Effect of ACE inhibition on pressor, renal vascular, and adrenal responses to infusion of angiotensin I in normal subjects eating a low-salt diet

Abstract

To examine the influence of angiotensin-converting enzyme (ACE) on pressor, renal vascular, and adrenal responses during angiotensin I (Ang I) infusion, we studied 10 normotensive, healthy men. Each was in balance with a 10-mEq sodium, 100-mEq potassium intake and was studied before and during ACE inhibition with enalapril. Ang I (3, 10, and 30 ng/kg/min) was infused in each subject. Then ACE inhibition was instituted with enalapril for 3 days, which induced the anticipated fall in blood pressure, plasma Ang II, and aldosterone concentration, and rise in renal plasma flow. During ACE inhibition only the 30-ng/kg/min Ang I dose raised plasma Ang II levels. There was a spectrum, however, in the end-organ response to Ang I during ACE inhibition. Responses of plasma aldosterone concentration and blood pressure were in excellent accord with the reduction in Ang II formation. On the other hand, responses of the renal blood supply were substantially less inhibited than anticipated. Under the conditions of this study, ACE inhibition led to nonuniform changes in the response to exogenous Ang I, suggesting intrarenal conversion of Ang I to Ang II.

ACE, Ang II, salt

The contribution of angiotensin-converting enzyme (ACE) inhibition to maintenance of renal function in processes ranging from heart failure to nephropathy because of diabetes mellitus and other forms of nephropathy is beyond debate.

Despite the clear success of this approach to treatment, a number of determinants of the response to ACE inhibition remain unclear.^1,,,,,,–8 Some of the factors that have been cited include differences in ACE inhibitor bioavailability, bioactivation of the drug, binding properties to the enzyme, and variation in tissue penetration of the drug.¹ Evidence supporting differential tissue penetration has been developed.^2,6 The magnitude of the reactive activation of the renin system in response to a fall in renal tissue angiotensin II (Ang II) must also be a determinant of the ultimate response.³ Moreover, somatic ACE contains two active centers, the N and C domains,⁷ with differing binding to both Ang I and to converting enzyme inhibitors.⁷ Finally, there is evidence for non–ACE-dependent Ang II generation in the human kidney, at least under conditions in which the renin system has been activated by restriction of salt intake.⁹ Much of the debate engendered by these considerations has involved the tissue renin-angiotensin system (RAS), as ACE inhibition in the circulating compartment has been much more straightforward.^5,6

A different perspective regarding the role for intrarenal tissue renin systems was reported by Vos et al.¹⁰ The study involved intravenous infusion of Ang I in healthy humans under conditions designed to minimize endogenous activity of the RAS. These conditions included ingestion of a very high salt intake, 340 mmol daily, combined with ACE inhibition induced by enalapril administration. Their study suggested that the renal vascular response to intravenous Ang I infusion could be accounted for entirely by Ang II delivery to the kidney. In this study we present rather different findings under conditions in which the renin system has been activated by restriction of salt intake, as in our earlier studies.⁹

Materials and methods

The study was performed in 10 normal men, who ranged in age from 18 to 55 years, and in body weight from 55 to 92 kg. The protocol was approved by the Human Subjects Committee of the Brigham and Women's Hospital. Written informed consent for the procedures was obtained after full description of the protocol. Each subject was admitted to the General Clinical Research Center of the Brigham and Women's Hospital, and given a constant isocaloric diet through the entire protocol. All studies were performed when external balance had been achieved on a daily intake of 10 mEq sodium, 100 mEq potassium, and 2500 mL of fluid. Twenty-four–hour urine samples were collected daily and analyzed for sodium, potassium, and creatinine. When urinary sodium excretion matched intake, usually in 4 to 6 days, the following protocols were performed.

Renal clearance studies

Each normal volunteer participated in a para-aminohippurate (PAH) (Merck Sharp & Dohme Research Laboratories, West Point, PA) clearance study after achieving metabolic balance eating the low-salt diet, as detailed earlier.¹¹ An intravenous catheter was placed in each of the subject's arms, one for infusion and the other for blood sampling. All subjects studied were supine and fasted for at least 8 h. A control blood sample was obtained and then a loading dose of PAH (8 mg/kg) was given. A constant infusion of PAH was initiated immediately at a rate of 12 mg/min using an IMED pump (IMED Corporation, San Diego, CA). This infusion rate achieved a plasma PAH concentration in the middle of the range in which tubular secretion dominates excretion.¹² At this plasma level of PAH, clearance is independent of plasma concentration and, when corrected for an individual surface area, represents about 90% of effective renal plasma flow (RPF). Basal PAH clearance was calculated from the plasma level and infusion rate. Plasma samples were obtained 50 and 60 min after the start of the constant infusion, when a steady state had been achieved.

