OUP user menu

Maintenance of adrenergic vascular tone by MMP transactivation of the EGFR requires PI3K and mitochondrial ATP synthesis

Prabhakara Reddy Nagareddy , Fung Lan Chow , Li Hao , Xiang Wang , Tamiko Nishimura , Kathleen M. MacLeod , John H. McNeill , Carlos Fernandez-Patron
DOI: http://dx.doi.org/10.1093/cvr/cvp230 368-377 First published online: 3 July 2009

Abstract

Aims G-protein-coupled receptors (GPCRs) modulate vascular tone, at least in part, via matrix metalloproteinase (MMP) transactivation of the epidermal growth factor receptor (EGFR). We previously have identified novel signalling pathways downstream of the EGFR suggestive of mitogen-activated protein kinase and mitochondrial redox control of vascular tone. In the present study, we examined whether MMP modulation of vascular tone involves phosphoinositide 3-kinase (PI3K) and mitochondrial ATP synthesis.

Methods and results To determine whether PI3K is required for the maintenance of adrenergic vascular tone, we first constricted rat small mesenteric arteries with phenylephrine (PE) and then perfused with PI3K inhibitors, LY294002 and wortmannin, both of which produced a dose-dependent vasodilatation. Next, to investigate whether MMPs modulate PI3K activity, we cultured rat aortic vascular smooth muscle cells (VSMCs) and stimulated them with GPCR agonists such as PE and angiotensin II. Inhibition of MMPs (by GM6001) or EGFR (by AG1478) or suppressing the expression of MMP-2 or MMP-7 or the EGFR by small interfering RNA blunted the PI3K phosphorylation of Akt induced by PE. Further, in VSMCs, PI3K inhibitors reduced the PE-induced increase in ATP synthesis and glucose transporter-4 translocation, an effect that was also observed with MMP and the EGFR inhibitors. Further, the PE-induced increase in ATP synthesis activated MMP-7 by mechanisms involving purinergic (P2X) receptors and calcium.

Conclusion These data suggest that the maintenance of adrenergic vascular tone by the MMP–EGFR pathway requires PI3K activation and ATP synthesis. Further, our data support the view that elevated levels of GPCR agonists exaggerate the MMP transactivation of EGFR response and contribute to enhanced vascular tone and development of cardiovascular disease such as hypertension.

  • Matrix metalloproteinase
  • Epidermal growth factor receptor
  • Phosphoinositide 3-kinase
  • ATP
  • Adrenergic tone

1. Introduction

Hypertension is a cardiovascular disease characterized by sustained high blood pressure with a complex and multifactorial aetiology.1 Factors initiating this condition in the general population remain unknown despite major research efforts. Whatever the initial cause, the common hallmarks of hypertension both in humans and in animal models are enhanced agonist-stimulated vasoconstriction, decreased vasodilatation, oxidative stress-associated endothelial dysfunction and hypertrophic growth, and remodelling of cardiac and vascular tissues.25 Although it is unclear how and why such apparently distinct pathological processes concur in hypertension, increasing evidence now suggests that elevated levels of vasoactive G-protein-coupled receptor (GPCR) agonists such as catecholamines, endothelin-1, and angiotensin II (AT II) may explain, at least in part, the development and progression of many hypertensive disorders.613

We have proposed that increased vascular tone, oxidative stress, and hypertrophic growth are interrelated processes largely signalled through a common pathway.11,14,15 An important step in this pathway is the transactivation of growth factor receptors, such as the epidermal growth factor receptor (EGFR), by matrix metalloproteinases (MMPs).16 MMPs act, at least in part, by shedding mature growth factors such as heparin-binding epidermal growth factor from transmembrane precursors, which then bind to the EGFR and phosphorylate tyrosine kinase. The activation of the EGFR leads to downstream activation of mitogen-activated protein kinases (MAPK) such as p38 (stress activated), NH2-terminal c-Jun kinase, extracellular signal-regulated kinase (ERK) 1 and ERK 2, and protein kinase B/Akt.13 Once activated, these kinases are translocated to the nucleus, where they bind to and phosphorylate nuclear transcription factors, stimulating gene transcription, protein synthesis, and cell growth.1720

