Through the Epidermal Growth Factor Agoni

Agonist-Induced Activation of Matrix Metalloproteinase-7 Promotes Vasoconstriction Through the Epidermal Growth Factor −Receptor Pathway Li Hao, Min Du, Ana Lopez-Campistrous and Carlos Fernandez-Patron Circ Res. 2004;94:68-76; originally published online December 1, 2003; doi: 10.1161/01.RES.0000109413.57726.91 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2003 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/94/1/68

Data Supplement (unedited) at: http://circres.ahajournals.org/content/suppl/2004/01/06/94.1.68.DC1.html

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/

Downloaded from http://circres.ahajournals.org/ by guest on June 7, 2013

Agonist-Induced Activation of Matrix Metalloproteinase-7 Promotes Vasoconstriction Through the Epidermal Growth Factor–Receptor Pathway Li Hao,* Min Du,* Ana Lopez-Campistrous, Carlos Fernandez-Patron Abstract—Matrix metalloproteinase (MMP)-dependent shedding of heparin-binding epidermal growth factor (HB-EGF) and subsequent activation of the EGF receptor (EGFR) in the cardiovasculature is emerging as a unique mechanism signaling growth effects of diverse G protein– coupled receptors (GPCRs). Among these GPCRs are adrenoceptors and angiotensin receptors that contribute to the pathogenesis of hypertension through their vasoconstrictive and growth effects. Focusing on ␣1b-adrenoceptors, we suggest here that MMP-dependent activation of the EGFR promotes vasoconstriction as well as growth. We identified MMP-7 as a major HB-EGF sheddase in rat mesenteric arteries and ␣1b-adrenoceptors, angiotensin receptors, and hypertension-stimulated MMP-7 activity. Adrenoceptors stimulated EGFR autophosphorylation in arteries, and this transactivation was opposed by the MMP-7 inhibitor GM6001 as well as MMP-7–specific antibodies. In isolated microperfused arteries, blockade of EGFR transactivation with inhibitors of the EGFR (AG1478 and PD153035), HB-EGF (CRM197 and neutralizing antibodies), or MMPs (doxycycline) inhibited adrenergic vasoconstriction. In spontaneously hypertensive rats but not in normotensive rats, the inhibition of MMPs with doxycycline (19.2 mg/d from week 7 until week 12) reduced systolic blood pressure and attenuated HB-EGF shedding in the mesenteric arteries. These findings suggest a previously unknown mechanism of vasoregulation whereby agonists of certain GPCRs (such as adrenoceptors and angiotensin receptors) activate MMPs (such as MMP-7) that shed EGFR ligands (such as HB-EGF), which then activate the EGFR, thereby promoting vasoconstriction as well as growth. Because this mechanism is triggered by agonists typically overexpressed in hypertension, its blockade may have therapeutic potential for simultaneously inhibiting pathological vasoconstriction and growth in hypertensive disorders. (Circ Res. 2004;94:68-76.) Key Words: matrix metalloproteinase 䡲 heparin-binding epidermal growth factor 䡲 epidermal growth factor receptor 䡲 vasoconstriction 䡲 hypertension

T

here is increasing evidence implicating matrix metalloproteinases (MMPs) and metalloproteinase disintegrins (ADAMs) in shedding of growth factors (eg, heparin-binding epidermal growth factor [HB-EGF]) and thereby transactivation of cognate growth factor receptors (eg, EGF receptor) in the development of hypertrophy associated with hypertension.1–7 In the heart, one mechanism of hypertrophy is transactivation of the EGF receptor by G protein– coupled receptors (GPCRs), such as adrenoceptors, angiotensin, and endothelin receptors, whose agonists (catecholamines, angiotensin II, and endothelins) are typically overexpressed as well as being historically implicated in the initiation, progression, and development of hypertensive disorders.4,8 –12 These GPCRs transactivate the EGF receptor (EGFR) in cardiomyocytes via a shared pathway, whereby ADAM 12 sheds membrane-

anchored HB-EGF, which then binds to the EGFR either directly or via an interaction with the proteoglycan matrix or cell membrane molecules (such as CD44) to signal growth.4,8 –10,13 In arteries, the HB-EGF sheddase responsible for EGFR transactivation is yet to be identified, and it is conceivable that more than one protease serves this function. Indeed, several matrix metalloproteinases present in arteries (MMP-2, MMP-3, and MMP-7) as well as metalloproteinase disintegrins (Kuzbanian/ADAM-10, ADAM-12, and TACE/ ADAM-17) shed HB-EGF in fibroblasts and epithelial and endothelial cells, contributing to EGFR transactivation in these systems.4,13–19 To date, most studies of EGFR transactivation by vasoactive GPCRs have centered on the growth effects of these GPCRs and their potential impact on development of cardio-

Original received June 4, 2003; first resubmission received August 7, 2003; second resubmission received October 22, 2003; revised second resubmission received November 11, 2003; accepted November 14, 2003. From the Department of Biochemistry, University of Alberta, Edmonton, Alberta, Canada. *Both authors contributed equally to this study. Correspondence to Dr Carlos Fernandez-Patron, Department of Biochemistry, 3-08 Medical Sciences Building, University of Alberta, Edmonton, Alberta T6G 2H7, Canada. E-mail [email protected] © 2004 American Heart Association, Inc. Circulation Research is available at http://www.circresaha.org

