Hypoxia-induced expression of cellular prion protein improves the therapeutic potential of mesenchymal stem cells

Mesenchymal stem cells (MSCs) are ‘adult' multipotent cells that promote regeneration of injured tissues in vivo. However, differences in oxygenation levels between normoxic culture conditions (21% oxygen) and both the MSC niche (2–8% oxygen) and ischemic injury-induced oxidative stress conditions in vivo have resulted in low efficacy of MSC therapies in both pre-clinical and clinical studies. To address this issue, we examined the effectiveness of hypoxia preconditioning (2% oxygen) for enhancing the bioactivity and tissue-regenerative potential of adipose-derived MSCs. Hypoxia preconditioning enhanced the proliferative potential of MSCs by promoting the expression of normal cellular prion protein (PrPC). In particular, hypoxia preconditioning-mediated MSC proliferation was regulated by PrPC-dependent JAK2 and STAT3 activation. In addition, hypoxia preconditioning-induced PrPC regulated superoxide dismutase and catalase activity, and inhibited oxidative stress-induced apoptosis via inactivation of cleaved caspase-3. In a murine hindlimb ischemia model, hypoxia preconditioning enhanced the survival and proliferation of transplanted MSCs, ultimately resulting in improved functional recovery of the ischemic tissue, including the ratio of blood flow perfusion, limb salvage, and neovascularization. These results suggest that Hypo-MSC offer a therapeutic strategy for accelerated neovasculogenesis in ischemic diseases, and that PrPC comprises a potential target for MSC-based therapies.

inferior differentiation, and death of grafted cells. 6,7 Notably, however, previous studies demonstrated that hypoxia-induced apoptosis can be prevented in certain cell types by preconditioning the cells, via exposure to less severe hypoxic conditions (1-3% O 2 ), for a period of time prior to exposing them to the severe ischemia at the injury site. [8][9][10] Cultivation under hypoxic conditions (1-3% O 2 ) may also be beneficial for MSCs, as this oxygen tension is similar to that present at the physiologic niche for MSCs in the bone marrow (2-7% O 2 ). Although hypoxia preconditioning induced the upregulation of hypoxia-inducible factor-1 alpha (HIF-1α), CXCR4, c-Met, and/or TWIST, it is unclear how this treatment affects selfrenewal, differentiation, and survival of various stem/progenitor cells in ischemic injured tissues.
The normal cellular isoform of prion protein (PrP C ) is a highly conserved glycoprotein that is monomeric in structure, protease-sensitive, and tethered to the cell membrane by a glycosyl phosphatidyl inositol anchor. 11 Meanwhile, the abnormal form of PrP is commonly multimeric, aggregated, and protease-resistant, and is associated with neurotoxicity and neurodegeneration. 12 Although several studies investigated the molecular mechanisms underlying the role of the abnormal form of PrP in neuronal diseases, understanding the physiological role of PrP C remains a challenge due to its numerous functions 13,14 and its roles in cellular processes such as proliferation, self-renewal, and differentiation. [15][16][17] In addition, a recent study indicated that cellular prion protein enhances neuronal survival and angioneurogenesis after post-ischemic injury. 18 However, it is unknown whether MSCs express PrP C or whether PrP C affects MSCs bioactivity, including survival and proliferation, under ischemic conditions.
In this study, we investigated whether cultivation under hypoxic conditions could be utilized to enhance the efficacy of MSCs for ischemic therapy relative to normoxic culturing. We first attempted to establish a general protocol for achieving increased MSC proliferation under hypoxic culture conditions by determining the effects of hypoxic preconditioning on MSC proliferative potential. We subsequently evaluated the ability of preconditioned cells to promote therapeutic vasculogenesis in ischemic limbs. Moreover, we investigated the mechanism by which PrP C regulation affects the proliferation and survival of MSCs subjected to hypoxic preconditioning.

