It has been proposed that, in skeletal muscle, the angiogenic response to exercise may be signaled by the increase in muscle blood flow, via biomechanical changes in the microcirculation (increased shear stress and/or wall tension). To examine this hypothesis, we compared the change in abundance of vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and transforming growth factor-beta1 (TGF-beta1) mRNA in skeletal muscles of the canine leg after 1 h of pump-controlled high blood flow alone (passive hyperperfusion; protocol A) and electrical stimulation of the femoral and sciatic nerves producing muscle contraction (protocol B). The increase in leg blood flow (5.4- and 5. 9-fold change from resting values, respectively) was similar in both groups. Passive hyperperfusion alone did not increase message abundance for VEGF (ratio of mRNA to 18S signals after vs. before hyperperfusion, 0.94 +/- 0.08) or bFGF (1.08 +/- 0.05) but slightly increased that of TGF-beta1 (1.14 +/- 0.07; P < 0.03). In contrast, as previously found in the rat, electrical stimulation provoked more than a threefold increase in VEGF mRNA abundance (3.40 +/- 1.45; P < 0.02). However, electrical stimulation produced no significant changes in either bFGF (1.16 +/- 0.13) or TGF-beta1 (1.31 +/- 0.27). These results suggest that the increased muscle blood flow of exercise does not account for the increased abundance of these angiogenic growth factor mRNA levels in response to acute exercise. We speculate that other factors, such as local hypoxia, metabolite concentration changes, or mechanical effects of contraction per se, may be responsible for the effects of exercise.