Metformin results in changes in gene expression levels in breast cancer cells
MDA-MB 468, MDA-MB 231 and MCF7 cells were treated with increasing doses of metformin (0,2; 0,4; 0,8; 1,6; 3,2 mM) in long term that they were incubated in each dose longer than a week. Then, we first analyzed the changes in transcriptional profile in that long-term metformin treated cells to see whether any cellular process was affected by the drug. The qPCR data showed that PIK3CA and mTOR mRNA expressions were downregulated in all cell lines. Even though 0,4mM and 0,8mM long term metformin treated MDA-MB-468 cells did not change the PIK3CA expressions significantly, PIK3CA was significantly downregulated in all other cells and in all treatment groups. Accordingly, we observed statistically significant downregulation pattern in mTOR expression in long term metformin treated cells excluding 0,4 mM metformin treated MDA-MB 468 cells and 0,8 mM and 1,6 mM treated MDA MB 231 cells. Interestingly, TOP2A and CCND1 gene expression levels exhibited increasing pattern in MDA-MB-468 cells which was the opposite of what was observed in MDA-MB231 and MCF7 cells. Whereas ZEB1, Vimentin and MMP9 expressions were upregulated in long term metformin treated MDA-MB 468 cells, that was not the case in the other cell lines. In MDA-MB 468 cells long term metformin treatment induced significant ZEB1 and VIM expression in all concentrations. Also, MMP9 was upregulated significantly in 0,4mM; 1,6mM and 3,2 mM metformin groups in MDA-MB-468 cells (Fig. 1).
The morphological changes are related with long term metformin treatment in a dose dependent manner
As metformin appears to induce proliferation and epithelial-mesenchymal transition (EMT) related gene expressions, we examined morphological changes in long term metformin treated MDA-MB 468 cells (will be referred to as metformin resistant [met-R] cells thereafter). As the cells got resistance to increasing doses of metformin, they lost their sphere like shape, and instead they gained spindle-like appearance (Fig. 2A). Especially the 1,6mM and the 3,2 mM metformin resistant cells exhibited mesenchymal phenotype which leaded us to investigate further the migratory capacity of these cells.
Metformin resistant cells exhibit significantly higher migratory capacity in a dose dependent manner
In the wound healing assay, we examined cell migration capacity of metformin resistant MDA-MB 468 cells by introducing mechanical scratch. Images of scratch areas from the time points 0 and 24 hours are illustrated in Fig. 2b. Wound healing assay revealed that high dose metformin resistant MDA-MB 468 cells had significantly more migration capacity compared to low dose resistant cells (Fig. 2B).
Metformin resistant MDA-MB 468 cells are involved in metastatic processes
The observation of increased migration capacity of these met-R MDA-MB -468 cells led us to investigate whether long term metformin use can also lead to EMT in TNBC. We tested this possibility by examining EMT related protein expressions in metformin resistant concentrations in MDA-MB 468 cells. As shown in Fig. 3, Zeb1, N-cadherin, Vimentin and Slug protein expressions were upregulated in all resistant concentrations. Accordingly, Claudin expression was downregulated in metformin resistant groups compared to parental control which demonstrates metastatic changes are compatible with microscopic images. Also, E-cadherin expression was downregulated in all resistant concentration groups except 0,4 mM metformin. While b-catenin was decreasing, snail was enhanced in 0,2 mM, 1,6 mM and 3, 2 mM metformin resistance cells. These distinct changes in EMT markers, clearly demonstrates that metformin resistant MDA-MB 468 cells has metastatic capacity compared to their parental counterparts.
Effects of metformin and LY294002 on cell viability in metformin resistant MDA-MB 468 cells
To check the proliferation pattern of metformin resistant MDA-MB-468 cells, MTT assay was used. Moreover, to assess whether metformin resistance will affect those cells’ sensitivity to PI3K inhibitor, we included PI3K inhibitor (LY294002) to the experimental set up as well (Fig. 4). We aimed to observe whether IC50 concentration of metformin induced cell death in metformin resistant MDA-MB 468 cells or not. As shown in Fig. 4, 10 mM metformin treatment was not effective on cell death in both 24 and 48 hours in all resistant groups compared to the control (parental) group. To determine the antiproliferative effects of a selective phosphatidylinositol 3-kinase (PI3K) inhibitor LY294002, all cells were treated with IC50 dose (10u) of LY294002 for 24 and 48 hours with or without metformin. After 24-hour treatment, 10 uM LY294002 exhibited higher cell proliferation significantly in 0,2mM, 0,4mM, 0,8mM metformin resistant MDA-MB468 cells compared to their control group. Also, at 48 hours, 10 uM LY294002 was significantly less effective in 0.2 and 3.2mM met-R cells in terms of reducing cell proliferation. Combined treatments led to a further decrease in cell viability compared to alone treatments of these agents. Therefore, these two agents have synergistic effects on reduction of metformin resistant MDA-MB 468 cell proliferation. However, the cell viability was lower in combinatorial treatment groups in metformin resistant cells compared to their parental controls. Especially, 0,2 mM at 24 hour and 3,2 mM metformin resistant MDA-MB 468 cells at 48 hours had significantly higher proliferation rates in combined treatments compared to control groups (Fig. 4).