INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE RENAL NERVES extensively innervate both vascular and tubular components of the kidney (3), and functionally they exert an important control over the level of renal blood flow, glomerular filtration rate, renin release, and tubular sodium reabsorption (15). Under normal conditions, renal sympathetic nerve traffic is at such a level as to exert an important control on renin secretion and sodium reabsorption, and, because of the dynamic nature of this neural regulation, it is likely to represent a major way by which fluid homeostasis is achieved (23).

The reflex regulation of renal sympathetic nerve activity is determined by sensory input from a number of systems. A primary source of reflex control of renal nerve activity is via the baroreceptors within both the high (14, 29)- and low (25, 30)-pressure regions of the cardiovascular system. Input from the higher centers is also important, and a number of investigations have shown, in conscious rat studies, that air-jet stress will lead to antinatriuresis and antidiuresis, which are more evident in hypertensive than in normotensive rats (26). The somatosensory system is another major source of afferent information that can modulate the neural control of the kidney. In our earlier studies using electrical stimulation of the brachial nerves in the rat, it was found that this maneuver increased renal nerve activity (13) and caused renin release (18) and a marked renal nerve-dependent antinatriuresis and antidiuresis (11). Importantly, the magnitude of these neural influences on the kidney could be influenced by ongoing activity within the high- and low-pressure baroreceptors of the cardiovascular system (11, 12).

There is good evidence that ANG II can facilitate neurotransmission by a presynaptic action (21) to increase norepinephrine release (5, 32) and can importantly determine the impact of sympathetic nerve traffic. At the level of the kidney, there are also a number of reports demonstrating that blockade of the action of ANG II decreases the magnitude of renal nerve-mediated antinatriuresis and antidiuresis (1, 2, 22, 35). A further dimension exists in that ANG II generated by the compartmentalized brain renin-angiotensin system (8) can exert a modulatory action on sympathetic outflow to the periphery. Thus there are a number of reports in the rat (7) and rabbit (27) that the baroreflex control of heart rate can be modulated by centrally generated ANG II, since after administration of either an angiotensin-converting enzyme inhibitor or an ANG II nonpeptide-receptor antagonist, the sensitivity of the baroreflex curves is enhanced.

The question arises as to whether the peripheral and/or central brain renin-angiotensin systems might be involved in mediating or modulating the somatosensory influences on the neural control of the kidney. This was approached in the present study by reflex activation of the renal nerves with subcutaneous capsaicin (37), blocking of the peripheral renin-angiotensin system by intravenous losartan, and then inhibition of the brain renin-angiotensin system by administration of losartan into the lateral cerebroventricles.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Male Wistar rats, 308 ± 5 g, were initially anesthetized with halothane in O₂-N₂O. The femoral vein was immediately cannulated, and chloralose-urethan was given intravenously at a dose of 12 and 180 mg/kg, respectively, over 45 min, followed by supplementary doses every 30 min. The left carotid and right femoral arteries were cannulated for measurements of systemic blood pressure and renal perfusion pressure. The left kidney was exposed through a flank incision, and its artery was cleared of connective tissue so that an electromagnetic flowmeter probe (Carolina Medical Electronic, King, NC) could be placed around it for the measurement of renal blood flow. The ureters of both the left and right kidneys were cannulated to enable collection of urine samples. A thread was placed around the aorta above the renal artery, which, when tightened against a plastic holder, caused the aorta to be constricted so that renal perfusion pressure could be maintained constant in the face of varying systemic pressures (9). Normal saline (150 mM NaCl) was infused at 3 ml/h as soon as the femoral vein had been cannulated and was continued throughout the experiment. A 2-ml bolus of inulin (15 mg/ml) in saline was given as a primer immediately after the surgery had been completed, followed by an infusion of saline containing 15 mg/ml inulin at 3 ml/h throughout the remainder of the experiment. The animals were allowed 2 h to stabilize before the experimental procedures were begun.

Central administration of drugs was made by injecting them into the right lateral cerebroventricle. For this purpose, a guide cannula, comprising a stainless steel tubing with diameter of 0.82 mm, was placed at the site 1.0 mm posterior to bregma, 2.5 mm lateral to midline, and 2.55 mm ventral to the surface of dura. Another stainless steel tube with diameter of 0.30 mm, being connected to a Hamilton syringe on a microinfusion pump, was inserted into the guide cannula so that it extended 0.2 mm beyond the tip of the guide cannula. In all experiments, Evans blue was injected at the end of the study to check that the fluid had been distributed throughout the ventricular system.