Among these kinases, p38 MAPK and phosphoinositide 3-kinase (PI3K) are known to participate in the signalling events leading to vasoconstriction subsequent to the activation of the EGFR by MMPs. This is because inhibitors of p38 MAPK and PI3K but not ERK1/2 produced vasorelaxation in phenylephrine (PE)-constricted arteries.11 Of specific interest is the PI3K pathway because PI3K has been implicated not only in the modulation of contraction21,22 but also in metabolism and growth.23 Further, a recent study from our laboratory suggested that following EGFR transactivation, the ensuing signalling events led to a significant increase in the generation of reactive oxygen species in mitochondria, which could impact mitochondrial ATP synthesis.15 These studies support the view that under sustained GPCR stimulation, growth-promoting and metabolic pathways culminate to increase oxidative stress, hypertrophic growth, and vascular tone, contributing to abnormal haemodynamic outcomes such as hypertension. Here, we tested the hypothesis that PI3K signalling modulates vascular tone by regulating mitochondrial ATP synthesis. These actions in turn are regulated by MMP transactivation of the EGFR secondary to the activation of GPCR by agonists such as PE.

2. Methods

2.1 Microperfusion experiments to study changes in vascular tone

Animal protocols were conducted in accordance with institutional guidelines issued by the Canada Council on Animal Care and conform to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). Small (OD: ∼500 µm, ID: 250–300 µm) mesenteric arteries from male Sprague–Dawley rats were used in studies of vascular tone. The arteries were dissected and mounted on a microperfusion arteriograph (Danish MyoTechnology, Aarhus, Denmark). Changes in arterial outer diameter in response to drugs were monitored using a video camera and processed using VediView software (Danish MyoTechnology). See Supplementary material online for details.

2.2 Cell culture studies

2.2.1 Preparation of rat aortic vascular smooth muscle cells

Rat aortic vascular smooth muscle cells (VSMCs) were grown in complete Dulbecco's modified Eagle medium (DMEM; Invitrogen Life Technologies, Carlsbad, CA, USA) with 10% foetal bovine serum and 100 units/mL penicillin–streptomycin at 37°C and 95% O2/5% CO2. Cells from passages 2–6 were used in all the experiments. When cells were 80–90% confluent, they were starved overnight using DMEM media without growth factors and used for experiments. See Supplementary material online.

2.2.2 EGFR, MMP-2, and MMP-7 suppression in VSMCs by small interfering RNA oligonucleotides

Small interfering RNAs (siRNAs) specific to rat EGFR, MMP-2, and MMP-7 were purchased from Santa Cruz Biotech. For optimal siRNA transfection efficiency, the manufacturer's protocol was followed. See Supplementary materials online for details.

2.3 Measurement of MMP activity by substrate zymography

MMP-2 and MMP-7 activity in tissue and cell culture releasates was measured using gelatin (for MMP-2) or carboxymethyl-transferrin substrate (for MMP-7) zymography assays. See Supplementary material online for details.

2.4 Western blotting

VSMCs were homogenized using either a sonicator or a homogenizer in modified RIPA buffer containing 50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 1 mM PMSF, 1 mM EDTA, 5 µg/mL aprotinin, 5 µg/mL leupeptin, 1% Triton x-100, 1% sodium deoxycholate, and 0.1% SDS at 4°C. The homogenate was centrifuged at 10 000g at 4°C for 15 min, and the protein content of the supernatants was determined by the Bradford protein assay. Equal amounts of protein (15–30 µg) from each sample were separated by 8 or 10% SDS–PAGE and transferred to polyvinylidene difluoride membranes. See Supplementary material online for details.

2.5 ATP measurements

ATP measurements were made using a commercially available luciferin-luciferase based assay kit. See supplementary material online for details.

2.6 Confocal immunofluorescence imaging of phospho-Akt (Ser 473)

Immunofluorescence imaging of phospho Akt was done as described previously.11 See Supplementary material online for details.

2.7 Preparation of detergent-resistant plasma membrane fractions for detection of glucose transporter-4 in VSMCs

See Supplementary material online for details.

2.8 Statistical analysis

All values are expressed as mean ± SEM. The sample size in each group is denoted by n. Statistical analysis was performed using one-way analysis of variance followed by the Newman–Keuls test for multiple comparisons. GraphPad Prism (GraphPad Software, CA, USA) software was used for statistical analysis. For all results, the level of significance was set at P < 0.05.

