J Am Soc Nephrol 14: 2255–2263, 2003 Nongenomic Vascular Action of Aldosterone in the Glomerular Microcirculation

SHUJI ARIMA,* KENTARO KOHAGURA,* HONG-LAN XU,* AKIRA SUGAWARA,* TAKAAKI ABE,* FUMITOSHI SATOH,* KAZUHISA TAKEUCHI,* and SADAYOSHI ITO* Division of Nephrology, Endocrinology, and Vascular Medicine, Tohoku University School of Medicine, Sendai, 980-8574, Japan.

Abstract. Aldosterone (Aldo) accelerates hypertension, pro- strictor actions are nongenomic. Pretreatment with neomycin teinuria, and glomerulosclerosis in animal models of malignant (a specific inhibitor of phospholipase C) abolished the vaso- hypertension or chronic renal failure. Aldo may exert these constrictor action of Aldo in both arterioles. In addition, the deleterious renal effects by elevating renal vascular resistance vasoconstrictor action of Aldo on Af-Arts was inhibited by and glomerular capillary pressure. To test this possibility, both nifedipine and efonidipine, whereas that on Ef-Arts was directly examined were the action of Aldo on the afferent (Af) inhibited by efonidipine but not nifedipine. The results dem- and efferent (Ef) arterioles (Arts). Examined were the effect of onstrate that Aldo causes nongenomic vasoconstriction by ac- Aldo added to both the bath and lumen on the intraluminal tivating phospholipase C with a subsequent calcium mobiliza- diameter (measured at the most responsive point) of rabbits. tion thorough L- or T-type voltage-dependent calcium channels Aldo caused dose-dependent constriction in both arterioles in Af- or Ef-Arts, respectively. These vasoconstrictor actions with a higher sensitivity in Ef-Arts. Vasoconstrictor action of on the glomerular microcirculation may play an important role Aldo was not affected by a mineralocorticoid receptor antag- in the pathophysiology and progression of renal diseases by onist spironolactone and was reproduced by membrane-imper- elevating renal vascular resistance and glomerular capillary meable albumin-conjugated Aldo, suggesting that the vasocon- pressure.