DOI: 10.1161/01.RES.0000109413.57726.91

68 Downloaded from http://circres.ahajournals.org/ by guest on June 7, 2013

Hao et al vascular hypertrophy.2,4,5 However, it is conceivable that EGFR transactivation modulates vascular tone as well as growth.20 –23 If this hypothesis were true, then blockers of EGFR transactivation (eg, MMP inhibitors) could have therapeutic potential for simultaneously inhibiting pathological vasoconstriction and growth in hypertensive disorders, where vasoconstrictive agonists capable of transactivating the EGFR, such as catecholamines and angiotensin II, are typically overexpressed and contribute to disease.1,2,4,11,12,24 Therefore, we investigated whether GPCR-mediated, MMPdependent transactivation of the EGFR modulates vascular tone of arteries. We first tested our hypothesis by focusing on ␣1b-adrenoceptors as a model of vasoactive GPCR. Next, we identified a major MMP capable of shedding HB-EGF in arteries and studied its regulation by vasoactive GPCRs involved in the pathogenesis of hypertension. Finally, we investigated the activity profile of this MMP in a genetic model of hypertension and explored whether MMP inhibitor therapy impacted blood pressure. Our findings implicate, for the first time, the MMP-dependent pathway of HB-EGF shedding (and thereby EGFR transactivation) in maintenance of vascular tone and suggest a role for this pathway in the modulation of blood pressure in hypertension.

Materials and Methods Microperfusion Bioassay Animal protocols were conducted in accordance with institutional guidelines issued by the Canada Council on Animal Care. Mesenteric arteries from male Sprague-Dawley rats (300 to 500 g; Charles River Laboratories, Wilmington, Mass) were used in studies of vascular tone. The arteries were dissected and mounted on a microperfusion arteriograph (Danish MyoTechnology and Living Systems Instrumentation). This perfusion system facilitated the study of vascular reactivity to luminal infusions of drugs as well as to drugs added to the arteriograph bath (adventitia side).18,19 The arteries were perfused at constant temperature (37°C) and flow rate (2 ␮L/min), with standard HEPES-PSS, pH 7.4, supplemented with glucose (5.5 mmol/L). In experiments involving the luminal administration of drugs, small volumes (5 ␮L) of specified drugs were injected into the perfusion line toward the artery. Changes in arterial outer diameter in response to drugs were monitored using a videocamera and processed using the VediView software (Danish MyoTechnology). In this bioassay, maximum of arterial reactivity (ie, the time-dependent change in arterial diameter) typically coincides with the infused drug being in the lumen of the artery. As the drug bolus leaves the artery, the arterial diameter returns to the magnitude it had before drug infusion. Maximum of reactivity was generally used for analysis of drug effects. The injection, without introducing flow rate-change–related artifacts, was facilitated by using a HPLC injection valve (Rheodyne Model 9725I, Mandel Scientific Co).18,19

Solid-Phase HB-EGF Shedding Assay An excess of purified pro-HB-EGF (⬇3 mg) was allowed to saturate heparin beads resuspended in HEPES-PSS. Ten microliters of pro-HB-EGF– coated heparin bead suspension was incubated under gentle vortexing (37°C) for 2 hours in the absence or presence of mesenteric artery lysate at a dilution (⬇1 ␮g of lysate protein) where endogenous HB-EGF immunoreactivity was virtually undetectable by Western blotting using enhanced chemiluminescence detection. The supernatant of the slurry containing HB-EGF shed from the heparin beads was applied to a 15% SDS-PAGE gel and analyzed by Western blotting with neutralizing HB-EGF antibodies.

Vasoconstriction via EGFR Transactivation

69

Substrate Gel Zymography For determining the molecular weight and tissue distribution of HB-EGF processing activities, lysates of rat mesenteric artery, aortae, heart, or kidney (50 ␮g each) were supplemented with heparin (10 ␮g) and subjected to electrophoresis on 7.5% SDSPAGE gels copolymerized with either reduced and carboxymethylated (CM) transferrin13 or pro-HB-EGF (2.5 mg/mL). After electrophoresis, the gel was washed with Triton X-100 (2.5%) (3⫻20 minutes). The gelatinolytic activity was developed by first incubating the gel for 16 hours (37°C) in enzyme assay buffer containing (in mmol/L) Tris 25, CaCl2 5, NaCl 142, and Na3N 0.5, supplemented with benzamidine (1 mmol/L, to inhibit potential serine proteases in the samples). Subsequently, the gel was stained with Coomassie blue. Proteases were detected as transparent bands against the background of Coomassie blue–stained undigested proHB-EGF (or CM transferrin). An expanded Materials and Methods section can be found in the online data supplement available at http://www.circresaha.org.

Results MMP-Dependent EGFR Transactivation Sustains Adrenergic Vasoconstriction The stimulation of rat mesenteric arteries with phenylephrine, a selective agonist of ␣1b-adrenoceptors, promoted EGFR autophosphorylation (at Tyr1173), as determined by immunofluorescence microscopy and Western blotting with EGFR phospho-specific antibodies (Figure 1). This transactivation of the EGFR was inhibited by the EGFR-selective tyrphostin AG 1478 as well as the MMP inhibitor GM 6001. Therefore, ␣1b-adrenoceptors use an MMP activity to transactivate the EGFR in arteries. To determine whether ␣1b-adrenoceptors use the EGFR for signaling of vasoconstriction, we mounted small mesenteric arteries on a microperfusion arteriograph18,19 (Figure 2). Phenylephrine (from 1 to 10 ␮mol/L), when added to the arterial bath (adventitia side), promoted a sharp and longlasting (⬇2-hour) contraction of the perfused arteries.19 In the presence of the EGFR kinase–specific tyrphostin AG 1478, the arteries constricted, but this constriction faded in a time-dependent manner (Figures 2A and 2C). Another EGFR kinase inhibitor, PD 153035, also concentration-dependently decreased phenylephrine vasoconstriction by 31.1⫾3.6% at 10 ␮mol/L and 100⫾0.98% at 50 ␮mol/L (P⬍0.05 versus phenylephrine, n⫽3). By contrast, inhibition of the EGFR with AG 1478 did not affect the contractile response to KCl that promotes vasoconstriction in a receptor-independent manner (Figure 2C). To determine whether, once transactivated, the EGFR contributes to maintenance of agonist-induced tone, we studied already-constricted, perfused arteries. When the arteries were constricted with phenylephrine, the luminal infusion of AG 1478 (online Figure 1a, available at http://www.circresaha.org), but not inactive tyrphostin AG 9 (online Figure 1b), stimulated vasorelaxation. Neither AG 1478 nor AG 9 infusions affected the tone of arteries constricted to a similar extent with KCl (online Figures 1c and 1d). Because KCl promotes vasoconstriction through a receptor-independent mechanism involving smooth muscle depolarization, the data suggest that the EGFR does not modulate receptor-independent vasoconstriction. However, once transactivated by ␣1b-adrenoceptors,


