Effect of Bilateral Nephrectomy on Active Renin, Angiotensinogen, and Renin Glycoforms in Plasma and Myocardium
Abstract In an attempt to clarify the relationship of the circulating and myocardial renin-angiotensin systems, active renin concentration, its constituent major glycoforms (active renin glycoforms I through V), and angiotensinogen were measured in plasma and left ventricular homogenates from sodium-depleted rats under control conditions or 2 minutes, 3 hours, 6 hours, and 48 hours after bilateral nephrectomy (BNX). Control myocardial renin concentration was 1.4±0.1 ng angiotensin I (Ang I) per gram myocardium per hour and plasma renin concentration was 6.7±1.1 ng Ang I per milliliter plasma per hour. Control myocardial angiotensinogen was 0.042±0.004 μmol/kg myocardium and plasma angiotensinogen was 1.5 μmol/L plasma. Two minutes after BNX and corresponding stimulation of renin secretion by anesthesia and surgery, plasma renin concentration was increased disproportionately compared with myocardial renin. Three, 6, and 48 hours after BNX, renin decay occurred significantly faster from the plasma than from the myocardium. Forty-eight hours after BNX, myocardial renin concentrations had fallen to 15% of control values, while myocardial angiotensinogen concentrations had increased 12-fold and plasma angiotensinogen concentrations had increased by only 3.5-fold. Myocardial renin glycoform proportions were identical in myocardial homogenates and plasma in control animals. At 6 hours BNX, the proportions of plasma active renin glycoforms I+II fell, while those in the myocardium significantly increased. We conclude that in control rats, active renin and active renin glycoforms are distributed as if in diffusion equilibrium between plasma and the myocardial interstitial space. After BNX, myocardial renin concentration falls dramatically, suggesting that most cardiac renin is derived from plasma renin of renal origin. After BNX, renin glycoforms I+II are preferentially cleared from the plasma but preferentially retained by the myocardium. Control myocardial angiotensinogen concentrations are too low to result from simple diffusion equilibrium between plasma and the myocardial interstitium.
Many RAS components have been found in mammalian myocardial tissues.1 These observations have suggested a local myocardial RAS that might act in concert with the circulating RAS. However, the relative importance of the myocardial RAS, the exact source of the myocardial RAS components, and the interplay between the circulating and local myocardial RAS remains controversial.1 2 3 4 Blockade of the RAS probably offers the greatest cardioprotective efficacy for pharmacological treatment of hypertension and heart failure.5 These cardioprotective effects of RAS blockade may operate through multiple physiological mechanisms that cause reductions of left ventricular preload, afterload, hypertrophy, sympathetic stimulation, and the ratio of myocardial oxygen demand to supply.6 Renin has been reported to be an independent risk factor for myocardial infarction in treated patients with essential hypertension.7
Therefore, the first purpose of this study was to assess myocardial and plasma renin and angiotensinogen concentrations under control and anephric conditions. BNX causes contrasting changes in plasma renin and plasma angiotensinogen concentrations, which should be mirrored within the myocardium if plasma is the major source of these myocardial components.
Active rat renin consists of five major glycoforms (I through V), which can be separated by high-resolution isoelectric focusing (renin glycoform profile). Plasma renin activity is equal to the sum of the activities of multiple circulating glycoforms. Renin glycoforms I and II possess high-mannose-type oligosaccharides and relatively small net negative charge.8 In addition, glycoforms I and II represent the predominant storage form of active renin in the kidney, are preferentially secreted by the kidney after acute stimulation, have relatively short plasma half-lives, and have very low vascular wall permeability compared with the other renin glycoforms.9 10 11 12 13 14
Thus, the second purpose of this study was to examine the hypothesis that active renin charge or oligosaccharide heterogeneity might confer a preferential distribution of specific renin glycoforms within the myocardium. Nephrectomy intervention provides an experimental opportunity to assess whether plasma is the major source of myocardial renin glycoforms and whether the myocardial permeability of renin glycoforms I and II is low compared with other renin glycoforms.
