A STUDY OF THE ACTION OF ANGIOTENSIN II ON PERFUSION THROUGH THE CORTEX AND PAPILLA OF THE RAT KIDNEY

Transcription

1 Experimental Physiology (1991) 76, Printed in Great Britain A STUDY OF THE ACTION OF ANGIOTENSIN II ON PERFUSION THROUGH THE CORTEX AND PAPILLA OF THE RAT KIDNEY CHUNLONG HUANG*, GERARD DAVIS AND EDWARD J. JOHNS Department of Physiology, The Medical School, Birmingham B15 2TT (MANUSCRIPT RECEIVED 31 JANUARY 1991, ACCEPTED 18 MARCH 1991) SUMMARY The effect of angiotensin IL on blood pressure and perfusion of blood through the cortex and papilla regions of the kidney was determined in pentobarbitone-anaesthetized rats which were subjected to laser-doppler flowmetry to estimate regional renal haemodynamics. Angiotensin II was infused at 10, 45 and 150 ng (kg body weight-1 min-m) which caused dose-related increases in blood pressure of 3, 12 and 24 %, respectively, and decreases in cortical perfusion of 9, 15 and 24 %, respectively. Papillary perfusion did not change at any dose of angiotensin II. This pattern and magnitude of responses to angiotensin II in blood pressure, cortical and papillary perfusions was essentially unaffected (a) following blockade of cyclo-oxygenase activity with indomethacin (1 3 mg kg-' plus 2 mg kg-' h-1), (b) during infusion of a bradykinin antagonist, at 1 3 ug minm, (c) when renal perfusion pressure was regulated at control levels and (d) following Methylene Blue administration to inhibit potential endothelial-derived relaxing factor production. By contrast, infusion of phenylephrine at 5, 10 and 20,ug kg-1 min-' caused dose-related increases in blood pressure and decreases in both cortical and papillary perfusions reaching some 28, 7 and 17 % respectively at the highest dose of phenylephrine used. These results showed that both cortex and papilla were sensitive to vasoconstrictor agents. They are compatible with the suggestion that angiotensin II regulates cortical but not papillary perfusion in the kidney, and that these responses do not depend on prostaglandin, bradykinin, renal perfusion pressure or endothelium-derived relaxing factor. INTRODUCTION The vasculature of the kidney varies markedly in anatomy and architecture between cortex, medulla and papilla and as a result may be differentially regulated by neural and humoral factors. The major proportion of total renal blood flow perfuses the cortex, where filtration, bulk reabsorption and secretion occur. Some % of blood flow reaches the deeper regions of the medulla and passes into the papilla and this appears to be involved in the maintenance of the high osmotic gradients necessary for the concentration of urine (Jamison & Kriz, 1982). At the middle and outer cortical glomeruli, the efferent arterioles quickly break up into the peritubular capillaries surrounding the convoluted tubular portions of the nephron (Zimmerhackl, Robertson & Jamison, 1985). The situation is different at the juxtamedullary glomeruli, in so far as the wide muscular efferent arteriole immediately turns and enters the medulla where it divides to form up to thirty descending vasa recta. These vasa recta vessels penetrate to varying levels within the medulla with only a few of the larger diameter descending to reach the tip of the papilla (Pallone, Robertson & Jamison, 1990). Measurement of regional blood flow in the cortex, medulla and papillary regions of the kidney is notoriously difficult and each approach is open to criticism (Chou, Porush & * Present address: Biology Department, Tianjin Normal University, Tianjin, People's Republic of China.