70

Circulation Research

January 9/23, 2004

Figure 1. ␣1b-Adrenoceptors use an MMP to transactivate the EGFR in arteries. A, Immunofluorescence microscopy of arteries showing EGFR phosphorylation at tyrosine 1173 after exposure to phenylephrine for 30 minutes. This phosphorylation was attenuated when the arteries were preincubated during 2 hours with the MMP and EGFR kinase inhibitors GM 6001 and AG 1478, respectively. B, Representative blots and quantitative Western blotting analysis confirming that transactivation of EGFR by ␣1b-adrenoceptors depends on MMP activity and EGFR kinase activity. Results (mean⫾SEM) are representative of 4 independent experiments for every data point. *Pⱕ0.05 vs control (HEPES-PSS). Rat strain: Sprague-Dawley.

the EGFR contributes to maintenance of vascular adrenergic tone. Phosphorylation of myosin light chain (MLC) is an important event in vascular smooth muscle contractile response. To assess whether EGFR transactivation affected phenylephrineinduced phosphorylation of MLC, we stimulated arteries with phenylephrine (10 ␮mol/L) in the absence (control) or presence of the EGFR kinase inhibitor AG 1478 (10 ␮mol/L) under conditions that resulted in EGFR transactivation (Fig-

Figure 2. EGFR signals ␣1b-adrenergic vasoconstriction. Representative trace of the time course (A) and quantitative analysis (B) of the contractile response of arteries constricted with phenylephrine (PE) in the absence and presence of EGFR kinase inhibitor AG 1478. 䡩, Representative time point when arterial diameter was acquired for quantitative analysis. C, Comparative analysis of the time course of contractile responses of arteries in the presence and absence of AG 1478 showed an involvement of the EGFR in ␣1b-adrenoceptor– dependent (PE), but not in receptor–independent (KCl-induced), contraction. Results (mean⫾SEM) are representative of 4 independent experiments for every data point. *Pⱕ0.05 vs control (PE). Rat strain: Sprague-Dawley.

ure 1 and online Figure 2). The arteries were snap frozen and lysed, and lysates containing equal amounts of protein were analyzed by Western blotting with antibodies to phosphoMLC (Thr18 and Ser19). The experiment indicated that adrenoceptors phosphorylate MLC, and this phosphorylation was not inhibited by the EGFR kinase inhibitor AG 1478 (online Figure 2). To confirm these data, we replaced phenylephrine with angiotensin II (100 nmol/L), reasoning that angiotensin II and phenylephrine would affect MLC phosphorylation through common pathways (because both bind to GPCRs as well as transactivate the EGFR in the vascular smooth muscle). Indeed, angiotensin II reproduced the observations we made with phenylephrine, promoting MLC phosphorylation that was not inhibited by AG 1478 (online Figure 2). Therefore, the data suggest that EGFR transactivation by agonists of GPCRs modulates vasoconstriction through MLC-phosphorylation–independent signaling events. To clarify signaling events downstream of the EGFR, we conducted pharmacological and immunofluorescence microscopy experiments aimed at measuring the contribution of kinases positioned downstream of the EGFR to adrenergic vasoconstriction (n⫽3 to 4, data not shown). These results indicated that, once transactivated, the EGFR triggered the phosphorylation (and thereby activation) of p38, Akt, and extracellular signal–regulated kinase (Erk) 1 and Erk2. Interestingly, when infused into phenylephrine-constricted arteries, inhibitors of p38 (SB 203580) and Akt (LY 294002 and wortmannin) but not MEK1/2, a kinase of Erk1/2 (PD98059 and U0126), promoted vasorelaxation. Thus, on EGFR transactivation, p38 and Akt are likely to contribute to adrenergic tone in the arteries studied as well as being involved in other EGFR reactions, such as transcription activation.

HB-EGF Signals Adrenergic Vasoconstriction We next studied whether EGFR ligands, such as HB-EGF, were present in the arteries and able to modulate vascular


Hao et al


71

Figure 3. Identification, tissue localization, and vascular tone effects of HB-EGF in rat mesenteric arteries. A, Western blotting revealed the presence of HB-EGF in arteries. B, Immunofluorescence microscopy with neutralizing antibodies showing the localization of active HB-EGF in arteries. IgG indicates control arterial section was incubated with nonimmune IgG and FITC-labeled secondary antibody (to determine background fluorescence). C, Luminally administrated recombinant human HB-EGF into basally perfused arteries promoted vasoconstriction. Results (mean⫾SEM) are representative of 4 independent experiments for every data point. *Pⱕ0.05 vs baseline. Rat strain: Sprague-Dawley.

