Preclinical Evaluation of Bone Marrow Mesenchymal Stem Cells Overexpressing MicroRNA-126 to Treat Critical Limb Ischemia

Background: Critical limb ischemia (CLI) considered as the most severe form of peripheral artery disease (PAD), and characterized by ischemic rest pain and non-healing ulcers. CLI being associated with a high risk of major amputation, cardiovascular events and death. Current therapy for CLI are surgical reconstruction and endovascular therapy or limb amputation (for patients with no treatment options). Neovasculogenesis induced by stem cells including mesenchymal stem cells (MSCs) therapy is a promising approach to treat CLI. But this method of treatment faces challenges that reduce its effectiveness, such as: MSCs survival and paracrine secretion. MicroRNAs are post transcriptional regulatory molecules that regulate many biological processes including VEGF pathway. Therefore, microRNAs could be used to increase viability and angiogenic potential of MSCs. This study was conducted to reinforce and increase the angiogenic potential of BM-MSCs by using microRNA-126 and evaluate the effect of this stem cell gene therapy on treatment of ischemic tissues in CLI mouse models. Results: Our cellular, molecular and functional tests indicated that during 28 days after transplantation, BM-MSCs and virus groups had an enhancing effect on angiogenesis. BM-MSCs miR-126 group had remarkable effect on endothelial cell migration, muscle restructure, functional improvements and neovascularization in ischemic tissues and led to more effective treatment. On the other hand, miR-126 could increase paracrine secretion of BM-MSCs. Additionally, in vivo evaluation showed that miR-126 could increase BM-MSCs survival and paracrine secretion of angiogenic factors such as VEGF, and led to remarkable functional improvements and neovascularization in ischemic tissues. Conclusion: According to the obtained results, it could concluded that combination of BM-MSCs and miR-126 leads to more effective recovery from critical limb ischemia compared to using them alone. In fact, miR-126 can be used as a strong modier to reinforce the angiogenic potential of BM-MSCs, leading to more effective treatment for CLI. Chromogen Chromogen DAB substrate distilled After dehydration CLI: Critical limb ischemia; PAD: Peripheral arterial disease; SCT: Stem cell therapy; BMMSCs: Bone marrow mesenchymal stem cells; MSCs: Mesenchymal stem cells; MiRNA: MicroRNA; VEGF: Vascular endothelial growth factor; SPRED1: Sprout-related EVH1 domain-containing protein 1; PIK3R2: Phosphoinositol-3 kinase regulatory subunit 2; LVs: Lentiviral vectors; GT: gene therapy; HIF-1: hypoxia inducible factor; SDF-1: Stromal Derived Factor-1; MMPs: matrix metalloproteinase; HUVECs: human vein endothelial cells.


