In the present study, we investigated the effects of omentin-1 on MSCs and found that omentin-1 can promote MSC proliferation, inhibit H2O2-induced apoptosis, enhance angiogenic growth factor secretion, and elevate the ability of MSCs to stimulate tube formation by HUVECs. Moreover, these effects of omentin-1 were shown to be mediated, at least partially, through the PI3K/Akt signaling pathway. To our knowledge, this is the first study to report the beneficial effects of omentin-1 on MSCs and the underlying mechanisms.

Based on promising preclinical and clinical data, MSC therapy has been suggested as a potential therapeutic strategy for tissue repair [38]. Moreover, accumulating evidence suggests that MSCs repair the damaged tissue mainly by secreting cytokines and other paracrine factors to promote angiogenesis, activating endogenous reparative responses [39, 40] rather than directly by differentiation and numerical replacement of lost cell mass [40]. However, the success of MSC therapy still has its limitations. One of them is that the small number obtained from the donors requires extensive expansion for therapeutic utility [7, 8]. The other is that MSC therapy is compromised by the decreased survival of transplanted MSCs mainly due to oxidative stress and hypoxia in the harsh environment [9, 10], which in turn inhibits angiogenesis and tissue repair [4143]. Therefore, to overcome these problems, researchers have attempted to enhance MSC proliferation, survival, and increase the secretion of cytokines for angiogenesis and other tissue repair function.

Omentin-1 is a new fat depot-specific secretory adipokine, the plasma levels of which are significantly decreased in patients with obesity, insulin resistance, diabetes, and cardiovascular diseases such as atherosclerosis, coronary artery disease, and ischemic heart disease [17, 20]. Although omentin-1 has been reported to exert proproliferation, prosurvival, and proangiogenic functions in various cells via an Akt-dependent mechanism [22, 23, 29], its specific role and underlying mechanism in MSCs remains unclear.

In the present study, the effects of different concentrations of omentin-1 (100, 200, 400, and 800 ng/ml) on MSCs were assessed. All of the concentrations of omentin-1, except 100 ng/ml, had a significant promoting effect on the growth of the MSCs in a time- and concentration-dependent manner. In addition, the number of cells in the S and G2 phases of the cell cycle significantly increased under omentin-1 exposure and the process was accompanied by upregulation of cyclin D1/E and downregulation of p21/p27. Of note, the concentration at 100 ng/ml was significantly lower than the physiological concentration in serum (200–400 ng/ml) [17, 44], suggesting that physiological and higher concentrations are required for MSC proliferation. Furthermore, we used H2O2 to induce oxidative stress to MSCs. The results showed that H2O2 induced higher intracellular ROS and more cellular apoptosis with a mitochondrial membrane potential decrease and caspase-3 activation. Moreover, excessive ROS could attack the mitochondrial membrane, leading to loss of the potential of the mitochondrial membrane and the consequent apoptosis of MSCs [45]; however, omentin-1 pretreatment could reduce the excessive ROS, restore mitochondrial membrane potential, and inhibit the apoptosis of MSCs in a concentration-dependent manner. In addition, the balance of Bax and Bcl-2 is responsible for the integrity of the mitochondrial membrane and ΔΨm stabilization [46]. Higher Bax and lower Bcl-2 levels resulted in a loss of ΔΨm and mitochondrial swelling or disruption [46]. In contrast, our experiment demonstrated that omentin-1 pretreatment upregulated Bcl-2 but downregulated Bax, which contributed to the preservation of mitochondrial function. Caspases are downstream of the Bcl-2 family in the apoptotic cascade and caspase-3 is one of the key effectors of apoptosis [47]. We further detected the activation of caspase-3 and found that omentin-1 pretreatment markedly reduced the levels of cleaved caspase-3 compared with the cells treated with H2O2 alone. Therefore, these findings suggest that omentin-1 may exert its protective property in oxidative stress-induced MSC apoptosis by reducing the excessive ROS and inhibiting mitochondria-dependent caspase cascades. Furthermore, our data showed that omentin-1 could increase MSC phosphorylated Akt levels in a time- and concentration-dependent manner. However, LY294002, the specific inhibitor of PI3K, not only blocked the activation of Akt induced by omentin-1, but also significantly attenuated the promoting effects of omentin-1 on the growth and survival of MSCs, suggesting that omentin-1 treatment activated Akt and led to MSC proliferation and survival.

