Prostacyclin metabolites in human plasma - ASCPT

1 downloads 0 Views 315KB Size Report
product of prostacyclin. When the material of this peak was derivatized to the methoxime methyl ester trimethylsilyl ether and analyzed by gas ...
Prostacyclin metabolites in human plasma The major metabolites of prostacyclin (PGI2) in human plasma have been determined after

intravenous infusion of tritium-labeled and unlabeled prostacyclin. Plasma was extracted and chromato graphed. On high-pressure liquid chromatography (HPLC), several radioactive peaks could be resolved. The major peak containing 41.6% of the radioactivity had the retention volume of authentic 6-keto-prostaglandin F1(6-keto-PGF), the stable in vitro hydrolysis product of prostacyclin. When the material of this peak was derivatized to the methoxime methyl ester trimethylsilyl ether and analyzed by gas chromatographymass spectrometry, the fragments ml: 508 and 598, which are characteristic of this derivative of 6-keto-PGF1 were detected. A much smaller peak representing 6.6% of the radioactivity eluted from HPLC with the same retention volume as dinor-6,15-diketo-13,14-dihydro-PGF1. On gas chromatographymass spectrometry this material resulted in the fragments ml: 527, 468, 437, and 347, which are characteristic for this prostanoid. Finally, 10.1% of the radioactivity with ions ml: 571, 481, 391, and 354 on mass spectrometric analysis could be identified as dinor-6,15-diketo-13,14-dihydro-20-carboxyl-PGF1. It is concluded that 6-keto-PGF1,, represents the major breakdown product of prostacyclin in human plasma. In addition, dinor-6,15-diketo-13,14-dihydro-PGF, and its (o-oxidized analog could be identified circulating metabolites.

B. Rosenkranz, M.D., C. Fischer, Ph.D., and J. C. FrOlich, M.D.

Stuttgart, West Germany Institute for Clinical Pharmacology

Prostacyclin is a recently discovered prostanoid that exerts many biologic effects. It dilates blood vessels, inhibits platelet aggregation, and induces renin release (for review see reference 10). It has been suggested that changes in endogenous prostacyclin production are involved in the pathogenesis of several diseases in which vascular or platelet abnormalities

Supported by the Robert Bosch Foundation, Stuttgart. Received for publication Sept. 8, 1980. Accepted for publication Nov. 24, 1980. Reprint requests to: Priv. Doz. Dr. med. J. C. Frolich, Institut fiir Klinische Pharmakologie, Auerbachstrasse 112, 7000 Stuttgart 50, West Germany.

420

observed* 6 Prostacyclin may have therapeutic uses. Prostacyclin is unstable at physiologic pH and cannot be measured with precision because of hydrolysis to 6-keto-prostaglandin (6-keto-PGF1,),7 which occurs within minutes after addition to plasma. To investigate prostacyclin synthesis in man, determination of its major urinary metabolite, dinor-6-ketoPGF1, '2 would be expected to be useful, but because of the long response time of urinary metabolite levels, rapid changes in prostacyclin synthesis might be reflected better in levels of plasma metabolites. Furthermore, unlike classical prostaglandins, prostacyclin has been postulated to be a circulating hormone.6 Levels of its are

0009-92361811030420+05$00.50/0 C) 1981 The C. V. Mosby Co.

Prostacyclin metabolites in human plasma

Volume 29

Number 3

421

(corn)

600-

400-

200-

20

30

50

4.0

60

70

E30

fraction number

Fig. 1. p,Porasil chromatogram of plasma extract after prostacyclin infusion. The flow rate was ml/min. One-milliliter fractions were collected and analyzed for radioactivity, and the materials corresponding to fractions 18 to 22, 38 to 40, and 41 to 48 were further analyzed by GC-MS (Figs. 2 to 4). 1

major plasma metabolite should be coupled tightly to its effects and rapidly reflect changes of endogenous production. Several studies on measurement of 6-keto-PGF1 in plasma have used radioimmunoassay and gas chromatography mass spectrometry (GC-MS)4' 5' 8 but no information is available on circulating metabolites of prostacyclin in man. We therefore examined the metabolic profile in plasma after infusion of prostacyclin. Material and methods