There is increasing evidence that renin-angiotensin-aldosterone found that captopril treatment prevented the development of system (RAAS) plays an important role in the pathogenesis and proteinuria and glomerular-renal vascular lesions with reduc- progression of renal diseases (1–5). Although angiotensin II ing endogenous Aldo levels, whereas subsequent Aldo infusion (AngII) has been identified as the primary mediator of the reversed these protective effects of captopril. Although the system, recent studies have raised the possibility that aldoste- underlying mechanisms for these actions are not well defined, rone (Aldo), independent of renin-angiotensin, also participates Aldo may exert these deleterious renal effects by elevating in mediating renal injury (5–10). Greene et al. (7) have eval- renal vascular resistance and glomerular capillary pressure uated the ability of Aldo to reverse the renal protective effects (PGC). However, to our knowledge, there is so far no study of RAAS blockade in the 5/6 nephrectomy model of hyperten- demonstrating the direct vascular action of Aldo in the kidney, sion and glomerulosclerosis. They found that pharmacologic although Aldo causes renal arteriolar injury (11). blockade of RAAS with angiotensin-converting enzyme inhib- To test the possibility that Aldo may elevate PGC,wemi- itor and AngII receptor blocker reduces plasma Aldo levels, croperfused rabbit afferent (Af) and efferent (Ef) arterioles systolic BP, proteinuria and renal lesions, all of which were (Arts), crucial vascular segments to the control of glomerular reversed when Aldo was infused concurrently. Rocha et al. (8) hemodynamics, and directly examined the vascular action of also demonstrated that renal-protective effects of captopril, an Aldo on these arterioles. angiotensin-converting enzyme inhibitor, were reversed by an Aldo infusion in stroke-prone spontaneously hypertensive rat Materials and Methods (SHRSP), an animal model of malignant hypertension. They This study was performed in accordance with the Guide for Animal Experimentation, Tohoku University School of Medicine. We used meth- ods similar to those described previously to isolate and microperfuse Af- Received April 18, 2002. Accepted June 10, 2003. and Ef-Arts (12–14). Briefly, young male New Zealand white rabbits (1.5 Correspondence to Dr. Shuji Arima, Division of Nephrology, Endocrinology, to 2.0 kg body weight), fed standard rabbit chow and tap water ad libitum, and Vascular Medicine, Tohoku University School of Medicine, 1-1 Seiryo- were killed with intravenous sodium pentobarbital (40 mg/kg) and given cho, Aoba-ku, Sendai, 980-8574, Japan. Phone: 81-22-717-7163; Fax: 81-22- an intravenous injection of heparin (500 U). The kidneys were removed 717-7168; E-mail: [email protected] and sliced along the corticomedullary axis. Slices were placed in ice-cold 1046-6673/1409-2255 MEM (Life Technologies BRL, Gaithersburg, MD) containing 5% BSA Journal of the American Society of Nephrology (Sigma Chemical, St. Louis, MO) and microdissected under a stereomi- Copyright © 2003 by the American Society of Nephrology croscope (SZH-10; Olympus, Tokyo, Japan) as described previously. DOI: 10.1097/01.ASN.0000083982.74108.54 From each rabbit, a single superficial Af- and/or Ef-Art with its glomer- 2256 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 2255–2263, 2003 ulus intact was microdissected. With a micropipette, the arteriole was Af- or Ef-Arts were treated with spironolactone at 10Ϫ5 M for 60 min, transferred to a temperature-regulated chamber mounted on an inverted which is considered to be a sufficient dose and time period to take effect microscope (IMT-2; Olympus) with Hoffman modulation. The arteriole (15). Then vasoconstrictor actions of Aldo were examined as in protocol was cannulated with an array of glass pipettes as described previously 1. We have confirmed that spironolactone vehicle has no effect on the (12–14) and perfused with oxygenated medium 199 (Life Technologies luminal diameter or vascular reactivity of Af- or Ef-Arts. BRL) containing 5% BSA. Intraluminal pressure was measured by Lan- dis’ technique, using a fine pipette introduced into the arteriole through Effect of Albumin-Conjugated Aldo on the Luminal the perfusion pipette, and was maintained at 60 mmHg in the case of Diameter Af-Arts. For Ef-Art perfusion, an Af-Art was microdissected together We found that pretreatment with spironolactone had no effect on with the glomerulus and attached Ef-Art (approximately 250 to 300 ␮m Aldo-induced vasoconstriction in either arteriole (see Results). To in length). The Af-Art was cut short (approximately 50 ␮m) and cannu- further confirm that intracellular MR does not mediate Aldo’s vaso- lated as described above, except that the perfusion pipette was advanced constrictor action, we next examined whether albumin-conjugated to the end of the Af-Art. The tip of the pressure pipette was placed just Aldo (Alb-Aldo), which will not enter the cytoplasm, may cause similar beyond the distal end of the Af-Art, and intraluminal pressure at this point time- and dose-dependent vasoconstriction as induced by free-form Aldo. was maintained at 50 mmHg throughout the experiment to eliminate the The Alb-Aldo (Peptide Institute, Osaka, Japan) was produced by binding hemodynamic influences of the Af-Art (13,14). In a previous study (13), one BSA (molecular weight, 66,000; Sigma) molecule per 12 molecules we found that pressure in the Ef-Art at a point 50 ␮m distal to the of Aldo (Sigma). Because both the bath and perfusate already contain glomerulus was approximately 35 mmHg, the physiologic level, under high amounts of BSA, Aldo-bound BSA is thought not to affect the these experimental conditions. vascular reactivity in our preparations. Effects of Alb-Aldo on the luminal The bath was identical to the arteriolar perfusate except that it diameter were examined as in protocol 1. contained 0.1% BSA, and was exchanged continuously. Microdissec- tion and cannulation of the arteriole were completed within 90 min at 8°C, after which the bath was gradually warmed to 37°C for the rest Effect of Actinomycin D or Cycloheximide on Aldo- of the experiment. Once the temperature was stable, a 30-min equil- Induced Constriction ibration period was allowed before taking any measurements. Images To examine whether Aldo causes nongenomic vasoconstriction in of the arteriole were displayed at magnifications up to ϫ1980 and renal arterioles, we next examined the effect of actinomycin D and recorded with a video system consisting of a camera (CS520MD; cycloheximide, inhibitors of transcription and protein synthesis, on Olympus), monitor (PVM1445MD; Sony, Tokyo, Japan), and video Aldo-induced vasoconstriction. After the equilibration period, Af- or recorder (HR-S101; Victor, Tokyo, Japan). When we analyzed the Ef-Arts were treated with actinomycin D (10Ϫ5 M, Sigma) or cyclo- data, we first observed the whole record of each experiment and heximide (10Ϫ5 M, Sigma) for 60 min. Then vasoconstrictor actions decided the site where the strongest constriction occurred in response of Aldo were examined as in protocol 1. It has been demonstrated that to Aldo. We consistently measured the intraluminal diameter of this this concentration of actinomycin D or cycloheximide prevents Aldo- site throughout the experiments (including the control period or pre- induced genomic nuclear signaling (16) or de novo protein synthesis treatment period) with an image-analysis system (VM-30; Olympus). in rat aorta (17), respectively. We used same vehicle for actinomycin D or cycloheximide with that for spironolactone. Effect of Aldo on Luminal Diameter After the equilibration period, increasing doses of Aldo (10Ϫ10 M, Effect of Neomycin on Aldo-Induced Constriction Ϫ9 ϫ Ϫ9 Ϫ8 10 M, 5 10 M and 10 M; Sigma) were added to both the We next examined a possible involvement of phospholipase C bath and lumen. Luminal diameter was measured immediately before (PLC), which mediates nongenomic actions of Aldo in cultured vas- adding Aldo and observed for at least 30 min at each dose. cular smooth muscle cells (VSMCs) (18,19), in the Aldo-induced vasoconstriction. After the equilibration period, Af- or Ef-Arts were Time Course and Reversibility of Aldo-Induced treated with neomycin (300 ␮M; Sigma), a specific inhibitor of PLC, Constriction for 60 min. Then vasoconstrictor actions of Aldo were examined as in We found that Aldo at the nanomolar level caused vasoconstriction protocol 1. It has been demonstrated that neomycin at this concentra- in both arterioles (see Results). To exclude the possibility that these tion inhibits Aldo-induced rapid (within 30 s) increase in intracellular constrictions were induced by delayed effects of lower concentrations inositol 1,4,5-triphosphate (IP3) in rat VSMCs (20). We used the same of Aldo, we examined the time course of Aldo action. After the vehicle for neomycin with that for spironolactone. equilibration period, vehicle or Aldo (at 5 ϫ 10Ϫ9 Mor10Ϫ8 M) was added to both the bath and lumen, and then luminal diameter was Effect of Calcium Channel Blockade on Aldo-Induced observed for 30 min. In addition, we also examined the reversibility of Constriction Aldo-induced constriction. After examining the 30-min time course, It has been demonstrated that calcium (Ca2ϩ) influx to VSMC both the bath and arteriolar perfusate were changed to the vehicle through L-type and/or T-type voltage-dependent calcium channels without Aldo, and then diameter was observed for additional 30 min. (VDCC) is required to elicit sustained vasoconstriction in Af- and Ef-Arts (14,21). Thus, we next examined the possible contribution of Effect of Spironolactone on Aldo-Induced Constriction Ca2ϩ influx through these calcium channels to Aldo-induced vaso- We next studied the possible involvement of intracellular miner- constriction. After the equilibration period, Af- and Ef-Arts were alocorticoid receptor (MR) in the Aldo-induced vasoconstriction. For treated with nifedipine (L-type calcium channel blocker, 1 ␮M; this, we examined the effect of spironolactone, a classic MR antago- Sigma) or efonidipine (both L- and T-type calcium channel blocker, 1 nist. On the day of the experiment, a fresh solution containing spi- ␮M; Nissan Chemical Industries, Tokyo, Japan) for 30 min. Then ronolactone (Sigma) at 10Ϫ3 M was prepared in physiologic salt vasoconstrictor actions of Aldo were examined as in protocol 1. We solution containing 0.1% DMSO (Sigma). After the equilibration period, (14,22) and Hayashi et al. (21) have previously reported that nifedi- J Am Soc Nephrol 14: 2255–2263, 2003 Aldosterone in Renal Arterioles 2257 pine or efonidipine at this concentration completely abolishes AngII– vasoconstrictor actions of Aldo were significantly stronger (P Ͻ and norepinephrine-induced vasoconstriction in Af- or Ef-Arts, re- 0.01) in Ef-Arts only at 10Ϫ9 M (which decreased the diameter of spectively. We used the same vehicle for nifedipine or efonidipine Af-Arts by 0.8 Ϯ 0.4 ␮mor5%Ϯ 2%). with that for spironolactone.

