Acta Physiol Scand 1979, 106: 307-312

Human forearm and kidney conversion of arachidonic acid to prostaglandins J. NOWAK and A. WENNMALM Department of Clinical Physiology, Karolinska Institute, Huddinge Hospital, Sweden

NOWAK, J. & WENNMALM, A.: Human forearm and kidney conversion of arachidonic acid to prostaglandins. Acta Physiol Scand 1979, 106: 307-312. Received I5 Jan. 1979. ISSN 0001-6772. Department of Clinical Physiology, Karolinska Institute, Huddinge Hospital, Sweden. The capacity of the human forearm and kidney to synthetize different prostaglandins (PGs) was studied, together with the quantitative relationship between the various PGs formed in these organs. 14C-labelledarachidonic acid (“C-AA) was infused in healthy male volunteers at a constant rate into the brachial or the renal artery, with simultaneous sampling of regional venous blood. The venous plasma content of 14C-PGs was extracted, separated with thin-layer chromatography (TLC) and quantified using fractionated liquid scintillation spectrometry. Most of the “C-AA infused was metabolized and radiopeaks parallel to unlabelled standards of E D 2 , WE2, PGF2,, bketo-PGF,, and 13,14-dihydro-15-ketoW E 2 (Me) were obtained. The chromatograms of both the forearm and the kidney plasma contained all the peaks described, but the relative distribution of the “C-PGs differed between the two tissues. In the cubital venous plasma, the main PG (apart from Me) was bketo-PGF,,, indicating a considerable synthesis of PGIz in the forearm. In the renal venous plasma, on the other hand, PGDz accounted for the largest part of the authentic “C-PGs found. Besides the tissue differences, large inter-individual variations were observed. The results demonstrate the existence of both a considerable tissue specificity and an appreciable inter-individual variation in the local conversion of AA to PGs in man. Key words: Arachidonic acid, kidney, prostaglandin, prostacyclin, renal vasculature,

skeletal muscle, vascular tissue

Endogenous prostaglandins (PG)have been proposed as mediators of several physiological and pathophysiological processes in man (Karim 1976) and their appearance in the peripheral venous blood has been studied intensively in recent years. Due to the rapid release of PG from blood cells and from injured tissue, the analysis of venous plasma PGconcentrations may yield falsely high values and may therefore (unless extraordinary precautions are taken during the sampling process) be a poor method for estimating the endogenous PG-synthesis. A better parameter than the parent prostaglandins in monitoring the endogenous formation of PGs is the stable 13,14-dihydro-15-ketoderivative of the respective compound (Samuelsson & GrCen 1974). However, since these derivatives are probably formed in the blood from the respective parent PGs, their plasma concentrations give no information about the specific origin of the PGs formed. Studies on

the local arachidonic acid (AA) metabolism in various parts of the circulatory system are therefore of interest, since the results may reveal a specific tissue differentiation of the endogenous PG synthesis, which in turn may contribute to an understanding of a possibly physiological role of the endogenous PGs. In the present investigation we have therefore studied the regional conversion of arachidonic acid to PGs in the human forearm and kidney.

MATERIAL AND METHODS The study was performed with the permission of the Ethical Committee at Karolinska Institute. 19 healthy volunteers, all men aged 19-43 years, were studied. They were fully informed about the aim, experimental procedure and possible risks of the investigation before giving their voluntary consent to participate. All experiments were performed in the morning in supine body position after an overnight fast. Actu Physiol Srand 106

