Vascular Smooth Muscle Overexpression of G Protein–Coupled Receptor Kinase 5 Elevates Blood Pressure, Which Segregates With Sex and Is Dependent on Gi-Mediated Signaling
Background— Essential hypertension involves an increase in sympathetic nervous system activity and an associated decrease in β-adrenergic receptor (AR)–mediated dilation. In addition, increased levels of G protein–coupled receptor (GPCR) kinases (GRKs), which regulate GPCR signaling, are associated with increased blood pressure (BP).
Methods and Results— We generated transgenic mice with ≈2-fold vascular smooth muscle (VSM)–specific overexpression of GRK5 to recapitulate a selective aspect of hypertension and understand the impact on GPCR regulation of BP. VSM-GRK5 mice were hypertensive, with a 25% to 35% increase in BP, whereas there was no concomitant cardiac or VSM hypertrophy. BP elevations were segregated with sex, with male mice having higher levels than female mice, and ovariectomy did not alter this phenotype. BP was restored to control values with pertussis toxin Gi-signaling inhibition or chronic β1AR inhibition after 7 days of CGP20712A, whereas the β2AR antagonist ICI 118,551 was ineffective. α1AR response was not altered, nor was βAR-mediated dilation in male blood vessels, whereas norepinephrine sensitivity was increased. In contrast, female VSM-GRK5 blood vessels have diminished βAR-mediated dilation and enhanced sensitivity to angiotensin II (Ang II).
Conclusions— Our data suggest that in both male and female mice, VSM-specific overexpression of GRK5 elevates BP mediated by Gi and, at least in part, by β1AR in males and Ang II receptors in females. Understanding mechanisms underlying an increase in VSM-GRK5 may have a profound influence on the use and development of antihypertensive therapeutics.
Received December 22, 2004; revision received April 6, 2005; accepted May 10, 2005.
The contraction state of blood vessels is a main determinant of blood pressure (BP) and is under tight regulatory control. A major component in the maintenance of contraction state is G protein–coupled receptors (GPCRs) present on the vascular smooth muscle (VSM). For example, impairment in β-adrenergic receptor (AR)–Gs heterotrimeric signaling has been documented in patients with hypertension, and studies from animal models provide further support for the critical role of βARs in hypertension.1–5 Other GPCRs, such as angiotensin II (Ang II) receptors, are also essential to BP regulation.6 Ang II receptors couple to both Gq and Gi, and activation of both these heterotrimeric G proteins mediates vasoconstriction. Gene targeting deletion of the Ang II AT1A receptor in mice results in a decrease in resting BP, and Ang II receptor antagonists are effective antihypertensive therapies.6 Therefore, regulation of GPCR signaling is important to BP control.
The predominant regulation of GPCRs occurs with the targeted phosphorylation of activated receptors, leading to G protein uncoupling, a process called desensitization. This process is initiated by phosphorylation of the agonist-occupied receptor by GPCR kinases (GRKs), a 7-member family (GRKs 1–7) of serine/threonine kinases.7,8 Studies in both humans and animal models of hypertension suggest that elevated GRK levels during high BP may somehow be involved in the regulation of VSM GPCRs. Two GRKs, GRK2 and GRK5, have been specifically addressed in studies of hypertension, and our previous data in VSM GRK2 transgenic mice9 suggest that VSM-targeted GRK2 overexpression is sufficient to increase BP. In addition, increased GRK2 protein expression has been correlated with increased BP in lymphocytes and VSM in rat models of hypertension as well as in lymphocytes of hypertensive patients.10–12 GRK5 levels are increased in VSM during both Ang II and norepinephrine (NE)-induced hypertension.13 This augmented GRK5 level was attenuated by concurrent hydralazine administration, which relaxes VSM in a non-GPCR manner by inhibiting accumulation of intracellular free Ca2+, suggesting that hemodynamic stress itself was responsible for the VSM-specific GRK5 increase.13
We were interested in determining the involvement of GRK5-mediated phosphorylation of VSM GPCR signaling and its contribution to hypertension. Our present data indicate that elevated levels of VSM GRK5 significantly increase resting BP, which is segregated by sex and not associated with a concomitant increase in cardiac or VSM hypertrophy. The VSM-GRK5–induced hypertension appears to be mediated primarily by β1AR-Gi in males and enhanced Ang II receptor sensitivity in females.