tone. HB-EGF immunoreactivity was detected in the wall of arteries, including endothelial, smooth muscle, and adventitia cells (Figures 3A and 3B). Direct activation of the EGFR by luminal bolus infusion of a soluble recombinant form of human HB-EGF promoted vasoconstriction of basally perfused arteries (Figure 3C). To determine whether endogenous HB-EGF contributed to adrenergic vasoconstriction, we used two unrelated HB-EGF inhibitors, neutralizing HB-EGF antibodies and CRM 197,10 a high-affinity ligand of HB-EGF. In the presence of CRM 197, phenylephrine-evoked vasoconstriction faded in a time-dependent manner (Figures 4A and 4B), mimicking the results previously observed with the EGFR kinase inhibitor AG 1478 (Figure 2). CRM 197 was a weak inhibitor of the contractile response to KCl, suggesting a relatively greater role for HB-EGF in ␣-adrenoceptor– dependent vasoconstriction than in receptor-independent (ie, KCl) vasoconstriction (Figures 4C and 4D). Like AG 1478, luminal injections of CRM 197 resulted in dose-dependent vasodilatation of phenylephrine-preconstricted arteries (Figures 5A and 5B). The luminal administration of a neutralizing HB-EGF antibody also relaxed phenylephrine-preconstricted arteries (Figure 5C). Therefore, similar to the EGFR, there

was a requirement for HB-EGF for maintenance of ␣1badrenergic tone in arteries. To our knowledge, these data are the first reported evidence suggesting that adrenoceptors use arterial HB-EGF to sustain vasoconstriction. Taken together, these data implicate an MMP-dependent, HB-EGF–mediated pathway of EGFR transactivation in maintenance of ␣1badrenergic tone.

MMP-7 Is a Major HB-EGF Sheddase in Arteries To determine whether arteries contained MMPs capable of processing HB-EGF, we first used a solid-phase HB-EGF shedding assay. Incubation of arterial lysates with pro-HBEGF immobilized on heparin beads resulted in the release of a soluble, small-molecular-weight HB-EGF (⬇10 to 19 kDa). This HB-EGF molecule contained the EGFR binding domain recognized by neutralizing HB-EGF antibody (Figure 6A). To determine the molecular weight of the HB-EGF–processing proteases in the arteries, we developed a substrate-gel zymography whereby pro-HB-EGF was copolymerized in polyacrylamide gel. The sheddases in arterial lysates were electrophoretically separated on pro-HB-EGF gel and then allowed to digest the pro-HB-EGF. This analysis revealed

Figure 4. HB-EGF signals ␣1b-adrenergic vasoconstriction. Representative trace (A) and quantitative analysis (B) of contractile response of arteries stimulated with phenylephrine in the absence and presence of HB-EGF inhibitor (CRM 197). For comparison, representative trace (C) and quantitative analysis (D) of contractile response of arteries stimulated with KCl in the absence and presence of CRM 197. Traces and results (mean⫾SEM) are representative of 3 or 4 independent experiments for every data point. 䡩, Representative time point when arterial diameter was acquired for quantitative analysis. *Pⱕ0.05 vs PE (or KCl). Rat strain: Sprague-Dawley.


72


January 9/23, 2004

Figure 5. Pharmacological inhibition of HB-EGF promotes vasorelaxation of agonist-preconstricted arteries. Representative trace (A) and quantitative analysis (B) of vasorelaxation of phenylephrine-constricted perfused arteries in response to the luminal infusions of CRM 197. C, Representative trace of vasorelaxation of phenylephrine-constricted arteries in response to luminal infusions of neutralizing HB-EGF antibody (antiHB-EGF) versus nonimmune IgG. 䡩, Representative time point when arterial diameter was acquired for quantitative analysis. Traces and results (mean⫾SEM) are representative of 3 or 4 independent experiments for every data point. *Pⱕ0.05 vs PE. Rat strain: Sprague-Dawley.

several activities capable of cleaving pro-HB-EGF to yield HB-EGF polypeptides small enough to diffuse through the pores of the polyacrylamide matrix (pore size cutoff, ⬇10 to 20 kDa) during zymographic development (Figure 6B). The strongest pro-HB-EGF processing activity was, however, a small protein of ⬇29 kDa (Figure 6B). Zymographic analysis with CM-transferrin (a substrate of MMP-7) showed that the molecular weight and tissue distribution of the 29-kDa pro-HB-EGF processing activity in arteries were indistinguishable from those of MMP-713 (Figure 6B). Indeed, it was strongly present in mesenteric arteries and aortae but not in the kidney, and its activity was weak in the heart, where ADAM-12 (92 kDa) is the major HB-EGF sheddase.4 Western analysis and immunofluorescence microscopy showed that, similar to HB-EGF (Figure 3B), MMP-7 localized to smooth muscle and adventitia of arteries (Figure 6D). Confirming that MMP-7 and HB-EGF interacted in the arteries, HB-EGF antibodies immunoprecipitated arterial MMP-7 (Figure 6E, left) and MMP-7 antibodies immunoprecipitated arterial HB-EGF (Figure 6E, right). Importantly, ␣1badrenoceptors stimulated MMP-7. Incubation of arteries with phenylephrine resulted in a rapid release of active MMP-7 (Figure 6F). This release was noticeable at 5 minutes (data not shown) and became significant after 30 minutes (Figure 6F). This observation suggested that MMP-7 activation may contribute to phenylephrine-induced transactivation of the EGFR. In line with this notion, the MMP inhibitor, GM 6001, attenuated phenylephrine-induced MMP-7 activation in arteries (Figure 6F) and blocked MMP-7– dependent cleavage of pro-HB-EGF in vitro (online Figure 3) as well as reducing the amount of EGFR transactivation in the arteries (Figure 1). Furthermore, MMP-7 antibodies, but not nonimmune IgG, also attenuated the phosphorylation of the EGFR induced by phenylephrine, confirming that MMP-7 indeed modulates EGFR transactivation by adrenoceptors (Figure 6G). Interestingly, the vasoconstrictive agonist, angiotensin II, also promoted MMP-7 release and activation, whereas vasodilatory agonists such as thrombin and calcitonin gene– related peptide did not (online Figure 4). Furthermore, MMP-7 was found to be upstream of the EGFR, because its

