Effect of prostanoids on renin release from rabbit afferent arterioles with and without macula densa

Kidney International, Vol. 35 (1989), pp. 1138—1144

Effect of pro stanoids on renin release from rabbit afferent arterioles with and without macula densa SADAYOSHI ITO1, OSCAR A. CARRETERO, KEISHI ABE, WILLIAM H. BEIERWALTES, and KAORU YOSHINAGA Hypertension Research Division, Henry Ford Hospital, Detroit, Michigan, USA and The Second Department of Internal Medicine, Tohoku University School of Medicine, Sendai 980, Japan

Effect of prostanoids on renin release from rabbit afferent arterioles with and without macula densa. There is evidence that cyclooxygenase products of arachidonic acid participate in the control of renin release. In this study we tested the hypothesis that prostaglandin (P0)12 and/or its metabolite(s), which are synthesized in the afferent arteriole (AF),

of renin release. On the other hand, POE2 appears to be the principal PG synthesized by renal tubular and medullary inter-

significantly (Pc 0.0001) from 1.04 0.21 to 3.12 0.86 (,i saM, N = 7), and from 0.45 0.14 to 1.48 0.53 (N = 9), respectively. During the recovery period, renin release increased even further, reaching 9.53 1.76 and 4.50 1.24, respectively. A P012 synthetase inhibitor, 9, 1 l-azoprosta-5,l3-dienoic acid blocked the effect of arachidonic acid. To examine whether the increases in renin release during the recovery period were due to metabolite(s) of P012, we tested the effect of both 6-keto-PGE1 (an active metabolite of P012) and carba-P012 (a synthetic analog that is metabolized differently from P012). Six-keto-PGE1 and carba-POI2 increased renin release only during the experimental period with no further increase during the recovery period. When POE2 (14 /LM) was added to AT alone, renin release did not change significantly; however, when it was added to AF-MD, renin release increased by 0.09 to 1.26 0.24 (N = 8, P c 0.02). These 133%, going from 0.54 results suggest that AF synthesizes PG!2 and possibly its metabolite(s), which in turn stimulate renin release by acting directly on AF, whereas POE2 stimulates renin release indirectly via the macula densa.

we microdissected rabbit afferent arterioles with and without macula densa attached [12—14], and incubated the structures in vitro. Using these two preparations, we examined first whether altering endogenous synthesis of P012 in the afferent arterioles affects renin release, second the effects of P0!2, 6-keto-PGE, (an active metabolite of P012) and carba-P012 (a synthetic analog that is metabolized differently from POT2) on renin release from afferent arterioles, and third the effect of POE2 on renin release from afferent arterioles both with and without macula densa attached.

also produce P012 which could in turn participate in the control

Arbor, Michigan, USA). Upjohn Pharmaceutical Co. supplied a

stitial cells [101, and it has been hypothesized that POE2 could stimulate renin release by altering electrolyte transport at the

stimulate renin release by acting directly on the AF while POE2 level of the macula densa [11]. However, there is no direct stimulates renin release indirectly via the macula densa. AF alone and evidence to support this hypothesis. AF with macula densa attached (AF-MD) were microdissected from In the present study, we tested the hypothesis that P012 and! rabbit kidneys and incubated in vitro. The renin release rate from a or its metabolite(s) are synthesized in the afferent arterioles and single AF (or an AF-MD) was calculated and expressed as ng act on them to stimulate renin release, whereas POE2 stimulates AL . hr' AW*hr (where AL is angiotensin I). When arachidonic acid (0.12 mM) or P012 (10 j.tM) was added to AF, renin release increased renin release indirectly via the macula densa. For this purpose,

Methods

Protocols 1 through 6 were performed in Detroit, Michigan, USA, and protocols 7 and 8 in Sendai, Japan. The following materials were obtained commercially and used in protocols 1 through 6: medium 199, Gibco Laboratories (Orand Island, There is evidence that cyclooxygenase products of arachi- New York, USA); bovine serum albumin (BSA), Schwarz! donic acid participate in the control of renin release [1—41. Mann (Orangeburg, New York, USA); sodium heparin (1,000 Although various prostaglandins (PGs) have been shown to U!ml), Elkins-Sinn, Inc. (Cherry Hill, New Jersey, USA); stimulate renin release both in vivo and in vitro, evidence sodium pentobarbital (Nembutal, 50 mg!mI), Abbott Laboratosupports PG!2 as the mediator of the PG-dependent renin ries (North Chicago, Illinois, USA); arachidonic acid (sodium releasing mechanism [5—7], PG!2 is the main PG synthesized by salt), Calbiochem-Behring Corp. (La Jolla, California, USA); the renal cortical blood vessesl [8, 9]. Thus afferent arterioles, POE2, PGI2 and 6-keto-PGE,, Sigma Chemical (St. Louis, where the juxtaglomerular cells synthesize and store renin, may Missouri, USA); and carba-PGI2, Cayman Chemical (Ann