Statistical Analyses Time Course and Reversibility of Aldo-Induced Values were expressed as mean Ϯ SEM, and all statistical analyses Constriction were carried out with absolute values. A paired t test was used to As shown in Figure 2, Aldo vehicle had no effect on the examine whether the diameter at a given concentration differed from luminal diameter of either arteriole. Before, 30 min, and 60 min the control or pretreated value within each group. ANCOVA was used after the treatment, the diameter was 17.0 Ϯ 0.6 ␮m, 17.2 Ϯ 0.6 to examine whether dose-response curves differed between groups, ␮m, and 17.0 Ϯ 0.6 ␮m in Af-Arts (n ϭ 7) and 15.0 Ϯ 0.2 ␮m, and a two-sample t test was used to examine whether the change in 15.0 Ϯ 0.2 ␮m, and 15.0 Ϯ 0.3 ␮m in Ef-Arts (n ϭ 7). Aldo at diameter at a given concentration differed between groups. A value of Ϫ Ϫ 5 ϫ 10 9 Mor10 8 M began to cause significant (P Ͻ 0.01) P Ͻ 0.0125 (0.05/4) was considered significant using Bonferroni’s adjustment for multiple comparisons. constriction within 5 min in both arterioles. At 10 min after the application of 5 ϫ 10Ϫ9 Mor10Ϫ8 M Aldo, luminal diameter Results decreased from 17.5 Ϯ 0.5 ␮m to 14.1 Ϯ 0.8 ␮m (by 3.4 Ϯ 0.5 ␮ Ϯ ϭ Ϯ ␮ Ϯ Effect of Aldo on Luminal Diameter mor19% 3%, n 8) or from 16.8 0.4 m to 12.5 0.4 ␮ Ϯ ␮ Ϯ ϭ Basal luminal diameter of Af- and Ef-Arts was 17.1 Ϯ 0.3 (n m (by 4.3 0.3 mor26% 2%, n 7) in Af-Arts, whereas Ϯ ␮ Ϯ ␮ Ϯ ␮ ϭ 8) and 14.9 Ϯ 0.4 ␮m(n ϭ 7), respectively. As shown in from 15.0 0.2 m to 12.1 0.2 m (by 3.0 0.3 mor20% Ϯ ϭ Ϯ ␮ Ϯ ␮ Figure 1, Aldo caused dose-dependent constriction in both 2%, n 7) or from 15.0 0.3 m to 11.2 0.3 m (by 3.7 Ϯ ␮ Ϯ ϭ arterioles (the constriction occurred within 10 min and persisted 0.3 m or 25% 2%, n 6) in Ef-Arts. However, thereafter, throughout the observation period). In Af-Arts, significant con- Aldo at each concentration did not cause further constriction until striction was observed from 5 ϫ 10Ϫ9 M, which decreased the 30 min in both arterioles. diameter by 3.4 Ϯ 0.7 ␮m (or 20% Ϯ 4%, P Ͻ 0.001). Aldo at Removing Aldo from the bath and lumen gradually dilated both 10Ϫ8 M similarly decreased the diameter by 4.9 Ϯ 0.5 ␮m (or arterioles, and the diameter tended to return to the basal level. 29% Ϯ 3%). On the other hand, Aldo began to cause significant Significant dilation was already observed at 10 min after remov- constriction from 10Ϫ9 M (by 2.2 Ϯ 0.3 ␮mor15%Ϯ 2%, P Ͻ ing Aldo in each case, and at 30 min, arteriolar diameters were not different from their basal value (at time 0) except for Af-Arts 0.01) in Ef-Arts. Thus, sensitivity to Aldo was higher in Ef-Arts Ϫ8 compared to Af-Arts. Aldo at 5 ϫ 10Ϫ9 Mor10Ϫ8 M similarly pretreated with 10 M Aldo. Thus, it became clear that vaso- decreased the diameter of Ef-Arts by 3.5 Ϯ 0.4 ␮m (or 23% Ϯ constrictor actions of Aldo on renal arterioles were reversible. 3%) or 3.9 Ϯ 0.2 ␮m (or 26% Ϯ 2%), respectively. In Ef-Arts, the Ϫ8 ϫ Ϫ9 decrease in diameter induced by 10 M Aldo (but not 5 10 Effect of Spironolactone on Aldo-Induced Constriction M Aldo) was significantly (P Ͻ 0.01) greater than that induced by Ϫ5 Ϫ9 Pretreatment with spironolactone at 10 M did not affect 10 M Aldo. In addition, when compared Af- and Ef-Arts, the luminal diameter of either arteriole; luminal diameter be- fore and after the treatment was 16.9 Ϯ 0.3 ␮m and 16.6 Ϯ 0.3 ␮m(n ϭ 6) or 15.0 Ϯ 0.3 ␮m and 15.2 Ϯ 0.4 ␮m(n ϭ 7) in Af- or Ef-Arts, respectively. As shown in Figure 3, spirono- lactone had no effect on Aldo-induced vasoconstriction in either arteriole. In spironolactone-treated arterioles, Aldo be- gan to cause significant constriction from 5 ϫ 10Ϫ9 M (by 3.2 Ϯ 0.6 ␮mor19%Ϯ 3%) in Af-Arts or from 10Ϫ9 M (by 2.0 Ϯ 0.2 ␮m or 14% Ϯ 2%) in Ef-Arts, and at 10Ϫ8 M the diameter decreased by 4.1 Ϯ 0.6 ␮m (or 24% Ϯ 3%) or 3.4 Ϯ 0.4 ␮m (or 22% Ϯ 2%), respectively.

