Glomerular Actions of a Free Radical-Generated

Transcription

1 Glomerular Ations of a Free Radial-Generated Novel Prostaglandin, 8-epi-Prostaglandin F2,,, in the Rat videne for Interation with Thromboxane A2 Reeptors Kihito Takahashi,* Tarek M. Nammour,* Megumu Fukunaga,* Joan bert,* Jason D. Morrow,I L. Jakson Roberts II,' Rihard L. Hoover,* and Kamal F. Badr* Departments ofmediine,* Pathology,* and Pharmaology,* Vanderbilt University Shool ofmediine, Nashville, Tennessee Abstrat 8-epi-prostaglandin F2. (8-epi-PGF2.) and related ompounds are novel prostanoid produed by a nonylooxygenase mehanism involving lipid peroxidation. Renal ishemia-reperfusion injury inreased urinary exretion of these ompounds by 3% over baseline level. Intrarenal arterial infusion at.5, 1, and 2,gg/kg per min indued dose-dependent redutions in glomerular filtration rate (GFR) and renal plasma flow, with renal funtion easing at the highest dose. Miropunture measurements (.5,g/kg per min) revealed a predominant inrease in afferent resistane, resulting in a derease in transapillary hydrauli pressure differene, and leading to redutions in single nephron GFR and plasma flow. These hanges were ompletely abolished or reversed by a TxA2 reeptor antagonist, SQ 29,548. ompetitive radioligand binding studies demonstrated that 8- epi-pgf2. is a potent ompetitor for I3HISQ 29,548 binding to rat renal arterial smooth musle ells (RASM) in ulture. Furthermore, addition of 8-epi-PGF2, to RASM or isolated glomeruli was not assoiated with stimulation of arahidonate ylooxygenase produts. Therefore, 8-epi-PGF2, is a potent preglomerular vasoonstritor ating prinipally through TxA2 reeptor ativation. These findings may explain, in part, the benefiial effets of antioxidant therapy and TxA2 antagonism observed in numerous models of renal injury indued by lipid peroxidation. (J. lin. Invest : ) Key words: glomerulonephritis * free radials * prostaglandins * thromboxane * renal blood flow Introdution The generation and pathophysiologi relevane of reative oxygen metabolites during glomerular injury is well established (1). We have reported the disovery of a series of novel prostaglandin F2 (PGF2)-like ompounds produed by a nonylooxygenase mehanism involving free radial atalyzed lipid peroxidation (2). These ompounds are easily deteted in normal This work was presented in abstrat form at the 23rd Annual Meeting ofthe Amerian Soiety ofnephrology, 2-5 Deember 199, Washington, D. Address orrespondene to Kihito Takahashi, M.D., S-3223 Medial enter North, Vanderbilt University Medial enter, Nashville, TN Reeivedfor publiation 4 September 1991 and in revisedform 3 Deember 1991 J. lin. Invest. The Amerian Soiety for linial Investigation, In /92/7/136/6 $2. Volume 9, July 1992, human plasma and urine, and their prodution is markedly enhaned in animal models with free radial-indued injury (2). Further, we examined the ations of one of these ompounds, 8-epi-prostaglandin F2, (8-epi-PGF2.), in normal rat kidney, and demonstrated extremely potent and seletive renal vasoonstritive effets in the low nanomolar onentration range (2). The fat that these prostanoids are present in urine and their formation inreases in situations involving free radial-indued injury, ombined with the potent renal onstritor ation demonstrated for 8-epi-PGFu,, suggests a potential role for this family of eiosanoids as novel mediators of the renal funtional deterioration that aompanies free radial-indued glomerular injury (1, 2). In the present study, using miropunture tehniques, we examined the preise responses of glomerular hemodynamis to intrarenal arterial administration of 8-epi-PGF2, in vivo in the normal rat. In addition, we assessed the role of thromboxane reeptors in mediating its renal effets both in vivo and in vitro. Furthermore, to investigate the pathophysiologi relevane of these ompounds, we assessed their urinary exretion rate in ishemi aute tubular nerosis, an oxygen free radialmediated experimental model of renal injury (3, 4). Methods 8-epi-PGF2, (gift from Dr. Gordon Bundy, The Upjohn o., Kalamazoo, MI). Aliquots of 8-epi-PGF2. were dissolved in ethanol and were stored at -7. Immediately before use, they were diluted with.1 N PBS to the required onentration. iosanoid purity and onentration in stok solutions and in syringes used for intrarenal injetions were verified periodially by gas hromatography/negative ion mass spetrometry as previously desribed (5). Physiologial measurements. learane studies were performed on anesthetized adult male Munih-Wistar rats weighing 2-23 g prepared aording to the protools desribed previously (6). In brief, after Inatin anesthesia (Andrew Lokwood & Assoiates, Sturtevant, WI), the left femoral artery atheter was used to sample blood and to monitor mean systemi arterial pressure (MAP)' by means of a pressure transduer (model P23Db, Statham Instrument; Oxnard, A) onneted to a diret writing reorder (Gould In. Instruments Div., leveland, OH). After a traheostomy, P-5 atheters were inserted into both jugular veins for infusions of plasma, inulin, and paraaminohip- 1. Abbreviations used in this paper: FF, filtration fration; GFR, glomerular filtration rate; Ht, hematorit; Kf, glomerular apillary ultrafiltration oeffiient; MAP, systemi arterial pressure; PAH, paraaminohippuri aid; P, efferent arteriolar pressure; PG, glomerular apillary pressure; PT, proximal tubular pressure; RA, afferent arteriolar resistane; RASM, renal arterial vasular smooth musle ells; RBF, renal blood flow; RPF, renal plasma flow; R, efferent arteriolar resistane; SNGFR, single nephron glomerular filtration rate; SNFF, single nephron filtration fration; SNPF, single nephron glomerular plasma; V, urine flow rate. 136 Takahashi, Nammour, Fukunaga, bert, Morrow, Roberts, Hoover, and Badr

2 puri aid (PAH) (7.5% and 1.6% solution, respetively, in.9% Nal at 1.2 ml/h), and other experimental solutions. The left kidney was exposed by a left subostal inision, separated from the surrounding fat, and suspended on a Luite holder. The kidney surfae was bathed with warm isotoni Nai. A 3-gauge needle was plaed in the abdominal aorta at the takeoff ofthe left renal artery, through whih a maintenane infusion of.9% NaI was initiated at a rate of.25 ml/min. Homologous rat plasma was administered intravenously at a rate of 1 ml/kg per h for 45 min, followed by a redution in an infusion rate to 1.5 ml/kg per h for the remainder of the experiment. Previously, this protool has been shown to adequately replae surgially indued plasma losses, thus maintaining euvolemia (7). In all experiments, measurements were started 6 min after the onset of plasma infusion and arried out as follows: one to three samples of urine from the experimental kidney were olleted, eah over 1 min, for the determination of urine flow rate (V), inulin and PAH onentrations, and for the alulation of whole kidney glomerular filtration rate (GFR) and renal plasma flow (RPF). For these urine olletions, indwelling ureteral atheters (P 1) were used. oinident with these urine olletions, two or three samples of femoral arterial blood were obtained in eah period for determination of hematorit (Ht) and plasma onentration ofinulin and PAH. Whole kidney inulin and PAH learanes were performed during baseline onditions and repeated during administration of 8-epi-PGF2, and other experimental onditions to be desribed. Inulin onentrations in plasma and urine were determined by the maroanthrone method (8). PAH onentrations in urine and plasma were determined aording to the method of Smith et al. (9). xperiments were performed on four groups of rats as follows. Group I: time ontrol (n = 7). In this group, baseline measurements were performed and repeated during a 3-min infusion of vehile (.1 N PBS). Group II: intrarenal arterial 8-epi-PGF2, dose response (n = 21). In this group, whole kidney measurements were performed during baseline onditions and during the infusion of 8-epi-PGF2, in the following doses: Group IIA:.5 Ag/kg/min (n = 7); group IIB: 1 ug/kg/min (n = 7); group II: 2 Asg/kg/min (n = 7). In our previous study (2), we demonstrated that the plasma onentration of 8-epi-PGF2, 3 min after the initiation of a 5-ytg/kg per min intravenous infusion was 34 nm (- I ng/ml). This onentration is equivalent to that whih is observed in free-radial related injuries in rats (14-indued liver injury and diquat treatment in selenium defiieny). In our urrent study, we hose the doses desribed above, whih theoretially deliver similar intrarenal onentrations of 8-epi-PGF2,,. Sine the experimental period in group II was assoiated with a time-dependent severe and progressive redution in renal perfusion, measurements were performed in one 15-min period. Group III: intrarenal arterial infusion of8-epi-pgf2. in thepresene ofa thromboxane A2 (TxA2) reeptor antagonist SQ 29,548 (1, 11) (n = 7). In this group, 8-epi-PGF2, was