This study examined the effects of renal arterial infusion of a selective cytochrome P-450 epoxygenase inhibitor, N-methylsulfonyl-6-(2-propargyloxyphenyl)hexanamide (MS-PPOH; 2 mg/kg plus 1.5 mg·kg−1·h−1), on renal hemodynamic responses to infusions of [Phe2,Ile3,Orn8]vasopressin and ANG II into the renal artery of anesthetized rabbits. MS-PPOH did not affect basal renal blood flow (RBF) or cortical or medullary blood flow measured by laser-Doppler flowmetry (CLDF/MLDF). In vehicle-treated rabbits, [Phe2,Ile3,Orn8]vasopressin (30 ng·kg−1·min−1) reduced MLDF by 62 ± 7% but CLDF and RBF were unaltered. In MS-PPOH-treated rabbits, RBF and CLDF fell by 51 ± 8 and 59 ± 13%, respectively, when [Phe2,Ile3,Orn8]vasopressin was infused. MS-PPOH had no significant effects on the MLDF response to [Phe2,Ile3,Orn8]vasopressin (43 ± 9% reduction). ANG II (20 ng·kg−1·min−1) reduced RBF by 45 ± 10% and CLDF by 41 ± 14%, but MLDF was not significantly altered. MS-PPOH did not affect blood flow responses to ANG II. Formation of epoxyeicosatrienoic acids (EETs) and dihydroxyeicosatrienoic acids (DiHETEs) was 49% lower in homogenates prepared from the renal cortex of MS-PPOH-treated rabbits than from vehicle-treated rabbits. MS-PPOH had no effect on the renal formation of 20-hydroxyeicosatetraenoic acid (20-HETE). Incubation of renal cortical homogenates from untreated rabbits with [Phe2,Ile3,Orn8]vasopressin (0.2–20 ng/ml) did not affect formation of EETs, DiHETEs, or 20-HETE. These results do not support a role for de novo EET synthesis in modulating renal hemodynamic responses to ANG II. However, EETs appear to selectively oppose V1-receptor-mediated vasoconstriction in the renal cortex but not in the medullary circulation and contribute to the relative insensitivity of medullary blood flow to V1-receptor activation.
- angiotensin II
- cytochrome P-450 enzyme system
- kidney cortex
- kidney medulla
- renal circulation
the role of the renal medullary circulation in long-term blood pressure regulation is well established. Numerous studies have reported that intrarenal infusion of vasoactive agents that reduce renal medullary blood flow (MBF) increase arterial pressure and that pharmacological agents that increase medullary blood flow reduce arterial pressure in several models of hypertension (10, 11, 25). However, the factors that control MBF in vivo remain to be determined.
There is strong evidence that vasoactive hormones can differentially influence cortical blood flow (CBF) and MBF. For example, MBF appears to be relatively insensitive to the vasoconstrictor effects of ANG II (3, 12, 14, 28, 29 33, 34, 36, 43, 49), whereas vasopressin V1-receptor agonists selectively reduce MBF at doses that have little effect on CBF or total renal blood flow (RBF) (9, 14, 18, 28, 29, 36). The differential effects of ANG II vs. vasopressin in the intrarenal control of blood flow likely underlie important mechanisms in long-term regulation of blood pressure and the control of urinary-concentrating ability, but the precise mechanism underlying these differences is unknown. There is evidence that ANG II stimulates the release of nitric oxide (NO) and prostaglandins in the medullary circulation and that the local release of these vasodilators contributes to the relative insensitivity of MBF to ANG II (3, 11, 29, 33, 34, 36, 43, 49). However, we have recently shown that inhibition of NO and prostaglandins does not contribute to the insensitivity of the renal cortical circulation to the V1 agonist [Phe2,Ile3,Orn8]vasopressin (28, 29, 36). Importantly, under conditions of NO synthase blockade (36), cyclooxygenase blockade (29), lipoxygenase blockade (28), or even combined blockade of all three of these enzyme systems (28), renal arterial administration of ANG II still reduces CBF more than MBF, whereas [Phe2,Ile3,Orn8]vasopressin has the opposite effect and reduces MBF more than CBF.
