Cardiovascular Influences of α1b-Adrenergic Receptor Defect in Mice
Background— The α1-adrenergic receptors (α1-ARs) play a key role in cardiovascular homeostasis. However, the functional role of α1-AR subtypes in vivo is still unclear. The aim of this study was to evaluate the cardiovascular influences of α1b-AR.
Methods and Results— In transgenic mice lacking α1-AR (KO) and their wild-type controls (WT), we evaluated blood pressure profile and cardiovascular remodeling induced by the chronic administration (18 days via osmotic pumps) of norepinephrine, angiotensin II, and subpressor doses of phenylephrine. Our results indicate that norepinephrine induced an increase in blood pressure levels only in WT mice. In contrast, the hypertensive state induced by angiotensin II was comparable between WT and KO mice. Phenylephrine did not modify blood pressure levels in either WT or KO mice. The cardiac hypertrophy and eutrophic vascular remodeling evoked by norepinephrine was observed only in WT mice, and this effect was independent of the hypertensive state because it was similar to that observed during subpressor phenylephrine infusion. Finally, the cardiac hypertrophy induced by thoracic aortic constriction was comparable between WT and KO mice.
Conclusions— Our data demonstrate that the lack of α1b-AR protects from the chronic increase of arterial blood pressure induced by norepinephrine and concomitantly prevents cardiovascular remodeling evoked by adrenergic activation independently of blood pressure levels.
Received December 18, 2001; revision received January 29, 2002; accepted January 29, 2002.
Essential hypertension is considered a family disease that results from environmental factors that accumulate over time in genetically susceptible persons. Interactions between an adverse environment and susceptible genes can induce stable derangement of blood pressure homeostasis. The sympathetic nervous system represents one of the main mechanisms by which the body organizes responses to environmental changes.1 The α1-adrenergic receptors (ARs) have a crucial role in the impact of sympathetic nervous activity on the regulation of arterial pressure, mediating potent vasoconstrictive response and changes in cardiovascular structure.2 The α1-AR family includes 3 α1-receptor subtypes (a, b, and d), and thus far, the role of these α1-AR subtypes in cardiovascular function is unclear. To elucidate this issue, we have generated knockout mice deficient in α1b-AR (KO mice).3 These mice have a reduced vascular contractility and blunted blood pressure response to acute infusion of phenylephrine (PHE), which suggests that α1b-AR may play an important role in the development of high blood pressure. The generation of a genetic-based mouse model of α1b-AR ablation allows further definition of the role of this AR subtype in the chronic regulation of arterial pressure, thus clarifying whether a selective defect in sympathetic nervous receptor signaling may play a key role in hypertension. Moreover, this mouse model represents a good method for exploring the influence of α1b-AR in the cardiovascular structural changes during chronic pressure overload induced by both humoral and mechanical stimuli.
Therefore, we first examined whether mice that have an ablated α1b-AR subtype are protected from chronic blood pressure increase and the consequent cardiovascular remodeling evoked by norepinephrine (NE), the main sympathetic neurotransmitter. Second, we explored whether the lack of α1b-AR may also interfere with heterologous stimuli that evoke chronic hypertension, such as angiotensin II (Ang II). Third, we tested whether the α1b-AR subtype may influence cardiovascular remodeling independently of changes in blood pressure. Finally, we examined whether ablation of the α1b-AR subtype may affect the cardiac remodeling evoked by a biomechanical stress caused by thoracic aortic constriction.
The generation of α1b-AR KO mice has been described previously.3 In particular, 129/C57 BL is the common genetic background for KO and their wild-type (WT) controls, which derive from the same germline of microinjected clones. For each strain, mice from different litters were randomly intercrossed to obtain WT and KO progenies and never intercrossed with other strains or mated with those from the same litters; thus, their genetic background should not be too different from that of the animals described in the previous study.3 Since 1997, at least 40 intercrosses have been made. WT mice were derived from the initial intercrosses between the heterozygous mice and bred in parallel with the KO progeny.
Mice were kept in standard cages under a 12-hour light/dark cycle, fed ad libitum, and cared for according to the guidelines of the Italian government; a local ethics committee approved all study protocols. All experiments were conducted in male mice that were 3 months old.