Experimental protocol. A series of six 15-min clearance periods was undertaken. Two clearances were taken to establish control levels. The cutaneous nociceptors were activated with capsaicin, which was given subcutaneously as five 0.1-mg injections every 3 min. The first injection was into the left front paw, and the third clearance period was begun 3 min later so that the urine remaining in the ureteral cannula could be washed out. The remaining doses of capsaicin were given into the left rear paw. Renal perfusion pressure was regulated at the level before capsaicin injection to prevent any change in systemic blood pressure from influencing kidney function. Immediately, a fourth 15-min clearance was taken. Animals were then allowed 30 min to recover from the injection before two recovery clearances were taken.

Group I (n = 7). The procedure described above was undertaken in rats in which the left kidney had an intact innervation.

Group II (n = 6). The procedure in this group was the same as that in group I except that all visible nerves to the left kidney were sectioned during the surgery.

Group III (n = 7). A series of eight 15-min clearance periods was performed in this group. Two hours after the surgery, two clearances were taken before a bolus dose of losartan, an ANG II AT₁ non-peptide-specific antagonist, was given intravenously at 3 mg/kg. Twenty minutes later, a further six clearances were collected in a pattern identical to that in groups I and II.

Group IV (n = 6). This group of animals was subjected to the same protocol as group III except that capsaicin challenge was not given. This represented a losartan time control.

Group V (n = 8). The procedure in this group was the same as that in group III except that losartan was given at 10 mg/kg.

Group VI (n = 8). In this group, saline was given intracerebroventricularly as an initial bolus of 2 µl, followed by a continuous infusion of 1 µl/h throughout the experiment. Capsaicin administration was exactly as for group I.

Group VII (n = 7). In this group, losartan was given intracerebroventricularly, initially as a bolus of 15 µg (in 2 µl of saline), followed by a continuous infusion of 7.5 µg/h (in 1 µl saline); i.e., the animals received a total of 30 µg over the 3-h period of the study. In groups VI and VII, ANG II (100 ng) was given intracerebroventricularly and then intravenously before, 30 min after the start of the intracerebroventricular infusion, and then at the end of the clearance protocol.

Group VIII (n = 6). Animals in this group were given saline intracerebroventricularly as an initial bolus of 2 µl, followed by a continuous infusion of 1 µl/h throughout the experiment. The sequence of clearances was identical to that of groups VI and VII except that capsaicin was not given. These animals acted as time controls.

Statistics. All data represent the average value calculated from individual rats and are expressed as means ± SE. The effect of capsaicin was taken as the difference between the value obtained during capsaicin injection and average value of the two control clearances. The influence of losartan was taken as a difference between the average values immediately before the administration and those obtained 20 min after losartan. Statistical analysis was performed using SuperANOVA Software (Abacus). One-way ANOVA was applied to the data obtained in each group of animals, and a Bonferroni-Dunn post hoc test was used to highlight the individual values that were different. Comparisons between groups were undertaken using a two-way ANOVA to differentiate treatment and time influences. Significance was taken when P < 0.05, and the particular test used is defined in the text.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

The cardiovascular and left renal hemodynamic responses of capsaicin are shown in Table 1, and it can be seen that there was a significant (P < 0.01, 1-way ANOVA) increase in mean arterial pressure of ~10% during the administration of capsaicin but which thereafter fell back to control levels and remained there during the recovery period. During this time, renal perfusion pressure was regulated at an unchanged level, and there were only small inconsistent alterations in renal blood flow and glomerular filtration rate. Figure 1 demonstrates that capsaicin challenge caused significant (P < 0.05 or 0.01, 1-way ANOVA) reductions in left kidney urine flow of 26%, absolute sodium excretion of 37%, and fractional sodium excretion of 36%, but in the subsequent clearance period, the variables had returned to levels not significantly different from control and remained at these values in the recovery periods taken 30 min later.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of capsaicin administration (0.5 mg sc) on urine flow, sodium excretion, and fractional sodium excretion of left kidney. Periods: control, average values of 2 clearances obtained before capsaicin injection; capsaicin, values obtained during capsaicin injection; after capsaicin, values obtained during clearance period after capsaicin injection; recovery, average values of 2 clearances obtained 30 min after capsaicin injection. Compared with control (1-way ANOVA): * P < 0.05; ** P < 0.01.