308

J . Nowuk and A. Wennmulm

a ) Catheterization and isotope infusion. In 10 subjects, The methanol was evaporated and the residue applied to aged 22-33 years, a teflon catheter (outer diameter 1.2 TLC plates (0.25 mm DC Fertigplatten Kieselgel F254. mm) was inserted into a brachial artery. Another similar Merck). Occasionally purified lipid extracts of forearm or catheter was inserted 5 0 6 0 mm in the retrograde direction renal venous plasma from more than one patient were from the ipsilateral antecubital vein so that the tip of the pooled before being subjected to TLC. The plates were catheter was positioned deep in the forearm muscle. This run 18 cm against unlabelled standards of bketo-PGF,,. catheter position yields blood that is representative of the PGFa, PGE,, PGD2, PGB,, PGA,, 13.14-dihydro-15forearm venous effluent (Wahren 1%6). After catheter keto-PGE, (Me) and Na-arachidonate. The following insertion, dynamic forearm work (handgrip) was per- solvents were used: I ) ethylacetate :acetic acid :2.2.4formed for I5 rnin on a hand ergometer (Piab. Sweden). trimethylpentane : water (90 :20 :50 : 100, vlv, organic The work load was 1 W and the contraction frequency 1 phase, AIX, Hamberg and Samuelsson 1966); 2) benHz. During the last 10 min of the forearm work 5 pCi zene :dioxane :acetic acid (20 :20 : I , v/v, A I, Ham"C-arachidonic acid (I4C-AA)(New England Nuclear, sp. berg and Samuelsson 1%6); 3) benzene : dioxane : act. 4C-60 mCilmmol), freshly prepared as the sodium salt acetic acid (10: 10: I , v/v, BDA, Isakson et al. 1977); in saline, was infused intra-arterially at a constant rate of chloroform : methanol : acetic acid : water (90 : 8 : 1 : 0.8, 0.5 pCi/min. Simultaneously forearm venous blood was vlv, C, Isakson et al. 1977). After development of the plates, the sample lanes were sampled into 10 ml syringes at a constant sampling rate of divided into 5 mm horizontal zones, which were scraped 15-20 ml/min. In 9 subjects, aged 1943 years, two teflon catheters off and eluted with ethanol. After addition of Instagel, the (outer diameter 1.66 mm) were inserted into a femoral activity of the ethanol eluates was counted in a liquid artery and a femoral vein, respectively. The arterial cathe- scintillation spectrometer (Intertechnique SL 4000). The ter was advanced under fluoroscopic control to one of the unlabelled PG standards were developed using renal arteries and the venous catheter was passed to the phosphomolybdic acid. ipsilateral renal vein. After the position of the arterial c) Calculations, The total "C-activity in the thin-layer catheter had been verified by the injection of a small chromatograms was estimated as the total area of all the amount of radiopaque medium (Isopaque Cerebral), the fractions after subtraction of background activity. The subjects rested comfortably and 10 pCi of "C-arachido- area of the various, clearly defined radiopeaks that cornate was infused during 10 rnin into the renal artery at a responded to authentic PGs was calculated and expressed constant rate of 1.0 pCilmin. During the entire infusion, in percent of the total area of ''C-PGs in the respective renal venous blood was continuously sampled as de- chromatograms. The entire "C-PG activity in the scribed above, to a final amount of 200 ml. chromatograms was also calculated in relation to the total All catheters were inserted percutaneously under local I4C-activity. All values are given as mean+S.E. unless anesthesia. The patency of the catheters, when not used otherwise stated. for infusion of isotope or sampling of blood, was maintained by intermittent flushing with saline. 6) Extraction and separation of plasma W'-PGs. All RESULTS subjects were pretreated with 3 g aspirin 3 days before the Extruction und purifcution investigation to avoid release of prostaglandins from the platelets during sampling of blood. Such treatment inhibits The efficiency of the lipid extraction procedure was PG-synthesis by platelets for 3 4 days after administration checked by determining recovery of :'H-labelled of the drug. (Kocsis et al. 1973) without affecting the PGFzo, WEz, PGAz, and PGBz added to external general PG-synthesis in the tissues (Hamberg 1972). The venous blood was collected in chilled vacutainers standards. The recovery ranged between 54t45% containing ethylene-diamine-tetraaceticacid (EDTA). It ( n = 3 ) for3H-PGAz and 7 3 f 2 % ( n = 3 ) for:'H-PGEz was kept on ice until the plasma was separated (within 30 and 3H-PGFw. The recovery of external PG rnin after sampling) by I5 rnin centrifugation at 2000Xg at standards in the methanol eluate of the silicic acid 4°C. After separation the plasma was pooled and im- columns was almost quantitative. mediately subjected to further analyses. After dilution of the plasma with 2 volumes of water (to avoid foaming during the lipid extraction), neutral lipids Thin layer chromutogruphic sepurution were removed by shaking with an equal volume of petrol Thin layer chromatographic separation of PGs was (40"-60"). Extraction of prostaglandins was performed studied in several solvent systems (Table I ) . In twice at pH 3.5 with equal volumes of ethylacetate (Fisher, E-145). After evaporation of the organic phase, some of the systems used, two or more compounds the residue was reconstituted in 5 ml of toluene: had almost identical R, values. Consequently, ethylacetate (9: I , vlv). Purification of the samples was calculation of the respective I4C-PG areas was not performed using silicic acid chromatography. Silicic possible in these cases and only the total I4C-PG-acacid (Mallincrodt, 100 mesh) was activated at 150°C. tivity of the chromatograms could be estimated. Microcolumns (5x45 mm) were prepared with toluene :ethylacetate (9 : I). After application of the sample, Complete separation of PGs of the D, E. F, and Fz the columns were washed with 15 ml of toluene : ethylace- series was only obtained with solvent system AIX. tate (9: I ) . Elution was performed with 20 ml of methanol. However, even in this system PGAz, PGBz and 13, Acfo Pliysiol Srond 106