Radiotelemetric Blood Pressure Measurement
After WT and KO mice were anesthetized with an intraperitoneal injection of a mixture of ketamine and xylazine at doses of 100 and 10 mg/kg body weight, respectively, the left femoral artery was exposed and cannulated, as distal as possible to preserve collateral circulation, with a 0.4-mm-diameter catheter connected to the radiotelemetric device (TA11PA-C20, Data Sciences International) anchored subcutaneously to the left side of the back. The radiotelemetric device, with a package volume of 2 cc and a weight of 3.36 g, imposed on the mice a load equivalent to ≈4% of their body weight. Each instrumented mouse was housed in a single cage and allowed 10 days’ recovery from surgical procedures, then blood pressure and heart rate were continuously recorded for 4 hours daily (from 8 am to noon) in basal conditions (4 days) and during the period of active treatment (18 days). From a receiver placed underneath each cage, the telemeter pressure signals were consolidated by the multiplexer and stored in a dedicated computed data acquisition system (Dataquest Acquisition and Analysis System, DQ ART 1.1 Gold, Data Sciences International). A dedicated software analysis program (Dataquest Acquisition and Analysis System, DQ ART 1.1 Gold) calculated the mean value during the acquisition time.
Chronic Receptor Agonist Infusions
NE (2.5 mg · kg−1 · d−1 in 0.2% ascorbic acid; WT n=15, KO n=16), Ang II (0.5 mg · kg−1 · d−1 in 0.9% NaCl; WT n=11, KO n=13), PHE (100 μg · kg−1 · d−1 in 0.2% ascorbic acid; WT n=14, KO n=15), and vehicle (WT n=33, KO n=36) were infused for 18 days through osmotic minipumps with a mean fill volume of 246 μL and a weight of 1.36 g (Alzet model 2004, Alza Corp). These latter were implanted subcutaneously on the right side of the back 14 days after implantation of the radiotelemetric devices. After the treatments, the animals were randomly sorted in the different phenotypic evaluations.
Cardiac and Vascular Remodeling
At the end of experimental series, the animals were weighed and then killed by decapitation. The heart was promptly dissected, and left ventricular weight/body weight ratio (LVW/BW, mg/g) was calculated. At the same time, from each mouse, mesenteric vessels corresponding to the second branch (≈140 to 200 μm of average diameter in relaxed conditions, 2 mm long) were obtained by dissection. The vessels were excised free of connective and adipose tissue, and 2 stainless steel wires of 40-μm diameter were threaded through the lumen. Total time for dissection and preparation was ≈45 minutes. Vessels were then equilibrated and relaxed for at least 30 minutes in physiological saline solution (PSS) with the following composition (in mmol/L): NaCl 119, NaHCO3 24, KCl 4.7, KH2PO4 1.18, MgSO4 1.17, CaCl2 2.5, and glucose 5.5, kept constantly at 37°C and bubbled with 5% CO2 in O2. After equilibration, the micromyograph was transferred to the stage of a light microscope with an immersion lens. The vessel was stretched slightly (wall tension ≈0.1 N/m), and structural characteristics of the vessels were evaluated. The following parameters were measured: wall thickness, media thickness, adventitia thickness, intima thickness, internal diameter, media/lumen ratio, and media cross-sectional area (MCSA). Then, the normalized internal circumference L1 was determined from the resting wall tension–internal circumference relation and Laplace equation (L1 is defined as 0.9×L100, where L100 is an estimate of the internal circumference that the vessel would have had in vivo when subjected to a transmural pressure of 100 mm Hg while relaxed).6 From L1, the normalized internal diameter l1 was calculated. With the assumption that the cross-sectional area remains constant when the vessel is extended to L1, the previously mentioned morphological parameters were also automatically calculated in a normalized condition.6
where MID (media internal diameter) = internal diameter plus 2 intimal thicknesses; MED (media external diameter) = internal diameter plus 2 intimal thicknesses plus 2 media thicknesses; n = normal (infusion of vehicle); and p = pathological (after infusion of PHE, NE, or Ang II). It quantifies how much of the vascular structural alteration may be explained by a rearrangement of the same material around a narrowed lumen, without cell growth.7 The growth index quantifies the component due to vascular smooth muscle cell growth, and it has been evaluated with the formula (MCSAp−MCSAn)/MCSAn.7 A remodeling index of 90% means that 90% of the observed change in the media/lumen ratio may be explained by eutrophic remodeling; a growth index of 20% means that 20% of the observed change in the media/lumen ratio may be explained by hypertrophic remodeling. Because of the different methods of calculation of remodeling and growth indexes, the sum of the values can be >100%. A remodeling index >100% usually means that all the structural changes are caused by eutrophic remodeling. In PHE- and vehicle-treated mice, a further evaluation of vascular smooth muscle cell morphology was performed through the dissector technique, as described previously.8
Transverse Aortic Constriction
After anesthesia was induced as described above, WT (n=21) and KO (n=23) mice were placed on thermal beds to prevent thermal dispersion, and respiration was assisted with a volume-cycled ventilator (Basile) connected to an 18-gauge cannula inserted into the trachea. Transverse aortic constriction (TAC) was performed as described previously.9 A further group of both mouse strains underwent the same surgical procedures without realizing aortic stenosis (sham: WT n=15, KO n=16).