Another candidate system that may modulate reactivity to vasoactive hormones is the cytochrome P-450 (CYP450)-dependent arachidonic acid (AA) metabolism cascade. CYP450 epoxygenase products of AA have been reported to modulate the renal vascular responses to ANG II (19, 21) and arginine vasopressin (42) in vitro. Furthermore, epoxyeicosatrienoic acids (EETs) are formed in the rabbit kidney (17), and arginine vasopressin stimulates the release of EETs from rabbit renal medullary tissue (39). However, whether EETs modulate the renal vasoconstrictor actions of vasopressin and other hormones in vivo remains to be determined.
The present study examined the hypothesis that the relative insensitivity of MBF to ANG II and the relative insensitivity of CBF to [Phe2,Ile3,Orn8]vasopressin arise from regional differences in the ability of these agents to stimulate the synthesis and/or release of CYP450 epoxygenase metabolites of AA. Therefore, we examined the effects of the selective CYP450 epoxygenase inhibitor N-methylsulfonyl-6-(2-propargyloxyphenyl)hexanamide (MS-PPOH) (4, 16) on the regional renal blood flow responses to infusion of ANG II and [Phe2,Ile3,Orn8]vasopressin directly into the renal artery of anesthetized rabbits. We confirmed the ability of MS-PPOH to selectively reduce epoxygenase activity after in vivo administration by measuring the metabolism of AA by CYP epoxygenase and ω-hydroxylase enzymes in homogenates prepared from the kidneys of vehicle- and MS-PPOH-treated rabbits. The results suggest that in rabbits in vivo, de novo synthesis of EETs does not modulate the renal hemodynamic responses to ANG II but does buffer the vasoconstrictor effects of V1-receptor agonists in the renal cortical circulation.
Experiments were performed on 22 male New Zealand White rabbits (2.73 ± 0.06 kg). Water was provided ad libitum, and they were fed once a day as previously described (15). Experiments were approved by the Monash University Department of Physiology/Central Animal Services Animal Ethics Committee and were performed in accordance with the American Physiological Society “Guiding Principles for Research Involving Animals and Human Beings” (2).
The rabbits were anesthetized with intravenous pentobarbital sodium (90–150 mg plus 30–50 mg/h Nembutal; Rhone Merieux, Pinkenba, QLD, Australia) and this was immediately followed by endotracheal intubation and artificial ventilation (Harvard Rodent Ventilator, model 683, Harvard Apparatus, South Natick, MA). Catheters were placed in both central ear arteries (22 gauge, Insyte; Becton Dickinson, Sandy, UT) and marginal ear veins (24 gauge) for measurement of arterial pressure and intravenous infusions, respectively.
The left kidney was exposed via a left flank retroperitoneal incision and denervated by stripping the nerves from the renal artery and vein. The rabbit was then placed in an upright crouching position, and a catheter (single-lumen PE-10: OD 0.61 × ID 0.28 mm; Datamasters, Ocean Grove, Victoria, Australia) was placed in a side branch (suprarenolumbar artery) of the renal artery for administration of vasoactive agents. Catheter patency was maintained by a continuous infusion of 154 mM NaCl (20 μl·kg−1·min−1). The kidney was placed in a stable cup, and a transit-time ultrasound flow probe (type 2SB; Transonic Systems, Ithaca, NY) was placed around the renal artery to measure total renal blood flow (RBF). CBF and MBF were estimated by laser-Doppler flowmetry (Moor Instruments, Millwey, Devon, UK). Medullary laser-Doppler flux (MLDF) was measured after implanting a needle probe (26 gauge, DP4s) 9 mm below the cortical surface using a micromanipulator (Narishige, Tokyo), A standard plastic straight probe (DP2b) was used to monitor cortical laser-Doppler flux (CLDF). We have previously shown that the percent changes in CLDF to renal arterial infusion of endothelin-1 measured with these two different probes are indistinguishable (22).