Male Sprague-Dawley rats (Harlan Inc, Madison, Wis) weighing 265 to 300 g were fed low-sodium (0.04%) chow (Tekland) for 3 weeks, with free access to food and tap water. Sodium depletion was used to increase plasma and myocardial renin levels so that after BNX, renin assays would still be above the lower limits of detection. In some experiments, normal-salt chow (0.4%) was used to compare renin levels and glycoform proportions to the low-salt animals. The concentration and isoelectric form profile of active renin and the concentration of angiotensinogen were measured both in plasma and the corresponding left ventricular homogenates under control conditions and after BNX. Control rats (n=7) were permitted to rest for 30 to 60 minutes before being rendered unconscious by CO2 inhalation, followed by decapitation and sample collection. BNX rats were anesthetized with pentobarbital (50 mg/kg IP) and both kidneys removed via a midline incision. Two minutes (n=6), 3 hours (n=7), 6 hours (n=7), or 48 hours (n=4) after BNX, anephric rats were rendered unconscious by CO2 inhalation (except for 2-minute BNX rats still under pentobarbital anesthesia), followed by weighing, decapitation, and sample collection. The procedures followed were in accordance with institutional guidelines and the guidelines of the American Association for the Accreditation of Laboratory Animals.
Approximately 7 mL of trunk blood was collected into tubes containing 200 μL of 5% EDTA. The blood was mixed, stored on ice, and centrifuged at 2000g for 15 minutes at 4°C. Separated plasma was snap frozen in liquid N2 and stored at −25°C for subsequent determination of plasma active renin, angiotensinogen, and renin glycoforms. Hearts were quickly excised, followed by retrograde perfusion of the aorta with 10 mL ice-cold heparinized saline and removal of pericardial tissue, atria, great vessels, and right ventricle. Both sides of the left ventricular wall were then blotted, weighed, snap frozen in liquid N2, and stored at −25°C. Total left ventricle preparation time was approximately 3 minutes.
Left ventricles were homogenized at 0°C in a hand-held 7-mL Kontes glass homogenizer at a ratio of 1 mg tissue to 5 μL PIB containing serine-, metallo-, and thiol-protease inhibitors and 0.15 mol/L sodium phosphate buffer (pH 7.5) containing EDTA (15 mmol/L), 8-hydroxyquinoline (2 mmol/L), sodium tetrathionate (10 mmol/L), benzamidine (20 mmol/L), N-ethylmaleimide (10 mmol/L), AEBSF hydrochloride (PEFABLOC, 1 mmol/L), leupeptin (10 μmol/L), aprotinin (450 μmol/L), and 1.0% BSA. Homogenates were split into three aliquots and stored at −25°C. Each set of three myocardial aliquots was thawed and centrifuged at 14 000g for 4 minutes at 4°C, and the supernatants were assayed for active renin, active renin glycoforms, or angiotensinogen.
Plasma and Myocardial Renin Measurement
Plasma was diluted between 7-fold and 30-fold in PIB to yield renin concentrations similar to myocardial homogenates, lesser dilutions being required for BNX plasma samples. Both plasma and homogenates were therefore diluted in the same solvent (PIB) and assayed identically as follows. Duplicate assay tubes contained 80 μL of diluted plasma, myocardial homogenate, or PIB (sample blanks), which was combined with 150 μL additional PIB and 75 μL 48-hour BNX unfractionated rat plasma (angiotensinogen source). In general, PIB represented at least 70% of the assay volume, so that all proteolytic inhibitors were well above their effective working minimum concentration. Two 100-μL aliquots were incubated at 0°C and another two 100-μL aliquots incubated at 37°C, both for 18 to 24 hours. Detectable average Ang I in the 0°C tubes (endogenous immunoreactive Ang I in sample and substrate source) was subtracted from the 37°C tube average to calculate generated Ang I, although the 0°C-tube Ang I concentrations were often undetectable in relatively high renin plasma and corresponding myocardial homogenate samples (see below for corresponding isoelectric focusing background subtraction). The corresponding difference in the sample blank tubes was also subtracted from the generated Ang I value to correct for endogenous renin activity in the substrate source. Sample blanks ranged from 3% to 30% of the generated Ang I, with the exception of the 48-hour BNX plasma samples, whose sample blank was nearly 75% of the generated Ang I. During the 37°C incubation, Ang I was generated in direct proportion to the sample renin concentration. Only 4% or less of the total available angiotensinogen was converted to Ang I, and linear generation of Ang I over time was verified throughout the study.15 16 Renin concentration was expressed as nanograms Ang I per milliliter plasma or gram myocardium per hour incubation.
Ang I was assayed by RIA, using a modified DuPont Ang I RIA kit. Major modifications included preparation of Ang I standards with the same solution contents as samples. RIA assay data were linearized with a log-logit transformation, resulting in a correlation coefficient of −.99 or better. The average labeled Ang I bound 60% to the primary antibody without competition. The average percent of error for predicting the standard concentration from the transformed curve was less than 6%. Ang I recovery from spiked samples was a function of the absolute Ang I concentration. In general, recoveries were between 94% and 106% at Ang I concentrations between 0.1 and 5 ng/mL, but above 5 ng/mL, recovery steadily decreased. As a result, all assays were kept between 0.1 and 5.0 ng Ang I per milliliter.
Angiotensinogen concentrations from left ventricular myocardial homogenate supernatants and plasma were measured by adding a large excess of exogenous porcine renin and subsequently converting all sample angiotensinogen to Ang I (verified by Ang I generation plateau). Ang I concentrations were then determined by RIA, and corresponding angiotensinogen concentrations were calculated on the basis of a 1:1 molar relationship between Ang I generation and angiotensinogen depletion. Plasma was diluted 70-fold in PIB to yield comparable angiotensinogen concentrations to myocardial homogenates. Both plasma and homogenates were therefore diluted in the same solvent (PIB) and assayed identically as follows. Porcine renin (7 mU, Sigma Chemical Company) in 325 μL PIB was added to 25 μL diluted plasma or myocardial homogenate at 0°C and then incubated in duplicate at 37°C for 20 minutes. Another set of duplicates was incubated for 40 minutes. Care was taken to ensure that samples were not warmed before the incubation step. Appropriate combinations of porcine renin and PIB, plasma and PIB, homogenates and PIB, or PIB alone yielded no detectable Ang I. The average percent difference between the Ang I concentration determined from the 20- and 40-minute time points was −0.36% and −4.04% for plasma and myocardial homogenates, respectively, indicating that an Ang I generation plateau had been reached by 20 minutes and that Ang I is relatively stable in the assay. The assay was repeated in selected samples with 2 mU or 70 mU of renin addition, and resulting angiotensinogen determinations at 20 and 40 minutes’ incubation were within 9.3% of the 7-mU renin addition, indicating that excess renin was being employed in the angiotensinogen assay. Furthermore, spiking ventricular homogenates with plasma yielded predictable angiotensinogen determinations equal to the sum of the homogenate plus plasma values determined separately. Addition of enalaprilat to the PIB yielded no differences in the amount of angiotensinogen measured. The minimal myocardial angiotensinogen detection limit was approximately equal to 15% of the control myocardial angiotensinogen values.
Shallow-Gradient Isoelectric Focusing
Rat active renin was resolved into five major glycoforms (I through V) by using shallow-gradient isoelectric focusing.8 Plasma (600 μL) was incubated with 15 mg SiO2 at 20°C for 30 minutes followed by centrifugation at 100 000g at 4°C for 30 minutes. Aliquots (100 to 200 μL) of the plasma supernatants were applied to the focusing gels. Ventricular homogenate (1200 μL) was incubated with 30 mg SiO2 at 20°C for 30 minutes followed by centrifugation at 100 000g at 4°C for 30 minutes. The homogenate supernatant was concentrated fivefold by dialysis (4000 to 6000 molecular weight cutoff) against solid polyethylene glycol (molecular weight 15 000 to 20 000) and again centrifuged at 100 000g at 4°C for 30 minutes. Samples (200 μL) of the homogenate supernatants were applied to focusing gels. Focusing was carried out as described previously.8 14
After the run, each gel was frozen and serially sliced into approximately 60 gel segments. Each gel segment was eluted in an individual polystyrene tube with 125 μL PIB at 4°C. Gel segments initially swelled and then reached diffusion equilibrium approximately 18 to 24 hours later. The gel segments were then carefully removed, leaving behind approximately 90 μL of eluant (with renin present, if the enzyme had focused in the corresponding gel segment).
Renin Glycoform Measurement
Each 90 μL of eluant was then combined with 30 μL of plasma from 48-hour bilaterally nephrectomized rats as an angiotensinogen source. The total 120 μL was then incubated at 37°C for 1 to 24 hours; longer incubation times were necessary for samples with relatively little renin activity. The Ang I generated in the entire 120-μL incubation mixture (assay tube) was then measured by RIA. Yields for this system are very close to 100%, assuming diffusion equilibrium between gel slice and eluant.8 16 Figures showing the resulting renin glycoforms separated by isoelectric focusing are raw data expressed as nanograms Ang I per total incubation time per milliliter assay tube.