Received for publication August 8, 1988 and in revised form November 14, 1988 Accepted for publication November 29, 1988

P012 synthetase inhibitor, 9,11 -azoprosta-5, 13-dienoic acid (azo analog 1). The following materials were obtained commercially and used in protocols 7 and 8: Eagle minimal essential medium (MEM), Nissui Pharmaceutical Co. (Tokyo, Japan); sodium heparin, Novo Industries (Denmark); sodium pentobarbital, Abbott Laboratories; bovine serum albumin (BSA), Wako Pure Chemical Industries L.T.D. (Osaka, Japan); and POE2,

© 1989 by the International Society of Nephrology

Sigma Chemical.

'In previous publications, the last name was spelled Itoh.

1138

Ito et al: Prostaglandin and renin release

Isolation and incubation of afferent arterioles with and without macula densa attached

1139

medium-0. 1% BSA and immediately frozen. After freezing and

thawing were repeated five times, the tissue samples were stored frozen until tissue renin content was determined.

The methods for the isolation and incubation of afferent arterioles, both with and without macula densa attached, have been described previously [12—14]. Young male, New Zealand white rabbits (N = 45) were used in protocols 1 through 6, and young Japanese white rabbits (N = 16) in protocols 7 and 8. We have previously shown that the basal renin release rate and renin response to isoproterenol in isolated afferent arterioles are similar in both strains [15]. Rabbits (1.5 to 2.5 kg) maintained on a standard rabbit chow and tap water ad libitum were anesthetized with sodium pentobarbital (40 mg/kg, i.v.) and given an intravenous injection of heparin. The left kidney was perfused in situ with cold, oxygenated (95% 02 and 5% C02) medium (medium 199 and MEM in protocols I to 6 and 7 and 8, respectively) containing 0.1%

Experimental protocol 1. Time control. Afferent arterioles were incubated in medium 199 containing 0.1% BSA for three 15-minute periods concurrently with each of the protocols 2 through 6. The proper

vehicle was added to the incubation medium used during the experimental period. Medium 199 had the following composition: Nat, 133 mEq/liter; K, 5.4 mEq/liter; C1, 125 mEq/liter;

Ca2, 3.6 mEqfliter; HC03, 14.9 mM; H2P04, 1.0 mM; HEPES 25 mM.

2. Arachidonjc acid. Arachidonic acid (final concentration 1.2 x l0— M) was added to the incubation medium used during the experimental period.

BSA. The kidney was then removed and sliced along the

3. PGI synthetase inhibitor plus arachidonic acid. In order to determine the importance of PGI2 in arachidonic acid-

corticomedullary axis. Slices were placed in ice-cold medium and microdissected at 4°C under a stereomicroscope at magnifications up to lOOx. Superficial afferent arterioles with or without macula densa attached, arising from the terminus of the interlobular arteries, were microdissected in 90 minutes. The afferent arterioles were

induced renin release, afferent arterioles were pretreated with a PGI2 synthetase inhibitor, 9,11 -azoprosta-5, 13-dienoic acid (azo analog 1) at a concentration of 2.8 X l0_6 M, which has been shown to block PG!2 synthesis [5, 7]. Azo analog 1 was added to the incubation media used during the preincubation, control, experimental and recovery periods. Arachidonic acid

severed from the interlobular arteries and glomeruli using (1.2 x l0 M) or its vehicle was added to the incubation hypodermic needles (26 to 30 gauge). Care was taken to avoid the distortion of arterioles or the disruption of vascular poles. During the microdissection of afferent arterioles with macula densa attached, the thick ascending limb of the 1oop of Henle and the distal convoluted tubule were each gently pulled off at a point less than 50 m from the macula densa. In protocols 1 through 6, approximately ten afferent arterioles alone were microdissected, and divided into control and experimental groups of five afferent arterioles each. The afferent arterioles in each group were transferred into a small plastic

medium used during the experimental period. 4. Prostaglandin '2 PG!2 (final concentration 1.0 x iO M) was added to the incubation medium used during the experimental period. It was dissolved in a solution containing nine parts of ethanol and one part of 0.05 M tris buffer (pH 9.0) and stored at —70°C (3.0 x lO— M). This stock solution was diluted