2 788 C. HUANG, G. DAVIS AND E. J. JOHNS Faubert, 1990). The techniques that have been applied are an accumulation of radiolabelled markers, which only offer an estimate of perfusion, and the videomicroscopic approach which analyses in detail flow through individual vasa recta vessels in the papilla. Recently, laser-doppler flowmeters have been developed and applied to the measurement of regional haemodynamics in the kidney, and Smits, Roman & Lombard (1986) and Roman & Smits (1986) have validated its use against conventional methodologies in both the renal cortex and papilla. One factor which is importantly involved in the regulation of renal haemodynamics is angiotensin II. At the whole-kidney level angiotensin II can reduce total renal blood flow, but it is also involved in a subtle way in the regulation of glomerular filtration rate by means of a differential effect on pre- and post-glomerular resistances (Johns, 1989). There is a general consensus that under normal conditions angiotensin II exerts a tonic influence on perfusion of the medulla and papilla, as when its action is blocked by converting enzyme inhibitors or receptor blockers perfusion in these regions increases (Pallone et al. 1990). By contrast, a preliminary report by Davis & Johns (1990) demonstrated that although systemically administered angiotensin II reduced perfusion of the cortex it had no effect on papillary perfusion. This phenomenon was examined further by assessing the potential role of other systems which are active within the kidney at the level of the microvasculature. METHODS Young male Wistar rats, weighing g (mean +s.e.m.), were anaesthetized (60 mg kg-' sodium pentobarbitone i.p. plus 6 mg kg-1 h-' i.v.) and cannulae were placed in the left jugular vein, for infusion of normal saline and drugs, and the right carotid artery, for blood pressure measurements (Statham P231D linked to a Grass model 7 polygraph). In group IV, (see below) blood pressure was measured from a femoral arterial cannula. The left kidney was exposed via an abdominal incision, cleared of peripheral fat, its artery, vein and ureter freed, and placed in a plastic holder to expose its dorsal surface. The ureter was sectioned and dissected towards the renal hilus to expose the papilla. Laser-Doppler perfusion was measured using a Perimed PF3 (Stockholm, Sweden) and a PF303 probe (Stockholm, Sweden) of 1 mm diameter. The laser-doppler flowmeter was calibrated such that one perfusion unit was equivalent to 10 mv. Perfusion measured in this way represents the product of the velocity of moving blood cells and the concentration of moving blood cells (CMBC) in the volume of tissue under the probe. Thus, the velocity is measured as the magnitude of the frequency shift whilst the intensity of the signal is proportional to the number of red cells moving within the tissue being illuminated. CMBC approximates to the haematocrit beneath the probe and with the PF3 flowmeter this component can be extracted from the perfusion signal and is expressed as volts (Nilsson, 1990). Six to eight measurements of 1-2 min duration were made at different sites over the cortex and one or two measurements were made over the papilla, and the average values were calculated. An intravenous infusion of normal saline was started at a rate of 2 ml h-' immediately after the venous cannula had been inserted, and was continued throughout the experiment. The initial measurements began 1 h after exposure of the papilla. One of the following experimental protocols was then conducted. Protocols Group I (n = 7): angiotensin II alone. In this group of animals baseline measurements of cortical and papillary perfusion were taken before and after an infusion of angiotensin II (Sigma, Dorset) at 10, 45 and 150 ng kg-1 min-1 i.v. in random order. Experimental measurements of regional flows were taken once they had stabilized at the new levels induced by angiotensin II, which took approximately 15 min. Group II (n = 8): angiotensin II during indomethacin infusion. The initial measurements of cortical

3 ANGIOTENSIN II ANO RENAL HAEMODYNAMICS 789 Table 1. The effect of angiotensin II (ng kg-1 min- ) on blood pressure and renal cortical and papillaril haemodynamics [Angiotensin II] A (n = 7) Systemic blood pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) Saline * ** 115+3*** * 160+9** 144+9*** ( * l1 * Indomethacin (1 3 mg kg-'+ 2 mg kg-' h-') [Angiotensin II] B (n= 8) Saline Control Systemic blood pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) * * *** *** *** *** * * [Angiotensin II] C (n = 7) Saline 45 Renal Perfusion pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) Control RPP 150 Uncontrolled RPP *** ** *** 120+8** * ** n = number of animals tested. * P < 0 05; P < 001; *** P < RPP = renal perfusion pressure. and papillary perfusion were taken to establish basal levels, and an intravenous infusion of the cyclooxygenase inhibitor indomethacin (Sigma, Dorset) was continued to the end of the experiment. Indomethacin (I mg ml-') was placed in normal saline and sodium bicarbonate (1-5 mg) was added slowly until the indomethacin dissolved. A bolus dose of 1 3 mg kg-1 in a volume of 0-25 ml was given followed by a steady infusion of 2 mg kg-' h-1, and the experiment was begun 30 min later. Measurements of cortical and papillary perfusion were taken before, during and after an infusion of angiotensin II at 10, 45 and 150 ng kg-1 min' given in random order. Group III (n = 4): angiotensin II during bradykinin antagonist infusion. After measurement of basal levels of renal haemodynamics the bradykinin antagonist (D-Arg, Hyp3, Thi5 8, D-Phe')-bradykinin