3. Results

3.1 Agonist (PE)-induced activation of PI3K is required for maintenance of adrenergic vascular tone in rat small mesenteric arteries

To determine whether GPCR agonists such as PE activate PI3K, we isolated resistance-size small mesenteric arteries and stimulated them with PE (10 µM) for 30 min. As shown (Figure 1A), PE produced a significant increase in the activation of PI3K, as determined by the phosphorylation of Akt at Ser473, a PI3K substrate. We next studied the effect of PI3K inhibitors such as LY294002 and wortmannin on PE-mediated adrenergic vascular tone. As shown, both wortmannin (Figure 1B) and LY294002 (Figure 1C) produced a concentration-dependent vasodilatation. These observations, which are consistent with our previous work on the vasodilatatory actions of MMP and EGFR inhibitors, suggest that the maintenance of adrenergic vascular tone depends on PI3K activity.

Figure 1

Maintenance of PE-induced adrenergic vascular tone requires PI3K activation. (A) Confocal immunofluorescence microscopy of small mesenteric arteries showing Akt phosphorylation (Ser473) after exposure to PE (10 µM) for 30 min. (B) Representative trace (top panel) and quantitative analysis (bottom panel) of vasorelaxation of PE-constricted arteries in response to the luminal injections of bolus doses of wortmannin (50–500 pmol). All data are expressed as mean ± SEM (n= 3–4 independent experiments). *Different from baseline, #different from PE (P < 0.05). (C) Representative trace (top panel) and quantitative analysis (bottom panel) of vasorelaxation of PE-constricted arteries in response to the luminal injections of bolus doses of LY294002 (50–500 pmol). All data are expressed as mean ± SEM (n= 3–4 independent experiments). *Different from baseline, #different from PE (P < 0.05).

3.2 Inhibition of the MMP–EGFR pathway suppresses PI3K activation of Akt in VSMCs

To investigate whether the maintenance of adrenergic vascular tone by PI3K involved MMP transactivation of the EGFR, we stimulated VSMCs with PE (10 µM) in the presence or absence of GM6001 (25 µM), a broad-spectrum MMP inhibitor, and AG1478 (10 µM), a selective inhibitor of EGFR tyrosine kinase, for 30 min. As observed in intact arteries, the stimulation of α1-adreneroceptors significantly increased the phosphorylation of Akt, and this was blunted in the presence of MMP and EGFR inhibitors (Figure 2). Further, in VSMCs, suppressing the expression of the EGFR (∼40%) by siRNA (Figure 3A) not only reduced its activation (Figure 3B) but also significantly prevented the increase in phosphorylation of Akt by PE (Figure 3C). Likewise, the suppression of MMP-2 expression by ∼60% (see Supplementary material online, Figure S1) or MMP-7 expression by ∼50% (Figure 4) by their corresponding siRNAs prevented PE-induced activation of Akt.

Figure 2

Inhibition of the MMP–EGFR pathway suppresses PE-induced activation of PI3K in VSMCs. Representative western blot showing the expression of phospho-Akt (Ser473) in VSMCs stimulated with PE (10 µM) for 30 min in the presence of MMP (GM6001, 25 µM) and EGFR (AG1478, 10 µM) inhibitors. The densitometric values of phospho-Akt (Ser473) were normalized to corresponding total Akt densitometric values, and the relative band intensities are expressed as mean ± SEM (n= 4 independent experiments). GAPDH served as a loading control. *P < 0.05 compared with all other groups, #different from PE group.

Figure 3

Effect of the EGFR suppression on PE-induced phosphorylation of Akt in VSMCs. Rat aortic VSMCs were transfected with either MOCK or rat-specific EGFR siRNA and stimulated with PE (10 µM) for 30 min. (A) Representative western blot showing total EGFR expression, with GAPDH as a loading control in VSMCs treated with MOCK or EGFR siRNA. The densitometric values of EGFR were normalized to corresponding total GAPDH densitometric values, and the relative band intensities are expressed as mean ± SEM (n = 4). *Different from MOCK siRNA-treated groups (P < 0.05). (B) Representative western blot showing phospho-EGFR expression, with GAPDH as a loading control in VSMCs treated with MOCK or EGFR siRNA and stimulated with PE (10 µM) for 30 min. The densitometric values of phospho-EGFR (Tyr1173) were normalized to corresponding total GAPDH densitometric values and the relative band intensities are expressed as mean ± SEM (n = 4). *Different from all other groups (P < 0.05), #different from MOCK siRNA- and PE-treated cells (P < 0.05). (C) Representative western blot showing phospho-Akt (Ser473) levels in VSMCs treated with MOCK or EGFR siRNA and stimulated with PE (10 µM) for 30 min. The densitometric values of phospho-Akt (Ser473) were normalized to corresponding total Akt densitometric values, and the relative band intensities are expressed as mean ± SEM (n= 4–6 independent experiments). *Different from all other groups (P < 0.05), #different from MOCK siRNA- and PE-treated group (P < 0.05).