Determination of Renin Glycoform Proportions
Zero time Ang I immunoreactivity and residual renin activity in the substrate represent background error subtracted from all gel segments containing renin activity to calculate glycoform proportions. This background error correction was determined from gel slices that clearly contained no sample renin activity due to their correspondingly high or low isoelectric points. The background error is equivalent to the sample blanks for the plasma and myocardial renin assays. Subtraction of background and calculation of the five major renin glycoform proportions resolved in the gels was performed as previously reported.8 16 Briefly, the amount of net renin activity (after background subtraction) in each of the five renin activity peaks (renin forms I throughV) was measured by summing the renin activity in each gel segment corresponding to an individual renin activity peak. Identification of each gel peak as a specific renin form was accomplished by identification of the corresponding isoelectric point of the gel segment from identical gels run in parallel and verification by observation of the relative position of a peak in relation to other peaks in the gel.
Focusing gel renin activity was then calculated by expressing the amount of renin activity (nanograms Ang I generated per total incubation time per milliliter assay tube) in each of five glycoform peaks as a percent of the total net renin activity recovered from the focusing gel. The activity of each of the five renin glycoforms was then expressed as a proportion (percentage) of the total net renin activity recovered from the gel.
Comparisons between plasma and myocardial concentrations of active renin and angiotensinogen and between low-salt and normal-salt rats were performed by using unpaired Student’s t tests. Renin glycoform comparisons were made by summing the proportions of the two least negatively charged renin glycoforms (highest isoelectric points, renin glycoforms I and II) and comparing the sum of renin glycoform proportions I+II in plasma and myocardium before and after BNX. Combining renin glycoform I and II proportions allowed assessment of statistical differences without reference to specific renin forms that are a function of the isoelectric pH gradient12 and also allowed direct analysis of the glycoforms that are known to participate in acute plasma concentration changes after stimulation or inhibition of renin secretion. Differences in proportions of renin forms were assessed by Student’s t test after arcsine transformation of proportions to equalize variances between groups. Statistical significance was assumed at P≤.05 (two-tailed).
Under control conditions (resting rats rendered unconscious by CO2 inhalation, followed by decapitation), plasma renin concentration was 6.7±1.1 ng Ang I per milliliter plasma per hour, while the corresponding myocardial renin concentration was 1.4±0.1 ng Ang I per gram myocardium per hour (Fig 1⇓). Approximately 2 minutes after BNX, and with the animals still under pentobarbital anesthesia, plasma renin increased almost 3.3-fold, to 22±6.6 ng Ang I per milliliter plasma per hour (P=.03), while myocardial renin increased 1.7-fold, to 2.4±0.3 ng Ang I per gram myocardium per hour (P=.002), as shown in Fig 1⇓ (0 hours). Subsequent plasma and myocardial renin measurements at 3, 6, and 48 hours post-BNX revealed that both plasma and myocardial renin concentration fell continuously, although the proportional changes in myocardial renin values were always less than plasma renin values (Fig 1⇓). Compared with control values, 48-hour BNX plasma and myocardial renin concentrations fell 95% and 85%, respectively, and both values were significantly different from controls (P<.001).
Control myocardial homogenates were prepared from a separate set of five low-salt rats and assayed as described above and also after an additional centrifugation at 100 000g for 1 hour at 4°C to pellet all myocardial membranes. This latter high-speed centrifugation resulted in pelleting of 12±3.8% of homogenate renin. However, after washing the pellet in PIB and again centrifuging at 100 000g for 1 hour at 4°C, all original pellet renin activity was present in the supernatant and none could be measured in the recentrifuged pellet with or without the presence of 1% Triton X-100.
Under control conditions, plasma angiotensinogen was 1.5±0.04 μmol/L, while the corresponding myocardial angiotensinogen concentration was only 0.042±0.004 μmol/kg myocardium (Fig 2⇓). Approximately 2 minutes after BNX and with the animals still under pentobarbital anesthesia, plasma and myocardial angiotensinogen concentrations were not different (Fig 2⇓, 0 hours) from control values. Subsequent plasma and myocardial angiotensinogen measurements at 6 and 48 hours post-BNX revealed that both plasma and myocardial angiotensinogen concentrations increased. Compared with control concentrations, 48-hour BNX plasma and myocardial angiotensinogen levels increased approximately 4-fold and 12-fold, to 5.5±0.2 μmol/L plasma and 0.49±0.08 μmol/kg myocardium, respectively. Both increases were statistically significant (P<.0002).
Assuming all myocardial renin and angiotensinogen measured in this study was confined to an interstitial space of 20 mL/100 g of left ventricle with a density of 1 mL/g (see “Discussion”), the myocardial interstitial Ang I generation rate derived from the control renin and angiotensinogen myocardial concentrations and in vitro incubation was 2.9 ng Ang I per milliliter interstitial space per hour. The corresponding control plasma Ang I generation rate was 16.0 ng Ang I per plasma per hour.