Effect of Alb-Aldo on Luminal Diameter Basal luminal diameter of Af- or Ef-Arts was 17.0 Ϯ 0.6 ␮m or 14.5 Ϯ 0.3 ␮m(n ϭ 7, each), respectively. As shown in Figure 3, Alb-Aldo caused similar dose-dependent vasocon- striction (occurring within 10 min) with free-form Aldo in both Figure 1. Effect of aldosterone on the luminal diameter of afferent arterioles; Alb-Aldo began to cause significant (P Ͻ 0.001) arterioles or efferent arterioles. Aldosterone caused rapid dose-depen- constriction from 5 ϫ 10Ϫ9 M (by 3.4 Ϯ 0.5 ␮mor20%Ϯ dent vasoconstriction in both arterioles. Significant constriction was Ϫ Ϫ Ϫ 9 Ϯ ␮ Ϯ observed from 5 ϫ 10 9 M and 10 9 M in afferent arterioles (open 2%) in Af-Arts or from 10 M (by 1.9 0.4 mor13% Ϫ8 circles, n ϭ 8) and efferent arterioles (solid circles, n ϭ 7), respec- 2%) in Ef-Arts, and at 10 M, the diameter decreased by 4.4 tively. Thus, sensitivity to aldosterone was higher in efferent arte- Ϯ 0.6 ␮m (or 25% Ϯ 3%) or 3.6 Ϯ 0.4 ␮m (or 24% Ϯ 3%), rioles. *P Ͻ 0.01 compared with prealdosterone value. respectively. 2258 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 2255–2263, 2003

Figure 2. Time course and reversibility of aldosterone-induced vasoconstriction in afferent arterioles (left) and efferent arterioles (right). Aldosterone at 5 ϫ 10Ϫ9 M (solid triangles) or 10Ϫ8 M (solid circles) caused time-dependent constriction until 10 min; however, thereafter, aldosterone at each concentration did not cause further constriction until 30 min in either arteriole. Removing aldosterone from the bath and arteriolar perfusate gradually dilated both arterioles, and the diameter tended to return to the basal level. Thus, it became clear that vasoconstrictor actions of aldosterone on renal arterioles were reversible. On the other hand, aldosterone vehicle (open circles) observed for 60 min had no effect on the luminal diameter of either arteriole. *P Ͻ 0.01 compared with 0 min, #P Ͻ 0.01 compared with 30 min.

Figure 3. Effect of spironolactone or albumin conjugation on the vasoconstrictor action of aldosterone in afferent arterioles (left) and efferent arterioles (right). In either arteriole, vasoconstrictor action of aldosterone was not affected by pretreatment with 10Ϫ5 M spironolactone (solid circles) and was reproduced by membrane-impermeable albumin-conjugated aldosterone (solid squares), suggesting that the vasoconstrictor actions are nongenomic. *P Ͻ 0.01 compared with prealdosterone value.

Effect of Actinomycin D or Cycloheximide on Aldo- and 17.3 Ϯ 0.3 ␮m for cycloheximide (n ϭ 6), whereas that of Induced Constriction Ef-Arts was 14.8 Ϯ 0.4 ␮m and 14.8 Ϯ 0.4 ␮m for actinomy- Pretreatment with actinomycin D or cycloheximide did not cinD(n ϭ 7) or 14.8 Ϯ 0.3 ␮m and 14.8 Ϯ 0.3 ␮m for affect the luminal diameter of either arteriole. The diameter of cycloheximide (n ϭ 7). As shown in Figure 4, neither actino- Af-Arts before and after the treatment was 17.0 Ϯ 0.6 ␮m and mycin D nor cycloheximide had effect on Aldo-induced vaso- 16.7 Ϯ 0.7 ␮m for actinomycin D (n ϭ 6) or 17.5 Ϯ 0.3 ␮m constriction in either arteriole. In actinomycin D–treated arte- J Am Soc Nephrol 14: 2255–2263, 2003 Aldosterone in Renal Arterioles 2259

Figure 4. Effect of cycloheximide and actinomycin D on the vasoconstrictor action of aldosterone in afferent arterioles (left) and efferent arterioles (right). In either arteriole, vasoconstrictor action of aldosterone was not affected by pretreatment with 10Ϫ5 M cycloheximide (solid circles) or 10Ϫ5 M actinomycin D (solid squares), further suggesting that the vasoconstrictor actions are nongenomic. *P Ͻ 0.01 compared with prealdosterone value.