309

Aruchidonic conversion to PG in forearm and kidney

Table 1. Thin layer chromatographic separation of PCs ~~

R, values ~~

Compound

System C

System BDA System A I

6-keto-PGF,, PGF, PGEz PGDl PGA,; PGB,; 13, 14, -2H, 15-keto-PGEz Na-AA

0.34-0.44 0.27 0.34-0.44 0.34-0.44 0.52 0.63

0.21 0.35 0.47 0.55-0.72 0.83

0.37-0.46 0.37-0.46 0.56

0.66-0.80 0.66-0.80 0.89

System A IX 0.13 0.17 0.25 0.36 0.43-0.57 0.78

ing their R, values in different solvent systems with the authentic compounds. The changes in chromatographic mobility of the radioactive products were similar to those of authentic markers. Forearm and kidney conversion of Y - A A The relative amount of the activity which apInfusion of I4C-AA through the renal and forearm peared during the I4C-AA infusion, identified as vasculature resulted in the appearance of 14C-la- I4C-PG, was almost the same in renal (41*5%, belled AA metabolites in the venous blood from n = 7 ) as in forearm venous plasma (38+3%, n = 8 ) . both regions. Thin layer chromatography of the The activity chromatographed usually in 5 peaks, purified lipid extracts of the venous plasma re- corresponding to PGA,IPGB,I 13,lCdihydro-ISvealed that the 14C activity chromatographed in keto-PGE, (Me), PGD2, PGE2, PGFm and 6-ketoparallel to authentic prostaglandin markers. Further PGF,,. However, the relative distribution of I4Cevidence that the radioactive products in these PGs in venous plasma differed between the two zones really were 14C-PGswas obtained by compar- vascular regions investigated. In the forearm venICdihydro- 15-keto-PGE2 run almost in parallel. The zone corresponding to these compounds was therefore considered as the metabolite zone (Me).

Relative distribution of “GPG inforearmvenous blood

Origin

Fig. 1. “C-PG in forearm venous plasma during infusion of “C-AA into the brachial artery in healthy males. The bars indicate mean+S.E. of the area of radiopeaks obtained in 4 expts. (TLC, system AIX) and are expressed as a percentage of the total area of the “C-PG peaks in the respective expts.

10 20 Fraction niunber

30

I

Front

Fig. 2 . Typical radiochromatogram (TLC, system AIX) of the lipid extract from forearm venous plasma collected during infusion of 14C-AA through the forearm vasculature in a healthy volunteer. Note the prominent Bketo-PGF,, peak, indicating that 14C-PGIzwas the main PG formed. Aclu Pliysiol Scatrd 106

310

J . Nowak and A . Wennmalm Relative distribution of I4C-PGin renal venous blood

7~

‘4c-AA metabolites in renal m

s blood

6

T

lo oriiin

20

Fraction number

30 Front

renal venous plasma during infusion of “C-AA into a renal artery in healthy males. The bars indicate mean 2 S.E.of the area of radiopeaks obtained in 5 expts. (TLC, system AIX) ane are expressed as a percentage of the total area of the L4C-PGpeaks in the respective expts.

