Kidney International, Vol. 19 (1981), pp. 755 759 Relevance of urinary &-keto-prostaglandin Fia determination BERND ROSENKRANZ, WAIHI KrrMIMA, and JURGEN. FROLIH Fischer-Bosch Institute of linical Pharmacology, Stuttgart, Germany Problems of assessment of endogenous prostaglandin production Endogenous prostaglandin production is often determined by measuring the plasma or urine concentrations of these prostanoids or their main metabolites [1]. The metabolism of most prostaglandins has been elucidated to ascertain which parameter is most suitable for determination of total body synthesis of the parent compound (for review, see Ref. 2). These studies demonstrated that metabolism of the "classical" prostaglandins E2 and F2a within the lung occurs very rapidly so that they are almost completely degraded during one passage through the circulation [31. It has been shown that 9 seconds after the i.v. administration of labeled PGE2 only about 3% of the injected dose was present in blood as PGE2, whereas the half-life of 1 l-hydroxy-9, 15-diketo-prost-5-enoic acid, one of its circulating metabolites, was about 8 mm [4]. Similar results were obtained for PGF2a [5]. This led to the conclusion that urinary excretion of PGE2 or PGF2a probably is not an indicator of total body synthesis. This was confirmed by the observation that after the i.v. administration of PGE, into man no increase in urinary PGE2 excretion could be measured [4]. The main urinary metabolite of PGE1 and PGE, in man is 7a-hydroxy-5,ll-diketo-tetranor-prosta- 1, l6-dioic acid [41 and that of PGF2 is 5,7-dihydroxy- 11 -keto-tetranor-prosta- 1,1 6-dioic acid [6]. These metabolites now are regarded as indicators of total body synthesis of PGE1 and PGE2 or PGF2a, respectively. A change in urinary excretion of one of these metabolites, however, does not necessarily prove a change in total body synthesis of the parent compound because the same effect also could be caused by a shift of its metabolic pattern. It would appear preferable, therefore, to determine several urinary metabolites of a prostanoid for assessment of its total body synthesis. The metabolic studies showed that excretion of PGE2 and PGF2a into urine does not result from circulating levels of these prostaglandins. In addition, stop-flow experiments carried out by our group have shown that PGE2 and PGF2 are secreted into the loop of Henle [7]. Therefore, it was concluded that urinary excretion of both prostanoids predominantly reflects their intrarenal synthesis [7, 8]. Numerous studies have since used the measurement of urinary PGE2 and PGF2a to assess the rate of renal synthesis of these prostanoids. Apart from significant methodologic problems showing discrepancies between values measured by the reference method (gas chromatography mass spectrometry) and radioimmunoassay [1], it can be expected that many factors including urinary ph and flow rate can affect this parameter. In contrast to the "classical prostaglandins," until recently no information existed about metabolism of prostacyclin (PGI2) in man. Studies performed in several animals have shown that this prostanoid is not metabolized during passage through the lung [9, 1]. Because of its chemical instability, prostacyclin is rapidly hydrolyzed to 6- keto-pgf1a under physiologic conditions [11]. Therefore, it has been assumed that determination of 6-keto-PGF1a in plasma or urine should reflect endogenous prostacyclin production. In addition to nonenzymatic hydrolysis to 6-keto-PGF1a, termination of the biological activity of prostacyclin might Received for publication December 11, 198 85-2538/81/19-755 $1. 1981 by the International Society of Nephrology 755
756 Rosenkranz et a! occur by metabolism within the kidney [12] (Rosenkranz et al, submitted for publication), the liver [13], or the vessel wall [14], resulting in metabolites other than 6-keto-PGF1a. This would lead to the conclusion that 6-keto-PGF1, like the "classical" prostaglandins, is not a useful parameter of endogenous prostacyclin production. The following discussion centers around the problem of assessment of renal and total body prostacyclin synthesis. Metabolism of prostacyclin and 6-keto-PGFj, in man To define the relevance of urinary 6-keto-PGF1a excretion, investigators have considered it necessary, like for PGE2 and PGF2a, to study the metabolism of prostacyclin