Generation of Transgenic Mice
The full coding region of the bovine GRK5 was fused to a portion of the SM22α promoter (−440 to +41, relative to transcription start) that has been shown to direct VSM-specific transgene expression.9,14 The construct also included an SV40 polyadenylation sequence for genome integration and mRNA expression determination. Three lines of GRK5 transgenic mice were generated by the Duke Transgenic Core Facility (Duke University, Durham NC) in a C57Bl/6 background. All mice were studied between the ages of 2 to 6 months of age and compared with their non-transgenic littermate controls (NLC).
Semiquantitative Real-Time Reverse Transcription PCR
Total RNA was extracted from the VSM layer of the aorta, from which endothelial cells were mechanically scraped, and adventitia was enzymatically removed. RNA was extracted from the VSM layer or isolated cardiac myocytes, converted to cDNA, and used as the template in a polymerase chain reaction (PCR) reaction that included the fluorophor SYBR Green. Increased fluorescence was detected by use of a real-time PCR machine (BioRad iCycler). PCR primers for GRK5 were designed to span introns that characterize its genomic DNA, thereby preventing genomic DNA contamination of the RNA and subsequent reverse transcription PCR (forward, 5′-GAAGGTTAAGCGGGAAGAGG-3′; reverse, 5′-TCCAGGCGCTTAAAGTTCAT-3′). A standard curve detected known amounts of GRK5 DNA ranging from 10 pg to 1 μg. GRK2 mRNA was determined (forward, 5′-CCCTCTCACCATCTCTGAGC-3′; reverse, 5′-CGGTTGGGGAACAAGTAGAA-3′). GRK3 mRNA was also detected (forward, 5′-GATATCGGCTCCTTCGATGA-3′; reverse, 5′-CAGTCCTTCCCCATAGCGTA-3′). Levels of β1AR (forward, 5′-TCGCTACCAGAGTTTGCTGA-3′; reverse, 5′-GGCACGTAGAAGGAGACGAC-3′) and β2AR (forward, 5′-ACCAAGAATAAGGCCCGAGT-3′; reverse, 5′-AGGGGCACGTAGAAAGACAC-3′) mRNA was determined by use of mouse specific primers and normalized to 18S (forward, 5′-TCAAGAACGAAAGTCGGAGG-3′; reverse, 5′-GGACATCTAAGGGCATCAC-3′).
Primary VSM Cell Isolation
VSM cells (VSMCs) were isolated from pooled samples of NLC or GRK5 thoracic aorta, plated, and cultured (less than passage 8) as described previously.9,14
Immunohistochemistry and Immunoblotting
Carotid arteries were isolated from NLC or GRK5 mice and embedded in paraffin, and 8-μm sections were taken. Immunoblotting and immunohistochemistry for GRK5 and GRK2 were performed as described previously using primary antibodies from Santa Cruz Technologies.9
Conscious BP was measured 24 hours after carotid artery catheter insertion as described previously.9 Additional BP measurements were determined on anesthetized mice, and it was found that if anesthetized heart rates (HRs) were >200 bpm, conscious and anesthetized BP were similar (data not shown).
Plasma Catecholamine Levels
Plasma catecholamine levels were determined by AniLytics (Gaithersburg, Md) as described previously.14
Female mice age 12 to 16 weeks were anesthetized with ketamine (100 mg/kg) and xylazine (5 mg/kg); a small lower abdominal incision was made, and the ovaries were exteriorized under a microscope and carefully maintained. Vessels attached to the ovaries were isolated and permanently tied twice with several millimeters between the ties with 6-0 silk suture. The vessels were cut between the 2 ties, the ovaries removed, and the abdomen closed. Sham-operated animals received the same treatment except that their ovaries were not removed. Three weeks later, animals were anesthetized as above, BP was determined, and thoracic and abdominal aorta sections were isolated for vascular reactivity studies.
The response of 2.5-mm aortic rings after mechanical endothelial cell denudation was determined as described previously.9,14
Chronic Inhibition of Gi-Mediated Signaling and β1AR or β2ARs
Pertussis toxin (PTx) was given as 1 intraperitoneal injection at a concentration of 15 μg/kg to inhibit Gi-mediated signaling. β1ARs were inhibited chronically for 7 days by use of miniosmotic pumps (Alzet) implanted subcutaneously and subscapularly using the antagonist CGP20712A (Sigma) at a concentration of 1.2 mg · kg−1 · d−1. β2ARs were inhibited by use of miniosmotic pumps filled with 0.2 mg · kg−1 · d−1 of the selective β2AR antagonist ICI 118,551 (Sigma).
