| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Howard Florey Institute of Experimental Physiology and Medicine, University of Melbourne, Parkville 3052, Australia
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Glucocorticoids increase renal blood flow
(RBF) and glomerular filtration rate, but the mechanisms are unclear.
We investigated whether the cortisol-induced increment in RBF is a
direct renal action or secondary to its systemic effects and whether
nitric oxide (NO) plays a role in this response. In conscious sheep, cortisol infused intravenously (5.0 mg/h) or into the
renal artery (1.3 mg/h) for 5 h increased RBF by 66 ± 8 and
53 ± 11 ml/min, respectively. Plasma glucose was
increased by intravenous cortisol (0.4 ± 0.1 mmol/l) but not by
intrarenal cortisol. Renal vein plasma cortisol levels were similar at
the end of each infusion (193 ± 31 intravenously; 151 ± 25 nmol/l intrarenal), but systemic levels were different (277 ± 31 intravenous; 69 ± 10 nmol/l intrarenal). Inhibition of NO synthesis
by N
-nitro-L-arginine infused
intravenously (10 mg/kg followed by 5 mg · kg
1 · h1)
or intrarenally (2 mg · kg1 · h1)
significantly reduced the cortisol-induced renal vasodilatation. In
contrast, constriction of the renal vasculature with intrarenal angiotensin (0.3 µg/h) did not prevent the cortisol-induced renal vasodilatation. These findings demonstrate that cortisol acts directly
on the kidney to cause renal vasodilatation and to increase RBF and
suggest that this response involves the endothelium-derived relaxing
factor NO.
mean arterial pressure; N-nitro-L-arginine
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
ADRENAL INSUFFICIENCY in the chronic adrenalectomized animal or in patients with Addison's disease results in a reduction in glomerular filtration rate (GFR) and renal blood flow (RBF), and replacement of glucocorticoids returns GFR and RBF toward normal (8, 15, 27). Increasing the circulating levels of glucocorticoids in intact animals, either by intravenous infusion of steroids or by stimulation of glucocorticoid release from the adrenal with corticotropin, also results in increases in GFR and RBF (4, 8, 9, 19, 33). This vasodilator action of glucocorticoids is selective for the kidney; infusion of cortisol does not cause vasodilatation in the coronary, mesenteric, or iliac vascular beds (18).
These earlier studies indicate that endogenous glucocorticoids play an important role in the maintenance of RBF and GFR, but there are only a few studies that have investigated this renal action of glucocorticoids, and the mechanisms of action remain obscure. Micropuncture studies have shown that in the normal rat kidney glucocorticoids cause vasodilatation of the afferent and efferent arterioles, resulting in an increase in glomerular plasma flow that is the sole factor accounting for the increase in GFR (4). It has been suggested that expansion of plasma volume, due to renal sodium retention induced by glucocorticoids, could lead to an increase in GFR (33). This explanation seems unlikely because glucocorticoids increase GFR in situations where blood volume is not increased (9, 13). A second possibility is that the renal hemodynamic effects of glucocorticoids are secondary to their metabolic actions. Glucocorticoids increase plasma glucose and amino acids (3), and infusions of amino acids or glucose cause renal vasodilatation and an increase in GFR (7, 16), but whether these actions mediate glucocorticoid-induced renal vasodilatation has not been investigated. A third possibility is that glucocorticoids have a direct action on the renal vasculature or alter the responsiveness to other vasoactive agents (1). The interactions with angiotensin and prostaglandins have been investigated, and at present there is no consensus as to whether reductions in sensitivity to angiotensin or altered prostaglandin production accounts for the increase in GFR induced by glucocorticoids (see Ref. 5 for review).
Recent studies in several species have demonstrated that the
endothelium-derived vasorelaxing factor nitric oxide (NO) plays a key
role in the regulation of renal hemodynamics (22, 29, 31). Inhibition
of NO synthesis by administration of various L-arginine analogs, such as
N-monomethyl-L-arginine (10) or
N-nitro-L-arginine
(L-NNA) (31), results in renal
vasoconstriction and a decrease in RBF, whereas infusion of agents,
such as acetylcholine or bradykinin, which cause NO release, results in
an increase in RBF (20). In addition, several isoforms of NO synthase
(NOS) have been localized in sites in the kidney where locally produced NO may act to regulate renal hemodynamics. Although an effect of
glucocorticoids on the constitutive form of NOS has not been demonstrated in aortic endothelial cells (21, 23), this may not reflect
the situation in the kidney, which is the only organ to show the
vasodilatory response to cortisol (18). These findings indicate that
basal release of NO plays an important role in the control of renal
vascular tone and raise the possibility that NO could mediate
glucocorticoid-induced renal vasodilatation.