Prostacyclin (100 pCi) that contained tritium at C-16 to C-19 (specific activity 0.73 Ci/mmole) was infused intravenously at the rate of ng/kg/min into a healthy man. Another subject was injected with 0.2 mg unlabeled prostacyclin at the rate of 5 ng/kg/min. The identity and purity of the labeled prostacyclin were established by two different systems of the thin-layer chromatography.' When labeled compound was given, 250 ml venous blood were drawn with sterile syringes 10 hr after beginning the infusion. It was assumed that the plasma levels of the metabolites had reached steady-state conditions at that time. Heparin was added to prevent coagulation and blood was centrifuged quickly and the plasma stored frozen. During 1

infusion of unlabeled compound, blood was collected under identical conditions. The plasmas drawn during infusion of the labeled and the unlabeled prostacyclin were combined, added to 600 ml of acetone, and centrifuged. The supernatant was partly evaporated, acidified to pH 3.0 with formic acid, and then extracted with ethyl acetate. The polar phase still contained 31% of the radioactivity obtained from plasma. This material was not worked up further. The organic phase was purified by use of 5 gm silicic acid with chloroform/acetone : 1 as solvent. Subsequently, the material that contained 42% of the radioactivity obtained from plasma was further chromatographed on an open-bed, reversed-phase column (10 gm of C-18-bonded HI-FLOSIL) using a gradient of 10%, 30%, 50%, and 80% acetonitrile in water (80 ml each). The flow rate was about 25 ml/hr. The recovery on this column was 93%. The fractions containing radioactivity were pooled and further chromatographed on high-pressure liquid chromatography (HPLC) (p,Porasil column, Waters).2 Radioactivity was determined with a Berthold Liquid Scintillation Counter (Betaszint BF 5000). An internal standard was used for calibration. The recovery during high-pressure liquid 1

422

Clin. Pharmacol. Titer. March 1981

Rosenkranz et al.

relative peak areas

Mt-31

598.3

Mt-31-90

m/z

M

t -31

527.3

11%

468.2

8%

437.3

29%

508 4 Fig. 2. Multiple-ion detection tracing of fractions 41 to 48 of the HPLC chromatogram shown in Fig. 1. The material was derivatized to the methoxime methyl ester trimethylsilyl ether and analyzed by GC-MS. These fragments and the retention time are characteristic for 6-keto-PGF1.

chromatography (HPLC) was 57%. The materials belonging to the prominent peaks were derivatized to the methoxime methyl ester trimethylsily1 ethers and analyzed by GC-MS. Because of the small amount of material, a multiple ionmonitoring technique was used. The fragments for monitoring were chosen according to the prostanoids expected because of the retention volumes on APorasil HPLC from previous investigations on the urinary metabolites of prostacyclin in man. The peak areas of the ions were related to the fragment with the maximum peak area, which was defined as 100%, and were compared with the corresponding peaks of the authentic prostanoids from other in vivo studies. GC-MS analysis was performed on a HewlettPackard 5985 A GC-MS equipped with a capillary column (10 m Supelco SP-2100) and a jet separator. Electron impact ionization was performed at an energy of 70 eV. The temperature of the injection port was kept at 2500 and that of the ion source at 200°. For gas chromatography a temperature program was chosen with the substances of interest eluting at a retention time between 7 and 13 min. Helium was used as carrier gas. Results

The chromatogram obtained from AP or asil HPLC of the plasma extract is shown in Fig. 1.

347.2

M

t -31- 2x 90

100

Fig. 3. Fragmentation pattern of fractions 18 to 22 (Fig. 1). The graph of each of the four peaks is calibrated separately. For each fragment the largest peak is taken for calibration of each tracing. Therefore, the tracings for each fragment are calibrated differently and cannot be compared with each other. The fragments and the retention time are characteristic for dinor-6 ,15-diketo-13 ,14-dihydro-PGF10.

