Key points• Angiogenic regulators respond to acute exercise with different temporal expression patterns (e.g. 2-4 h versus 12-24 h) creating a complex multifaceted response that must be considered in studies using a single time point for post-exercise analyses.• In response to chronic training there appears to be a complex coordination in the proteomic responses of both positive and negative angiogenic factors that correspond with training-induced muscle capillary adaptation, such that altered basal expression and acute responses to exercise appear to withdraw or reduce the stimulus of angiogenic regulators in an expanding capillary bed with active angiogenesis.• These are the first data to show that nucleolin (a protein responsible for transcriptional processing and transportation of proteins from the cytoplasm to the nucleus) is responsive to acute exercise. We speculate that nucleolin may work in concert with vascular endothelial growth factor-A (VEGF) and endostatin.• Temporal responses observed in mice, particularly for VEGF, MMP-2 and MMP-9, may not be directly comparable to humans.Abstract Angiogenesis is controlled by a balance between positive and negative angiogenic factors, but temporal protein expression of many key angiogenic regulators in response to exercise are still poorly defined. In C57BL/6 mice, we evaluated the temporal protein expression of several pro-angiogenic and anti-angiogenic factors in response to (1) a single acute bout of exercise and (2) chronic exercise training resulting from 3, 5, 7, 14 and 28 days of voluntary wheel running. Following acute exercise, protein levels of vascular endothelial growth factor-A (VEGF), endostatin and nucleolin were increased at 2-4 h (P < 0.05), whereas matrix metalloproteinase (MMP)-2 was elevated within a 12-24 h window (P < 0.05). Training increased muscle capillarity 11%, 15% and 22% starting with 7, 14 and 28 days of training, respectively (P < 0.01). Basal VEGF and MMP-2 were increased by 31% and 22%, respectively, compared to controls (P < 0.05) after 7 days (7d) training, but decreased to back to baseline after 14d training. After 28d training VEGF fell 49% below baseline control (P < 0.01). Basal muscle expression of thrombospondin 1 (TSP-1) was ∼900% greater in 14d-and 28d-trained mice compared to either 5d-and 7d-trained mice (P < 0.05), and tended to increase by ∼180-258% compared to basal control levels (P < 0.10).The acute responsiveness of VEGF to exercise in untrained mice (i.e. 161% increase, P < 0.001) was lost with capillary adaptation occurring after 7, 14 and 28d training. Taken together, these data support the notion that skeletal muscle angiogenesis is controlled by a balance between positive and negative mitogens, and reveals a complex, highly-coordinated, temporal scheme whereby these factors can differentially influence capillary growth in response to acute versus chronic exercise.

These results suggest a link through which flow-mediated endothelial-derived signals may promote myocyte production of VEGF-A. In turn, myocyte-derived VEGF-A is required for appropriate flow-mediated microvascular remodelling. This highlights the importance of the local environment and paracrine interactions in the regulation of tissue perfusion.

Key pointsr Skeletal muscle capillary regression is associated with elevated thrombospondin-1 and vascular endothelial growth factor protein expression with detraining.r Vascular endothelial growth factor and nucleolin responses to acute exercise are blunted in the triceps surae muscles following exercise training. In the plantaris and soleus muscles, this blunted response persists up to 28 days after cessation of training.r Effects of detraining on skeletal muscle microvascular density appears to be similar among skeletal muscle of varying oxidative potential and is poorly associated with expression of matrix metalloproteinases-2 and -9 and endostatin.Abstract Temporal expression of positive and negative angiogenic factors in response to detraining is poorly understood. We report the protein expression of anti-angiogenic peptides (thrombospondin-1, TSP-1; and endostatin) as well as pro-angiogenic factors (vascular endothelial growth factor, VEGF; matrix metalloproteinases-2 and -9), and nucleolin (a nuclear protein involved with synthesis and maturation of ribosomes) in response to detraining in triceps surae muscles of C57BL/6 mice. Male mice were allowed to exercise voluntarily for 21 days, and then basal and acute response to exercise were evaluated at 1, 7, 14 and 28 days detraining (D1, D7, D14, D28, respectively, n = 12/group). As seen in the D1 mice, training resulted in the increased muscle capillary-to-fibre ratio (C/F), increased maximal running time and elevated basal expression of VEGF and matrix metalloproteinase-9 (P < 0.05). After 7 days of detraining (D7), C/F levels were similar to control levels, but both basal VEGF and TSP-1 were elevated (P < 0.05). At D14 and D28, TSP-1 protein was not different compared to baseline levels; however, VEGF was elevated in gastrocnemius (GA), but not the soleus (SOL) or plantaris (PLT) muscles, of D14 mice. Endostatin tended to decrease in D14 and D28 compared to controls. Timing of nucleolin protein expression differed between muscle groups, with increases at D1, D7 and D14 in the PLT, SOL and GA muscles, respectively. The response of VEGF and nucleolin to acute exercise was blunted with training, and remained blunted in the PLT and SOL even after 28 days of detraining, at a time point long after muscle capillarization was observed to be similar to pre-training levels. These data suggest that TSP-1 may be a mediator of capillary regression with detraining, even in the face of elevated VEGF, suggesting that pro-angiogenic regulators may not be able to prevent the regression of skeletal muscle capillaries under physiological conditions. The responses of matrix metalloproteinases, endostatin and nucleolin poorly correlated with detraining-induced capillary regression.