Background
Critical limb ischemia (CLI) is the most severe manifestation of peripheral arterial disease (PAD). CLI is the result of PAD with spoiled blood flow in peripheral tissues (1,2). The rst-line treatment is revascularization. Revascularization may be surgical via bypass or endovascular techniques. In the absence of revascularization, up to 40% of patients will require lower limb amputation by 1 year (3). Also, nearly half of patients are not suitable for revascularization procedures because of undesirable endovascular anatomy and high operative risk (4,5). Today, the use of stem cell therapy (SCT) and gene therapy (GT) in order to stimulate angiogenesis emerged as a promising treatment for disorders related to limb ischemia. The concept of "therapeutic angiogenesis" for CLI, performed by gene, protein, and cell therapy, has recently raised an excellent deal of hope for CLI patients who cannot undergo standard treatment (6,7). In fact the new proposal for CLI treatment is using growth factors to stimulate formation of new vessels and/or remodeling of dysfunctional vessels. These factors, which are proteins, can be administered in their protein form, or through genes or cells that express these factors and the treatment modality is known as therapeutic angiogenesis (6,7).
Gene therapy (GT) represents a promising alternative treatment for CLI. In fact, expression of speci c angiogenic growth factors is induced in the ischemic tissue to achieve a more sustained local delivery (8). The genes usually used for angiogenic therapy include the genes for vascular endothelium growth factor (VEGF), broblast growth factor (FGF), hepatocyte growth factor (HGF), Stromal Derived Factor-1 (SDF-1), and hypoxia inducible factor (HIF-1) (9). Gene therapy faces with some important challenges, including: 1) delivery route 2) duration of gene expression 3) complexity of angiogenesis process: in fact, formation of new blood vessels is a complex process that involving proliferation and differentiation of precursor cells under the control of many regulatory molecules. Thus, using a single factor that speci cally acts on endothelial cells is possibly insu cient to form a mature vessel in an ischemic and inflamed environment (10,11).
Stem cell therapy (SCT) shows a grate promising alternative treatment for CLI. Whereas SCs play an important role in neovascularization process, the therapy involving these cells can theoretically be more e cient compared with the protein or gene therapy due to their direct self-renewal and differentiation into an organ-speci c cell, and also to their paracrine effect (6,12,13). Numerous trials have tested the administration of mesenchymal stem cells (MSCs) for treatment of CLI. The characteristics that make these cells ideal for this application are, ease of isolation and expansion, low immunogenicity in allogeneic settings, differentiation to speci c cells, and ability to secrete paracrine factors that stimulate regeneration, demonstrated safety, and tropism toward sites of hypoxia (14)(15)(16). While MSCs therapy have shown great results in animal models, the nite success in human clinical trials indicates that more potent treatment strategies or cell formulations are needed. In fact, by combining cell and gene therapy, MSCs that are genetically modi ed to over express angiogenic genes could be an optimal approach, for treatment of CLI (11,17).
Investigations about stem cell therapy for CLI focus on the use of bone marrow mesenchymal stem cells (BMSCs), and support by various clinical evidences (18,19). Mesenchymal stem cells (MSCs) consist of nearly 0.001% to 0.08% of cells within bone marrow and are nonhematopoietic stromal cells be able to differentiate into mesenchymal lineages, such as muscle, bone, cartilage and fat. They can be cultureexpanded easily to harvest a large number of cells. In addition these cells can be stored for both autologous and allogeneic use (20). In multiple preclinical studies, the therapeutic angiogenesis of BMSCs were proved and also there are strong evidences that BMSCs have the potential to produce growth factors, angiogenesis-related cytokines, chemokines and extracellular matrix molecules, and also have similar cellular and molecular properties with pericytes (14,21).
Although, stem cell therapy (SCT) faces with some important challenges, including: 1) identify the most ideal cell type 2) standardization of procedures of MSC isolation and expansion 3) in vivo administration route 4) dosage optimization 5) survival of the cells 6) administration timing (22,23). MicroRNAs (miRNAs) are a class of small (~18-22 bp), endogenous single-stranded and non-coding RNA molecules with conserved structures. They constitute nearly 1% to 2% of the eukaryotic transcriptome. MicroRNAs have signi cant roles as posttranscriptional gene regulators in different cellular processes such as, cell proliferation, apoptosis, differentiation of cells including endothelial cells (ECs) and stem cells, metabolism, tissue homeostasis, embryonic development and even tumor formation (24,25). The biologic roles of miRNAs in vascular diseases have led to an increased interest in miRNA regulation as a therapeutic and diagnostic approach. Recent studies demonstrated that miRNAs show promising potential in therapeutic interventions directed toward ischemic diseases (26,27).
MicroRNA-126 is a major positive regulator of vascular integrity and angiogenic signaling in vivo, in fact it has a distinguished role in the control of vascular integrity and angiogenesis through distribution of angiogenic factors, such as vascular endothelial growth factor (VEGF), Hypoxia Inducible Factor 1-alpha (HIF-1α), and matrix metalloproteinase (MMPs) (28). As a matter of fact one of the most vigorous microRNA functions on angiogenesis is revealed by microRNA-126 (29,30). MicroRNA-126-3p and microRNA-126-5p are the two mature strands of the pre-microRNA-126 that have cell-type and strandspeci c function in angiogenesis. MicroRNA-126 regulates the response of ECs to angiogenic factors such as VEGF, through directly repressing negative regulators of VEGF pathway, including the Sproutrelated EVH1 domain-containing protein 1 (SPRED1) and phosphoinositol-3 kinase regulatory subunit 2 (PIK3R2/p85-b), miR-126 regulates SPRED1 and PIK3R2 by directly targeting the 3-UTR (31,32).
The ideal angiogenic treatment for limb ischemia would be one that can act on all or a majority of the cells and molecules that participate in angiogenesis and control of inflammation. Some ways to achieve this ideal treatment are: 1. using more angiogenic genes 2. using genetically modi ed stem cells (9,11).
The purpose of this study was to reinforce the angiogenic potential of Bone Marrow Mesenchymal Stem Cells (BM-MSCs) by using miR-126. According to the previous studies new therapeutic ways to treat CLI including gene therapy and stem cell therapy, facing with some critical challenges that cause these methods do not have acceptable results in treating CLI. On the other hand, using these methods alone will not show the desired results. It seems that due to the gene regulatory potential of miR-126 and angiogenic potential and also its effects on BM-MSCs functions and survival, the use of miR-126 and BM-MSCs as a combination therapy will have better results in CLI treatment.