Fig. 4. Typical radiochromatogram (TLC, system AIX) of the lipid extract from renal venous plasma collected

ous plasma (Fig. I ) the major radiopeak (41+8%) was usually found in parallel to Me. Large peaks were also found in parallel to 6-keto-PGF,, (25f10%) and PGD, (18+2%). Two small radiopeaks displayed the same chromatographic behaviour as the PGF, (9f1%) and PGE, (8f2%) standards. The ratio of PGD2/PGE2/PGF%/ 6 keto-PGF,, was 2.39 : 1 .O : I . 17 : 3.36. The most constant finding was the peak of radioactivity in parallel to PGD, (coefficient of variation 23 %), whereas other I4C-PGs showed larger quantitative inter-individual variations. The most pronounced interindividual differences occurred in the radioactive zone corresponding to 6keto-PGF1, (coefficient of variation 79%). Thus, in one of the chromatograms (Fig. 2) 6-keto-PGF1, was found to constitute 50% of the I4C-PG activity, compared with only 9% in the chromatogram from another experiment. Also in the renal venous plasma (Fig. 3) the Mepeak usually displayed the largest amplitude (40+5%). Major peaks were also found in parallel to PGD, (20+5%) and 6-keto-PGF,, (17f3%). PGF* and PGE, appeared to be released in somewhat smaller and almost equal amounts (14+3% and 13f2% respectively). The ratio of PGD2/

PGEz/PGF&-keto-PGF,, was 1 S4: 1 .O: 1.07: 1.25. Large inter-individual variations occurred in the relative distribution of I4C-PGs.Thus, PGD2, which apart from Me was the main PG formed constituted 32% of the I4C-PG activity in one chromotogram (Fig. 4) compared with only 8% in another (coefficient of variation 50%).

Fig. 3 . ‘ C P G in

ACIU Physiol Srctnd

IW

during infusion of “C-AA through the renal vasculature in a healthy volunteer. Note the major peak running parallel

to PGD,.

DISCUSSION The aim of the present investigation was to find out whether regional differences in the conversion of I4C-AA to I4C-PGs occur in the vascular system. Such differences would imply that locally formed PGs may play different roles in different parts of the vascular bed. As suitable regions we chose the kidney and the forearm vasculare, not only because these regions are relatively simple to study but also since they represent vascular beds that receive a considerable part of the cardiac output. In the current experiments ‘‘C-labelled AA was infused intraarterially. The concentration of AA infused was low and the infusion rate did not exceed 8 and 17 nmol/min in the brachial and renal artery respectively. Since the arterial plasma concentration of AA in healthy young men is about 10 pmol/l