The present study was designed to investigate whether the renal vasodilator action of cortisol was due to a direct action on the kidney by comparing the responses to intravenous and intrarenal infusions in conscious sheep. The dose of cortisol infused into the renal artery was chosen to produce similar intrarenal cortisol levels but significantly lower systemic levels than the intravenous cortisol infusion. The involvement of NO in this cortisol-induced renal vasodilatation was investigated by studying the effect of cortisol in the presence of a blocker of NO synthesis.
| |
METHODS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
General Methods
Mature merino-cross ewes were used in all experiments and were individually housed in metabolism cages. Access to water was allowed ad libitum, and 800 g of oaten chaff was offered daily (containing 90-120 mmol/kg of Na+ and 270-380 mmol/kg of K+). Sheep were surgically prepared in two stages. In each stage anesthesia was induced with thiopentone sodium (0.4 ml/kg) and was maintained with 1.5-2.0% halothane-O2 mixture. The first stage involved oophorectomy, uninephrectomy, and construction of external carotid artery loops. Oophorectomy is routinely performed on all sheep to prevent any influence of ovarian hormones on fluid balance, and uninephrectomy is performed to prevent any confounding effects of the contralateral kidney. Three to four weeks later a transit-time flow probe (4 mm, Transonic Systems) was implanted on the renal artery of the remaining kidney (6), and Silastic cannulas (0.64 mm ID, 1.19 mm OD) were inserted into the renal artery and the renal vein. Heparinized saline (25 U heparin/ml) was infused at 3 ml/h to maintain patency. Cannulas and flow probe leads were protected by a jacket worn by the animal. Animals were allowed 2 wk of recovery after surgery before experiments commenced.Arterial pressure was measured via a Tygon cannula (1.0 mm ID, 1.5 mm OD) inserted 15 cm into a carotid artery loop. Patency was maintained by infusing heparinized saline (25 U/ml) at 3 ml/h. The cannula was connected to a pressure transducer (TDXIII, Cobe) (6) tied to the wool on the sheep's back. Two polyethylene cannulas (1.20 mm ID, 1.70 mm OD) were inserted into the jugular vein for intravenous infusion. The signal from the pressure transducer was amplified and calibrated daily against a mercury manometer. The pressure was corrected to compensate for the height of the transducer above the level of the heart. Heart level was taken as 64% of the distance from the back to the sternum, which is the level of the junction of the left atrium with the left atrial appendage (6).
Renal blood flow was measured with a transit-time flow probe connected to a transit-time flowmeter (T201CDS, Transonics Systems) either directly or via a four-probe sequential scanner (TM04, Transonic Systems). Mean arterial pressure (MAP), heart rate (HR), RBF, and renal conductance (RC = RBF/MAP) were monitored using a personal computer 486 data-acquisition system with custom-written software (6). After analog-to-digital conversion (DT 2811 Board, Data Translation) the data were collected at 100 Hz for 10 s at intervals of 5 min.
Experimental Methods
Comparison of the responses to intravenous and intrarenal infusions of cortisol. In six conscious sheep, RBF, RC (RC = RBF/MAP), MAP, and HR were recorded every 5 min. After a 1-h control period, cortisol (5 mg/h iv or 1.3 mg/h into the renal artery) or vehicle (5% dextrose containing 4% ethanol at 12 ml/h iv; 5% dextrose containing 1% ethanol at 12 ml/h ir) was infused for 5 h. The intravenous dose of cortisol reproduces the plasma cortisol levels seen during intravenous infusion of corticotropin (5 µg · kg1 · h1)
(11), and this dose of cortisol has been shown to cause selective renal vasodilatation (18). The intrarenal dose was calculated to
perfuse the kidney with the same concentration of cortisol as the
intravenous cortisol infusion. The calculation took into account the
renal blood flow and the contribution from recirculated steroid
calculated from the blood clearance rate (30). Blood samples from the
carotid artery and renal vein were collected hourly for measurement of
hematocrit and plasma levels of sodium, potassium, osmolality, total
protein, glucose, and cortisol.
Dose response to intravenous cortisol. To determine the dose response to intravenous cortisol, either vehicle (12 ml/h) or cortisol (1.0, 2.5, and 5.0 mg/h) was infused for 5 h; each dose was given on a separate day to four conscious sheep. Cortisol was made up from a stock solution in ethanol (9 mg/ml) and diluted in 5% dextrose before intravenous infusion, giving a final concentration of ethanol of 4%. During a 1-h control period and for the 5-h infusion periods, MAP, RC, and RBF were recorded every 5 min on computer. To determine whether the renal vasodilator and the hyperglycemic effects of cortisol could be disassociated, arterial blood samples were collected for measurement of plasma glucose levels before and at the end of each infusion. In addition, these data were used to compare the responses to intrarenal cortisol and a low intravenous dose of cortisol to determine whether spillover of cortisol from the kidney during the intrarenal infusion could account for the response.
Effect of intravenous L-NNA on
the renal hemodynamic response to intravenous cortisol.
After a 1-h control period L-NNA
(10 mg/kg) dissolved in normal saline (50 ml) was given as
an intravenous bolus to six conscious sheep followed by an intravenous
infusion of L-NNA (5 mg · kg1 · h1).