Elmsford, New York, USA), and the ladles were preincubated at 37°C in 7 ml of oxygenated medium containing 0.1% BSA (medium-0. 1% BSA) for 35 minutes. After preincubation, each ladle was rinsed, blotted, and transferred into a small plastic microtube containing 100 d of oxygenated medium-0. 1% BSA. The gas layer above the incubation medium was replaced with 95% 02 and 5% C02, and the microtubes were covered tightly. The afferent arterioles were incubated in medium-0. 1% BSA for the first 15 minutes (control period), after which the ladles were transferred to fresh medium-0. 1% BSA containing either the

period than during the experimental period (Results), we tested

drug to be tested or its vehicle for the second 15-minute

concentration of 5.0 x l0 M with oxygenated medium-0. 1% BSA, and it was used immediately. 6. Six-keto-prostaglandin E,. Six-keto-PGE1, an active metabolite of PGI2, was added to the incubation medium of the experimental period (final concentration 1.0 x l0— M). It was

to 1.0 x l0— M with ice-cold 0.5 M tris buffer, pH 9.0 and further diluted to 1.0 x 10 M with oxygenated medium-0. 1% BSA immediately before use.

5. Carba-prostaglandin '2 Since renin responses to both ladle that had a nylon mesh bottom (54 sm; Tetko, Inc., arachidonic acid and PG!2 were greater during the recovery

incubation (experimental period). Following the experimental period, arterioles were again placed in medium-0. 1% BSA for the 15-minute recovery period. In protocols 7 and 8, approximately five afferent arterioles both with and without macula densa attached were microdissected and incubated for control and experimental periods as described above; each incubation period was 20 minutes, and the recovery period was omitted. Incubation medium left in the microtube was frozen (—20°C) until the renin assay was performed. After serial incubations

whether the continued increases in renin release during the recovery period are mediated by metabolite(s) of PGI2. CarbaPGI2, a synthetic analog of PG!2. was added to the incubation medium used during the experimental period (final concentra-

tion 5.0 x l0 M). Although carba-PGI2 may be metabolized [16, 17], at least some of the metabolites of PGI2, such as 6-keto-PGF1 and 6-keto-PGE1, will not be formed from carbaPG!2. Carba-PG!2 was dissolved in 100% ethanol (5.0 x 102 M) and stored at —20°C. This stock solution was diluted to the final

dissolved in 100% ethanol, stored at —20°C and protected from

exposure to light. Before use, ethanol was blown out and the residue was dissolved with medium-0.1% BSA. 7. Time control. Afferent arterioles with and without macula densa were incubated in MEM containing 0.1% BSA for two were completed, it was confirmed microscopically that all 20-minute periods; the recovery period was omitted. MEM had microdissected structures remained in the ladle. For the deter- the following composition: Na, 141 mEq/liter; K, 5.4 mEq/ mination of tissue renin content, these incubated afferent arte- liter; Cl, 125 mEq/liter; Ca2, 3.6 mEq/liter: HC03, 23.8 mM; rioles with or without macula densa were placed in 1 ml of H2P04, 1.0 mM.

1140

Ito et a!: Prostaglandin and renin release 12

8. Prostaglandin E2. Afferent arterioles, both with and without macula densa, were incubated in MEM containing 0.1% BSA for the control period and then placed in medium containing PGE2 (final concentration 1.4 X 10 M) for the experimental period.

Determination of renin concentration 9

Renin concentration in the incubation media and tissue samples was measured as previously described [12]. Briefly, incubation media and solutions of tissue samples were incubated with partially purified rabbit substrate equivalent to 1,000 ng angiotensin I (Al) at 37°C for three hours (pH 6.5). Gener-

ated Al was measured by radioimmunoassay. Substrate was prepared by ammonium sulfate fractionation of the plasma obtained from rabbits nephrectomized 48 hours previously. Substrate had specific activity ranging from 300 to 700 ng Al/mg protein, and had no detectable renin or angiotensinase activity. AT generation was linear with respect to renin concentration and incubation time up to four hours, and was not affected by any of the drugs tested in this study. Renin release rate was calculated as ng Al generated per hour per arteriole per hour incubation of arteriole (with or without macula densa), and was expressed as ng A! . hr' 'AF'/hr.

6

.

C—

C

3

Tissue renin content of a single arteriole (with or without macula densa) was calculated and expressed as ng Al. hr 1/ AF.

a

Statistics All data were expressed as mean SEM. For protocols 1 through 6, profile analysis was used to test whether changes from control to experimental to recovery periods were different between groups 1181. Analysis of variance was used to test whether the values during the recovery period were different between protocol 6 (carba-PGI2) and protocol 4 (PGI,). Student's unpaired t-test was used for the statistical evaluation in groups 7 and 8. P < 0.05 was considered significant.