Skeletal muscle overload induces the expression of angiogenic factors such as vascular endothelial growth factor (VEGF) and matrix metalloproteinase (MMP)-2, leading to new capillary growth. We found that the overload-induced increase in angiogenesis, as well as increases in VEGF, MMP-2 and MT1-MMP transcripts were abrogated in muscle VEGF KO mice, highlighting the critical role of myocyte-derived VEGF in controlling this process. The upstream mediators that contribute to overload-induced expression of VEGF have yet to be ascertained. We found that muscle overload increased angiotensinogen expression, a precursor of angiotensin (Ang) II, and that Ang II signaling played an important role in basal VEGF production in C2C12 cells. Furthermore, matrix-bound VEGF released from myoblasts induced the activation of endothelial cells, as evidenced by elevated endothelial cell phospho-p38 levels. We also found that exogenous Ang II elevates VEGF expression, as well as MMP-2 transcript levels in C2C12 myotubes. Interestingly, these responses also were observed in skeletal muscle endothelial cells in response to Ang II treatment, indicating that these cells also can respond directly to the stimulus. The involvement of Ang II in muscle overload-induced angiogenesis was assessed. We found that blockade of AT1R-dependent Ang II signaling using losartan did not attenuate capillary growth. Surprisingly, increased levels of VEGF protein were detected in overloaded muscle from losartan-treated rats. Similarly, we observed elevated VEGF production in cultured endothelial cells treated with losartan alone or in combination with Ang II. These studies conclusively establish the requirement for muscle derived VEGF in overload-induced angiogenesis and highlight a role for Ang II in basal VEGF production in skeletal muscle. However, while Ang II signaling is activated following overload and plays a role in muscle VEGF production, inhibition of this pathway is not sufficient to halt overload-induced angiogenesis, indicating that AT1-independent signals maintain VEGF production in losartan-treated muscle.

No abstract

Angiogenesis is an important adaptation to exercise, occurring in response to a multitude of different stimuli including: shear stress, mechanical stretch, ischemia, electrical stimulation, and exercise. Current thinking suggests skeletal muscle angiogenesis is a temporal process controlled by a balance between positive and negative angiogenic proteins. But there is limited information on what molecular mediators control skeletal muscle angiogenesis in this time line, creating a critical need to clarify how individual protein responses regulate physiologic skeletal muscle angiogenesis in response to exercise training and/or physical deconditioning. Our objective is to characterize the temporal expression of several key positive (VEGF, MMP-2, MMP-9, nucleolin) and negative (TSP-1, endostatin) angiogenic factors under basal conditions, after acute exercise, in response to training, and after detraining. The central hypothesis is that training and deconditioning will cause temporally coordinated changes in positive and negative angiogenic regulators in response to exercise training, which will be reversed during detraining.

Vascular endothelial growth factor (VEGF) is a multifunctional protein critical in vascular growth/development and essential for angiogenesis, particularly in response to exercise. Using microarray analysis, we examined skeletal muscle (gastrocnemius) gene responses in sedentary (sed) and 8‐week treadmill trained (tra; 1 hr, 5 d/wk) muscle‐specific VEGF deficient mice (KOSed and KOTra) compared to littermate control mice (WTSed and WTTra) to gain insight into the potential non‐angiogenic mechanisms involving VEGF. When comparing sedentary groups (KOSed vs WTSed) and trained groups (KOTra vs WTTra) there were 349 and 126 differentially expressed genes, respectively. Ingenuity pathway analysis revealed potential novel functions of VEGF in association with lipid and carbohydrate metabolism between KOSed and WTSed. Between KOTra and WTTra there were associations with gene networks involved in cell signaling, protein synthesis, endocrine system disorders and metabolic disease. Data also confirm the vital association of VEGF in skeletal muscle structure and function, but more importantly identify potential novel relationships between VEGF and metabolism, endocrine and metabolic disorder not previously recognized. Future studies examining the role of VEGF as a metabolic regulator may open new avenues linking angiogenesis and metabolic diseases.

Skeletal muscle function is critical in exercise capacity. There is growing evidence that thrombospondin‐1 (TSP‐1), a potent anti‐angiogenic protein, is an important regulator of skeletal muscle function. We tested the hypothesis that chronic exposure to a TSP‐1 mimetic (ABT‐510) would decrease skeletal muscle capillarity and alter the balance between angiogenic proteins. Osmotic minipumps with ABT‐510 were implanted subcutaneously in the sub‐scapular region of C57/BLK6 mice for 14 days. This resulted in a 12% decrease in maximal running speed in the ABT‐510 treated group (P=0.028) with no significant change in the vehicle treated mice. When compared to the vehicle mice, the ABT‐510 group had a 13% decrease in capillarity in the gastrocnemius muscle (GA, P=0.055), 10% decrease in the plantaris muscle (PLT, P=0.051), and a 28% decrease in the soleus muscle (SOL, P<.001). ABT‐510 mice had decreased VEGF in both the GA and SOL muscles (GA; −30%, P=0.016; SOL; −38%, P=0.055), but no change in the PLT. There were no significant differences in endogenous TSP‐1 protein in the GA between groups. These data show that a TSP‐1 mimetic decreases whole body exercise and skeletal muscle capillarity, and alters the balance between VEGF and TSP‐1; and that chronic elevation of the TSP‐1 pathway can directly alter skeletal muscle structure and function.Support: NIH 5T32‐HL090610, AHA 10BGIA3630002.