Figure 5. Effect of neomycin, a phospholipase C (PLC) inhibitor, on the vasoconstrictor action of aldosterone in afferent arterioles (left) and efferent arterioles (right). Pretreatment with 300 ␮M neomycin (solid circles) abolished the vasoconstrictor action of aldosterone in either arteriole, indicating the critical role of PLC activation in it. *P Ͻ 0.01 compared with prealdosterone value.

rioles, Aldo began to cause significant constriction from 5 ϫ 2%) in Ef-Arts, and at 10Ϫ8 M, the diameter decreased by 4.2 10Ϫ9 M (by 2.7 Ϯ 0.6 ␮mor16%Ϯ 3%) in Af-Arts or from Ϯ 0.7 ␮m (or 24% Ϯ 4%) or 3.8 Ϯ 0.6 ␮m (or 25% Ϯ 4%), 10Ϫ9 M (by 2.4 Ϯ 0.4 ␮mor16%Ϯ 3%) in Ef-Arts, and at respectively. 10Ϫ8 M, the diameter decreased by 3.9 Ϯ 0.5 ␮m (or 23% Ϯ 2%) or 3.9 Ϯ 0.4 ␮m (or 26% Ϯ 2%), respectively. In cyclo- Effect of Neomycin on Aldo-Induced Constriction heximide-treated arterioles, Aldo began to cause significant Pretreatment with neomycin at 300 ␮M did not affect the constriction from 5 ϫ 10Ϫ9 M (by 3.2 Ϯ 0.8 ␮mor19%Ϯ luminal diameter of either arteriole; luminal diameter before 5%) in Af-Arts or from 10Ϫ9 M (by 1.9 Ϯ 0.4 ␮mor12%Ϯ and after the treatment was 17.7 Ϯ 0.5 ␮m and 17.3 Ϯ 0.5 ␮m 2260 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 2255–2263, 2003

(n ϭ 6) or 15.1 Ϯ 0.5 ␮m and 15.2 Ϯ 0.5 ␮m(n ϭ 6) in Af- constriction until the concentration reached 10Ϫ8 M, which or Ef-Arts, respectively. As shown in Figure 5, Aldo did not slightly decreased the diameter by 1.7 Ϯ 0.3 ␮m (or 11% Ϯ cause any constriction in such arterioles; the decrease in diam- 2%, P Ͻ 0.01). eter induced by Aldo at 10Ϫ8 M was 0.5 Ϯ 0.3 ␮m and 0.8 Ϯ 0.5 ␮m in Af- and Ef-Arts, respectively. Thus, pretreatment Discussion with neomycin abolished the vasoconstrictor action of Aldo in Development of proteinuria and elevated PGC in the miner- either arteriole. alocorticoid-salt hypertension model (23) suggests that miner- alocorticoids cause hemodynamic injury to the glomerulus. Effect of Calcium Channel Blockade on Aldo-Induced However, their vascular action on the glomerular hemodynam- Constriction ics has not been defined. In the study presented here, we Pretreatment with nifedipine at 1 ␮M did not affect the examined the possible vascular action of Aldo, a representative luminal diameter of either arteriole; luminal diameter before mineralocorticoid, on the renal microvessels and found that and after the treatment was 17.3 Ϯ 0.5 ␮m and 17.7 Ϯ 0.4 ␮m Aldo at the nanomolar level causes dose-dependent vasocon- (n ϭ 8) or 15.0 Ϯ 0.3 ␮m and 14.9 Ϯ 0.3 ␮m(n ϭ 7) in Af- striction in both Af- and Ef-Arts, with a higher sensitivity in or Ef-Arts, respectively. As shown in Figure 6, Aldo did not the Ef-Art. Although the decrease in diameter (approximately cause any constriction in nifedipine-treated Af-Arts; the de- 15% to 30%) induced by Aldo appears to be weaker than that crease in diameter induced by Aldo at 10Ϫ8 M was 0.6 Ϯ 0.3 induced by other vasoconstrictors such as AngII or norepineph- ␮m (3% Ϯ 1%). On the other hand, nifedipine had no effect on rine (13), it represents more than twice increase in vascular Aldo-induced vasoconstriction in Ef-Arts. In nifedipine-treated resistance (because the vascular resistance is proportional to Ef-Arts, Aldo began to cause significant constriction from the reciprocal of fourth power of radius). Thus, from a physi- 10Ϫ9 M (by 2.1 Ϯ 0.4 ␮mor14%Ϯ 2%), and at 10Ϫ8 M, the ological point of view, Aldo may play an important role in the diameter decreased by 4.0 Ϯ 0.4 ␮m (or 27% Ϯ 3%). regulation of vascular resistance in the renal microcirculation. Pretreatment with efonidipine at 1 ␮M did not affect the The concentrations of Aldo required for significant constric- luminal diameter of either arteriole; luminal diameter before tion were higher than physiologic plasma levels (approximate- and after the treatment was 17.0 Ϯ 0.6 ␮m and 17.7 Ϯ 0.4 ␮m ly 0.1 nM in human (24) or approximately 0.5 nM in rabbit (n ϭ 6) or 15.0 Ϯ 0.3 ␮m and 14.9 Ϯ 0.3 ␮m(n ϭ 7) in Af- (25)); however, under some pathologic conditions, plasma or Ef-Arts, respectively. In contrast to nifedipine, efonidipine Aldo concentration (PAC) rises above 1 nM, which causes inhibited Aldo-induced vasoconstriction in both arterioles. As significant constriction in Ef-Arts in the study presented here. shown in Figure 6, Aldo did not cause any constriction in Berl et al. (26) have reported that mean PAC of eight patients efonidipine-treated Af-Arts; the decrease in diameter induced with chronic renal failure were 523 pg/ml (approximately 1.4 by Aldo at 10Ϫ8 M was 0.7 Ϯ 0.4 ␮m (4% Ϯ 2%). In nM), which increased to 2393 pg/ml (6.6 nM) when patients efonidipine-treated Ef-Arts, Aldo did not cause significant were placed on a low-sodium diet. Greene et al. (7) reported