Fig. 1. Effects of pretreat,nent with 9,II-azoprosta-5,13-dienoic acid

Results

experimental period. The change in renin release caused by arachidonic acid in no pretreatment group (•; N = 7) was significantly (P < 0.0002) different from that in the azo analog I pretreatment group.

Time control. Since the vehicles caused no change in renin release in any of the time control groups, the data from each time control experiment were combined. Renin release rates from afferent arterioles during the control, experimental and recovery periods were 0.65 0.07, 0.76 0.10 and 0.88 0.11 . AFVhr, respectively (N = 45). Arteriolar renin ng AT . content was 24.3 1.9 ng Al . hr'/AF. Renin release during the control period was 3.67 0.86% of arteriolar renin content per hour.

hr

Arachidonic acid. When arachidonic acid was added to afferent arterioles, renin release rate increased from 1.04 0.21 to 3.12 0.82 ng A! . hr' . AF/hr and continued increasing to 9.53 1.76 ng A! . hr' . AF'/hr during the recovery period (P < 0.0001 compared with the time control group; Fig. 1). Arteriolar renin content was 19.5 1.5 ng Al . hr'/AF. PGI2 synthetase inhibitor plus arachidonic acid. Pretreatment with azo analog 1 did not affect basal renin release rate (0.40 0.06, 0.50 0.12 and 0.48 0.07 ng Al . hr' . AF'/ hr in the control, experimental and recovery periods, respectively; these values are not statistically significant from those of

0 Control

Experimental

Recovery

Incubation period

(azo analog 1), a PG!2 s)'nthetase inhibitor, on basal and arachidonic acid-induced renin release .from afferent arterioles. Azo analog I at 2.8 )( 106 M was present in the incubation media used during the preincubation, control, experimental and recovery periods. Arachidonic acid (•; N 6; final concentration 1.4 x l0- M) or its vehicle (0; N = 5) was added to the incubation medium used during the

Prostaglandin '2 When PG!2 was added to afferent arte-

rioles, renin release increased from 0.45 0.14 to 1.48 0.53 ng Al . hr' AF'/hr and continued increasing to 4.50 1.24 ng A! hr . AF/hr during the recovery period (P < 0.0001 compared with the time control group; Fig. 2). Arteriolar renin content was 35.5 5.5 ng A! . hr'/AF.

Carba-prostaglandin '2 When carba-PGI2 was added to

afferent arterioles, renin release increased from 0.88 0.31 to 0.49 ng A! . hr' . AF1/hr; renin release stayed at the 1.75

same level during the recovery period (1.81

hr'

0.59 ng

. AF'/hr; Fig. 2). This change was significant (P < A! . 0.0001) compared with the time control group. The increase (A) in renin release from the control to the experimental period was

0,20 ng Al. hr ' . AF/hr in the carba-PGI2 group, which was not different from that caused by PG!2 (1.04 0.45 ng A!. hr . AF/hr). However, the changes in renin release over the entire time course were significantly different between the carba-PGI2 and PG!, groups (P < 0.0001, by profile analy0.87

time control). However, it prevented the increase in renin release induced by arachidonic acid at 1.2 x iO M (Fig. 1). sis). Furthermore, renin release rates during the recovery

1141

Ito et al: Prostaglandin and renin release Table 1. Effect of prostaglandin E2 on renin release from afferent arterioles with and without attached macula densa

Renin release rate

ng Al. hr' AF'/hr No.

Group Afferent arterioles Time control

POE2 (1.4 x 1O M)

Control

Experimental

8 8

1.13 1.02

0.33 0.35

1.32 1.17

0.40

7 8

0.35 0.54

0.06* 0.09

0.36

0.06* 0.24

0.37

P

Tissue renin content

ng Al. hr'/AF

NS

33.4 35.8

5.9

<0.02

49.0 51.0

4.9

11.1

Afferent arterioles with

macula densa Time control

PGE2 (1.4 x iO M)

1.26

13.0

No. refers to the number of experiments in each group. P values were obtained by comparing the changes in renin release from the control to the experimental period between the time control and PGE2 groups. aP < 0.05 compared with afferent arterioles alone.

0.0004) compared with the time control group. Arteriolar renin content was 118.3 58.0 ng Al. hr '/AF. This apparently high arteriolar renin content was due to one experiment (287.0 ng

6

5

Al. hC'/AF), and the value of this group was not statistically different from that of the time control group.