Figure 4

Effect of MMP-7 suppression on PE-induced phosphorylation of Akt in VSMCs. VSMCs were transfected with either MOCK- or MMP-7-specific siRNA and stimulated with PE (10 µM) for 30 min. (A) Representative western blot showing total MMP-7 expression, with GAPDH as a loading control in VSMCs treated with MOCK or MMP-7 siRNA. The densitometric values of MMP-7 were normalized to corresponding GAPDH densitometric values, and the relative band intensities are expressed as mean ± SEM (n = 4). *Different from MOCK siRNA-treated groups (P < 0.05). (B) Representative western blot showing phospho-Akt (Ser473) expression in VSMCs treated with MOCK or MMP-7 siRNA and stimulated with PE (10 µM) for 30 min. The densitometric values of phospho-Akt (Ser473) were normalized to corresponding total Akt densitometric values, and the relative band intensities are expressed as mean ± SEM (n= 4–6 independent experiments). *Different from all other groups (P < 0.05), #different from MOCK siRNA PE-treated group (P < 0.05).

3.3 Maintenance of adrenergic vascular tone requires mitochondrial ATP synthesis in VSMCs

We next examined the potential mechanisms involved in the maintenance of adrenergic vascular tone downstream of PI3K. We specifically wanted to determine whether the stimulation of α1-adrenoceptor signalling triggers mitochondrial ATP synthesis in VSMCs and whether the inhibition of PI3K and/or the MMP–EGFR pathway can modulate this response. The stimulation of α1-adrenoceptors with PE (10 µM) increased intracellular ATP levels in a concentration- and time-dependent manner, peaking at 10 min (see Supplementary material online, Figure S2). To confirm the generality of this finding, we compared PE (0.1–10 µM) effects with those of another GPCR agonist (AT II) on ATP synthesis. Similar to PE, AT II (0.1–10 µM) increased ATP synthesis in a time- and concentration-dependent manner (see Supplementary material online, Figure S2C). Oligomycin (1 µM), a mitochondrial ATP synthase inhibitor, not only inhibited PE- and AT II-induced increases in ATP synthesis, but also decreased basal ATP synthesis in resting cells (see Supplementary material online, Figure S2). These findings suggest that mitochondria are the major source of ATP in VSMCs both in normal resting and in agonist-stimulated working conditions. Interestingly, the PE-induced initial surge in ATP synthesis (until 10 min) was not reduced by inhibitors of PI3K (LY294002, 10 µM), the MMPs (GM6001, 25 µM), or the EGFR (AG1478, 10 µM). However, ATP levels declined more rapidly in the presence of PI3K and MMP–EGFR inhibitors compared with ATP levels in PE-stimulated cells in the absence of these inhibitors (Figure 5). These data are consistent with our observations that the inhibition of either PI3K (Figure 1) or MMPs or the EGFR11 had no effect on the initial rapid contractile response to PE, but relaxed the sustained vasoconstrictor response to this agonist. Furthermore, the inhibition of mitochondrial ATP synthesis (Figure 5B) also produced vasodilatation in PE-constricted arteries, suggesting that the maintenance of adrenergic vascular tone requires ATP synthesis by mechanisms involving the activation of MMPs, the EGFR, and PI3K. The vasodilatation produced by PI3K inhibitors or oligomycin was reversible and did not result from any damage or death of VSMCs due to the inhibition of mitochondrial ATP synthesis. This was confirmed by methacholine, which produced a vasorelaxant effect in PE-constricted arteries following the washout of PI3K inhibitors and oligomycin (data not shown).