Figure 6. Effect of calcium channel blockade on the vasoconstrictor action of aldosterone in afferent arterioles (left) and efferent arterioles (right). Vasoconstrictor action of aldosterone on afferent arterioles was inhibited by both nifedipine (L-type calcium channel blocker; solid circles) and efonidipine (both L- and T-type calcium channel blocker; solid squares), whereas that on efferent arterioles was inhibited only by efonidipine. Thus, calcium mobilization thorough L- or T-type calcium channels seems to be required for aldosterone-induced constriction in afferent or efferent arterioles, respectively. *P Ͻ 0.01 compared with prealdosterone value. J Am Soc Nephrol 14: 2255–2263, 2003 Aldosterone in Renal Arterioles 2261 that PAC increases 10-fold in five of six nephrectomized rats ment of PLC (and the phosphoinositide pathway) in the intra- from 50 to 526 pg/ml (approximately 1.5 nM). Thus, Aldo may cellular signaling for nongenomic Aldo action on the VSMCs cause vasoconstriction in Ef-Arts under some pathologic con- (18,19). As expected, we found that pretreatment with neomy- ditions such as chronic renal failure, in which hyperaldoste- cin (a specific inhibitor of PLC) completely abolishes Aldo- ronism (several- to tenfold increase in PAC) is a common induced constriction in both Af- and Ef-Arts, demonstrating finding in both experimental animals (7) and humans (26,27). the involvement of PLC and phosphoinositide pathway in the Through its vasoconstrictor action on the postglomerular Ef- intracellular signaling for Aldo-induced constriction of renal