Hypertensive Mice Generated by Miniosmotic Pumping With Ang II
Miniosmotic Alzet pumps releasing 0.7 mg · kg−1 · d−1 Ang II were implanted subcutaneously, subscapularly, for 7 days as described previously.14
Adenylate Cyclase Activity
cAMP accumulation was measured by use of a Bridge-It cAMP assay kit according to the manufacturer’s instructions (Mediomics, LLC).
All data are presented as mean±SEM. For simple comparisons between 2 groups, an unpaired, 2-tailed Student t test was used. One-way ANOVA with post hoc Bonferroni multiple comparisons was used to compare multiple groups. Two-way ANOVA was used when a dose-response was analyzed.
Transgene Expression in VSM-GRK5 Mice
To understand how elevations in VSM GRK5, as seen in hypertension,13 impact GPCR signaling, we generated 3 lines of transgenic mice using a portion of the SM22α promoter that has previously been shown to direct expression specifically in VSM9,15–19 to overexpress GRK5. Thoracic aortas were isolated from NLC and GRK5-overexpressing mice. By use of real-time PCR analysis, we determined a 2-fold overexpression of GRK5 mRNA compared with endogenous levels (Figure 1A). Results were verified in all 3 lines of VSM-GRK5 mice, and overexpression levels were similar between the 3 different lines (data not shown) and comparable to levels we have seen for other genes we have previously expressed using this portion of the SM22α promoter.9,14 Furthermore, as confirmation of previously published studies,13 we used miniosmotic pumps to administer 0.7 mg · kg−1 · d−1 Ang II to NLC mice for 7 days and determined GRK5 levels in the VSM of the thoracic aorta. Similar to the overexpression levels in our transgenic mice, there was a 2-fold increase in GRK5 mRNA after chronic Ang II treatment (Figure 1A). To ensure promoter specificity, we also isolated adult cardiac myocytes and determined GRK5 expression. As expected, there was no change in cardiac myocyte GRK5 levels between the NLC and VSM-GRK5 mice (Figure 1A). Importantly, there were no differences in either GRK2 or GRK3 mRNA levels as determined by use of real-time PCR between NLC and VSM-GRK5 mice (data not shown). In addition, GRK2 protein levels were similar between NLC and VSM-GRK5 mice.
We verified in vivo protein expression using immunohistochemistry of thoracic aorta from NLC and GRK5 mice. A low endogenous level of GRK5 expression was detected. The VSM-GRK5 mice had a more prominent signal, illustrating effective transgenic expression of the GRK5 protein (Figure 1B).
VSMCs were enzymatically isolated and cultured from thoracic aorta. Immunoblots were performed to confirm transgene overexpression. Similar to mRNA data, GRK5 overexpression was verified in the VSMCs isolated from the transgenic compared with the control mice (Figure 1C).
VSM-GRK5 Overexpression Induces High BP That Is Greater in Male Mice
Conscious BP was determined in freely moving mice by use of an indwelling fluid-filled carotid artery catheter. Mean arterial pressure was increased in all 3 lines of VSM-GRK5 transgenic mice (Figure 2A). Interestingly, although there was no difference in BP between control male (93±1, n=4) and female (95±3, n=6) mice, BP was increased in female VSM-GRK5 mice compared with control (112±3, n=3) and increased to an even greater extent in male VSM-GRK5 mice (138±2, n=4) compared with control and female VSM-GRK5 mice (Figure 2B). Importantly, the increase in VSM-GRK5–induced BP was not accompanied by a change in plasma catecholamines (Figure 2C). In addition, there were no differences in HR between NLCs and VSM-GRK5 mice for either males or females (HR in male NLC, 547±31 bpm; female NLC, 575±35 bpm; VSM GRK5 male mice, 545±15 bpm; female VSM GRK5, 568±10 bpm). Furthermore, there was no change in heart size, as measured by wet heart-weight–to–body-weight ratio, concomitant with this increase in afterload (NLC: male, 4.48±0.09, n=5; female, 4.57±0.09, n=10; GRK5: male, 4.59±0.06, n=8; female, 4.63±0.07, n=9). Furthermore, we examined the thickness and area of the VSM layer in NLC and VSM-GRK5 common carotid arteries after perfusion fixing, and there was no change in hypertrophic status of the VSM layer in the blood vessel wall (NLC: 27.0±2.7 μm, n=9; VSM-GRK5: 26.0±1.9 μm, n=10).