The ear artery catheters were connected to pressure transducers to measure arterial pressure (Cobe, Arvarda, CO). Heart rate (HR) was measured using a tachometer activated by the pulse pressure. Throughout the experiment mean arterial pressure (MAP, mmHg), HR (beats/min), RBF (ml/min), CLDF (perfusion units), and MLDF (perfusion units) were recorded at 2-s intervals using a digital recording system.
After surgery and a 30- to 60-min equilibration period, MS-PPOH [2 mg/kg plus 1.5 mg·kg−1·h−1 (4, 16)] or vehicle (45% hydroxypropyl-β-cyclodextran; Sigma Chemical, St Louis, MO) was infused into the renal artery. Thirty minutes later the responses to graded renal arterial infusions of ANG II (2, 6, and 20 ng·kg−1·min−1; 10 min for each dose; Auspep, Parkville, Victoria, Australia) and [Phe2,Ile3,Orn8]vasopressin (3, 10, and 30 ng·kg−1·min−1; 15 min for each dose; Peninsula Laboratories, Belmont, CA) were determined in random order. At the end of each experiment the left kidney was removed, frozen in liquid nitrogen, and stored at −80°C for later measurement of the renal metabolism of AA in vitro.
CYP450 ω-Hydroxylase and Epoxygenase Activity
Microsomes were prepared from the renal cortex and medulla of six untreated rabbits. Briefly, the renal cortex and inner and outer medulla were homogenized in 3 vol of a 10 mM potassium phosphate buffer containing 250 mM sucrose and 1 mM EDTA. The microsomes were prepared by sequential centrifugation of the homogenate (at 3,000 g for 5 min, 11,000 g for 15 min, and 100,000 g for 60 min). Microsomal pellets were resuspended in 100 mM potassium phosphate buffer containing 0.5 mM EDTA, 1 mM DTT, 30% glycerol, and 0.1 mM phenylmethylsulfonyl fluoride.
Renal homogenates rather than microsomes were prepared from the kidneys of the rabbits treated in vivo with MS-PPOH or its vehicle because it remains to be determined whether MS-PPOH acts as a competitive, noncompetitive, or mixed-mode inhibitor of epoxygenase activity. Thus these kidneys were homogenized in only 2 vol of the 10 mM potassium phosphate buffer to minimize dilution of MS-PPOH retained in the tissue and centrifuged at low speed (at 3,000 g for 5 min). An aliquot of the homogenate was taken to determine the CYP450 activity as described below.
The renal metabolism of AA was determined by incubating 0.5 mg of microsomal protein or 2 mg of homogenate protein with [14C]AA (Amersham, Arlington Heights, IL) in 1 ml of a 100 mM potassium phosphate buffer (pH 7.25) containing 10 mM sodium isocitrate, 0.16 U/ml isocitrate dehydrogenase, and 1 mM β-NADPH. The microsome reactions were performed using a saturating concentration of AA (0.1 μCi, 42 μM) whereas the homogenates were studied using a low concentration of substrate (AA, 0.1 μCi, 2 μM) to maximize the conversion rate of labeled AA and minimize competition of MS-PPOH with excess substrate. Incubations were carried out in a shaking water bath at 37°C for 30 min (microsomes) or 60 min (homogenates) in an atmosphere of 100% oxygen. The reaction was terminated with 1 M formic acid, and AA metabolites were extracted with 3 ml of ethyl acetate. The metabolites were separated using reverse-phase HPLC equipped with an online radioactive flow detector as previously described (47). Rates of product formation are expressed as picomoles formed per minute per milligram protein (pmol·min−1·mg−1). The rate of formation of 20-hydroxyeicosatetraenoic acid (20-HETE) provided a measure of ω-hydroxylase activity, while the rate of formation of dihydroxyeicosatrienoic acids (DiHETEs) and EETs provided a measure of epoxygenase activity.