4

Time control. Renin release rates from afferent arterioles with and without macula densa were initially 0.35 0.06 and 1.13 0.33 ng A! . AF/hr, respectively, and they remained stable during the next incubation period (Table 1). Although the

hr'

I-

tissue renin content of afferent arterioles with macula densa attached was not different from that of afferent arterioles alone,

renin release rate was significantly less in afferent arterioles with macula densa attached (P < 0.05). Thus, the ratio of renin release to tissue renin content was significantly smaller (P < 0.05) in the presence (0.58 0.11%) than in the absence of the macula densa (4.32 1.51%). Prostaglandin E2. When PGE2 was added to afferent arterioles alone, there was no significant change in renin release (P > 0.05 compared with the time control group, Table 1). However, when PGE2 was added to afferent arterioles with macula densa attached, renin release rate increased by 133% (P < 0.02 compared with the time control group; Table 1, Fig. 3).

Co.

icJ a)-.

I.

3

2

1'

Discussion U

Control

Recovery Experimental Incubation period = 9), 6-keto-PGE, (a metabolite ofPGI2;

Fig. 2. Effects of PG!2 (; N N = 4) and carba-PGI2 (a synthetic analog of PGI2; 0; N = 10) on renin release from afferent arterioles. PG!2 (1.0 x io— M), 6-keto-PGE

(1.0 x l0— M) or carba-PG!2 (5.0 X l0 M) was added to the

We have previously shown that microdissected afferent arterioles with and without macula densa attached can be used for the study of renin release [12—151. The preparation of afferent arterioles alone allows us to study renin release in the absence of glomeruli, tubules and macula densa, while the preparation

with macula densa can be used to study the influence of attached macula densa on renin responses to various subinduced by PG!2 was significantly different from that induced by 6-keto-PGE1 or carba-PGI2 (P < 0.0001 by profile analysis). *JJ < 0.01 stances. In previous studies [13, 14], we reported that the incubation medium used during the experimental period. The change compared with carba-PGI2 group by analysis of variance.

period were also significantly (P < 0.01) different between the two groups by analysis of variance (Fig. 2). Arteriolar renin content of the carba-PGI2 group was 20.6 4.85 ng Al . hr 1/ AF. Six-keto-prostaglandin E1. When 6-keto-PGE1 was added to afferent arterioles, renin release increased from 0.54 0.07 to

attached macula densa has an inhibitory action on renin release, which appears to be mediated by its endogenous production of adenosine. The present study was undertaken first to examine the interaction between PGs and renin release at the local level

of afferent arterioles, and second to examine whether the macula densa mediates PGE2-induced renin release. Our results suggest that PG!2 and possibly its metabolite(s), synthesized in the afferent arterioles, stimulate renin release by acting directly

on the juxtaglomerular cells, whereas POE2 stimulates renin . AF/hr and returned toward the release indirectly via the macula densa. 0.33 ng Al . Three major renal tissues have been reported to contain 0.13 ng basal value during the recovery period (0.77 Al. hr' . AF'Ihr; Fig. 2). This change was significant (P < cyclooxygenase and be capable of PG biosynthesis: 1) the 1.80

hr

Ito et a!: Prostaglandin and renin release

1142

chloride transport at the macula densa. However, no previous study has reported direct evidence for the involvement of the macula densa portion" of the distal tubule in PGE2-induced

400

renin release. In the present study, PGE2 increased renin

*

300 -

a (V..,,

cn,CV

a

release only in the presence of the attached macula densa. This observation may be considered the first direct evidence that the macula densa mediates PGE,-induced renin release. The mechanism by which PGE2 stimulates renin release could be due to the inhibition of sodium chloride transport at the macula densa. In order to suppress PGI2 synthesis, we pretreated afferent arterioles with azo analog 1 which is reported to inhibit both

P012 and thromboxane A2 synthetases [5, 7, 22]. In rabbit

200 -

kidney slices [71 and isolated rat glomeruli [5], it is reported that azo analog 1 suppressed PGI2 synthesis induced by arachidonic

P

acid, while production of PGE2 and PGF remained intact. !n the present study, the pretreatment with azo analog 1 prevented

the increase in renin release induced by arachidonic acid. 100

0—

I

I

Control

PGE2

1.4 x iOM Incubation period

3. Effect of pro staglandin E2 on renin release from afferent arterioles alone (0; N 8) and afferent arterioles with nacula densa attached •; N = 8). *p < 0.02 compared with the time control group. Fig.

