Abstract We studied the long-term effects after withdrawal of enalapril, an angiotensin-converting enzyme inhibitor, on tail systolic pressure and cardiovascular structural properties in spontaneously hypertensive rats (SHR) and Wistar-Kyoto rats (WKY). Observations in control rats were from 4 to 35 weeks of age, whereas treated rats received enalapril from 4 to 20 weeks and were studied for a further 15 weeks. We measured blood pressure and the ratio of left ventricle weight to body weight and derived methoxamine log dose–perfusion pressure curves in the isolated hindquarter bed. From the changes in resistance properties we also estimated the changes in structure using a model developed previously. During therapy, blood pressure was depressed to a common value in both strains. After drug withdrawal, by age 35 weeks, previously treated SHR developed only mild hypertension, whereas blood pressure of WKY had recovered to the corresponding control level. At 21 weeks, soon after enalapril was stopped, left ventricular development was depressed in both strains; the depression was slightly greater in SHR, but that of vascular resistance was proportionately similar in each strain. Late cardiovascular development between 21 and 35 weeks was attenuated in the previously treated groups. For the left ventricle, it was similar in each strain, but for the vasculature, late development was relatively smaller in SHR than WKY. In the former, the pattern of development between 21 and 35 weeks was the same as in untreated controls and appeared to be mediated in response to the rise in blood pressure. In previously treated WKY, the rise in blood pressure was smaller, so that their relatively greater late vascular development was largely through another mechanism, probably involving a direct action on smooth muscle growth, through a product of the converting enzyme. This non–blood pressure–dependent component of late development was absent in previously treated SHR. We conclude that the products of angiotensin-converting enzyme play an important role in the normal development of each strain. The between-strain differences in late development contribute to the relatively greater restoration of blood pressure in previously treated WKY compared with SHR and for the mildness of the hypertension in the latter. In normal cardiovascular development, the action of converting enzyme is relatively nonspecific, suggesting that it does not play other than a general role in the pathogenesis of hypertension.
In the spontaneously hypertensive rat (SHR), vascular resistance is elevated even at full dilatation, suggesting narrowing of the small arteries.1 2 Such narrowing, in conjunction with an increase in their wall-lumen ratio, makes the vessels into “amplifiers” of constrictor and dilator stimuli.2 3 4 5 6 Another structural change, concentric hypertrophy of the left ventricle (LV), makes the latter into a stroke volume amplifier, which helps the heart maintain cardiac output against an increased blood pressure (BP).7 8 9 The structural amplifiers contribute to the maintenance of the elevated BP in virtually every type of hypertension.2 4 10 In SHR, LV hypertrophy and the characteristic vascular resistance properties are already present early in life, as assessed from comparisons with normotensive Wistar-Kyoto rats (WKY), suggesting that they may play a role in the pathogenesis of hypertension.11 12 One candidate for causing hypertension is the sympathetic nervous system, because after extensive suppression of sympathoadrenal activity early in life, the subsequent hypertension and structural changes were completely prevented.12
The candidacy of the renin-angiotensin system is based largely on studies with angiotensin-converting enzyme (ACE) inhibitors in young SHR. ACE inhibitors can lower BP by reducing angiotensin II (Ang II) production and decreasing bradykinin degradation, but it is thought that most of the actions on BP and cardiovascular hypertrophy are through Ang II.13 14 15 16 Earlier work by ourselves and others17 18 19 20 21 22 23 has shown that after a relatively short period of treatment in young SHR, the hypertension that develops after drug withdrawal is attenuated when compared with that of untreated SHR. In most previous studies on the long-term effects of ACE inhibitors on SHR, including our own,21 treatment has been confined to the hypertensive strain, so little information is available on the effects on WKY or other normotensive strains. There are advantages of including the latter in the protocol for determining the specificity of the response. Indeed, attenuation of hypertension has not been observed in every strain with genetic hypertension.24
Our aim was to determine the long-term effects on BP and cardiovascular structure after withdrawal of enalapril in SHR and WKY and the specificity of its actions. The timing of therapy, from 4 to 20 weeks, extended over the period of rapid BP rise in SHR.11 We measured BP, the ratio of LV weight to body weight (LV/BW ratio), and the resistance properties of the isolated hindquarter (HQ) bed in untreated and enalapril-treated rats from each strain and continued observations until the animals were 35 weeks old. From the observed changes in resistance properties after enalapril, we estimated the corresponding average structural changes using a model of the vasculature in which these interrelationships have been studied previously.6
Animals and Protocols
We used male SHR and WKY bred at the Baker Medical Research Institute from stock from the Shimane Institute of Health Science, kindly supplied in 1985 by Prof Y. Yamori. Housing and feeding were as described previously.11 The study was approved by the Baker Institute–Alfred Hospital Animal Ethics Committee and conformed to the guidelines of the Australian National Health and Medical Research Council.