To evaluate the hemodynamic overload imposed on the left ventricle, the systolic pressure gradient was measured by selective cannulation of left and right carotid arteries 7 days after TAC.9 After hemodynamic evaluations, mice were weighed, then the hearts were excised, and LVW/BW ratio was calculated. Thereafter, left ventricular tissues were processed for further analysis. Total perioperative and postoperative mortality was not significantly different in WT and KO mice (20% and 16%, respectively).
Seven days after TAC and after 18 days of PHE infusion, echocardiographic analysis was performed in WT and KO mice anesthetized with tribromoethanol (350 mg/kg IP) and ketamine (50 mg/kg IP). Echocardiographic examination was performed with an HP Sonos 100 (Hewlett Packard Co) equipped with a 7.5-MHz imaging transducer. After good-quality 2D short-axis images of the left ventricle were obtained, M-mode freeze frames were printed on common echocardiographic paper. End-diastolic interventricular septum and posterior wall thickness and end-systolic (ESD) and end-diastolic (EDD) left ventricular internal diameters were measured by a National Institutes of Health image-analysis system. Percent fractional shortening was calculated as [(EDD−ESD)/EDD]×100. Estimated echocardiographic left ventricular mass (LVMi) was calculated as LVMi=[(EDD−end-diastolic interventricular septum+posterior wall thickness)3−EDD3]×1.055, where 1.055 is the density of mouse myocardium.
Left Ventricular Atrial Natriuretic Factor mRNA
Left ventricular atrial natriuretic factor (ANF) mRNA expression was evaluated in WT and KO TAC and PHE-treated mice as described previously.9
All data are expressed as mean±SD unless otherwise stated. One-way ANOVA and Bonferroni correction for multiple comparisons were used to evaluate differences among groups. A nonparametric approach (Mann-Whitney rank sum test) was adopted for the variable (cell layers) that was not normally distributed. Two-way ANOVA for repeated measures was used for blood pressure and heart rate (group×time; BMDP Statistical Software programs 7D, 3S, 1V and 2V, BMDP Statistical Software Inc).
As shown in Figure 1, in the days preceding the various treatments, basal systolic blood pressure, diastolic blood pressure, and heart rate were similar between WT and KO mice. Chronic NE infusion induced a progressive and slow increase (from day 4) in both systolic and diastolic blood pressure in WT but not in KO mice (Figure 1a and Figure 2). In contrast, Ang II evoked a significant and progressive increase (from day 2) in systolic and diastolic blood pressures, which were comparable in both mouse strains (Figure 1b). As expected, PHE did not affect the blood pressure profile in either WT or KO mice (Figure 1c). Basal heart rate was similar in WT and KO mice and was not significantly influenced by any treatment (Figure 1).
As shown in Table 1, at the end of all experimental series, WT and KO mice showed a BW similar to that observed in basal conditions. NE and PHE induced a significant increase in absolute LVW and subsequently LVW/BW in WT mice but not in KO mice. In contrast, Ang II raised LVW and LVW/BW to a similar extent in both mouse strains.
Seven days after the surgical procedures, both mouse strains had increased systolic blood pressure in the right carotid artery compared with that recorded in the left carotid artery (WT 174±12 versus 105±8 mm Hg, P<0.05; KO 171±13 versus 100±10 mm Hg, P<0.05). Consequently, pressure overload, evaluated as systolic pressure gradient, was similar in both WT and KO mice (69±4 versus 71±5 mm Hg, P=NS). In response to TAC, absolute LVW and LVW/BW increased markedly to a similar extent in both WT and KO mice compared with their sham controls (Table 1).
