The therapeutic effect of adipose-derived lipoaspirate cells in femoral head necrosis by improving angiogenesis

Abstract

Femoral head necrosis (FHN), one of the most popular joint diseases in the musculoskeletal system, is usually attributed to local ischemia of the femoral head. Thus, regenerating the vascularization capacity and restoring the local perfusion of the femoral head becomes an efficient therapeutic approach for FHN. We investigated the function of autologous lipoaspirate cells (LPCs) in regenerating circulation in FHN animal models and human subjects in this study. We also explored the mechanisms of why LPCs show a superior effect than that of the bone marrow-derived stem cells (BMSCs) in vascularization. Thirty-four FHN patients were recruited for the randomized clinical trial. Harris Hip Score (HHS) and digital subtraction arteriography (DSA) and interventional technique were used to compare the efficacy of LPCs treatment and vehicle therapy in improving femoral head circulation and hip joint function. Cellular mechanism that underlies the beneficial effect of LPCs in restoring blood supply and rescuing bone architecture was further explored using canine and mouse FHN animal models. We found that LPCs perfusion through the medial circumflex artery will promote the femoral head vascularization and bone structure significantly in both FHN patients and animal models. The HHS in LPCs treated patients was significantly improved relative to vehicle group. The levels of angiogenesis factor secreted by LPCs such as VEGF, FGF2, VEC, TGF-β, were significantly higher than that of BMSCs. As the result, LPCs showed a better effect in promoting the tube structure formation of human vascular endothelial cells (HUVEC) than that of BMSCs. Moreover, LPCs contains a unique CD44⁺CD34⁺CD31^- population. The CD44⁺CD34⁺CD31^- LPCs showed significantly higher angiogenesis potential as compared to that of BMSCs. Taken together, our results show that LPCs possess a superior vascularization capacity in both autonomous and paracrine manner, indicating that autologous LPCs perfusion via the medial circumflex artery is an effective therapy for FHN.

Keywords: angiogenesis; femoral head necrosis; lipoaspirate; stem cell; theraputic.

FIGURE 1

Vascularization and promotion of hip function in FHN patients received multiple LPCs infusion treatments. (A–D) DSA images perform the branches of the medial circumflex artery before treatment in different groups (zero-time, one time, three times, seven times, respectively). The white rectangle selects the area where distribute the branches of the medial circumflex artery or should be. SRA: superior retinacular artery, IRA: inferior retinacular artery. (E–H) DSA images perform the branches of the medial circumflex artery after treatment. The white rectangle selects the area where the new branches of the medial circumflex artery growth or will be. (I) The quantitative analysis of the blood vessel area measured in femoral head in different groups before and after treatment. (J) The quantitative analysis of Harris Hip score measured in four different groups before and after treatment. (K–N) Representative DSA images that represent the capillary network priority regeneration type along with three times treatment. Red area: original blood area; Yellow area: regenerated area after the first treatment; Blue area: regenerated area after the second treatment. (O–R) Representative DSA images that represent the inferior priority regeneration type along with three times treatment. Blue area: original blood area; Yellow area: regenerated area after the first treatment; Green area: regenerated area after the second treatment.

FIGURE 2

LPCs promote angiogenesis of endothelial cells by paracrine. (A–D) Quantitative analysis of the mRNA expression of VEGF, FGF2, VEC, TGF-β in BMSCs and LPCs. **P＜0.0001. (E) Representative microphotographs of capillary-like tubes formed by induced HUVEC cells on Matrigel in the presence of BMSCs or LPCs conditioned media. (F) Quantitative analysis of the total length of the tubes structure in BMSCs or LPCs conditioned media treated groups (Pixels). **p = 0.0015.

FIGURE 3

The characteristic angiogenic properties of LPCs and BMSCs. (A) Representative microphotographs of capillary-like tubes formed by induced LPCs and BMSCs on Matrigel in the presence of ECGS or vehicle media. (B) Quantitative analysis of the total length of tubes structure in the ECGS or vehicle treated groups (Pixels). **p = 0.0027.

FIGURE 4

The surface markers of LPCs. (A–G) Flow cytometry analysis for CD29, CD44, CD73, CD90, CD34, CD31, CD45 in LPCs. Green color histogram represents isotype negtive control, red color histogram represents cells stained with fluorescent antibodies. And the positive range selected performs the representative percentage of the positive population for each marker.

FIGURE 5

Angiogenic function of CD44⁺CD34⁺CD31⁻ LPCs. (A) Representative flow cytometry plots show the percentage of CD44⁺CD34⁺CD31⁻ LPCs. (B) Representative microphotographs of capillary-like tubes formed by induced CD44⁺CD34⁺CD31⁻ LPCs and CD44⁺CD34⁻CD31⁻ LPCs on Matrigel in the presence of ECGS media. (C) Total length of tubes structure (Pixels). ****P＜0.0001. (D) Statistic analysis of the frequency of CD44⁺CD34⁺CD31⁻ population in LPCs and BMSCs. ***P＜0.0001. (E) Representative flow cytometry plots show the percentage of CD44⁺CD34⁺CD31⁻ BMSCs.

FIGURE 6

The function of LPCs in bone structure improvement. (A) T2WI images of the femoral head from control group and LPCs-treated group. (B) Quantitative analysis of edema area of two groups. ****P＜0.0001. (C) HE-staining show the edema degree in the medial position of femoral head in control group and LPCs-treated group, and the black arrow point at the edema area. (D) Quantitative analysis of edema area in control group and LPCs-treated group. **p = 0.0039. (E) Representative 3-dimensional micro-computed tomography (μCT) image of coronal view of the femoral head in control group and LPCs-treated group. The white frame in the subchondral bone region, point out the necrotic region, and sclerotic region, respectively. (F) Quantitative analysis of bone volume /tissue volume (BV/TV) ratio, **p = 0.001, (G) trabecular pattern factor (Tb. Pf), *p = 0.0133, (H) trabecular thickness in control group and LPCs-treated group. **p = 0.0023, n = 3.

FIGURE 7

The effect of LPCs in vascularization and osteocyte generation. (A) Representative three-dimensional microcomputed tomography angiography (μCTA) images in the femoral head in control group and LPCs-treated group. (B) HE staining and immunohistochemical staining for Osteocalcin for both vehicle and LPCs cells treated groups. The black arrow shows osteocyte and lacunae. (C–D) Quantitative analysis of empty lacunae number and positive area of OCN. **p = 0.0023, *p = 0.0389, n = 3.