BP Increase Is Mediated by Gi-Coupled GPCRs
To test the involvement of Gi-mediated signaling, such as may be mediated through Ang II receptors as well as numerous other receptors, including the βARs,20,21 we used a single dose of PTx, known to inhibit Gi-mediated signaling22 and shown previously to inhibit high BP in other animal models of hypertension.23 After PTx treatment and subsequent Gi inhibition, BP was restored to control values in VSM-GRK5 mice, whereas it was unaffected in NLC mice (Figure 3A). The ability of PTx to inhibit BP was similar in both male and female mice (data not shown).
Inhibition of β1AR but Not β2AR Is Sufficient to Decrease BP in VSM-GRK5 Mice
βAR inhibition is a successful antihypertensive therapeutic strategy in many hypertensive patients.24 To understand how βAR inhibition would affect hypertension in VSM-GRK5 mice, we used miniosmotic pumps to chronically administer the β1AR antagonist CGP20712A or the β2AR antagonist ICI 118,551 for 7 days. BP was decreased after treatment with CGP20712A but not ICI 118,551 (Figure 3B). Importantly, before βAR antagonist treatment, there were no differences in HR (see above), and not unexpectedly, because there were no basal increases in HR with VSM-GRK5 overexpression, βAR blockade did not alter this.
βAR-Mediated Vasodilation in Aorta of Male GRK5 Transgenic Mice
Impairment in βAR signaling has been documented in patients with hypertension,25 and there is a downregulation of the aorta βAR-adenylate cyclase system in spontaneously hypertensive and Dahl salt-sensitive hypertensive rats.4 In addition, we have also generated another line of transgenic mice with VSM-GRK2 overexpression, and these mice are hypertensive because of abnormal (uncoupled) βAR signaling leading to loss of dilatory control.9 Therefore, we examined the relaxation in response to βAR stimulation in male thoracic and abdominal aorta, which is a typical physiological response caused by βAR-Gs activation (Figure 4). Preconstriction was established by phenylephrine (PE), an α1AR agonist. Responses to PE at the doses used were not different between male and female mice or between NLC and GRK5 transgenic mice (data not shown). Dilation in response to increasing doses of isoproterenol (ISO), a βAR agonist, were similar in male NLC and GRK5 thoracic and abdominal aorta (Figure 4, A and B). In addition, after pretreatment of the vessels with 2×10−7 mol/L ICI 118,551, a specific β2AR antagonist, there was a similar degree of inhibition of relaxation in the male thoracic aorta (Figure 4C).
KCl-Induced K+ Channel Inhibition Uncovers Greater ISO-Induced Constriction in Male VSM-GRK5 Mice
We did not detect a difference in ISO-mediated dilation in male thoracic or abdominal aorta (Figure 4). In contrast, when we examined the ISO response in thoracic aorta in the presence of preconstriction mediated by 60 mmol/L KCl, ISO induced a dose-dependent constriction (Figure 4D). In contrast to the predicted βAR-mediated dilation, ISO-induced constriction and tension was significantly greater in VSM-GRK5 vessels than NLC vessels (Figure 4D), such that there was a significant decrease in the Kd for ISO-induced constriction after K+ channel inhibition in VSM-GRK5 mice (NLC: 4.8±0.8×10−6 mol/L, n=5; VSM-GRK5: 1.3±0.2×10−6 mol/L, n=5, P=0.01). Thus, in the presence of K+ channel inhibition mediated by KCl, ISO induces a greater constriction in male VSM-GRK5 vessels, perhaps through enhanced coupling to or activation of L-type Ca2+ channels. Importantly, as described above, response to PE was not different between NLC and VSM-GRK5 mice, and therefore, it is unlikely that nonspecific effects of ISO binding to α1ARs at the higher concentrations is conferring this enhanced sensitivity in the VSM-GRK5 blood vessels.
NE-Mediated Vasoconstriction Is Enhanced in Male VSM-GRK5 Mice
Rats made hypertensive by use of a Goldblatt 2-kidney/1-clip model have increased VSM sensitivity to NE stimulation26 and decreased βAR responsiveness.27 NE stimulates αARs as well as β1ARs, and we know from the isolated ring data that the PE response is similar in male and female NLC and GRK5 thoracic and abdominal aorta. Furthermore, βAR-mediated dilation in males is not altered with GRK5 overexpression (Figure 4, A–C). Constriction in response to increasing doses of NE was similar in NLC and female VSM-GRK5 thoracic aorta (Figure 4E). In contrast, there was a significant increase in the tension generated by male VSM-GRK5 compared with both NLC and female VSM-GRK5 thoracic aorta (Figure 4E). These data further support in vivo data (Figure 3B) that β1AR inhibition decreased GRK5-induced hypertension and suggest that β1AR stimulation is mediating constriction rather than dilation.