Figure 5

(A) Effect of the inhibitors of MMP, the EGFR, and PI3K on PE-stimulated mitochondrial ATP synthesis in VSMCs. Serum-starved rat aortic VSMCs were grown in 96-well bioluminescent compatible plates and treated with inhibitors of MMP (GM6001, 25 µM), the EGFR (AG1478, 10 µM), and PI3K (LY294002, 10 µM) for 20 min, followed by stimulation with PE (10 µM) for an additional 60 min. Next, the cells were lysed using a somatic cell lysis reagent, and ATP was measured in the lysates using a luminometric ATP detection assay. All data are expressed as mean ± SEM (n= 4–6 independent experiments). *Different from control group at 10 min (P < 0.05), #different from PE-stimulated group at 20 min (P < 0.05). (B) Rat small mesenteric arteries were mounted on the microperfusion system. PE (10 µM) was added to the bath (advential side) to cause vasoconstriction, followed by luminal injection of bolus doses of oligomycin (50–500 pmol). Representative trace (top panel) and quantitative analysis (bottom panel) of vasorelaxation of PE-constricted arteries in response to the luminal injections of bolus doses of oligomycin (50–500 pmol). All data are expressed as mean ± SEM (n= 4 independent experiments). *Different from PE group in the absence of oligomycin (P < 0.05).

3.4 PE-induced glucose transporter-4 translocation is reduced by the inhibition of PI3K and the MMP–EGFR pathway in VSMCs

Having determined that agonist-induced increase in vascular tone is maintained by MMP transactivation of the EGFR via the activation of PI3K and mitochondrial ATP synthesis, we investigated the link between PI3K and ATP production in mitochondria. Because glucose is the predominant source of energy during vascular contraction,24,25 we tested whether PE induced the translocation of glucose transporter-4 (GLUT4), the major glucose transporter,26 to the plasma membrane. Further we examined whether PE-induced GLUT4 translocation was affected by GM6001, AG1478, and LY294002. We found that the stimulation of VSMCs with PE significantly increased the translocation of GLUT4 to plasma membrane, as determined by the increased expression of GLUT in detergent insoluble membrane fractions. However, the presence of MMP or EGFR or PI3K inhibitors reduced this translocation, as observed by decreased expression of GLUT4 in the membrane compared with cytosol fraction (Figure 6).

Figure 6

PE-stimulated GLUT4 translocation is reduced by the inhibitors of MMPs, the EGFR, and PI3K in VSMCs. Cells were incubated with inhibitors of MMPs (GM6001, 25 µM), the EGFR (Ag1478, 10 µM), or the PI3K (LY294002, 10 µM) for 20 min, followed by stimulation with PE (10 µM) for additional 40 min. Representative western blot showing GLUT4 expression in the membrane and cytosol fractions and pan cadherin as a marker of membrane fractions. The extent of GLUT4 translocation was measured by determining the ratio of GLUT4 expression in the membrane and cytosol fractions. All data are expressed as mean ± SEM (n = 4). *Different from all other groups (P < 0.05), #different from PE group (P < 0.05).

3.5 PE-induced mitochondrial ATP synthesis activates MMP-7 in arteries

Previously, we have reported that GPCR stimulation by agonists such as PE and AT II increases MMP-7 activity in arteries. Since these agonists also triggered ATP synthesis in VSMCs, we next examined whether ATP synthesis in response to PE has any influence on further activation of MMP-7 in vascular tissues. When either mitochondrial ATP synthesis was inhibited (by oligomycin, 1 µM) or when extracellularly released ATP was scavenged using apyrase (5 U/mL), the ability of PE to activate MMP-7 was significantly reduced in arteries (Figure 7A), suggesting a novel pathway of MMP activation in vascular tissues. Mimicking the effects of PE, the direct application of ATP (1 µM) exogenously to arteries promoted a rapid release and the activation of vascular MMP-7 (Figure 7B). Because ATP can be rapidly degraded to ADP, AMP, and other metabolites in the extracellular space by ectoenzymes, we next wanted to clarify whether the observed effect on MMP-7 activation was due to ATP or to its metabolites. The direct application of ADP (0.01–100 µM) failed to activate MMP-7 in vascular tissues (data not shown), suggesting that the activation of MMP-7 is by ATP and not by its metabolites.