MicroRNA design
In order to evaluate the angiogenic potential of miR-126, shRNA-lentiviral vector construct was made. At rst step shRNA was designed. In the present study, annealed oligos method was used. However, some changes were made in the shRNA design according to the purpose of the study. The entire procedure involves the following steps: 1. The sequence of miR-126-3p and miR-126-5p was determined according to sequences from http://www.mirbase.org/. The 3p arm of shRNA is known as guide strand and the 5p arm is known as passenger strand. The shRNA sequence (from 5′ to 3′) will be in the order of passenger strand, loop, and then guide strand.
Loop sequence: CTCGAG 3. In order to clone the shRNA to pLKO.1-TRC vector the complementary sequences of restriction endonuclease Age1 and EcoR1 enzymes were designed at the 5´ ends of the oligonucleotides.

BM-MSCs isolation and culture
To isolate bone marrow, 6-week-old C57BL/6 mice were killed by cervical dislocation, then the ends of tibia and femur bones were cut to expose the marrow. 5-ml syringe containing complete media used to extract the cells via ushing the marrow plug out of the cut end of the bone with 1 ml of complete media and collect in a 15-ml tube. Strong ushing is necessary during marrow cell preparation. Then the cell pellet derived from 2 tibia and 2 femur bones was suspended in growth medium containing Dulbecco's modi ed Eagle's medium (DMEM) high glucose (Gibco, USA) and 10% fetal bovine serum (FBS; Gibco, USA) with 100 U/mL penicillin-streptomycin (Gibco, USA) and cultured in a 75-cm2 culture ask and were maintained at 37C and 5% CO2. Nonadherent cells were removed after 24 h and the ask was washed with phosphate-buffered saline (PBS; Gibco, USA). The medium was changed regularly every 3-4 days, and at approximately 70% con uence, the cells were detached using trypsin-EDTA (Gibco, USA) and transferred to new asks, these cells were considered as passage 1 (Figure 1).

Differentiation of BM-MSCs
Osteogenesis: for differentiation to osteocyte, passage 3 BM-MSCs were incubated to differentiate into osteoblasts in corresponding induction medium for 2 weeks. The cells were maintained in control and osteogenic media. The control medium consisted of DMEM (Gibco, USA), supplemented with 10% FBS (Gibco, USA), 1% penicillin/streptomycin (Gibco, USA). The osteogenic medium contained DMEM, 15% FBS (Gibco, USA), 100 µM L-ascorbic acid, 10 mM glycerol 3-phosphate, and 100 nM dexamethasone (Sigma-Aldrich). Medium was replaced 2 times a week. After 2 weeks the cultured cells were stained with Alizarin Red solution to visualize calcium deposits. The cells were xed in 4% paraformaldehyde (PFA) for 20 minutes and stained using Alizarin Red solution for 20 minutes at room temperature.

Transduction of BM-MSCs and puromycin reporter gene detection
Cells were seeded in six-well plates at 5 × 104 cells per well and different concentrations of virus were added and incubated at 37°C for 72 h. The volume of lentivirus to be used was determined from titer results. After incubation, 1.5 mL of fresh DMEM complete containing 2µg puromycin (Sigma) was added to each well. It was the beginning of the selection process. After 2 days, the culture medium was removed from each well and fresh medium was added without puromycin. The resistant cells (stable cell line) were cultured and expanded into T75 ask.