release by agonists (phenylephrine and angiotensin II) was not affected by inhibition of the EGFR with AG 1478 (data not shown). Taken together, the data strongly suggest that MMP-7 is a major HB-EGF sheddase in arteries, which is supported by a previous report that MMP-7 is a major HB-EGF sheddase in rat uterus.13 Furthermore, MMP-7 activity is stimulated by agonists of certain vasoconstrictive GPCRs, such as adrenoceptors and angiotensin receptors. Our analysis on HB-EGF substrate zymography (Figure 6B) suggested the presence in arteries of minor bands of proteinases capable of processing pro-HB-EGF but distinct from MMP-7. Because their apparent molecular weights ranged from 40 to 100 kDa, we considered two potential HB-EGF sheddases, MMP-2 (62 kDa gelatinase, HB-EGFsheddase in epithelial, and endothelial cells14) and ADAM 12 (92 kDa, major HB-EGF sheddase in cardiomyocytes4). We compared the stability of pro-HB-EGF in vitro using arterial lysates supplemented with nonimmune IgG or antibodies against MMP-2 or ADAM 12. After a period of incubation (2 hours at 37°C), the arterial lysates were analyzed by Western blotting with HB-EGF–specific antibodies. We observed that the MMP-2 and ADAM 12 antibodies, but not nonimmuneIgG, protected HB-EGF from proteolytic processing (online Figure 5). These data suggested that HB-EGF processing in arteries may involve the concerted activity of various MMPs (such as MMP-7 and MMP-2) as well as ADAMs (such as ADAM 12).

Differential Regulation of Systolic Blood Pressure Through the MMP/HB-EGF System in Normotension and Hypertension We next compared the activity profile of HB-EGF–processing enzymes in arteries from hypertensive and normotensive rats. Zymographic analysis on pro-HB-EGF gels (Figure 7A) revealed an increased activity of pro-HB-EGF processing enzymes, including MMP-7, in mesenteric arteries from 12-week-old spontaneously hypertensive rats, as opposed to the arteries of two age-matched normotensive rat strains (WKY and Sprague-Dawley) (Figure 7A). Additional analysis (online Figure 6) indicated that pro-HB-EGF processing


Hao et al


73

Figure 6. Identification of MMP-7 as a major HB-EGF sheddase in arteries. A, Solid-phase HB-EGF shedding assay reveals that mesenteric artery lysates (mes. art. lys.) have an activity capable of shedding HB-EGF from pro-HB-EGF immobilized on heparin beads. B, Relative molecular weight and tissue distribution of major HB-EGF sheddases in lysates of mesenteric arteries, aortae, hearts, and kidneys, as determined using zymography with pro-HB-EGF (a substrate of HB-EGF sheddases) and CM-transferrin (a substrate of MMP7). HMW indicates high-molecular-weight bands with pro-HB-EGF processing activity; arrowhead, low-molecular-weight activity consistent with pro-MMP-7. C, Western blotting showing the presence of MMP-7 in lysate of mesenteric arteries. Comparison with partially activated recombinant human (rh) MMP-7. D, Immunofluorescence microscopy revealing MMP-7 immunoreactivity in vascular smooth muscle and adventitia of rat mesenteric arteries. FITC-labeled secondary antibody. E, left, HB-EGF antibodies immunoprecipitate (IP) MMP-7 from mesenteric artery lysates. CM-transferrin zymography was used to detect MMP-7 activity pulled down using protein G-agarose beads coated with neutralizing HB-EGF (⫹) or nonimmune IgG (⫺). Right, MMP-7 antibodies (⫹) effectively immunoprecipitated HB-EGF from arteries, whereas nonimmune IgG (⫺) only resulted in background binding of HB-EGF to the protein G-agarose beads. F, Phenylephrine induced the release and activation of arterial MMP-7. This activation was attenuated by the broad-spectrum MMP inhibitor GM 6001. *Pⱕ0.05 vs control; #Pⱕ0.05 vs PE⫹GM 6001. G, MMP-7 antibodies, but not nonimmune IgG, attenuated phenylephrine-induced EGFR transactivation in arteries. *Pⱕ0.05 vs control; #Pⱕ0.05 vs PE⫹MMP-7 antibodies. Results (mean⫾SEM) are representative of 4 independent experiments. Rat strain: Sprague-Dawley.

activity and MMP-7 were high at the onset of hypertension in SHR (5 weeks) and in early stages of the hypertension (9 weeks) but dramatically decreased by week 15 (ie, 10 weeks after the onset of the hypertension), approaching the levels of activity seen for age-matched normotensive rats. These data suggest a role for pro-HB-EGF processing activity as well as MMP-7 in the development and progression of hypertension. To determine whether activity of MMPs such as MMP-7 could impact vascular tone, we first studied the effects of a broad-spectrum inhibitor, doxycycline, on adrenergic tone of isolated, microperfused mesenteric arteries (Figure 7B). Like with blockers of the EGFR (online Figure 1) or HB-EGF

(Figure 5) and in line with our previous report,19 bolus infusions of doxycycline opposed adrenergic tone in isolated arteries (Figure 7B). This result confirmed that MMP activity is involved in maintenance of vascular adrenergic tone of arteries. We therefore investigated whether MMP inhibitor therapy would have antihypertensive action in hypertensive rats. In spontaneously hypertensive rats, the administration of doxycycline (19.2 mg/d IP from weeks 7 to 12) reduced the systolic blood pressure of the rats (Figure 7C) as well as attenuated HB-EGF shedding in arteries at week 12 (online Figure 7). In normotensive rats, however, doxycycline did not decrease blood pressure from week 7 to 10 (Figure 7C).