administered at a rate of 2 ltg/kg per min (n = 7) in the presene of TxA2 reeptor antagonist SQ 29,548 administered by a sustained infusion at 3 mg/kg per h. Treatment with SQ 29,548 was started 3 min before the initial baseline measurements. Additionally, three rats in group II reeived SQ 29,548 (3 mg/kg per h) 15 min after the initiation of 8-epi-PGF2, infusion (2,ug/kg per min), an experimental period during whih measurements were performed in the presene of 8-epi-PGF2<, alone. For ontinuous monitoring of renal blood flow (RBF), an eletromagneti flow probe was plaed around the left renal artery and onneted to a flow meter (arolina Medial letronis, King, N). Group IV(n = 7): miropunture measurements during8-epi-pgf2. infusion. In this group, glomerular miropunture measurements were performed during baseline onditions and repeated during 8-epi-PGF2, administration at.5 ug/kg per min. All miropunture experiments were performed on anesthetized adult male Munih-Wistar rats that weighed g and were prepared for miropunture aording to the protools desribed previously (12). The preparation is idential to that desribed above for whole kidney learane studies, exept that 3H-inulin (3 gi/experimental period in.9% Nal at 1.2 ml per h) from New ngland Nulear (Boston, MA) was used instead of nonradiolabeled inulin. Miropunture measurements were started 6 min after the onset of plasma infusion and arried out as follows: 2-min samples of fluid were olleted from surfae proximal onvolutions to determine the flow rate and inulin onentration. onomitantly, femoral arterial blood samples were obtained during eah period to determine Ht and plasma protein and inulin onentrations. Three samples of blood were obtained from surfae efferent arterioles (star vessels) to determine efferent arteriolar protein onentration. Time-averaged hydrauli pressures were measured in surfae glomerular apillaries (Po), proximal tubules (PT), and surfae efferent arterioles (P) using a servo-null miropipette transduer system (model 5; Instrumentation for Physiology & Mediine, In., San Diego, A) and miropipettes with outer tip diameters of 2-3,um ontaining 2 M Nal. olloid osmoti pressures of plasma entering and leaving glomerular apillaries, single nephron glomerular filtration rate (SNGFR), single nephron filtration fration (SNFF), glomerular apillary ultrafiltration oeffiient (Kf), resistanes of single afferent (RA) and efferent (R) arterioles, and single nephron glomerular apillary plasma flow (SNPF) were determined using equations desribed in detail elsewhere (13). The onentrations ofinulin in tubular fluid and plasma were determined by measuring the radioativity of 3H-inulin in a sintillation ounter (Bekman Instruments, In., Fullerton, A). Protein onentrations in efferent arteriolar and femoral arterial blood plasmas were determined using a fluorometri method developed by Viets et al. (14). Group V (n = 4): ishemi aute tubular nerosis model. In this group, urine was olleted from the atheter in the left ureter for 15 min under baseline onditions and the left renal artery was lamped for 3 min. 2 min after relief of the lamp from the renal artery, urine was again olleted for 15 min. PGF2<,-ompounds in these samples was analyzed after thin layer hromatography purifiation by gas hromatography/negative ion mass spetrometry as desribed in (5), and the urinary exretion rate of PGF2,-ompounds was expressed in pg/min. from isolated glomeruli iosanoids prodution by 8-epi-PGF2, and rat renal arterial vasular smooth musle ells (RASM). Glomeruli were isolated from rat kidneys as desribed previously (15). RASM were ultured from medial explants of rat main renal arteries by the modified method of Ross (16). The resulting ells were grown in RPMI 164 medium with 1% fasting blood sugar, peniillin (1 U/ml) and streptomyin (1,g/ml) from Gibo Laboratories, Gibo Div. (St. Lawrene, MA), plated into 1 mm ell ulture petri dishes, and inubated at 37 in a humidified atmosphere of95% 2/5% O2. Smooth musle ell olonies were subultured in 6 mm ulture dishes, and experiments were arried out on ells from passages The isolated glomeruli (3, glomeruli per inubation) and RASM (16 ells per inubation) were inubated in RPMI 164 maintained at 37 and 5% O2 for 15 min in the absene or presene of inreasing amount of 8-epi-PGF2.,, (19 to 1-6 M). The preparation was entrifuged and the supernatant was frozen at -7 for TxA2, prostaglandin 12 (PGI2) or prostaglandin 