In vivo efficacy of endothelial growth medium stimulated mesenchymal stem cells derived from patients with critical limb ischemia

Abstract

Background: Cell therapy has been proposed for patients with critical limb ischemia (CLI). Autologous bone marrow derived cells (BMCs) have been mostly used, mesenchymal stem cells (MSCs) being an alternative. The aim of this study was to characterize two types of MSCs and evaluate their efficacy.

Methods: MSCs were obtained from CLI-patients BMCs. Stimulated- (S-) MSCs were cultured in endothelial growth medium. Cells were characterized by the expression of cell surface markers, the relative expression of 6 genes, the secretion of 10 cytokines and the ability to form vessel-like structures. The cell proangiogenic properties was analysed in vivo, in a hindlimb ischemia model. Perfusion of lower limbs and functional tests were assessed for 28 days after cell infusion. Muscle histological analysis (neoangiogenesis, arteriogenesis and muscle repair) was performed.

Results: S-MSCs can be obtained from CLI-patients BMCs. They do not express endothelial specific markers but can be distinguished from MSCs by their secretome. S-MSCs have the ability to form tube-like structures and, in vivo, to induce blood flow recovery. No amputation was observed in S-MSCs treated mice. Functional tests showed improvement in treated groups with a superiority of MSCs and S-MSCs. In muscles, CD31+ and αSMA+ labelling were the highest in S-MSCs treated mice. S-MSCs induced the highest muscle repair.

Conclusions: S-MSCs exert angiogenic potential probably mediated by a paracrine mechanism. Their administration is associated with flow recovery, limb salvage and muscle repair. The secretome from S-MSCs or secretome-derived products may have a strong potential in vessel regeneration and muscle repair. Trial registration NCT00533104.

Keywords: Angiogenesis; Cell therapy; Critical limb ischemia; Mesenchymal stem cells.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

Study design. a BMCs were obtained from 7 CLI patients. MSCs were selected and expanded in a CFU-F medium for 28 days (n = 7). Stimulated MSCs were cultured in EGM-2 medium for 14 days (n = 7). b Hindlimb ischemia was induced by femoral artery ligation in Nude mice. Cells (BMCs, MSCs, S-MSCs) or vehicle were injected in the gastrocnemius. Hindlimb perfusion, Necrose detection and functional tests were performed during 28 days. At day 28 muscles were harvested for biological analysis

Fig. 2

Cells observation and real-time RT-PCR analysis. a BMCs morphology in May Grünwald Giemsa staining. b MSCs morphology in phase contrast microscopy. c S-MSCs morphology in phase contrast microscopy. d mRNA expression of six genes of interest were quantified using simultaneous real-time RT-qPCR experiments in MSCs (black box, n = 7), in S-MSCs (empty box, n = 7) and cb-ECFC (grey box, n = 5) used as control. x: aberrant distribution values as indicated by Box and Whiskers plots Medcalc version 7.3 software (*p < 0.05 and †p < 0.01)

Fig. 3

Cells characterization by FC. a BMCs characterization. HSCs (CD34+ cells) analysis was performed by FC according to the ISHAGE reference method. b Immunophenotype characterization of isolated MSCs and S-MSCs. MSCs and S-MSCs (grey) were stained with the following human monoclonal antibodies: CD105 (endoglin), CD90 (Thy-1), CD73 (5′ nucleotidase), CD34, CD45, CD11b (integrin alpha M), HLA-DR (human leucocyte antigen-D related), CD146 (MCAM), CD140a (PDGF RA), CD140b (PDGF RB), CD31 (PECAM1), CD144 (Ve-Cadherin), VEGF R1, VEGF-R2, CD184 (CXCR4) and CD106 (VCAM1). The isotype-matched mouse IgG1-FITC, IgG1-PE, IgG1-PCy5 and IgG1- PCy7 were used as negative controls (empty)

Fig. 4

Angiogenic effect of culture supernates. a MSCs and S-MSCs culture supernates and cell-free media (CFU-F and EGM-2) were recovered at day 28. HMEC-1 were suspended in endothelial cell growth medium MV (n = 3), in cell-free CFU-F medium (n = 3), in cell-free EGM-2 medium (n = 3), in MSCs culture supernates (obtained from 5 CLI-MSCs, n = 2 each group), in S-MSCs culture supernates (obtained from 5 CLI-S-MSCs, n = 2 each group) and incubated on Matrigel during 24 h. The extend of the network of the capillary-like tubes was appreciated at 3:30 h. b Quantification of the loops number at 3:30 h. c Quantification of total tube length (µm) at 3:30 h. x: aberrant distribution values as indicated by Box and Whiskers plots Medcalc version 7.3 software (*p < 0.05, †p < 0.01 and ‡p < 0.001)