cAMP Accumulation Is Diminished in Male VSM GRK5 VSMCs After βAR Stimulation
To examine the classic coupling of the βAR to Gs, we determined cAMP accumulation. We measured cAMP accumulation in cultured, low-passage male NLC and GRK5 thoracic aorta VSMCs (Figure 5). cAMP accumulation in response to 10−7 mol/L ISO was diminished in male VSM-GRK5 cells compared with NLC, suggesting that cAMP is a more sensitive assay than isolated ring baths to detect subtle signaling alterations. After treatment of the cells with 10−7 mol/L ICI 118,551, all cAMP accumulation in response to ISO was abrogated in both NLC and VSM-GRK5 cells (Figure 5C); therefore, the majority of βAR-Gs activation is mediated via the β2AR.
To determine whether βAR expression is altered in the male mice, we used real-time PCR to measure mRNA expression. For both NLC and VSM-GRK5 mice, β2AR was in greater abundance than β1AR mRNA (Table). Although VSM-GRK5 expression did not alter β1AR expression, there was an almost 50% reduction in β2AR mRNA, changing the ratio of β2AR to β1AR mRNA from approximately 340-fold to 6-fold.
βAR-Mediated Vasodilation in Aorta of Female GRK5 Transgenic Mice
We also examined the βAR-mediated dilation in thoracic and abdominal aorta isolated from female control and transgenic mice. Dilation was attenuated in the VSM-GRK5 thoracic aorta compared with NLC (Figure 6A). Interestingly, and in contrast to the male abdominal aorta, the female NLC abdominal aorta did not respond to ISO (Figure 6B). In addition, the GRK5 abdominal aorta tended to constrict in response to ISO (Figure 6B). After treatment with 2×10−7 mol/L ICI 118,551, similar to what we observed in the male thoracic aorta, there was an almost 10-fold shift in the Kd for maximum relaxation in response to ISO (NLC: 3.7±0.5×10−7 mol/L, n=5; NLC+ICI: 3.9±2.1×10−6 mol/L, n=4). After pretreatment with ICI 118,551, relaxation is mediated primarily by the β1AR. The contribution to relaxation mediated by the β1ARs is more significant in the VSM-GRK5 thoracic aorta, and the NLC thoracic aorta maintains more of an ability to relax in the presence of β2AR inhibition compared with the VSM-GRK5 thoracic aorta (Figure 6C).
Ang II–Stimulated Constriction Is Increased in Female but Not Male VSM-GRK5 Mice
We examined constriction in response to Ang II in abdominal aorta, because we were unable to detect any appreciable response in thoracic aorta, as has been described previously28,29 (Figure 7). Constriction mediated by Ang II receptors in male NLC and VSM-GRK5 mice was similar (Figure 7A). Ang II is coupled to both Gq and Gi. Our data show that in general, Gi signaling is important in conferring hypertension in the VSM GRK5 mice (Figure 3). Gi but not Gq directly stimulates c-Src,30 and it has been established in other animal models that c-Src is an important signaling molecule downstream of Ang II in VSM31 by directly inhibiting iberiotoxin-sensitive Ca2+-sensitive K+ channels32 and increasing intracellular Ca2+.33 To determine whether signaling molecules downstream of Ang II–Gi activation were altered in either male or female mice with VSM-GRK5 overexpression, we tested the involvement of c-Src using PP2, a pharmacological inhibitor. Pretreatment with 10−6 mol/L PP2 significantly blunted constriction in male abdominal aorta, and overexpression of GRK5 had no effect (Figure 7A, 2-way ANOVA, P<0.001). In contrast to the Ang II response in VSM GRK5 male mice, constriction mediated by Ang II in female abdominal aorta was significantly increased in VSM-GRK5 mice (Figure 7B), although c-Src inhibition also attenuated Ang II–induced constriction in female abdominal aorta to a similar extent in NLC and VSM-GRK5 vessels (Figure 7B). Therefore, vasoconstriction in response to Ang II is dependent primarily on Gi–c-Src and is enhanced in female VSM-GRK5 mice.
