Paracrine‐derived vascular endothelial growth factor (VEGF) is established as an indispensable contributor to the angiogenic cascade. Autocrine VEGF, in contrast, is solely attributed to endothelial cell survival signaling, and its potential regulation of other endothelial cell functions has not been studied. We hypothesized that the deletion of VEGF would disrupt endothelial cell migration and shear stress adaptation. Endothelial cells were isolated from transgenic mice having VEGF exon 3 flanked by loxP sites (VEGFL/L). Cells were transduced with adeno‐Cre recombinase to induce VEGF deletion, or adeno‐β‐galactosidase as a control. When subjected to a scrape migration assay, cells with Cre induced VEGF deletion (VEGFΔ) exhibited impaired cell motility compared VEGFL/L cells (12% compared to 64% wound closure after 48 hours; p<0.05). Shear stress stimulation (15 dynes; 2hr) caused significant increases in phospho‐Akt in VEGFL/L cells, which was repressed in VEGFΔ cells (2.4 vs. 1.5 fold above static, respectively). Shear‐induced increases in phospho‐p38MAPK were unaltered in VEGFΔ cells (3.9 vs 3.3 fold above static, respectively). Shear‐induced endothelial cell alignment (15 dynes; 24hr) also was repressed in VEGFΔ cells. Taken together these results provide novel evidence that autocrine VEGF contributes to multiple aspects of endothelial cell function.

Increased capillary shear stress induces angiogenesis in skeletal muscle, but the signaling mechanisms underlying this response are not known. We hypothesize that shear stress-dependent activation of vascular endothelial growth factor receptor 2 (VEGFR2) causes p38 and ERK1/2 phosphorylation, which contribute to shear stress-induced angiogenesis. Skeletal muscle microvascular endothelial cells were sheared (12 dynes/cm(2), 0.5-24 h). VEGFR2-Y1214 phosphorylation increased in response to elevated shear stress and VEGF stimulation. p38 and ERK1/2 phosphorylation increased at 2 h of shear stress but only p38 remained phosphorylated at 6 and 24 h of shear stress. VEGFR2 inhibition abrogated p38, but not ERK1/2 phosphorylation. VEGF production was increased in response to shear stress at 6 h, and this increased production was abolished by p38 inhibition. Male Sprague-Dawley rats were administered prazosin (50 mg/L drinking water, 1, 2, 4, or 7 days) to induce chronically elevated capillary shear stress in skeletal muscle. In some experiments, mini-osmotic pumps were used to dispense p38 inhibitor SB203580 or its inactive analog SB202474, to the extensor digitorum longus (EDL) of control and prazosin-treated rats. Immunostaining and Western blotting showed increases in p38 phosphorylation in capillaries from rats treated with prazosin for 2 days but returned to basal levels at 4 and 7 days. p38 inhibition abolished the increase in capillary to muscle fiber ratio seen after 7 days of prazosin treatment. Our data suggest that p38 activation is necessary for shear stress-dependent angiogenesis.

Luminal splitting is a form of angiogenesis caused by increased shear stress. A feature of luminal splitting is a lack of extracellular matrix proteolysis correlated with decreased production of matrix metalloproteinase (MMP)‐2. We hypothesized that an MMP inhibitor, tissue inhibitor of matrix metalloproteinase (TIMP)‐1, is upregulated by shear stress. TIMP‐1 mRNA increased in the EDL of prasozin‐treated rats (p<0.05, n=4) as measured by RT‐PCR. Rat skeletal muscle microvascular endothelial cells were exposed to shear stress (12 dyne/cm2). TIMP‐1 mRNA increased in cells after 2 and 24 hours of shear (p<0.05, n=3), and TIMP‐1 protein increased at 2 and 24 but not 4 hours (p<0.05, n=3) as assessed by Western. TIMP‐1 secretion decreased at 2 hours of shear and returned to control levels at 24 hours (p<0.01, n=3) as measured by reverse zymography of concentrated media. Similarly, MMP‐2 activity increased in media at 2 but not 24 hours of shear (p<0.05, n=7) as assessed by an in vitro MMP‐2 activity assay (Anaspec). 2 hours of shear caused no change in TIMP‐1 surface staining (ns, n=3) as assessed by confocal microscopy, but did cause an increase in intracellular TIMP‐1 staining (p<0.05, n=3). Our results show that intracellular TIMP‐1 protein is increased by shear stress, due to increased production and decreased secretion. The function of intracellular TIMP‐1 is unknown.