Cell migration assay (wound healing assay)
Wound-healing assay on endothelial cells, human umbilical vein endothelial cell (HUVECs), was used to study the paracrine effect of BM-MSCs and BM-MSCs miR-126 on endothelial cell migration potential. After 24-h FBS starvation of MSCs, the supernatants from cells were collected. The wound-healing assay was employed by scratching the monolayer HUVEC con uent cell cultures using a sterile plastic micropipette tip. The cells were washed by DMEM and PBS twice for smoothing the edges of the scratch and removing oating cells. After that, wound healing initiated by adding the collected supernatants from BM-MSCs and BM-MSCs miR-126 on HUVEC cultures in different wells of a 6-well plate. After incubating the cells at 37 °C in 5% CO2 for 0, 15, 18, 20, 22 and 24 h, the migration images of the cells were observed under a microscope with 10x (Nikon, ECLIPSE, TS100), the distance of cell migration was measured and images taken using image-J Analysis software.

Mouse Critical limb Ischemia model
All animal experiments were performed according to the guidelines approved by the Shiraz University of Medical Sciences ethics committee (1397.430).
Male C57BL/6 mice, weighing 25-30 g (6-8 week-old) purchased from Comparative and Experimental Medicine center at Shiraz University of Medical Sciences and were selected randomly. Animals were kept in standard conditions (12 h light and 12 h darkness, temperature of 18-22 °C, and 55 ± 5% humidity).
They were fed standard mouse diet and given water ad libitum.
For creating critical limb ischemia (CLI) model, all animals in each group were anesthetized intraperitoneal (IP) administration using ketamine 10% (100 mg/kg, Alfasan CO., Netherlands) and xylazine 2% (5 mg/kg, Alfasan CO., Netherlands). Animals were placed in dorsal recumbency. A skin incision (about 10 mm) was done along the center of the medial thigh from the abdomen towards the knee. Subcutaneous fat tissue super cial was gently pushed away to exposed femoral neurovascular bundle. The femoral artery pushed away from the femoral vein and nerve. Subsequently, the femoral artery was isolated by blunt dissection from the femoral vein at the ligation sites between the proximal caudal femoral artery and the bifurcation of the deep femoral artery and saphenous artery. Then two sutures were passed using 6-0 silk, transected and cauterized. The incision was closed using continues 5-0 vicryl sutures. The operation was done under a surgical microscope (Zeiss OP-MI6 SD; Carl Zeiss, Goettingen, Germany) ( Figure 2).

Animal groups and transplantation
The male mice were randomly divided into the following four groups (n = 12/group): (1) control group: CLI was performed and received PBS into ischemic site, (2) BM-MSCs group: CLI was performed and received BM-MSCs, (3) virus-miR-126 (recombinant vector) group: CLI was performed and received recombinant vector, (4) BM-MSCs miR-126 group: CLI was performed and received BM-MSCs that transfected with virus-miR-126. To transfer the cells and recombinant vector into ischemic leg, intramuscularly (IM) injection of 5×10 5 BM-MSCs and BM-MSCs miR-126 that resuspended in PBS and 4.7×10 6 virus particles as treatment groups and PBS as a control group were done at four different sites of gastrocnemicus (GC) muscle 24 hours after surgery.
Additionally to identify the presence of grafted BM-MSCs and BM-MSCs miR-126 in the gastrocnemicus (GC) muscle the fth and sixth groups were added as, survival groups: CLI was performed on female mice (n = 5/group) and received intramuscularly (IM) injection of 5×10 5 male BM-MSCs and BM-MSCs miR-126 at four different sites of gastrocnemicus (GC) muscle 24 hours after surgery.

Functional scoring
Functional grading was performed according to the Tarlov, ischemia, modi ed ischemia, function and the grade of limb necrosis scoring system at pre-operation, post-operation and at days 3, 7, 14, 21 and 28 after transplantation. The degree of ischemic damage was evaluated through indicators, such as skin color changes, swelling, and grade of the limb necrosis ( Table 1).