One hour later, cortisol (5 mg/h iv) or vehicle (5% dextrose containing 4% ethanol at 12 ml/h) was infused intravenously for 5 h.
Measurements of RBF, RC, MAP, and HR were made every 5 min and averaged
over a 1-h period. Treatments were given in random order, and 1 wk was
allowed between experiments in which
L-NNA was administered.
Effect of intrarenal L-NNA on the
renal hemodynamic response to intrarenal cortisol.
To reduce the systemic actions of intravenous
L-NNA and intravenous cortisol,
the effect of lower doses infused directly into the renal artery were
studied. After a 1-h control period,
L-NNA (2.0 mg · kg1 · h1)
or vehicle (normal saline 12 ml/h) was infused directly into the renal
artery in five conscious sheep.
L-NNA was infused for 3 h before
the start of the cortisol infusion. Cortisol (1.3 mg/h) or vehicle
(dextrose containing 1% ethanol, 12 ml/h) was infused directly into
the renal artery for 5 h together with
L-NNA. Measurements of RBF, RC,
MAP, and HR were made every 5 min and averaged over a 1-h period.
Treatments were given in random order, and 1 wk was allowed between
experiments in which L-NNA was
administered.
Analytic Methods
Analysis of cortisol in blood was carried out by radioimmunoassay (28). Plasma glucose and electrolytes were measured using a Beckman Synchron CX5 (Beckman Instruments, Brea, CA). Osmolality was measured by freezing-point depression using an osmometer (Advanced Instruments, Needham Heights, MA.)Statistical Analysis
Data are presented as means ± SE. Statistical analysis was performed in three steps. First, the effect of infusion of L-NNA or vehicle before the start of the cortisol infusion (from
1 to 0 h for intravenous
infusions and from 3 to 0 h for intrarenal infusions) on RBF,
RC, MAP, and HR was tested for significance by paired
t-tests. Secondly, the effect of
infusion of L-NNA and/or cortisol (5 h) was compared with the pretreatment period (0 h) by
paired t-tests. Finally, the last 5-h
infusion for each treatment (i.e., from the beginning of the cortisol
infusion) was collectively compared using repeated-measures analysis of
variance with the Greenhouse-Geisser correction. Analysis of the
responses to intravenous and intrarenal infusion of cortisol compared
treatment (cortisol vs. vehicle), route of administration (intravenous
vs. intrarenal) and the interaction between treatment and route of
administration. Analysis of the effect of
L-NNA on the responses to
cortisol compared the response to infusions of
L-NNA and cortisol with the
responses to vehicle infusion and the interaction between
L-NNA and cortisol infusions.
Responses were considered significant at the level of
P < 0.05.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Comparison of the Responses to Intravenous and Intrarenal Infusions of Cortisol
Cortisol infused intravenously (5 mg/h; n = 6) or intrarenally (1.3 mg/h; n = 6) caused similar, progressive increases in RBF and RC, but neither treatment had any effect on MAP or heart rate over the 5-h infusion period (Fig. 1). Cortisol infused intravenously caused significant increases in RBF (from 337 ± 19 to 403 ± 15 ml/min; P < 0.001) and RC (from 4.40 ± 0.28 to 5.21 ± 0.28 ml · min1 · mmHg1;
P < 0.001) and were significantly
different (P < 0.001) compared with
intravenous vehicle (Fig. 1, A and
B). With intrarenal infusion of
cortisol (1.3 mg/h) the increases in RBF (from 347 ± 24 to 399 ± 26 ml/min, P < 0.001) and RC
(from 4.51 ± 0.33 to 5.09 ± 0.32 ml · min1 · mmHg1,
P < 0.001) were similar to
the increases with intravenous cortisol and were significantly
different (P < 0.001) from the
changes with intrarenal infusion of vehicle (Fig. 1,
A and
B).
|
Renal vein cortisol levels increased by similar amounts with intravenous cortisol (from 18 ± 3 to 193 ± 31 nmol/l; n = 5; P < 0.05) and intrarenal cortisol (from 15 ± 7 to 151 ± 25 nmol/l; n = 5; P < 0.05) (Table 1). The increase in systemic cortisol levels with intravenous cortisol (20 ± 5 to 277 ± 31 nmol/l; P < 0.05) was significantly greater (P < 0.001) than with intrarenal cortisol (31 ± 9 to 69 ± 10 nmol/l; P < 0.05). Plasma glucose levels increased from 3.6 ± 0.1 to 4.0 ± 0.1 mmol/l (P < 0.001) with intravenous cortisol, but were unchanged with intrarenal cortisol (Table 1). Neither intravenous nor intrarenal infusion of cortisol altered plasma sodium, potassium, osmolality, total protein, or hematocrit (Table 1). Infusion of vehicle via either route had no effect on plasma cortisol or other biochemical variables measured (Table 1).
|
Dose Response to Intravenous Cortisol
At the end of the 5-h infusion of vehicle, RBF had decreased by 20 ± 8 ml/min and RC had decreased by 0.15 ± 0.11 ml · min1 · mmHg1
(Fig. 2). This decline in RBF and RC
reflects the normal diurnal changes seen over this time of day.
