In the present study we report the effects of a single bout of exercise on the expression of pro- and antiangiogenic factors in diabetic skeletal muscle. The results also provide novel data about exercise- and diabetes-induced changes in the expression of VEGF-A and TSP-1 separately in capillaries and muscle fibers. Our results showed that in the healthy mice the mRNA expression of VEGF-A and its receptor VEGFR-2 were increased, by 25% and 29% respectively, 6 h post exercise in the skeletal muscle homogenates. VEGF-B expression also showed a similar trend (15%, P = 0.08). Increases in VEGF-A and VEGFR-2 have been demonstrated earlier in healthy muscles e.g. [16, 25–27]. The present results suggest that the expression of VEGF-B may also be affected by exercise; however the increase was not statistically significant. VEGF-B belongs to the same protein family as VEGF-A, and it plays a role in the vascularization of adult and embryonic tissues, particularly in metabolically highly active tissues such as heart and skeletal muscle . However, it has not been studied to the same extent as VEGF-A and its exact role remains yet to be determined. In diabetic mice there was also a trend for increased expression of VEGF-A (17%) and VEGFR-2 (20%) 6 h post exercise, but these changes were not statistically significant. This suggests that angiogenic responses to a single bout of exercise in skeletal muscle are reduced in diabetic compared to healthy mice. The expression of angiogenic growth factors in diabetic muscles may, nevertheless, be upregulated by exercise to some extent. This is supported by the results obtained from our previous exercise training study, where basal mRNA expression of VEGF-A and VEGF-B were increased in trained compared to sedentary diabetic mice . In that study, sedentary diabetic mice exhibited decreased basal expression of VEGF-A and VEGF-B compared to the healthy mice. In the present experiment, there was no difference in the basal expression after 10 days of diabetes, which can be considered as acute diabetes. Contradictory findings exist in the literature about the effects of diabetes or hyperglycemia on the expression of VEGF-A. Several studies have shown decreased expression of VEGF-A in diabetic skeletal muscle or myocardium [9, 29–31]. Some studies have reported no change in the expression of VEGF-A due to hyperglycemia [8, 32], while others have shown increased expression compared to healthy muscle or myocardium [10, 33]. This demonstrates the complexity of the regulation of angiogenic signaling, and how it could be affected by diabetes. The discrepancies between these findings may arise from different models of diabetes used (type 1, type 2 or hyperglycemia), and from the use of different species (human, mouse, rat, rabbit). The severity and duration of diabetes may also have influenced the results. However, despite of the differences between the findings from these studies, it seems that the VEGF signaling pathway is somehow affected in skeletal muscle by diabetes and/or hyperglycemia.
The observed increases in the expression levels of VEGF-A and VEGFR-2 after the exercise bout in the present study were modest compared to several previous studies, where the highest expression (from 2 to 5-fold increase) has been observed within the first hour after the exercise [16, 17, 27, 34–36]. We chose to study time points at 3 h and 6 h post exercise to focus on the recovery period from exercise. At these time points VEGF-A and VEGFR-2 expressions have been shown to remain elevated (see refs above), but for the other proteins studied in the present experiment the data from the recovery period is very limited. We also wanted to study the mRNA expression changes in capillaries when blood flow has returned to basal levels after the exercise bout and angiogenesis processes may proceed. Both animal and human studies have shown that VEGF-A mRNA expression remains elevated up to 4–6 h post exercise, and returns to baseline levels by 8–24 h [16, 26, 27, 37, 38]. Different intensity, mode and duration of exercise may also induce variation between the findings of different studies.
The exercise intensity in the present study was chosen so that the diabetic mice were also able to complete the one-hour running bout. Thus, the relative intensity was probably higher for the diabetic mice than for the healthy mice, as their exercise capacity was decreased due to diabetes. However, the changes in the angiogenic gene expression were greater in the healthy mice than in diabetic mice. This further supports the suggestion that angiogenic responses in skeletal muscle are reduced in diabetic mice compared to healthy mice.
The strong correlations found between the proangiogenic factors revealed that mice showing high expression of angiogenic growth factors VEGF-A and VEGF-B, also showed high expression of their receptors, especially VEGFR-2. This suggests that the expression of angiogenic growth factors and their receptors are co-ordinately regulated in response to exercise or elevated blood glucose, and that their expression is controlled by a common upstream regulator. This idea needs to be further studied in the future.