The major peak containing 41.6% of the radioactivity recovered from HPLC had the same retention volume as authentic 6-ketoPGF1 (41 to 48 m1). The two fragments m/z 508 (Mt-31-90) and m/z 598 (Mt could be obtained from GC-MS (Fig. 2). GC-MS analysis of the methoxime methyl ester trimethylsilyl ether of authentic 6-keto-PGF1 resulted in identical fragments with the same retention time on GC as the material from HPLC. The relative peak area of the ion m/z 598 was 53% of the area of m/z 508. For authentic 6-keto-PGF1,, we obtained a corresponding value of 44% (data not shown). Six and six-tenths percent of the radioactivity could be recovered from HPLC with the same retention volume as dinor-6,15-diketo-13,14dihydro-PGF,,, (18 to 22 m1). On GC-MS this

31)

Volume 29 Number 3

material resulted in fragments m/z 527, 468, 437, and 347, which correspond to W-31, Mt-90, W-31-90, and W-31-2 x 90 of dinor-6,15-diketo-13,14-dihydro-PGF,0 as shown" (Fig. 3). The relative peak areas of the monitored ions are 11%, 8%, 29%, and 100%. The corresponding values of dinor-6,15-diketo13,14-dihydro-PGF, could be detected in urine after prostacyclin infusion by our group" are 13%, 7%, 43%, and 100%. The fragments m/z 571, 481, 391, and 354 could be obtained from material that eluted at the retention volume of 38 to 40 ml and contained 10.1% of the radioactivity (Fig. 4). The retention volume and the characteristic fragments corresponding to Mt 31, M-31 90, M1-31-2 x 90 and Mt-158-90 (miz 158 representing the top chain) are identical with the GC-MS parameters of dinor-6,15-diketo-13,14dihydro-20-carboxyl-PGF, detected in human urine by our group.13 The relative peak areas represent 6%, 43%, 100%, and 74%, whereas the reference material had values of 7%, 43%, 100%, and 100%. The other peaks, which each contained less than 10% of the radioactivity, could not be identified.

Discussion Our results show that 6-keto-PGF1 is the major breakdown product in plasma after infusion of prostacyclin. It represents 41.6% of the radioactivity recovered from HPLC. Prostacyclin is unstable during acidic extraction and is hydrolyzed chemically to 6-keto-PGFI. Hence, the amount of labeled 6-keto-PGF1, detected in plasma after prostacyclin infusion may result from circulating 6-keto-PGF,, from unchanged prostacyclin, or from both. Our observation is at odds with that suggesting that no 6-ketoPGF1,,could be detected in human plasma after intravenous infusion of prostacyclin at the rates of 2.5 to 20 ng/kg/min by radioimmunoassay. From the amount of radioactivity in plasma during infusion, the relative amount of 6-ketoPGF1, and the specific activity of the infused prostacyclin, it can be calculated that the plasma concentration of infused 6-keto-PGF, amounted to 120 pg/ml during prostacyclin infusion at the rate of 1 ng/kg/min in our experiment. Since this level is above the detection

Prostacyclin metabolites in human plasma

423

relat lye pea',

area,

m/z

Mt-31

6%

5714 M

4-31-90

M

t -31-2890

M

t -158-90

43%

481.3

100%

391.2

354 2

11

74%

12

time (mm)

Fig. 4. Fragments of dinor-6,15-diketo-13,14-dihydro-20-carboxyl-PGF1 obtained from fractions 38 to 40 (Fig. 1). For further explanation, see Fig. 3.

limit of the radioimmunoassay of 100 to 200 pg/ml, the reason for this discrepancy is not clear. It cannot be explained by the suggestion that a differentiation between circulating prostacyclin and 6-keto-PGF1 a might be obtained by radioimmunoassay, because the work-up procedure used for this method also results in chemical degradation of prostacyclin.8 The amount of 120 pg/ml of 6-keto-PGF, calculated from our results is very close to the rise in the concentration of 6-keto-PGF, in plasma of 138 pg/ml detected by Hensby et al.5 with GC-MS after infusion of prostacyclin at the rate of 1 ng/kg/min into six normal subjects. Since the basal levels of 6-keto-PGF, in human plasma reported by that group were in the same range (about 190 pg/ml), it can be speculated that the endogenous production of prostacyclin is about 1 ng/kg/min. That plasma levels of 6-keto-PGF,0 do not increase linearly after infusion of increasing doses of prostacyclin5 has to be taken into account, and therefore,