Cortisol (1.0, 2.5, and 5.0 mg/h iv) infused for 5 h significantly
increased RBF and RC (Fig. 2). Plasma glucose levels were unchanged
with vehicle or the low dose of cortisol (1 mg/h), but increased from
3.8 ± 0.1 to 4.3 ± 0.3 and from 3.9 ± 0.1 to 4.5 ± 0.3 mmol/l with the 2.5 and 5.0 mg/h infusions of cortisol, respectively.
There were no changes in MAP or HR with any of the infusions.
|
Effect of Intravenous L-NNA on the Renal Hemodynamic Response to Intravenous Cortisol
There was no significant difference in the control values of RBF, RC, MAP, or HR for the four different treatments. Administration of L-NNA, given intravenously as a bolus followed by continuous infusion, increased MAP from 81 ± 4 to 96 ± 4 mmHg (P < 0.05) and decreased HR from 66 ± 3 to 55 ± 2 beats/min (P < 0.05) over the first hour (Fig. 3, C and D). Control levels of RBF and RC were 357 ± 43 ml/min and 4.55 ± 0.64 ml · min1 · mmHg1,
and after 1 h of intravenous
L-NNA they were 291 ± 24 ml/min (not significant) and 3.12 ± 0.35 ml · min1 · mmHg1
(P < 0.05), respectively (Fig. 3,
A and
B). During the remainder of the 5-h
infusion period over which L-NNA
and vehicle were infused, there were no significant changes in MAP, HR,
RBF, or RC compared with control values (0 vs. 5 h) or when compared
with the change during vehicle alone (Fig. 3). Cortisol treatment did
not alter the changes in MAP and HR produced in response to
L-NNA.
|
The cortisol-induced renal vasodilatation was inhibited by treatment
with L-NNA (Fig. 3). Cortisol
increased RC by 0.81 ± 0.07 ml · min1 · mmHg1
(P < 0.001) over 5 h (Fig.
3B), and the response was
significantly different when compared with that to vehicle alone
(0.25 ± 0.23 ml · min1 · mmHg1,
P < 0.001). In contrast,
simultaneous infusion of cortisol and L-NNA did not change RC over 5 h
(0.08 ± 0.27 ml · min1 · mmHg1),
and this interaction was significant
(P < 0.05). Infusion of L-NNA alone did not change RC
(0.07 ± 0.20 ml · min1 · mmHg1),
and the response was not different to that of vehicle alone. Cortisol
infused intravenously significantly increased RBF by 66 ± 8 ml/min
(P < 0.001) (Fig.
3A), and when cortisol was infused with L-NNA there was no
significant increase in RBF (10 ± 19 ml/min); however, there was no
significant interaction.
Effect of Intrarenal L-NNA on the Renal Hemodynamic Response to Intrarenal Cortisol
L-NNA infused directly into the renal artery at 2.0 mg · kg1 · h1
significantly reduced RBF from 352 ± 46 to 298 ± 37 ml/min
(P < 0.05) and RC from 4.54 ± 0.73 to 3.32 ± 0.35 ml · min1 · mmHg1
(P < 0.05) after 3 h infusion (Fig.
4, A and
B). Initially, MAP was 80 ± 4 mmHg and HR was 57 ± 2 beats/min, and after 3 h of L-NNA infusion they were 89 ± 3 ml/min and 50 ± 4 beats/min, respectively. During the
remainder of the 5-h infusion period over which
L-NNA and vehicle were infused,
there were no significant changes in RBF, RC, MAP, or HR compared with
control values (0 vs. 5 h) or compared with vehicle alone (Fig. 4).
|
Infusion of L-NNA into the renal
artery inhibited the renal vasodilatation in response to an intrarenal
infusion of cortisol (1.3 mg/h) (Fig. 4,
A and
B). Intrarenal infusion of cortisol caused a significant increase in RC from 4.90 ± 0.60 to 5.98 ± 0.63 ml · min1 · mmHg1
(P < 0.05) and in RBF from
373 ± 46 to 447 ± 48 ml/min (P < 0.05), and these changes were significantly different from the
response of vehicle infusion. In contrast, intrarenal infusion of
cortisol during treatment with intrarenal
L-NNA did not significantly
increase RC (from 3.91 ± 0.78 to 4.05 ± 0.78 ml · min1 · mmHg1)
over 5 h, and this interaction was significant
(P < 0.05). During this combined
treatment, RBF increased from 326 ± 53 to 364 ± 55 ml/min
(P < 0.05), and there was no
significant interaction. Albeit, the rise in RBF with
L-NNA and cortisol infusion was
blunted compared with the rise in RBF seen with cortisol infusion (37 ± 12 vs. 74 ± 10 ml/min, respectively).