Similar effects of azo analog 1 on renin release have also been reported in rabbit kidney slices [7] and isolated rat glomeruli [5]. While azo analog 1 suppresses thromboxane A2 synthesis, its effect on renin release in the present study is likely due to the inhibition of PG!2 synthesis since rabbit renal cortex does not synthesize detectable levels of thromboxane A2, even in the presence of exogenous arachidonic acid [7]. Despite the fact that PGI2 is rapidly hydrolyzed to 6-ketoPGF1,,, an inactive PG, in an aqueous solution at a physiolog-

ical pH and temperature [23], prolonged duration of PGI2induced renin release was observed in kidney slices [6]. It has been proposed that such prolonged duration may be mediated by metabolite(s) of PG!2 [24, 25]. In the present study, renin

release from afferent arterioles continued to increase during the vasculature including glomeruli, 2) the collecting tubule and 3) recovery period in both arachidonic acid and P012 groups. This the medullary interstitial cells [19]. The renal vasculature is the is in contrast with the observation that renin release induced by main tissue that synthesizes PG!2 [8, 9], and it has been 6-keto-PGE, in the present study or isoproterenol in our previsuggested that compartmcntalization of PG!2 synthesis within ous study [13] returned toward the basal value during the the vasculature is important for the regulation of renin release recovery period. In order to examine whether these continued [10], Beierwaltes et al [51 reported that in isolated rat glomeruli, increases during the recovery period are mediated by metabolic the stimulation of renin release elicited by arachidonic acid transformation of PG!2, we tested the effect of carba-PG!2, a correlates with the production of PG!2 but not PGE2. However, synthetic analog of PG!2. It has been shown that carba-PG!2 since only 3 to 10% of glomeruli have their attendant afferent mimics the action of PGI2 in various preparations although it is

arterioles in this preparation [20], most of the PGs would be produced by the glomeruli, which could in turn influence renin release. In the present study, the preparation of afferent arterioles alone is free from all other renal tissue, and, therefore, the results show a direct interaction between PGs and renin release at the local level of afferent arterioles. Renin release from microdissected afferent arterioles increased with the administration of arachidonic acid, and this increase was completely blocked by pretreatment with azo analog 1, a PG!2

less potent than PG!2 (approximately one-tenth) [26—28]. During

the experimental period, carba-PGI2 (5.0 X l0— M) increased renin release to a level similar to that caused by PG!2 (1.0 X

i0 M); however, during the recovery period renin release did not increase further. Thus renin response induced by carbaPG!2 was significantly less than that induced by PG!2 during the recovery period. This result may suggest that at least part of the

renin response to PGI, is mediated by its metabolite(s). It is not known from the present study which metabolite(s) synthetase inhibitor. This may suggest that renin release may be may be responsible for the continued increase in renin release regulated by PGI2 which is locally synthesized in the afferent induced by arachidonic acid and PG!2. The only active metabolite known thus far is 6-keto-PGE1, which is formed from P012 arterioles. PGE2 is a principal PG synthesized by renal tubular and and probably from 6-keto-PGF1,, via the 9-hydroxyprostamedullary interstitial cells [10]. It may be possible that PGE2 of glandin dehydrogenase pathway [29]. The activity of this entubular and medullary origins could participate in the regulation zyme is greatest in the cortex and negligible in the papilla, of renin release by altering the electrolyte transport at the corresponding to the zonal distribution of renin in rabbits [24, macula densa. It has been shown that PGE2 inhibits chloride 30]. Since 6-keto-PGE1 is more stable [29, 31] and its pharmatransport in the medullary thick ascending limb of the loop of cological actions are similar to PG!2 both quantitatively and Henle [21], and Kotchen, Welch and Ott [11] proposed that qualitatively [24, 25, 3 1—33], it has been suggested that 6PGE2 may stimulate renin release by inhibiting absorptive keto-PGE1 may mediate a part of the renin-releasing action of

Ito et a!: Prostaglandin and renin release

PGI2 [24, 25]. In the present study, 6-keto-PGE1 increased renin release from afferent arterioles, reaching a maximum of 1.80 ng

Al. hr' . AF

6. WHORTON AR, MisoNo K, HOLLIFIELD J, FROLICH JC, INAGAMI

T, OATES JA: Prostaglandins and renin release: I. Stimulation of renin release from rabbit renal cortical slices by PG!2. Prostaglan-

'/hr during the experimental period, while the

maximum induced by the same concentration of P012 was 4.50 ng Al . hr' . AF/hr (during the recovery period). Therefore,

even if we assumed that PGI2 was totally converted to 6keto-PGE1, the large increase in renin release during the

1143

7.