424

Rosenkranz et al.

it is difficult to come to quantitative conclusions

on the exact amount of endogenous prostacyclin

production from the amount of circulating

6-keto-PGF1. Dinor- 6 ,15-diketo-13 ,14 -dihydro-20-carboxyl-PGF1 and dinor-6 ,15-diketo-13,14-dihydroPGF, were also identified in the plasma. These metabolites also have been identified as metabolites of prostacyclin and 6-keto-PGF1 in human urine. They result from 0- and w-oxidation, 15-hydroxy-dehydrogenation, and reduction of the double bond at C-13 of 6-keto-PGR, The latter is formed from prostacyclin by nonenzymatic hydrolysis. It may be concluded that determination of 6-keto-PGF1 in plasma represents a valuable parameter for measurement of endogenous prostacyclin production in man. It has been shown that endothelial cells are a major source of prostacyclin.'° It may be speculated that the levels of 6-keto-PGF1 determined in plasma are often too high because of stimulation of prostacyclin release from the endothelial cells during blood sampling. We therefore suggest that additional determination of one of the circulating enzymatic metabolites of prostacyclin, e.g., of dinor-6,15-diketo-13,14-dihydro-20-carboxyl-PGF,, be performed to detect this possible error. We are grateful for excellent technical help provided by U. Bloschies.

References I. Dembinska-Kiec A, Gryglewska T, Zmuda A, Gryglewski RJ: The generation of prostacyclin by arteries and by coronary vascular bed is re-

duced in experimental atherosclerosis in rabbits. Prostaglandins 14:1025-1034, 1977. 2. Green K, Hamberg M, Samuelsson B, Frolich JC: Extraction and chromatographic procedures for purification of prostaglandins, thromboxanes,

Clin. Pharmacol. Ther. March 1981

prostacyclin and their metabolites, in Frolich JC, editor: Methods in prostaglandin research. New York, 1978, Raven Press, p. 15. Green K, Hamberg M, Samuelsson B, Smigel M, Frolich JC: Measurement of prostaglandins, thromboxanes, prostacyclin and their metabolites by gas liquid chromatographymass spectrometry, in Frolich JC, editor: Methods in prostaglandin research, New York, 1978, Raven Press, p. 39. Hensby CN, Barner PJ, Dollery CT, Dargie H: Production of 6-oxo-PGF10 by human lung in vivo. Lancet 2:1162-1163, 1979. Hensby CN, Fitzgerald GA, Friedman LA, Lewis PJ, Dollery CT: Measurement of 6-oxo-PGF1 in human plasma using gas chromatographymass spectrometry. Prostaglandins 18:731-736, 1979. Johnson M, Harrison HE, Raftery AT, Elder JB: Vascular prostacyclin may be reduced in diabetes in man. Lancet 1:325-326, 1979. Johnson RA, Morton DR, Kinner JH, Gorman RR, McGuire JC, Sun FF, Whittacker N, Bunting S, Salmon J, Moncada S, Vane JR: The chemical structure of prostaglandin X (prostacyclin). Prostaglandins 12:915-928, 1976. Mitchell MD: A sensitive radioimmunoassay for 6-keto-prostaglandin F,: preliminary observations on circulating concentrations. Prostaglandins Med 1:13-21, 1978. Moncada S, Korbut R, Bunting S, Vane JR: Prostacyclin is a circulating hormone. Nature 273:767-768, 1978. Moncada S, Vane JR: Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2 and prostacyclin. Pharmacol Rev 30:293-331, 1979. Patrono C, Ciabattoni G, Cinotti GA, Pugliese F, Maseri A, Chierchia S: Prostacyclin and renin release in man. Clin Res 27:426A, 1979. (Abst.) Rosenkranz B, Fischer C, Reimann I, Weimer KE, Beck G, Frolich JC: Identification of the major metabolite of prostacyclin and 6-keto-prostaglandin F, in man. Biochim Biophys Acta 619:207-213, 1980. Rosenkranz B, Fischer C, Weimer KE, Frolich JC: Metabolism of prostacyclin and 6-keto-prostaglandin F1, in man. J Biol Chem 255:1019410198, 1980: