As a stable, inexpensive and highly effective oral drug, metformin has been used for the treatment of type 2 diabetes for several decades. It stimulates glucose uptake and increases fatty acid oxidation in muscle and liver with no adverse effects [3, 17]. Recent data indicate that metformin can protect from cancer and inhibit the proliferation of several types of cancer cells in vitro and in vivo, such as breast cancer , gastric cancer , pancreatic cancer , and thyroid cancer . The anti-tumor effects of metformin have been investigated in different types of adenocarcinoma; however, its effects on squamous cell carcinoma, a malignant tumor of epidermal keratinocytes that invades the dermis, have not yet been well defined. Adenocarcinoma and squamous cell carcinoma can differ significantly in their symptoms, natural history, prognosis, and response to treatment owing to differences in cellular origin. In the present study we focused on the effects of metformin on OSCC, a common squamous cell carcinoma of the head and neck. The present findings are significant because 1) we demonstrate for the first time that metformin exerts potent anti-OSCC effects both in vitro and in vivo; 2) metformin induces cell cycle arrest at the G0/G1 phase and apoptosis of OSCC cells associated with the modulation of cell cycle-regulatory and apoptosis-related protein expression. CDK inhibitors such as p21 and p27 have been shown to play an important role in the inhibitory effects of metformin in previous studies [18, 20]. However, in the present study, we did not observe significant changes of these proteins in OSCC cells following metformin treatment. This discrepancy could be due to the differences in the properties of the different types of cancer cells.
A previous study showed that specific cyclin/CDK complexes are activated at different intervals during the cell cycle and complexes of CDK4 and CDK6 with cyclin D1 are required for G1 phase progression . Down-regulation of cyclin D1 in response to metformin has been shown in several cancer cell lines including breast cancer  and prostate cancer  cells. The effects of metformin on the catalytic subunits of cyclin D1, CDK4 and CDK6 in OSCC cells, however, remain unknown. In the present study, metformin blocked cell cycle progression at the G0/G1 phase, which was correlated with a remarkable decrease in the expression of cyclin D1 and phosphorylation of pRb, two major cell-cycle regulators. Cyclin D1 binds to and activates CD4/CDK6, which then phosphorylates pRb. Upon phosphorylation, pRb releases the transcription factor E2F, which activates the transcription of genes required for G1/S phase transition . Cyclin D1 gene amplification and overexpression are observed in several types of human cancer including OSCC [23–25]. Furthermore, overexpressed cyclin D1 is associated with enhanced tumor growth and chemotherapy resistance [24, 26]. Thus, cyclin D1 is a potential molecular target for the treatment of OSCC. In addition to its effect on cyclin D1, metformin strongly inhibits the phosphorylation of pRb in OSCC cells, blocking the activation of E2F. Activation of E2F by disruption of the Rb tumor suppressor pathway is a key event in the development of many human cancers. Increased expression of E2F is associated with malignant transformation in OSCC, and down-regulation of this transcription factor is associated with induction of apoptosis and cell cycle arrest in OSCC cells [27, 28]. Therefore, our results suggest that metformin could be developed as a potential therapeutic agent to block the progression of OSCC.
In the present study, metformin activated the AMPK pathway and inhibited S6K and mTOR phosphorylation in OSCC cells, suggesting that the mTOR pathway may be involved in mediating the effect of metformin in these cells. However, the role of AMPK in the activation of mTOR signaling is the subject of controversy. Using siRNA against the two catalytic subunits of AMPK, Ben Sahra et al. demonstrated that the anti-proliferative effect of metformin was mediated by the mTOR pathway independently of AMPK . On the other hand, Zakikhani et al. showed that metformin inhibited cell growth via the α1 AMPK subunit in MCF-7 breast cancer cells . Although our results clearly showed the growth inhibitory effect of metformin in OSCC, the involvement of the AMPK pathway in the anti-tumor effect of metformin on OSCC remains to be elucidated. Moreover, because metformin is known to play a role in the control of cell metabolism, it would be interesting to determine whether the metabolic consequences of metformin are related to its anti-proliferative effects.
In addition to the effect of metformin on the cell cycle, we examined whether the anti-neoplastic effect of this agent is mediated by the induction of apoptosis. Our flow cytometry results demonstrated that metformin significantly induced apoptosis in all three OSCC cells lines. These findings were further confirmed by our western blot results showing a significant down-regulation of the anti-apoptotic proteins Bcl-2 and Bcl-xL and the up-regulation of the pro-apoptotic protein Bax. Several death and survival genes, such as Bcl-2 or Bax, which are regulated by extracellular factors, are involved in apoptosis . When the ratio of pro-apoptotic Bcl-2 family members (Bax, Bad) to anti-apoptotic Bcl-2 family members (Bcl-2, Bcl-xL and Mcl-1) increases, pores form in the outer mitochondrial membrane, liberating apoptogenic mitochondrial proteins to activate caspases and induce apoptosis . Data concerning the effect of metformin on apoptosis in cancer cells are limited and controversial. A recent study indicated that metformin suppressed the growth of human head and neck squamous cell carcinoma mainly via G1 arrest, which coincided with a decrease in the protein levels of CDKs, cyclins and CDK inhibitors . Ben Sahra et al. also showed that metformin blocked the cell cycle in the G0/G1 phase in prostate cancer cells and did not induce apoptosis . In contrast, metformin has been shown to promote apoptosis in pancreatic cancer  and melanoma  cells. This discrepancy between studies regarding the effect of metformin on apoptosis may be the result of variations in experimental conditions, cell-specific functions and/or different cell origin, and suggests that further investigation is necessary. Moreover, Hirsch et al.  reported that low doses of metformin could inhibit cellular transformation and selectively kill cancer stem cells in four genetically different types of breast cancer, thus inhibited the tumor growth both in vitro and in vivo. Whether similar mechanisms also contribute to the anti-cancer effect of metformin in OSCC still needs to be identified in our further study.
Although the doses of 20 mM metformin used in our in vitro study are similar to those used in prior studies on gastric cancer , melanoma  and breast cancer , one can still argue that these doses are above physiological levels. Indeed, the concentration of metformin in the blood of type 2 diabetic patients treated with the drug is approximately 30 ~ 60 μmol/L , which indicates that the doses used in our study exceeded the therapeutic level by 300-fold. However, it has been reported that metformin accumulates in tissues at concentrations similar to the dose used in our experiments [35, 36]. Moreover, tumor cells in culture are grown under high concentrations of glucose and 10% FBS, which results in excessive growth stimulation. This may also contribute to the high dose of metformin required to exert anti-tumor effects in a cell culture system compared to the dose used in patients with diabetes. Furthermore, according to the study of Ben Sahra et al., the doses of 1 to 3 mg/day metformin caused no side effect in mice, which was equal to the dosage used for patients , we obtained a strong inhibition of OSCC tumor growth in vivo. 200 μg/ml metformin administered orally significantly decreased OSCC growth in a xenograft model. This result is of particular importance as it is the first time that metformin is shown to inhibit OSCC tumor growth in vivo.