As shown in Table 2, echocardiographic analysis showed that at the end of the infusion period, PHE induced a significant increase in cardiac wall thickness and LVMi in WT mice compared with vehicle-treated animals. This effect was absent in KO mice. No significant changes in cardiac diameters or fractional shortening were observed in either mouse strain during PHE infusion compared with vehicle-treated mice. Finally, TAC induced a similar increase in cardiac wall thickness and LVMi in both WT and KO mice compared with sham.
As expected, LV ANF mRNA expression was undetectable in control conditions in both WT and KO mice. PHE evoked a significant increase in LV ANF mRNA expression compared with vehicle in WT but not in KO mice. In contrast, TAC induced a similar increase in LV ANF mRNA expression compared with sham in both mouse strains (Figure 3b).
As shown in Table 3, after infusion of NE, WT mesenteric arteries showed a tendency to reduction of internal diameter and a significant increase in media/lumen ratio, which was not seen in KO mice. Similarly, PHE induced a reduction in internal diameter and an increase in media/lumen ratio in WT but not in KO mice (Figure 4). The increase in media/lumen ratio evoked by both PHE and NE in WT mice was caused by inward eutrophic remodeling, as indicated by the calculation of the remodeling index (>100%) and growth index (close to 0%), as well as by the lack of differences in MCSA. No significant difference in cell volume, cell length, cell cross-sectional area, number of cells per segment length, or number of cell layers was observed between WT and KO mice after infusion of vehicle or PHE (Table 4). Therefore, no evidence of hypertrophic remodeling was present, in terms of hypertrophy or hyperplasia, which again supports the presence of eutrophic remodeling after infusion of PHE.
In contrast, the infusion of Ang II was accompanied by a trend toward hypertrophic remodeling as suggested by a slight increase in MCSA (P=0.052 in WT and P=0.080 in KO mice versus infusion of vehicle). In addition, calculation of the growth index suggested the presence of 20% to 30% cell growth.
The main finding of this study is that α1b-AR–deficient mice are selectively protected from the chronic increase in arterial blood pressure induced by NE, the main sympathetic neurotransmitter. Moreover, we provide evidence that α1b-ARs are involved in the cardiovascular remodeling evoked by adrenergic agonists, independently of blood pressure levels.
It is well known that the sympathetic nervous system is involved in the pathogenesis of chronic arterial hypertension. It has been reported that catecholamines are among the leading candidates in initiating both the rise in blood pressure and the trophic mechanism that maintains hypertension.1 In the present study, we demonstrated that the hypertensive effects of catecholamines were mediated by the α1b-AR, because NE chronic infusion failed to increase blood pressure in mice deficient in this receptor subtype. The lack of the α1b-AR does not appear to interfere with other conditions that cause chronic hypertension, such as that induced by Ang II.
In the present study, no significant changes in heart rate were observed in response to agonist-induced hypertension, although a baroreflex drop in heart rate could have been expected. However, this latter phenomenon is mainly observed when an acute increase in blood pressure is realized, whereas in the present study, NE and Ang II attained their pressor effect slowly and progressively, likely activating long-term blood pressure counterbalancing mechanisms rather than a reflex decrease in heart rate. Moreover, NE and, in small part, Ang II have direct or indirect positive effects on heart rate that could have concealed heart rate baroreflex.
The role of the sympathetic nervous system in cardiovascular homeostasis also involves its action on cardiovascular remodeling, mediated mainly through α1-ARs. This experimental evidence prompted us to explore the role of α1b-ARs in the cardiovascular structural changes consequent to chronic infusion of NE, mimicking hyperactivity of the sympathetic nervous system. Our data demonstrate that mice deficient in the α1b-AR do not show NE-evoked cardiac hypertrophy compared with control animals. This effect cannot be attributed solely to the fact that KO mice are protected from NE-evoked hypertension, because a similar impairment in cardiac hypertrophic development has also been observed after a subpressor dose of PHE, the homologous receptor agonist. Our findings are fully supported by previous observations demonstrating that a targeted cardiac overexpression of a constitutively active α1b-AR mutant in transgenic mice, without influencing blood pressure, results in activation of biochemical mechanisms that lead to the development of cardiac hypertrophy.10,11⇓ Thus, the α1b-AR pathway is the main receptor signaling that mediates catecholamine-induced cardiac hypertrophic remodeling, beyond its effect on blood pressure homeostasis.