The main series of experiments consisted of control and enalapril-treated SHR and WKY (SHRc, WKYc and SHRen, WKYen). Enalapril maleate (Merck, Sharp & Dohme) was given in the rats’ drinking water at a dose of 25 to 30 mg/kg per day between the ages of 4 and 20 weeks as described previously.21 BP and BW were measured from 4 weeks of age; HQ resistance properties, LV weight, and LV/BW ratio were measured at 21 and 35 weeks.
BP, BW, and LV Weight
Tail systolic BP and BW were measured twice weekly in conscious rats, with the results averaged for each week. For BP measurements, the rats were placed in a test chamber maintained at approximately 28°C.11 LV weight (including the septum) was measured at 21 and 35 weeks. We have reported both unadjusted LV weights and LV/BW ratios. However, LV/BW ratio is the preferred index for assessment of relative LV mass, because we previously found that it makes adequate allowance for BW-related differences in LV weight in age-matched groups with different BW values.12
HQ resistance properties were assessed by the method of Folkow et al,1 modified as described previously.11 12 21 Rats were anesthetized with sodium pentobarbital (veterinary Nembutal, Abbott, 60 mg/kg IP) and then were heparinized. The lower aorta was cannulated, the vessels were ligated, and the HQ was perfused with 1.5% dextran T-70 (Pharmacia) in Tyrode’s solution (for composition, see Reference 1212 ) aerated with 95% O2/5% CO2. Soon after the start of perfusion, the HQ bed was dilated fully with boluses of papaverine HCl (2.4 mg/10 mL perfusate). The fully dilated bed was perfused at three flow rates (5, 10, and 15 mL/min per 100 g HQ weight), and we calculated the regression equation relating perfusion pressure (PP) to flow from the pooled data of each group.
Subsequently, we determined the relationship between the dose of methoxamine HCl (Sigma Chemical Co) and PP of the bed. The HQ bed was perfused at 10 mL/min per 100 g HQ weight, estimating HQ weight from regression equations relating HQ weight to BW previously determined for SHR and WKY of the same colony.11 Each methoxamine log dose–PP curve was obtained starting at minimal PP (PPmin) and progressively increasing the methoxamine concentration in the perfusate from 0.01 to 50 to 100 μg/mL until maximal PP (PPmax); the latter was verified by injection of a 25 to 50 μg/mL bolus of Ang II (CIBA-GEIGY). A logistic function was fitted to the data of each rat with the use of a computer program developed by J. Ricketts and G.A. Head at the Baker Institute. From this we obtained the parameters PPmin, PPmax, the average slope, and EC50 (drug concentration needed to reach 50% of the range from PPmin to PPmax). The differences in the resistance properties between SHR and WKY are related to the structural differences between them.1 2 3 5 6 25
One-way ANOVA, orthogonally partitioned into individual degrees of freedom, was performed to assess the significance of the various differences.26 For each comparison, we assessed the significance of differences between SHR and WKY caused by enalapril and the changes at the different times within a given treatment group in each strain. In addition, we calculated linear regression equations relating PPmin to HQ flow in the fully dilated bed under control conditions and after enalapril. We used ANCOVA to compare the responses of similarly treated age-matched SHR and WKY and to assess the significance of the differences in their regression coefficients and in the magnitude of the intercepts between the regression lines.26
BP and BW
Systolic BP was higher in SHRc than in WKYc through the entire observation period from 4 to 35 weeks (Fig 1⇓, bottom left). At 4 weeks, BP in SHRc was 126±4.5 mm Hg, compared with 109±3.8 mm Hg in WKYc (P<.01). BP increased with age in each strain: at 20 and 21 weeks it was 185±3.0 mm Hg in SHRc and 139±1.7 mm Hg in WKYc; during the last 5 weeks of observations, from 30 to 35 weeks, it was 200±3.9 mm Hg in SHRc and 136±2.8 mm Hg in WKYc (Fig 1⇓).
From the start of enalapril treatment, the BP of each strain was below the corresponding control value. During the last 5 weeks of treatment, from 15 to 20 weeks of age, BP was closely similar in each strain, averaging 120 mm Hg (Fig 1⇑, bottom right). In SHRen, this was 63 mm Hg (34%) below the BP of SHRc, and in WKYen, it was 19 mm Hg (14%) below that of WKYc. When enalapril was stopped at 20 weeks of age, BP rose in each strain, with the rise greater in SHRen than in WKYen (Fig 1⇑). Between 30 and 35 weeks, it reached 146.4±2.4 mm Hg in SHRen and 129±1.8 mm Hg in WKYen. Thus, the difference between strains was approximately one quarter that between untreated controls.
Growth tended to be more rapid in WKYc than in SHRc (Fig 1⇑), with the BW difference between strains approximately 14% averaged over 21 and 35 weeks (Table 1⇓, P<.001). Enalapril retarded growth to a somewhat greater degree in WKY than SHR compared with their respective controls (14% versus 2.7%, P<.01; Table 1⇓). Soon after the end of treatment, BW was closely similar in SHRen and WKYen (Fig 1⇑, top right). By 35 weeks of age, BW of SHRen was close to that of SHRc, but in WKYen it was approximately 8% less than in WKYc (P<.01, Table 1⇓).