The second study compared the renal responses to the subcutaneous capsaicin administration in the group of animals subjected to a left denervation with the right kidney remaining innervated. Capsaicin increased blood pressure and had no effect on renal perfusion pressure, renal blood flow, or glomerular filtration rate (Table 2), which was a pattern of responses similar to that of the first group of rats (Table 1). Thereafter (Table 2), blood pressure returned toward control levels, and all variables remained at those values in the recovery period 30 min later. The excretory responses are given in Fig. 2, and it can be seen that the control levels of urine flow and absolute and fractional sodium excretions in the left denervated kidney were some twofold higher than those of the right innervated kidney (P < 0.001, 2-way ANOVA), consistent with a denervation diuresis and natriuresis. It is evident that capsaicin administration had no effect on the excretory variables during, immediately after, or over the recovery periods of the left kidney (Fig. 2). By contrast, in the right kidney, there were significant (P < 0.05 or 0.01, 1-way ANOVA) reductions in urine flow and absolute and fractional sodium excretion of 30-50% both during and immediately after capsaicin administration but had returned to control levels when the recovery measurements were made. The pattern and magnitudes of the responses were markedly different between the two kidneys.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effect of capsaicin administration (0.5 mg sc) on urine flow, sodium excretion, and fractional sodium excretion of both kidneys with left kidney denervated. Periods as in legend to Fig. 1. Open columns, denervated kidney; filled columns, innervated kidney. Compared with control (1-way ANOVA): * P < 0.05; ** P < 0.01.

The influence of losartan (3 mg/kg iv) on both blood pressure and kidney function and the responses to capsaicin challenge are shown in Table 3. Administration of losartan decreased both mean blood pressure and renal perfusion pressure by some 9% (both P < 0.01, 1-way ANOVA), had no effect on renal blood flow or glomerular filtration rate, but was associated with reductions in urine flow and absolute and fractional sodium excretions of 33-41% (P < 0.05 or 0.01, 1-way ANOVA). Capsaicin challenge (Table 3) caused a significant increase in mean arterial pressure of 11%, the magnitude of which could not be distinguished from that obtained when losartan was not present (Table 1), renal perfusion pressure was regulated at an unchanged level, and both renal blood flow and glomerular filtration rate were unaltered during and after capsaicin and tended to be at slightly higher levels in the recovery period. Similarly, urine flow and absolute and fractional sodium excretions were not altered by capsaicin administration and, if anything, reached slightly higher levels after capsaicin, remaining there during the recovery phase. The pattern and magnitudes of these excretory responses were very different (P < 0.001, 2-way ANOVA) from those obtained when losartan was not given (Table 1). In a time control study, losartan (3 mg/kg iv) was given but no capsaicin challenge occurred (Table 4). The losartan decreased blood pressure and renal perfusion pressure by some 11 mmHg (P < 0.05, 1-way ANOVA), associated with falls in urine flow and absolute and fractional sodium excretions. However, all variables remained at these new values for the remainder of the experiment.

A further group of rats, to which losartan was given at 10 mg/kg (Table 5), was studied. This dose of losartan reduced both mean arterial pressure and renal perfusion pressure by 21% (P < 0.01, 1-way ANOVA), had no effect on renal blood flow or glomerular filtration rate, but was associated with marked reductions (P < 0.05, 1-way ANOVA) in urine flow and absolute and fractional sodium excretions of 62, 70, and 63%, respectively. Administration of capsaicin increased mean arterial pressure by ~18% (P < 0.01, 1-way ANOVA), a response comparable in magnitude to that obtained in the absence of losartan (Tables 1 and 2). When renal perfusion pressure was regulated at control levels, there was no change in renal blood flow and a fall in glomerular filtration rate (17%, P < 0.05, 1-way ANOVA), but the inconsistent decreases in the excretory variables did not reach statistical significance. In the period immediately after capsaicin and in the recovery period 30 min later, the cardiovascular and excretory variables were at levels not different from those measured in the control periods.