Control myocardial renin was 24.1±2% of plasma renin. This distribution percentage was not significantly different from experiments done with rats fed a normal (0.4% sodium)-salt diet (Table⇓). Control myocardial angiotensinogen was 2.7±0.25% of plasma angiotensinogen values and was significantly different from normal-salt rats, although both angiotensinogen distribution percentages were quite low relative to the distribution percentages for renin (Table⇓).
Five major active renin glycoforms were found in both plasma and the myocardium (renin glycoforms I through V, with corresponding average isoelectric points of 5.7, 5.4, 5.2, 5.0, and 4.8; Fig 3⇓). Under control conditions, renin glycoforms I+II represented 45.2±3.0% of plasma renin and 49.9±4.2% of myocardial renin and were not significantly different (Figs 3⇓ and 4⇓). In additional experiments with normal (0.4% sodium)-salt rats (Table⇑), plasma glycoform proportions I+II were 49.7±0.91% (n=4) and were not significantly different from the low-salt rats. Corresponding normal-salt myocardial renin glycoform proportions I+II were 47.4±3.51% (n=4) and were not significantly different from low-salt rats. Normal-salt rats did display significantly lower plasma renin and angiotensinogen concentrations, as well as a lower myocardial renin concentration (Table⇑).
Six hours after BNX, the proportions of plasma active renin glycoforms I+II fell significantly to 24.7±2.7% (Fig 4⇑), consistent with preferential hepatic extraction of these circulating glycoforms, as reported previously. However, the proportions of active renin glycoforms I+II in the myocardium had increased significantly to 62.6±3.2% of active renin 6 hours after BNX (Figs 4⇑ and 5⇓), during a time when the concentration of active renin in the myocardium was decaying and the proportions of circulating glycoforms I+II were decreasing significantly. The absolute levels of glycoforms I+II (nanograms Ang I generated per total incubation time per milliliter assay tube) in control and BNX plasma were 34.1±5.0 and 11.6±3.8, respectively. The absolute levels of glycoforms I+II (nanograms Ang I generated per total incubation time per milliliter assay tube) in control and BNX myocardium were 9.0±1.4 and 7.5±1.0, respectively. Myocardial glycoforms III, IV, and V were barely detected 6 hours after BNX (Fig 5⇓).
The left ventricular interstitial space has been estimated to be 18 to 25 mL/100 g wet weight,17 18 19 and the corresponding density of control ventricles is approximately 1.04 g/mL (measured using left ventricles from pilot experiments as blotted wet weight mass per volume of saline displaced). If the myocardial renin in this study was excluded from the intracellular compartment and was not membrane bound, and if all left ventricular blood was removed during the 10-mL ice-cold heparinized saline perfusion, then the measured myocardial renin would be confined to the interstitial space. Providing that control plasma and myocardial interstitial renin were in diffusion equilibrium, then myocardial renin, when expressed as nanograms Ang I per gram wet weight per hour, would be expected to be approximately 18% to 25% of plasma renin. This estimation corresponds to the actual renin measurements in the present study, since control myocardial renin was 23.2±2.6% (n=7) of plasma renin. Care was taken to keep control animals in a steady state, and myocardial and plasma renin concentrations were assayed in an identical fashion (see “Methods”) so that their values could be compared. Therefore, in control animals, plasma renin could have been in simple diffusion equilibrium with myocardial interstitial renin, although it is also possible that the homogenization protocol did not result in complete recovery of cardiac renin.
The myocardial renin measured in this study was apparently not membrane bound. Although 12±3.8% of the myocardial homogenate renin was ultracentrifuged and pelleted with cardiac membranes, this renin activity could be separated from the membrane fraction with a single wash step, indicating nonspecific pelleting of renin along with cardiac membranes in the first ultracentifugation run. A previous study3 also observed about 12% of the myocardial renin associated with the cardiac membrane fraction, but in that study, the renin activity consistently stayed with the membrane fraction after washing. It is possible that renin binding was not demonstrable in the present study due to differences in animal species, degree of renin stimulation, or buffer systems used in the preparation of the myocardial homogenates.