Ovariectomy Was Not Sufficient to Change BP or Vascular Reactivity
We removed the ovaries from female mice and allowed them to recover for 3 weeks. Subsequently, we determined BP and found that although sham-treated VSM-GRK5 mice had elevated BP (121±4 mm Hg, n=9) compared with sham NLC (100±6 mm Hg, n=9), ovariectomy (OVx)-treated NLC mice had normal BP (100±7 mm Hg, n=6), and OVx-treated VSM-GRK5 mice had a BP (115±7 mm Hg, n=5) similar to that of sham-treated VSM-GRK5 mice. Therefore, OVx was not sufficient to result in a female mouse now having a similar phenotype to male mice. In addition, vascular responses to NE, ISO, and Ang II in isolated aortic segments were not different in OVx mice compared with sham-treated mice (data not shown).
In the present study, we confirm that VSM-GRK5 overexpression induces high BP that is greater in male than female mice without concomitant cardiac or VSM hypertrophy (Figure 8). In both male and female VSM-GRK5 mice, Gi signaling is critical to the development of hypertension. In males, the increased BP is also mediated by a β1AR mechanism. βAR antagonists are widely used as an antihypertensive therapeutic strategy. Importantly, these data may provide evidence to suggest that the success of β-blockers in the treatment of hypertension may not be solely a result of actions on the heart. Our data suggest that direct VSM β1AR signaling inhibition, in high-BP models with elevated GRK5 expression and activity, is also partially responsible for attenuation of high BP. Importantly, there were no differences in HR between our transgenic VSM-GRK5 mice and control mice, suggesting that the decrease in BP seen in the transgenic mice was not because of differential effects on HR. In female mice, VSM-GRK5 overexpression enhances Ang II signaling. Ang II receptor blockers and ACE inhibitors are also key antihypertensive therapies, although their effectiveness is limited in some patients, especially when administered as a monotherapy. These data illustrate that changes in expression of GRKs, as are seen in hypertension, can have profound consequences on the signaling pathways used by VSM GPCRs. In addition, it suggests that understanding the levels of such molecules as GRK5 in hypertensive patients may provide insight into more effective therapeutic antihypertensive strategies. Furthermore, it reiterates the importance of understanding data in the context of sex, especially in transgenic mice, in which there is often a tendency to study both sexes grouped together.
It is interesting and unexpected that there is a sex difference between the GPCR signaling in response to GRK5 overexpression. Whether this is because of the presence of estrogen or lack of testosterone is unknown, but the effects are clearly seen at the individual GPCR level (βARs, Figures 4 and 5⇑; Ang II, Figure 7). Nebivolol is a highly selective β1AR antagonist with vasodilating characteristics attributed partially to its interaction with the estrogen receptor and the subsequent generation of nitric oxide.34 Therefore, it is possible that there is direct interaction between the estrogen receptor and β1AR. Furthermore, it has been described that estradiol treatment in OVx rats was sufficient to increase AT1 and AT2 receptors in the kidney.35 In our mice, OVx was not sufficient to result in a “male” phenotype in “female” mice. This might be because we removed the ovaries of fully mature female mice. Other studies have documented decreases in L-type Ca2+ channels and β1AR in the heart with castration36 and an increase in β1AR with OVx.36 It has also been documented that there is a greater abundance of L-type Ca2+ channels, ryanodine receptors, and sodium–Ca2+ exchange protein in female versus male rats.37,38 Further study is needed examining the role of estrogen and testosterone and other signaling proteins associated with vasoconstriction during development to fully understand the mechanisms underlying the sex differences in our VSM-GRK5 mice. Data derived from these studies could lend further insight into the sex differences in human cardiovascular disease.
It is well appreciated that although structurally similar, β1- and β2ARs play very distinct roles in cardiac myocytes.39 The PDZ motif on β1AR is needed to restrict coupling of this receptor to Gs, whereas mutation of this motif or inhibition of the interaction between the β1AR and the PDZ-domain protein permits the receptor to couple to both Gs and Gi in vitro.40 It has been suggested that elevated levels of GRK5 disrupt binding of PSD-95, a PDZ-binding domain protein, to the C-terminal portion of β1AR.41 Therefore, it is reasonable to propose that elevated VSM-GRK5 downregulates β2AR, decreasing dilation mediated via this receptor, and at the same time, permits β1AR to couple to Gi via disruption of the PSD-95 binding, thereby mediating vasoconstriction. Importantly, we demonstrate that inhibition of β1AR signaling was sufficient to decrease elevated BP in the VSM-GRK5–overexpressing mice, suggesting that β1AR plays a role in the hypertension. Whether GRK5 also facilitates the ability of the Ang II receptor to bind to Gi versus Gq or other G proteins remains to be determined.