Renal Vascular Response to Cortisol in Kidneys Preconstricted With ANG II
In four sheep, the effect of intravenous infusion of cortisol on kidneys preconstricted with an intrarenal infusion of ANG II was examined. Intrarenal infusion of ANG II decreased RBF from 372 ± 15 to 324 ± 8 ml/min after 1 h (P < 0.05) and to 308 ± 4 ml/min at the end of a further 5-h infusion of ANG II with vehicle (Fig. 5A). During intrarenal ANG II, RC fell from 4.64 ± 0.21 to 4.14 ± 0.19 ml · min1 · mmHg1
(P < 0.05) after 1 h and to 4.07 ± 0.13 ml · min1 · mmHg1
after 5 h (Fig. 5B). Infusion of ANG
II, with or without cortisol, did not alter MAP or HR (Fig. 5,
C and
D).
|
Infusion of cortisol, starting 1 h after the start of intrarenal ANG II
and continuing for 5 h together with ANG II, caused similar increases
in RBF and RC to the changes in normal kidneys with basal vascular
tone. Over the first hour of ANG II, RBF decreased from 379 ± 2 to 339 ± 18 ml/min (P < 0.05), and over 5-h combined infusion of ANG II and cortisol, RBF
increased to 404 ± 13 ml/min (P < 0.05), and this increase was significantly different from the
response of ANG II alone (P < 0.05;
Fig. 5A). During infusion of
cortisol and ANG II, RC increased from 4.38 ± 0.23 to 5.19 ± 0.12 ml · min1 · mmHg1
(P < 0.05), and this increase was
significantly different from the response of ANG II alone
(P < 0.05; Fig.
5B). The increases in RBF and RC to
cortisol in the presence of ANG II were not different from the
increases to cortisol alone, indicating that constriction of the renal
arterioles, per se, does not alter the ability of cortisol to dilate
the renal vasculature.
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Glucocorticoid hormones have been shown to play an important role in the maintenance of normal GFR and RBF in several species, but the mechanisms involved remain unclear. The present studies have examined the vascular responses of the kidney in vivo to infusion of the endogenous glucocorticoid cortisol in conscious sheep. We have demonstrated that cortisol has a direct renal action, resulting in vasodilatation and an increase in RBF. The renal vasodilatation was inhibited by treatment with L-NNA, an NOS inhibitor, suggesting a role for NO in this response.
In the present study, the renal vascular response to intrarenal (1.3 mg/h) and intravenous (5 mg/h) infusions of cortisol were examined. These two treatments produced similar renal vein cortisol levels, but the systemic levels with intrarenal infusion were fourfold lower than with intravenous infusion. The finding that infusion of cortisol into the renal artery caused a similar degree of renal vasodilatation to the higher systemic dose indicates that the response to cortisol resulted from a direct action on the kidney and that the higher systemic levels during intravenous infusion did not contribute to the renal vasodilatation. During intrarenal infusion of cortisol there was a small increase in circulating cortisol, so a systemic component to this response cannot be excluded entirely; however, as discussed below, our findings provide no support for an indirect action.
It has been proposed that this response to cortisol could be secondary to hemodynamic changes or to systemic metabolic actions that result in increased plasma glucose and amino acid levels. However, cortisol did not alter arterial pressure over the infusion period, and cortisol does not alter central venous pressure (18), indicating that the renal vasodilatation is not secondary to hemodynamic actions of cortisol. The present finding that intrarenal infusion of cortisol did not increase plasma glucose levels, but did cause renal vasodilatation, indicates that this does not depend on the hyperglycemic action of cortisol. This is supported by our finding that intravenous infusion of cortisol at 1 mg/h increased RBF and RC but did not increase plasma glucose, indicating that the renal response to cortisol is more sensitive than the hyperglycemic response. Evidence against a role for amino acids in the renal response to cortisol comes from studies in humans in which increases in plasma concentrations of various amino acids are observed only after 24-48 h of exposure to pharmacological doses of glucocorticoids (26, 34), whereas we found that renal vasodilatation occurred within 2-3 h.
The finding that cortisol-induced renal vasodilatation was inhibited by treatment with L-NNA suggests that NO, an endothelium-derived vasodilator, is involved in mediating the renal hemodynamic responses to glucocorticoids. Studies of the effects of NO synthase blockers have demonstrated that the tonic release of NO plays an important role in the control of renal hemodynamics (22, 24, 29). Blockade of NO synthesis decreases RBF and GFR, and there is evidence that this results from vasoconstriction in both the afferent and efferent arterioles (10, 15). This evidence that tonic NO release maintains both afferent and efferent arteriolar tone and that NOS is present in the afferent and efferent arterioles, together with our finding that L-NNA inhibited the renal vasodilatation caused by cortisol, raises the possibility that the renal response to glucocorticoids is mediated in part via NO-mediated vasodilatation of these vessels.
