Despite normal arteriogenic and angiogenic responses, hind limb perfusion recovery and necrotic and fibroadipose tissue clearance are impaired in matrix metalloproteinase 9-deficient mice

Abstract

Objective: The relative contributions of arteriogenesis, angiogenesis, and ischemic muscle tissue composition toward reperfusion after arterial occlusion are largely unknown. Differential loss of bone marrow-derived cell (BMC) matrix metalloproteinase 9 (MMP9), which has been implicated in all of these processes, was used to assess the relative contributions of these processes during limb reperfusion.

Methods: We compared collateral growth (arteriogenesis), capillary growth (angiogenesis), and ischemic muscle tissue composition after femoral artery ligation in FVB/NJ mice that had been reconstituted with bone marrow from wild-type or MMP9(-/-) mice.

Results: Laser Doppler perfusion imaging confirmed decreased reperfusion capacity in mice with BMC-specific loss of MMP9; however, collateral arteriogenesis was not affected. Furthermore, when accounting for the fact that muscle tissue composition changes markedly with ischemia (ie, necrotic, fibroadipose, and regenerating tissue regions are present), angiogenesis was also unaffected. Instead, BMC-specific loss of MMP9 caused an increase in the proportion of necrotic and fibroadipose tissue, which showed the strongest correlation with poor perfusion recovery. Similarly, the reciprocal loss of MMP9 from non-BMCs showed similar deficits in perfusion and tissue composition without affecting arteriogenesis.

Conclusions: By concurrently analyzing arteriogenesis, angiogenesis, and ischemic tissue composition, we determined that the loss of BMC-derived or non-BMC-derived MMP9 impairs necrotic and fibroadipose tissue clearance after femoral artery ligation, despite normal arteriogenic and angiogenic vascular growth. These findings imply that therapeutic revascularization strategies for treating peripheral arterial disease may benefit from additionally targeting necrotic tissue clearance or skeletal muscle regeneration, or both.

Figure 1

BMC-specific deletion of MMP9 impairs perfusion recovery after FAL. A) Laser Doppler foot perfusion recovery curve for WT→WT and MMP9^−/−→WT groups (n=8 and 7, respectively). B) Representative laser Doppler perfusion images of mice (ligation on left leg, L: stabilized with two-sided tape, dark squares). I p<0.05 between WT→WT and MMP9^−/−→WT.

Figure 2

Arteriogenesis after FAL. A) microCT images of mouse upper hindlimbs at 14 days post-FAL. Arrows indicate primary collateral arteries in the gracilis muscle. B) Whole mounts of smooth muscle α-actin+ (SMαA) collateral arteries. Bars indicate 500μm. C) Bar graphs of collateral artery diameter (n=5-8 per group). D,E) Bar graphs of wall nuclei and vessel thickness (n=4-7 per group). *p<0.05 between ligated and unligated limbs within WT→WT or MMP9^−/− →WT mice. I p<0.05 between WT→WT versus MMP9^−/−→WT mice within ligated or unligated limbs.

Figure 3

BMC-derived MMP9 is not involved in collateral artery matrix remodeling. A) Immunofluorescent staining showed minimal presence of MMP9 around collateral arteries or within Mac+ macrophages (see Fig SI for positive control) at day 3 post-FAL. B) Polarized light birefringence images of picosirious red stained collagen was used to quantify mean peri-collateral collagen intensity (C) and area (D). Bars indicate 25 μm. *p<0.05 between ligated versus unligated limbs within WT→WT or MMP9^−/−→WT mice (n=6-7 per group).

Figure 4

Capillary growth is unaltered in MMP9^−/−→WT mice. A-E) Capillary growth was analyzed in calf muscle cross sections at day 14 post-FAL. A) Images of the 4 ischemic tissue morphologies. Bars indicate 25 μm. B,C) Bar graphs of CD31+ area based on tissue type (B) or averaged across all tissue types (C) (n=7,8 mice grouper WT→WT and MMP9^−/−→WT group, respectively). D, E) Bar graphs of CD31+ vessels within regions of viable muscle (regenerating or mature fibers) on a per fiber basis for capillary density (D) or capillary to muscle fiber ratio (E) (n=7, 8 mice). F, G) Early time-point analysis of capillary density (F) and capillary to muscle fiber ratio (G) at day 3 post-FAL (n=4 mice per group). *p<0.05 between ligated versus unligated limbs within WT→WT or MMP9^−/−→WT mice. Ip<0.05 between WT→WT versus MMP9^−/−→WT mice within ligated or unligated limbs.

Figure 5

MMP9^−/−→WT mice exhibit impaired skeletal muscle repair within damaged regions. A) Images of H&E stained calf muscle cross-sections. Bars indicate 200 μm. B, C) Tissue composition line graphs. *p<0.05 for total regenerating tissue vs. WT→WT (n=5-8 per group). Ip<0.05 vs. WT→WT across all time points.

Figure 6

Perfusion is related to skeletal muscle repair, but not vascular remodeling. A, C, E) Scatterplots relating perfusion (LDPI ratio) to percent total fibro-adipose and necrotic tissue area, collateral diameter, and approximate CD31+ vascular area. B, D, F) Scatterplots from A, C, and E with linear regressions.