2 (PG2) assay at a later time. The amount of TxB2, 6-keto-PGF,1, the stable metabolites of TxA2, PGI2, respetively, and PG2 in the supernatants were determined by radioimmunoassay (Amersham orp., Arlington Heights, IL). Binding studies. Studies of [3H]SQ 29,548 binding were performed on RASM grown to onfluene in 24-well luster dishes as desribed elsewhere (17, 18). ells were washed one with PBS and then exposed to the appropriate onentration of [3H]SQ 29,548 in Hanks' medium, ph 7.6, ontaining.1% bovine serum albumin. At the ompletion of the experiment, the experimental medium was removed and the ells were washed five times with ie-old PBS. The ells were then dissolved with 1 ml ofn NaOH, neutralized with H1, dissolved in 1 ml of Aquasol (New ngland Nulear, Boston, MA), and the bound radioativity was determined using a sintillation ounter (Bekman Instruments, In., Fullerton, A). Nonspeifi binding was determined by measuring the amount of [3H]SQ 29,548 bound in the presene of 1,-fold exess of Glomerular Ations of8-epi-pgf2,,, 137

3 unlabeled SQ 29,548. ell density was determined by ounting ells from repliate wells, using a oulter ounter (model ZBi, oulter letronis In., Hialeah, FL). The time ourse of speifi [3H]SQ 29,548 binding to RASM was assessed in three experiments in whih 5 nm [3H]SQ 29,548 was inubated with RASM as desribed above and speifi binding assessed at 1 min to 24 h after addition of agonist. ompetitive binding-inhibition studies were arried out by inubation of RASM with 5 nm [3H]SQ 29,548 and addition, at equilibrium, of inreasing onentrations of 8-epi-PGF2,, or unlabeled SQ 29,548. Statistial analyses. Student's paired t test was performed to ompare within eah group the hanges in various whole kidney and miroirulatory indies that ourred from the baseline to the experimental periods. Intergroup multiple omparisons were performed with a oneway analysis of variane followed by the Newman-Keuls test. Differenes were onsidered signifiant at P <.5. All values are reported as means±s. Results In vivo studies. Values reported in this setion represent the means of repeated measurements performed during the baseline and experimental periods, unless otherwise stated in Methods. The miropunture values reported are the means of repeated measurements performed early and late in the ourse of observed responses. In group I animals (time ontrols), administration of 8-epi- PGF2,, vehile for a 3-min period was not assoiated with signifiant hanges in any ofthe physiologi parameters monitored. hanges in Ht, MAP, GFR, and RPF from the baseline to the experimental period were: Ht, 48.1±.6 to 47.5±.6 vol% (NS); MAP, 18±3 to 16±4 mmhg (NS); GFR,.93±.7 to.92±.7 ml/min (NS), whih is orrespondent to 1.3±1.2% fall; and RPF, 4.44±.29 to 4.38±.32 ml/min (NS) (1.6±2.2%). In previous studies, it was shown that repeated measurements of single nephron funtions do not reveal any signifiant hanges over this period of time in the presene of stable systemi and whole kidney hemodynamis (19). In group II rats, administration of inreasing doses of intrarenal arterial 8-epi-PGF2,, infusion was not assoiated with signifiant hanges in Ht and MAP, exept for a slight inrease in group II animals. Ht hanged from 46.7±.6 to 46.4±.5 vol% in group IIA (NS), from 45.8±.7 to 45.5±.6 vol% in group IIB (NS), and from 46.8±.5 to 46.5±.3 vol% in group II (NS). MAP hanged from 99±3 to 98±3 mmhg in group IIA (NS), from 17±4 to 19±4 mmhg in group IIB (NS), and from 16±3 to 112±4 mmhg in group II (P <.1). Despite the absene of marked hanges in these systemi parameters, the administration of 8-epi-PGF2,t was assoiated with signifiant and dose-dependent redutions in GFR and RPF (Fig. 1). GFR fell from.98±.7 to.69±.9 ml/min (P <.5), whih is orrespondent to 25.8±3.% fall in groups IIA; from 1±.5 to. 51±.4 ml/min (P <.5) (49.1±4.1% fall) in group IIB; from.95±.4 to.27±.5 ml/min (P <.5) (72.4±5.1% fall) in group II. RPF fell from 4.43±.36 to 3.46±.31 ml/min (P <.5) (2.8±3.4% fall) in group IIA; from 4.85±.39 to 2.89±.41 ml/min (P <.5) (43.3±4.7% fall) in group IIB; from 4.83±.16 to 1.76±.29 ml/min (P <.5) (62.4±6.