Forty-eight hours after BNX, plasma and myocardial renin concentrations fell 95% and 85%, respectively, suggesting that most myocardial renin is derived from plasma renin of renal origin. Myocardial renin at 48 hours BNX was still above the lower level of detection for all animals studied (significantly above sample blanks). The low myocardial renin concentration at 48 hours BNX may have been due to slow myocardial renin diffusion into the plasma compartment secondary to hepatic clearance of plasma renin, as well as preferential myocardial trapping of renin glycoforms I+II, as discussed below. In addition, the possibility that myocardial renin synthesis is responsible for the 15% of myocardial renin remaining at 48 hours BNX cannot be excluded. Although controversial,2 myocardial gene expression for all constituents of the RAS have been demonstrated,20 and myocardial renin gene expression has been reported to increase secondary to volume overload (cardiac stretch).21 It should also be noted that in normal-salt animals, renin concentration after BNX decayed to levels that were close to the lower limits of detection. Therefore, we used chronic low-salt rats. It may be that a low-salt diet altered the percentage of myocardial renin seen after 48 hours BNX (15% of low-salt controls).
After BNX and careful attention to cathepsin D inhibition, Fordis et al22 were unable to measure renin activity in rat aorta and microvessels. In addition, Schwieler et al23 were unable to demonstrate local de novo synthesis of Ang II in gracilis muscle vasculature. In this report, control myocardial renin per gram tissue was much higher than previous vascular reports14 and much higher after 48 hours BNX than comparable vascular BNX studies.14 22 This may be due to a larger myocardial versus vascular interstitial space14 or myocardial synthesis of renin.
This study attempted to rule out two potential artifacts that may have been present in previous studies. The first artifact is the possible measurement of myocardial reninlike activity not due to renin. In the present study, this possibility appears unlikely for four reasons: (1) Unfractionated BNX plasma was used as an angiotensinogen source at pH 7.5 during Ang I generation, conditions which inhibit cathepsin D activity.2 In our assay, cathepsin D activity is over 97% inhibited when assayed at pH 7.5 versus pH 4.5. (2) The isoelectric focusing profile of myocardial renin activity matched the focusing profile of plasma renin and did not match the focusing profile of rat cathepsin D (rat cathepsin D isoelectric points range from 6.0 to 6.2 in a focusing system that focuses rat renin between 5.35 and 5.6524 ). (3) A combination of serine-, metallo-, and thiol-protease inhibitors was employed during Ang I generation, preventing interference by many other nonspecific proteases. (4) Myocardial renin activity fell after BNX, an unlikely result for a nonrenin myocardial enzyme.
The second potential artifact concerns contamination of myocardial homogenates by plasma-derived renin or angiotensinogen. In this study, plasma contamination was minimized by retrograde perfusion of the aorta with 10 mL ice-cold heparinized saline and blotting both sides of the left ventricular wall before freezing and assay. The fact that plasma and myocardial renin glycoform proportions were significantly different after 6 hours BNX also indicates that plasma renin contamination of myocardial homogenates was probably not significant.
If myocardial angiotensinogen was confined to the interstitial space, its concentration would be expected to be approximately 18% to 25% that of plasma, assuming simple diffusion equilibrium between plasma and myocardial interstitial fluid. In this study, the ratio of control myocardial to plasma angiotensinogen was only 2.7±0.25%. Therefore, in control animals, cardiac angiotensinogen concentration was lower than predicted on the basis of diffusional equilibrium.
In vivo enzymatic cleavage of angiotensinogen by renin in the myocardium probably contributed to the low control myocardial angiotensinogen levels measured in this study. Both plasma and myocardial angiotensinogen increased after BNX, but myocardial angiotensinogen increased more. Presumably, the disproportionate increase in myocardial angiotensinogen was secondary to both elevated levels of plasma angiotensinogen entering the myocardial interstitium after BNX and less available myocardial renin, causing less enzymatic destruction of angiotensinogen after BNX. At 48 hours BNX, the myocardial concentration of angiotensinogen relative to plasma was significantly elevated above the control value of 2.7±0.25% and equaled 8.9±1.5% of plasma angiotensinogen. After 48 hours BNX, myocardial angiotensinogen distribution was somewhat similar to myocardial albumin. Albumin has similar charge and mass to angiotensinogen and undergoes no myocardial enzymatic breakdown. Previous measurements of myocardial albumin distribution have ranged from 6% to 16% of plasma and are lower than smaller-solute interstitial space distributions due to size restriction.25 Therefore, low control myocardial angiotensinogen is probably the result of ongoing myocardial renin enzymatic depletion of angiotensinogen and also because the available volume of distribution for angiotensinogen within the myocardial extracellular matrix may be similar to albumin.
