In the current study, we isolated WJ-MSCs and investigated their differentiation into IPCs using two differentiation protocols; with or without exendin-4. Our results showed that WJ-MSCs indeed fulfilled all MSC characteristics, and showed good potential to differentiate into IPCs. Additionally, exendin-4 was found to have profound beneficial effects on the differentiation outcome of those WJ-MSCs into IPCs. Besides, although our results showed that exendin-4 alone can induce the expression of some β-cell markers, namely Pdx-1, Nkx2.2 and MafA but not Isl-1, there still exists a synergistic effect on the induction of such markers upon addition of nicotinamide and β-mercaptoethanol with overall improvement of differentiation outcome.

Strategies to create surrogate β cells for therapeutic replacement have ignited significant excitement over the last decade, which has fueled profound interest in deriving functioning IPCs from stem cells [32]. Our results regarding the observed ease of isolation and relatively homogeneous population of isolated WJ-MSCs came in accordance with all of the reports highlighting the important potential of WJ-MSCs for various regenerative medicine applications [28].

Despite all efforts so far regarding the differentiation of stem cells to IPCs, current protocols are not optimized for different reasons, including the pleotropic effects induced by individual morphogens, together with the complexity of the signaling pathways involved and differences between the various types of used stem cells [18]. Further exploration into the differentiation of several stem cells to IPCs is thus indeed warranted together with investigating the various effects of incorporated inducing factors in the differentiation process [17, 33].

In the current study, we investigated two differentiation protocols, namely protocol A (without exendin-4) and protocol B (with exendin-4). Actually, both protocols induced the cells to detach and form clusters, and both of them showed positive DIZ staining; no apparent morphological differences were thus observed in IPCs generated from both protocols. Interestingly, the IPCs derived from protocol B, with exendin-4, showed a profound induction for the expression levels of all investigated β-cell-related genes in the current study, Pdx-1, Nkx2.2, Isl-1 and MafA, as compared with their control uninduced WJ-MSCs. These beneficial effects were further extended to our observations in the GSIS assay in which IPCs generated by protocol B demonstrated a reasonable response to HG as compared with LG concentrations, while those generated from protocol A totally lacked such response. Moreover, when carrying out immunocytochemistry to detect Pdx-1 in IPCs generated by both protocols, those generated by protocol B clearly expressed Pdx-1 but those derived from protocol A did not show such clear expression. Those findings also imply failure of protocol A lacking exendin-4 to induce proper differentiation, and further highlight the beneficial effects of exendin-4 on the differentiation outcome. Moreover, exendin-4 could enhance the viability of the generated IPCs as shown in Additional file 1: Figure S2. Actually, this might be attributed to the beneficial effects which have been reported recently for exendin-4 regarding improving MSC proliferation and cell survival [34].

Our findings regarding the enhancement of differentiation of WJ-MSCs into IPCs under the effect of exendin-4 are in agreement with previous studies which reported beneficial effects of exendin-4 for the differentiation of ESCs and increasing the insulin release by IPCs [22], MSCs from adipose tissue [35] or BM-MSCs [36].

Interestingly, apart from its beneficial effects on IPC differentiation, exendin-4 was previously reported to increase β-cell mass by stimulating β-cell proliferation and neogenesis in diabetic rats [26]. However, the exact mechanism of action of exendin-4 during the differentiation process and even its anti-diabetic actions or effects on β cells in clinical settings (exenatide) are not understood completely [37]. Among these suggested mechanisms are the induction of Pdx-1, and acting through several intracellular pathways such as the phosphatidylinositol-3-kinase (PI3K), the hedgehog or the MAPK/ERK pathway [18]. The suggested mechanism through inducing Pdx-1 was exhibited in our study by the significantly elevated expression of Pdx-1 transcript in IPCs generated from protocol B as compared with its very low expression levels in those obtained from protocol A.

Although exendin-4 has been implicated extensively in generation of IPCs from various forms of MSCs including BM-MSCs [36], adipose MSCs [38] and even WJ-MSCs [13], the exact mechanism of exendin-4 on differentiation of MSCs into IPCs is far from complete elucidation and studies involved in understanding this mechanism are scarce. Accordingly, we decided to study the effect of exendin-4 alone and in combination with nicotinamide and β-mercaptoethanol in this differentiation process.

Interestingly, our results clearly demonstrated different effects of exendin-4 on various β-cell differentiation markers. First, we found that exendin-4 can alone induce Pdx-1. Moreover, adding nicotinamide and β-mercaptoethanol caused further induction of this transcription factor. Several reports have shown that exendin-4 upregulated Pdx-1 during β-cell regeneration and that the effect of exendin-4 on β cells is mediated by Pdx-1 [39]. Actually, Pdx-1 is important for early pancreatic development and formation of β cells, and also acts as an insulin secretion activator [40]. Moreover, overexpression of Pdx-1 has been implicated in the differentiation of ESCs into IPCs [41]. Our results therefore imply that the improvement of differentiation outcome induced by exendin-4 might be mediated by Pdx-1.

