INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE RENIN ANGIOTENSIN SYSTEM (RAS) plays an important role in the regulation of renal function. It used to be considered a circulating hormone system with ANG II as an effector. Although recent data emphasize its role also as a paracrine system (reviewed in Ref. 14), the functional importance of circulating ANG II has not been questioned. To differentiate the endocrine and paracrine roles of the RAS, it is essential to determine turnover rates of ANG II and its precursor ANG I in the systemic circulation. Earlier studies had shown that circulating ANGs are rapidly degraded and that conversion of ANG I to ANG II occurs predominantly in the lung. However, compared with the huge body of investigations dealing with the RAS, only a few studies were done to quantify conversion and degradation rates of ANGs in the circulation. They focused on arteriovenous differences of endogenous and exogenous ANGs in different organs with the use of biochemical techniques (7, 9, 13, 17, 19, 22, 23). These studies could be done only in large animals that supplied a sufficient quantity of blood for a biochemical assay. The results were highly variable and influenced by the sensitivity of the analytical technique and the experimental protocol employed.

In this study, we used a different approach to quantify the conversion rate of ANG I in the lung and the kidney and degradation rates of ANG I and ANG II in the renal, pulmonary, and systemic circulation in rats. Levels of circulating ANG II were increased by infusing ANG I or ANG II either intravenously, into the left ventricle (LV), or into the left renal artery (RA). Corresponding changes in renal blood flow (RBF) were used as a bioassay to determine renal concentrations of ANG II. Conversion and degradation rates were then determined by mathematical analysis of the comparative data. Until recently, this approach was not viable, because it required intrarenal infusion with its inherent bias. Drugs infused intravenously or into the left ventricle are thoroughly mixed with blood during the passage through the left or both ventricles with turbulent flow. However, because of laminar flow in arteries, agents infused into a renal artery are carried to only a small fraction of the kidney tissue at an ill-defined concentration, and the corresponding change in RBF cannot be used to quantify their vasoactive potency for comparative purposes. This problem was overcome with a recently developed device (18) that ensures homogeneous intrarenal distribution of infused agents. Furthermore, the roles of aminopeptidases A (EC 3.4.11.7) and N (identical with M, EC 3.4.11.2) in intrarenal degradation of ANG I and ANG II were addressed by infusing their inhibitor amastatin into the kidney.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Surgical Procedures

For one set of experiments, a cannula was inserted into the left ventricle for intracardiac infusions. The cannula was connected to a pressure transducer and inserted into the right carotid artery to record arterial pressure. It was then advanced via the ascending aorta up to the aortic valve, where the pulsatile pressure amplitudes declined. From this point the cannula was moved back and forth until the large ventricular pulse could be observed. An infusion of 1 µl/min saline was administered in series with the pressure gauge to keep the orifice patent. During the intervals needed for intracardiac infusions, the cannula was disconnected from the pressure device and connected to infusion pumps.

For the second set of experiments, a Teflon cannula was inserted into the left renal artery via the right femoral artery for intrarenal infusions. The cannula was connected with several lines to infusion pumps and with one line to a device for generating periodic pressure waves (18). The cannula was first inserted into the abdominal aorta above the renal arteries. After exposing the aorta and the left renal artery, a 1% solution of Lissamine green was infused through this catheter system to locate the cannula in the aorta. Then the cannula tip was pulled back a few millimeters above the renal artery and inserted into it by tilting the artery over the tip with two glass hooks. Saline containing 5 U heparin/ml was infused at a rate of 5 µl/min through the cannula throughout the experiment. During intrarenal infusions the pressure wave generator was operated to ensure a thorough mixing of drugs with renal arterial blood. Periodically, it sucked ~4 µl blood from the renal artery within 700 ms and ejected it back within 70 ms. The blood oscillation could be observed in the cannula outside the animal.