4 790 C. HUANG, G. DAVIS AND E. J. JOHNS (Peninsula Laboratories Europe Ltd, Merseyside) was infused at a dose of 1 39,tg min-' and continued to the end of the experiment. Measurements of cortical and papillary perfusion were taken 30 min after the start of antagonist infusion, and during an infusion of angiotensin II at 45 and 150 ng kg-' min-m. Bradykinin (Peninsula Laboratories Europe Ltd, Merseyside) was given as an intravenous bolus of 333 ng in a volume of 0 2 ml before the antagonist infusion, and at the end of the experiment whilst the antagonist was still being infused. Group IV (n = 7): angiotensin II during controlled renal perfusion pressure. In this group of animals, the arterial cannula was placed in the right femoral artery and a silk ligature was loosely placed around the aorta rostral to the left renal artery and attached to a screw device such that the aorta could be constricted, thereby reducing pressure at the kidney (Davis & Johns, 1991). After the initial measurements, angiotensin II was infused at 45 and then at 150 ng kg-' min-', and as blood pressure began to rise under the influence of angiotensin II, the suture around the aorta was tightened such that perfusion pressure was returned to its value immediately prior to administration of the drug, and was kept at that level for the duration of the angiotensin infusion. The ligature was then released during angiotensin II infusion at 150 ng kg-1 min-' to allow perfusion pressure to rise, and measurements were then taken at the higher pressure. The infusion of angiotensin II was stopped and baseline measurements were taken once blood pressure had stabilized. Group V (n = 4): angiotensin II and Methylene Blue. In this group, angiotensin II was only infused at 150 ng kg-1 min-'. Initially the renal haemodynamic responses to angiotensin II were obtained. Methylene Blue (Sigma Chemical Company, Dorset), an inhibitor of the endothelium-dependent relaxing factor, was given intravenously as a bolus of 1 15 x 10-3 M in 0-2 ml and measurements were taken 20 min later. Two further doses of Methylene Blue (1-15 x 10-3 M in 0-2 ml) were given i.v. with measurements being taken 20 min after each dose. Group VI (n = 7): phenylephrine administration. The infusion and measurement protocol in this group was similar to that in group I. Phenylephrine hydrochloride (Boots Company plc, Nottingham), an ax1-adrenergic receptor agonist, was infused at 5, 10 or 20,ug kg-1 min-1 for 20 min at increasing doses and once a stable blood pressure had been achieved, papillary and cortical perfusion measurements were undertaken. Statistics The mean values+ standard errors of the means are presented and the absolute and percentage changes quoted are the means of individual values obtained from each group. The effect of indomethacin, bradykinin or Methylene Blue on baseline levels was calculated as the difference between the value obtained immediately before application of the compound and that obtained after its administration. The response to angiotensin II and phenylephrine was taken as the difference between the mean values recorded before and after administration with that obtained during administration of these vasoactive agents. Regression analysis was carried out to give the gradient and correlation values. Differences were taken as significant at the 5 % level using Student's t test and analysis of variance (Fastat Software and Apple Macintosh) as appropriate. RESULTS The initial study was undertaken to define the blood pressure and renal haemodynamic responses to angiotensin II given intravenously (Table 1 A). Angiotensin II caused doserelated increases in blood pressure with a gradient of mmhg (ng angiotensin II (kg body weight)-' min-')-' (P < 0 001). This was associated with a graded reduction in perfusion within the cortex which reached 24% at the highest dose of angiotensin II, with a gradient of PU (ng angiotensin II kg-' min-m') (P < 0 001). Cortical CMBC was only slightly, but significantly (P < 0 05) reduced at the two highest rates of infusion. By contrast, angiotensin II had no effect on papillary perfusion or CMBC. The potential role of prostaglandins in these responses was determined by blocking their production with the cyclo-oxygenase inhibitor indomethacin, and the results are presented