LV/BW Ratio and Unadjusted LV Weight
At 21 weeks (1 week after drug withdrawal), the LV/BW ratio of both SHRc and WKYc was the same as at 35 weeks (Table 1⇑). In SHRc it averaged 3.08 mg/g, which was 33% greater than in WKYc (2.32 mg/g, P<.001).
After enalapril treatment, the LV/BW ratio was significantly lower in each strain than the corresponding value in untreated controls (Table 1⇑). At 21 weeks, the difference from control was greater in SHRen (0.74±0.10 mg/g) than in WKYen (0.41±0.04 mg/g). At this time, the LV/BW ratio of SHRen was 23% greater than that of WKYen (P<.01).
Between 21 and 35 weeks, the LV/BW ratio increased to a similar degree in both strains, with a rise of 0.22±0.06 mg/g in SHRen and 0.17±0.05 mg/g in WKYen, but the difference between strains remained at 23% at 35 weeks (Table 1⇑). Thus, previous treatment with enalapril affected both strains, with the effect somewhat greater in SHRen, resulting in slight to moderate attenuation of LV hypertrophy compared with untreated controls.
When the unadjusted LV weights were used for the various comparisons, the changes caused by enalapril were qualitatively similar to those observed with LV/BW ratio (Table 1⇑). At both times, significant differences remained between strains in both control and enalapril-treated rats.
HQ Resistance Properties
Methoxamine log dose–PP curves were derived at 21 and 35 weeks of age (Figs 2⇓ and 3⇓, Table 2⇓). In the curves of untreated control rats, the three main parameters (PPmax, PPmin, and slope) were all significantly greater in SHRc than WKYc (Table 2⇓). The ratio of the slope (S) of the dose-response curve in SHRc over that obtained in WKYc, which is a measure of the potency of the vascular amplifier,6 averaged 1.54 at 21 weeks of age and 1.26 at 35 weeks (.1>P>.05). In SHRc, the only parameter to alter significantly between 21 and 35 weeks was PPmax, which increased by 42 mm Hg (P<.001) (Fig 3⇓, Table 2⇓). In WKYc, the dose-response curves were closely similar at both ages, with no significant changes in curve parameters (Fig 3⇓, Table 2⇓).
After enalapril treatment, all three curve parameters were depressed below the corresponding control values, with the percent changes closely similar in each strain (Figs 2⇑ and 3⇑, Table 2⇑). At 21 weeks, PPmax and PPmin were depressed by approximately 20% below control, and the depression in slope was approximately 50% (Fig 3⇑, Table 2⇑). At that time, PPmax, PPmin, and slope were all greater in SHRen than WKYen; the between-strain differences were similar to those between SHRc and WKYc, including the ratio of slopes of SHRen/WKYen, which was 1.46 (Table 2⇑).
Between 21 and 35 weeks, the only parameter to change in SHRen was PPmax, which increased by the same amount as in SHRc (40 mm Hg) (Fig 3⇑, Table 2⇑). This suggests that the rise was unrelated to previous enalapril treatment. By contrast, in WKYen all parameters increased significantly between the above ages, with PPmax and PPmin rising by approximately 10% and slope increasing by approximately 65% from the values at 21 weeks (Table 2⇑). The increase in slope was responsible for the reduction in the amplifier ratio of slopes SHRen/WKYen to only 1.06. Thus, at 35 weeks of age, all parameters of WKYen were closer to those of their age-matched controls than those of SHRen. However, the resistance parameters of both strains were still well below those of their corresponding controls (Fig 3⇑, Table 2⇑).
Pressure-Flow Relationship in the Dilated Bed
Linear regression equations relating PP to the three levels of flow in the fully dilated HQ bed were derived from the pooled data of each group. In the eight rat groups shown in Fig 4⇓, the correlation coefficients ranged from .83 to .97 (mean r=.90±.023) (Table 3⇓).
In each group of SHR and WKY, the regression lines were approximately parallel (Fig 4⇑), with the differences between the regression coefficients not significant but with significant differences between intercepts (Table 3⇑). In the controls (Fig 4⇑, left), the regression coefficients were similar at 21 and 35 weeks of age, with mean values (averaged over strains) of 1.44 and 1.38 mm Hg/mL flow, respectively (Table 3⇑). The age-related differences in intercepts were the same as the differences in PPmin and were greater at 21 weeks (3.8±1.0 mm Hg, P<.001) than at 35 weeks (1.7±0.8 mm Hg, P=.05).