Experiments were designed to compare the effects of ANG I and ANG II (Sigma Chemicals) given at different sites (intravenous, LV, or RA) on RBF. Precautions were taken to achieve a high reproducibility of the drug effects needed for this purpose. To minimize the loss of peptides caused by adsorption on the glass wall of the infusion syringe, the peptides were dissolved in saline containing 10⁵ M BSA. Aliquots of stock solutions prepared and frozen (20°C) at the beginning of the study were used within 1 h after being thawed. Contents of connecting catheters were renewed by a short infusion 5 min before experimental infusions. Identical solutions were used to compare renal effects of a peptide administered at different sites, and alternate measurements were repeated two to three times to verify their reproducibility. Peptide concentrations were adjusted to 10⁷-10⁶ M to obtain the required doses by infusion of 5-50 µl/min and for comparison with equimolar dose by infusion of 10 µl/min. Because the effects of peptides were reversible, up to three different interventions were done on the same animal.

Experimental Procedures

Comparison of intravenous ANG I vs. intravenous ANG II. Infusions of ANG I and ANG II (10 pmol/min) were separated by a control period of 5 min. The reproducibility of the results was tested with two different aliquots of each peptide.

Comparison of LV vs. intravenous ANG I and ANG II. The peptides were infused either into the left ventricle or intravenously at a rate of 10 pmol/min.

Comparison of RA vs. intravenous ANG II. The intrarenal dose was chosen to be one tenth of that for intravenous infusion, 1 pmol/min RA vs. 10 pmol/min iv. The rationale for using a one tenth-dose for RA infusion was derived from the consideration that RBF is approximately one tenth of the cardiac output. In these and the following experiments MBF was also measured.

Comparison of RA vs. intravenous ANG I. An identical dose of 5 pmol/min ANG I was infused into the renal artery and intravenously.

Effects of amastatin coinfusion on RA ANG I and ANG II: These experiments were done to determine the roles of aminopeptidases A and N, which are inhibited by amastatin, in renal degradation of ANG I and ANG II. We compared the effects of RA infusions of 5 pmol/min ANG I or 1 pmol/min ANG II before and during RA infusion of amastatin at a rate of 5 nmol/min (Calbiochem). This dose of amastatin had been found to produce a close to maximum effect in preliminary experiments. The infusion of amastatin was started at least 5 min before the infusion of peptides to obtain a new baseline value. In additional experiments (n = 4) we tested NH₂-ANG II (Hypertensin, Ciba), which is resistant to hydrolysis by aminopeptidase A.

The data are presented as means ± SE. Statistical comparisons were done by paired t-test. P values below 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The compiled data are shown in Table 1. Figure 1 shows the dose-response curves for the effects of intravenous ANG II on RBF. Corresponding experiments done with ANG I produced similar results (Table 1). Figure 2 shows that intravenous infusion of ANG I had an effect on RBF similar to that of ANG II (0.05 < P < 0.1). It also shows comparable effects of LV vs. intravenous infusion of ANG II (0.05 < P < 0.1). However, LV infusion of ANG I was distinctly less effective than intravenous infusion. In following experiments we compared effects of ANG II, 1 pmol/min RA vs. 10 pmol/min iv, and effects of equimolar ANG I, 5 pmol/min RA vs. iv. Absolute values of MBF in these experiments ranged between 8 and 12 ml/min. As shown in Fig. 3, reduction in RBF by low-dose RA ANG II was smaller than that for intravenous infusion, and the effect of RA ANG I was larger than that for intravenous infusion. However, the renal effects of 1 pmol/min RA ANG II and 5 pmol/min RA ANG I in the two series of experiments were not different. RA infusions of peptides had either no (ANG II) or a much lesser (ANG I) effect on MBF and BP than did intravenous infusions. Next we compared the effects of RA ANG II and ANG I under control conditions and during coinfusion of amastatin, an inhibitor of aminopeptidases A and N. Amastatin reduced RBF by 2.2 ± 0.6% but had no significant effect on BP. As shown in Fig. 4, under control conditions 1 pmol/min RA ANG II and 5 pmol/min RA ANG I reduced RBF to a similar extent, and amastatin increased the effects by 50%. Amastatin had no significant effect on the corresponding changes of MBF and BP. Four additional experiments revealed that amastatin did not accentuate effects of NH₂-ANG II (1 pmol/min RA), which is resistant to cleavage by aminopeptidase A [percent change in RBF (%RBF), 23.8 ± 0.9 vs. 25.2 ± 0.9 after amastatin].

