Melanoma is one of the most complex and heterogeneous cancers [9, 10, 54]. Despite significant progress in research on melanoma heterogeneity, the identification of individual sub-populations of cells is still extremely difficult. These diagnostic limitations are influenced not only by the high genetic diversity of cells (in melanoma is the highest frequency of mutation from all cancers), but also acquired heterogeneity, resulting from cell reaction to changes in the environment of growing tumor. The literature data indicate that about 55% of solid tumors (including melanomas) are characterized by areas of hypoxia or anoxia in where the oxygen concentration doesn't excel 1.5% [51, 55, 56].
The available research shows that DPPIV glycoprotein can play an important role in development and progression of neoplastic diseases probably through the enzymatic and non-enzymatic mechanisms. An analysis of the available literature indicates differential CD26/DPP4 expression in different cancer types [26–38]. Our research confirmed differential expression of DPPIV in well-known cancer cell lines PC-3, DU-145, SK-Mel-28, A-375, G-361, 769-P, HepG2 representing 4 distinct tumor types. Our research showed that the greatest changes in the released glycoprotein level were observed between three different melanoma cell lines SK-MEL-28, A-375 and G-361. The human melanoma cell lines used in our studies have been previously described as lines of high and low metastatic potential, respectively . It was already determined that the sensitivity of tumor cells to changes in microenvironmental conditions like oxygen or nutrient deficiency can be diverse (from adaptation up to cells death) [57–60].
Research shows melanoma cells are capable of adapting quickly to changing microenvironmental conditions by leading to cellular metabolism changes (from oxidative phosphorylation to aerobic glycolysis for sufficient ATP production) [51, 56, 58, 59]. Our research indicates diverse sensitivity of melanoma cells to oxygen deficiency both in terms of cell viability and DPPIV glycoprotein levels. The SK-Mel-28 line known for high metastatic potential, was distinguished from other melanoma lines by a high viability but also level of DPPIV under hypoxia conditions. It should be noted that there was no research prior to ours which would evaluate the level of DPPIV in this melanoma cell line under conditions of hypoxic growth that mimics the tumor microenvironment.
Based on the analysis of the available literature, it can be concluded that hypoxia changes the pH in the intracellular and extracellular space. This could trigger an adaptive response among melanoma cells and activate glycolysis which stimulates intense growth of cells despite the seemingly unfavorable environmental conditions [61, 62]. The glycolytic phenotype of melanoma cells appears to be closely associated with higher metastatic potential and resistance to anti-cancer therapy.
Widmer et al.  showed that under hypoxic conditions the invasiveness of proliferative melanoma cell may be increased from 2- to 4-fold. Feige et al.  reached similar conclusions that hypoxia, through HIF1a, alters the gene-expression in proliferative melanoma cells, making them more invasive in in vitro assays. Cheli et al. showed that hypoxic conditions lead to a decrease in microphthalmia-associated transcription factor (MITF) expression which leads to increase in metastatic potential of melanoma cells in vivo. On the other hand, significantly lower cell viability of the A-375 and G-361, known for their low metastatic potential [60, 67], may suggest that hypoxia is not a factor which can trigger these cells adaptive switch to highly proliferative phenotype.
In our studies we noticed that, DPPIV level may depend not only on type of melanoma cell line but also on culture conditions. The change in the microenvironmental conditions for SK-Mel-28 cells determined the level of glycoprotein released by them. We didn't observe such dependencies for other tested melanoma lines. Regardless of the aforementioned observations, we were surprised with high level of glycoprotein released by examined melanoma cells. The analysis of limited number of available previous studies pointed to decrease or loss of CD26/DPPIV expression in the course of neoplastic transformation of melanocytes [21–23, 43]. In turn, the noticeable differences in the level of DPPIV released by SK-Mel-28 in the environmental conditions tested, may be due to activation of glycolysis by hypoxia which leads to extracellular acidosis of melanoma. This process could eventually result in lower DPPIV level compared to the level of this parameter under normoxic conditions (the optimal pH for DPPIV is 7.8) [61, 62].
Our studies found that DPPIV inhibitors, especially linagliptin, can inhibit melanoma cell proliferation in hypoxia conditions. Already the lowest concentrations of drugs dynamically reduced viability of the most sensitive to hypoxia cell lines: A-375 and G-361. Linagliptin at a concentration of 157.59 µM or 188.67 µM, respectively, significantly inhibited recruitment of the above-mentioned melanoma cells from sub-G1 phase to further stages of the cell cycle. Interestingly, the most hypoxic-resistant SK-Mel-28 line has undergone cytotoxic effects of linagliptin (IC50 78.26 µM) considered to be the most potent and selective dipeptidylpeptidase IV (DPPIV) inhibitors in this class of antidiabetic agents . Linagliptin triggered early and late apoptosis of the above-mentioned melanoma cells. A 2- fold increase of early apoptotic cells and over a 10-fold enhancement of late apoptotic cells were noticed after treatment with this gliptin. Also, Li et al.  revealed, the cytotoxic action of linagliptin in colorectal cancer cell line. The cytotoxic effect of linagliptin was dependent on the dose and the time of exposure of cancer cells. Linagliptin significantly inhibited HCT116 cell proliferation by cell cycle arrest at G2/M and S phase and the induction of cell apoptosis. In our study, we observed the increased population of SK-Mel-28 cells in G1 phase after treatment with aforementioned gliptin. Yang et al. confirmed the anti-cancer activity of sitagliptin in their research. Sitagliptin limited cell proliferation and invasiveness of endometrial cancer through regulation of HIF-1α and VEGFA signaling dependent on DPPIV expression.
Our results showed that lowering the level of DPPIV in all melanoma cell lines correlated with apoptogenic potential of linagliptin and saxagliptin. Despite extensive research into the new molecules which can initiate and regulate apoptosis in melanomas, still there is no enough information on how to effectively limit their chemoresistance. The gliptins we tested, especially linagliptin, turned out to be strong inducers of apoptosis in melanoma cells which are known not only from apoptotic pathway evasion strategy, but also from the unchecked proliferation. Linagliptin and saxagliptin in melanoma cells, induced both early and late apoptotic mechanisms [43, 75]. Moreover, the activation of apoptosis by DPPIV inhibitors led to a significant decrease in the number of viable melanoma cells. This may suggest that DPPIV is a promising therapeutic target for melanomas treatment.
In the future study, we want to provide information on the role of DPPIV inhibitors in modulation of melanoma cells resistance in available anti-cancer therapies. We want to check whether modulation of DPPIV expression translates to delay of melanoma cells resistance of anti-cancer therapy.