Alternatively, low control myocardial angiotensinogen levels relative to plasma may have resulted from artificially low determinations of myocardial angiotensinogen. During the 3 minutes’ time for removal and cleaning of the myocardium and subsequent homogenization, it is possible that some myocardial angiotensinogen was enzymatically destroyed by renin. However, in vitro incubations simulating control in vivo renin and angiotensinogen concentrations yielded only a negligible fall of myocardial angiotensinogen in 3 minutes. Assay error also appears unlikely, because assay conditions for plasma and myocardium were nearly identical, facilitating comparisons between the two compartments. In addition, the preassay dilution of plasma and ventricular homogenates with PIB containing extensive serine-, metallo-, and thiol-protease inhibitors prevented short-term Ang I breakdown, as demonstrated by a rapidly developed and sustained plateau of generated Ang I and by Ang I spiking of samples. Also, demonstration of a large renin excess for rapid conversion of angiotensinogen to Ang I, excellent blanks, small percent differences between the Ang I concentration at 20- and 40-minute time points, and the fact that when plasma angiotensinogen increases and myocardial renin falls, the measured cardiac angiotensinogen increases as predicted do not make assay artifact seem likely.
Plasma angiotensinogen results in this study are in good agreement with previous determinations.26 However, Danser et al3 reported that angiotensinogen concentrations in pig cardiac tissue were 10% to 25% of those in plasma, versus 2.7±0.25% in the present study. Although the methodology of angiotensinogen assay in the two studies was similar, we briefly perfused hearts with ice-cold saline and thus do not have plasma angiotensinogen contamination, which could have resulted in relatively elevated myocardial angiotensinogen determinations. Significant loss of myocardial angiotensinogen to the perfusate also seems unlikely, since Lindpaintner et al4 showed that significant amounts of angiotensinogen do not leave perfused hearts until well over 20 minutes of constant perfusion.
In the rat, circulating active renin is composed of five major glycoforms (I through V), each with similar enzyme specific activities. The chronic low-salt diet used in this study probably did not alter the proportions of circulating or myocardial control rat renin glycoforms, since very similar renin glycoform proportions were found in a separate group of four normal-salt animals. In addition, a previous study found that chronic high-salt rats and chronic low-salt rats secrete similar renin form proportions from corresponding in vitro kidney slices.27 Since only acute changes in renin secretion stimuli have been previously shown to cause renin glycoform proportional changes,9 12 13 28 it seems unlikely that there are major differences between chronic normal-salt and the low-salt renin glycoform proportions in plasma or myocardium.
In this study, plasma renin and myocardial renin in control rats were found to be composed of the same glycoforms in similar proportions, indicating that all major forms of active renin, regardless of charge or oligosaccharide content, are present in the myocardium. However, high-mannose renin glycoforms possessing relatively small net negative charge (renin glycoforms I and II), are preferentially retained by the myocardium after BNX. At 6 hours BNX, when the concentration and proportion of the high-mannose renin glycoforms I and II were significantly decreased in plasma due to preferential hepatic degradation via the action of the hepatic mannose receptor8 9 10 11 13 and myocardial renin was also decreased, the proportion of renin glycoforms I and II was significantly increased in the myocardium. Measurement of renin glycoforms in the myocardium at 48 hours BNX could not be made due to the overall low level of myocardial renin. Preferential retention of renin glycoforms I and II suggests that myocardial wall exit of (and presumably access to) active renin is a function of net charge and/or oligosaccharide attachments to renin. Thus, myocardial renin derived from plasma renin is unlikely to reflect rapid acute changes in plasma renin for two reasons. First, proportional changes in myocardial renin were always less than plasma renin, indicating that the permeability of the myocardial capillaries to renin was sufficiently low to retard diffusional movement between the plasma and myocardial interstitium. Second, only renin glycoforms I and II participate in acute renin changes in plasma renin concentration,9 12 13 28 yet these same renin glycoforms were preferentially retained by the myocardium after BNX and therefore appear to possess the lowest permeability across the myocardial circulation.
These results are qualitatively similar to a previous study that examined renin glycoform exit from the rabbit carotid artery after BNX.14 In that study, renin glycoforms I and II were also shown to be preferentially retained by the vascular wall. However, the distribution of renin within the carotid wall and the rate of renin exit were both markedly less in the carotid wall than in the myocardium. The role of plasma renin uptake in cardiac interstitial biochemical processes obviously remains obscure. However, these data clarify the active interchange of renin from the kidney to the heart, where sustained local renin activity could subserve chronic functional roles.29
Selected Abbreviations and Acronyms
|PIB||=||proteolytic inhibitor buffer|
This work was supported by grants from the American Heart Association (Minnesota Affiliate), the National Kidney Foundation of the Upper Midwest, and the National Heart, Lung, and Blood Institute (1-KO7-HL03435-01).