Our previous study with GRK2 overexpression in VSM9 also resulted in a hypertensive phenotype. However, these models have very distinct properties, probably as a result of in vivo receptor selectivity of both GRK2 and GRK5. Sex differences were not apparent in the GRK2 model. Hypertension in the GRK2 model was associated with concomitant cardiac and VSM hypertrophy, whereas GRK5 overexpression was not. Hypertension in the GRK2 model appears to be more a result of decreased Gs-mediated dilation and increased Gq signaling, whereas GRK5 overexpression is reliant primarily on Gi. Interestingly, in human and animal models of hypertension, different models have one or the other GRK elevated.10–13 In addition, GRK5 but not GRK2 contains a DNA-binding nuclear localization sequence and may play a signaling role in the nucleus,42 although the importance of this in the development of VSM-GRK5 hypertension remains to be investigated. This suggests that both of these models provide us with powerful tools to help dissect important GPCR signaling pathways in hypertension with different pathogeneses. In addition, it suggests that GRKs may be useful as an indicator of action for the most appropriate efficacious therapeutic antihypertensive strategy.
VSM overexpression of GRK5 is sufficient to increase resting BP via a Gi-dependent mechanism without concomitant hypertrophy. Elevated levels of GRK5 have been associated with different animal models of hypertension,13 and the mice described here provide us with a powerful tool to help dissect the important signaling pathways that are perturbed during increases in GRK5 expression in both males and females. In addition, studies involving the VSM GRK5 mice may help us to develop novel therapeutic strategies and/or biomarkers to combat hypertension.
We thank Dr Walter Koch for guidance, thoughtful discussion, and valuable input. These studies were supported in part by American Heart Association Scientist Development Grant 0230060N and National Institutes of Health, National Heart, Lung, and Blood Institute grant RO1-HL-69847 (both to Dr Eckhart).
Bray MS, Krushkal J, Li L, Ferrell R, Kardia S, Sing CF, Turner ST, Boerwinkle E. Positional genomic analysis identifies the β2-adrenergic receptor gene as a susceptibility locus for human hypertension. Circulation. 2000; 101: 2877–2882.
Goto K, Fujii K, Abe I. Impaired β-adrenergic hyperpolarization in arteries from prehypertensive spontaneously hypertensive rats. Hypertension. 2001; 37: 609–613.
Sitzler G, Zolk O, Laufs U, Paul M, Bohm M. Vascular β-adrenergic receptor adenylyl cyclase system from renin-transgenic hypertensive rats. Hypertension. 1998; 31: 1157–1165.
de Gasparo M, Catt KJ, Inagami T, Wright JW, Unger T. International union of pharmacology, XXIII: the angiotensin II receptors. Pharmacol Rev. 2000; 52: 415–472.
Inglese J, Freedman NJ, Koch WJ, Lefkowitz RJ. Structure and mechanism of the G protein-coupled receptor kinases. J Biol Chem. 1993; 268: 23735–23738.
Eckhart AD, Ozaki T, Tevaearai H, Rockman HA, Koch WJ. Vascular-targeted overexpression of G protein-coupled receptor kinase-2 in transgenic mice attenuates β-adrenergic receptor signaling and increases resting blood pressure. Mol Pharmacol. 2002; 61: 749–758.
Gros R, Chorazyczewski J, Meek MD, Benovic JL, Ferguson SS, Feldman RD. G-protein–coupled receptor kinase activity in hypertension: increased vascular and lymphocyte G-protein receptor kinase-2 protein expression. Hypertension. 2000; 35: 38–42.
Ishizaka N, Alexander RW, Laursen JB, Kai H, Fukui T, Oppermann M, Lefkowitz RJ, Lyons PR, Griendling KK. G protein-coupled receptor kinase 5 in cultured vascular smooth muscle cells and rat aorta. J Biol Chem. 1997; 272: 32482–32488.
Keys JR, Greene EA, Koch WJ, Eckhart AD. Cardiac hypertrophy induced by Gq-coupled receptor agonists is mediated via the vasculature. Hypertension. 2002; 40: 660–666.
Li L, Miano JM, Mercer B, Olson EN. Expression of the SM22α promoter in transgenic mice provides evidence for distinct transcriptional regulatory programs in vascular and visceral smooth muscle cells. J Cell Biol. 1996; 132: 849–859.
