Bone mesenchymal stem cells pretreated with erythropoietin accelerate the repair of acute kidney injury

Background Mesenchymal stem cells (MSCs) represent a promising treatment option for acute kidney injury (AKI).The main drawbacks of MSC therapy including the lack of specific homing following systemic infusion and early death of the cells in the inflammatory microenvironment, directly affect the therapeutic efficacy of MSCs. Erythropoietin (EPO)-preconditioning promotes the therapeutic effect of the MSCs, although the underlying mechanism remains unknown. In this study, we sought to investigate the efficacy and mechanism of EPO on bone marrow mesenchymal stem cells (BMSCs) for the treatment of AKI. Results We found that incubation of BMSCs with ischemia/reperfusion(I/R)-induced AKI kidney homogenate supernatant (KHS) caused apoptosis in the BMSCs, which was decreased following EPO pretreatment indicating that EPO protected the cells from apoptosis. Further, we found that EPO upregulated SIRT1 and Bcl-2 expression, and downregulated p53 expression. The EPO-mediated anti-apoptotic mechanism in pretreated BMSCs may be mediated though the SIRT1 pathway. In a rat AKI model, our data showed that 24 h following intravenous infusion, GFP-BMSCs were predominantly in the lungs. However, EPO pretreatment reduced the lung entrapment of BMSCs, and increased the distribution of the BMSCs to the target organs. AKI rats infused with EPO-BMSCs had significantly lower levels of serum IL-1β and TNF-a and significantly higher level of IL-10 compared to rats infused with BMSCs. The administration of EPO-BMSCs after reperfusion was more effective in reducing serum creatinine, blood urea nitrogen, and pathological scores in the I/R-AKI rats than BMSCs. Conclusions Our data suggest that EPO pretreatment enhances the efficacy of BMSCs in improving renal function and pathological presentation in I/R-AKI rats. we reveal several novel findings regarding the efficacy and mechanism of EPO-BMSCs in I/R-AKI. Firstly, the I/R-AKI microenvironment clearly caused apoptosis in the BMSCs, but EPO pretreatment protected the cells against this apoptotic effect. We found that EPO upregulated SIRT1 and Bcl-2 expression, and downregulated p53 expression. The anti-apoptotic mechanism

resulted in apoptosis of the BMSCs, but EPO pretreatment protected the cells against this apoptotic effect. The apoptotic rate in EPO-BMSCs/AKI-KHS group was significantly lower compared to that in the BMSCs/AKI-KHS group (Fig. 2B).

Western blot analysis
Results of the flow cytometric analysis demonstrated that AKI-KHS treatment had an adverse effect on the survival of BMSCs while the EPO pretreatment improved BMSCs survival in the AKI microenvironment. To further investigate the anti-apoptotic mechanism of EPO pretreatment of the BMSCs, we analyzed the expression of the anti-apoptotic factor Bcl-2, apoptotic factor p53, and silent information regulator 1 (SIRT1). All cytokines were examined using western blot analyses (Fig. 2C).

SIRT1 and Bcl-2 protein expression in both BMSCs/AKI-KHS and EPO-BMSCs/AKI-KHS groups was
significantly increased compared to that in the control group (p < 0.05), especially in the EPO-BMSCs/AKI-KHS group (p < 0.05). The expression of p53 protein was significantly higher in the BMSCs/AKI-KHS group compared to the other two groups (p < 0.05, Fig. 2D).

Treatment of AKI in rats Blood biochemical indicators
To compare the effect of the treatment with BMSCs and EPO-BMSCs in AKI rats, blood urea nitrogen (BUN) and serum creatinine (SCr) were measured on days 1 and 5 after treatment. There was a significant increase in BUN and SCr levels on days 1 and 5 after the induction of AKI. Treatment of AKI rats with EPO, BMSCs, and EPO-BMSCs showed varying therapeutic effects. All three treatments reduced the level of BUN and SCr on days 1 and 5 compared to the control group treated with the vehicle, however only BMSCs and EPO-BMSCs treatment showed significant effect (p < 0.05). Rats in the EPO-BMSCs group had significantly lower BUN and SCr levels compared to those in the BMSCs group (p < 0.0, Fig. 3A and B).

Detection of cytokines
Twenty-four hours after the treatment of the AKI rats with vehicle, EPO, BMSCs and EPO-BMSCs, the level of serum proinflammatory cytokines IL-1β and TNF-α, and that of the anti-inflammatory cytokine, IL-10, were tested using enzyme-linked immunosorbent assay (ELISA). There was a significant increase in the level of IL-1β, TNF-α, and IL-10 on day 1 after the induction of AKI. The level of the cytokines in the EPO group was not significantly different compared to the model group. Infusion of BMSCs significantly reduced serum IL-1β (Fig. 3C), and TNF-α ( Fig. 3D) level, while it increased serum IL-10 level (Fig. 3E). Furthermore, the AKI rats that underwent EPO-BMSCs infusion had significantly lower serum IL-1β and TNF-α levels, and significantly higher IL-10 levels compared to the rats that received BMSCs (Fig. 3C-E).