Increased capillary shear stress in skeletal muscle causes a form of angiogenesis termed luminal splitting. An increased number of capillaries are observed at 7 days of chronic shear stress that is reliant on p38 signaling. Others have shown that VEGF is essential for luminal splitting. We hypothesize that VEGF production is p38 dependent and is increased by p38 dependent transcription factors Creb, ATF2 and MEF2C in response to shear stress. Male Sprague‐Dawley rats were administered prazosin (50 mg/L drinking water) to increase shear stress, and the extensor digitorum longus was extracted at 1, 2, 4 or 7 days. By Western blot, p38 was activated only at 2 days. Skeletal muscle endothelial cells were sheared (12 dynes/cm2) with or without 10 μM SB203580 (p38 inhibitor), then lysed and analyzed by RT‐PCR or Western blot. In vitro, shear stress induced VEGF mRNA and protein elevation at 2 and 6 hours, respectively. These increases were abolished by SB203580. Shear stress increased both ATF2 phosphorylation and MEF2C production at 2 hours, which were abrogated by SB203580. Creb phosphorylation was unchanged by shear stress. Our data show that shear stress‐mediated p38 activation is required for VEGF production. ATF2 phosphorylation and MEF2C production are increased by shear stress, but further investigation is necessary to determine if they mediate p38‐induced VEGF production. Supported by HSF and NSERC.Grant Funding SourceThe Heart and Stroke Foundation

Angiogenesis (the growth of new capillaries) occurs in adults in response to physiological stimuli such as wound healing and exercise. The mitogen-activated protein kinase c-jun N-terminal kinase (JNK) has a controversial role in the process of angiogenesis, with previous evidence supporting JNK as both a positive and negative regulator of blood vessel growth. The purpose of this study was to clarify the role of JNK in the angiogenesis process. Phosphorylated JNK was observed in cultured endothelial cells, and levels were constant regardless of extracellular matrix composition. Using SP600125, inhibition of JNK attenuated sprout growth in 3D capillary sprout cultures. Inhibition of JNK reduced endothelial cell proliferation and migration in vitro. JNK inhibition and siRNA knockdown of c-jun (a downstream target of JNK) decreased protein levels of the transcription factor Egr-1, a regulator of genes involved in proliferation and migration. Matrix metalloproteinase-2 (MMP-2) production, and activity also, was reduced in sprout cultures treated with SP600125. c-Jun silencing decreased both MMP-2 and membrane type-1 (MT1)-MMP mRNA in endothelial cells, implicating both JNK and c-jun as activators of proteolysis. Taken together, these results provide evidence that JNK and its downstream target c-jun positively regulate angiogenesis via activation of endothelial cell proliferation, migration and proteolysis.

Luminal splitting via internal division of capillaries is a form of angiogenesis caused by increased shear stress. A feature of luminal splitting is a lack of extracellular matrix proteolysis, which correlates with a decreased production of matrix metalloproteinase‐2 (MMP‐2). We hypothesized that the MMP inhibitor, tissue inhibitor of matrix metalloproteinase‐1 (TIMP‐1), is upregulated in response to shear stress. TIMP‐1 mRNA expression increased 3.8‐fold in the EDL of prasozin‐treated rats (p<0.05, n=4). Skeletal muscle endothelial cells were exposed to 2, 4, or 24 hours of shear stress (12 dyne/cm2). TIMP‐1 mRNA expression increased 7.0 and 9.0‐fold in cells after 2 and 24 hours of exposure to shear stress respectively (p<0.05, n=3), and TIMP‐1 protein increased 1.5 and 2.2‐fold after 2 and 24 hours of shear stress (p<0.05, n=3). TIMP‐1‐mediated inhibition of MMP‐2 also was increased 1.5 and 1.3‐fold after 2 and 24 hours of shear stress (n=2 and p<0.05, n=3 respectively). Our results show that TIMP‐1 is upregulated by shear stress, and suggest a mechanism for the absence of extracellular matrix proteolysis observed during luminal splitting.Funded by the Heart and Stroke Foundation of Canada.