The possible site and mechanism of the interaction between glucocorticoids and NO are unknown. The time course of the cortisol-induced renal vasodilatation suggests that it is a classical genomic response, probably resulting from an action on glucocorticoid receptors, because synthetic glucocorticoids, but not aldosterone, induce a similar response (18, 19) within 5 h. Glucocorticoid binding sites exist throughout the kidney, with the greatest density occurring in the proximal tubule and cortical collecting duct (17). In the normal rat kidney it has been demonstrated that glucocorticoids increase GFR by causing vasodilatation of the afferent and efferent arterioles (4), and although specific glucocorticoid receptors have been found in glomeruli (12), we are unaware of any studies of the distribution of glucocorticoid binding sites in afferent and efferent arterioles. The constitutive isoforms of NOS, including the neuronal isoform (NOS I) and the vascular endothelial isoform (NOS III), have been localized in macula densa cells, endothelium of cortical and medullary vessels, and efferent and afferent endothelium (2, 21). The finding that both glucocorticoids and NO have actions on the afferent and efferent arterioles and the location of NOS in these vessels further suggests that this could be the site of the putative interaction.
Previous studies investigating the effects of glucocorticoids on the activity of NOS in porcine aortic endothelial cells have not demonstrated an interaction with the constitutive form of NO synthase, the form most likely to mediate the vascular effects (21, 23). However, the vasodilator effect of glucocorticoids is specific to the renal vasculature (18), whereas in the whole animal there is evidence that cortisol increases vascular reactivity to pressor agents (1, 32), presumably reflecting actions in vascular beds other than the kidney. Thus it is unlikely that analysis of the action of glucocorticoids on NOS in nonrenal vessels will reflect changes that occur in the renal vasculature.
It could be argued that inhibition of the renal vasodilator action of cortisol by L-NNA results from nonspecific actions, such as the renal vasoconstriction or the increase in arterial pressure. The finding that cortisol caused vasodilatation in kidneys preconstricted with angiotensin indicates that the attenuation of the renal vasodilatation in response to cortisol by inhibition of NOS is not because cortisol is unable to cause vasodilatation in a kidney with preconstricted vessels. It is also clear that inhibition of NOS does not cause a nonspecific attenuation of the action of all vasodilators, as it has been shown that NG-nitro-L-arginine methyl ester does not inhibit the non-endothelium-dependent vasodilatation in response to prostaglandin E2 (27). It is unlikely that the small increase in arterial pressure with intrarenal infusion of L-NNA altered the response to cortisol, because the renal vasodilatation in response to glucocorticoids is not altered in situations in which arterial pressure is increased, for example during infusion of a combination of adrenal steroids or corticotropin (33).
In summary, these studies have examined the renal vasodilator action of cortisol in vivo in conscious animals. By comparing the responses to intravenous and intrarenal infusion of doses of cortisol that produced similar renal vein cortisol levels, it has been demonstrated that cortisol acts directly on the kidney to cause vasodilatation and an increase in RBF. The response to cortisol was antagonized by inhibition of NOS, suggesting that NO is a mediator of this renal hemodynamic response to glucocorticoids.
Perspectives
Glucocorticoids are known to play an important role in the maintenance of normal renal function, as demonstrated by the reduced RBF and GFR in patients with Addison's disease and in animals with adrenal insufficiency. The present findings that inhibition of NOS markedly attenuates the renal vasodilator response to cortisol suggest that this action of glucocorticoids to maintain renal function depends on release of NO. This vasodilator action of glucocorticoids is selective to the renal vasculature, suggesting that glucocorticoids act on specific renal sites such as the afferent and efferent arterioles or the macula densa, sites that contain NOS and play a major role in the control of renal hemodynamics. It is unknown whether glucocorticoids increase NO release by a direct action on the NO synthesis pathway or whether they act indirectly via release of other humoral factors such as renal kinins. An interrelationship between NO and prostaglandins in the control of vascular tone has been demonstrated, and it would be of interest to determine whether NO and prostaglandins act in concert to mediate glucocorticoid-induced renal vasodilatation.| |
ACKNOWLEDGEMENTS |
|---|
The authors acknowledge Rod Patterson and Dr. Simon Potocnik for surgical assistance, Jason Thompson for measurement of plasma cortisol, and Professor J. Ludbrook and Dr. N. Yates for statistical advice.
| |
FOOTNOTES |
|---|
This work was supported by an institute grant to the Howard Florey Institute from the National Health and Medical Research Council.
Address reprint requests to C. May.
Received 3 June 1997; accepted in final form 2 September 1997.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
1.
Altura, B. M.,
B. T. Altura,
and
S. G. Hershey.
Pharmacodynamic actions of corticosteroids on the microcirculation and vascular smooth muscle.
In: Steroids and Shock, edited by T. M. Glenn. Baltimore, MD: University Park Press, 1974, p. 67-88.
2.
Bachman, S.,
H. M. Bosse,
and
P. Mundel.
Topography of nitric oxide synthases by localizing constitutive NO synthases in mammalin kidney.
Am. J. Physiol.
268 (Renal Fluid Electrolyte Physiol. 37):
F885-F898,
1995
3.
Baxter, J. D.,
and
P. H. Forsham.
Tissue effects of glucocorticoids.
Am. J. Med.
53:
573-589,
1972[Medline].