Solway J, Seltzer J, Samaha FF, Kim S, Alger LE, Niu Q, Morrisey EE, Ip HS, Parmacek MS. Structure and expression of a smooth muscle cell-specific gene, SM22α. J Biol Chem. 1995; 270: 13460–13469.
Schneider DJ, Hayes M, Wadsworth M, Taatjes H, Rincón M, Taatjes DJ, Sobel BE. Attenuation of neointimal vascular smooth muscle cellularity in atheroma by plasminogen activator inhibitor type 1 (PAI-1). J Histochem Cytochem. 2004; 52: 1091–1099.
Schuh K, Quaschning T, Knauer S, Hu K, Kocak S, Roethlein N, Neyses L. Regulation of vascular tone in animals overexpressing the sarcolemmal calcium pump. J Biol Chem. 2003; 278: 41246–41252.
Moessler H, Mericskay M, Li Z, Nagl S, Paulin D, Small J. The SM22 promoter directs tissue-specific expression in arterial but not in venous or visceral smooth muscle cells in transgenic mice. Development. 1996; 122: 2415–2425.
Fields TA, Casey PJ. Signalling functions and biochemical properties of pertussis toxin-resistant G-proteins. Biochem J. 1997; 321: 561–571.
Li Y, Anand-Srivastava MB. Inactivation of enhanced expression of Gi proteins by pertussis toxin attenuates the development of high blood pressure in spontaneously hypertensive rats. Circ Res. 2002; 91: 247–254.
Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC). Bethesda, Md: US Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute; 2004. Publication No. 04-5230.
Zhou Y, Chen Y, Dirksen WP, Morris M, Periasamy M. AT1b receptor predominantly mediates contractions in major mouse blood vessels. Circ Res. 2003; 93: 1089–1094.
Russell A, Watts S. Vascular reactivity of isolated thoracic aorta of the C57BL/6J mouse. J Pharmacol Exp Ther. 2000; 294: 598–604.
Alioua A, Mahajan A, Nishimaru K, Zarei MM, Stefani E, Toro L. Coupling of c-Src to large conductance voltage- and Ca2+-activated K+ channels as a new mechanism of agonist-induced vasoconstriction. Proc Natl Acad Sci U S A. 2002; 99: 14560–14565.
Che Q, Carmines PK. Src family kinase involvement in rat preglomerular microvascular contractile and [Ca2+]i responses to ANG II. Am J Physiol Renal Physiol. 2005; 88: F658–F664.
Xiang Y, Devic E, Kobilka B. The PDZ binding motif of the β1 adrenergic receptor modulates receptor trafficking and signaling in cardiac myocytes. J Biol Chem. 2002; 277: 33783–33790.
Hu LA, Chen W, Premont RT, Cong M, Lefkowitz RJ. G protein-coupled receptor kinase 5 regulates β1-adrenergic receptor association with PSD-95. J Biol Chem. 2002; 277: 1607–1613.
Johnson LR, Scott MGH, Pitcher JA. G protein-coupled receptor kinase 5 contains a DNA-binding nuclear localization sequence. Mol Cell Biol. 2004; 24: 10169–10179.
Antihypertensive therapy is complicated, and patients are often required to be on multiple drugs to control their blood pressure. A more thorough understanding of the pathogenesis of the disease is needed to tailor antihypertensive therapy specifically and appropriately. Signaling by G protein–coupled receptors, which can mediate both vascular smooth muscle (VSM) constriction and dilation, are tightly regulated by G protein–coupled receptor kinases (GRKs). We recapitulated a particular aspect of hypertension by creating transgenic mice with VSM-specific overexpression of GRK5. We found not only that these mice were hypertensive but also that blood pressure was increased to a greater extent in male versus female mice without concomitant cardiac or VSM hypertrophy. Gi-mediated signaling is critically involved. In male mice, β1-adrenergic receptors appear to contribute, whereas in female mice, Ang II receptor sensitivity is increased. These data suggest that VSM GRK5 levels can have a significant effect on G protein–coupled receptor signaling. In addition, they would suggest that β-blockers would be a more effective antihypertensive therapy in males, whereas Ang II receptor blockers would be more effective in females with elevated GRK5 expression. Thus, data derived from these studies could lend further insight into the sex differences in human cardiovascular disease. In addition, understanding mechanisms underlying an increase in VSM GRK5 may have a profound influence on the use and development of antihypertensive therapeutics.
↵*The first 2 authors contributed equally to this work.