Effects of MS-PPOH In Vivo
Baseline levels of hemodynamic variables.
There were no significant differences in baseline systemic or renal hemodynamics between rabbits subsequently treated with vehicle or MS-PPOH, so these data were pooled. Baseline MAP and HR averaged 77 ± 2 mmHg and 277 ± 3 beats/min, respectively, while RBF, CLDF, and MLDF averaged 31 ± 2 ml/min, 405 ± 25 U, and 63 ± 9 U, respectively. MAP had increased by 13 ± 3%, and HR and RBF had reduced by 6 ± 1 and 10 ± 3%, respectively, 30 min after infusion of vehicle, but CLDF and MLDF were not significantly altered. MS-PPOH treatment had a similar effect, and the changes seen after administration of MS-PPOH were not significantly different from those seen in rabbits given vehicle (Fig. 1).
Responses to renal arterial infusions of [Phe2,Ile3,Orn8]vasopressin.
A comparison of the effects of [Phe2,Ile3,Orn8]vasopressin on renal hemodynamics in rabbits treated with vehicle and MS-PPOH is presented in Fig. 2. In vehicle-treated rabbits, infusion of the V1 agonist dose dependently reduced MLDF, but it had no effect on RBF or CLDF (Fig. 2). The V1 agonist did reduce HR (by 17 ± 2% at 30 ng·kg−1·min−1) but did not significantly affect MAP (data not shown). In contrast, [Phe2,Ile3,Orn8]vasopressin dose dependently reduced RBF and CLDF in MS-PPOH-treated rabbits. For example, at a dose of 30 ng·kg−1·min−1, RBF and CLDF fell by 51 ± 8 and 59 ± 13%, respectively, in MS-PPOH-treated rabbits but changed only by −13 ± 15 and +3 ± 9% in vehicle-treated rabbits. The MLDF response to [Phe2,Ile3,Orn8]vasopressin was not significantly altered by MS-PPOH (Fig. 2). Thus responses of MLDF to [Phe2,Ile3,Orn8]vasopressin were significantly greater than those of CLDF in vehicle-treated rabbits but not in MS-PPOH-treated rabbits.
Responses to renal arterial infusion of ANG II.
These results are presented in Fig. 3. ANG II dose dependently reduced RBF and CLDF, but it had no significant effect on MLDF (Fig. 3) or MAP or HR (data not shown). At the highest dose studied (20 ng·kg−1·min−1) RBF was reduced by 45 ± 10% and CLDF was reduced by 41 ± 14%. The renal hemodynamic responses to ANG II were similar in rabbits given MS-PPOH.
CYP450 ω-Hydroxylase and Epoxygenase Activity
Because the renal metabolism of AA is highly influenced by genetics and environmental conditions, we first characterized the renal metabolism of AA in microsomes prepared from the renal cortex and medulla of the strain of rabbits used in this study. These results are presented in Fig. 4. Basal production of 20-HETE, EETs, and DiHETEs averaged 120 ± 16, 18 ± 3 and 43 ± 9 pmol·min−1·mg protein−1, respectively, in the renal cortex. The production of 20-HETE was 59 ± 9% less in microsomes prepared from the medulla of these rabbits, whereas production of EETs and DiHETEs was not significantly different from that in the cortex.
The effects of in vivo infusion of MS-PPOH on the metabolism of AA by renal cortical homogenates are presented in Fig. 5. Production of EETs and DiHETEs was 38 and 65% less, respectively, and total epoxygenase activity was 49% less, in cortical homogenates prepared from the kidneys of rabbits treated with MS-PPOH than in those from vehicle-treated rabbits. In contrast, ω-hydroxylase activity was similar in homogenates prepared from the kidneys of rabbits treated with vehicle and MS-PPOH.