The data for the group of rats given intracerebroventricular saline as a vehicle are given in Table 6. It can be seen that administration of the bolus of saline plus continuous infusion of saline had no effect on blood pressure, renal perfusion pressure, or any renal hemodynamic or excretory variable. Furthermore, capsaicin challenge increased (P < 0.05, 1-way ANOVA) blood pressure to the same extent as in the control group of rats (Table 1), had no effect on renal blood flow or glomerular filtration rate, and, as shown in Fig. 3, reduced urine flow and absolute and fractional sodium excretions by 29-38% (P < 0.05 or 0.01, 1-way ANOVA), returning toward control levels immediately after capsaicin and during the recovery periods. The magnitude and patterns of these excretory responses were very comparable to those of the control studies (Figs. 1 and 2).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Changes in urine flow, sodium excretion, and fractional sodium excretion in intracerebroventricular (icv) saline-time control group (), icv saline-capsaicin group (), and icv losartan-capsaicin () groups over experimental period. * P < 0.05, comparing saline icv-capsaicin group against icv saline-time control groups.  P < 0.05 comparing icv saline-capsaicin against icv losartan-capsaicin group (2-way ANOVA).

The influence of intracerebroventricular losartan on the hemodynamic responses to capsaicin challenge is given in Table 7 and on excretory responses in Fig. 3. It can be seen that there were small but not statistically significant falls in blood pressure and fluid excretion after the bolus and continuous infusion of losartan (Table 7). The administration of capsaicin increased blood pressure by 15% (P < 0.05, 1-way ANOVA), renal perfusion pressure was regulated at control levels, and, although neither renal blood flow nor glomerular filtration rate was changed during the challenge, renal blood flow increased slightly immediately after capsaicin and remained at this elevated level in the recovery period. Under these conditions, there were only minor nonsignificant changes in urine flow and absolute and fractional sodium excretions in response to capsaicin administration, indicating blockade of capsaicin-induced excretory responses (Fig. 3). The magnitude and pattern of the responses in urine flow and absolute and fractional sodium excretions were significantly different (P < 0.05, 2-way ANOVA) from those obtained when only saline was given intracerebroventricularly (Fig. 3).

A control study in which the animals received saline intracerebroventricularly but were not given capsaicin (group VIII) was undertaken, and it can be seen (Table 8) that over the time course of the study neither blood pressure nor renal perfusion pressure changed, whereas there was a small inconsistent increase in renal blood flow and glomerular filtration rate. Furthermore, urine flow and absolute fractional sodium excretions, shown in Fig. 3, were stable throughout the period of measurement, a finding that was significantly different (P < 0.05, 2-way ANOVA) from the reversible reductions in the variables which occurred when capsaicin was given.

Table 9 (top) shows that 100 ng ANG II given intracerebroventricularly caused a slowly developing increase in blood pressure of some 13 mmHg (P < 0.01, 1-way ANOVA) but had no effect on renal blood flow. The same dose of ANG II given 3 h after the start of the intracerebroventricular saline infusion caused a similar vasopressor response, but, in the group given losartan intracerebroventricularly, the rise in blood pressure was blocked significantly (P < 0.001, 1-way ANOVA) compared with the response at the beginning of the experiment. In Table 9 (bottom), it can be seen that ANG II (100 ng iv) induced a sharp rise in blood pressure and reduction in renal blood flow and that the magnitude of these responses was the same at the end of the study 3 h later regardless of whether losartan or saline was given intracerebroventricuarly.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

It is recognized that the renin-angiotensin system is able to influence cardiovascular homeostasis in a multifactorial way by acting at a number of levels in the brain and at the periphery at both neural and endocrine levels (10). The aim of this study was to examine how information arising from the somatosensory system, which causes a reflex activation of the sympathetic nervous system, could influence the sympathetic neural control of kidney hemodynamic and tubular functions and whether this reflex was in any way dependent on ANG II.