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Dose-response relationship between intravenously infused ANG II and reduction in renal blood flow (%RBF). Values are means ± SE.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Comparative effects (filled vs. open bars) of 10 pmol/min ANG II and ANG I infused iv and into left ventricle (LV) on %RBF and blood pressure (%BP). First pairs of columns show effects of iv ANG I vs. ANG II, second pairs show effects of LV ANG II vs. iv ANG II, and third pairs show effects of LV ANG I vs. iv ANG I. * P < 0.05.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Comparative effects of renal artery (RA) vs. iv infusion (filled vs. open bars) of ANG II and ANG I on renal and mesenteric blood flow (%BF), and %BP. First pairs of columns show effects of 1 pmol/min ANG II RA vs. 10 pmol/min ANG II iv; second pairs show effects of equimolar dose (5 pmol/min) ANG I RA vs. iv. * P < 0.05 RA vs. iv.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Comparative effects of RA infusion of ANG II and ANG I under control conditions (CON, open bars) and during coinfusion of amastatin (AMA, filled bars) on renal and mesenteric %BF and %BP. First pairs of columns show effects of 1 pmol/min ANG II, and second pairs show effects of 5 pmol/min ANG I. * P < 0.05 AMA vs. CON.

Theoretical Considerations and Conclusions

Conclusion 1: conversion of ANG I in the lung. Comparison of equimolar intravenous ANG I vs. ANG II showed that a marginally smaller effect of ANG I on RBF corresponded to a U ratio (ANG I:ANG II) of 0.83 ± 0.08, which was not significantly different from unity (0.05 < P < 0.1). This can be explained by an almost complete conversion (C_L  1) into vasoactive ANG II during the first circulation passage through the lung. It cannot be ruled out, however, that in the lung both peptides degrade to an identical degree to inactive products (D_{I L}  D_{II L}), and the systemic conversion is not addressed.

Conclusion 2: degradation of ANG I and ANG II in the lung. If part of ANG II is degraded in the lung, LV infusion should be more effective than intravenous infusion. However, similar renal effects found for intravenous and LV infusions indicate only minimal degradation of ANG II in the lung. The U ratio (intravenous:LV) of 0.86 ± 0.06 was not significantly different from unity (0.05 < P < 0.1), indicating that degradation of ANG II in the lung, if any, could only be marginal (D_{II L}  0). Considering that intravenous ANG I is almost entirely converted in the lung and is equipotent to intravenous ANG II, the conclusion for ANG II should be also true for ANG I (D_{I L}  0).

Conclusion 3: degradation of ANG I and ANG II in the kidney. Infusion of 5 pmol/min ANG I into the left kidney caused a small but significant reduction in MBF (3.2 and 4.2% in 2 different series). For the kidney such a reduction would correspond to an intravenous dose of ~0.5 pmol/min (10% of RA dose). For the MBF, which was more sensitive to intravenous ANGs than was RBF (Table 1), this would correspond to a somewhat smaller dose. Hence the degradation of ANG I during a single renal passage is >90% (D_{I K} > 0.9).

ANG II (1 pmol/min) given into the renal artery had no detectable effect on the flow in the mesenteric artery, indicating a substantial renal degradation. A higher dose of RA ANG II necessary to quantify the change in MBF could not be used without causing renal ischemia. In analogy to ANG I, however, it is reasonable to assume that ANG II is also degraded by >90% during a single renal passage (D _II,K > 0.9).