and 6-keto-PGF1a. Therefore, our group investigated the urinary metabolites of prostacyclin and 6-keto-PGF1a in man [15, 16]. Tritiated and tetradeuterated prostacyclin, as well as unlabeled prostacyclin, was infused i.v. into healthy male volunteers at the rates of I and 5 ng/kg/min, respectively. In separate experiments, tritiated or unlabeled 6-keto-PGF1a was infused at the rates of.4 and 42 ng/kg/min. When compared with the infusion rates necessary for saturation of the metabolizing enzymes of PGE2 or PGF2a, the infusion rates of I and.4 ng/kg/min chosen for the labeled compounds were below this level [15]. Our study demonstrated that after the administration of labeled prostacyclin, labeled 6- keto-pgf1a could be detected in urine after extraction and chromatography by use of gas chromatography mass spectrometry (GMS) (Fig. 1). In addition to the fragments m/z 614 (M-l5), 598 (M-31), 558 (M-7l), 58 (M-31 9), 468 (M-71 9), 418 (M-31-2 x 9), and 378 (M-71-2 x 9), the deuterated ions mlz 618, 62, 512, and 422 were present, showing that this material resulted from the infused deuterated prostacyclin and did not reflect endogenous renal prostacyclin synthesis. The fragments obtained after cleavage of the lower chain (M -71) did not show a corresponding deuterated fragment because the molecule was labeled in the lower chain. The material from which this mass spectrum was obtained represented 14.2% of the radioactivity recovered from high-pressure liquid chromotography (HPL). Thus, circulating prostacyclin is partly excreted into urine as 6-keto-PGF1a. This study further revealed that the major urinary metabolite of prostacyclin in man is not 6-keto- PGF1a, but rather its f3- and o-oxidized analogues, dinor-6-keto-pgf1 and dinor-6, 1 5-diketo- 13, 14-dihydro-2-carboxyl-PGF1, corresponding to 2.5 and 19.7% of the radioactivity, respectively. In 1 8 6 4 2 1 8 6 4 2 (1 Mt 71 2 X 9 378 Mt 31 2 X 9 418 422 I.... 34 36 38 4 42 44 46 48 M 31 9 512 Mt 71 58 M 71 9 468 M 31 598 62 Mt 15 hjii 614 618 5 52 54 56 58 6 62 64 Fig. 1. Mass spectrum of deuterated6-keto-pgf1, obtainedfrom human urine after the Lv. infusion of tetradeuterated and tritiated PGI2. The urine was chromatographed, and the resulting peaks were derivatized to the methoxime methyl ester trimethyl silyl ether and were analyzed by gas chromatography mass spectrometry using a glass capillary column. addition, dinor-6, I 5-diketo- 13, 14-dihydro-PGFj and 6, 15-diketo-13, 14-dihydro-PGF1 (6.8 and 1.4%, respectively) could be detected in urine [15]. The same metabolites also could be identified from urine after infusion of 6-keto-PGF1a with the relative amounts of 6.8% (6-keto-PGF1,), 22.4% (dinor- 6-keto-PGF1a), 7.% (dinor-6, I 5-diketo- 13,1 4-dihydro-2-carboxyl-PGF1a), 5.4% (dinor-6, 1 5-diketo- 13, l4-dihydro-pgf1), and 5.7% (6, 15-diketo-l3, 14- dihydro-pgfj) [15]. The similarity of both patterns of metabolites suggests that circulating prostacyclin first is hydrolyzed to 6-keto-PGF1 and partly metabolized further by 13- and to-oxidation, 15-hydroxydehydrogenation, and reduction of the double bond at l3. These enzymatic mechanisms have also been described for several other prostanoids [121, including prostacyclin in the rat [17, 18] These metabolic studies demonstrate that urinary 6-keto- PGF1a can result from circulating prostacyclin or 6- keto-pgf1a. Other studies have shown, however, that the renal cortex [19], the glomeruli [2, 21], and the collecting duct [22] are able to convert arachidonic acid to prostacyclin. Therefore, it is still not clear whether 6-keto-PGF1a detected in urine under physiologic conditions really represents total body prostacyclin synthesis or whether it represents renal prostacyclin production. Specificity of urinary prostacyclin metabolites for intrarenal synthesis Studies in the perfused rabbit kidney [12] have shown that dinor-6-keto-pgf and dinor-6, 15-diketo- 13, l4-dihydro-pgf1, could be detected in the effluent collected from the renal vein after addition