Angiotensin i infusion

After completing the basal PAH clearance assessment, the subjects received an infusion of Ang I (Clinalfa AG, Laufelfingen, Switzerland) in successive doses of 3.0, 10, and 30 ng/kg/min for 45 min each, using an electronic infusion pump (Harvard Apparatus Co., Inc., Millis, MA). The constant infusion of PAH continued throughout the Ang I infusion to assess changes in RPF with increasing Ang I doses. Blood pressures were recorded every 10 min during the basal clearance studies and then every 2 min during the Ang I infusion with an indirect recording sphygmamanometer (Arteriosonde, Roche Diagnostics Div., Hoffman-La Roche Inc., Nutley, NJ), with the cuff positioned over the brachial artery of the arm containing the sampling catheter. Blood samples were obtained at the end of the basal PAH clearance and after each incremental infusion dose of Ang I and analyzed for PAH, aldosterone, cortisol, plasma renin activity (PRA), and Ang II. Serum sodium and potassium were measured at the initiation and termination of the Ang I infusion. Serum creatinine was measured at the initiation of the clearance study.

Treatment with enalapril

After completion of the initial determination of PAH clearance, and of the response to Ang I, treatment was initiated with enalapril. The doses employed were 2.5 mg twice a day on the first day, 5 mg twice a day on the second, and 10 mg twice a day on the third. The next morning, the clearance studies were then repeated, about 4 h after a third 10-mg enalapril dose was administered at 4 AM, so that the two clearance studies were performed at the same time of day.

Laboratory procedures

All blood samples were collected on ice, spun immediately, and the plasma was separated and frozen until the time of assay. Serum and urine sodium and potassium levels were measured by flame photometry using lithium as an internal standard. Serum creatinine and PAH¹² were measured by an autoanalyzer technique. Ang II, PRA, and aldosterone were assayed by radioimmunoassay (RIA) techniques that have been described.¹³ The antibody used for measurement of Ang II has less than 0.1% affinity for Ang I.

Statistics

Group means have been presented with the standard error of the mean as the index of dispersion. Statistical probability was assessed with the paired or unpaired data t test for normally distributed data, linear regression, Dunnett's t test for multiple comparisons from a single baseline value, a form of ANOVA. The Fisher Exact Test and χ² Test were employed to assess nonhomogenously distributed data. We rejected the null hypothesis when P achieved an alpha level of < .05.

Results

Restriction of sodium intake resulted in the anticipated fall in sodium excretion and activation of the renin-angiotensin-aldosterone system (RAAS) in the subjects (Table 1).

Table 1.

Baseline values when balance achieved with a low-salt intake before angi infusion

	Baseline	Enalapril
N	10	10
24-h urine Na (mEq)	12 ± 4	10 ± 3
24-h urine K (mEq)	88 ± 7	93 ± 4
Serum Na (mEq/L)	137 ± 1	139 ± 2
Serum K (mEq/L)	4.4 ± 0.1	4.4 ± 0.1
PRA (ng Ang I/mL/hr)	4.4 ± 0.5	21.0 ± 3.3†
Plasma Ang II (pg/mL)	45 ± 5	29.0 ± 1.4**
Plasma aldosterone (ng/dL)	29 ± 1.4	13.2 ± 2.1**
PAH clearance (mL/min/1.83 m2)	593 ± 18	646 ± 11*
Systolic BP (mm Hg)	109 ± 5	99 ± 5*
Diastolic BP (mm Hg)	72 ± 2	67 ± 3*
Body weight (kg)	72.4 ± 3.2	72.2 ± 3.6

*

P < .05;

**

P < .001.

Ang I, angiotensin I; ang II, angiotensin II; PRA, plasma renin activity; PAH, para-aminohippurate; BP, blood pressure.

Infusion of Ang I before enalapril administration resulted in a dose-related and robust increase in plasma Ang II concentration, fall in RPF, increase in plasma aldosterone concentration, and pressor response (Figure 1).

Basal responses to angiotensin i (ang i) infusion of plasma angiotensin ii (ang ii) concentration, renal plasma flow (RPF), and plasma aldosterone concentration and diastolic blood pressure.

Figure 1.