- Received June 27, 1996.
- Revision received August 21, 1996.
- Accepted January 29, 1997.
Stock P, Lutz L, Paul M, Ganten D. Local renin-angiotensin systems in cardiovascular tissues: localization and functional role. Cardiology. 1995;86(suppl 1):2-8.
von Lutterotti N, Catanzaro DF, Sealey JE, Laragh JH. Renin is not synthesized by cardiac and extrarenal vascular tissues. Circulation. 1994;89:458-470.
Danser AHJ, van Kats JP, Admiraal PJJ, Derkz FHM, Lamers JMJ, Verdouw PD, Saxena PR, Schalekamp MADH. Cardiac renin and angiotensins: uptake from plasma versus in situ synthesis. Hypertension. 1994;24:37-48.
Lindpaintner K, Jin M, Niedermaier N, Wilhelm MJ, Ganten D. Cardiac angiotensinogen and its local activation in the isolated perfused beating heart. Circ Res. 1990;67:564-573.
Lonn EM, Yusuf S, Jha P, Montague TJ, Teo KK, Benedict CR, Pitt B. Emerging role of angiotensin-converting enzyme inhibitors in cardiac and vascular protection. Circulation. 1994;90:2056-2069.
Katz SA, Opsahl JA, Abraham PA, Gardner MJ. The relationship between renin isoelectric forms and renin glycoforms. Am J Physiol. 1994;267:R244-R252.
Abraham PA, Katz SA, Opsahl JA, Miller R, Stanchfield W, Andersen RC. Renal secretion and hepatic clearance of human multiple renin forms. Hypertension. 1990;16:669-676.
Sessler FM, Jacquez JA, Malvin RL. Different production and decay rates of six renin forms isolated from rat plasma. Am J Physiol. 1986;250:E551-E557.
Shier DN, Malvin RL. Differential secretion and removal of multiple renin forms. Am J Physiol. 1985;249:R79-R84.
Opsahl JA, Abraham PA, Shake JG, Katz SA. Role of renin isoelectric heterogeneity in renal storage and secretion of renin. J Am Soc Nephrol. 1993;4:1054-1063.
Katz SA, Opsahl JA. Biochemistry and physiology of multiple renin forms. In: Laragh JH, Brenner BM, eds. Hypertension. New York, NY: Raven Press; 1994:1489-1502.
Katz SA, Opsahl JA, Forbis LM, Ayenew W. Active renin and renin glycoform dynamics in the carotid artery. Am J Physiol. 1996;271:H184-H191.
Reil GH, Frombach R, Kownatzki R, Quante W, Licthlen PR. Ascorbic acid: a nonradioactive extracellular space marker in canine heart. Am J Physiol. 1987;253:H1305-H1314.
Lindpaintner K, Ganten D. The cardiac renin-angiotensin system: an appraisal of present experimental and clinical evidence. Circ Res. 1991;68:905-921.
Boer PH, Ruzicka M, Lear W, Harmsen E, Rosenthal J, Leenen FHH. Stretch-mediated activation of cardiac renin gene. Am J Physiol. 1994;267:H1630-H1636.
Fordis CM, Megorden JS, Ropchak TG, Kesier HR. Absence of renin-like activity in rat aorta and microvessels. Hypertension. 1983;5:635-641.
Schwieler JH, Nussberger J, Kahan T, Hjemdahl P. Angiotensin II overflow from canine skeletal muscle in vivo: importance of plasma angiotensin I. Am J Physiol. 1994;266:R1664-R1669.
Menard J, Clauser E, Bouhnik J, Corvol P. Angiotensinogen: biochemistry. In: Robertson JIS, Nicholls MG, eds. The Renin-Angiotensin System. London, UK: Grover Medical Publishers; 1993:8.1-8.10.
Opsahl JA, Goldberg MR, Katz SA. Effect of acute and chronic losartan therapy on active and inactive renin and active renin glycoforms. Am J Hypertens. 1995;8:1090-1098.
Everett AD, Tufro-McReddie A, Fisher A, Gomez RA. Angiotensin receptor regulates cardiac hypertrophy and transforming growth factor-β1 expression. Hypertension. 1994;23:587-592.