% fall) in group II. A proportionately greater derease in GFR ompared to RPF resulted in a fall in filtration fration (FF) that was not statistially signifi- & o epi-PGF2a, gg/kg/min,. I. -2- a.4- - A -.) Bepi.PGF2a, pg/kg/min Figure 1. Dose-dependent redutions in GFR and RPF aused by intrarenal arterial infusion of 8-epi-PGF2,, at a rate of.5, 1, and 2,ug/kg/min. Asterisks indiate signifiant differene from values observed in vehile-infused rats as shown gg/kg/min (*P <.5, **p <.1). ant at the lower dose of 8-epi-PGF2, (groups IIA and IIB, in whih FF hanged from.23±.3 to.21±.3, NS, and.19±.1 to.17±.2, NS respetively), but was signifiant at the maximum dose in II (.2±.1 to.15+.1, P <.25). xamination of renal responses to this ompound at the single-nephron level in group IV (Fig. 2) revealed that the dereases in whole kidney GFR and RPF were aompanied by parallel redutions in SNGFR and SNPF, the former falling from 47.1±2.2 to 34.8±2.5 nl/min (P <.5), and the latter fell from 24.3±12.6 to 146.3±11.7 nl/min (P <.5). The hange in FF was not signifiant (from.25±.1 to.24±.1 (NS). Aompanying these redutions in SNGFR and SNPF were simultaneous 8-epi-PGF2<,-indued augmentations of both RA and R, whih inreased from 1.9±.11 to 1.79±.19 1' dyn * sm5 (P <.5), and from.8±.9 to 1.11±.11 1' dyn * s * m-5 (P <.5), respetively. The proportionately greater inrease in RA led to a signifiant redution in PG, whih dereased from 45.3±.4 to 4.1±.7 mmhg (P <.5). PT also fell slightly during 8-epi-PGF2,a administration, from 13±1 to 111 ± mmhg (P <.5). Mean net transapillary hydrauli pressure differene (AP) was redued signifiantly from 33.6±.8 to 27.7±.5 mmhg (P <.5). The presene of filtration pressure disequilibrium during 8-epi-PGF2, administration allowed for the alulation of unique values for Kf in all the experimental animals. The administration of 8-epi-PGF2o, was not assoiated with a signifiant hange in the mean value for this parameter, whih hanged from.64±. 1 to.64±.9 nl/(s. mmhg) (NS). When 8-epi-PGF2a was administered in the presene of a speifi TxA2 reeptor antagonist, SQ 29,548, a marked modifiation of the former's renal ations was noted. In the presene ofsq 29,548, 8-epi-PGF2< administration was without a signifiant effet on either Ht or MAP. Ht hanged from 46.5±.8 to 46.3±.7 vol% (NS), and MAP hanged from 13±4 to 15±4 mmhg (NS). In ontrast to the effet of 8-epi-PGF2, infusion alone (group II), however, 8-epi-PGF2, administration (2 jig/kg per min) in the presene of SQ 29,548 was not assoiated with any signifiant hanges in GFR and RPF. GFR hanged from.92±.5 to.89±.4 ml/min (1.8±5.9% fall), and RPF hanged from 4.85±.14 to 5.7±.22 ml/min (3.2±3.9% inrease (Fig. 3). FF was unhanged between the ontrol (.19±.1) and the 8-epi-PGF2, infusion period (.18±.1) (NS). Additionally, baseline measurements in this group (i.e. measurements performed in the presene of SQ 29,548, but before the initiation of 8-epi-PGF2, infusion) were 138 Takahashi, Nammour, Fukunaga, bert, Morrow, Roberts, Hoover, and Badr

4 7-6 - SNGFR 5- so 4-* AP Ln Up D n , 2. 1 ~~~R SNPF RA & RA /* not signifiantly different from those observed in groups I and II, suggesting lak of an independent effet for the TxA2 reeptor antagonist on renal or systemi hemodynamis. In addition, three rats in group II reeived SQ 29,548 (3 mg/kg per h) onomitantly with 8-epi-PGF2, (2 ag/kg per min), after a 15- min experimental period with 8-epi-PGF2, alone. The redutions in GFR (.22±.1, ompared to.92±.8 ml/min in ontrol period, P <.5) and RPF (1.59±.13, ompared with 4.74±.26 ml/min in ontrol period, P <.5) indued by 8-epi-PGF2a were ompletely reversed to baseline level (GFR:.87±.5 ml/min, NS vs. baseline value; RPF: 4.74±.26 ml/min, NS vs. baseline value). FF, redued by 8- epi-pgf2, (.14±.1 vs..19±.1 in ontrol period, P <.5), also returned to baseline level (.19±.1, NS vs. baseline value). Fig. 4 illustrates a representative traing of RBF monitoring from this group. In the four other animals in group II, GFR and RPF deteriorated to naught 3 min after starting 8-epi-PGF2, infusion. Of note, the slightly elevated MAP m %U) ~~~~~~~~~~~~~~~~~~~~... SNFF _.8 T TFigure 2. Summary of the hanges.6 i j NS in glomerular miroirulation be-.4. fore () and after () the adminis-.2 tration of 8-epi-PGF2, at a rate of.5,ug/kg/min. Asterisks indiate. signifiant differene from baseline values (*P <.5, **P <.5). (116±4 vs. 113±4 mmhg in ontrol period, P <.25) was also redued to baseline level (111±4 mmhg, NS vs. baseline value). Additionally, the presene of a ylooxygenase inhibitor, ibuprofen (2 mg/kg i.v.), had no effet on the renal responses to 8-epi-PGF2a (2,g/kg per min) observed in group II animals desribed above (data not shown). In group V animals, the urinary exretion rates of PGF2, ompounds