GFP fluorescence in frozen sections
Our data showed that 24 h following the intravenous infusion, majority of the GFP-BMSCs were trapped inside the lungs, with minimal fluorescence detected in the spleen, liver, and kidneys in the BMSCs group. However, in the EPO-BMSCs group, the fluorescence intensity was significantly lower in the lungs, while the other organs, especially the kidneys showed increased number of fluorescent cells. In both the groups, GFP-BMSCs were found predominantly in the lungs, and EPO preconditioning reduced the lung entrapment of the BMSCs, and increased the distribution of the fluorescent BMSCs to the target organs (Fig. 4).

Histological analysis
Hematoxylin and eosin (HE) staining of kidney sections revealed tubular necrosis, cast formation, tubular dilation, and loss of brush border in the model group (Fig. 5A). These parameters were assessed using the pathological scores following treatment with EPO, BMSCs, and EPO-BMSCs on days 1 and 5. The pathological scores in the BMSCs and EPO-BMSCs treatment groups were significantly lower than that in the model group on days 1 and 5 (p < 0.05). Both BMSCs and EPO-BMSCs infusion reduced tubular injury. EPO-BMSCs administration led to significantly better improvement of the tubular injury on day 5 compared to infusion with BMSCs (Fig. 5B).

Discussion
In the present study, we reveal several novel findings regarding the efficacy and mechanism of EPO-BMSCs in I/R-AKI. Firstly, the I/R-AKI microenvironment clearly caused apoptosis in the BMSCs, but EPO pretreatment protected the cells against this apoptotic effect. We found that EPO upregulated SIRT1 and Bcl-2 expression, and downregulated p53 expression. The anti-apoptotic mechanism mediated by EPO pretreatment of BMSCs may occur though the SIRT1 pathway. Secondly, EPO pretreatment reduced pulmonary entrapment, and increased the number of cells reaching the target organ. Furthermore, EPO pretreatment reduced the expression of disease-related inflammatory cytokines. The AKI rats that underwent EPO-BMSCs infusion had significant lower serum IL-1β and TNF-α levels, and significant higher IL-10 levels compared to the rats that received BMSCs. Our data suggest that EPO pretreatment enhanced the efficacy of BMSCs in the improving the renal function and pathological presentation of I/R-AKI rats.
The protective effect of MSCs in acute and chronic renal injury may be through a paracrine/autocrine mechanism, which is related to immune regulation, anti-apoptosis, and reduction of disease-related inflammation. Several studies have shown that the survival and retention of MSCs in target organs or tissues is closely related to the therapeutic effect mediated by MSCs. The main drawbacks of MSCs therapy are the early pulmonary entrapment and the lack of specific homing after systemic infusion.
Most of the MSCs undergo cell death after transplantation, mainly due to local adverse microenvironment, and this directly affects the therapeutic efficacy of the MSCs [14,15].
Increasing the dose of the MSCs infusion represents a viable option to achieve better therapeutic effect. However, this has some unfavorable side effects, such as microvascular embolization and potential risk of long-term tumor differentiation. For these reasons, current studies on MSCs-related therapies have focused on minimal infusion amounts to achieve the maximum therapeutic effect possible. Cell-free treatments including the microvesicles, exosomes, specific cytokines, miRNA, and the pretreatment of MSCs represent viable alternatives to address the issue of long-term negative effects [25][26][27]. Application of pretreated-MSCs is a novel strategy to enhance the capacity of MSCs to migrate and promote tissue repair in the treatment of kidney injury [28]. Pretreatment with cytokines such as transforming growth factor β1 (TGF-β1), interleukin-17A (IL-17A), and melatonin before infusion can increase the number of MSCs that home to the injured kidney, promote the recovery of renal function, and ameliorate impairments to the renal structure [3,17,29]. Similar result was obtained by pretreatment of MSCs with EPO.
EPO is a glycoprotein hormone that promotes the proliferation and differentiation of bone marrow marrow HSCs are homologous and express the EpoR. Many other cell types, including neurons, endothelial cells, cardiomyocytes, and renal tubular cells also express the EpoR and respond to EPO treatment [30]. Studies have shown that EPO is suitable for the treatment of a variety of diseases, including cerebral ischemia, myocardial infarction, and chronic congestive heart failure, as well as renal injury [20,31]. Moreover, few studies suggest that overexpression of EPO by gene-transfection in MSCs could further enhance the protective effect [32,33]. Compared to other pretreatments or improvement strategies aimed at improving the therapeutic effect of MSCs, in vitro pretreatment with EPO has significant advantages in terms of clinical feasibility. Firstly, EPO is a commonly used therapeutic drug with few side effects and is widely used in the clinical treatment of anemia, especially in patients with CKD. Secondly, it has anti-oxidative and anti-inflammatory effects.