After enalapril, the regression lines derived from SHRen and WKYen were also approximately parallel (Fig 4⇑, right). The elevation of the regression line was less steep than in controls, with the average regression coefficients at 21 and 35 weeks (0.99 and 1.13 mm Hg/mL flow, respectively) significantly below those of the corresponding controls. However, the intercepts between the lines were close to those of the corresponding controls (Fig 4⇑, Table 3⇑). The difference in slopes has been taken to reflect increased distensibility caused by enalapril (see “Discussion”).
Model Estimates of Structure From Resistance Properties
We used a model of the vascular bed in which Korner and Angus6 recently examined some of the relationships between changes and resistance properties. The bed is represented as a single vessel in which Poiseuille’s law applies, so that the average internal radius (r) is inversely related to the fourth root of the vascular resistance (R) by the formula
where k is a constant and l is the length of the bed (see Reference 66 for detailed assumptions). Under in vivo conditions the bed has intrinsic resting tone and, starting from resting, the dose-R (or r) relationship for constrictor agonists is represented by a rectangular hyperbola, and another rectangular hyperbola gives the dose-R (or r) relationship for dilator agonists. The two curves are combined into a single logistic function, the parameters of which define the resistance properties over the full range of vasomotor tone. Under in vitro conditions (as in the HQ bed of the present experiments), the logistic function is derived in one operation from full dilatation to maximal constriction. From the viewpoint of the formulas relating structure to R (see Reference 66 ), we have assumed that in this preparation, “resting” values lie halfway between the R plateaus of the dose-response curve (R at midrange, Rmid). In the present application, we have also assumed that the vascular length l at a given age is the same in SHR and WKY and, furthermore, that changes in l between the ages of 21 and 35 weeks can be neglected within each strain, since by 21 weeks the rapid growth phase is largely over and the resistance properties are fully developed in WKY (Figs 1⇑ and 4⇑).
Korner and Angus6 found that each resistance parameter was affected by all the structural properties of the bed. For example, the slope (S) of the curve and r at full dilatation (rmin) were each influenced independently (1) by the ratio of wall thickness to internal radius (w/r), (2) by the internal radius (r), and (3) by a factor E, which is an inverse measure of relative wall “stiffness” and is directly related to relative wall distensibility. In the normal circulation, E=1.0; it is less than 1.0 with a stiffer (less distensible) bed and greater than 1.0 in a less stiff, more distensible bed.6 This was also the case for R at maximal constriction (rmax) (Korner and Angus, unpublished data).
In the same way, the radius values at the different levels of vasomotor tone in each group are also affected by all the structural properties. Table 4⇓ summarizes the mean values obtained from individual rats of the different preparations, which were examined by one-way ANOVA. In SHRc at 21 weeks, rmin, rmid, and rmax were all approximately 3% narrower than corresponding values in WKYc (P<.01) (see Table 4⇓ footnote for determination of mean±SEM percent changes). Similarly, in 21-week-old rats previously treated with enalapril, the narrowing of SHRen compared with WKYen was approximately 2.5% to 3% at all radii (P<.01), despite the overall widening of the bed immediately after the drug was stopped (see below). By 35 weeks, rmin in SHRc was only 1.4% narrower than in WKYc, which is in accord with the reduced difference in intercepts in the PP-flow lines obtained at full dilatation (Fig 4⇑). However, rmid and rmax were 6% narrower in SHRc than in WKYc. This narrowing in the presence of vasomotor tone was greater than in 21-week-old SHRc (P<.01) because of the structural changes between the two ages (see below). Similarly, in 35-week-old rats previously treated with enalapril, rmin in SHRen was only 1.8% narrower than in WKYen, whereas rmid and rmax were approximately 3.5% narrower (P<.05 for relative difference from rmin).
Immediately after enalapril was stopped, the average r of the bed was wider than under control conditions at each level of tone. In 21-week-old SHRen and WKYen, the values of rmin, rmid, and rmax in each strain were approximately 6% wider than in the corresponding controls (P<.001). By 35 weeks, rmin in SHRen and WKYen was still approximately 4% wider than in controls. However, in SHRen, rmid and rmax were approximately 5.7% wider than in SHRc, which was similar to the relative differences at 21 weeks. By contrast, in WKYen these radii were only approximately 2.5% wider than those of corresponding controls, suggesting that the relative narrowing between 21 and 35 weeks was greater in this strain than in SHRen.
The equation derived by Korner and Angus,6 relating S to w/r, internal radius r (ie, rmid), and the distensibility factor E, is log S=1.734+1.891w/r−3.596r+1.154E, indicating that S varies approximately as the square of w/r, inversely as the fourth power of r, and linearly with E. We used this equation to estimate the relative differences in average structure of the HQ resistance vessels between SHRc and WKYc and the changes with age and treatment in each group (Table 5⇓). In SHRc, the suggested structural changes from 21 to 35 weeks in Table 5⇓ are consistent with the observed changes in resistance properties (Table 2⇑): from the equation, an increase in w/r and narrowing of rmid will increase S, so that to keep the latter the same as at 21 weeks in accordance with experimental observations, distensibility has to be reduced. In 21-week-old SHRen, the suggested structure corresponding to the resistance properties included a wider r (Table 4⇑), a substantially reduced w/r, and an increased distensibility (see “Discussion”). In SHRen, the changes from 21 to 35 weeks in r and w/r were proportionately similar to the corresponding changes in SHRc, but we assumed that restoration of distensibility was only slight. The structural changes are illustrated schematically in Fig 5⇓, which also indicates that at 35 weeks there was still a considerable gap in structure between SHRen and SHRc.