Results
Characterization of MSC differentiation and assessment of MSC proliferation under hypoxic culture conditions. There was no significant difference in MSC differentiation when cultivated under normoxic and hypoxic (2% O 2 ) conditions, as determined by oil red O (adipogenesis), alkaline phosphate (osteogenesis), and safranin O (chondrogenesis) staining analyses, and fabp4 gene (adipogenesis), osteopontin gene (osteogenesis), and sox9 gene (chondrogenesis) expression did not change significantly (Figure 1a and b). To investigate their proliferative potential under hypoxic conditions, MSCs were evaluated using a single-cell expansion assay ( Figure 1c). Notably, hypoxiapreconditioned MSCs (Hypo-MSC) exhibited significantly greater levels of expansion than MSCs cultivated under normoxic conditions (Nor-MSCs; Figure 1d and e). These data suggested that hypoxia preconditioning enhances MSC proliferation.
Hypoxia preconditioning induces HIF-1α-mediated PrP C expression. To explore whether hypoxia preconditioning induces PrP C expression in MSCs, we evaluated the expression levels of HIF-1α and PrP C in hypoxiapreconditioned cells by western blot analysis. Hypoxia preconditioning resulted in a time-dependent increase in HIF-1α expression (Figure 2a). Meanwhile, maximal levels of PrP C expression were observed after hypoxic stimulation for 12 h (Figure 2a). Notably, however, this hypoxia-induced expression of PrP C was attenuated by treatment of MSCs with HIF-1α-specific siRNAs (Figure 2b). These findings suggest that hypoxia preconditioning promotes the expression of PrP C via hypoxia-mediated increases in HIF-1α expression.
PrP C regulates hMSC proliferation via the JAK2/STAT3 pathway. A previous study demonstrated that hypoxia preconditioning enhances cell proliferation through the JAK2/STAT3 signaling pathway. 19 To confirm whether hypoxic conditions promote MSC proliferation through the JAK2/ STAT3 pathway, hypoxia-induced phosphorylation of JAK2 and STAT3 were investigated by western blot analysis. Although hypoxia preconditioning for 12 h resulted in increased JAK2 and STAT3 phosphorylation, these effects were attenuated by treatment of MSCs with PrP C -specific siRNA molecules (Figure 2c and d). Likewise, we observed increased expression of cyclin D1 and c-Myc, which are encoded by STAT3 downstream genes, after hypoxic stimulation for 12 h, and decreased expression of these proteins upon siRNA-mediated knockdown of PrP C expression ( Figure 2d).
To explore the role of enhanced PrP C expression on the proliferative potential of MSCs under hypoxic conditions, single-cell expansion assays were performed ( Figure 2e). Notably, the increase in proliferation observed in Hypo-MSC, compared with that of MSCs cultivated under normoxic conditions, was inhibited by siRNA-mediated PrP C knockdown ( Figure 2f). Together, these results suggest that hypoxia preconditioning enhances MSC proliferation potential through PrP C -mediated activation of JAK2/STAT3 pathway.
Hypoxia preconditioning-mediated increases in PrP C expression provide a protective effect against oxidative stress-induced apoptosis in MSCs. To examine the effects of increased PrP C expression on ROS-scavenging, the superoxide dismutase (SOD) and catalase activity of Hyp-MSCs were evaluated. Although hypoxia preconditioning had no significant effect on SOD activity, downregulation of PrP C resulted in significantly decreased SOD activity in Hypo-MSCs compared with normoxic control group ( Figure 3a). Conversely, MSCs subjected to hypoxic preconditioning exhibited significantly increased catalase activity, compared with that of MSCs cultivated under normoxic conditions, and this enhancement was blocked by treatment with the PrP Cspecific siRNA (Figure 3b). To investigate whether PrP C provides a protective effect against oxidative stress in MSCs, we assessed the levels of apoptosis in MSCs subjected to oxidative stress conditions, including long-term hypoxia (96 h) or treatment with H 2 O 2 by western blot analysis and flow cytometry for detection of PI and Annexin V staining. In MSCs exposed to long-term hypoxia, siRNA-mediated downregulation of PrP C resulted in increased levels of phosphorylatednuclear factor kappa B (p-NF-κB) and caspase-3 cleavage (Figure 3c and d), and increased numbers of apoptotic cells (Figure 3e and f). Similar results were observed in MSCs subjected to H 2 O 2 -induced oxidative stress conditions (Figure 3g and h). These data suggest that PrP C protects MSCs against induction of oxidative stress-induced apoptosis through ROS-scavenging activity.