Conclusion 4: degradation of ANG I and ANG II in the systemic circulation. Comparison of RBF effects of ANG II (10 pmol/min iv vs. 1 pmol/min RA) showed a larger change in response to intravenous infusion, which corresponds to a 1.6-fold higher effective dose (U ratio 1.58 ± 0.22). After starting intravenous infusion the initial concentration of ANG II increases by a factor of 1/D_S as a result of accumulation of recirculating ANG II [1 + R_S + R²_S + R³_S + . . . = 1/(1  R_S) = 1/D_S]. Thus in the steady state a kidney obtaining 10% of cardiac output would get 16% of the intravenous dose (1.6 pmol/min) at D_{II S} = 0.63 (1/0.63 = 1.6, 63% systemic degradation). It should be pointed out that this estimation includes a possible minor degradation in the lung; renal circulation is included as part of the system.

The reliability of the above conclusion indicating two-thirds degradation in the systemic circulation was tested in three additional experiments for the hind limb circulation. In these experiments a threefold higher ANG II dose (30 pmol/min) infused into the aorta below the renal arteries produced comparable effects on RBF as intravenous infusions (10 pmol/min).

Comparison of LV vs. intravenous infusions of ANG I (10 pmol/min) showed that the LV infusion was less effective, corresponding to a 0.45-fold dose (U ratio 0.45 ± 0.05). Of the intravenously infused ANG I converted in the lung, 16%, i.e., 1.6 pmol/min, reached the kidney as ANG II. Therefore, the effective renal dose during LV infusion should have been 0.45 + 1.6 = 0.72 pmol/min ANG II. As will be shown in Conclusion 5, 0.2 pmol/min of this is the result of renal conversion during the first cycle (20% conversion in 10% of cardiac output). Therefore, the remaining ANG II (0.72  0.20 = 0.52 pmol/min) can now be accounted for by assuming R_{I S} = 0.33 so that of LV ANG I (3.3 pmol/min; 10 × 0.33 pmol/min) reaches the lung and is converted there, whereas 16% of it (3.3 × 0.16 = 0.52 pmol/min) goes to the kidney. Accordingly, the rate of systemic degradation of ANG I is 67% (1  R_{I S}), which is identical to that for ANG II (63%).

Conclusion 5: conversion of ANG I in the kidney. As shown in Fig. 4, in untreated kidney RA infusions of ANG I (5 pmol/min) and ANG II (1 pmol/min) had similar effects on RBF, with a U ratio of 1.04 ± 0.10 (ANG I:ANG II). Thus 20% of ANG I is converted in the kidney. Similar results were also obtained from two different series of experiments shown in Fig. 3. For the integrated analysis, however, we used a U ratio of 1.85 ± 0.13 obtained for ANG I (5 pmol/min) RA vs. intravenous infusion.

Conclusion 6: integrated analysis. Next we optimized numerically the mean values of all U ratios with a mathematical model combining the above single step considerations, assuming degradation rates of ANG I and ANG II to be identical for any vascular bed (D_{I L} = D_{II L}, D_{I K} = D_{II K}, D_{I S} = D_{II S}). This contention is justified by the findings that most of the peptidases degrade both peptides in a similar manner (1, 2). It should be pointed out that this model does not need measured changes in MBF.

For the integrated analysis, fractional doses of the infused ANG I and ANG II reaching the kidney were calculated by the following equations with the use of an additional abbreviation, N, for nonconverted (1  C) The values for independent variables C_L, C_K, D_L, D_S, and D_K were optimized to fit the determined U ratios. Optimization of the parameters was done by minimizing the normalized absolute sum of the difference between measured and predicted U ratios. The measured and calculated U ratios, respectively, were: ANG II:ANG I (intravenous), 0.83 ± 0.08 and 0.85; iv:LV (ANG II), 0.86 ± 0.06 and 0.95; iv:RA (ANG II), 1.58 ± 0.22 and 1.36; LV:iv (ANG I), 0.45 ± 0.05 and 0.50; RA:iv (ANG I), 1.85 ± 0.13 and 1.85. Conversion and degradation parameters obtained by the model are shown in Table 2; the values are within the range of those obtained by single-step considerations.

