Figure 1.
Phenotype of pericytes in wildtype and CD248 knockout skeletal muscle.
Immunofluorescence and confocal microscopy of EDL muscle sections from wildtype mice stained with antibodies to either (A) Collagen IV (green), CD31 (blue) and pericyte markers αSMA, NG2 and PDGFRβ (all red) or (B) CD248 (green), CD31 (blue), pericyte markers αSMA, NG2 and PDGFRβ (all red) and nuclei (grey). CD248 was detected alone (green – marked with arrow) or co-localised with the other pericyte markers (yellow – marked with star). Enlarged region (dashed box) shows a pericyte (red) surrounded by the collagen IV basement membrane. (C) Pericyte coverage expressed as a percentage of CD31 positive vessels positive for ≥1 pericyte marker. (D) Expression of individual pericyte markers, expressed as percentage of CD31 positive vessels positive for either PDGFRβ, NG2 or αSMA. Empty bars are WT, filled bars are CD248-/-. Data are mean ± SEM from 3 independent animals. N.E = not expressed. ns = no significant difference assessed by (C) t-test or (D) ANOVA with Bonferroni post-test. Scale bars are 50 microns.
Figure 2.
Effects of CD248-/- on the architecture of skeletal muscle in control animals.
(A) Average capillary density (capillary/µm2) and (B) capillary to fibre ratio were determined from tissue sections of untreated EDL muscle by immunofluorescence. In (A) and (B) each point represents the mean value obtained for 3–5 sections per animal. The mean for all animals is represented by the wide line and the error bars show the 95% confidence interval. (C) In both genotypes a non-linear relationship is seen between muscle fibre size and capillary density. Empty circles are WT, filled circles are CD248-/-. Each circle represents the mean value for one animal, subjected to control, extirpation or prazosin treatment. (D) Immunofluorescence and confocal microscopy of EDL tissue sections from untreated 8-12 week old wildtype (top) or CD248-/- (bottom) mice. Sections were stained with antibodies to Collagen IV (blue) to mark the basement membrane and CD31 (red/magenta) to mark the blood vessels. Scale bars are 50 microns. ns = no significant difference by t-test.
Figure 3.
Effect of Prazosin and Extirpation treatment on muscle angiogenesis.
Effects of (A) prazosin treatment (Pra) and (B) extirpation (Ext) on the capillary to fibre ratio (C:F) in wildtype and CD248-/-mice. The capillary to fibre ratio (C:F) was calculated for each group. Alternatively mice underwent (C) prazosin treatment or (D) extirpation surgery either alone or in the presence (+) or absence (-) of the PDGF inhibitor Imatinib (150 mg/kg/day). In each case, control animals received no treatment. Empty bars are WT, filled bars are CD248-/-. (E) Western blot of EDL muscle from WT and CD248-/- mice undergoing extirpation (Ext) surgery either alone or in the presence (+) or absence (-) of the PDGF inhibitor Imatinib (Imat) at 150 mg/kg/day. Upper blot shows levels of phospho-ERK expression whilst the lower blot is a loading control showing levels of total ERK. (F) Top: Representative images of phospho-ERK expression by confocal microscopy from control, extirpated (Ext) and extirpated plus Imatinib (Imat). Phospho-ERK (green) nuclei (grey). Scale bar is 50 microns. Below: Pixel counts (expressed as fold change from genotype-matched control) from confocal immunofluorescence images of phosphorylated ERK from muscle sections with (+) and without (-) extirpation (Ext) and/or Imatinib (Imat) treatment. Data are mean ± SEM from 6 animals (A-D) and 3 animals (E–F). (A–D) ANOVA with Bonferroni post-test shows a significant effect of treatment and genotype on the response to stimulus. (F) Students t-test was used to identify significance * = P<0.05, ** = P<0.01, *** = P<0.001, ns = non-significant.
Figure 4.
Effect of CD248 genotype on transcriptional response to Prazosin and Extirpation treatment.
mRNA analysis of EDL muscle tissue by RT-PCR from WT and CD248-/- mice following either no treatment control (-), extirpation (ext) or prazosin (pra) treatment. Gene transcription data were acquired for angiopoietin2 (Ang2: A), endothelial tyrosine kinase (TEK: B), platelet-derived growth factor B (PDGF-B: C), platelet-derived growth factor receptor β (PDGFRβ: D), vascular endothelial growth factor A (VEGF-A: E) and hypoxia-inducible factor 1α (HIF1α: F). Data are shown as relative expression units (2−ΔCt) relative to 18S. Data are mean ± SEM from 6 animals. ANOVA with Bonferroni post-test shows a significant effect of treatment and genotype on the response to stimulus * = P<0.05; ** = P<0.01, *** = P<0.001, ns = non-significant.
Figure 5.
Effect of PDGF signalling inhibition by Imatinib on gene expression following Prazosin and Extirpation treatment.
mRNA analysis of EDL muscle tissue by RT-PCR from WT and CD248-/- mice plus no treatment control (-), extirpation (ext) or prazosin (pra) treatment. In addition, all mice were treated with Imatinib throughout the experiment. Gene transcription data were acquired for angiopoietin2 (Ang2: A), endothelial tyrosine kinase (TEK: B), platelet-derived growth factor B (PDGF-B: C), platelet-derived growth factor receptor β (PDGFRβ: D), vascular endothelial growth factor A (VEGF-A: E) and hypoxia-inducible factor 1α (HIF1α: F). Data are shown as relative expression units (2−ΔCt) relative to 18S. Data are mean ± SEM from 6 animals. ANOVA with Bonferroni post-test was performed and no significant differences were observed between any treatments or genotypes.
Figure 6.
Model of the proposed role of CD248-mediated PDGF signalling in vessel sprouting.
When endothelial cells are subjected to mechanical stretch, such as that seen following extirpation, the transcription factor HIF-1α is up-regulated and can bind to the HIF-binding site on the angiopoietin 2 gene. Angiopoietin 2 then binds to its receptor TEK on endothelial cells, inducing its phosphorylation. This results in upregulation of PDGF-B protein which dimerises and binds to its receptor PDGFRβ on pericytes. In the presence of CD248 this induces downstream signalling that stimulates migration of the pericyte to the endothelium and subsequent stabilisation of the capillary. When sprouting angiogenesis is successful the stretch on the capillary is reduced and HIF1α and angiopoietin gene expression can return to baseline.