Art, Aldo would elevate PGC. Glomerular hypertension is now arterioles. In addition, we have also found that both nifedipine believed to be responsible, at least in part, for the development (L-type calcium channel blocker) and efonidipine (L/T-type of glomerular dysfunction and glomerular structural damages calcium channel blocker) inhibited the vasoconstrictor action in renal diseases (28–30). Thus, this vasoconstrictor action of Aldo on Af-Arts, whereas efonidipine but not nifedipine may account for the pathogenic role for hyperaldosteronism in inhibited that on Ef-Arts. Thus, our results indicate that Aldo the progression of renal diseases (5–10). causes nongenomic vasoconstriction by activating PLC with a In Af-Arts, further higher concentrations of Aldo were re- subsequent calcium mobilization thorough L- or T-type VDCC quired for significant constriction. However, because PAC is in Af- or Ef-Arts, respectively. However, the precise mecha- known to elevate in SHRSP (PAC of M-strain SHRSP at age nism of post-PLC activation for Aldo-induced constriction is of 12 wk reaches approximately 4.5 nM (31)), Aldo may also still unclear. Activation of PLC induces hydrolytic breakdown cause vasoconstriction in Af-Arts under some other pathologic of phosphoinositides into IP3 and diacylglycerol (DAG); IP3 conditions, such as severe hypertension. Because the Af-Art induces calcium release from the sarcoplasmic reticulum and accounts for most of the preglomerular vascular resistance and DAG activates protein kinase C (PKC) (38). Indeed, Christ et increase in its vascular resistance contributes to the pathogen- al. (20,39) have demonstrated that Aldo, presumably through esis of essential hypertension (32), the vasoconstrictor action its nongenomic action, stimulates the production of both IP3 on Af-Arts may play a role in the BP elevating effect of Aldo and DAG-PKC in VSMC. Thus, IP3 and/or DAG-PKC path- besides its “volume-retaining” actions in severe to malignant ways may act to elicit Aldo-induced vasoconstriction in renal hypertension. arterioles. It is generally accepted that Aldo exerts its action by binding With regard to this possibility, using AngII as a vasocon- to high-affinity intracellular MR, which act as transcription strictor agent, Nagahama et al. (40) have studied the relative factors and mediate delayed genomic effects (18,33). This role of IP3 and DAG-PKC pathways in PLC-mediated renal delay (generally hours) would be a consequence of the time arteriolar vasoconstriction. They found that IP3-induced cal- necessary for transcribing new mRNA and subsequent trans- cium release from the sarcoplasmic reticulum with a subse- lation of that message into cell-specific proteins. On the other quent extracellular calcium entry through L-type VDCC plays hand, several recent studies have demonstrated the rapid non- an important role in the Af-Art constriction-mediated by PLC genomic effects of Aldo in cardiovascular tissues. In cultured activation, whereas DAG-PKC pathway which requires extra- VSMC, physiologic concentrations (subnanomolar) of Aldo cellular calcium entry independent of L-type VDCC plays a induces a prompt (Ͻ1 min) increase in intracellular Ca2ϩ predominant role in the Ef-Art. However, post-PLC mecha- concentration, which was not inhibited by a MR antagonist nism for Aldo may be different from that of AngII. Thus, spironolactone (19). Moreover, in humans, intravenous appli- further studies that more precisely characterize the intracellular cation of Aldo significantly increases systemic vascular resis- mechanism or mechanisms by which Aldo causes vasocon- tance within a few minutes (34,35), suggesting a nongenomic striction are required. In addition, the biochemical nature of the vasoconstrictor action because of the short time frame. In the receptors mediating Aldo-induced vasoconstriction is also to study presented here, vasoconstrictor action of Aldo was also be established. compatible with nongenomic actions; the major characteristics In addition to such nongenomic vascular actions, genomic was its relatively early onset (apparent within 5 min), which renal deleterious actions of Aldo have also been reported. was not affected by spironolactone at 1000-fold excess con- Rocha et al. (11,41) have demonstrated that spironolactone or centrations and was reproduced by membrane-impermeable eplerenone, a selective Aldo receptor blocker, reduces protein- Alb-Aldo. Furthermore, neither actinomycin D nor cyclohexi- uria and malignant nephrosclerotic lesions without altering BP mide had effect on Aldo-induced vasoconstriction in either in saline-drinking SHRSP or in AngII–infused and saline- arteriole. Taken together, these results strongly suggest that in drinking Wistar rats with chronic nitric oxide synthesis inhi- the glomerular microcirculation Aldo causes direct non- bition. Thus, it is also possible that Aldo may exert genomic genomic vasoconstriction, which may be mediated by specific vascular actions on the glomerular microvessels when applied membrane receptors distinct from the intracellular MRs. This much longer (several hours or more) than this study (30 min for idea is supported by recent studies indicating the presence of each dose). Alternatively, both genomic and nongenomic path- membrane receptors that mediate rapid nongenomic actions of ways may synergistically contribute to the vascular actions of Aldo (18,36,37). Aldo on the glomerular microcirculation because it is proposed To study the mechanism or mechanisms for Aldo-induced that rapid modulation of intracellular signaling and second vasoconstriction, we first examined the possible contribution messengers by nongenomic actions in turn influences the of PLC because several studies have demonstrated the involve- genomic actions of Aldo (15,18,42). Further studies examining 2262 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 2255–2263, 2003 possible genomic vascular actions of Aldo on the renal mi- 7. Greene E, Kren S, Hostetter TH: Role of aldosterone in the crovessels, together with their possible interactions with non- remnant kidney model in the rat. J Clin Invest 98: 1063–1068, genomic actions, are clearly required. 1996 In summary, we found that Aldo causes dose-dependent 8. Rocha R, Chander PN, Zuckerman A, Stier CT: Role of aldo- vasoconstriction in both Af- and Ef-Arts, with a higher sensi- sterone in renal vascular injury in stroke-prone hypertensive rats. tivity in Ef-Arts. Thus, in addition to its tubular and fibrotic Hypertension 33: 232–237, 1999 actions, Aldo may contribute to the pathophysiology and pro- 9. Hostetter TH, Rosenberg ME, Ibrahim HN, Juknevicius I: Aldo- gression of renal diseases by elevating renal vascular resistance sterone in renal disease. Curr Opin Nephrol Hypertens 10: 105– 110, 2001 and P through its hemodynamic actions on the glomerular GC 10. Epstein M: Aldosterone as a mediator of progressive renal dis- microcirculation. We also found that these vasoconstrictor ease: Pathogenic and clinical implications. Am J Kidney Dis 37: actions were presumably nongenomic and mediated via mem- 677–688, 2001 brane-bound receptors because they were not affected by spi- 11. Rocha R, Stier CT Jr, Kifor I, Ochoa-Maya MR, Rennke HG, ronolactone, actinomycin D or cycloheximide, and were repro- Williams GH, Adler GK: Aldosterone: A mediator of myocardial duced by the membrane-impermeable conjugate of BSA-Aldo. necrosis and renal arteriopathy. Endocrinology 141: 3871–3878, In addition, our results indicate that the vasoconstrictor 2000 actions are mediated through activation of PLC with a subse- 12. Ito S, Johnson CS, Carretero OA: Modulation of angiotensin quent calcium mobilization via L- or T-type calcium channels II-induced vasoconstriction by endothelium-derived relaxing fac- in Af- or Ef-Arts, respectively. However, our results obtained tor in the isolated microperfused rabbit afferent arteriole. J Clin in isolated arterioles may represent an in vitro phenomenon and Invest 87: 1656–1663, 1991 may have no direct relevance to the in vivo Aldo action. 13. Ito S, Arima S, Ren Y, Juncos LA, Carretero OA: Endothelium- Although Romagni et al. (43) recently demonstrated that Aldo derived relaxing factor/nitric oxide modulates angiotensin II ac- at physiologic concentrations causes direct and rapid (presum- tion in the isolated microperfused rabbit afferent but not efferent ably nongenomic) vasoconstriction in human resistance arter- arteriole. J Clin Invest 91: 2012–2019, 1993 ies in vivo, to our knowledge, no study has provided direct 14. Arima S, Ito S, Omata K, Tsunoda K, Yaoita H, Abe K: Diverse evidence that Aldo causes renal vasoconstriction in vivo. Thus, effects of calcium antagonists on the glomerular hemodynamics. to clarify the physiologic and pathophysiologic significance of Kidney Int 49[Suppl 55]: S132–S134, 1996 the possible (nongenomic) vascular actions of Aldo in the 15. Christ M, Gunther A, Heck M, Schmidt BMW, Falkenstein E, Wehling M: Aldosterone, not estradiol, is the physiological ag- kidney, studies investigating the renal hemodynamic effects of onist for rapid increases in cAMP in vascular smooth muscle Aldo in vivo (for example, effects of Aldo on BP, renal blood cells. Circulation 99: 1485–1491, 1999 flow, renal vascular resistance, and GFR) are required. 16. Schafer C, Shahin V, Albermann L, Hug MJ, Reinhardt J, Schill- ers H, Schneider SW, Oberleithner H: Aldosterone signaling Acknowledgments pathway across the nuclear envelope. Proc Natl Acad SciUSA This work was supported in part by research grants for cardiovas- 99: 7154–7159, 2002 cular research (11C-4, 13C-5) from the Ministry of Health and Wel- 17. Browne M, Connolly C, Dicherty JR: Vascular actions of 17b- fare of Japan, and grants (12470208, 12671020, 13770594, and oestradiol in rat aorta and mesenteric artery. J Auton Pharmacol 15590840) from the Ministry of Science, Education, and Culture, 19: 291–299, 1999 Japan. We thank Miss Hiroko Kato for her experimental assistance 18. Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M: and Miss Mari Goko for her secretarial assistance. Multiple actions of steroid hormones—A focus on rapid, non- genomic effects. Pharmacol Rev 52: 513–555, 2000 19. Wehling M, Neylon CB, Fullerton M, Bobik A, Funder JW: References Nongenomic effects of aldosterone on intracellular calcium in 1. Anderson S, Rennke H, Brenner BM: Therapeutic advantage of vascular smooth muscle cells. Circ Res 76: 973–979, 1995 converting enzyme inhibitors in arresting progressive renal dis- 20. Christ M, Douwes K, Eisen C, Bechtner G, Theisen K, Wehling eases associated with systemic hypertension in the rat. J Clin M: Rapid effects of aldosterone on sodium transport in vascular Invest 77: 1993–2000, 1986 smooth muscle cells. Hypertension 26: 117–123, 1995 2. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD: The effect of 21. Hayashi K, Nagahama T, Oka K, Epstein M, Saruta T: Disparate angiotensin-converting-enzyme inhibition on diabetic nephropa- effects of calcium antagonists on renal microcirculation. Hyper- thy. N Engl J Med 329: 1456–1462, 1993 3. Rosenberg ME, Smith LJ, Correa-Rotter R, Hostetter TH: The tens Res 19: 31–36, 1996 paradox of the renin-angiotensin system in chronic renal disease. 22. Arima S, Ito S, Omata K, Tsunoda K, Yaoita H, Abe K: Diverse Kidney Int 45: 403–410, 1994 hemodynamic effects of calcium blockade in renal microcircu- 4. Remuzzi G, Ruggenenti P, Benigni A: Understanding the nature lation [Abstract]. J Hypertens 14[Suppl 1]:S45, 1996 of renal disease progression. Kidney Int 51: 2–15, 1997 23. Dworkin LD, Hostetter TH, Rennke HG, Brenner BM; Hemo- 5. Ibrahim HN, Rosenberg ME, Hostetter TH: Role of renin-angio- dynamic basis for glomerular injury in rats with deoxycortico- tensin-aldosterone system in the progression of renal disease: A sterone-salt hypertension. J Clin Invest 73: 1448–1461, 1984 critical review. Semin Nephrol 17: 431–440, 1997 24. Al-Dujaili EAS, Edwards CRW: The development and applica- 6. Quan ZY, Walser M, Hill GS: Adrenalectomy ameliorates abla- tion of direct radioimmunoassay for plasma aldosterone using tive nephropathy in the rat independently of corticosterone main- 131-I–labelled ligand-comparison of three methods. J Clin En- tenance level. Kidney Int 41: 326–333, 1992 docrinol Metab 46: 105–113, 1978 J Am Soc Nephrol 14: 2255–2263, 2003 Aldosterone in Renal Arterioles 2263