Forkhead box O3 (FoxO3a), a transcription factor, is the downstream target of Akt [48]. Moreover, FoxO3a can promote antiproliferation or proapoptosis signaling through either increasing the protein levels of cyclin-dependent kinase inhibitors [49], or regulating the expression of Bcl-2 family proteins [50]. Activated Akt-mediated phosphorylation and inactivation of FoxO3a improves cell proliferation and survival. Likewise, glycogen synthase kinase-3β (GSK-3β) also exists downstream of Akt and acts as a key regulator of multiple processes that are critical for the proliferation and apoptosis in various cells, including MSCs [51, 52]. In this present study, we found that FoxO3a and GSK-3β were phosphorylated by omentin-1, which might be a direct result of Akt activation, and this was reversed in the presence of LY294002. This suggested a potential role of FoxO3a and GSK-3β in omentin-1-induced MSC proliferation and survival.

In addition, omentin-1 has also been observed to exhibit antiproliferative and proapoptotic effects via the PI3K/Akt pathway in other types of cells, such as human osteoblasts [27], mouse neural stem cells [53], cardiomyocytes [23], and endothelial cells [22]. In contrast, Zhang and colleagues [54] provided evidence that omentin-1, as an anticancer factor, inhibited human hepatocellular carcinoma cell proliferation and survival via the JNK signaling pathway. It was also found that omentin-1 attenuated neointimal formation after arterial injury and suppressed vascular smooth muscle cell proliferation through AMPK-ERK-dependent mechanisms [55]. One possible explanation for this discrepancy could be that the activity of these signaling pathways is differentially controlled in specific cells in response to omentin-1 treatment, and the specificity of the regulation is dependent on cell type and/or its exposure environment.

During the past decade, it has been demonstrated that MSCs can promote angiogenesis by the secretion of proangiogenic factors (e.g., VEGF, IGF-1, HGF, and FGF-2) that contribute to tissue repair and enhance the reparative process [41, 56, 57]. Thus, various strategies have been adopted to upregulate the secretion of angiogenic factors in MSCs, such as gene transfer [41] and culture under hypoxic conditions [58]. Although it has also been reported that omentin-1 could directly promote endothelial cell function and revascularization in the hind limb ischemia mouse model [22], the indirect effect of omentin-1 on angiogenesis through MSCs remains unclear. In this study, we observed that conditioned medium from omentin-1-pretreated MSCs significantly increased the tube length relative to the conditioned medium from untreated MSCs and basal medium. Furthermore, we confirmed that proangiogenic factors (VEGF, FGF-2, HGF, and IGF-1) from omentin-1-pretreated MSC conditioned medium had a higher level than that from untreated MSCs; thus we concluded that omentin-1 could augment endothelial cell capillary tube forming capacity in vitro at least partially due to the upregulation of the secretion of angiogenic factors by MSCs. Additionally, a blockade of the Akt pathway caused a reduction in the tube-forming capacity and angiogenic cytokines of conditioned medium from MSCs pretreated with omentin-1. Together, these data revealed that the proangiogenic effects of omentin-1 on MSCs are mediated at least partially through the PI3K/Akt signaling pathway. However, these results do not rule out the proangiogenic effects of other factors that are secreted by MSCs, such as Ang-1, Ang-2, IL-6, and PLGF. It will be interesting to acquire more information on the angiogenic factors found in the MSC secretome induced by omentin-1. In addition to proangiogenic factors, we also detected several main anti-inflammatory (IL-1ra and TSG-6) and proinflammatory (IL-6 and IL-8) cytokines from the MSC medium; however, there was no significant change between groups, suggesting that omentin-1 might at least have no effects on the secretion of the inflammation-associated cytokines mentioned above by MSCs.

Of note, the signaling pathways mediating the protective effects of omentin-1 are complex and never exclusive. Several studies have argued that AMPK could be activated by omentin-1 and involved in the regulation of proliferation and apoptosis [22, 23, 59]; whether this pathway also contributes to the beneficial influence of omentin-1 and its possible crosstalk with Akt require further research. Moreover, contributions of other adipokines, especially those that exert positive effects on health such as C1q/TNF-related protein 9 (CTRP9), must be elucidated in future studies.