4.
Baylis, C.,
and
B. M. Brenner.
Mechanism of the glucocorticoid-induced increase in glomerular filtration rate.
Am. J. Physiol.
234 (Renal Fluid Electrolyte Physiol. 3):
F166-F170,
1978
5.
Baylis, C.,
R. K. Handa,
and
M. Sorkin.
Steroid reponses along the nephron. Glucocorticoids and control of glomerular filtration rate.
Semin. Nephrol.
10:
320-329,
1990[Medline].
6.
Bednarik, J. A.,
and
C. N. May.
Evaluation of a transit-time system for the chronic measurement of regional blood flow in conscious sheep.
J. Appl. Physiol.
78:
524-530,
1995
7.
Brøchner-Mortensen, J.
The glomerular filtration rate during moderate hyperglycemia in normal man.
Acta Med. Scand.
194:
31-37,
1973.
8.
Davis, J. O.,
and
D. S. Howell.
Comparative effect of ACTH, cortisone and DCA on renal function, electrolyte excretion and water exchange in normal dogs.
Endocrinology
53:
245-255,
1953.
9.
De Bermudez, L.,
and
J. P. Hayslett.
Effect of methylprednisolone on renal function and the zonal distribution of blood flow in the rat.
Circ. Res.
31:
44-52,
1972
10.
Deng, A.,
and
C. Baylis.
Locally produced EDRF controls preglomerular resistance and ultrafiltration coefficient.
Am. J. Physiol.
264 (Renal Fluid Electrolyte Physiol. 33):
F212-F215,
1993
11.
Fan, J. S. K.,
J. P. Coghlan,
D. A. Denton,
C. J. Oddie,
B. A. Scoggins,
and
A. A. Shulkes.
Effect of intravenous infusion of corticosteroids on blood pressure, electrolytes, and water metabolism in sheep.
Am. J. Physiol.
228:
1695-1701,
1975.
12.
Farman, N.,
A. Vandewalle,
and
J. P. Bonvalet.
Autoradiographic study of aldosterone and dexamethasone binding in isolated glomeruli of rabbit kidney.
Am. J. Physiol.
243 (Renal Fluid Electrolyte Physiol. 12):
F235-F242,
1982.
13.
Hall, J. E.,
C. L. Morse,
M. J. Smith,
D. B. Young,
and
A. C. Guyton.
Control of arterial pressure and renal function during glucocorticoid excess in dogs.
Hypertension
2:
139-148,
1980
14.
Ito, S.
Role of nitric oxide in glomerular arterioles and macula densa.
News Physiol. Sci.
9:
115-119,
1994.
15.
Kleeman, C. R.,
J. Levi,
and
O. Better.
Kidney and adrenocortical hormones.
Nephron
15:
261-278,
1975[Medline].
16.
Lee, K. E.,
and
R. A. Summerill.
Glomerular filtration rate following administration of individual amino acids in conscious dogs.
Q. J. Exp. Physiol.
67:
459-465,
1982
17.
Marver, D.
Evidence of corticosteroid action along the nephron.
Am. J. Physiol.
246 (Renal Fluid Electrolyte Physiol. 15):
F111-F123,
1984.
18.
May, C. N.,
and
J. A. Bednarik.
Regional hemodynamic and endocrine effects of aldosterone and cortisol in conscious sheep. Comparison with the effects of corticotropin.
Hypertension
26:
294-300,
1995
19.
McDougall, J. G.,
A. Butkus,
J. P. Coghlan,
D. A. Denton,
P. J. Meehan,
S. T. Potocnik,
B. A. Scoggins,
and
R. D. Wright.
Dissociation of the biological effects and classical receptor binding of C-21 steroids in the kidney.
In: The Adrenal Gland and Hypertension, edited by F. Mantero,
E. G. Biglieri,
J. W. Funder,
and B. A. Scoggins. New York: Raven, 1985, p. 303-306.
20.
Moncada, S.,
R. M. J. Palmer,
and
E. A. Higgs.
Nitric oxide: physiology, pathophysiology, and pharmacology.
Pharmacol. Rev.
43:
109-142,
1991[Medline].
21.
Palmer, R. M. J.,
L. Bridge,
N. A. Foxwell,
and
S. Moncada.
The role of nitric oxide in endothelial cell damage and its inhibition by glucocorticoids.
Br. J. Pharmacol.
105:
11-12,
1992[Medline].
22.
Radermacer, J.,
U. Förstermann,
and
J. C. Frölich.
Endothelium-derived relaxing factor influences renal vascular resistance.
Am. J. Physiol.
259 (Renal Fluid Electrolyte Physiol. 28):
F9-F17,
1990
23.
Radomski, M. W.,
R. M. J. Palmer,
and
S. Moncada.
Glucocorticoids inhibit the expression of an inducible but not the constitutive nitric oxide synthase in vascular endothelial cells.
Proc. Natl. Acad. Sci. USA
87:
10043-10047,
1990
24.
Raij, L.,
and
C. Baylis.
Glomerular actions of nitric oxide.
