Toward a mouse model of hind limb ischemia to test therapeutic angiogenesis

Abstract

Introduction: Several clinical trials are currently evaluating stem cell therapy for patients with critical limb ischemia that have no other surgical or endovascular options for revascularization. However, these trials are conducted with different protocols, including use of different stem cell populations and different injection protocols, providing little means to compare trials and guide therapy. Accordingly, we developed a murine model of severe ischemia to allow methodic testing of relevant clinical parameters.

Methods: High femoral artery ligation and total excision of the superficial femoral artery was performed on C57BL/6 mice. Mononuclear cells (MNCs) were isolated from the bone marrow of donor mice, characterized using fluorescence-activated cell sorting, and injected (5×10(5) to 2×10(6)) into the semimembranosus (proximal) or gastrocnemius (distal) muscle. Vascular and functional outcomes were measured using invasive Doppler imaging, laser Doppler perfusion imaging, and the Tarlov and ischemia scores. Histologic analysis included quantification of muscle fiber area and number as well as capillary density.

Results: Blood flow and functional outcomes were improved in MNC-treated mice compared with controls over 28 days (flow: P<.0001; Tarlov: P=.0004; ischemia score: P=.0002). MNC-treated mice also showed greater gastrocnemius fiber area (P=.0053) and increased capillary density (P=.0004). Dose-response analysis showed increased angiogenesis and gastrocnemius fiber area but no changes in macroscopic vascular flow or functional scores. Overall functional outcomes in mice injected proximally to the ischemic area were similar to mice injected more distally, but muscle flow, capillary density, and gastrocnemius fiber area were increased (P<.05).

Conclusions: High femoral ligation with complete excision of the superficial femoral artery is a reliable model of severe hind limb ischemia in C57BL/6 mice that shows a response to MNC treatment for functional and vascular outcomes. A dose response to the injection of MNCs appears to be present, at least microscopically, suggesting that an optimal cell number for stem cell therapy exists and that preclinical testing needs to be performed to optimally guide human trials. Injection of MNCs proximal to the site of ischemia may provide different outcomes compared with distal injection and warrants additional study.

Figure 1

Methods. (A) Modified hind limb ischemia model; Arrow (↓) denotes site of femoral ligation; Asterisks (*) denote length of SFA excision. (B) Representative image of digit gangrene (black arrow). (C–D) Expression of mononuclear cell (MNC) markers. (E–F) Expression of endothelial progenitor cell (EPC) markers. (G–H) Co-expression of MNC and EPC markers.

Figure 2

MNC treatment (5×10⁵) improves hind limb ischemia. n=21 (MNC-treated) and 20 (control). (A) Tarlov score: MNC-treated mice show improved functional outcomes (*, P = .0004). (B) Ischemia score: MNC-treated mice had greater improvement of ischemia (P = .0002). (C) Modified ischemia score: no significant differences were observed (P = .4587). (D–E) Invasive Doppler measurements show MNC-treated mice to have improved (D) proximal flow (*, P < .0001) and (E) distal flow (*, P < .0001). (F) Representative laser Doppler images show improved perfusion in MNC-treated mice. Arrows show improved perfusion in MNC-treated mice at days 7 and 14. P vales represent statistical analysis by ANOVA.

Figure 3

Gastrocnemius muscle histology. n=3 (MNC-treated) and 3 (control). (A) Histological comparison of MNC treated and control mice over 28 days. (B) MNC-treated mice demonstrated increased muscle fiber area at day 7, 14 and 28 (*, P = 0.0053). (C) MNC-treated mice showed decreased number of muscle fibers (*, P < .0001). (D) MNC-treated mice demonstrated higher capillary density at day 14(*, P < .0001). P values represent statistical analysis by ANOVA.

Figure 4

MNC dose-response. n=4–6 per group. (A) Representative laser Doppler perfusion images; quantification showed no differences between MNC-treated groups at day 7 (P = .7501, ANOVA). (B) MNC-treated mice show increased perfusion (day 7) compared to control mice (P = .0018, ANOVA); 5×10⁵ and 1×10⁶ MNC groups showed increased distal flow versus control (*, P < .05; post-hoc test). **, P > .05 compared to other groups. (C) Functional outcomes: No significant differences between MNC treatment groups and control (P = .5520, ANOVA). (D) Increased capillary density (day 14) in MNC-treated mice compared to control mice (P < .0001, ANOVA). *, P < .05 vs. control, post-hoc test. **, P < .05 vs. control, post-hoc test; P > .05 vs. each other, post-hoc test). n=3. (E) Increased gastrocnemius muscle fiber area (day 14) in MNC-treated mice compared to control mice (P < .0001, ANOVA). *, P < .05 vs. control, post-hoc test. **, P < .05 vs. control, post-hoc test; P > .05 vs. each other, post-hoc test). n=3.

Figure 5

Proximal MNC injection response. n=21 (MNC-treated) and 6 (control). (A) Functional measurements with the Tarlov scale (day 7) showed no difference between control and MNC-treated groups (P = .9821). (B) Functional measurements with the ischemia scale (day 7) showed no difference between groups (P = .6594). (C) Increased semimembranosus flow with MNC-treated mice (P = .0041, ANOVA); there was no significant difference in flow between proximal and distal treatment (P > .05; post-hoc test). *, P < .05 vs. control, post-hoc test. (D) Distal flow was increased in proximally and distally treated mice versus control (P = .0086, ANOVA); there was no significant difference in flow between proximal and distal MNC treatment (P > .05; post-hoc test). *, P < .05 vs. control, post-hoc test. (E) MNC-treated mice showed increased capillary density compared to both control and baseline (P < .0001, ANOVA); there was no significant difference in capillary density between proximally and distally treated mice (P > .05; post-hoc test); *, P < .05 vs. control, post-hoc test. n=3. (F) MNC-treated mice showed increased gastrocnemius muscle fiber area compared to control mice (P < .0001, ANOVA); the difference between proximal and distal injection was significant (P < .05; post-hoc test). *, P < .05 vs. control, post-hoc test. n=3. (G) MNC-treated mice showed decreased gastrocnemius muscle fiber number compared to control mice (P = .002, ANOVA); the difference between proximal and distal injection was not significant (P > .05; post-hoc test). *, P < .05 vs. control, post-hoc test. n=3.