In 21-week-old WKYc, the structure simulating the observed resistance properties relative to SHRc included a wider r, smaller w/r ratio, and more distensible resistance vessels than in the hypertensive strain (Table 5⇑, Fig 5⇑). Between 21 and 35 weeks, the estimated changes in structure were small, consistent with the observed changes in resistance properties. In 21-week-old WKYen, the observed changes in S and r were proportionately similar to those in SHRen, so that we have assumed pro rata changes in w/r and vascular distensibility compared with WKYc. However, in view of the more pronounced rises in all resistance parameters in WKYen between 21 and 35 weeks (Fig 3⇑), late development of the vascular wall and narrowing of the lumen were estimated to be relatively greater than in SHR (Table 5⇑, Fig 5⇑). Despite this, there was still a significant gap in 35-week-old WKYen compared with WKYc.
These findings suggest that ACE plays an important role in normal cardiovascular development in both SHR and WKY. When the drug is stopped after prolonged ACE inhibition in young rats, subsequent development is retarded in older rats. In this discussion, we consider first the significance of the depression in structural properties immediately after enalapril is stopped and the mechanisms underlying the subsequent long-term responses. Next, we consider the various factors that alter BP, including the structural changes. Last, we discuss the advantage of including WKY in the experimental design.
The depression of structural properties by prolonged enalapril treatment was far in excess of the general depression of body growth, particularly in SHR. We have assumed first that at 21 weeks the structural properties were not greatly different from those at the end of treatment, since inhibition of tissue ACE is relatively long lasting.13 Second, we have assumed that the changes in structural properties between 21 and 35 weeks are a measure of the capacity of the older rats to make up for the earlier suppression of development by enalapril. Under vascular “structural properties,” we refer to both the resistance properties and the corresponding transformation of the latter into average units of structure. Because of the nonlinear relationship between resistance properties and structure, this is a convenient way of visualizing the meaning of the changes in resistance parameters. From the simplifying assumptions (see “Model Estimates of Structure From Resistance Properties” above and Reference 66 ), it should be clear that these transforms are not intended as a substitute for morphological analysis. In the event, the estimates of relative structural differences within and between strains are in reasonable agreement with the directional changes observed in morphological studies (see below).
The present differences between SHRc and WKYc in LV/BW ratios, resistance properties, and estimates of structure were all in accord with previous findings.2 3 5 11 23 27 28 29 30 The greater w/r ratio and narrower lumen in 21-week-old SHRc are thought to be partly due to medial hypertrophy/hyperplasia and partly due to remodeling, ie, rearrangement of wall constituents not involving net growth.31 32 33 In the present study, the vascular changes in SHRc between 21 and 35 weeks (Figs 3⇑ and 5⇑) suggest additional medial hypertrophy with little or no further luminal narrowing at full dilatation (Table 4⇑). The changes coincide with a rise in BP and are probably caused by it. The pattern resembles the medial hypertrophy of large conducting arteries in hypertension (eg, see Reference 3434 ). There was no associated late LV hypertrophy, but this has previously been observed in 40- to 50-week-old SHRc.11 12
BP of SHRen and WKYen was depressed from the start of enalapril therapy. Since the between-strain differences in LV and vascular structural properties are normally present from a very early age,11 12 it is also likely that structural development was depressed by ACE inhibition through much of the treatment period. At 21 weeks, the depression of the LV/BW ratio was somewhat greater in SHRen than in WKYen, but the percent depression of the vascular resistance parameters was the same in each strain, so that the between-strain differences were similar to those in controls (Table 2⇑). In treated rats of each strain, the vascular resistance at the midpoint of the dose-response curve was approximately 25% below the corresponding control value, and the estimated structure (Table 5⇑, Fig 5⇑) suggested a lower w/r ratio and wider r, which is in accord with morphological studies.20 23 35 36 Thybo et al23 have found that ACE inhibition with perindopril produces similar structural changes in the small arteries of most major beds. ACE inhibitors also increase the distensibility of large arteries37 and prevent the deposition of arterial collagen.38 In our experiments, the smaller rise in PP per unit increase in flow in the dilated HQ bed (Table 3⇑, Fig 4⇑) suggests that the increase in distensibility also affected the resistance vasculature.