In addition to Pdx-1 as an early pancreatic development factor, exendin-4 also induced the expression of Nkx2.2 transcription factor in WJ-MSCs. Nkx2.2 is implicated in early pancreatic development and, again, its overexpression induced β-cell differentiation from ESCs [42, 43]. In contrast to Pdx-1 and Nkx2.2, exendin-4 failed alone to induce Isl-1, another transcription factor implicated in later endocrine differentiation during pancreatic development [44]. However, when exendin-4 was added with nicotinamide and/or β-mercaptoethanol, all of them showed some sort of synergism and induced the expression of Isl-1 significantly as compared with control noninduced cells and as compared with induction using exendin-4 alone. Collectively, these results indicate that exendin-4 can be playing a role in the early development of IPCs from MSCs through its effect on Pdx-1 and Nkx2.2, and possibly through Isl-1 in the presence of nicotinamide and β-mercaptoethanol, but not alone. We assume this is considered an interesting finding that confirms not only the need for several factors to achieve complete differentiation to IPCs, but also reflects the complicated steps of such a process.

Interestingly, our results demonstrated that MafA transcript levels could be induced by exendin-4. MafA controls and activates the expression and secretion of insulin [45]. It is noteworthy here that exendin-4 at the dose used in our protocols was reported previously to increase the insulin secretion by β cells differentiated from ESCs [22]. Moreover, a recent study showed that proper activation of MafA at a specific stage of β-cell differentiation is critical for achieving optimum maturation of β cells [46]. Actually, the induced expression of MafA in the current study was enhanced with the addition of nicotinamide and β-mercaptoethanol, indicating a synergistic effect of these three factors for better differentiation of WJ-MSCs into IPCs, with the ultimate goal of attaining fully differentiated IPCs. This was clearly reflected in the far better GSIS of the IPCs generated from protocol B, with exendin-4, than that of protocol A, as well as the better expression of Pdx-1 protein in IPCs generated from protocol B as compared with those derived from protocol A. Collectively, these results clearly demonstrate the importance of exendin-4 in attaining better mature IPCs from WJ-MSCs and, therefore, improving the differentiation outcome of these cells.

It is noteworthy here that nicotinamide and β-mercaptoethanol used together in protocol A could induce some transcription factors, namely Nkx2.2, Isl-1 and MafA. On the contrary, nicotinamide and β-mercaptoethanol showed a synergistic effect with exendin-4 in induction of Pdx-1, Isl-1 and MafA. From these results one can assume that these extrinsic factors are acting in a complementary way in the induction of the complex process of differentiation. Accordingly, one can recommend, with confidence, using these extrinsic factors together in order to achieve better functioning IPCs. Also, we can portray exendin-4 as an indispensable factor in the differentiation protocols for WJ-MSCs into IPCs based on the levels of induction of β-cell markers in protocol B and better functionality reflected in better GSIS of this protocol.

Collectively, it seems that exendin-4 can induce several stages in the complex differentiation process, including the early stages exemplified by Pdx-1 and Nkx2.2, in addition to late stages of insulin secretion exemplified by MafA.

This study is one of the few that dissect the role of exendin-4 on various β-cell markers upon differentiation of WJ-MSCs. This will be of great help in understanding the mechanism behind the effect of exendin-4 on the generation of IPCs from MSCs of different sources. Interestingly, a recent study by Xin et al. [47] showed that IPCs differentiated from human BM-MSCs could ameliorate diabetes in streptozotocin-induced diabetic nude mice. This implicates that differentiated cells can find their ways into diabetes regenerative medicine. However, the existing induction strategies of IPCs from MSCs need to be modified and improved. Our study is a step toward better understanding and developing a competent protocol of differentiation that ensures better functioning IPCs through the utilization of exendin-4 side by side with nicotinamide and β-mercaptoethanol.

In fact, the synergistic effect between exendin-4 with nicotinamide and β-mercaptoethanol observed in the induction of expression of several β-cell markers like Pdx-1, Isl-1 as well as MafA sheds lights on the fact that the effects of exendin-4 could indeed be affected greatly by other extrinsic factors. This opens the door for a wide array of further investigations to study possible interactions between exendin-4 and other extrinsic factors incorporated during the course of various differentiation protocols in the literature, with the ultimate goal of achieving fully mature functional β cells for cell therapy of diabetes.

Our results demonstrate that exendin-4 greatly enhanced the differentiation of WJ-MSCs into IPCs. In doing so, exendin-4 induced several β-cell markers with some synergy with nicotinamide and β-mercaptoethanol. This is considered a step forward toward better understanding the differentiation process and achieving better functioning IPCs from MSCs that can be of value in diabetes regenerative medicine.