Figure 7

PE-stimulated activation of MMP-7 requires mitochondrial ATP synthesis and release. (A) Rat superior mesenteric arteries (SMA) were incubated in the presence of oligomycin (1 µM) or apyrase (5 U/mL) for 20 min, followed by stimulation with PE (10 µM) for another 40 min. MMP-7 activity in the releasates from PE-incubated arteries was assessed by CM-transferrin zymography and quantitated as the ratio of active MMP-7 to pro-MMP-7. All data are expressed as mean ± SEM (n= 4–6 independent experiments). *Different from all other groups (P < 0.05), #different from PE-stimulated VSMCs in the absence of inhibitors (P < 0.05). (B) Rat arteries (SMA and aorta) were incubated in the presence of ATP (1 µM) for 40 min, followed by the measurement of MMP-7 activity in the releasates using CM-transferrin zymography. Quantitative analysis of MMP-7 activity as measured by the ratio of active MMP-7 to pro-MMP-7. All data are expressed as mean ± SEM (n= 4–6 independent experiments). *Different from control groups (P < 0.05).

3.6 Activation of MMP-7 by ATP involves purinergic P2X receptors and calcium

ATP activates a family of ionotropic receptors (P2X) as well as metabotropic receptors (P2Y), collectively called purinergic receptors. P2Y receptors couple to intracellular second messenger systems through heterotrimeric G-proteins,27 whereas the P2X receptors are ligand (ATP)-gated ion channel receptors.28 To clarify the involvement of purinergic receptor signalling in ATP-induced activation of MMP-7, we first tested the effect of ATP in the presence of suramin (100 µM), a non-selective blocker of P2X and P2Y receptors, and then in the presence of pyridoxal-phosphate-6-azophenyl-2′,4′-disulfonate (PPADS, 100 µM), a selective P2Y blocker. ATP-mediated MMP-7 activation was abolished when vascular tissues were incubated with suramin but not with PPADS, suggesting a possible role for P2X receptors (see Supplementary material online, Figure S3A). ATP binding to cell surface P2X receptors produces a rapid influx of divalent cations such as Ca2+. Since calcium is an important cofactor for a number of secretory events, we next studied the effect of Ca2+ inhibition in ATP-induced MMP-7 activation. In separate experiments, under conditions of either a Ca2+-free HEPES buffer (containing EDTA, 50 µM + EGTA, 50 µM) or in the presence of BAPTA-AM (100 µM), an intracellular Ca2+chelating agent, ATP did not activate MMP-7 (see Supplementary material online, Figure S3B). These observations suggest a novel mechanism by which ATP activates MMP-7 in vascular tissues involving Ca2+ and purinergic P2X receptors.

4. Discussion

Hypertension is a disease characterized by enhanced agonist-induced vasoconstriction, reduced vasodilatation, and increased smooth muscle cell growth. We hypothesized that the MMP–EGFR pathway modulates vascular tone, at least in part, via the activation of PI3K and the modulation of mitochondrial ATP synthesis. In the present study, we investigated the mechanisms by which MMP transactivation of the EGFR modulates adrenergic vascular tone in VSMCs. The key findings in the present study are (i) PE-induced stimulation of α1-adrenergic receptors triggers mitochondrial ATP synthesis and activates PI3K via the MMP–EGFR pathway in VSMCs; (ii) maintenance of adrenergic vascular tone requires PI3K activation and mitochondrial ATP synthesis; (iii) PE-induced ATP synthesis promotes further activation of MMPs such as MMP-7. These new findings are consistent with and expand our recent research on the role of MMP–EGFR pathway on adrenergic signalling in VSM.11,15

Various mechanisms have been proposed to explain the missing link between the transactivation of the EGFR by MMPs and the intracellular factors that regulate the contractile function in the VSMC. In our previous investigations, we examined the involvement of mitochondrial respiratory chain complexes I and III and proposed a possible role for mitochondrial ATP synthesis in the modulation of adrenergic vascular tone by the MMP–EGFR pathway.11,15 In the present study, we studied whether agonist-induced mitochondrial ATP synthesis is required for the maintenance of adrenergic vascular tone and whether this is modulated by PI3K. We found that the stimulation of vascular GPCRs by PE or AT II significantly increased intracellular ATP levels in a concentration- and time-dependent manner. The initial phase of ATP synthesis, characterized by a rapid and profound increase (<10 min) in ATP levels, was probably to meet the immediate energy requirements of actin–myosin filaments during contraction, and the later phase may help to sustain the contraction initiated by the agonist.