74


January 9/23, 2004

Figure 7. Upregulation of HB-EGF sheddases in hypertensive rats and their role in modulation of vascular tone and systolic blood pressure. A, Pro-HB-EGF zymography revealed an increased activity of HB-EGF–processing enzymes in mesenteric arteries of hypertensive rats (SHR) compared with 2 normotensive rat strains (WKY or Sprague-Dawley). BP indicates mean value of systolic blood pressure⫾SEM. Analysis at week 12. B, Quantitative analysis of vasorelaxation of phenylephrine-constricted arteries in response to the bolus infusion of broad-spectrum MMP inhibitor doxycycline (Dox). Results (mean⫾SEM) are representative of 4 independent experiments. C, Time course and quantitative analysis of the systolic blood pressure of spontaneously hypertensive rats (SHR) versus normotensive (WKY) rats treated with the broad-spectrum MMP inhibitor, doxycycline (Dox), or placebo (PBS). Results (mean⫾SEM) are representative of 5 independent experiments. *Pⱕ0.05 vs SHR, placebo; #Pⱕ0.05 vs WKY, placebo. Rat strains: Sprague-Dawley, WistarKyoto, and spontaneously hypertensive rats.

Doxycycline resulted in a decrease of blood pressure between weeks 10 and 12 (Figure 7C) without affecting HB-EGF (online Figure 7). These data suggest that activity of MMPs promotes vasoconstriction. Moreover, activity of MMPs differentially impacts blood pressure in hypertension and normotension. A mechanism of the vasodilatory effects of doxycycline in hypertension may be inhibition of MMPs that shed HB-EGF in arteries.

Discussion Our results are consistent with a novel mechanism contributing to the vascular tone effects of ␣1b-adrenoceptors, whereby stimulation of ␣1b-adrenoceptors in arteries results in activation of HB-EGF sheddases (such as MMP-7) and subsequent EGFR-dependent signaling. To demonstrate this pathway, we conducted studies of vascular tone in vivo and in isolated rat mesenteric arteries. To confirm the presence of each of the steps in the pathway, we combined immunochemical, zymo-

graphic, and vasocontractility techniques with two independent pharmacological tools that selectively inhibited mediators of the pathway (ie, MMPs, HB-EGF, and EGFR). The classical pathway of adrenergic vasoconstriction purports that agonists such as epinephrine, norepinephrine, and phenylephrine bind to adrenoceptors stimulating coupling to Gq/11 protein, which activates phospholipase C, thus initiating a contractile response.12 We confirmed that this pathway indeed operates in the arteries studied (data not shown). However, we also found that adrenergic vasoconstriction was sustained; at least in part, through a novel mechanism involving MMP-7– dependent regulation of arterial HB-EGF and subsequent activation of the EGFR by HB-EGF. Because the EGFR mediates the growth effects of diverse GPCRs, including angiotensin receptors (that also activated MMP-7), it is likely that MMP-7– dependent shedding of HB-EGF and subsequent transactivation of the EGFR modulates the vascular tone effects of these GPCRs as well.


Hao et al We observed a differential regulation of MMP-7 by different vasoactive GPCRs, because adrenoceptors and angiotensin receptors, but not calcitonin gene-related peptide (CGRP) or thrombin receptors, stimulated MMP-7 in arteries. These results suggest that MMP-7 may be differentially regulated by vasoconstrictive and vasodilatory GPCRs, because both thrombin and CGRP dilate intact arteries whereas phenylephrine and angiotensin II are vasoconstrictors. Although present data strongly implicate MMP-7 in modulation of HB-EGF in arteries, we cannot exclude that proteinases other than MMP-7 contribute to regulation of HB-EGF as well. In line with this notion, our pro-HB-EGF substrate zymography analysis revealed the presence of minor bands of yet-unidentified proteases that were capable of processing pro-HB-EGF as well as MMP-7. We considered two candidates that have molecular weights in the range of the activities we detected by zymography, MMP-2 and ADAM 12. MMP-2 (62 kDa) is a major gelatinase in arteries and sheds HB-EGF in endothelial and epithelial cells.14 ADAM 12 (92 kDa) is the major HB-EGF sheddase in cardiomyocytes.4 The antibodies against MMP-2 and ADAM 12, but not nonimmune IgG, protected HB-EGF from proteolytic processing in arterial lysates. This result suggested that MMP-2 and ADAM 12 may also contribute to regulation of HB-EGF in arteries. However, additional studies are needed to determine the importance of these proteinases in modulation of HB-EGF or other ligands of the EGFR. Taken together, our data are consistent with the concept that MMPs are important HB-EGF sheddases in arteries. Moreover, by shedding HB-EGF, MMPs connect vasoactive GPCRs (such as adrenoceptors and angiotensin receptors) with the EGFR. Because the EGFR is a receptor tyrosine kinase and signals through mitogen-activated protein kinases such as p38, Akt, and Erk2,4,5,8 –10 (also data not shown), the present observations additionally suggest a possible solution for the long-standing, controversial question of how protein tyrosine kinases interact with vasoactive GPCRs to modulate vascular tone, suggesting a role for HB-EGF sheddases in the cross talk between vasoactive pathways and growth factor signaling.20 –24 It is tempting to speculate that mitogen-activated protein kinases ultimately modulate vascular contractility by interacting with the cytoskeleton directly or through scaffold proteins.25 Increasing evidence now implicates the activity of MMPs, HB-EGF, and the EGFR in development and progression of the cardiac hypertrophy and fibrosis associated with cardiomyopathies and hypertension, and inhibition of MMPs, HBEGF, or the EGFR is a promising strategy for the treatment of these disorders.2,4 –7,18,19,24,26 –29 In spontaneously hypertensive rats, a model where the agonists of adrenoceptors and angiotensin receptors as well as the EGFR contribute to the development and progression of the hypertension,1,2,7,11,12 the activity of HB-EGF sheddases including MMP-7 was high throughout weeks 5 through 12. Moreover, MMP inhibitor therapy attenuated HB-EGF shedding in arteries as well as decreased the blood pressure of hypertensive SHR rats. These data suggest that activity of MMPs is implicated in the promotion of vascular tone in hypertension, at least in the early stages of the condition, and indicate a novel mechanism