25. Mihailidou AS, Bundgaard H, Mardini M, Hansen PS, Kjeldsen 35. Schmidt BMW, Montealegre A, Janson CP, Martin N, Stein- K, Rasmussen HH: Hyperaldosteronemia in rabbits inhibits the Kemmesies C, Scherhag A, Feuring M, Christ M, Wehling M: ϩ ϩ cardiac sarcolemmal Na -K pump. Circ Res 86: 37–42, 2000 Short term cardiovascular effects of aldosterone in healthy male 26. Berl T, Katz FH, Henrich WL, de Torrente A, Schrier RW: Role volunteers. J Clin Endocrinol Metab 84: 3528–3533, 1999 of aldosterone in the control of sodium excretion in patients with 36. Wehling M, Christ M, Theisen K: Membrane receptors for advanced chronic renal failure. Kidney Int 14: 228–235, 1978 aldosterone: A novel pathway for mineralocorticoid action. Am J 27. Hene RJ: Boer P, Koomans HA, Mees EJD: Plasma aldosterone con- Physiol 263: E974–E979, 1992 centrations in chronic renal disease. Kidney Int 21 :98–101, 1982 37. Christ M, Sippel K, Eisen C, Wehling M: Non-classical receptors 28. Hostetter TH, Rennke HG, Brenner BM: The case for intrarenal for aldosterone in plasma membranes from pig kidneys. Mol Cell hypertension in the initiation and progression of diabetic and Endocrinol 99: R31–R34, 1994 other glomerulopathies. Am J Med 72: 375–380, 1982 38. Kamm KE, Stull JT: Regulation of smooth muscle contractile 29. Zats R, Dunn R, Meyer TW, Anderson S, Rennke HG, Brenner elements by second messengers. Annu Rev Physiol 51: 299–313, BM: Prevention of diabetic glomerulopathy by pharmacological 1989 amelioration of glomerular capillary hypertension. J Clin Invest 39. Christ M, Meyer C, Sippel K, Wehling M: Rapid aldosterone 77: 1925–1930, 1986 signaling in vascular smooth muscle cells: Involvement of phos- 30. Neuringer JR, Brenner BM: Hemodynamic theory of progressive pholipase C, diacylglycerol and protein kinase C. Biochem Bio- renal disease: A 10-year update in brief review. Am J Kidney Dis phys Res Commun 213: 123–129, 1995 22: 98–104, 1993 31. Horiuchi M, Nishiyama H, Hama J, Takenaka T, Hondo H, Kino 40. Nagahama T, Hayashi K, Ozawa Y, Takenaka T, Saruta T: Role H, Nagata S, Sugimura K, Katori R: Characterization of renal of protein kinase C in angiotensin II-induced constriction of renal aldosterone receptors in gebnetically hypertensive rats. Am J microvessels. Kidney Int 57: 215–223, 2000 Physiol 264:F286–F291, 1993 41. Rocha R, Chander PN, Khanna K, Zuckerman A, Stier CT Jr: 32. Imig JD, falck JR, Gebremedhin D, Harder DR, Roman RJ: Mineralocorticoid blockade reduces vascular injury in stroke- Elevated renovascular tone in young spontaneously hypertensive prone hypertensive rats. Hypertension 31: 451–458, 1998 rats: Role of cytochrome P-450. Hypertension 22: 357–364, 1993 42. Lim-Tio SS, Fuller PJ: Intracellular signaling pathways confer 33. Funder JW: Aldosterone action: Fact failure and the future. Clin specificity of transactivation by mineralocorticoid and glucocor- Expt Pharmacol Physiol 25: S47–S50, 1998 ticoid receptors. Endocrinology 139: 1653–1661, 1998 34. Wehling M, Spes CH, Win N, Anson CP, Schmidt BMW, 43. Romagni P, Rossi F, Guerrini L, Quirini C, Santiemma V: Theisen SK, Christ M: Rapid cardiovascular action of aldoste- Aldosterone induces contraction of the resistance arteries in man. rone in man. J Clin Endocrinol Metab 83: 3517–3522, 1998 Atherosclerosis 166: 345–349, 2003