Satellite cell-mediated angiogenesis in vitro coincides with a functional hypoxia-inducible factor pathway

Abstract

Muscle regeneration involves the coordination of myogenesis and revascularization to restore proper muscle function. Myogenesis is driven by resident stem cells termed satellite cells (SC), whereas angiogenesis arises from endothelial cells and perivascular cells of preexisting vascular segments and the collateral vasculature. Communication between myogenic and angiogenic cells seems plausible, especially given the number of growth factors produced by SC. To characterize these interactions, we developed an in vitro coculture model composed of rat skeletal muscle SC and microvascular fragments (MVF). In this system, isolated epididymal MVF suspended in collagen gel are cultured over a rat SC monolayer culture. In the presence of SC, MVF exhibit greater indices of angiogenesis than MVF cultured alone. A positive dose-dependent effect of SC conditioned medium (CM) on MVF growth was observed, suggesting that SC secrete soluble-acting growth factor(s). Next, we specifically blocked VEGF action in SC CM, and this was sufficient to abolish satellite cell-induced angiogenesis. Finally, hypoxia-inducible factor-1alpha (HIF-1alpha), a transcriptional regulator of VEGF gene expression, was found to be expressed in cultured SC and in putative SC in sections of in vivo stretch-injured rat muscle. Hypoxic culture conditions increased SC HIF-1alpha activity, which was positively associated with SC VEGF gene expression and protein levels. Collectively, these initial observations suggest that a heretofore unexplored aspect of satellite cell physiology is the initiation of a proangiogenic program.

Fig. 1.

Promotion of angiogenesis by satellite cells. A and B: image of rat microvascular fragment (MVF) cultured alone. C: image of rat satellite cells (RSC) cocultured with MVF. MVF and myotubes are indicated by red and black arrows, respectively. In each image, a bar equivalent to 20 μm in length is provided.

Fig. 2.

Quantification of satellite cell- or muscle-derived fibroblast stimulation of in vitro angiogenesis. MVF embedded in collagen were layered over monolayer satellite cell- (48 h after isolation) or skeletal muscle-derived fibroblast cultures. After 5 days of coculture, pictures of MVF were taken and angiogenesis was quantified. Bars with different letters differ at P < 0.01.

Fig. 3.

Effect of satellite cell conditioned medium (CM) on in vitro angiogenesis. A: MVF cultures were grown for 5 days in the presence of control (DMEM + 2% FBS), 5× satellite cell culture CM in DMEM + 2% FBS, and 10× satellite cell culture CM in DMEM + 2% FBS. B: CM (concentrated 10×) from satellite cells and muscle-derived fibroblast was prepared, handled in the same manner, and compared in MVF angiogenesis assays. Bars with different letters differ at P < 0.01.

Fig. 4.

Effect of VEGF inhibitors on satellite cell-mediated angiogenesis. MVF were cultured with satellite cell-derived CM, CM with a soluble VEGF receptor cocktail of sFlt and sFlk [VEGF receptors 1 and 2 (R1, R2)], or control medium (DMEM). Left: bars represent mean sprout length ± SE. Right: bars represent mean sprout number ± SE. Bars with different letters differ at P < 0.05.

Fig. 5.

A: expression of hypoxia-inducible factor-1α (HIF-1α) in cultured RSC. RT-PCR was used to detect the presence of HIF-1α message in cultured satellite cells from 24 h through 144 h. The β-actin message served as a control for equal amounts of starting RNA among samples. Two representative samples are shown at each time point. B: immunolocalization of HIF-1α in activated (48 h) RSC in culture. Cells were counterstained with DAPI to identify nuclei.

Fig. 6.

Modulation of HIF-1 and VEGF gene expression and protein in primary RSC. For each experiment, subconfluent RSC were cultured in six-well plates (100,000 cells/well) containing DMEM with 20% FBS before experimental treatment. A: RSC were cultured in DMEM with 2% FBS and subjected to normoxic (20% O₂), hypoxic (1% O₂), or cobalt chloride (CoCl₂; 150 μM) conditions for 24 h before RNA isolation and real-time PCR analysis of HIF-1α gene expression. B: RSC were transfected with 0.25 μg hypoxia response element (HRE) luciferase reporter plasmid (pHRE) before normoxic, hypoxic, or CoCl₂ exposure. After 24 h, cell lysates were harvested and assayed for luciferase activity. C: RSC were treated as described in A before real-time PCR analysis of VEGF gene expression. D: RSC were cultured in DMEM with 2% FBS and subjected to normoxic, hypoxic, or CoCl₂ conditions. At the end of 48 h, conditioned media were collected and VEGF protein concentration was determined by ELISA and normalized to DNA content per well. Bars with different letters differ at P < 0.05.

Fig. 7.

Immunolocalization of VEGF (D and H), HIF-1α (A and G), and Pax7 (B and E) in rat muscle 48 h after stretch, a process that activates quiescent satellite cells. C, F, and I: composite immunolocalization images plus DAPI staining to indicate nuclear position. Arrows indicate colocalization of proteins within putative satellite cells. I: HIF-1α and VEGF colocalize in mononucleated cells in frozen sections. J and K: images correspond to secondary antibody use alone. In each image, a bar equivalent to 20 μm in length is provided for size reference.