Hypo-MSC yield improved functional recovery in a murine hindlimb ischemia model. To determine whether MSCs subjected to hypoxia preconditioning exhibit enhanced vascular repair in vivo, blood perfusion and tissue repair were investigated following transplantation of PBS, Nor-MSCs, Hypo-MSCs, sh-PRNP Hypo-MSCs, and Scr shRNA Hypo-MSC into ischemic thigh muscles of murine hindlimbs. We used PRNP shRNA for stable PrP C downregulation in MSCs. PrP C expression was confirmed by western blotting (Figure 4a).
Blood perfusion was assessed by LDPI at postoperative days 0, 5, 10, 15, 20, and 25 ( Figure 4b). Notably, mice receiving Hypo-MSCs exhibited significantly increased blood perfusion ratios than those in the PBS, Nor-MSC, and sh-PRNP Hypo-MSC groups ( Figure 4c). Meanwhile, transplantation of Hypo-MSCs also resulted in significantly reduced levels of toe loss and foot necrosis compared with that observed in the other mouse groups (Figure 4d and e). These data indicate that transplantation of Hypo-MSC yield enhanced functional recovery in the murine hindlimb ischemia model through improvement of blood flow.
Hypoxia preconditioning enhances MSC survival, proliferation, and neovascularization in ischemic tissues. To explore whether hypoxia preconditioning augments the survival and proliferation of transplanted MSCs, ischemic tissues were harvested at postoperative day 3 and subjected to western blot analysis for p-NF-κB and cleaved caspase-3, and immunofluorescence staining for HNA, cleaved caspase-3, and Ki67 ( Figure 5). The phosphorylation of NF-κB and expression of cleaved caspase-3 was significantly decreased in tissue lysates injected with hypo-MSCs compared within other groups (Figure 5a (Figure 5f). To confirm angiogenic cytokine expression after the transplantation of MSCs, we assessed the expression of human epidermal growth factor (hEGF), human vascular endothelial growth factor (hVEGF), human fibroblast growth factor (hFGF), and human hepatocyte growth factor (hHGF) at postoperative day 3. The expression levels of hEGF, hVEGF, hFGF, and hHGF were significantly higher in ischemic tissue injected with Hypo-MSCs than in tissues from other groups (Figure 6a). To evaluate the effects of Hypo-MSC on neovascularization, we quantified the number of CD31-positive capillaries and α-SMApositive arterioles within the ischemic tissues of mice in each group at postoperative day 28 by immunofluorescence staining. There was a significantly greater density of capillaries and arterioles within the ischemic tissues of mice transplanted with Hypo-MSCs than in those transplanted with Nor-MSCs or sh-PRNP Hypo-MSCs (Figure 6b-e). These findings therefore imply that hypoxia preconditioning promotes the survival and proliferation of transplanted MSCs in ischemic tissues in vivo, thereby yielding increased transplanted MSC-mediated neovascularization.

Discussion
MSCs have important roles in the maintenance and repair of bone and blood tissues. In principle, MSCs can be readily isolated from patient tissues, including bone marrow and adipose tissues, and then expanded in culture until the large numbers required for clinical application are obtained. However, the application of MSCs for stem cell-based therapies is hindered by several limitations, including low proliferative potential, restricted life span, and gradual loss of 'stemness' during ex vivo expansion. [20][21][22] In addition, exposure to stressful conditions such as oxidative stress, ischemia, and inflammation can induce apoptosis and cell death, which has resulted in low efficacies of MSCs transplanted into injured tissues. 20,[23][24][25] Notably, although the tissues from which MSCs are isolated are associated with variable oxygen levels (1-7 % in bone marrow, 10-15% in adipose tissue, and 2-8% at the MSC niche), 26,27 MSCs are typically cultured under normoxic conditions, resulting in the induction of oxidative stress and the alteration of DNA, biomolecule structures, and metabolism. 28 To address this issue, several studies have demonstrated that hypoxia preconditioning of MSCs results in reduced apoptosis, chromosomal abnormality, and senescence, and enhanced paracrine activity and regenerative potential. 20,28,29 Although there is increasing evidence highlighting the beneficial effects of hypoxia preconditioning on MSCs, the precise mechanism through which preconditioning affects these cells remains unclear. Likewise, although there is a consensus that PrP C enhances stem cell proliferation in vitro and in vivo, 30 the molecular mechanism through which PrP C influences the proliferative potential of MSCs remains unclear. Here, we provide the first evidence that hypoxia treatment results in enhanced expression of PrP C , which in turn promotes MSC expansion via activation of the JAK2/STAT3 pathway. Previous reports revealed that STAT3 contributes to hypoxia-induced enhancement of self-renewal in embryonic and other adult stem cells. [31][32][33] Furthermore, our previous study indicated that hypoxia increases the proliferation of endothelial colony-forming cells through the activation of JAK2/STAT3 pathway. 19 These findings suggest that hypoxia-induced PrP C enhances MSC proliferation via JAK2/ STAT3 pathway activation, and that PrP C is key regulator of JAK2/STAT3 signaling in MSCs.