The expression of thrombospondin-1, a known inhibitor of angiogenesis , was increased both in the skeletal muscle homogenates and in the capillaries of diabetic mice. Our results concur with the findings by Stenina et al., who found increased TSP-1 expression in the vessel wall of diabetic Zucker rats , which is a model for metabolic syndrome. Recently, hexosamine pathway of glucose catabolism was identified to mediate the upregulation of TSP-1 activated by high glucose in human aortic smooth muscle cells . It has been proposed that increased TSP-1 in blood vessels could be a direct response of vascular cells to increased glucose levels and, thus, a link between diabetes and atherosclerotic complications [40, 41]. This idea is supported by the present findings, as capillaries from diabetic mice expressed more TSP-1 than capillaries from healthy mice. In addition to atherosclerosis in larger blood vessels, TSP-1 may play a role in capillary rarefaction in skeletal muscles, as has been observed in obese Zucker rats  and in type 1 diabetic mice . Interestingly, Olfert et al. recently showed that a single exercise bout increased TSP-1 mRNA in healthy rat skeletal muscle, but that after 3 consecutive days of exercise this response was ablated . In their study the peak expression was detected 1 h after the cessation of exercise, and the increase was no longer statistically significant 2 h post exercise. This concurs with our findings, since we did not observe any changes in TSP-1 expression in healthy mice either 3 h or 6 h after the exercise bout in muscle homogenates. The healthy exercised mice showed even decreased capillary expression of TSP-1 6 h post exercise compared to the sedentary controls (Fig. 3). In addition to increased angiogenic growth factors, the decrease in TSP-1 in endothelial cells may further facilitate angiogenesis processes post exercise. If the decrease in TSP-1 expression occurs only in endothelial cells and not in muscle fibers, it is very difficult to observe from muscle homogenate samples.
The expression of Cyr61 was also increased in diabetic muscles. Cyr61 (also known as CCN1) is an important regulator of angiogenesis, endothelial cell function and ECM modulation [44, 45]. It also regulates the activity and production of other angiogenic proteins like VEGF-A [45, 46]. Recently, in diabetic retina advanced glycation end products were shown to induce the expression of Cyr61 and connective tissue growth factor (CTGF/CCN2), both of which belong to the same CCN protein family . In the diabetic retina the expression of Cyr61 and CTGF seem to be related to the thickening of capillary basement membrane . If this is also the case in skeletal muscle, it may lead to excess accumulation of extracellular matrix components and endothelial cell death, leading to dysfunctional capillaries and capillary rarefaction. There was a strong positive correlation between blood glucose concentration and the expression of TSP-1 and Cyr61 in the muscle homogenates, indicating that the expression of these genes is related to the level of hyperglycemia. In the present study, exercise did not attenuate the increased expression of either TSP-1 or Cyr61 in diabetic muscle.
On the basis of the present LCM results it seems that the mRNA expression of VEGF-A and TSP-1 is more affected by diabetes and exercise in capillaries than in skeletal muscle fibers. The expression of VEGF-A in capillaries was increased in the healthy exercised mice compared to controls, and thus agreed well with the observations from the muscle homogenates. However, the magnitude of the increase was higher in capillaries. In the muscle fibers extracted with LCM the expression of VEGF-A was increased about 2-fold, but due to the small number of mice in the analysis, it did not reach statistical significance. This increase is similar to what Birot et al. observed in rat single muscle fibers after a bout of endurance exercise (~90%) . Capillary TSP-1 expression was reduced in the healthy exercised group and increased in both diabetic groups compared to healthy controls. These increases in capillary TSP-1 due to diabetes also resemble the changes detected in the muscle homogenates. Taken together, these findings show that diabetes- and exercise-induced effects on angiogenic factors are stronger in capillary endothelial cells than in muscle fibers. This is interesting, since recently Lee and co-workers showed that autocrine VEGF-A signaling in endothelial cells is required for vascular homeostasis, and could not be compensated with paracrine VEGF-A (e.g. from skeletal muscle fibers) . Capillaries in the mouse skeletal muscle are small, and it is possible that cells other than endothelial cells located close to capillaries were also collected in the LCM procedure. However, the majority of the cells in the assay were endothelial cells, and the differences observed in mRNA expression between muscle fibers and capillaries indicate that representative cell populations were collected. This was controlled by setting the laser pulse diameter always smaller than 10 μm and examining the collecting cap microscopically after removing it off the sample.
The reduced capability of diabetic compared to healthy endothelial cells to increase VEGF-A production in response to exercise stimulus may be related to the impaired vascular function and capillary rarefaction in diabetic skeletal muscle. In skeletal muscle and myocardial ischemia studies, diabetic animals exhibit impaired angiogenesis, and this impairment has been linked to several proteins in angiogenesis signaling [5, 8–11]. Gene transfer with VEGF-A was able to promote angiogenesis in the myocardium of type 2 diabetic mice , but so far angiogenic gene transfer studies in humans have not been very successful . Exercise may positively affect angiogenic signaling by stimulating the whole signaling pathway and thus, be effective in the prevention and treatment of peripheral vascular problems in diseases such as diabetes.