5 ANGIOTENSIN II AND RENAL HAEMODYNAMICS Change in systemic 30 blood pressure (mmhg) 20 Lg [Angiotensin II] (ng kg-1 min-1) -20-Vx Change in cortical perfusion IVA Change in papillary l'v ' 1 perfusion -lo0l &Irj Fig. 1. The absolute responses in systemic blood pressure, cortical and papillary perfusion (expressed as perfusion units, PU) in two groups of rats given increasing doses of angiotensin II (10, 45 and 150 ng kg-1 min-') in the absence (C1; group I rats) and presence of indomethacin (1 3 mg kg-1 plus 2 0 mg kg-1 h-i; 0; group II rats). Error bars show S.E.M. in Table 1 B. Administration of indomethacin (I 3 mg kg-' plus 2 mg kg-1 h-1) had no effect on blood pressure or cortical perfusion but caused a significant (P < 0O05) reduction in papillary perfusion of PU. Neither cortical nor papillary CMBC were affected. Under these conditions, angiotensin II infusion caused graded increases in blood pressure, with a gradient of mmhg (ng angiotensin II kg-' min-')-' (P < ), reaching mmhg at the highest dose, and decreases in cortical perfusion, having a gradient of PU (ng angiotensin II kg-' min-')-' (P < 0-001) with a maximum decrease of 29 %. The gradient of cortical perfusion to the angiotensin II in the presence of indomethacin was significantly (P < 005) greater than that obtained when indomethacin was not present. However, papillary perfusion was not changed by any dose of angiotensin II. Infusion of angiotensin II decreased cortical and increased papillary CMBC (P < 0 05) only when given at the highest dose. A comparison of the absolute changes in blood pressure, cortical and papillary perfusions in the absence and in the presence of indomethacin is given in Fig. 1. A second small study was performed in which a bradykinin antagonist was given intravenously, and the data are presented in Table 2A. A bolus dose of bradykinin (333 ng) was given and this caused a transient fall in blood pressure with a peak response of mmhg. A similar dose of bradykinin given at the end of the period of bradykinin antagonist application decreased blood pressure by only mmhg, an almost 50 % inhibition of the response. Administration of the bradykinin antagonist had no effect on

6 792 Table 2. C. HUANG. G. DAVIS AND E. J. JOHNS The effect ofangiotensin II (ng kg-1 min- ) and phenylephrine on bloodpressure and renal cortical and papillary haemodynamics Bradykinin antagonist (1 39 pg min-1) [Angiotensin II] A (n = 7) Saline Control Systemic blood pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) B (n = 4) Systemic blood pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) Saline ** 138+4** ** ** * * Methylene Blue 3-45 x 10-3 M [Angiotensin II] ** Phenylephrine (aug kg-1 min-1) C (n = 7) Saline * * Systemic blood pressure (mmhg) Cortical perfusion Papillary perfusion Cortical CMBC (V) Papillary CMBC (V) ** 117+6** 127+7** * ** ** * ** Abbreviations, see Table 1. blood pressure, cortical perfusion or cortical or papillary CMBC, but caused a significant (P < 0 05) reduction in papillary perfusion of PU. Only two doses of angiotensin II were given, 45 and 150 ng kg-1 min-1, and these increased blood pressure significantly (both P < 001) by and 28+3 mmhg respectively, and significantly (both P < 0-01) decreased cortical perfusion by 12 % and 15 %/o respectively, while papillary perfusion was not changed. Under these conditions angiotensin II had no effect on the CMBC of either cortex or papilla. The blood pressure and renal haemodynamic responses in the absence and presence of bradykinin is given in Fig. 2. A further potential complicating factor was the rise in blood pressure during angiotensin II adminstration. Therefore, a study was undertaken in which perfusion pressure at the

7 ANGIOTENSIN II AND RENAL HAEMODYNAMICS 793 Change in systemic 30 - blood pressure (mmhg) ~~~~~~~~~ [Angiotensin II] (ng kg-1 min-) -10 Change in cortical t - perfusion _30 _ _1 10 Change in papillary perfusion 0 _10L Fig. 2. The absolute changes in systemic blood pressure, cortical and papillary perfusion (expressed as perfusion units, PU) in two groups of rats given two doses of angiotensin II, 45 and 150 ng kg-' min-m in the absence (D-; group I rats) and in the presence of bradykinin antagonist (1 39 lug min-';1; group III rats). kidney was regulated at constant levels, and the results are shown in Table 1 C. In this group of animals blood pressure was effectively regulated during both the 45 and 150 ng kg-1 min-1 angiotensin II infusions, but there were significant reductions in cortical perfusion of 13 and respectively, while papillary perfusion was not changed at either dose of angiotensin II. Papillary CMBC was not changed following administration of either dose of angiotensin II, although there was a significant reduction in cortical CMBC at the higher dose. When perfusion pressure at the kidney was allowed to rise to prevailing levels (from 96+4 to mmhg) by loosening the aortic constriction, cortical perfusion remained at its depressed level and there was a small, but not significant, increase in papillary perfusion, while cortical and papillary CMBC remained unchanged. These responses are illustrated in Fig. 3. The effect of blocking the effects of potential EDRF formation with Methylene Blue was studied in a small group of animals and the results are given in Table 2B. Only one dose of angiotensin II was given, 150 ng kg-1 min-'. Angiotensin II at 150 ng kg-1 min-1 increased blood pressure by mmhg (P < 0-05) and decreased cortical perfusion by 20 % (P < 0 01) but had no effect on either papillary perfusion or CMBC in either cortex or papilla. The average responses over each of the three doses of Methylene Blue were calculated. Administration of Methylene Blue itself had no effect on the basal levels of any of the measured variables and while the average responses over the three doses of