Our results showing increased ATP synthesis peaking at 10 min following the stimulation of α1-adrenoceptors are consistent with our previous observations of increased mitochondrial membrane potential, mitochondrial ROS production, phosphorylation of the EGFR, and vasoconstriction, all of which also peaked at 10 min.15 Because oligomycin, a mitochondrial ATP synthase inhibitor, completely abolished both the agonist-induced and basal ATP syntheses, the source of ATP could be attributed to mitochondria. Interestingly, MMP–EGFR and PI3K inhibitors did not inhibit the initial surge in mitochondrial ATP synthesis triggered by PE. Further in our previous studies, the inhibition of the EGFR did not prevent the phosphorylation of myosin light chain in response to adrenergic stimulation.11 Taken together, these data suggest that, under basal conditions, the MMP–EGFR pathway influences neither the initiation of vascular contraction nor the surge in ATP synthesis triggered by vasoconstrictors. Subsequent to initial increase in ATP synthesis, PI3K and MMP–EGFR inhibitors caused a rapid decline in the intracellular ATP levels. Therefore, the MMP–EGFR–PI3K pathway may influence vascular tone as well as ATP synthesis by way of providing substrates for sustained ATP synthesis. It is likely that the MMPs are activated secondary to the initial contraction of VSMCs or mitochondrial ATP synthesis by mechanisms involving intracellular calcium, generation of ROS, phosphorylation of Src, and ATP production.14,2932

The activation of P13-kinase is one of the downstream events associated with the phosphorylation of the EGFR in VSMCs that potentially could modulate mitochondrial ATP synthesis via Akt-GLUT4 signalling. PI3K is a lipid kinase that converts phosphatidylinositol 4,5-diphosphate to a more potent second messenger, phosphatidylinositol 3,4,5-triphosphate. This compound is essential for the translocation of Akt to the plasma membrane, where it is phosphorylated and activated by phosphoinositide-dependent kinases. Akt is known to regulate energy metabolism by multiple mechanisms, including increased expression and translocation of glucose transporters such as GLUT4.3335 The phosphorylation of Akt increases the expression and translocation of GLUT4 to the plasma membrane, which in turn facilitates the uptake of glucose, a major substrate in vascular metabolism. Indeed, our data demonstrate that adrenergic stimulation in isolated rat arteries and VSMCs causes the activation of PI3K downstream of the EGFR, as detected by the increased levels of phospho-Akt. The blockade of MMPs, the EGFR, or PI3K using both pharmacological inhibitors and siRNA blunted the phosphorylation of Akt (at Ser473) as well as the synthesis of ATP downstream of adrenergic receptors. Our data are consistent with the previous studies that have reported lower ATP levels in Akt-deficient cells.36 Further, we have also found that the inhibition of PI3K or ATP synthesis concentration-dependently inhibited adrenergic vascular tone in rat small mesenteric arteries. Vasorelaxation produced by P13-kinase inhibitors in the arteries seems to be independent of nitric oxide and other endothelium-derived vasoactive factors since we did not detect any difference in vascular responses either to PE or PI3K inhibitors (wortmannin and LY294002) between arteries with denuded and intact endothelium (data not shown). Specifically, the vasorelaxation profiles of PI3K inhibitors in endothelium-denuded arteries were similar in magnitude and time course as in intact arteries. These data suggest that PI3K inhibitors promote vasorelaxation by directly interacting with VSMCs independent of NO system. In a nutshell, all these findings can be explained by a mechanism (Figure 8) whereby, downstream of the adrenoceptors, MMP transactivation of the EGFR results in PI3K phosphorylation of Akt, recruitment and translocation of GLUT4 to the plasma membrane, and increased glucose uptake to provide a continuous source of substrates for ATP synthesis and to maintain vascular tone. In support of our interpretation, previous studies using GLUT4 inhibitors have demonstrated decreased glucose uptake and attenuated vascular responses to adrenergic stimulation in vascular tissues.25,37 The detailed study of glucose transport is, however, beyond the scope of the present investigation.

Figure 8

Proposed mechanism for agonist-induced signalling of vascular tone, hypertrophic growth, and remodelling. Activation of vascular Gq-protein-coupled receptors (such as α1-adrenergic receptors or angiotensin receptors) by cognate agonists transactivates MMPs (such as MMP-7) and, thereby, growth factor receptors (such as EGFR). These transactivation events activate PI3K. Activation of PI3K results in the phosphorylation of Akt and the recruitment of glucose transporters such as GLUT4, which, in turn, provides a continuous supply of substrates (glucose) for mitochondrial ATP synthesis. The provision of substrates helps sustain an elevated vascular tone initiated by vasoactive agonists for longer periods of time. Additionally, the ATP synthesized and released into the extracellular milieu activates MMPs such as MMP-7 to maintain the feed-forward cycle, resulting in sustained ATP synthesis and vascular tone. Activation of MMPs and ATP synthesis either independently or synergistically contribute to the regulation of vascular tone, growth, and remodelling processes initiated by vasoconstrictory GPCR agonists.

