and a reduction of apoptosis (Duyao et al., 1992; Mori et al., 1999; Ariumi et al., 2000; Haller et al., 2000; Yoshida, 2001; Sun and Yamaoka, 2005; Krueger et al., 2006). In the present study, we found that the mRNA levels of RelA, CDK4, and c-MYC significantly decreased after treatment with 20µM GBA. However, when combined with 4µM ATO, only the expression levels of CDK4 was downregulated, although, the reduction in CDK4 levels in treatment with GBA alone was greater than the reduction mediated by the GBA+ATO combination. Cyclin dependent kinase 4 (CDK4) mediates G1/S cell cycle progression (Sherr and Roberts, 1995). In HTLV-1 infected cells, Tax interacts with inhibitors of CDK activity such as p16INK4A, leading to uncontrolled cell proliferation (Suzuki et al., 1996; Haller et al., 2000; Yoshida, 2001). GBA-mediated downregulation of CDK4 might inhibit transition from G1phase of cell cycle to S and therefore induces sub-G1phase cells accumulation. In addition to CDK4, we also observed reduced expression levels of c-FLIPL and c-FLIPS in cells treated with the GBA+ATO combination, which might associate with induced apoptotic cell death in MT-2 cells. Several studies have reported that high expression levels of c-FLIP supports the cancer cells to elude the immunosurveillance and promotes the tumor cells growth (Djerbi et al., 1999; Medema et al., 1999). Tax expressing HTLV-1 infected T cells express high levels of c-FLIPL and c-FLIPS which contributes to inhibition of apoptosis in infected cells, blocking the receptor mediated cell death, evading the host immune response and supports development of HTLV-1 associated diseases (Krueger et al., 2006; Okamoto et al., 2006). GBA-induced downregulation of c-FLIPL and c-FLIPS in MT-2 cells treated with the GBA+ATO combination could augment death receptor mediated apoptosis in MT-2 cells and make them more susceptible to apoptotic cell death.
The GBA+ATO combination unexpectedly decreased the expression levels of the tumor suppressor p53 in MT-2 cells. p53 plays a crucial role in cancer prevention by controlling the cell cycle progression (Ozaki and Nakagawara, 2011; Xu-Monette et al., 2012). In the current study, a significant positive correlation was found between the expression of c-FLIPS and p53. The GBA-mediated downregulation of p53 might be related with the involvement of p53 in c-FLIPS expression (Bartke et al., 2001). c-FLIP contributes to apoptosis resistance. Therefore, treatment with the GBA+ATO combination resulted in downregulation of p53 expression in MT-2 cells and consequently inhibition of the p53-induced upregulation of c-FLIPS and therefore suppression of the c-FLIPS apoptosis inhibitory effects. On this basis, we conclude that apoptosis induction in MT-2 cells seems to be independent of p53 activation. Taken together, GBA+ATO mediated downregulation of p53 and c-FLIPS might lead to induction of death receptor-mediated apoptosis in MT-2 cells.