Between 21 and 35 weeks, late development of the LV/BW ratio was similar in SHRen and WKYen. However, late vascular development differed in the two strains. In SHRen the changes were the same as in SHRc, with a similar associated rise in BP over this time but at a lower level (Fig 1⇑). Thus, the changes in SHRen appear to be unrelated to previous ACE inhibition, and we hypothesize that in both groups of SHR they are directly caused by the rise in BP. By contrast, in WKYen the resistance parameters and associated estimates of structure recovered to a relatively greater extent than in SHRen (Figs 3⇑ and 5⇑). However, the rise in BP between 21 and 35 weeks was considerably smaller, suggesting only a small contribution through the BP-related component to late vascular development. Thus, much of the late development is through a non–BP-dependent mechanism. We hypothesize that the latter acts directly on vascular smooth muscle growth and that the mechanism is mediated at least in part through a product of ACE, eg, Ang II.
It has been difficult to know whether a particular intervention depresses BP because of its primary action on structure or whether structure is depressed because treatment primarily lowers BP. The different patterns of late development after enalapril in the two strains suggest that both mechanisms can be involved. Because attenuation of resistance properties immediately after treatment was stopped was proportionately similar in both strains, we hypothesize that normal vascular development in young SHRc and WKYc involves both BP-related and non–BP-related mechanisms, with the latter increasing growth through one of the products of ACE. Moreover, the close similarity of the changes in resistance properties between 21 and 35 weeks in SHRen and SHRc suggests that after treatment with enalapril, the BP-related mechanism is not depressed. If this also applies to WKYen, the moderate attenuation of late vascular development in this strain must be through the non–BP-related mechanism. The latter is completely suppressed in SHRen. In both strains, expression of the relevant genes is presumably reduced in the older rats compared with their expression in youth.
By contrast, late development of the LV/BW ratio after withdrawal of enalapril was similar in both strains. Hence, it is not possible to say which mechanisms are involved, but we suggest that both BP-dependent and non–BP-dependent mechanisms play a role.
The hypothesis that both direct and indirect mechanisms are involved in structural development is in accord with the well-known adaptive BP-related hypertrophy in cardiac and vascular muscle2 33 and with the direct actions of Ang II on growth in myocardial and vascular muscle cell culture, where it usually acts in conjunction with other growth factors.15 39 The between-strain differences in late development could be due to differences in cellular signaling mechanisms and in related gene expression. An example of a strain-specific difference in responsiveness to a particular growth factor comes from an in vitro study in which a large number of growth factors caused similar degrees of expression of the transforming growth factor-β1 (TGF-β1) gene in both strains, but the autoinduction of this gene by TGF-β1 occurred only in WKY and not in SHR.40 In the heart in an in vivo model of Ang II–mediated LV hypertrophy, stimulation of AT1 receptors led to enhanced expression of the genes regulating these receptors and TGF-β1.41
We do not know whether further late cardiovascular development would have occurred after 35 weeks of age. BP appeared to be leveling off in each strain (Fig 1⇑), and we have assumed that by then the greater part of late development had occurred.
Determinants of BP
During enalapril administration and at 21 weeks, the BP difference between SHRc and SHRen was approximately 2.5 to 3 times as great as that between WKYc and WKYen. Since the potency of the vascular amplifier was similar in treated and untreated SHR and that of the LV amplifier was altered only slightly, the “dilator” stimulus provided by treatment will lower BP and absolute vascular resistance more in SHR than in WKY.42 Assuming that the HQ bed reflects changes in other major beds, structure can account for up to about two thirds of the above BP difference between treated and untreated rats of the two strains, but not for all. Although we did not measure sympathetic activity in the present study, we have previously found that it is greater in SHRc, than in WKYc11 and that during enalapril administration it is reduced,21 which will reinforce the effect of structure on BP reduction in SHR.
Soon after withdrawal of enalapril, the greater rise in BP in SHRen than in WKYen is largely due to the cardiovascular amplifier properties of SHRen, and this will be reinforced by the known return to control of sympathetic activity.21 By 35 weeks, the average vascular cross section of SHRen had recovered only slightly from the value at 21 weeks, which must contribute to the mildness of their hypertension. Similarly, the more pronounced structural recovery of WKYen must be important in the restoration of their BP close to that of WKYc. However, the structural development of WKYen at that time was still well below that of WKYc. Hence, additional homeostatic factors must also contribute to the restoration of their BP.
In the present study, the hypertension developed by SHRen after enalapril was stopped was milder than in any previous report with ACE inhibitors.17 20 21 22 23 35 It is not clear why, after drug withdrawal, more hypertension developed after other ACE inhibitors (mainly captopril and perindopril). Possible reasons include pharmacokinetic differences between drugs and differences in dosage. Our rats were derived from inbred Japanese SHR and WKY, which may have somewhat different characteristics compared with rats from American or European sources.43
However, the hypertension in the present group of SHRen was also milder than in our own previous study.21 In both, the daily dose of enalapril and the source of rats were the same, but there were differences in treatment duration. In the present study, treatment was from 4 to 20 weeks, and the steady-state BP after enalapril was 146 mm Hg. Corresponding BP values from the earlier study were 158 mm Hg (treatment from 4 to 14 weeks), 172 mm Hg (treatment from 4 to 9 weeks), and 176 mm Hg (treatment from 14 to 20 weeks).21 Taking both studies together, one can see an inverse relationship between the duration of preceding treatment and the extent of long-term attenuation of hypertension. This suggests that the capacity of young SHR for cardiovascular development through the non–BP-dependent mechanism is gradually lost with prolonged ACE inhibition, age, or both. Presumably, the non–BP-dependent component is suppressed less completely after shorter periods of ACE inhibition in young SHR.