Treatment with enalapril induced a sharp increase in baseline PRA (P < .001) and RPF (P < .05), and fall in plasma Ang II concentration (P < .01) and blood pressure (P < .05) (Table 1). Treatment with enalapril also reduced Ang II generation during Ang I infusion, as anticipated (Table 2,(Figure 2). Indeed, only the 30-ng/kg/min Ang I dose resulted in a significant increase in plasma Ang II concentration (P < .01; Table 2,Figure 2). In accord, the rise in plasma aldosterone concentration and diastolic blood pressure achieved statistical significance only with a 30-ng/kg/min Ang I dose during enalapril treatment (P < 0.001; Table 2). The renal vascular response to Ang I, on the other hand, was better sustained during ACE inhibition Figure 1). A significant fall in RPF occurred at the lowest Ang I dose, 3 ng/kg/min (P < .025), with a dose-related fall thereafter.

Table 2.

Responses to angiotensin i (angi) infusion before and during ace inhibition

	Ang I (3 ng/kg/min)	Ang I (10 ng/kg/min)	Ang I (30 ng/kg/min)
Change in plasma ang II
Basal	22 ± 4	68 ± 12	99 ± 16
Enalapril	0 ± 5	0 ± 9	31 ± 8
Change in RPF
Basal	−57 ± 12	−112 ± 12	−162 ± 3
Enalapril	−25 ± 9	−39 ± 9	−110 ± 16
Change in plasma aldosterone (ng/dL)
Basal	7.7 ± 2.8	20.1 ± 3.3	58 ± 10.1
Enalapril	1.0 ± 0.9	4.9 ± 4.0	20 ± 9.2
Change in diastolic BP (mm Hg)
Basal	2 ± 1	7 ± 2	18 ± 3
Enalapril	0 ± 1	1 ± 0.6	7 ± 2

ACE, angiotensin-converting enzyme; RPF, renal plasma flow.

Percent inhibition of responses to angiotensin I (Ang I) induced by enalapril. Angiotensin-converting enzyme (ACE) inhibition blocked an increase in plasma angiotensin II (Ang II) concentration completely at the two lowest Ang I doses employed and inhibited the response substantially at the high dose. The pressor and adrenal responses were in approximate accord with the reduction in Ang II generation. The responses of the renal blood supply, on the other hand, were blocked substantially less than anticipated from the reduction in Ang II concentration in plasma. These findings support a quantitatively important non-ACE pathway for Ang II generation in the kidney.

Figure 2.

To facilitate comparison, the percent inhibition of responses to Ang I induced by enalapril are presented for plasma Ang II concentration, plasma aldosterone concentration, blood pressure, and RPF in Figure 2. The percent inhibition of the role in plasma Ang II concentration fell with increasing Ang I infusion rate, from 100% ± 14.5% at 3 ng/kg/min to 68% ± 10.7% at the 30-ng/kg/min Ang I dose. The inhibition of the rise in plasma aldosterone and blood pressure induced by Ang I paralleled the inhibition of Ang II formation. The inhibition of renal vascular responses to Ang I induced by enalapril was strikingly less, ranging from 56% ± 16% at the 3-ng/kg/min Ang I infusion rate, to 32% ± 9.5% at the highest Ang I infusion rate. At each Ang I infusion rate the percent inhibition of renal vascular responses to Ang I was significantly less than the inhibition of Ang II formation (P < .01).

Discussion

Angiotensin-converting enzyme inhibition resulted in a series of consequences. As anticipated, conversion of Ang I to Ang II was sharply blunted by enalapril. As a result, there was a fall in basal blood pressure, plasma Ang II, and aldosterone concentration, and a rise in RPF. Indeed, during Ang I infusion after enalapril a significant rise in plasma Ang II concentration did not occur at doses of 3 and 10 ng/kg/min, and the increase in plasma Ang II concentration at 30 ng Ang I/kg/min during ACE inhibition was only 27% of the level achieved during the baseline study. The plasma aldosterone and pressor responses to Ang I infusion were in excellent accord. Neither the plasma Ang II level nor plasma aldosterone concentration rose significantly until the 30-ng/kg/min Ang I dose was infused, and the adrenal response was sharply reduced at that dose. Not in accord, on the other hand, was the response of the renal blood supply to Ang I during ACE inhibition. Although the renal response was clearly blunted after enalapril, a significant fall in RPF occurred with the 3-ng/kg/min Ang I dose, and larger falls occurred with increasing doses. The renal vascular response at 30 ng/kg/min was 70% of the baseline response—clearly differing from all of the other systems.