The somatosensory system was activated by giving small doses of capsaicin subcutaneously to selectively stimulate the sensory receptors of the skin (19). This approach differed from that used in previous studies (12, 13) in which the sensory nerves of the brachial plexus were electrically stimulated, having the drawback that a range of sensory fibers was activated, including not only those of the skin but also those of the muscle and joints of the limb, which might have somewhat different effects. Administration of capsaicin in this way led to a vasopressor response of relatively short duration (2-3 min), but to determine an effect on renal function, which was dependent on 15-min urine collection periods, a longer period of sensory activation was required. Therefore the pattern of five small doses of capsaicin given 3 min apart, starting 3 min before the clearance period was begun, was developed in preliminary studies and used in the present investigations. A second feature was that efforts were made to ensure that renal perfusion pressure was maintained at an unchanged value during the experimental challenge, since it is recognized that perfusion pressure, via renal interstitial hydrostatic pressure (17), is a major determinant of sodium and hence water excretion (33).

From these approaches, it was evident from the initial study that, in the period when capsaicin was administered, there was a small vasopressor effect but no measurable effect on either renal blood flow or glomerular filtration rate even though there was a fall in water and sodium excretion of some 30-40%. The vasopressor action was probably due to an increase in vascular resistance in response to a sympathetically mediated vasoconstriction in a number of vascular beds (10, 22). It was of interest that a vasoconstriction was not observed in the kidney, but one possibility is that the powerful autoregulatory responses might have overcome any modest increase in renal vascular tone resulting from activation of the renal nerves. Alternatively, this apparent separation of vascular and tubular responses could reflect a differential central activation of groups of neurons subserving selective functions within the kidney, which would support the histological evidence of Luff et al. (28) and electrophysiological data of DiBona et al. (16). The falls in water and sodium excretions during capsaicin challenge were similar in magnitude to those observed in previous studies using brachial nerve stimulation (11, 12). It was clearly evident that removal of the influence of the nerves in the left kidney resulted in a higher basal level of fluid excretion, a denervation diuresis and natriuresis, which did not change during the period of somatosensory activation. By contrast, in the innervated kidney, basal fluid excretion was much lower and decreased in a reversible fashion during the period of capsaicin challenge. These results have been taken as reflecting the renal sympathetic nerves having a direct action on the tubular epithelial cells at the proximal tubule and thick ascending limb of the loop of Henle, stimulating them to increase sodium reabsorption (15). Moreover, in a recent report, we had shown that administration of capsaicin subcutaneously was able to cause a transient increase in renal sympathetic nerve activity (37).

The role of the renin-angiotensin system in this somatorenal reflex was initially investigated by using losartan, one of the first ANG II nonpeptide-receptor antagonists (36), given intravenously to block the action of ANG II. The two doses of losartan given (3 and 10 mg/kg iv) have been shown by others to block the vasopressor actions of ANG II (36). Administration of the higher dose of losartan was associated with a large fall in blood pressure and water and sodium excretion, even though renal blood flow remained unaltered, which was indicative of efficient autoregulation against this pressure reduction. The decreased fluid excretion during losartan was most probably due to the influence of the large fall in pressure at the kidney acting via the mechanisms described above (33). Preliminary studies showed that the 10-mg/kg dose of losartan achieved a >90% blockade of the vasopressor action of systemically administered ANG II (100 ng iv), which was maintained over the 3-h period of the experiment. Nonetheless, the fact that urine flow and sodium excretion increased so that the recovery values were larger than the control values suggests that some other influences were coming into play to determine the rate of fluid excretion. However, in an attempt to avoid the large falls in blood pressure, a study was also undertaken in which a lower dose of losartan (3 mg/kg iv) was used. Although a smaller fall in blood pressure was observed with the 3-mg/kg intravenous dose of losartan, at the end of the experiment some 3 h later, the vasopressor action of the ANG II was only blocked by 40-50%. It must be recognized that the renal functional measurements took place over 2-3 h, with the recovery measurements only being taken in the last 30 min; it is therefore likely that at the time when capsaicin was given a more complete blockade was probably present.