Mitochondrial ATP synthesis in response to agonist stimulation was also found to activate MMPs such as MMP-7, although the mechanisms remain unclear. The observation that apyrase, an ATP scavenger, inhibits the activation of MMP-7 suggests that the activation of MMP-7 requires ATP to be released to the extracellular milieu. In spite of very high concentrations in the cytosol (∼3–10 mM), ATP and other nucleotides cannot penetrate cell membranes because of their negative charge. However, because of its high concentration gradient (106-fold across the membrane), ATP can activate certain nucleotide transporters and channels to release ATP.3840 Further, the released ATP has a very short life and is rapidly degraded to ADP, AMP, and adenosine by ectonucleotidases. Our data suggest that it was ATP and not the degradation products that activated MMP-7 because the direct stimulation of arteries with ADP did not activate MMP-7.

We next studied the potential mechanisms by which ATP activates MMP-7. Because ATP activates both ionotropic P2X and metabotropic P2Y receptors,41 we studied the role of these receptors in MMP-7 activation. P2X purinoceptors are found on VSMCs, where they mediate vasoconstriction resulting from ATP released as a co-transmitter with norepinephrine from sympathetic nerves. P2Y purinoceptors, however, are located on the vascular endothelium and mediate vasorelaxation to locally produced ATP.42 Although, our data suggest that MMP-7 activation is mediated by P2X receptors and involves calcium, it is possible that the inhibition of MMP-7 activity may be due to the non-specific effects of the calcium antagonists used. Since MMPs require metal ions such as calcium and zinc for their catalytic activity, the calcium-chelating agents used in the present study might have inhibited MMP activity.43 Further studies are required to elucidate the mechanisms of MMP-7 activation by ATP. Regardless of the mechanism involved, the activation of MMP-7 by ATP released as a result of agonist stimulation of vascular GPCRs is of prime importance because it initiates a feed-forward cycle involving GPCR–MMP–EGFR–ATP and MMP to sustain vascular tone11,15 as well as growth and remodelling.44 In addition, extracellular ATP is mitogenic and stimulates several pathways either directly or synergistically with other polypeptide growth factors such as platelet-derived growth factor, insulin-like growth factor-1, EGF, and insulin resulting in VSM proliferation and hypertrophy.44

In conclusion, our data suggest that agonist-induced stimulation of GPCRs such as adrenergic receptors causes vasoconstriction possibly as a result of the initial surge in mitochondrial ATP synthesis. This is followed by the activation of PI3K via the mechanisms involving MMPs and subsequent transactivation of the EGFR. The activation of PI3K results in the phosphorylation of Akt and the recruitment of glucose transporters such as GLUT4, which, in turn, provides a continuous supply of substrates for mitochondrial ATP synthesis. The provision of substrates such as glucose might help to sustain the vascular tone initiated by vasoactive agonists for longer periods of time. Additionally, the ATP synthesized and released into the extracellular milieu activates MMPs such as MMP-7 to maintain the feed-forward cycle of sustained ATP synthesis and hence vascular tone.

In hypertension, the characteristically elevated levels of GPCR agonists such as NE, AT II, and ET-1 may result in exaggerated transactivation of the vascular MMP–EGFR pathway culminating in pathological features such as enhanced vascular tone and hypertrophic growth. The selective inhibition of the vascular MMP–EGFR transactivation pathway could have therapeutic potential for decreasing agonist-induced activation of PI3K and for attenuating vasoconstriction and hypertrophic growth in hypertensive disorders.

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Conflict of interest: none declared.

Funding

This work was supported by the Canadian Institutes of Health Research (No. G118160401) and new investigator awards from CIHR to C.F.-P and the Heart and Stroke Foundation of Canada (HSFC) to C.F.-P and J.H.M. P.R.N. was supported by doctoral research awards from the HSFC and the MSFHR.

References

View Abstract