Conclusion 7: degradation of ANG I and ANG II by aminopeptidases A and N in kidney. Comparison of the effects of RA ANG I (5 pmol/min) and RA ANG II (1 pmol/min) under control conditions and during coinfusion of the aminopeptidase inhibitor amastatin (5 nmol/min) showed that in the presence of amastatin the peptides were more effective in reducing RBF. The U ratio of the effective dose without vs. with amastatin was 0.39 ± 0.04 for ANG II and 0.44 ± 0.04 for ANG I. Accordingly, in the absence of amastatin, the peptides were degraded by 60% in the kidney before they could elicit their vasoactive properties either directly or after intrarenal conversion.

Amastatin has been shown to be an inhibitor of aminopeptidases A (EC 3.4.11.7) and N or M (EC 3.4.11.2) (1, 20, 24). Aminopeptidase A cleaves N-terminal acidic amino acids (asparagine in ANG II and ANG I) (2) to produce ANG III (2-8 ANG II) and 2-10 ANG I. Subsequently, aminopeptidase N cleaves arginine in position 2 (2), leading to ANG IV (3-8 ANG II) and 3-10 ANG I. ANG III generated from infused ANG II or after intrarenal conversion of infused ANG I has been shown to be a potent renal vasoconstrictor (6), but the peptide loses its constrictor property after cleavage of arginine to ANG IV (2, 11). That the above steps are involved in the intrarenal degradation of the peptides was further substantiated by additional experiments using hypertensin (NH₂-Asp ANG II), which is resistant to the first step of hydrolysis by aminopeptidase A (1, 2, 24). In these experiments amastatin failed to potentiate the effect of RA hypertensin.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study we quantified the fractional conversion and degradation of exogenous ANG I and ANG II in the pulmonary, systemic, and renal circulation in rats by comparing effects of ANGs infused at different sites and by using RBF as a bioassay parameter for circulating ANG II.

Three decades ago Biron et al. (3) used BP as an assay parameter to quantify degradation rates of ANG II in rats and dogs. They adjusted doses of ANG II given intravenously, into the ascending aorta, and into the renal artery to produce identical changes in systemic blood pressure measured with a mercury manometer. They found for both species 5% degradation in the lung and 70% degradation in the systemic circulation, which was confirmed in our study. However, the reported intrarenal degradation rate of 70% was much less than in our study. It should be noted that the change in RBF is more reproducible than that in BP. Contemporary studies by Hodge et al. and Ng and Vane (12, 15, 16), who used contraction of rat colon and stomach strip superfused with dog blood in an extracorporal circuit to measure circulating ANG II, also reported a comparable (65%) degradation in the systemic circulation (16), a negligible degradation in the lung, and a degradation of 75% (12, 16) in the kidney. Additionally, they demonstrated that the conversion of ANG I occurs predominantly in the lung (15, 16). We could confirm all conclusions from the above studies except for rates of renal degradation of the peptides. This difference is very likely related to the high RA dose of ANG II needed in those experiments, which increased systemic blood pressure by 20-25 mmHg. This would increase the concentration of ANG II in renal plasma to extremely high values as a result of concomitant reduction of RBF almost to zero and thereby override the degrading capacity of the kidney.

Infusions or bolus injections of equipotent doses of ANG I and ANG II in the dog kidney for reducing RBF were used to demonstrate the ability of the kidney to convert ANG I. However, intrarenal conversion of ANG I to ANG II estimated from the reported dose ratio varied between ~3% (4, 8) and ~20% (10, 21). Intrarenal conversion of 9% (17) and 22% (22) has been determined by infusing ANG I into the dog kidney and measuring the concentration ratio of ANG II:ANG I in renal venous blood samples. Our results correspond to the upper estimates from the above studies.