Long-term attenuation of hypertension after treatment with ACE inhibitors has not been observed in every strain with genetic hypertension. Mulvany et al24 treated Milan hypertensive rats with perindopril for 20 weeks and observed the expected retarded structural development. However, 16 weeks after drug withdrawal, BP had risen to the level of untreated controls, which is similar to the restoration of BP in WKYen. No data on structure were available for the latter time, but assuming that recovery was greater than in the present group of SHRen, long-term suppression after drug withdrawal of the non–BP-dependent mechanism would appear to be specific for particular genetic hypertensive strains.
It is also interesting to compare the long-term effects of enalapril with those following neonatal sympathectomy combined with α-adrenoceptor blockade.12 Treatment over the first 8 weeks of life completely prevents subsequent development of hypertension and structural changes in SHR. This is specific to SHR, because the regimen has little effect on WKY. Although the actions of enalapril in the present study after drug withdrawal were less specific than those of sympathectomy, the long-term therapeutic effect on BP after drug withdrawal was only slightly less effective, although the period of treatment required to bring this about was considerably longer.
The Place of Normotensive Controls
The aim of broadening the experimental design by including more than one rat strain is to provide information about the specificity of treatment.44 In the present study, the effects on vascular structural properties soon after enalapril was stopped were similar in both strains, and this relatively nonspecific action makes it unlikely that ACE plays other than a general role in the pathogenesis of hypertension in SHR. Some between-strain differences occurred subsequently during late development. Together with the earlier study on duration of treatment, the present study provides a model for a preventive strategy in human hypertension. How successful such a strategy will be depends on whether cardiovascular development in human primary hypertension resembles that of SHR or Milan hypertensive rats.
The very proper concerns that have been expressed about relating too readily any differences in traits between SHR and WKY to the genes that affect BP45 46 should not be taken to mean that it is inappropriate or unnecessary to study WKY or other normotensive rats; this would adversely affect the level of physiological analysis. The use of several strains in sorting out the mechanisms affected by particular interventions is likely to sharpen the focus of subsequent detailed genetic experiments.
In young SHR and WKY, ACE normally plays an important role in the development of cardiovascular structure, and we hypothesize that this is mediated through a BP-dependent mechanism and a non–BP-dependent mechanism that acts directly on the myocardium and vascular smooth muscle. The present long-term effects after drug withdrawal in conjunction with our earlier study with enalapril21 suggest that the duration of treatment in young rats determines the extent to which the non–BP-dependent mechanism is attenuated in older rats. Between-strain differences in the capacity of the latter mechanism to respond after drug withdrawal account for the larger gap in BP between treated and control rats in SHR compared with WKY and for the mildness of the hypertension in SHRen.
This work was supported by grants from the National Heart Foundation of Australia, the National Health and Medical Research Council, and the Alfred Hospital Research Fund. We are grateful to Peter Kanellakis for technical assistance with the experiments.
Reprint requests to Prof P.I. Korner, 3, Hunter St, Woolwich, NSW 2110, Australia.
- Received July 15, 1994.
- Revision received September 8, 1994.
- Accepted December 14, 1994.
Folkow B. Physiological aspects of primary hypertension. Physiol Rev. 1982;62:347-504.
Folkow B. Cardiovascular structural adaptation: its role in the initiation and maintenance of primary hypertension. Clin Sci Mol Med. 1978;55:3s-22s.
Korner PI. Causal and homeostatic factors in hypertension. Clin Sci. 1982;63:5s-26s.
Mulvany MJ, Aalkjaer C. Structure and function of small arteries. Physiol Rev. 1990;70:921-962.
Korner PI. The role of the heart in hypertension. In: Robertson JIS, ed. Handbook of Hypertension, Volume I: Clinical Aspects of Essential Hypertension. Amsterdam, the Netherlands: Elsevier; 1983:97-132.
Broughton A, Korner PI. Left ventricular pump function in renal hypertensive dogs with cardiac hypertrophy. Am J Physiol. 1986;251:H1260-H1266.
Korner PI. Cardiac structure and function in animal models and in human hypertension: basic concepts. In: Safar M, Fouad-Tarazi F, eds. The Heart in Hypertension. Dordrecht, the Netherlands: Kluwer Academic Publishers; 1989:145-164.