Investigation of donor cell survival
To identify the presence of grafted male cells in the female gastrocnemicus (GC) muscle (the survival groups), the Y chromosome-speci c SRY gene was assessed by polymerase chain reaction. At days 3, 7, 14, 21, and 28 after treatment, the GC muscle was collected, genomic DNA was extracted using Trizol (Sigma-Aldrich) and then SRY analysis was performed by PCR. The forward primer sequence was 5′-TTTATGGTGTGGTCCCGTGGTGAG-3′ and the reverse primer sequence was 5′-TTGGAGTACAGGTGTGCAGCTCTAC-3′.

RNA isolation and quantitative real-time PCR
In order to evaluate the expression of target genes (VEGF, SPRED1, PIK3R2 and GAPDH as housekeeping gene), after transplantation at days 7, 14, 21 and 28, the GC muscle was collected, total RNA was extracted using Trizol (Sigma-Aldrich T9424) according to the manufacturer's protocol. The concentration of total RNA was quanti ed by the absorbance at 260 nm. Extracted RNA was then reverse-transcribed using the SMOBIO cDNA Synthesis Kit RP1300. Then gene expression analysis was performed by Realtime PCR. qRT-PCR was performed on an ABI Prism 7500 PCR system using RealQ Plus 2X Master Mix Green Low Rox (Amplicon). PCR was performed in a total reaction volume of 15 ml consisting of appropriate cDNA. For semi-quantitative RT-PCR, the following primers were used ( Table 2).
The two-step reaction was performed triplet for all samples. The PCR was performed under the following cycling conditions: denaturation at 95 °C for 15 min, followed by 40 cycles of 30 sec at 95°C and 30 sec at 62°C (annealing), and extension at 72 °C for 30 sec. Data were analyzed by the comparative cycle threshold (Ct) method and normalized against GAPDH controls.

Histopathological evaluation
Histological analysis of limb tissues were done at days 7, 14, 21 and 28 after transplantation. To evaluate that the target cells (BM-MSCs and BM-MSCs miR-126 ) and recombinant vector can induce angiogenesis in the gastrocnemius muscle, immunohistologic staining was performed with an anti-CD31 antibody (Biocare Medical), a marker for vascular endothelial cells. The mice were euthanized at predetermined times and gastrocnemius was removed and xed in 10% formalin. After xation, the tissue was embedded in para n, and the tissue sections were prepared and mounted on slides. Then tissue section slides were stained with hematoxylin-eosin (H&E) and assessed by microscopy (Olympus BX43, Shinjuku, Japan). The tissue samples were depara nized by xylene and ethanol and then antigen retrieval was performed with proteinase K treatment. After this treatment, the tissue samples were incubated with an anti-CD31 monoclonal primary antibody at 100 µL for 15 minutes at room temperature. After being washed with TBS, the samples were incubated with a master polymer plus HRP at room temperature for 30 minutes. Then samples were washed with TBS. Chromogen solution was prepared, 1 drop of DAB Chromogen concentrate to 1 ml of DAB substrate buffer. Samples were treated with DAB and incubated at room temperature for 5 minutes. The samples were covered with hematoxylin for 15 seconds and then washed with distilled water. After that dehydration step was performed by alcohol solutions at room temperature for 30 seconds. Finally the samples was cleared in xylene and mounted with permanent mounting medium. Tissue sections were analyzed using the vascular hotspot technique to obtain MVD.
Sections were scanned at low power to determine areas of highest vascular density. Within this region, individual microvessels were counted in three separate random elds at high power (0.142-mm2 eld size). The mean vessel count from the three elds was used. A single countable microvessel was de ned as any endothelial cell or group of cells that was clearly separate from other vessels, stroma, or tumor cells without the necessity of a vessel lumen or RBC within the lumen. Areas of gross hemorrhage and necrosis were avoided. MVD counts were measured by counting CD31 IHC hot spots in three separate 400× elds.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 7 Software. Results are expressed as the mean standard error of the mean. Comparisons between multiple groups were performed using one-way analysis of variance (ANOVA), with post hoc testing performed with Bonferroni analysis or unpaired ttests, as appropriate. Also comparisons between two groups were performed using independent sample T-test. P values ≤ 0.05 were considered statistically signi cant (*p < .05, **p < .01, ***p < .001, ****p < .0001). Data were presented as mean ± standard deviation (SD).
The degree of ischemic damage was assessed through indicators, such as swelling, skin color changes and grade of the limb necrosis. Necrotic changes were macroscopically evaluated using the grade of limb necrosis score, in BM-MSCs and recombinant vector treated mice groups, ischemic damage recovery was observed at day 7 (0.4±0.5, p < 0.0001) and (0±0, p < 0.0001) respectively. For BM-MSCs miR-126 group, no signs of necrosis were observed ( Figure 6).