Other studies, designed to assess the metabolism of ANGs, analyzed blood samples collected at different sites under basal conditions and during infusion of ANGs. Except for one study done on rats (13), in which 5 ml blood was collected from the carotid artery, the experimental animals were dogs, pigs, or sheep. Fractional conversion of ANG I to ANG II in the lung determined in these studies varied considerably, being 25% in pigs (7), 32% in sheep (9), 66% in rats (13), and 72% in dogs (23). Degradation of ANG I in the lung to peptides other than ANG III varied between 0 and 18% (7, 9, 13). In two studies the production of ANG III from ANG I was also determined, and the fractional metabolism was found to be close to 0% in one study (13) and 17% in the other (9). Our results do not allow us to differentiate between ANG II and ANG III, which are similarly potent constrictors in the renal circulation (6). Systemic degradation of ANG II has been reported to be 70% in sheep (9), and that of ANG I has been reported to be 50% in pigs (7). Renal degradation of ANG I was 90% in pigs (7), and that of ANG II was 70% in dogs (19). The large variations in data from the above studies seem to reflect methodological rather than species differences. Problems in determining peptide concentrations in circulating plasma may arise from changes occurring during blood collection from catheters, nonquantitative separation with HPLC (5), and changes in peptide metabolism at high concentrations. Szidon et al. (23) have demonstrated that fractional pulmonary conversion of ANG I can be reduced from 72 to 55% by increasing circulating ANG I from tracer concentrations to 1,000× a physiological level. In our experiments the rise in plasma concentrations of ANG II, calculated from RBF and renal infusion rate, was about 300 pM above the basal value.

Our study does not allow us to determine peripheral conversion of ANG I to ANG II except in the kidney. In one study, fractional conversion in peripheral circulation was found to be 10% (7), whereas other studies could not detect any conversion in the systemic circulation (15) or in the circulation of the liver and hind limb (17). It should also be noted that anesthetized operated rats, as used in our study, have higher levels of circulating renin and ANGs than do conscious animals, which might have modulated the calculated rates of conversion and degradation. Furthermore, systemic and intrarenal infusions of the peptides would increase their concentration in blood additionally as a result of concomitant reductions in cardiac output and RBF, respectively. These correction factors were neglected in calculating our dose ratios RA:iv, because they would cancel out each other, at least partially.

The present study also addresses peptidases involved in intrarenal degradation of ANG I and ANG II in vivo. Coinfusion of amastatin increased the effective dose of RA-infused peptides 2.5-fold, indicating that inhibited peptidases were responsible for a 60% degradation of the peptides. Amastatin inhibits aminopeptidase A (EC 3.4.11.7) with IC₅₀ of 8 µM and aminopeptidase N (EC 3.4.11.2) with IC₅₀ of 0.2 µM (1). As discussed above, ANG II is converted by aminopeptidase A to ANG III and then by aminopeptidase N to vasoinert ANG IV. The higher potency of amastatin to inhibit aminopeptidase N suggests that this enzyme played the crucial role in our experiments. The important role of aminopeptidase A in this cascade was demonstrated with the use of hypertensin (NH₂-ANG II), which is resistant to aminopeptidase A; its effect was not influenced by amastatin. It is interesting to note that amastatin inhibited a degradation of ANG I and ANG II by 60% for ANG II receptors in renal vessels, but apparently did not considerably increase the concentration of the peptides in renal venous effluent. Amastatin increased the effect of ANG I on BP and MBF nonsignificantly, indicating that the rise in concentration of the peptide in the renal vein, if any, was only marginal. Therefore, it is conceivable that the degradation of ANG II is compartmentalized, and other peptidases are also involved in renal degradation (2).

In conclusion, our study quantifies the conversion and degradation of circulating ANG I and ANG II in rats with a uniform model. It also pinpoints the role of aminopeptidases A and N in the renal degradation of the peptides.

Perspectives