Importantly, we also demonstrated that maintenance of PrP C expression protected MSCs against oxidative stressinduced apoptosis by promoting increased SOD and catalase activity. Our data are therefore in agreement with previous studies showing that PrP C confers enhanced resistance to oxidative stress by contributing to cellular SOD and catalase activity, whereas attenuated PrP C expression results in enhanced sensitivity to oxidative stress. [34][35][36] Interestingly, PrP C has a pro-apoptotic role during ER stress, but an antiapoptotic role during oxidative stress-induced cell death. 37 Our data indicate that hypoxia preconditioning-mediated PrP C expression prevents oxidative stress-induced apoptosis and cell death through the regulation of SOD and catalase activity.
Although MSCs differentiate into smooth muscle, endothelial cells, or myocardial muscle in vivo, the predominant mechanism of MSC-mediated tissue repair occurs through angiogenic factor secretion. Accumulating evidence suggests that hypoxia enhances the secretion of VEGF, IGF-1, SDF-1, and EPO, indicating that hypoxic preconditioning improves the tissue repair potential of MSCs. [38][39][40] However, successful cell  is determined by several factors, including the rates of cell apoptosis and death, the extent of mechanical loss, and the rates at which the cells escape from the region of transplantation, influence the rates of cell engraftment and proliferation in ischemic sites. 1 Our current study shows that hypoxia preconditioning yields enhanced functional recovery in a murine hindlimb ischemia model via augmentation of survival, proliferation, and neovascularization of transplanted MSCs. Moreover, the beneficial effects of hypoxia preconditioning on tissue repair and vascularization appears to be mediated by regulation of PrP C expression. Consistent with these findings, overexpression of PrP C was shown to provide a protective effect against focal cerebral ischemia via the activation of ERK, whereas downregulation of PrP C increased the exacerbation of ischemic brain injury through postischemic activation of caspase-3. 41,42 Moreover, PrP C was found to promote neuroprotection, neurogenesis, and angiogenesis in ischemic brain injuries. 18

Conclusion
In this study, we investigated the efficacy of ex vivo hypoxic preconditioning for enhancing the proliferative potential of MSCs, as well as their resistance to ischemic stress-induced apoptosis. The optimized culture conditions examined here promoted robust cell survival, proliferation, and protection against oxidative stress, resulting in improved therapeutic efficacy of MSCs in a murine hindlimb ischemia model, suggesting that hypoxic preconditioning could comprise a new therapeutic strategy for overcoming ischemic diseases. In addition, our results indicate that PrP C might represent a novel target for MSC-based therapies for several ischemic diseases.

Materials and Methods
Cell cultures. Human adipose-derived MSCs were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and were pathogen-(HBV, HCV, HIV, and Syphilis) and mycoplasma-negative. The supplier certified that the MSCs exhibited expression of the cell surface markers CD73 and CD105, but not of CD31, as well as adipogenic and osteogenic differentiation potential when cultured in specific differentiation media. Cells were cultured in alpha-minimum essential medium (Thermo-Fisher Scientific, Waltham, MA, USA) supplemented with 10% (v/v) fetal bovine serum (Thermo-Fisher Scientific), 100 U/ml of penicillin, and 100 mg/ml streptomycin. Cells were growing a humidified incubator at 37°C with 5% CO 2 .
MSC differentiation. For MSC differentiation, cells were grown in StemPro adipogenic, osteogenic, or chondrogenic culture medium (Thermo-Fisher Scientific). To detect adipocytes, cells were cultivated in adipogenic medium for 21 days and stained with oil red O (Sigma-Aldrich, St. Louis, MO, USA) for 15 min. To detect osteoblasts, cells were grown in osteogenic medium for 21 days and subjected to alkaline phosphatase staining with an alkaline phosphatase kit (Sigma-Aldrich). Chondrocytes were detected by staining cells grown in chondrogenic medium for 14 days with safranin O (Sigma-Aldrich).