Our data showed that 24 h following intravenous infusion, GFP-BMSCs were found predominantly in the lungs. EPO preconditioning reduced the lung entrapment of BMSCs, and increased the distribution of the fluorescent BMSCs to the target organs. In our previous study, we showed that pretreatment of BMSCs with an optimal concentration of EPO for an appropriate time induced a marked change in their proliferation rate and cytoskeletal rearrangement. After incubation with EPO, most of the cells exhibited parallel-oriented filaments organized along the axis of the cells. We observed that CXCR4 which was a pivotal mediator of migration and engraftment of MSCs was upregulated following EPO treatment. These changes enhanced the migration ability of BMSCs [18].
Homing of MSCs to the injured tissues is very critical in cell therapy. There are various methods of MSCs infusion, such as peripheral intravenous infusion, arterial infusion of target organs, and local injection. Local injection increase risks and side effects such as bleeding and tissue injury, while direct arterial administration can result in occlusion and embolization. Following these observations, MSCs are mostly administered through a standard intravenous route. Pulmonary cell entrapment is a major problem after intravenous infusion. Harting et. al. demonstrated that less than 4% of the infused cells were likely to traverse the pulmonary microvasculature and reach the arterial circulation, a phenomenon termed the "pulmonary first-pass effect", which limits the efficacy of this therapeutic approach [34]. Some studies showed that smaller microspheres (4-5 µm) can pass through the pulmonary system while the majority of the 20 µm microspheres and MSCs (15-19 µm) were trapped inside the lungs. Lung entrapment may be due to the small capillary size, large capillary network, and strong adhesion properties of the MSCs [35][36][37]. A variety of molecules may have a role in the lung entrapment of systemically infused cells and the composition of the cell surface molecules like α4, α5, and α6 integrins likely affect the migratory behavior of the therapeutic cells [13,37]. We found that following intravenous infusion, most of the stem cells were trapped inside the lungs, but EPO pretreatment reduced pulmonary entrapment, and increased the number of cells reaching the target organ.
We also found that EPO upregulated SIRT1 and Bcl-2 expression, and downregulated p53 expression. SIRT1 is NAD+-dependent deacetylase belonging to the class Ⅲ histone deacetylases, and is known as the longevity protein in mammals. SIRT1 regulates a variety of cellular signaling pathways by modifying the acetylation status of target proteins, including p53, members of the forkhead family of transcription factors (FOXO), nuclear factor NF-κB, among others. Following activation, it participates in various cellular processes such as cell senescence, apoptosis, DNA damage repair, cell cycle, antioxidative stress, energy metabolism regulation, tumor generation, and other physiological and pathological processes [38][39][40]. P53 plays a vital role in the apoptotic signaling pathways, including membrane apoptotic signaling and mitochondrial apoptotic pathways, and affects the transcription and expression of many apoptosis-related cytokines in the nucleus [41]. SIRT1 reduces the transcriptional activity of p53, and blocks p53-dependent cell apoptosis caused by DNA damage [42].
In tumor studies, SIRT1 inhibits the apoptosis of tumor cells by regulating p53 and Bcl-2 [43]. EPO has been demonstrated to protect against chemotherapy drug doxorubicin (DOX)-induced cardiotoxicity by activating SIRT1 to enhance mitochondrial function [20]. Hong et. al. revealed a novel mechanism of alleviation of hepatic steatosis by EPO by activation of autophagy through SIRT1-dependent deacetylation of LC3 in the treatment of hepatic steatosis [38]. The data from the present study demonstrated that the anti-apoptotic effect following EPO pretreatment of BMSCs may be mediated though the SIRT1 pathway.
The mechanism through which EPO-pretreated BMSCs accelerate the repair of AKI involves three facets. Firstly, after incubation with EPO, most of the BMSCs exhibit parallelly-oriented filaments organized along the axis of the cells, and showed increased CXCR4 expression. These changes reduce the lung entrapment of the BMSCs, and increase the homing of the BMSCs to the target organs.
Secondly, EPO upregulated SIRT1 and Bcl-2 expression, and downregulated p53 expression in the BMSCs. SIRT1 inhibited the apoptosis of BMSCs cells by regulating p53 and Bcl-2. Thirdly, the direct anti-inflammatory effect of EPO-BMSCs is also likely to be involved in the process. However, there are some limitations in the present study. Our study did not elucidate the mechanism underlying the EPOmediated activation of SIRT1 signaling in BMSCs, which needs future investigations.

Conclusion
In conclusion, the administration of EPO-BMSCs was more effective in the reduction of SCr, BUN, and pathological scores in I/R-AKI rats after reperfusion than untreated BMSCs. These results suggest that EPO pretreatment may be a potential novel alternative to untreated BMSCs for the management of AKI.  Anti-apoptotic factor Bcl-2, apoptotic factor p53, and SIRT1 were examined using western blot analyses. *p < 0.05 versus BMSCs+N-KHS; #p < 0.05 versus BMSCs+AKI-KHS; Figure 3 To compare the effect of the treatment with BMSCs and EPO-BMSCs in AKI rats, SCr(A) and BUN(B) were measured on days 1 and 5 after treatment. Twenty-four hours after the treatment of the AKI rats with vehicle, EPO, BMSCs and EPO-BMSCs, the level of serum proinflammatory cytokines IL-1β(C), and TNF-α(D), and that of the anti-inflammatory cytokine, IL-10(E), were tested using ELISA.*p < 0.05 versus Model group; #p < 0.05 versus BMSCs group;