inreased from 21.6±1.4 under basal ondition to 68.2±47.6 pg/min (P <.5) during the reperfusion period (Fig. 5). In vitro studies Seondary release of eiosanoids by 8-epi-PGF2a. The alulated ross-reativity to 8-epi-PGF2, of the antibody for TxB2 was.5%. The results obtained from the experiments were 1 min U- o 9 ;. i * i De K RBF 6 (mi/min) 3 '~~~4 ~ ~~~~ 8-epl-PGF2a 8-epl-PGF2naSO B-e: PGF2,. P3hi:a-P_ --e: _ Figure 3. ffets of a thromboxane reeptor antagonist on the renal ations of 8-epi-PGF2,. In the presene of SQ 29,548 (3 mg/kg per h), the redutions in GFR and RPF aused by 2 gg/kg/min of 8-epi- PGF2a were abolished. Asterisks (**) indiate signifiant differene (P <.1) between hanges aused by 8-epi-PGF2. with and without SQ 29,548. Figure 4. ffet of SQ 29,548 (3 mg/kg per h) on the dereasing RBF by ontinuous infusion of 8-epi-PGF2. (2 ug/kg/min). The dereasing RBF by 8-epi-PGF2_(8-epi indiates the start of 8-epi-PGF. infusion) was immediately reversed by the initiation of onomitant SQ 29,548 infusion (indiated by SQ). Glomerular Ations of8-epi-pgf2. 139

5 I" Figure 5. Urinary exre- 1 D - _tion rate of PGF2a-om- 8G - pounds, indiated as 8-6 epi-pgf2,,, (pg/mmn) j during ontrol (olumn g/, and after release of 4g ),; ~~~~~~~~mn 2 i <so~> - renal artery lamp (ol- R). An asterisk in-..._..i _ diates signifiant dif- ferene from ontrol R values (*P<.5). orreted based on this ross-reativity. The ross-reativities of the antibodies for PG2 and 6-keto-PGFa, were negligible. As shown in Fig. 6, inubation of isolated glomeruli with inreasing amounts of 8-epi-PGF2,, was not assoiated with signifiant hanges in the amount oftxb2 (Fig. 6 A) and PG2 (Fig. 6 B) in the supernatant. No signifiant hange was again noted in the value for 6-keto-PGFIa, (Fig. 6 ) in the supernatant of RASM inubated with inreasing onentrations of 8-epi- PGF2a. TxB2 was undetetable in the RASM supernatant inubated even with the highest onentration of 8-epi-PGF2, (lo-5 M). Binding study. As shown in Fig. 7 A, speifi binding of 5 nm [3H]SQ 29,548 to RASM inreased progressively with time, reahing a maximum at 6 h and remaining stable for up to 24 h. The properties of SQ 29,548 binding sites on RASM with a dissoiation onstant (Kd) of43 nm and speifi binding sites (Bma.) of 2.1 X IO' sites/ell agreed with those observed in the studies using rat aorti smooth musle ells reported previously (17, 18). Fig. 7 B displays the result ofompetitive inhibi- 3 o 2 o 1 A 4 _B, 3-1 d). -, D 's epi -PGF2a Log onentration (M) Figure 6. Glomerular and RASM eiosanoids generation following 15-min inubation of inreasing amounts of 8-epi-PGF2a. (A) Glomerular PG2 generation. (B) Glomerular TxA2 generation. () RASM PGI2 generation. o Time (hours) 1 - n 8- o 6- -,, 4- 'K 2- e Log onentration (M) Figure 7. Reeptor binding studies. (A) Saturation binding of [3HJSQ 29,548 to RASM. losed square: total binding. losed irle: nonspeifi binding. Open irle: speifi binding. (B) ompetitive inhibition of speifi binding of [3HJSQ 29,548 to RASM by 8-epi-PGF2. and non-labelled SQ 29,548. The ontrol value (1%) was defined as the speifi binding of [3H]SQ 29,548 in the absene of these ompounds. ah point is the mean oftripliate measurements in a typial experiment. losed square: 8-epi-PGF2,,. Open square: SQ 29,548. tion studies. Unlabeled 8-epi-PGF2,, and SQ 29,548 ompeted for the speifi binding of [3H]SQ 29,548 to RASM in a onentration-dependent manner and with equivalent poteny. Half-maximal inhibition of binding by 8-epi-PGF2,, was at 1 nm. Disussion The present experiments onfirmed that intrarenal arterial administration of 8-epi-PGF2, exerted dose-dependent redutions ofgfr and RPF. This was assoiated with only minimal hanges in MAP at the highest dose. Of note, a predominant effet of this ompound on GFR, as ompared to RPF, was observed as the administered dose was inreased: this was evidened by a signifiant fall in FF in group II (2 Atg/kg per min). It is interesting to note that the inversion of the stereohemistry at -8 renders 8-epi-PGF2,, a potent renal vasoonstritor, as ompared to PGF2a, whih, at similar doses, has minimal effet, if any, on renal vasular tone in the rat (2, 21). To identify the preise effetor loi of this prostanoid, we performed miropunture measurements only at the lowest dose (.5,Ag/kg per min), beause severe progressive vasoonstrition indued by higher doses of this ompound prevented the performane of reliable miropunture measurements. This