The effects of [Phe2,Ile3,Orn8]vasopressin on CYP450 activity in vitro are shown in Fig. 6. Addition of [Phe2,Ile3,Orn8]vasopressin (0.2, 2, and 20 ng/ml) had no significant effects on the formation of EETs, DiHETEs or 20-HETE by cortical homogenates in vitro.
This study examined whether regional differences in the formation and release of EETs contribute to differences in the regional kidney blood flow responses to ANG II and [Phe2,Ile3,Orn8]vasopressin in rabbits. The results indicate that inhibition of epoxygenase activity with MS-PPOH markedly enhances the sensitivity of the renal cortical circulation to the vasoconstrictor actions of the V1 agonist [Phe2,Ile3,Orn8]vasopressin. In contrast, MS-PPOH had no significant effect on the changes in MLDF elicited by the V1 agonist, nor did it affect the responses to a different vasoconstrictor agonist, ANG II. These results suggest that products of CYP450 epoxygenation buffer V1-receptor-mediated reductions in CBF but not MBF, and so in large part account for the relative insensitivity of the cortical circulation to V1-mediated vasoconstriction. Our results are consistent with previous observations indicating that neither NO (36) nor products of cyclooxygenase-dependent (28, 29) or lipoxygenase-dependent (28) AA metabolism make major contributions to the relative insensitivity of the cortical circulation to the vasoconstrictor effects of V1-receptor activation.
We are not aware of any previous studies examining the effects of selective inhibition of epoxygenase activity on renal vasoconstrictor response to ANG II in vivo. Our observation that MS-PPOH did not significantly affect renal hemodynamic responses to ANG II does not allow us to exclude roles for EETs in modulating renal hemodynamic responses to this peptide because 20-HETE and EETs can be prestored in membrane phospholipids and released in response to hormonal stimuli (37). There is also evidence that ANG II stimulates the release of P-450 metabolites of AA from rabbit isolated perfused kidneys (8) and that EETs modulate the vasoconstrictor responses of in vitro-perfused afferent arterioles of rabbits (21) and rats (19). Nevertheless, the results of the present study indicate that the renal hemodynamic effects of ANG II in vivo in rabbits are not modulated by the de novo synthesis of EETs.
Measurements of CYP450 activity (as assessed by the production of EETs and DiHETEs from AA) in renal cortical homogenates prepared from the rabbits studied in vivo indicated that the dose of MS-PPOH was sufficient to selectively reduce CYP450 epoxygenase activity in vitro. Indeed, it may be that our in vitro assay underestimated the degree of epoxygenase inhibition in vivo, because preparation of the homogenates necessitated considerable dilution of the MS-PPOH within the kidney. As expected, MS-PPOH had no effect on the renal production of 20-HETE. These findings are consistent with those of previous in vivo (4) and in vitro (44) studies of rat kidney indicating that MS-PPOH is a selective epoxygenase inhibitor that is effective at blocking this pathway in vivo at doses comparable to those used in the present study.
We also characterized the production of EETs, DiHETEs, and 20-HETE in the kidneys of the rabbits used in this study because previous studies have indicated that there are large differences in renal AA metabolism and CYP450 isoform expression between species, and even between strains within a single species, and that the production of these compounds is dependent on diet and other environmental conditions (6, 26, 38, 45). The results indicate that the rate of formation of EETs and 20-HETE in renal cortical microsomes of the rabbits used in the present study was similar to that reported in the rat when studied under similar conditions (1) and much higher than that seen in mice (41). Our present data also indicate that CYP450 ω-hydroxylase and epoxygenase activity is two to four times greater in the renal cortex than outer medulla of the rabbit, similar to what has generally been reported in studies of rats (1, 41, 48).