8 794 C. HUANG, G. DAVIS AND E. J. JOHNS Change in systemic 15 - blood pressure (mmhg) [Angiotensin II] (ng kg-l min-') C Change in cortical perfusion Change in papillary 0 perfusion _ L Fig. 3. The absolute changes in systemic blood pressure, cortical and papillary perfusion (expressed as perfusion units, PU) during angiotensin II infusion at 45 and 150 ng kg-' min-' during controlled renal perfusion pressure (O; group IV rats) and when renal perfusion pressure was allowed to rise to systemic levels during 150 ng kg-' min-m angiotensin II infusion (3). Methylene Blue for blood pressure were not significantly different from that obtained before the compound was given, the fall in cortical perfusion was significantly (P < 0 01) attenuated. In the presence of Methylene Blue, angiotensin II had no effect on papillary perfusion or CMBC in either cortex or papilla. A dose-response curve to phenylephrine was drawn up for blood pressure and renal haemodynamics and the results are presented in Table 2C. Phenylephrine, 5, 10 and 20,ug kg-1 min-', caused increases in blood pressure, giving a gradient of f13 mmhg (phenylephrine jtg kg-1 min-1)-1 (P < 0 01) with a rise of mmhg at the highest dose, and decreases in cortical perfusion with a gradient of PU (,ug phenylephrine kg-' min-1)-' (P < 0 01) with a maximum reduction of 7 %. It also caused a reduction in papillary perfusion, with a gradient of PU (,ug phenylephrine kg-' min-1)-' (P < 0 05) with a fall of 17 % at the highest dose of phenylephrine. Neither cortical nor papillary CMBC changed during the phenylephrine infusion. Figure 4 shows the absolute responses of these variables to phenylephrine.

9 ANGIOTENSIN II AND RENAL HAEMODYNAMICS Change in systemic 30 blood pressure (mmhg) 20 []T Fig Change in cortical -10 perfusion Change in papillary -20 perfusion _ [Phenylephrine] (ug kg-' min--) 0 - The absolute changes in systemic blood pressure, cortical and papillary perfusion (expressed as perfusion units, PU) during phenylephrine infusion at 5, 10 and 20,tg kg-1 min-1 (group VI rats). DISCUSSION In a previous report (Davis & Johns, 1990) we observed that vasopressin at doses which increased blood pressure decreased perfusion at both cortex and papilla, whereas angiotensin II, which caused similar dose-dependent increases in blood pressure, decreased cortical perfusion but had no effect on perfusion through the papilla. The intention of the present study was to explore potential mechanisms for this differential action of angiotensin II on regional haemodynamics. A laser-doppler flowmeter was used, as its applicability for use in the kidney has been rigorously validated at the renal cortex by Smits et al. (1986) using electromagnetic flowmeter measurements, and at the renal papilla by Roman & Smits (1986) using 51Cr-labelled erythrocyte accumulation in this region. The control levels for cortical and papillary perfusions in the various groups of animals reported herein are closely comparable to those obtained in previous work by ourselves (Davis & Johns, 1990; Huang, Davis & Johns, 1991) and others (Roman, Cowley, Garcia-Estan & Lombard, 1988). The experimental design was such that angiotensin II was given systemically at rates which caused small to modest (approximately 25 mmhg), increases in blood pressure. There were concomitant dose-related decreases in cortical perfusion, reaching a maximum of about 25 % with the largest dose of angiotensin II, which was very similar to the