examination revealed a predominant augmentation of RA, leading to redutions in SNPF and PG. onsequently, the 8-epi-PGF2a,-indued fall in GFR at a low dose, was most likely beause of a fall in SNPF, as evidened by insignifiant hanges in both FF and Kf. The redution of GFR at a high dose of 8-epi-PGF2,, (2,ug/kg per min), however, is likely to result from another mehanism, probably the fall in Kf, in addition to the fall in plasma flow (simple vasoonstrition, in the light of predominant effets of this eiosanoid on GFR at higher doses, as evidened by the fall in FF in group II). Interestingly, the presene of the speifi thromboxane reeptor antagonist was assoiated with omplete abolition ofthe redutions in GFR, RPF, and FF, and the inrease in MAP observed during 8-epi-PGF2,a, infusion alone (2,g/kg per min). These results indiate strongly the involvement of thromboxane reeptors in the mehanism(s) by whih this prostanoid redues renal perfusion and GFR. In the light ofthe high seletivity of TxA2 reeptor to SQ 29,548 (1), it is not likely that other PG reeptors play a major role in mediating the hemodynami effets of 8-epi-PGF2,,. The partiipation of thromboxane reeptors in the latter mehanism, oupled with the pre- 14 Takahashi, Nammour, Fukunaga, bert, Morrow, Roberts, Hoover, and Badr

6 vious demonstration ofthe apaity oftxa2 to ontrat mesangial ells (22), and the improvement by thromboxane reeptor blokade oflow Kf values in previous studies (15, 23), supports the hypothesis that 8-epi-PGF2a-indued redutions in GFR at higher doses may be partly beause of a signifiant fall in Kf, an effet thought to be mediated through redutions in glomerular apillary surfae area seondary to ontration of mesangial ells (24). Further, subsequent administration of SQ 29,548 during ontinuous infusion of 8-epi-PGF2, prompt and omplete normalizations of GFR, RPF, FF, and MAP. This observation, interpreted as the displaement by a thromboxane reeptor antagonist of 8-epi-PGF2, from thromboxane reeptors on vasular smooth musle and probably on mesangial ells, argues additionally in favor ofthe partiipation of thromboxane reeptors in mediating renal and systemi vasoonstrition exerted by 8-epi-PGF2,. was assoiated with To assess this hypothesis, we performed the ompetitive binding study using vasular smooth musle ells from rat renal artery (RASM). 8-epi-PGF2, was a potent ompetitor for [3H]SQ 29,548 binding in a manner similar to unlabeled SQ 29,548 (Fig. 7 B). This result, oupled with the failure of 8-epi- PGF2, to stimulate seondary release of arahidonate ylooxygenase produts from isolated glomeruli and RASM (Fig. 6), strongly supports the in vivo observations, suggesting that diret interation of 8-epi-PGF2, with TxA2 reeptors is the prinipal mehanism underlying its renal onstritor ations. The formation of these prostanoids atalyzed by free radials (2), oupled with the potent renal vasular ativity demonstrated for 8-epi-PGF2, at pathologially relevant doses in the present study, suggests a potential pathophysiologial role for those ompounds as novel mediators in oxidant-related renal injury. In fat, our data demonstrated that during the reperfusion period of ishemi aute tubular nerosis, an oxygen free radial-mediated experimental model ofrenal injury (3, 4), urinary exretion rate ofthese eiosanoids inreased by more than 3%. In summary, we examined the responses of the glomerular miroirulation in the rat to 8-epi-PGF2a, a novel nonylooxygenase-derived prostanoid produed as a result of free radial atalyzed lipid peroxidation. Our experiments demonstrate that the primary sites of ation of 8-epi-PGF2, are in preglomerular and possibly mesangial smooth musle, where it exerts potent onstritor effets. These ations result in redutions in renal perfusion and probably in glomerular apillary surfae area, thereby depressing GFR. We have also sown that in rat renal irulation, the onstritor effets of 8-epi-PGF2, are aused by its ativation of thromboxane reeptors. These observations, oupled with the inreased prodution of these prostanoids in ertain oxygen free radial-related experimental models, suggest a pathophysiologial role for free radial-generated prostaglandins in mediating the renal funtional derangements observed in oxidant-related renal injury. In addition, our findings might provide an explanation for the salutary effets of TxA2 antagonism demonstrated in models of lipid peroxidation-indued renal pathophysiology (25). Aknowledgments This work was supported by National Institutes of Health grants DK 39261, DK 43883, and GM Jason D. Morrow is a Howard Hughs Medial Institute Physiian Researh Fellow. Referenes 1. Shah, S. V Oxidant mehanisms in glomerulonephritis. Semin. Nephrol. 