Although our experiment provides no information regarding the precise identity of the AA epoxygenase products that appear to blunt V1-receptor-mediated cortical vasoconstriction, 11,12-EET, 8,9-EET, and 5,6-EET must be prime candidates because these (or their derivatives) have been demonstrated to be potent vasodilators of the renal circulation in rats and rabbits under in vitro conditions (7, 35, 46). In contrast, other studies have shown these EETs to produce vasoconstriction when infused directly into the renal artery (27, 41) and have also provided evidence that endogenous EETs (or their metabolites) at least partly mediate renal vasoconstrictor responses to NO synthase inhibition in rats (27). Most evidence suggests that the vascular actions of EETs depend heavily on the relative activity of downstream metabolic pathways (e.g., cyclooxygenase), which seem capable of converting EETs both into vasoconstrictor (41) and vasodilator (7) products. Thus the apparent discrepancy between studies identifying vasoconstrictor (27, 41) and vasodilator (present study) roles of EETs within the kidney in vivo might reflect differential activation of these downstream metabolic pathways under different experimental conditions.
Our results also do not allow us to determine whether V1 receptor activation stimulates release of EETs within the cortex or whether basal EET production is sufficient to blunt V1-mediated cortical vasoconstriction. However, the latter scenario seems unlikely, given our observation of similar CYP450 epoxygenase activity in cortical compared with medullary microsomes. The hypothesis that V1 receptor activation increases de novo synthesis of EETs is further supported by the observation that arginine vasopressin increases release of CYP450 metabolites in the isolated perfused rat kidney and that this effect is blunted by treatment with CYP450 inhibitors (30, 42).
We also found that [Phe2,Ile3,Orn8]vasopressin had little or no effect on CYP450 epoxygenase or ω-hydroxylase activity per se. Thus our results are consistent with the hypothesis that V1 receptor activation increases EET production in the cortex by increasing substrate availability. This hypothesis remains to be tested but is at least consistent with the established effect of arginine vasopressin to stimulate phospolipase A2 activity in vascular smooth muscle (5, 20). Regardless of the precise mechanism involved, our observations implicate CYP450-dependent EET/DiHETE formation as an important counterregulatory vasodilator mechanism, buffering renal vasoconstrictor responses to V1 receptor activation in the cortex, but not in vascular elements controlling MBF.
MS-PPOH had little effect on resting systemic and renal hemodynamics. Similarly, we previously found in anesthetized rabbits that renal arterial infusion of the nonselective CYP450 inhibitor 17-octadecynoic acid had no significant effects on RBF (13), and others have shown little effect of miconazole, a selective epoxygenase inhibitor, on RBF (31, 32). Thus these data do not support a role for EETs in control of basal renal vascular tone.
The present study confirms previous findings (3, 8, 11, 12, 14, 18, 28, 29, 33, 34, 36, 43, 49) that the medullary microcirculation is refractory to ANG II-induced vasoconstriction and that the cortical circulation is refractory to V1-receptor-mediated vasoconstriction. This differential effect of vasoactive hormones on regional kidney perfusion likely represents an important regulatory mechanism in long-term blood pressure control, but we are only just beginning to understand the mechanisms underlying it. Our present results indicate that EETs do not contribute to the relative insensitivity of the medullary circulation to ANG II, which has been previously attributed to the actions of NO and prostaglandins (3, 11, 29, 33, 34, 36, 43, 49). However, the relative insensitivity of the renal cortical circulation to V1-mediated vasoconstriction appears to be largely attributable to the actions of epoxygenase products of AA.
This study was supported by grants to R. G. Evans from the National Health and Medical Research Council of Australia (143785, 143603), the National Heart Foundation of Australia (G 00M 0633), and the Ramaciotti Foundations (A6370, RA159/98); to J. R. Falck from the National Institute of Diabetes and Digestive and Kidney Diseases (DK-38226) and the Robert A. Welch Foundation; and to R. J. Roman from the National Heart, Lung, and Blood Institute (HL-36279). N. W. Rajapakse is a recipient of a Commonwealth Postgraduate Scholarship (Australia).
We thank M. Aebly for assistance with the in vitro experiments of this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society