Recently, it has been reported that transgenic mice overexpressing tissue-specific α1b-AR show hypotension and a decreased pressor response to α1-adrenergic agonists, which suggests that this AR subtype is not involved in the adrenergic-mediated blood pressure increase.10 These findings could appear to be in conflict with the results of the present study. However, the results from the 2 studies cannot be compared directly, because mice overexpressing α1b-AR showed important neurodegenerative phenomena, with a global autonomic dysfunction and blunted sympathetic activity, which can indirectly affect blood pressure homeostasis. On the other hand, these mice, which overexpress the transgene exactly where α1b-ARs are distributed physiologically, show myocardial hypertrophy despite hypotension, which indicates the involvement of α1b-AR in the mechanisms of cardiac growth. This evidence fully supports our results that demonstrate an involvement of α1b-AR in the development of cardiac hypertrophy independently of hypertensive conditions.
Another aim of the present study was to determine whether α1b-ARs are generally involved in all conditions that result in cardiac hypertrophic remodeling or whether they are specific mediators of the cardiac structural changes induced by adrenergic agonists. Our results indicate that a heterologous receptor stimulus, such as Ang II, as well as aortic coarctation, a well-characterized model of biomechanical stress,12 induced cardiac hypertrophic remodeling to a similar extent in both KO and WT mice. These data demonstrate that α1b-ARs are selectively involved in catecholamine-induced cardiac hypertrophic remodeling, and their signaling does not participate significantly in the cardiac hypertrophy evoked by other mechanisms.
The protective effect of α1b-AR deficiency seen in catecholamine-induced cardiac hypertrophy is also observable at the vascular level. In particular, whereas WT mice treated with pressor doses of NE or subpressor doses of PHE develop a eutrophic remodeling (a rearrangement of normal tissue around a narrowed lumen) in the mesenteric artery, KO mice are protected against these vascular structural changes. Even in this case, as discussed previously for cardiac remodeling, we can conclude that the eutrophic remodeling observed in mesenteric vessels must be ascribed to adrenergic rather than blood pressure signaling, because Ang II evoked a blood pressure elevation more pronounced that NE without clear vascular remodeling.
The fact that Ang II evoked only modest changes of borderline statistical significance in media/lumen ratio and in MCSA in both mouse strains is likely due to a low sensitivity of mouse arteries to Ang II. However, this fact does not weaken the evidence that intact α1b-AR signaling is required for the development of adrenergic-evoked vascular eutrophic remodeling.
Although the precise mechanisms of eutrophic vascular remodeling are unknown, it is considered to be a slowly developing process unassociated with shrinkage of intima or adventitia. Accordingly, a rearrangement of structural elements may occur in which changes in connections between extracellular matrix proteins and vascular smooth muscle cells are not fixed but rather labile, which allows individual cells or cell layers to move relative to each other. This process has been suggested to be integrin mediated.13 Alternatively, but less probably, is the explanation that there is a simultaneous ongoing process of apoptosis/hyperplasia with no changes in total cell number that nonetheless results in a change in the architecture of the vessel wall.14 The main parameters necessary to define vascular eutrophic remodeling are an increase in media/lumen ratio and no change in MCSA.14 Eutrophic remodeling has been described in human essential hypertension and in several animal models of genetic or experimental hypertension.7,14,15⇓⇓
Indeed, in vitro studies have observed that catecholamines also exert a growth effect on smooth muscle cells.16 Our data showing that chronic NE infusion in mice does not evoke any effect on vascular growth could appear to be in conflict with these previous observations. However, Rizzoni et al17 recently observed eutrophic vascular remodeling in patients with pheochromocytoma similar to that observed in essential hypertensives, with no changes in the vascular smooth muscle cell component, which, in contrast, is significantly increased in patients with renovascular hypertension. Thus, it must be emphasized again that because the vasculature is a highly differentiated organ system composed of multiple cell types, a number of other factors, which are lacking in the cell culture system, can concomitantly participate in vascular remodeling, which confirms the limitations of comparing in vitro and in vivo experiments.
A potential limitation of the present study is that our analysis was focused only on the mesenteric vascular bed and was not extended to other resistance vessels. Unfortunately, the micromyographic technique, although miniaturized for mice studies, does not enable extension of our structural analysis to other resistance vessels because of technical limitations. However, several studies conducted in other species clearly indicate that mesenteric vascular remodeling is strongly related to structural changes of other resistance vessels18 in which the presence of α1b-AR has also been reported.19 Thus, it is unlikely that catecholamine-evoked vascular remodeling is selective for the mesenteric vascular bed.