10 796 C. HUANG, G. DAVIS AND E. J. JOHNS magnitude and pattern of that reported previously (Davis & Johns, 1990). The most striking observation was that, over the whole of this dose range of angiotensin II, papillary perfusion did not change. Comparable observations were reported by Cupples, Sakai & Marsh, (1988), using dual slit videomicroscopy in the rat, who found that systemic infusion of angiotensin II at doses which reduced total renal blood flow by 25 % (equivalent to 63 ng kg-1 min-1) did not alter vasa recta red cell velocity, and a preliminary report by Mathson, Raff & Roman (1988) using laser-doppler flowmetry found that angiotensin II infusion into the rat at 20 ng kg-1 min-1 failed to affect papillary perfusion. Together these two reports, one of which used a different methodology, provide supportive evidence for our own findings that angiotensin II has a minimal effect on perfusion of blood through the papilla, at least in the rat. By contrast, Faubert, Chou & Porush (1987), using the dog, found that direct intra-renal perfusion of angiotensin II, which had no effect on total renal blood flow or glomerular filtration rate, decreased [125I]albumin accumulation by 66%, while Carmines & Navar (1989), using videomicroscopic measurements in the rat with topical application of angiotensin II, demonstrated both afferent and efferent arteriolar constriction of juxtamedullary glomeruli. The reason for these conflicting responses is not clear but may reside in the different nature of the experimental measurements employed and the high local concentrations of angiotensin II achieved by these workers. An attempt has been made to define the mechanism(s) which could underlie the inability of angiotensin II to influence papillary haemodynamics, possibly by the generation of a vasodilator compound locally. Cortical constriction could have generated prostaglandin production (Cooper, Shaffer & Malik, 1985), could have stimulated renal bradykinin production (Chou et al. 1990) or involved the production of endothelium-derived relaxing factor (Furchgott & Vanhoutte, 1989), each of which could have acted on vessels regulating papillary perfusion to overcome any potential vasoconstrictor action of angiotensin II. Administration of blocking doses of indomethacin (Roman, Kauker, Terragno & Wong, 1978), a bradykinin antagonist (Roman, Kaldunski, Scicli & Carretero, 1988) or Methylene Blue (Martin, Villani, Jothianandan & Furchgott, 1985) caused small reductions or no change, respectively, in basal levels of papillary perfusion, but had no effect on either the cortical or papillary perfusion responses to angiotensin II. The conclusion from these studies was that neither prostaglandins, bradykinin nor endothelium-derived relaxing factors were involved in mediating the angiotensin II responses in perfusion at the cortex and papilla. Administration of angiotensin II was associated with rises in blood pressure which itself would have called into play autoregulatory myogenic responses of the renal vasculature which could have overridden a potential vasoconstrictor action to decrease papillary perfusion. However, the magnitude of the angiotensin II-induced decreases in cortical perfusion when renal perfusion pressure was servo controlled was identical to that observed in the group in which renal perfusion pressure rose with systemic pressure. Further, cortical perfusion remained unchanged when the aortic constriction was released and the pressure at the kidney was allowed to increase to systemic levels during infusion of the highest dose of angiotensin II. Once more, papillary perfusion was unaffected by angiotensin II and was in no way dependent upon whether perfusion pressure at the kidney was controlled at preangiotensin II levels or allowed to rise to the elevated systemic pressure. A final study was undertaken to investigate whether papillary perfusion was sensitive to other vasoconstrictor agents. The a1-adrenoceptor agonist phenylephrine was used, as this is known to cause vasoconstriction in many vascular beds. Phenylephrine induced doserelated increases in blood pressure and decreases in both cortical and papillary perfusions.