Fig. 5

Hindlimb blood flow recovery. a Representative LASCA images show improved perfusion in vehicle-, BMCs-, MSCs- and S-MSCs- treated mice at days 0 after ligation, 3 and 28. Low or no blood perfusion was displayed as dark blue, whereas the highest perfusion was displayed as red. b Hindlimb blood flow recovery after femoral artery ligation and BMCs (n = 40), MSCs (n = 35), S-MSCs (n = 40), or vehicle (n = 20) injection. CT by BMCs transplantation (blue) was effective in comparison with vehicle (green) injection for blood flow recovery at day 28 but BMCs did not restore completely the blood flow. In contrast MSCs (red) provided complete recovery at day 20 whereas S-MSCs (violet) were effective to completely restore blood flow at day 14. c Hindlimb blood flow recovery after BMCs (approximately 6 mice per patient), MSCs (approximately 6 mice per patient) and S-MSCs (approximately 6 mice per patient) infusion. BMCs were obtained from 7 different CLI-patients (Table 1) and were considered as “gold standard” (*comparison of MSCs group versus BMCs group; #S-MSCs versus BMCs; ¥S-MSCs versus MSCs; §BMCs versus vehicle; ΔMSCs versus vehicle; ∞S-MSCs versus vehicle)

Fig. 6

Limb survival. a S-MSCs provide complete limb salvage in comparison with MSCs and BMCs. b In comparison with vehicle (n = 20), BMCs (n = 40) provided better limb salvage (p < 0.001). There was no difference between BMCs and MSCs (n = 35) treatment. S-MSCs (n = 40) were the most protective compared to BMCs and MSCs (p < 0.05)

Fig. 7

Functional tests (static and dynamic). a, b Static tests: MSCs and S-MSCs were significantly more effective than BMCs and vehicle to improve IDS. MSCs (red) and S-MSCs (violet) provided significantly higher ratios at day 28 in comparison with BMCs (blue). There was no difference between the MSCs and S-MSCs groups. c Dynamic test: S-MSCs, MSCs and BMCs were effective to restore a normal walk in comparison with vehicle [*comparison of MSCs group (n = 35) versus BMCs group (n = 40); #S-MSCs (n = 40) versus BMCs; ¥S-MSCs versus MSCs; §BMCs versus vehicle (n = 20); ΔMSCs versus vehicle; ∞S-MSCs versus vehicle]

Fig. 8

Neovessels visualization with CD31 and αSMA labelling. CD31 + staining (red) detects both angiogenic capillaries and arterioles. αSMA+ (green) detects arterioles. a Visualization of capillaries and arterioles in the semimembranosus in normal and ischemic legs. b Visualization of capillaries and arterioles in the gastrocnemius in normal and ischemic legs. c Angiogenesis analysis: fluorescence intensity ratio (CD31+ labelling) in the ischemic gastrocnemius in comparison with controlateral muscle. d Arteriogenesis analysis: Fluorescence intensity ratio (αSMA+/CD31+ labelling) in the ischemic gastrocnemius in comparison with controlateral muscle (*comparison of MSCs group versus BMCs group; #S-MSCs versus BMCs; ¥S-MSCs versus MSCs; §BMCs versus vehicle; ΔMSCs versus vehicle; ∞S-MSCs versus vehicle)

Fig. 9

Evaluation of muscle repair. a Histological analysis of semimembranosus muscle in normal and ischemic leg. b Histological analysis of gastrocnemius muscle in normal and ischemic leg. c Evaluation of muscle repair: percentage of muscle fibers with a central nucleus quantified in the ischemic gastrocnemius in comparison with its controlateral muscle (*comparison of MSCs group versus BMCs group; #S-MSCs versus BMCs; ¥S-MSCs versus MSCs; §BMCs versus vehicle; ΔMSCs versus vehicle; ∞S-MSCs versus vehicle)