Arachidonic conversion to PG in forearm and kidney

3 I1

and the plasma turnover rate of AA is about 1.6 to generate PGs of the D, E, F. and I series from pmol/min (Hagenfeldt & Wennmalm 19751, the cur- exogenously administered arachidonate. These rerent I4C-AA infusion should not have altered the sults are mainly in accordance with data from prebasal AA turnover in the tissues studied. This is vious studies. However, the appearance of 6-ketoimportant, since an elevation of the plasma con- PGF,, , indicating formation of PGI,, is remarkable, centration of PG precursor might well induce such since all subjects were pre-treated with aspirin at a changes in the tissue conversion of I4C-AA that the dose that should be sufficient to block the converresulting 14C-PG production does not reflect the sion of AA to PG endoperoxides in the platelets. It has been suggested that vascular endothelium is normal basal synthesis in the organ. Apart from this, the validity of the results as unable to carry out conversion of AA to PG enstandards of regional PG formation rests on the doperoxides and therefore is obliged to "borrow" assumption that recirculating 14C-PGof the D,E, F, preformed endoperoxides from platelets adhering to and I-series-formed from I4C-AA in other organs the vascular wall (Bunting et a]. 1976). The current or tissuesaid not influence the present data, however, indicate that formation of PGI, in chromatograms to any considerable degree. It has the vascular bed is not completely dependent upon been established that PGs are rapidly metabolized an intact function of the cyclo-oxygenase in the in the circulatory system by dehydration of the 15- platelets. This is further supported by a recent hydroxy group and reduction of the AI3 double demonstration that human endothelial cells synthbond (Hamberg & Samuelsson 1971, Granstrom etize PGI, directly from exogenous arachidonate 1972). This degradation is probably common to (Weksler et al. 1977). It has been reported that inhibition of PG synthPGD,, PGE2, PGF2,, and PGIz and results in formation of the respective 13,14-dihydro-15-keto-deriva- esis by indomethacin decreases functional hypertives. Since these derivatives differ in emia in the human forearm (Kilbom & Wennmalm chromatographic mobility from their parent com- 1976). Recently, increased concentrations of pounds, it does not appear likely that tissues beside radioimmunoactive PGE material were demonthose specifically studied interfered-via recirculat- strated in the venous blood from working skeletal ing 14C-PGs-with the present chromatograms. On muscle in man (Nowak & Wennmalm 1978b). On the other hand, the 13,14-dihydro-15-keto-deriva- the basis of these and other observations, endogentives can be expected to occur in considerably high- ously formed and released vasodilating PGs have er recirculating concentrations, due to the rapid been proposed to modulate local vascular remetabolism of the parent compounds (Samuelsson sistance (Hedqvist 1972, Staszewska-Barczak & & Grken 1974). It is therefore conceivable that the Vane 1975, Nowak & Wennmalm 1978a). The cur14C-Me zone in present chromatograms reflects rent data are compatible with such a role for enalso, if not mainly, the metabolites of I4C-PGgener- dogenous PGs and focus interest on PGI, as the quantitatively most important PG in the forearm ated in other organs or tissues. When estimating the relative amounts of IT-PGs vascular bed. In the kidney the major prostaglandins have been formed from the infused I4C-AA, the efficiency of the lipid extraction and purification procedures reported to be PGE,, PGF2,, and recently also must be considered. The recovery of the respective PGD, (Lee et al. 1967, Hamberg 1969, Blackwell et PGs differed in this respect, which may raise dif- al. 1975, Friesinger et al. 1978), all of which are ficulties in defining precisely the relative amounts represented in the current chromatograms. Since of the 14C-PGsformed. However, these differences PGI, has been shown to be synthetized by human were not large compared to the inter-individual renal cortical microsomes in vitro (Whorton et al. variations and probably did not influence the 19771, it seems likely that this PG is also synthetized characteristic chromatographic PG-profiles to any in intact kidney. The occurrence of a radiopeak significant extent. Furthermore, this objection is corresponding to 6-keto-PGFI, in the present not relevant when comparing the relative distribu- chromatograms may provide further evidence for tion of PG biosynthetic pathways in the vascular PGIz formation in the renal tissue, although it is not yet known to what extent the renal vasculature regions studied. accounts for this synthesis. The present chromatograms demonstrate that In summary, the present data show that human both vascular regions studied possess the capacity

skeletal muscle and kidney convert exogenous AA into PGs of the D2,E2, Fa and l2 series. The relative distribution of the PGs formed is not the same in the forearm and in the kidney, indicating the existence of tissue specificity in their PG biosynthesis. Furthermore, the results demonstrate the occurrence of considerable inter-individual variations in the quantitative relations between the various PGs formed. In the light of the pronounced and varying biological effects of different PGs, the present data, demonstrating tissue specificity and interindividual differences in formation, support the hypothesis that PGs may contribute to physiological regulatory processes in human tissues. This study was supported by grants from the Swedish Medical Research Council (project 14X-4341) and from "Forenade Liv", Mutual Group Insurance Company, PGD2, Stockholm, Sweden. 13,14-dihydr0-IS-keto-PGE~~ PGF,, PGE2 and PG12 were kindly supplied by Dr

