The CSC-inhibiting activity of salinomycin has previously been demonstrated in a variety of tumors including those of the breast, lung, and colon. Here we have extended these studies by showing that salinomycin induces apoptosis and chemosensitivity while inhibiting cell proliferation, invasion, stem cell marker expression and sphere formation in putative HNSCC stem cells. Ultimately, these results suggest that salinomycin or its derivatives may be an effective novel treatment for HNSCC, especially when administered in combination with standard treatments. Our results are consistent with a previous study by Busa et al. reporting the ability of salinomycin to eradicate treatment-resistant phenotypes in HNSCC. However, Basu et al. report no observed synergistic efficacy between salinomycin and cisplatin in HNSCC in vitro, speculating a possible overlap of the individual drugs’ cytotoxic mechanisms . Although the method of quantifying drug interactions is not specified, we are not surprised by this finding given the relatively weak synergy observed between cisplatin and salinomycin in JLO-1. In contrast, combination treatment of paclitaxel with salinomycin resulted in strong synergy for all tested drug ratios, emphasizing the potential of this drug pair in the treatment of HNSCC. Salinomycin was also observed in our system to activate Akt signaling and induce changes in gene expression indicative of EMT. These results are quite unusual and potentially worrisome given that Akt signaling and EMT are both heavily implicated in cell proliferation, invasion and acquisition of CSC properties.
At this time of writing there appears to be no other study which documents the effect of salinomycin on Akt, leaving open for investigation whether salinomycin also activates Akt in other cancers. Drugs including cisplatin, etoposide, doxorubicin, and tamoxifen have been shown to induce Akt phosphorylation leading to chemoresistance in some cancers [28–30]. Similarly, it is possible that pro-survival mechanisms within HNSCC stem cells activate Akt in the presence of salinomycin in attempt to overcome drug-induced cell death. Further investigation is required to elucidate the mechanisms that are responsible for drug-induced phosphorylation. What is clear, however, is that salinomycin is ultimately capable of inducing apoptosis and inhibiting cell proliferation in HNSCC stem cells. Since apoptosis occurs despite the activation of Akt, it is likely that salinomycin targets apoptotic pathways that are downstream of Akt. We report an induction of Bax and constant expression of Bcl-2 in salinomycin-treated JLO-1 despite increased Akt kinase activity. Previous studies have also shown that salinomycin is capable of inducing apoptosis through a variety of targets including Bcl-2, P-glycoprotein, 26S proteasome, calpain and cytochrome C, all of which are downstream or independent of Akt .
EMT has been nearly synonymous with the acquisition of an invasive and metastatic phenotype and its link to cancer stem cell properties is also becoming well-established [15, 32]. Furthermore, salinomycin was originally identified as a cancer stem cell inhibitor by screening for drugs with specific toxicity against mesenchymally transdifferentiated breast cancer cells . Likewise, Basu et al. demonstrated in vivo depletion by salinomycin of the vimentin-positive subpopulation and enrichment of the E-cadherin-positive subpopulation in primary tumor-derived xenografts, possibly through selective cytotoxicity, promotion of MET, or inhibition of EMT . Thus, it is interesting that salinomycin induces gene expression changes indicative of EMT in HNSCC stem cells, especially while inhibiting invasion and stemness. This surprising observation requires a reassessment of the link between EMT and cancer stem cells, and strongly suggests that EMT may not in all cases lead to an invasive or stem cell-like phenotype.
Although this study marks the first instance in which salinomycin is shown to induce EMT, it is not the first to show a disconnection between EMT and stem cells. In fact, it is well known that embryonic stem cells (ESCs) resemble epithelial cells and have high expression of E-cadherin, which is crucial for pluripotency in ESCs and may even be used in place of Oct-4 during somatic cell reprogramming . It is also well established that the reverse of EMT, mesenchymal-epithelial transition (MET), is a critical step for reprogramming mouse fibroblasts to induced pluripotent stem cells (iPSCs) [34, 35]. In terms of cancer, it was recently demonstrated that prostate cancer stem cells are characterized by high E-cadherin expression, are highly invasive, and exhibit high expression of stem cell markers Oct-3/4 and Sox2 compared to cells with low E-cadherin expression [36, 37]. It has been hypothesized that cancer stem cells possess a high degree of plasticity, and that following EMT, E-cadherin expression may be restored without losing stem cell function or invasive capacity .
Further research is required to reconcile the apparent inconsistency between contexts in terms of the relationship between EMT and stemness. Contrary to the data presented here, a previous report has confirmed the ability of salinomycin to reverse EMT in colorectal cancer . Thus, whether salinomycin promotes or inhibits EMT varies between cases and may be highly dependent on cell type. In any case, the data presented here make it clear that EMT does not always correspond to stem cell phenotype or invasion, and that salinomycin may induce loss of stemness through pathways that are independent of EMT.
Investigating changes in microRNA expression may offer additional insight into the mechanism of salinomycin. In particular, microRNAs have previously been shown to regulate invasion via EMT-independent pathways . MiR-328 has been shown to negatively regulate the expression of ABCG2 in human cancer cells, while miR-199a-3p has been shown to induce cell cycle arrest, reduce invasion, and increase doxorubicin sensitivity by negatively regulating mTOR and c-Met [25, 26]. Interestingly, we report an increase in activation of mTOR upon salinomycin treatment despite induction of miR-199a-3p. However, it is known that individual microRNAs can target multiple components within one signaling pathway. MiR-199a-3p has also been shown to inhibit proliferation by negatively regulating the cancer stem cell marker CD44 . The upregulation of these miRs may explain some of the effects of salinomycin including the downregulation of CD44, decrease in invasion, and the synergy observed between salinomycin and cisplatin or paclitaxel. The ability of salinomycin to induce EMT in HNSCC stem cells may be explained by the dose-dependent downregulation of miR-203, which has been shown to inhibit EMT in prostate cancer . Further characterization of these microRNAs and other potential pathways affected by salinomycin will provide a greater understanding of how to target cancer stem cells.