In conclusion, our results provide the first evidence that the α1b-AR plays a key role in mediating the blood pressure response and cardiovascular structural adaptation to chronic adrenergic stimulation. Therefore, the α1b-AR may be a target of novel therapeutic strategies focused on modulation of blood pressure responses and specific features of cardiovascular remodeling in conditions of sympathetic nervous system activation.
The authors thank engineer Gianfranco Lauria for his expert technical assistance.
↵*Drs Vecchione and Fratta contributed equally to this work.
- ↵Mark AL. The sympathetic nervous system in hypertension: a potential long term regulator of arterial pressure. J Hypertens. 1996; 14: S159–S165.
- ↵Chen L, Xin X, Eckhart AD, et al. Regulation of vascular smooth muscle growth by α-adrenoreceptor subtypes in vitro and in situ. J Biol Chem. 1995; 270: 30980–30988.
- ↵Cavalli A, Lattion AL, Hummler E, et al. Decreased blood pressure response in mice deficient of the alpha1b adrenergic receptor. Proc Natl Acad Sci U S A. 1997; 94: 11589–11594.
- ↵Laycock SK, McMurray J, Kane KA, et al. Effects of chronic norepinephrine administration on cardiac function in rats. J Cardiovasc Pharmacol. 1995; 26: 584–589.
- ↵Daugherty A, Cassis L. Chronic angiotensin II infusion promotes atherogenesis in low density lipoprotein receptor -/- mice. Ann N Y Acad Sci. 1999; 892: 108–118.
- ↵Rizzoni D, Castellano M, Porteri E, et al. Delayed development of hypertension after short-term nitrendipine treatment. Hypertension. 1994; 24: 131–139.
- ↵Heagerty AM, Aalkjaaer C, Bund SJ, et al. Small artery structure in hypertension: dual process of remodeling and growth. Hypertension. 1993; 21: 391–397.
- ↵Korsgaard N, Mulvany MJ. Cellular hypertrophy in mesenteric resistance vessels from renal hypertensive rats. Hypertension. 1988; 12: 162–167.
- ↵Lembo G, Rockman HA, Hunter JJ, et al. Elevated blood pressure and enhanced myocardial contractility in mice with severe IGF-1 deficiency. J Clin Invest. 1996; 98: 2648–2655.
- ↵Zuscik MJ, Chalothorn D, Hellard D, et al. Hypotension, autonomic failure, and cardiac hypertrophy in transgenic mice overexpressing the αb-adrenergic receptor. J Biol Chem. 2001; 276: 13738–13743.
- ↵Milano CA, Dolber PC, Rockman HA, et al. Myocardial expression of a constitutively active alpha 1B-adrenergic receptor in transgenic mice induces cardiac hypertrophy. Proc Natl Acad Sci U S A. 1994; 91: 10109–10113.
- ↵Hirota H, Chen J, Betz UA, et al. Loss of a gp130 cardiac muscle cell survival pathway is a critical event in the onset of heart failure during biomechanical stress. Cell. 1999; 97: 189–198.
- ↵Lee RT, Berditchevski F, Cheng GC, et al. Integrin-mediated collagen matrix reorganization by cultured human vascular smooth muscle cells. Circ Res. 1995; 76: 209–214.
- ↵Intengan HD, Schiffrin EL. Vascular remodeling in hypertension: role of apoptosis, inflammation and fibrosis. Hypertension. 2001; 38: 581–587.
- ↵Korsgaard N, Aalkiaer C, Heagerty AM, et al. Histology of subcutaneous small resistance arteries from patients with essential hypertension. Hypertension. 1993; 22: 523–526.
- ↵Yu SM, Tsai SY, Guh JH, et al. Mechanism of catecholamine-induced proliferation of vascular smooth muscle cells. Circulation. 1996; 94: 547–554.
- ↵Rizzoni D, Porteri E, Guelfi D, et al. Cellular hypertrophy in subcutaneous small arteries of patients with renovascular hypertension. Hypertension. 2000; 35: 931–935.
- ↵Mulvany MJ, Aalkjaer C. Structure and function of small arteries. Physiol Rev. 1990; 70: 921–961.
- ↵Hrometz SL, Edelmann SE, McCune DF, et al. Expression of multiple alpha1-adrenoceptors on vascular smooth muscle: correlation with the regulation of contraction. J Pharmacol Exp Ther. 1999; 290: 452–463.