PCR for SRY gene
Polymerase chain reaction for Y chromosome gene SRY was performed. Results indicated that at days 3,

MiR-126 overexpression increased VEGF expression, and induced angiogenesis
In order to understand the angiogenic potential of BM-MSCs, recombinant vector and BM-MSCs miR-126 , gastrocnemius muscles were collected at days 7, 14, 21, and 28 after transplantation and real-time PCR was performed to determine the expression of VEGF.
To understand the regulatory mechanism of miR-126 on VEGF pathway, the expression of PIK3R2 and SPRED1 (miR-126 targets in VEGF pathway) in virus and PBS groups were measured. The analysis of increased rate during 28 days (38.3 ± 2.5, 45 ± 2.6, 49.3 ± 2.5, 51.6 ± 3.5, respectively, P < 0.0001 for all time points). In fact CD31-MVD in BM-MSCs miR-126 treated mice was signi cantly higher than the other groups ( Figure 11).

Discussion
Critical limb ischemia (CLI) considered as the most severe form of peripheral artery disease (PAD), being associated with a high risk of major amputation, cardiovascular events and death. CLI is considered the "end stage" of peripheral arterial disease (PAD). It is associated with considerable mortality, morbidity, and increased use of health care resources (33). On the other hand one of the most important property of MSCs is their capacity for promoting angiogenesis, and proangiogenic factor VEGF is the main growth factor for this process that secret by MSCs during pathophysiological conditions (37,39).
MiR-126 plays a critical role in regulating stem cell functions. MiR-126 can regulate the BM-MSCs proliferation, apoptosis, survival, migration, invasion, para-secretion and differentiation to endothelial cells via regulating the PI3K/AKT and MAPK/ERK1 signaling pathways. These pathways are an intracellular signal transduction pathways that promote gene transcription, apoptosis, metabolism, proliferation, differentiation, cell survival, growth and angiogenesis in response to extracellular signals.
Two important targets of miR-126 including PIK3R2 (P85β) and SPRED1 play an important role as an inhibitor of PI3K (in PI3K/AKT signaling pathway) and RAF1 (in MAPK/ERK1 signaling pathway) respectively (40,41). In BM-MSCs miR-126 , miR-126 targets these two negative regulators of pathways and leading to promote signaling pathways that results in MSCs proliferation and increase paracrine secretion (42,43). As a result, the BM-MSCs miR-126 conditioned medium contains more effective concentration of angiogenic growth factors compared to untransformed MSCs, leads to increase remarkable proliferation and migration in human umbilical vein endothelial cells (HUVECs).
The BM-MSCs, recombinant vector and BM-MSCs miR-126 were then assessed for their ability to treat critical limb ischemia. We used a model of limb ischemia that was induced by the complete removal of the femoral artery and the closing of its branches. At the rst step, CLI induced mice were evaluated for functional recovery and ischemia using the Tarlov, ischemia, modi ed ischemia, function and the grade of limb necrosis scoring system at days 3, 7, 14, 21 and 28 after transplantation (33,44). The results showed that BM-MSCs transplantation improved the treatment e cacy of hindlimb ischemia compared to control group. Remarkable improvements were observed in function and foot mobility with BM-MSCs treatment according to Tarlov and Function scores. In fact muscle function and muscle strength, which are a very important parameter to evaluate the angiogenic effects of MSCs therapy, improved during 28 days after BM-MSCs treatment. Additionally the mice showed restructuring of skeletal muscles at the injured sites and functional improvements. Mice in BM-MSCs group improved necrosis of the hindlimb and also showed reduced swelling and edema. In fact according to observational tests results, BM-MSCs compared to control group resulted in a faster and more effective recovery from limb ischemia because of its therapeutic properties. Recombinant vector treated mice showed similar effect to BM-MSCs group.
But in BM-MSCs miR-126 group functional improvements including muscle function and muscle strength and also restructuring of skeletal muscles were much better than the other groups.
In the next step, to evaluate angiogenesis at the cellular level, H&E and immunohistochemical staining was done for the evaluation of muscle regeneration and detection of CD31 endothelial cells marker.
Assessment of neovascularization and capillary density was done by MVD score (45,46). Results from H&E and IHC con rmed the results obtained from the behavioral tests. Muscle regeneration in BM-MSCs group had a signi cant increase compared to control group during 28 days, starting from day 14. On the other hand it con rmed that neovascularization and blood supply were increased at GC muscle. In recombinant vector group, similar effects to BM-MSCs group obtained. But BM-MSCs miR-126 group showed signi cant increase in muscle regeneration compared to other groups. Furthermore IHC results indicated that formation of new blood vessels at day 7 was not signi cantly differ in BM-MSCs and recombinant vector groups, but at days 14, 21, and 28 neovascularization in 2 groups was greater than control groups. In BM-MSCs miR-126 group according to MVD score neovascularization was increased after day 7 and signi cantly increased during 28 days compared to the other groups. According to results, MVD in BM-MSCs group was increased after day 14 compared with control group. This property of BM-MSCs in promoting angiogenesis is related to the vascular differentiation capacity of these cells as well as secretion of angiogenic growth factors and cytokines (47,48). In fact these results showed that histologic ndings were associated with vascular and functional outcomes. On the other hand according to obtained results from behavioral, H&E, and IHC tests improvements were observed in muscle function and muscle strength with BM-MSCs treatment as well as formation of new blood vessels and muscle regeneration. BM-MSCs can contribute to angiogenesis via secretion of VEGF, angiogenin, IL-8, HGF, TNFalpha, PD-ECGF, FGF-2, and MMP-9. VEGF plays a critical role in angiogenesis through stimulating endothelial cell proliferation, migration, and organization into tubules. Furthermore, VEGF can increase the number of circulating endothelial progenitor cells (49)(50)(51)(52).