Urinary 6-keto-prostaglandin Fia 757 of labeled prostacyclin to the perfusion medium. But not enough radioactivity could be collected from the ureteral effluent to determine the metabolic profile. Therefore, we performed experiments (Rosenkranz et a!, submitted for publication) to determine whether a metabolite could be identified that is specific for the urinary compartment and originates from within the kidney. Isolated rabbit kidneys were perfused with a solution to which prostacyclin was added continuously at the rate of.23 pgimin for 1 hour. Subsequently, the kidney was perfused for another 15 mm with the perfusate only. By this technique, 2.3% of the total radioactivity could be recovered from the ureteral effluent. The renal venous and the ureteral effluents were extracted and chromatographed. The materials belonging to the peaks obtained by Porasil HPL were derivatized to the methoxime methyl ester trimethylsilyl ether and analyzed by GMS. Of the radioactivity recovered from the ureteral and venous effluent, 9.6% and 9.1%, respectively, resulted in the fragmentation pattern mlz 586, 57, 53, 48, 44, 421, 39, and 35, corresponding to M -15, M -31, M ;f -71, M -9 31, M-9 71, M-2x9, M-2x9-31, and M- 2 x 9 71 of dinor-6-keto-pgf1a [16]. In addition, 39.2% and 58.2% of the ureteral and venous effluent, respectively, could be identified as 6-keto- PGF1a by use of GMS. Further smaller peaks were obtained on HPL, which all contained less than 1% of the radioactivity each. Thus, in this experiment, no metabolite of prostacyclin specific for intrarenal synthesis of prostacydin could be identified from the ureteral effluent. In addition, these results show that urinary dinor-6- keto-pgf1a, one of the major urinary metabolites of prostacyclin after its Lv. administration in man, can partly result from renal metabolism of prostacyclin. Origin of urinary 6-keto-PGF,,. Our studies about metabolism of prostacyclin and 6-keto-PGF, in man have demonstrated that circulating 6-keto-PGF1 can be excreted into urine. But, the question of the origin of 6-keto-PGF1a detected in urine under physiologic conditions still remains unanswered. Besides glomerular filtration, the most likely mechanism for the excretion of 6-keto-PGF1a would be an active secretion by the organic acid secretory pathway because it has been shown that this route plays an important role in the excretion of PGE2, reaching the kidney via the renal artery [23]. We performed experiments in the dog to determine the importance of the acid secretory pathway for 6-2, 15, 1, 5, 15, 1, 5, 2 4 6 8 1 12 14 Fraction number Fig. 2. Urinary excretion of a bolus of tritiated PGE2 together with inulin administered into one renal artery of the anesthetized dog. Urine was collected in 2-second fractions and analyzed for radioactivity and inulin. Top: control; bottom: after pretreatment with probenecid. keto-pgfjq (Kitajima et al, submitted for publication). A bolus of inulin (1 mg) and tritiated PGE2 or 6-keto-PGF1a (.5 pi each) was administered into one renal artery of the anesthetized dog before and after pretreatment with probenecid (25 mg/kg), a blocker of the organic acid secretory pathway. Urine was collected in 2-second periods and analyzed for radioactivity and inulin. When PGE2 was administered, excretion of radioactivity was delayed with respect to the inulin peak (Fig. 2). Inulin served as a marker for glomerular filtration. Thus, some mechanism that is more time consuming than mere filtration is involved in the excretion of PGE2. As pretreatment with probenecid abolished this delayed portion of the radioactive peak, it was concluded that PGE2 is partly excreted via the organic acid secretory pathway as previously described [23]. The same experiment was carried out for 6-keto- PGF1a in the anesthetized dog. The radioactive material appeared in the same fractions as the inulin peak (Fig. 3). Probenecid affected neither excretion of radioactivity significantly nor excretion of endogenous 6-keto-PGF1a as determined by GMS. These studies show that 6-keto-PGF1a, in contrast to PGE2, is not excreted via the organic acid secretory pathway. (N x