Single-cell cultivation assay. A limited-dilution assay was used to aliquot single MSCs into individual wells of 96-well culture plates. In brief, cell suspensions containing 1 × 10 3 cells in 10 ml complete medium were diluted 1:10 (cells: complete medium), and 100 μl of the diluted sample (approximately 1 cell/100 μl) was seeded into 96-well plates. Cells were then cultured under normoxic or hypoxic conditions in a humidified incubator, and each well was examined for MSC grow that day 10.
Hypoxia preconditioning. MSCs were incubated in a modular incubator chamber (IB Science, Daejeon, Korea) maintaining a hypoxic gas mixture composed of a 2% O 2 , 5% CO 2 , and balanced N 2 for 0, 6, 12, and 24 h at 37°C. siRNA transfection. hMSCs were grown to 70% confluence in culture dishes and were transfected for 48 h with SMART pool siRNAs (100 nM/l) specific to HIF-1α or PrP C mRNA using Lipofectamine 2000 reagent (Thermo-Fisher Scientific), according to the manufacturer's protocols.
SOD activity assay. The protein concentration of cells lysate was determined using bicinchoninic acid assay (Sigma). The SOD activity assay is based on water soluble tetrazolium reduction by superoxide anion (produced from xanthine by xanthine oxidase) to a colored water soluble tetrazolium formazan product and was performed the total SOD activity measurement at 480 nm.
Catalase activity assay. Catalase activity was based on the rate of degradation of H 2 O 2 .The cells were removed from the culture dish by scraping with a policeman, and cells lysate protein extracts (40 μg) reaction sample contained the indicated quantities of catalase and 20 mM H 2 O 2 in 0.1 M Tris-HCl, and incubated for 30 min, and 50 mM Amplex red reagent and 0.2 U/ml of horseradish peroxidase was added and incubated for 30 min at 37°C. The decomposition of the substrate was recorded by the decrease in absorbance at 563 nm.
PI/Annexin V flow cytometry analysis. To examine the levels of cell apoptosis, cells were stained with Annexin V-FITC and propidium iodide (PI) (De Novo Software, Los Angeles, CA, USA) and evaluated using a Cyflow Cube 8 kit (Partec, Münster, Germany). Data were analyzed using standard FSC Express software (De Novo Software).
PrP C knockdown by shRNA. The lentiviral PRNP shRNA constructs were obtained from OriGene Technologies (Rockville, MD, USA). At 24 h after cell seeding, 293T cells (1.5 × 10 6 ) were transfected with a lentiviral vector encoding PRNP shRNA (3 μg) and packaging vector pGFP-C-shLenti (0.3 μg; OriGene Technologies) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Fortyeight hours after transfection, the supernatant harboring lentiviruses was collected and filtered using a sterilized 0.45-μm syringe filter. The PRNP shRNA lentiviral supernatant with polybrene (5 μg/ml) was added to the hMSCs for infection. After selection in growth medium containing puromycin (2 μg/ml) for 2 weeks, individual resistant cells were expanded in growth media without puromycin, and protein expression was evaluated via western blot analysis. PRNP shRNA No. 3 showed the most effective PrP C knockdown among the four PRNP shRNA clones (No.1-4). The corresponding knockdown clone was used for further studies. The PRNP shRNA No. 3 sequence was 5′-AACCAACATGAAGCACATGGCTGGTGCTG-3′.
Cell transplantation in a murine hindlimb ischemia model. Cell transplantation experiments were performed on 8-week-old nude male BALB/c mice (Biogenomics, Seoul, Korea) maintained under a 12-h light/dark cycle, and in accordance with the regulations of Soonchunhyang University, Seoul Hospital. All animal procedures were approved by the Institutional Animal Care and Use Committee of Soonchunhyang University, Seoul Hospital, Korea (IACUC2013-5). To induce ischemia and oxidative stress and to assess neovascularization, we utilized a hindlimb ischemia model, as previously described but with minor modifications. 43,44 Ischemia was induced by ligation of the proximal femoral artery