11 ANGIOTENSIN II AND RENAL HAEMODYNAMICS 797 These responses clearly demonstrated that these two regions of the kidney vasculature were sensitive to vasoconstrictor peptides. Indeed, this supports our previous observation with another peptide vasoconstrictor, vasopressin (Huang et al. 1991) in which both cortical and papillary perfusions were sensitive to and decreased by vasopressin. This study aimed to describe the action of vasopressor doses of angiotensin II on perfusion in the renal cortex and papilla. Whereas angiotensin II caused dose-related reductions in cortical perfusion, papillary perfusion did not change. This selective pattern of renal responses to angiotensin II did not appear to depend on the generation of prostaglandins, bradykinin or endothelium-derived relaxing factor or the level of renal perfusion pressure. This pattern of haemodynamic changes appeared to be a characteristic of angiotensin II, as vasoactive agents such as phenylephrine and vasopressin appeared equally effective at both cortex and papilla. The mechanisms underlying this selective action of angiotensin II on renal haemodynamics remains to be elucidated. The financial support of Roche Products Limited is gratefully acknowledged. C. H. is a scholar funded by the Chinese Education Commission. REFERENCES CARMINES, P. K. & NAVAR, L. G. (1989). Disparate effects of Ca channel blockade on afferent and efferent arteriolar responses to angiotensin II. American Journal of Physiology 256, F CHOU, S.-Y., PORUSH, J. G. & FAUBERT, P. F. (1990). Renal medullary circulation: hormonal control. Kidney International 37, COOPER, C. L., SHAFFER, J. E. & MALIK, K. U. (1985). Mechanism of action of angiotensin II and bradykinin on prostaglandin synthesis and vascular tone in the isolated rat kidney. Effect of Ca++ antagonists and calmodulin inhibitors. Circulation Research 56, CUPPLES, W. A., SAKAI, T. & MARSH, D. J. (1988). Angiotensin II and prostaglandins in control of vasa recta blood flow. American Journal of Physiology 254, F DAVIS, G. & JoHNs, E. J. (1990). The effects of angiotensin II and vasopressin on renal haemodynamics. Journal of Medical Engineering and Technology 14, DAVIS, G. & JOHNs, E. J. (1991). Effect of somatic nerve stimulation on the kidney in intact, vagotomized and carotid sinus-denervated rats. Journal of Physiology 432, FAUBERT, P. F., CHOU, S.-Y. & PORUSH, J. G. (1987). Regulation of papillary plasma flow by angiotensin II. Kidney International 32, FURCHGOTT, R. F. & VANHOUTTE, P. M. (1989). Endothelium derived relaxing and contracting factors. FASEB Journal 3, HUANG, C., DAVIS, G. & JOHNs, E. J. (1991). An investigation into the influence of vasopressin on perfusion of the cortex and papilla of the rat kidney. Journal of Experimental Physiology 76, JAMISON, R. L. & KRIZ, W. (1982). Urinary Concentrating Mechanism: Structure and Function. Oxford University Press, New York and Oxford. JOHNS, E. J. (1989). Role of angiotensin and the sympathetic nervous system in the control of renal function. Journal of Hypertension 7, MARTIN, W., VILLANI, G. M., JOTHIANANDAN, D. & FURCHGOTT, R. F. (1985). Selective blockade of endothelium-dependent and glyceryl trinitrate-induced relaxation by haemoglobin and by methylene blue in the rabbit aorta. Journal of Pharmacology and Experimental Therapeutics 232, MATHSON, D. L., RAFF, H. & ROMAN, R. J. (1988). Effect of captopril and angiotensin II (AII) on cortical and papillary blood flow in rats. FASEB Journal 2, A5674. NILSSON, G. E. (1990). Perimed's LDV flowmeter. In Laser-Doppler Blood Flowmetry, chap. 4, ed. SHEPHERD, A. P. & OBERG, T. A., pp Kluwer Academic Publishers, Boston, Dordrecht & London. PALLONE, T. L., ROBERTSON, C. R. & JAMISON, R. L. (1990). Renal medullary microcirculation. Physiological Reviews 70, EPH 76 28E8 7

12 798 C. HUANG, G. DAVIS AND E. J. JOHNS ROMAN, R. G., COWLEY, A. W., GARCIA-ESTAN, J. & LOMBARD, J. H. (1988). Pressure diuresis in volume expanded rats. Hypertension 12, ROMAN, R. J., KALDUNSKI, M. L., SCICLI, A. G. & CARRETERO, 0. A. (1988). Influence of kinins and angiotensin II on the regulation of papillary blood flow. Anmerican Journal of Phy'siologv 255, F ROMAN, R. J., KAUKER, M. L., TERRAGNO, N. A. & WONG, P. Y.-K. (1978). Inhibition of renal prostaglandin synthesis and metabolism by indomethacin in rats. Proceedings of the Society for Experimental Biology and Medicine 159, ROMAN, R. J. & SMITS, G. J. (1986). Laser-Doppler determination of papillary blood flow in young and adult rats. American Journal of Physiology 251, F SMITS, G. J., ROMAN, R. J. & LOMBARD, J. H. (1986). Evaluation of laser-doppler flowmetry as a measure of tissue blood flow. Journal of Applied Physiology 61, ZIMMERHACKL, B., ROBERTSON, C. R. & JAMISON, R. L. (1985). The microcirculation of the renal medulla. Circulation Research 57,