11: Morrow, J. D., K.. Hill, R. F. Burk, T. M. Nammour, K. F. Badr, and L. J. Roberts, II A series of prostaglandin F2-like ompounds are produed in vivo in humans by a non-ylooxygenase, free radial-atalyzed mehanism. Pro. Natl. Aad. Si. USA. 87: Hansson, R.,. Jonsson, S. Lundstam, S. Pettersson, T. Shersten, and J. Waldenstrom ffets of free radial savengers on renal irulation after ishemia in the rabbit. lin. Si. (Lond.). 65: Paller, M. S., J. R. Hoidal, and T. F. Ferris Oxygen free radials in ishemi aute renal failure in the rat. J. lin. Invest. 74: Morrow, J. D., T. M. Harris, and L. J. Roberts II Nonylooxygenase oxidative formation of a series of novel prostaglandins: analytial ramifiations for measurement of eiosanoids. Anal. Biohem. 184: Takahashi, K., J. apdevila, J. R. Falk, H. R. Jaobson, and K. F. Badr ytohrome P45 arahidonate in the rat kidney: haraterization and hemodynamis responses. Am. J. Physiol. 258 (Renal and letrolyte Physiol. 27):F781-F Maddox, D. A., D.. Prine, and F.. Retor, Jr ffet of surgery on plasma volume and salt and water exretion in rat. Am. J. Physiol. 233 (Renal Fluid letrolyte Physiol. 6):F6-F Fuhr, J., J. Kazmazyk, and. D. Krfittgen ine einfahe olorimetrishe Methode zur Inulinbestimmung fur Nieren-learaneuntersuhungen bei Stoffwehselgesunden und Diabetikern. Klin. Wohenshr. 33: Smith, H. W., N. Finkelstein, and L. Aliminosa The renal learane of substituted hippuri aid derivatives and other aromati aids in dog and man. J. lin. Invest. 24: Ogletree, M. L., D. N. Harris, R. Greenberg, M. F. Haslanger, and M. Nakane Pharmaologial ations of SQ 29,548, a novel seletive thromboxane antagonist. J. Pharmaol. xp. Ther. 234: Stahl, G. L., H. Darius, and A. M. Lefer Antagonism of thromboxane ations in the isolated perfused rat heart. Life Si. 38: Badr, K. F., D. K. DeBoer, K. Takahashi, R.. Harris, A. Fogo, and H. R. Jaobson Glomerular responses to platelet ativating fator in the rat: role of thromboxane A2. Am. J. Physiol. 256 (Renal Fluid letrolyte Physiol. 25) F35-F Deen, W. M., J. L. Troy,. R. Robertson, and B. M. Brenner Dynamis of glomerular ultrafiltration in the rat. IV. Determination ofthe ultrafiltration oeffiient. J. lin. Invest. 52: Viets, J. W., W. M. Deen, J. L. Troy, and B. M. Brenner Determination of serum protein onentration in nanoliter blood samples using fluoresamine or o-phthaldehyde. Anal. Biohem. 88: Takahashi, K., G. F. Shreiner, K. Yamashita, B. W. hristman, I. Blair, and K. F. Badr Predominant funtional roles for thromboxane A2 and prostaglandin 2 during late nephrotoxi serum glomerulonephritis in the rat. J. lin. Invest. 85: Ross, R The smooth mustle ell. II. Growth of smooth mustle in ulture and formation of elasti fiber. J. ell Biol. 5: Hanasaki, K., K. Nakano, H. Kasai, H. Arita, K. Ohtani, and M. Doteuhi Speifi reeptors for thromboxane A2 in ultured vasular smooth mustle ells of rat aorta. Biohem. Biophys. Res. ommun. 15: Hanasaki, K., K. Nakano, H. Kasai, and H. Arita Biohemial haraterization and omparison of rat thromboxane AJprostaglandin H2 reeptors in platelets and ultured aorti smooth mustle ells. Biohem. Pharmaol. 38: Badr, K. F., B. M. Brenner, and I. Ihikawa ffets of leukotriene D4 on glomerular dynamis in the rat. Am. J. Physiol. 253 (Renal Fluid letrolyte Physiol. 22):F239-F Harris, R.., K. A. Munger, K. F. Badr, and K. Takahashi Mediation of renal vasular effets ofepidermal growth fator by arahidonate metabolites. FASB (Fed. Am. So. xp. Biol.) J. 4: Stier,. T., Jr., L. Jakson Roberts, II, and P. Y-K. Wong Renal response to 9a,,-prostagalandin F2 in the rat. J. Pharmaol. xp. Ther. 2 17: Mene, P., and M. J. Dunn iosanoids and ontrol of mesangial ell ontration. ir. Res. 62: Badr, K. F., V.. Kelley, H. G. Renneke, and B. M. Brenner Roles for thromboxane A2 and leukotrienes in endotoxin-indued aute renal failure. Kidney Int. 3: Shlondorff, D The glomerular mesangial ell: an expanding role for a speialized periyte. FASB (Fed. Am. So. xp. Biol.) J. 1: Kaplan, R., H. S. Aynedjian, D. Shlondorff, and N. Bank Renal vasoonstrition aused by short-term holesterol feeding is orreted by thromboxane antagonist or probuol. J. lin. Invest. 86: Glomerular Ations of8-epi-pgf2a, 141