Korner P, Bobik A, Jennings G, Angus J, Anderson W. Significance of cardiovascular hypertrophy in the development and maintenance of hypertension. J Cardiovasc Pharmacol. 1991;17(suppl 2):S25-S32.
Adams MA, Bobik A, Korner PI. Differential development of vascular and cardiac hypertrophy in genetic hypertension: relation to sympathetic function. Hypertension. 1989;14:191-202.
Korner P, Bobik A, Oddie C, Friberg P. Sympathoadrenal system is critical for structural changes in genetic hypertension. Hypertension. 1993;22:243-252.
Gohlke P, Linz W, Schölkens BA, Kuwer I, Bartenbach S, Schnell A, Unger T. Angiotensin-converting enzyme inhibition improves cardiac function: role of bradykinin. Hypertension. 1994;23:411-418.
Johnston CI. Tissue angiotensin converting enzyme in cardiac and vascular hypertrophy, repair, and remodeling. Hypertension. 1994;23:258-268.
Skidgel R, Erdös E. Biochemistry of angiotensin I-converting enzyme. In: Robertson JIS, Nicholls MG, eds. Textbook of Hypertension. London, England: Gower Medical Publishing; 1993:10.1-10.10.
Guidicelli JF, Freslon JL, Glasson S, Richer C. Captopril and hypertension: development in SHR. Clin Exp Hypertens A. 1980;2:1083-1096.
Richer C, Doussau MP, Guidicelli JF. Prevention du developpement de l’hypertension genetique par le MK421 chez le SHR. Arch Mal Coeur. 1982;75:55-58.
Adams MA, Bobik A, Korner PI. Enalapril can prevent vascular amplifier development in spontaneously hypertensive rat. Hypertension. 1990;16:252-260.
Harrap SB, Van der Merwe WM, Griffin SA, Macpherson F, Lever AF. Brief angiotensin converting enzyme inhibitor treatment in young spontaneously hypertensive rats reduces blood pressure long-term. Hypertension. 1990;16:603-614.
Thybo N, Korsgaard N, Eriksen S, Christensen K, Mulvany M. Dose-dependent effects of perindopril on blood pressure and small-artery structure. Hypertension. 1994;23:659-666.
Snedecor G, Cochran W. Statistical Methods. 8th ed. Ames, Iowa: Iowa State University Press; 1989:503.
Suwa N, Takahashi T. Morphological and Morphometrical Analysis of Circulation in Hypertension and Ischemic Kidney. Munich, Germany: Urban & Schwarzenberg; 1971:132.
Baumbach GL, Heistad DD. Remodeling of cerebral arterioles in chronic hypertension. Hypertension. 1989;13:968-972.
Heagerty A, Aalkjaer C, Bund S, Korsgaard N, Mulvany M. Small artery structure in hypertension: dual processes of remodeling and growth. Hypertension. 1993;21:391-397.
Mulvany M. Resistance vessels in hypertension. In: Swales JD, ed. Textbook of Hypertension. London, England: Blackwell Scientific Publications; 1994:103-119.
Christensen KL, Jespersen LT, Mulvany MJ. Therapeutic regression of resistance vessel structure in spontaneously hypertensive rats does not necessarily inhibit redevelopment of blood pressure on withdrawal of treatment. J Hypertens. 1988;6(suppl 4): S701-S702.
Mulvany MJ. Resistance vessel structure: effects of treatment. J Cardiovasc Pharmacol. 1991;17(suppl 2):S58-S63.
Barenbrock M, Spieker C, Hoeks APG, Zidek W, Rahn K-H. Effect of lisinopril and metoprolol on arterial distensibility. Hypertension. 1994;23(suppl I):I-161-I-163.
Albaladejo P, Bouaziz H, Duriez M, Gohlke P, Levy BI, Safar ME, Benetos A. Angiotensin converting enzyme inhibition prevents the increase in aortic collagen in rats. Hypertension. 1994;23:74-82.
Agrotis A, Saltis J, Bobik A. Transforming growth factor-β1 gene activation and growth of smooth muscle from hypertensive rats. Hypertension. 1994;23:593-599.
Everett A, Tufro-McReddie A, Fisher A, Gomez RA. Angiotensin receptor regulates cardiac hypertrophy and transforming growth factor-β1 expression. Hypertension. 1994;23:587-592.
Wright CE, Angus JA, Korner PI. Vascular amplifier properties in renovascular hypertension in conscious rabbits. Hypertension. 1987;9:122-131.
Nabika T, Nara Y, Ikeda K, Endo J, Yamori Y. Genetic heterogeneity of the spontaneously hypertensive rat. Hypertension. 1991;18:12-16.
Fisher RA. The Design of Experiments. 4th ed. Edinburgh, Scotland: Oliver and Boyd; 1947:240.
Rapp J. A paradigm for identification of primary genetic causes of hypertension in rats. Hypertension. 1983;5(suppl I):I-198-I-203.
Rapp JP. Use and misuse of control strains for genetically hypertensive rats. Hypertension. 1987;10:7-10.