In BM-MSCs miR-126 group these results (improvements in muscle function, muscle strength and formation of new blood vessels and muscle regeneration) was much better than in BM-MSCs group. As mentioned earlier, miR-126 can regulate PI3K/AKT and MAPK/ERK1 signaling pathways through targeting PIK3R2 and SPRED1. By inhibition of these two negative regulators, activation of pathways leads to MSCs proliferation, differentiation to ECs, increase paracrine secretion of angiogenic factors including VEGF, bFGF, and angiogenesis. In fact because of the angiogenic potential of miR-126, VEGF expression increases and consequently VEGF pathway and angiogenesis promotes (53,54). In recombinant vector group, de ned number of virus particles were injected intramuscularly (IM) at GC muscle. Consequently at skeletal muscle, miR-126 leads to increase VEGF expression and induces angiogenesis, muscle regeneration and functional improvements. In BM-MSCs miR-126 group, the combination of angiogenic potential of BM-MSCs and miR-126 resulted in signi cant neovascularization, muscle regeneration and functional improvements compared to the other groups.
At day 28 after the treatment, the presence of donor cells were assessed by PCR for SRY gene. The results showed that both BM-MSCs and BM-MSCs miR-126 were survived during 21 days at GC muscle and also BM-MSCs miR-126 were detected at day 28. These results indicated important information concerning immunologic response after allogeneic MSC transplantation to the GC muscle. One of the most important limitations of MSCs transplantation to treat limb ischemia is poor survival of the transplanted cells because of adverse microenvironment in injured site such as hypoxia, ischemia and also excessive in ammation (55,56). Previous studies demonstrated that BM-MSCs secrete various cytokines that have immunomodulatory, anti-in ammatory and anti-apoptotic functions in the target position of the transplantation such as IL-10, IL-6, PGE2 and TGF-β (57,58). Some strategies are to reinforce the longevity of transplanted cells such as ways of delivery, number of injected cells, preconditioning of cells, genetic modi cation and combined cell therapy (59). In this study we used IM injection as delivery way and de ned number of cells and also used miR-126 to increase the survival of BM-MSCs. According to the results these methods were effective in cell survival. MiR-126 can increase the resistance of the BM-MSCs to apoptosis via regulating the PI3K/AKT and MAPK/ERK1 signaling pathways. By inhibiting the negative regulators (PIK3R2 and SPRED1) of pathways, miR-126 increases the survival of the cells (31,60).
Finally, the evaluation of VEGF expression in 4 groups showed that in BM-MSCs and virus groups VEGF over expression was similar but in miR-126-BM-MSCs group over expression of VEGF was remarkable. In BM-MSCs group over expression of VEGF is due to the paracrine effect of MSCs. MSCs have been extensively investigated as a source for promising proangiogenic cell therapy in angiogenesis dependent diseases such as CLI (61). BM-MSCs can secrete growth factors and cytokines related to angiogenesis such as VEGF. In virus group over expression of VEGF is due to the angiogenic potential of miR-126. As mentioned earlier, miR-126 targets two regulatory molecules (PIK3R2 and SPRED1) in PI3K/AKT and MAPK/ERK1 signaling pathways, leading to promote VEGF pathway and increasing secretion of VEGF (53,62). On the other hand, miR-126 exerts angiogenic potential through regulating of stimulatory signaling pathways and inhibits the negative regulators.
It was demonstrated that both BM-MSCs and miR-126 have angiogenic potential. Therefore, to increase the angiogenic effect of MSCs, miR-126 was transferred to the cells and obtained results indicated that over expression of VEGF was higher than MSCs and virus groups separately. In other words, the angiogenic potential of BM-MSCs and miR-126 were combined together in order to achieve more effective therapy to treat CLI.
It was mentioned that miR-126 exerts angiogenic potential through inhibiting of PIK3R2 and SPRED1. In order to verify this regulatory mechanism, the expression of PIK3R2 and SPRED1 were evaluated in virus and control groups. Regulatory mechanism of miR-126 leads to decrease the expression of PIK3R2 and SPRED1 (31, 63). As expected, the expression of PIK3R2 and SPRED1 were decreased at days 14 and 21 in virus group compared to control group. Although the expression of PIK3R2 and SPRED1 on 28 day did not decrease as much as on 14 and 21 days. Additionally, the expression of VEGF in virus and MSCs groups on 28 day did not increase as much as on 14 and 21 days. It seems that after day 21, because of the progress in curing the disease (improvements in angiogenesis, muscle regeneration, muscle function, muscle strength and edema) and also decreased expression of miR-126, VEGF expression begins to decline and on the other hand the expression of PIK3R2 and SPRED1 begins to increase.
The purpose of this study was to investigate the potential of miR-126 on increasing angiogenic capacity as well as increasing BM-MSCs viability. According to the results, miR-126 has signi cant effect on mesenchymal stem cells functions to increase paracrine secretion, proliferation, differentiation, survival, migration and angiogenic potential.