75

of vasoregulation whereby the agonists of certain receptors (such as adrenoceptors and angiotensin receptors) may stimulate MMPs capable of HB-EGF shedding (such as MMP-7) to promote vascular tone in hypertension via transactivation of the EGFR in arteries. The inhibition of MMPs with doxycycline did not result in complete abrogation of the hypertension. Therefore, MMP-dependent EGFR transactivation may be but one mechanism contributing to the development of hypertension in the model of SHR.1,2 In line with this, the zymographic analysis of pro-HB-EGF processing activities and MMP-7 in arteries revealed that these proteolytic activities were greater during the initial stages of the development and progression of the hypertension (from weeks 5 through 12), sharply decreasing at some point between week 12 and week 15 (ie, 10 weeks after the onset of the condition). We could not exclude possible interstrain differences, because MMP levels were higher in SHR than WKY at all time points tested, except long after the onset of the hypertension (week 15). However, these observations explain, at least in part, why doxycycline antihypertensive effects were maximal at week 9. We are presently implementing proteomic tools to better understand the molecular basis of the doxycycline-resistant part of the hypertension in this model as well as other potential mechanisms of antihypertensive actions of doxycycline. Previously, we proposed that MMPs may modulate vascular tone through the regulation of peptidic agonists such as big endothelin and CGRP.18,19 The receptors for endothelins and CGRP are GPCRs and may transactivate the EGFR in arteries.2,4,8 –10 Therefore, the data indicate that MMPs modulate vascular tone via regulation of HB-EGF as well as vasopeptides, two mechanisms that may converge in growth factor receptors such as the EGFR. Taken together, our findings suggest a previously unknown mechanism of vasoregulation whereby the agonists of certain GPCRs (including adrenoceptors and angiotensin receptors) promote vasoconstriction as well as growth via activation of HB-EGF sheddases (such as MMP-7) and, thereby, transactivation of the EGFR. This mechanism may be important for the development and progression of hypertensive disorders, because it is triggered by agonists that are typically overexpressed in hypertension. Blockers of EGFR transactivation have therapeutic potential for simultaneously inhibiting pathological vasoconstriction as well as growth in hypertensive disorders.

Acknowledgments This work was supported by start-up funds from the University of Alberta and research grants from the Natural Sciences and Engineering Research Council of Canada (NSERC), Heart and Stroke Foundation of Canada, and Canadian Institutes of Health Research (to C.F.-P.). M.D. is a postdoctoral fellow from NSERC. C.F.-P. is a Heart and Stroke Foundation of Canada Scholar. We are indebted to Drs J. Filep (University of Montreal), W. Cole, and R. Loutzenhiser (University of Calgary) and G. Lopaschuk and M. Michalak (University of Alberta) for invaluable comments on this manuscript.

References 1. Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hypertension. Cell. 2001;104:545–556.


76


January 9/23, 2004

2. Touyz RM. Recent advances in intracellular signalling in hypertension. Curr Opin Nephrol Hypertens. 2003;12:165–174. 3. Oeckler RA, Kaminski PM, Wolin MS. Stretch enhances contraction of bovine coronary arteries via an NAD(P)H oxidase-mediated activation of the extracellular signal-regulated kinase mitogen-activated protein kinase cascade. Circ Res. 2003;92:23–31. 4. Asakura M, Kitakaze M, Takashima S, Liao Y, Ishikura F, Yoshinaka T, Ohmoto H, Node K, Yoshino K, Ishiguro H, Asanuma H, Sanada S, Matsumura Y, Takeda H, Beppu S, Tada M, Hori M, Higashiyama S. Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metalloproteinase inhibitors as a new therapy. Nat Med. 2002;8:35– 40. 5. Kalmes A, Daum G, Clowes AW. EGFR transactivation in the regulation of SMC function. Ann N Y Acad Sci. 2001;947:42–54. 6. Florian JA, Watts SW. Epidermal growth factor: a potent vasoconstrictor in experimental hypertension. Am J Physiol. 1999;276:H976 –H983. 7. Fujino T, Hasebe N, Fujita M, Takeuchi K, Kawabe J, Tobise K, Higashiyama S, Taniguchi N, Kikuchi K. Enhanced expression of heparin-binding EGF-like growth factor and its receptor in hypertrophied left ventricle of spontaneously hypertensive rats. Cardiovasc Res. 1998; 38:365–374. 8. Carpenter G. EGF receptor transactivation mediated by the proteolytic production of EGF-like agonists. Sci STKE. 2000;15:1. 9. Gschwind A, Zwick E, Prenzel N, Leserer M, Ullrich A. Cell communication networks: epidermal growth factor receptor transactivation as the paradigm for interreceptor signal transmission. Oncogene. 2001;20: 1594 –1600. 10. Prenzel N, Zwick E, Daub H, Leserer M, Abraham R, Wallasch C, Ullrich A. EGF receptor transactivation by G-protein-coupled receptors requires metalloproteinase cleavage of proHB-EGF. Nature. 1999;402:884 – 888. 11. Reja V, Goodchild AK, Pilowsky PM. Catecholamine-related gene expression correlates with blood pressures in SHR. Hypertension. 2002; 40:342–347. 12. Guimaraes S, Moura D. Vascular adrenoceptors: an update. Pharmacol Rev. 2001;53:319 –356. 13. Yu WH, Woessner JF Jr, McNeish JD, Stamenkovic I. CD44 anchors the assembly of matrilysin/MMP-7 with heparin-binding epidermal growth factor precursor and ErbB4 and regulates female reproductive organ remodeling. Genes Dev. 2002;16:307–323. 14. Razandi M, Pedram A, Park ST, Levin ER. Proximal events in signaling by plasma membrane estrogen receptors. J Biol Chem. 2003;278: 2701–2712. 15. Yan Y, Shirakabe K, Werb Z. The metalloprotease Kuzbanian (ADAM10) mediates the transactivation of EGF receptor by G proteincoupled receptors. J Cell Biol. 2002;158:221–226. 16. Lee DC, Sunnarborg SW, Hinkle CL, Myers TJ, Stevenson MY, Russell WE, Castner BJ, Gerhart MJ, Paxton RJ, Black RA, Chang A, Jackson