758 Rosenkranz et at E I 2 4 6 8 1 12 14 16 18 Fraction number Fig. 3. Urinary excretion of tritiated 6-keto-PGF1,, and inulin. For further description, see Fig. 2. onclusion These studies lead to the conclusion that urinary 6-keto-PGF1a can originate from circulating prostacyclin or 6-keto-PGF1a. Renal excretion of 6-keto- PGF1c. obviously occurs mainly by glomerular filtration, whereas the organic acid secretory pathway does not play an important role. But no information exists so far about the excretion of 6-keto-PGF1a at another site of the tubule. Therefore, further studies are required to define whether 6-keto-PGF1a detected in urine mainly results from circulating prostacydin as suggested by our results or from the kidney. Finally, the data presented strongly suggest that the determination of dinor-6-keto-pgf1a or dinor- 6, 15-diketo- 13, 14-dihydro-2-carboxyl-PGF1, the major metabolites of prostacyclin in human urine, represents better parameters for total body prostacyclin production in man than does 6-keto-PGF1. The determination of these metabolites together with 6-keto-PGF1a can give more precise information than does the determination of a single parameter because these three compounds reflect about 5% of the metabolites of prostacyclin in human urine [14]. x 9 6 = Acknowledgments This work was supported by the Robert Bosch Foundation, Stuttgart. Drs. K. E. Weimer, G. Beck, K. Hofbauer, and. Fischer, respectively, synthesized the prostacyclin, prepared the perfused kidneys, and performed the GMS analysis. Mrs. B. Grozinger prepared the manuscript. Reprint requests to Dr. B. Rosenkranz, Fischer-Bosch Institute of linical Pharmacology, Auerbachstr. 112, D-7 Stuttgart 5, Federal Republic of Germany References 1. Methods in Prostaglandin Research, edited by FROLIH J, New York, Raven Press, 1978 2. SAMUELSSON B, GRANSTROM E, GREEN K, HAMBERG M, HAMMARSTROM S: Prostaglandins. Ann Rev Biochem 44:669-695, 1975 3. PIPER PJ, VANE JR, WYLLIE JH: Inactivation of prostaglandins by the lungs. Nature 225:6 64, 197 4. HAMBERG M, SAMUELSSON B: On the metabolism of prostaglandins E1 and E2 in man. JBiol hem 246:6713 6721, 1971 5. GRANSTROM E: On the metabolism of prostaglandin F2, in female subjects: Structures of two metabolites in blood. Eur J Biochem 27:462 469, 1972 6. GRANSTROM E, SAMUELSSON B: On the metabolism of prostaglandin F2, in female subjects. J Biol hem 246:5254 5263, 1971 7, WILLIAMS WM, FROLJH J, NIES AS, OATES JA: Urinary prostaglandins: Site of entry into renal tubular fluid. Kidney mt 11:256 26, 1977 8. FROLIH J, WILSON TW, SWEETMAN BJ, SMIGEL M, NIES AS, ARR K, WATSON JT, OATES JA: Urinary prostaglandins: Identification and origin. J /in Invest 55:763 77, 1975 9. DUSTING J, MONADA 5, VANE JR: Recirculation of prostacyclin (PGI2) in the dog. Br J Pharmacol 64:315 32, 1978 1. WALDMAN HM, ALTER I, KOT PA, ROSE J, RAMWELL PW: Effect of lung transit on systemic depressor responses to arachidonic acid and prostacyclin in dogs. J Pharmacol Exp Ther 24:289 293, 1978 11. HO MJ, ALLEN MA: hemical stability of prostacyclin (PGI2) in aqueous solutions. Prostaglandins 15:943 954, 1978 12. WONG PY-K, MGIFF J, AGEN L, MALIK KU, SUN FF: Metabolism of prostacyclin in the rabbit kidney. J Biol hem 254:12 14, 1979 13. WONG PY-K, MALIK KU, DESIDERIUS DM, MGIFF J, SUN FF: Hepatic metabolism of prostacyclin (PGI2) in the rabbit: formation of a potent novel inhibitor of platelet aggregation. Biochem Biophys Res ommun 93:486 494, 198 14. WONG PY-K, SUN FF, MGIFF J: Metabolism of prostacydin in blood vessels. J Biol hem 253:5555 5557, 1978 15. ROSENKRANZ B, FISHER, WEIMER KE, FROLIH J Metabolism of prostacyclin and 6-keto-prostaglandin F1,, in man. J Biol hem 255:1194 1198, 198 16. ROSENKRANZ B, FISHER, REIMANN I, WEIMER KE, BEK G, FROLIH J: Identification of the major metabolite of prostacyclin and 6-keto-prostaglandin F1,, in man. Biochim Biophys Acta 619:27 213, 198 17. SUN FF, TAYLOR BM, SUTTER DM, WEEKS JR: Metabolism of prostacyclin: III. Urinary metabolite profile of 6-keto- PGFI,, in rat. Prostaglandins 17:753 759, 1979
Urinary 6-keto-prostaglandin F1, 759 18. SUN FF, TAYLOR BM: Metabolism of prostacyclin in rat. Biochemistry 17:496 411, 1978 19. WHORTON AR, SMIGEL M, OATES JA, FROLIH J: Regional differences in prostaglandin formation by the kidney: Prostacyclin is a major prostaglandin of renal cortex. Biochim Biophys Ada 529:176 18, 1978 2. FOLKERT VW, SHLONDORFF D: Prostaglandin synthesis in isolated glomeruli. Prosraglandins 17:79 86, 1979 21. HA5sID A, KONIEZKOWSKI M, DUNN MJ: Prostaglandin synthesis in isolated rat kidney glomeruli. Proc Nan Acad Sd USA 76:1155 1159, 1979 22. GRENIER F, SMITH W: Formation of 6-keto-PGF1, by collecting tubule cells isolated from rabbit renal papillae. Prostaglandins 16:759 772, 1978 23. ROSENBLATT 5G. PATAK RV, LIFSHITZ MD: Organic acid secretory pathway and urinary excretion of prostaglandin E in the dog. Am J Fhysiol 235:F473 F479, 1978