dins 14:1095—1104, 1977 WHORTON AR, LAZAR JD, SMIGEL MD, OATES JA: Prostaglandin-

mediated renin release from renal cortical slices, in Advances in Prostaglandin and Thromboxane Research, edited by B SAMUELSSON, PW RAMwELL, R PAOLETT, New York, Raven Press, vol 7,

1980, pp. 1123—1 129 recovery period in the PG!2 group may not be accounted for 8. TERRAGNO NA, TERRAGNO A, EARLY JA, ROBERTS MA, MCGIFF solely by formation of 6-keto-PGE1. However, it could be JC: Endogenous prostaglandin synthesis inhibitor in the renal possible that PG!2 and formed 6-keto-PGE1 acted synergisticortex: Effects on production of prostacyclin by renal blood vessels. C/in Sci Mo! Med 55: l99S—202S, 1978 cally in activating the intracellular mechanism of renin release, leading to the larger renin release during the recovery period. 9. CHAUDHARI A, KIRSCHENBAUM MA: A rapid method for isolating rabbit renal microvessels. Am J Physiol 254:F291—F296, 1988 As an alternative or additional mechanism, increased renin 10. MCGIFF JC, WONG PY-K: Compartmentalization of prostaglandins release during the recovery period may reflect prolonged acand prostacyclin within the kidney: Implications for renal function. tions of PGI2 on intracellular levels of cyclic AMP and/or Fed Proc 38:89—93, 1979

calcium.

11. KOTCHEN TA, WELCH WJ, OTT CE: The renal tubular signal for

renin release. J Hypertens 2(Suppl I):35—42, 1984 In contrast to the present study, McGiff, Spokas and Wong [24] and Spokas, Wong and McGiff [25] reported that in rabbit 12. ITOH S. CARRETERO OA, MURRAY RD: Renin release from isolated afferent arterioles. Kidney mt 27:762—767, 1985 kidney slices, 6-keto-PGE1 -mediated renin release remained 13. IToH S, CARRETERO OA: Role of the macula densa in renin release. elevated during the recovery period whereas PGI2-mediated Hypertension 7 (Suppl !):!49—154, 1985 release did not, which was more evident when endogenous 14. IToH 5, CARRETERO OA, MURRAY RD: Possible role of adenosine in the macula densa mechanism of renin release in rabbits. J C/in prostaglandin synthesis was inhibited. The discrepancy beInvest 76: 1412—1417, 1985 tween their studies and ours may be due to the difference in the 15. ITOH S, ABE K, NU5HIR0 N, OMATA K, YASUJIMA M, YOSHINAGA experimental preparations employed; especially the presence of K: Effect of atrial natriuretic factor on renin release in isolated tubular components may not only hamper the rinsing of added afferent arterioles. Kidney mt 32:493—497, 1987 prostaglandins but also affect their metabolism. 16. MCGUIRE JC, MORTON DR. BROKAW FC, SUN FF: Metabolism of In conclusion, the results of the present study suggest that (5E)-6a-carbaprostaglandin '2 by rhesus monkey lung 15-OH prostaglandin dehydrogenase. Prostaglandins 21:609—614, 1981 arachidonic acid stimulates renin release through the formation of PG!2 and possibly its metabolite(s) in the afferent arterioles. 17. JARABAK J, BERKOWITZ D, SUN FF: Oxidation of prostacyclin and its analogs by three 15-hydroxyprostaglandin dehydrogenases. These arachidonic acid metabolites act directly on the afferent Prostaglandins 28:509—516, 1984 arterioles where they are synthesized. On the other hand, PGE2 18. MORRISON DF: Multivariate Statistical Methods (Profile Analysis). stimulates renin release indirectly via the macula densa. This 2nd ed, New York, McGraw-Hill Co., 1976 may suggest that PGE2 synthesized in renal tubular or medul- 19. SMITH WL, GRENIER FC, BELL TG, WILKIN GP: Cellular distribution of enzymes involved in prostaglandin metabolism in the lary interstitial cells could participate in the control of renin mammalian kidney, in Prostaglandin in Cardiovascular & Renal release by means of the macula densa. Function, edited by A SCRIABINE, AM LEFER, SA KUEHL JR,

Acknowledgments

This study was supported by National Institutes of Health grant number HL28982 and by grants from the Ministry of Education, Science and Culture of Japan (61132005 and 62870011). The Upjohn Pharmaceutical Company kindly donated the PGI2 synthetase inhibitor,

9,11-azoprosta-5,13-dienoic acid, used in this study. The authors are grateful to Dr. Frank F. Sun, Upjohn Company, Kalamazoo, Michigan, USA, for the information about the metabolism of carba-PGL2, and to Dr. Mike VanRollins, Hypertension Research Division, Henry Ford Hospital, Detroit, Michigan, USA, for helpful discussion. Reprint requests to Sadayoshi Ito, M.D., Ph.D., Hypertension Research Division, Henry Ford Hospital, 2799 W. Grand Boulevard, Detroit, Michigan 48202, USA.