Kidney Int.
48:
20-32,
1995[Medline].
25.
Salazar, F. J.,
A. Alberola,
T. Nakamura,
and
J. P. Granger.
Role of nitric oxide in the renal hemodynamic response to a meat meal.
Am. J. Physiol.
267 (Regulatory Integrative Comp. Physiol. 36):
R1050-R1055,
1994
26.
Simmons, P. S.,
J. M. Miles,
J. E. Gerich,
and
M. W. Haymond.
Increased proteolysis. An effect of increases in plasma cortisol within the physiological range.
J. Clin. Invest.
73:
412-420,
1984.
27.
Stanton, B.,
G. Giebisch,
G. Klein-Robbenhaar,
J. Wade,
and
R. A. DeFronzo.
Effects of adrenalectomy and chronic adrenal corticosteroid replacement on potassium transport in rat kidney.
J. Clin. Invest.
75:
1317-1326,
1985.
28.
Tangalakis, K.,
F. E. Roberts,
and
E. M. Wintour.
The time-course of ACTH stimulation of cortisol synthesis by the immature ovine foetal adrenal gland.
J. Steroid Biochem. Mol. Biol.
42:
527-532,
1992[Medline].
29.
Tolins, J. P.,
R. M. J. Palmer,
S. Moncada,
and
L. Raij.
Role of endothelium-derived relaxing factor (EDRF) in regulation of renal hemodynamics.
Am. J. Physiol.
258 (Heart Circ. Physiol. 27):
H655-H662,
1990
30.
Tresham, J.,
J. P. Coghlan,
and
C. N. May.
Evidence against a central pressor mechanism for adrenocortical steroid hypertension in sheep.
Clin. Exp. Hypertens.
18:
831-849,
1996.
31.
Tresham, J. J.,
E. J. Cooper,
J. A. Bednarik,
G. J. Dusting,
and
C. N. May.
Prolonged regional vasoconstriction produced by N-nitro-L-arginine in conscious sheep.
J. Cardiovasc. Pharmacol.
24:
144-150,
1994[Medline].
32.
Whitworth, J. A.,
J. M. C. Connell,
A. F. Lever,
and
R. Fraser.
Pressor responsiveness in steroid induced hypertension in man.
Clin. Exp. Pharmacol. Physiol.
13:
353-358,
1986[Medline].
33.
Whitworth, J. A.,
J. P. Coghlan,
D. A. Denton,
J. S. K. Fan,
T. S. Humphery,
J. G. McDougall,
and
B. A. Scoggins.
Effect of ACTH and adrenal corticosteroids on renal function in sheep.
Renal Physiol.
1:
275-282,
1978.
34.
Wise, J. K.,
R. Hendler,
and
P. Felig.
Influence of glucocorticoids on glucagon secretion and plasma amino acid concentration in man.
J. Clin. Invest.
52:
2774-2782,
1973.
This article has been cited by other articles:
|
|
 |
|
  N. A. Bobadilla, E. Tapia, F. Jimenez, L. G. Sanchez-Lozada, J. Santamaria, A. Monjardin, A. Bolio, G. Gamba, and J. Herrera-Acosta Dexamethasone increases eNOS gene expression and prevents renal vasoconstriction induced by cyclosporin Am J Physiol Renal Physiol, September 1, 1999; 277(3): F464 - F471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. de Matteo and C. N. May Inhibition of prostaglandin and nitric oxide synthesis prevents cortisol-induced renal vasodilatation in sheep Am J Physiol Regulatory Integrative Comp Physiol, April 1, 1999; 276(4): R1125 - R1131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Anwar, M. Schwab, L. Poston, and P. W. Nathanielsz Betamethasone-mediated vascular dysfunction and changes in hematological profile in the ovine fetus Am J Physiol Heart Circ Physiol, April 1, 1999; 276(4): H1137 - H1143. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
), intravenous cortisol 5.0 mg/h (n = 6; 
), intrarenal vehicle
(n = 6; 
), and intrarenal cortisol
1.3 mg/h (n = 6; 
). Infusions
started at time 0 and lasted for 5 h. Statistical
analysis was performed in 2 steps.
1) # Significant effect of
infusion of cortisol or vehicle (0 vs. 5 h;
P < 0.05).
2) Curves were compared using
repeated-measures analysis of variance (ANOVA) for the last 5 h of
infusion. Statistics compared effects of treatment (cortisol vs.
control) and route of administration (intravenous vs. intrarenal).
Significant difference compared with 5 h of vehicle
(P < 0.05). There was no significant
interaction between route and treatment. bpm, Beats/min.
),
and intravenous L-NNA + intravenous cortisol 5.0 mg/h (n = 6;
). Treatment with L-NNA (10 mg/kg bolus and 5 mg · kg
significant
interaction between L-NNA and
cortisol (P < 0.05).
),
intrarenal L-NNA + intrarenal
cortisol 1.3 mg/h (n = 5; 
).
L-NNA or vehicle infusion
started at 