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

LF. TACE/ADAM17 processing of EGFR ligands indicates a role as a physiological convertase. Ann N Y Acad Sci. 2003;995:22–38. Sunnarborg SW, Hinkle CL, Stevenson M, Russell WE, Raska CS, Peschon JJ, Castner BJ, Gerhart MJ, Paxton RJ, Black RA, Lee DC. Tumor necrosis factor-␣ converting enzyme (TACE) regulates epidermal growth factor receptor ligand availability. J Biol Chem. 2002;277: 12838 –12845. Fernandez-Patron C, Radomski MW. Davidge ST. Vascular matrix metalloproteinase-2 cleaves big endothelin-1 yielding a novel vasoconstrictor. Circ Res. 1999;85:906 –911. Fernandez-Patron C, Stewart KG, Zhang Y, Koivunen E, Radomski MW, Davidge ST. Vascular matrix metalloproteinase-2-dependent cleavage of calcitonin gene-related peptide promotes vasoconstriction. Circ Res. 2000;87:670 – 676. Hollenberg MD. Tyrosine kinase-mediated signal transduction pathways and the actions of polypeptide growth factors and G-protein-coupled agonists in smooth muscle. Mol Cell Biochem. 1995;149 –150:77– 85. Fernandez-Patron C, Zhang Y, Radomski MW, Hollenberg MD, Davidge ST. Rapid release of matrix metalloproteinase-2 by thrombin in the rat aorta: modulation by protein tyrosine kinase/phosphatase. Thromb Haemost. 1999;82:1353–1357. Walsh MP, Horowitz A, Clement-Chomienne O, Andrea JE, Allen BG, Morgan KG. Protein kinase C mediation of Ca2⫹-independent contractions of vascular smooth muscle. Biochem Cell Biol. 1996;74: 485–502. Watts SW, Florian JA, Monroe KM. Dissociation of angiotensin II-stimulated activation of mitogen-activated protein kinase kinase from vascular contraction. J Pharmacol Exp Ther. 1998;286:1431–1438. Kagiyama S, Qian K, Kagiyama T, Phillips MI. Antisense to epidermal growth factor receptor prevents the development of left ventricular hypertrophy. Hypertension. 2003;41:824 – 829. Soderling SH, Binns KL, Wayman GA, Davee SM, Ong SH, Pawson T, Scott JD. The WRP component of the WAVE-1 complex attenuates Rac-mediated signalling. Nat Cell Biol. 2002;4:970 –975. Axisa B, Loftus IM, Naylor AR, Goodall S, Jones L, Bell PR, Thompson MM. Prospective, randomized, double-blind trial investigating the effect of doxycycline on matrix metalloproteinase expression within atherosclerotic carotid plaques. Stroke. 2002;33:2858 –2864. Peterson JT, Hallak H, Johnson L, Li H, O’Brien PM, Sliskovic DR, Bocan TM, Coker ML, Etoh T, Spinale FG. Matrix metalloproteinase inhibition attenuates left ventricular remodeling and dysfunction in a rat model of progressive heart failure. Circulation. 2001;103:2303–2309. Cowan KN, Jones PL, Rabinovitch M. Elastase and matrix metalloproteinase inhibitors induce regression, and tenascin-C antisense prevents progression, of vascular disease. J Clin Invest. 2000;105:21–34. Flamant M, Tharaux PL, Placier S, Henrion D, Coffman T, Chatziantoniou C, Dussaule JC. Epidermal growth factor receptor transactivation mediates the tonic and fibrogenic effects of endothelin in the aortic wall of transgenic mice. FASEB J. 2003;17:327–329.





CIRCRESAHA/2003/072272 Agonist-induced activation of MMP-7 promotes vasoconstriction through the EGFreceptor pathway Li Hao, Min Du, Ana Lopez-Campistrous, and Carlos Fernandez-Patron

Legend to Online supplement Fig.1: Pharmacological inhibition of the EGFR promotes vasorelaxation of agonist-, but not KCl-, preconstricted arteries. Maximum of arterial diameter in response to luminal bolus infusions of increasing amounts of the EGFR kinase inhibitor, AG 1478 (diagrams a and c), or the inactive tyrphostin, AG 9 (diagrams b and d), injected in the perfusion line towards arteries maintained in the presence of phenylephrine or KCl (added to the bath, i.e., adventia side) to produce a tone. AG 1478 relaxed phenylephrine, but not KCl, constricted arteries. Results (mean ± sem) are representative of 4 independent experiments for every data point. (*): P