Jamaica, New York, Spectrum Publications Inc. 1980, pp. 71—91 20. KREISBERG JI, KARNOVSKY MJ: Glomerular cells in culture. Kidney Jut 23:439—447, 1983 21. STOKES JB: Effect of prostaglandin E2 on chloride transport across the rabbit thick ascending limb of Henle. J Clin Invest 64:495—502, 1979 22. FITZPATRICK FA, GORMAN RR: A comparison of imidazole and

9,ll-azoprosta-5,13-dienoic acid. Two selective thromboxane synthetase inhibitors. Biochim Biophys Acta 539:162—172, 1978 23. JOHNSON RA, MORTON DR, KINNER JH, GORMAN RR, MCGUIRE JC, SUN FF, WHITTAKER N, BUNTING S, SALMON J, MONCADA S,

VANE JR: The chemical structure of prostaglandin x (prostacyclin). Prostaglandins 12:915—928, 1976 24. MCGIFF JC, SPOKAS EG, WONG PY-K: Stimulation of renin release

by 6-oxo-prostaglandin E1 and prostacyclin. Br J Pharmaco! 75: 137—144, 1982

25. SPOKAS EG, WONG PY-K, MCGIFF JC: Prostaglandin-related renin

References

release from rabbit renal cortical slices. Hypertension 4 (suppl II): 1196—11100, 1982

I. FRANCO-SAENZ R, SUZUKI S, TAN SY: Prostaglandins and renin

production. A review. Prostaglandins 20:1131—I 143, 1980 2. FREEMAN RH, DAvis JO, VILLARREAL D: Role of renal prostaglan-

dins in the control of renin release. Circ Res 54:1—9, 1984 3. HENRICH WL: Role of prostaglandins in renin secretion. Kidney mt 19:822—830, 1981

4. KEETON TK, CAMPBELL WB: The pharmacologic alteration of renin release. Pharmacol Rev 32:81—227, 1980 5. BEIERWALTES WH, SCHRYVER 5, SANDERS E, STRAND J, ROMERO

JC: Renin release selectively stimulated by prostaglandin 12 in isolated rat glomeruli. Am J Physiol 243:F276—F283, 1982

26. WHITTLE BJR, MONCADA 5, WHITING F, VANE JR: Carbacyclin—

a potent stable prostacyclin analogue for the inhibition of platelet aggregation. Prostaglandins 19:605—627, 1980 27. AIKEN JW, SHEBUSKI Ri: Comparison in anesthetized dogs of the

anti-aggregatory and hemodynamic effects of prostacyclin and a chemically stable prostacyclin analog, 6a-carba-PGI2 (carbacyclin). Prostaglandins 19:629—643, 1980 28. WHITTLE BJR, MONCADA 5: Platelet actions of stable carbocyclic analogues of prostacyclin. Circulation 72:1219—1225, 1985 29. WONG PY-K, MALIK KU, DE5IDERI0 DM, MCGIFF JC, SUN FF: Hepatic metabolism of prostacyclin (PGI2) in the rabbit: Formation

1144

Ito et at: Prostaglandin and renin release

of a potent novel inhibitor of platelet aggregation. Bioc/iem Biophys Res Comm 93:486—494, 1980

30. MooRE PK, HOULT JRS: Prostaglandin metabolism in rabbit kid-

ney: Identification and properties of a novel prostaglandin 9hydroxydehydrogenase. Biochim Biophys Ada 528:276—287, 1978

31. QUILLEY CP, McGwF JC, LEE WH, SUN FF, WONO PY-K: 6-keto-PGE1: A possible metabolite of prostacyclin having platelet antiaggregatory effects. Hypertension 2:524—528, 1980

32. JACKSON EK, HERZER WA, ZIMMERMAN JB, BRANCH RA, OATES

JA, GERKENS JE: 6-keto-prostaglandin F1 is more potent than prostaglandin 12 as a renal vasodilator and renin seeretagogue. J PharmacolExp Ther2l6:24—27, 1981 33. SCHWERTSCHLAG U, STAHL T, HACKENTHAL F: A comparison of

the effects of prostacyclin and 6-keto-prostaglandin F1 on renin release in the isolated rat and rabbit kidney. Prostaglandins 23:129— 138, 1982