It was evident that, in the presence of the low dose of losartan, capsaicin-induced renal nerve-dependent antinatriuretic and antidiuretic responses were blocked. A similar situation pertained when the high dose of losartan was used, but in this instance there was a variable nonsignificant fall in fluid output in response to capsaicin with a slight rise in output in the recovery period. Together these observations are consistent with the view that ANG II has a modulatory role at the neuroeffector junctions at the tubular level to facilitate sympathetic transmission (5, 21, 22, 34). There is a body of evidence, drawn from a number of tissues and vascular beds, supporting this contention, but there has been debate as to whether presynaptic or postsynaptic receptors or a combination of influences at both sites are involved (38). At the level of the kidney, there have been relatively few reports of this function of ANG II, but, in early studies using angiotensin-converting enzyme inhibitors or high-salt intake (1, 2, 21) in the rat to block or depress the renin-angiotensin system, the adrenergic influences on sodium handling were inhibited. More recently, Veelken et al. (35), using the conscious rat, were also able to show that air-jet stress caused a reflex renal nerve-dependent antinatriuresis and antidiuresis, which was also blocked by the intravenous administration of an ANG II nonpeptide-receptor antagonist, ZD-7155. They argued that this finding was consistent with the involvement with ANG II receptors at neuroeffector junctions in the kidney. The findings of the present study, in which the somatorenal reflex was found to be dependent on ANG II, parallel the observations obtained in the study of the air-jet stress reflex (30). Nonetheless, other reports examining the reflex neural regulation of sodium reabsorption in the dog after hemorrhage (31) or bilateral carotid artery occlusion (32) found that these responses were not dependent on the presence of ANG II. The reasons for these different findings remain unclear and require further study.

The question that arises is whether ANG II within the brain itself might have had some role to play in modifying this reflex. There is good evidence that there is a compartmentalized and independent renin-angiotensin system within the central nervous system (8) and that ANG II can modulate one reflex, the baroreflex control of heart rate by which it seems to exert a tonic inhibitory action (7, 27). It is apparent that losartan can cross the blood-brain barrier, and it is likely that, at the high but not the low intravenous doses used in the present study, losartan could have blocked an action of ANG II within the central nervous system. Indeed, in a recent report (6), we were able to show that losartan at 10 mg/kg caused a marked alteration of the power spectral pattern within the renal nerve signal, which might have functional implications. The approach taken in the present study was to administer the losartan locally into the lateral cerebroventricles at a dose sufficient to block the vasopressor effect of ANG II within the brain but having no blocking action on systemic ANG II receptors. It was apparent that these criteria were achieved. The data generated under these circumstances quite clearly showed that capsaicin-induced renal nerve-dependent antinatriuresis and antidiuresis were inhibited even though the vasopressor response to capsaicin was unaltered. These findings were reinforced by the time control study in which capsaicin was not given and showed the stability of the experimental preparation.

The maintenance of the vasopressor response to capsaicin suggests that the sympathetically mediated increase in vascular resistance in tissues and organs other than the kidney were only minimally influenced by ANG II in the brain. By contrast, the specific somatorenal reflex, particularly in relation to the renal sympathetic nerve control of sodium reabsorption by the epithelial cells, seemed to a large extent to be dependent on ANG II within the central nervous system. Exactly where and how this action of ANG II might be exerted are unclear at the present time. The components of the renin-angiotensin system have been shown to be present at a number of loci within the brain, particularly at those loci involved with the control of the cardiovascular system (8), and therefore have the potential of modulating the generation of sympathetic outflow. How this action of ANG II might take place is unclear; whether it is at the level of transmission of afferent signals across the central nervous system, in the integrative areas of the hypothalamus, or during onward transmission to the descending pathways remains uncertain at the present time.

This study set out to examine the role of ANG II in the somatosensory reflex. Capsaicin was given subcutaneously to activate sensory receptors in anesthetized rats and resulted in a rise in blood pressure, no change in renal hemodynamics, but reversible reductions in sodium and water excretion. This antinatriuretic and antidiuretic response was shown to be dependent on the renal sympathetic nerves. Administration of low and high doses of losartan systemically to block ANG II receptors had no effect on capsaicin-induced vasopressor response but blocked the reductions in sodium and water output. However, blockade of the brain renin-angiotensin system, by infusion of losartan into the lateral cerebroventricles, also blocked the renal excretory responses to the somatosensory challenge. These data show that ANG II acts at two levels to modulate the somatorenal reflex: one at the periphery, where it is likely to be involved in the classical manner of facilitating neurotransmission at the level of the kidney, and a second at the level of the central nervous system, where it can determine the magnitude of the impact of somatosensory activation on renal sympathetic nerve-dependent sodium and water output. The manner in which this influence of ANG II is exerted within the brain remains to be established.

Perspectives