Several possible explanations could account for the difference in the influence of the ACE inhibitor on renal vascular response to Ang I. One possibility is that Ang I exerts an influence on the renal blood supply that does not depend on conversion to Ang II. Although at one time evidence was presented to suggest that a number of smooth muscle preparations would respond to Ang I,¹⁴ the subsequent demonstration of converting enzyme in the tissues¹⁵ made that conclusion unlikely. A second possibility is that conversion of Ang I to Ang II occurs in the kidney, and that renal ACE might be more resistant to ACE inhibition than is systemic conversion. Evidence for intrarenal conversion of Ang I to Ang II exists,¹⁶ but the relation between systemic ACE inhibition and ACE inhibition induced in the kidney is complex; generally, however, ACE inhibition is less effective in the kidney.^5,6 The rise in RPF that followed ACE inhibition certainly suggests a substantial influence on the kidney. The third possibility is that the ACE inhibitor-induced reduction in Ang II formation was offset, in part, by non–ACE-dependent Ang II generation.⁹

This interpretation of our findings is not in accord with the study performed by Vos et al¹⁰ in healthy humans. They infused Ang I intravenously and came to the conclusion that the renal vascular response reflected primarily Ang II delivery to the kidney. Their findings were in accord with earlier observations of the kinetics of tracer-labeled Ang I conversion to Ang II, which suggested minimal conversion of Ang I delivered to the kidney via the renal blood supply.¹⁷ A similar conclusion was reached by Krekels et al from their comparisons of renal, adrenal, and pressor responses to Ang I and to Ang II infused intravenously.¹⁸ They found no difference in the renal vascular response to Ang I and to Ang II infused intravenously in 0.3 and 1.0 pmol/kg/min concentrations, but the response to Ang II substantially exceeded the responses to Ang I at the 3-pmol/kg/min dose. The adrenal and pressor responses to the two hormones, on the other hand, were essentially identical. Taken together, these studies suggest that conversion of Ang I to Ang II when Ang I is delivered to the kidney via the renal blood supply is limited. In this study, on the other hand, at all doses of Ang I infused during ACE inhibition, the renal vascular response substantially exceeded that anticipated from the circulating Ang II levels. One possible explanation for the difference involves the fact that our studies were performed when the endogenous renin system was activated by restriction of salt intake to a very low level, 10 mEq/day. The possibility exists that the conversion of intrarenal Ang I Ang II becomes quantitatively more important when the renin system is activated. Although no direct evidence favoring this possibility has been reported, certainly gene expression of other elements of the renin cascade has been found to vary with the state of the renin system.¹⁵

This study has a number of limitations. All subjects were studied at baseline first and then during exposure to ACE inhibition. Although this is a common practice, a randomized exposure to ACE inhibitor and placebo first would have made a more ideal study design, although rather more costly. We have shown that serial renal vascular responses to Ang II are remarkably reproducible.¹⁹ Although we measured immunoreactive Ang II rather than authentic Ang II after HPLC separation of fragments, the method we employed was adequate to show the directional effect of ACE inhibition and a very good correlation between Ang I infusion rate and resultant Ang II concentration. The likelihood that a distortion based on measurement of immunoreactive Ang II contributed to our findings is exceedingly unlikely in view of the close parallelism between the resultant plasma Ang II concentration during ACE inhibition and the response of plasma aldosterone and blood pressure (Figure 2). With only 0.1% cross-reactivity between the antiserum used for Ang II measurement and Ang I, it is unlikely that this provided a source of error. Direct measurement of plasma Ang I concentration during the infusion studies would have provided a better measure of the conversion rate, but this was not a goal of the study. Although it would have been of interest to assess the response to graded doses of enalapril, we were fortunate in that the enalapril dose employed provided the index required, which was complete blockade at the lower Ang I infusion rates, and partial blockade thereafter.

Longstanding interest in the possibility that tissue formation of Ang II contributed to tissue responses was addressed in a recent study, which suggested that the renal response to Ang I depended entirely on systemic conversion of Ang I to Ang II, and delivery to the kidney.¹⁰ That study was designed to minimize activity of the endogenous renin-angiotensin system, both by means of a very high salt intake and prior ACE inhibition.¹⁰ In this study, in which the endogenous renin-angiotensin system was activated by a low-salt diet, unambiguous evidence for renal vasoconstriction induced by intrarenal formation of Ang II was presented. Whether or not the formation of Ang II from Ang I involved ACE-dependent or -independent pathways must be the subject of further research.

Acknowledgments

We thank Ms. Charlene Malarick, RN, Diane Passan, MT, Diana Capone, and the nurses and dieticians of the General Clinical Center for their assistance in various aspects of this study. The study was performed at the General Clinical Research Center.