Elevated shear stress within the skeletal muscle microvasculature is implicated in the induction of a longitudinal splitting form of angiogenesis, which is characterized by the lack of basement membrane breakage. We investigated whether the transcriptional regulator, Ets‐1, is responsive to changes in hemodynamic forces and if so, whether Ets‐1 controls microvascular endothelial cell integrity by inducing the expression of inhibitors of matrix degrading proteases. Rats were treated with prazosin for 2, 4, and 7 days to increase in microvascular shear stress in hindlimb skeletal muscles. In complimentary in vitro experiments, rat microvascular skeletal muscle endothelial cells were exposed to laminar shear stress (15 dyne/cm 2 ) for 0.5, 2, and 24 h. TaqMan PCR analysis of laser microdissected capillaries isolated from EDL muscles demonstrated transient (after 2 days) induction of Ets‐1 gene expression. In cultured cells, a transient up‐regulation of Ets‐1 mRNA was observed after 2 h shear stimulation, accompanied by increased phosphorylation of Ets‐1 and enhanced Ets‐1 DNA binding activity. This response was modulated by ERK1/2 and p38 MAP kinases, but was not dependent on NOS or COX‐2 activity. PAI‐1, TIMP‐1 and TIMP‐3 mRNA were elevated significantly in prazosin treated EDL, and in response to shear stimulation in vitro. In cultured endothelial cells, Ets‐1 RNA interference abolished the shear‐induced increases in Ets‐1, PAI‐1, TIMP‐1, and TIMP‐3 mRNA expression. These results suggest that enhanced laminar shear stress may act to preserve the integrity of microvascular walls in part through Ets‐1‐dependent induction of protease inhibitors. J. Cell. Physiol. 217: 502–510, 2008. © 2008 Wiley‐Liss, Inc.

Increased capillary shear stress in skeletal muscle causes luminal splitting. Signaling mechanisms initiating this form of angiogenesis are not known. We hypothesize that shear stress-dependent activation of vascular endothelial growth factor receptor 2 (VEGFR2) causes p38 and ERK1/2 phosphorylation. Skeletal muscle endothelial cells were sheared (12 dynes/cm2, 2 or 6 hrs) in the presence or absence of 10 μM VEGFR2 kinase inhibitor I or 30 μM LNNA (nitric oxide synthase inhibitor) then lysed. Cells were treated with 20 ng/ml vascular endothelial growth factor 165 (VEGF) for 30 minutes then lysed. Male Sprague-Dawley rats were administered prazosin (50 mg/L drinking water) to increase shear stress, the extensor digitorum longus was extracted at 2, 4 and 7 days. Lysates were analyzed by Western blot. In vitro, p38 and ERK1/2 phosphorylation increased at 2 hrs of shear stress but only p38 remained phosphorylated at 6 hrs. Sustained phosphorylation of p38 was not blocked by LNNA. VEGFR2 inhibition abrogated p38 but not ERK1/2 phosphorylation. Similarly, VEGF treatment of static cultures increased phosphorylation of p38 but not ERK1/2. In vivo, p38 phosphorylation was increased significantly in muscles exposed to increased shear stress for 7 days. Our data imply that VEGFR2 is a shear stress sensitive receptor necessary for p38 phosphorylation, and it may play a role in luminal splitting. Supported by CIHR.

VEGF and MMP protein production are both required for exercise‐induced capillary growth in skeletal muscle. The underlying process by which muscle activity initiates an angiogenic response is not established, but it is known that mechanical forces such as muscle stretch are involved. We hypothesized that stretch of skeletal muscle microvascular endothelial cells induces production of MMP‐2 and VEGF through a common signal pathway. Endothelial cells were grown on Bioflex plates and exposed to 10% static stretch for up to 24 h. MMP‐2 protein level was measured by gelatin zymography and VEGF, MMP‐2, and MT1‐MMP mRNA levels were quantified by real‐time quantitative PCR. ERK1/2 and JNK phosphorylation and VEGF protein levels were assessed by Western blotting. Effects of mitogen‐activated protein kinases (MAPKs) (ERK1/2, JNK) and reactive oxygen species (ROS) on stretch‐induced expression of MMP‐2 and VEGF were tested using pharmacological inhibitors. Stretching of endothelial cells for 24 h caused significant increases in MMP‐2 protein and mRNA level, but no change in MT1‐MMP mRNA. While MMP‐2 protein production was enhanced by H2O2 in unstretched cells, ROS inhibition during stretch did not diminish MMP‐2 mRNA or protein production. Inhibition of JNK suppressed stretch‐induced MMP‐2 protein and mRNA, but inhibition of ERK had no effect. In contrast, inhibition of ERK but not JNK attenuated the stretch‐induced increase in VEGF mRNA. Our results demonstrate that differential regulation of MMP‐2 and VEGF by MAPK signal pathways contribute to stretch‐induced activation of microvascular endothelial cells. J. Cell. Biochem. 100: 750–761, 2007. © 2006 Wiley‐Liss, Inc.