J. Pike (Upjohn Company).

REFERENCES BLACKWELL, G. J., FLOWER, R. J. & VANE, J. R. 1975. Some characteristics of the prostaglandin synthetizing system in rabbit kidney microsomes. Biochim Biophys Acta 398: 178-190. BUNTING, S., GRYGLEWSKI, R., MONCADA. S. & VANE, J. R. 1976. Arterial walls generate from prostaglandin endoperoxides a substance (prostaglandin X) which relaxes strips of mesenteric and coeliac arteries and inhibits platelet aggregation. Prostaglandins 12: 897-913. FRIESINGER, G. Ch., OLEZ, 0.. SWEETMAN, B. J., NIES, A. S. & DATA, J. L. 1978. Prostaglandin D2, another renal prostaglandin? Prostaglandins IS: %9981. GRANSTROM. E. 1972. On the metabolism of prostaglandins Fa in female subjects. Structures of two metabolites in blood. Eur J Biochem 27: 462469. HAGENFELDT, L. & WENNMALM, A. 1975. Turnover of a prostaglandin precursor, arachidonic acid, in rheumatoid arthritis. Europ J Clin Invest 5 : 235-239. HAMBERG, M. 1969. Biosynthesis of prostaglandins in the renal medulla of rabbit. FEBS Lett 5 : 127-130. HAMBERG, M. 1972. Inhibition of prostaglandin synthesis in man. Biochem Biophys Res Commun 49: 720-726.

HAMBERG. M. & SAMUELSSON, B. 1966. Prostaglandins in human seminal plasma. J Biol Chem 241: 257263. HAMBERG, M: & SAMUELSSON, B. 1971. On the metabolism of prostaglandins E, and E, in man. J Biol Chem 22:67134721. HEDQVIST. P. 1972. Prostaglandin-induced inhibition of vascular tone and reactivity in the cat's hindleg in vivo. Eur J Pharmacol 17: 157-162. ISAKSON, P. C., RAZ. A., DENNY, S. E., PURE, E. & NEEDLEMAN, Ph. 1977. A novel prostaglandin is the major product of arachidonic acid metabolism in rabbit heart. Proc Natl Acad Sci USA 1: 101-105. KARIM, S. M. M. 1976. Prostaglandins: physiological pharmacological and pathophysiological aspects. MTP Press Ltd. KILBOM, A. & WENNMALM, A. 1976. Endogenous prostaglandins as local regulators of blood tlow in man: Effect of indomethacin on reactive and functional hyperaemia J Physiol257: 109-121. KOCSIS, J. J.. HERNANDOVICH, J., SILVER, M. J., SMITH, J. B. & INGERMAN, C. 1973.Duration of inhibition of platelet prostaglandin formation and aggregation by ingested aspirin or indomethacin. Prostaglandins 3: 141-144. LEE, J. B.. CROWSHAW, K., TAKMAN. B. H. & ATTREP, K. A. 1967. The identification of prostaglandins El, Fa and A, from rabbit kidney medulla. Biochem J 105: 1251-1260. NOWAK, J. & WENNMALM, A. 1978a. Influence of indomethacin and of prostaglandin E, on total and regional blood flow in man. Acta Physiol Scand 102:484-491. NOWAK, J. & WENNMALM, A. I978 b. Effect of exercise on human arterial and regional venous plasma concentrations of prostaglandin E. Prostaglandins and Medicine 1: 489497. SAMUELSSON, B. & GREEN, K. 1974. Endogenous levels of 15-keto-dihydro-prostaglandins in human plasma. Parameters for monitoring prostaglandin synthesis. Biochem Med 1 I: 298-303. STASZEWSKA-BARCZAK, J. & VANE, J. R. 1975. The role of prostaglandins in the local control of circulation. Clin Exptl Pharmacol Physiol, Suppl. 2: 71-78. WAHREN, J. 1966. Quantitative aspects of blood flow and oxygen uptake in the human forearm during rhythmic exercise. Acta Physiol Scand, Suppl. 269. WEKSLER, B. B.. MARCUS, A. J. & JAFFE, E. A. 1977. Synthesis of prostaglandin I, (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Acad Sci USA 9: 3922-3926. WHORTON, R.. FROLICH, J. C. & OATES, J. A. 1977. Prostacyclin is produced in renal cortical microsomes. Annual Winter Conference Prostaglandins, Vail, Colorado.