Conclusions
According to the obtained results, it can be concluded that combination use of BM-MSCs and miR-126 leads to more effective recovery from critical limb ischemia compared to using them alone. In fact, miR-126 due to its ability to regulate gene expression, especially genes involve in angiogenesis pathway, it can be used as a strong modi er to reinforce the angiogenic potential and survival of BM-MSCs, leading to more effective treatment for CLI.    Figure 1 Images of bone marrow derived mesenchymal stem cell collection procedures (A) First the mouse was sacri ced by cervical dislocation (B) Tibia and femur bones were dissected and then muscles, ligaments, and tendons were removed (C) Bones were transferred to a 100-mm sterile culture dish with 5 mL phosphate-buffered saline containing penicillin-streptomycin (D) The dish was transferred into the biosafety cabinet and the two ends of bones below the end of the marrow cavity were cut with scalpel (E) A 5-ml syringe was inserted into the bone cavity and used to slowly ush the marrow out. The bone cavities were washed until the bones became pale. (F) Finally all the bone pieces were removed and the media transferred to a new ask that incubated at 37C in a 5% CO2 incubator (G).   Functional scoring. All transplanted mice groups were evaluated for functional recovery according to Tarlov score and Function score (A-B). Evaluation of ischemia was performed by ischemia score and modi ed ischemia score (C-D). And improvements of necrotic changes was determined according to the grade of limb necrosis score (E). Data presented as (mean ± SD) (P ≤ 0.05) (*p< 0.05, **p <0.01, ***p< 0.001, ****p < .0001).