Sep 25, 2015 - secretion, red cells transiently concentrate the polar compound intracellularly. Onset and reversal of inhi- bition of cyclic AMP export by PGAl ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY :c) 1985 hy The American Society of Biological Chemists, tne.
Vol. 260, No, 21, Issue of September 25, pp. 11514--11519,1985 Printed in U.S.A.
Prostaglandin A1 Metabolism andInhibition of Cyclic AMP Extrusion by Avian Erythrocytes* (Received for publication, November 26, 1984)
Lynn E. Heasley and Laurence L. Brunton$ From the Divisions of Pharmacology and Cardiology, M-O13N, University of California, Sun Diego*La Jolla, California 92092
Prostaglandins (PG) inhibit active cyclic AMP export from pigeon red cells, PGAl and PGAz most potently (Brunton, L, L., and Mayer, S . E. (1979) J. Biol. Chem. 254, 9714-9720). To probe the mechanism of this action of PGA,, we have studied the interaction of [3H]PGAI with suspensions of pigeon red cells. The interaction of PGAl with pigeon red cells is a multistep process of uptake, metabolism, and secretion. 13H] PGA, rapidly enters red cells and is promptly metabolized (V,,, 5: 1 nmol/min/107 cells) to a compound(s) that remains in the aqueous layer after ethylacetate extraction. The glutathione-depleting agent, diamide, inhibits formation of the PGA, metabolite. In agreement with the order of potency of other prostaglandins to inhibit cAMP efflux (A >> E = B > F), PGA2 forms a polar adduct whereas prostaglandins E2,B,, and Fz, do not. The red cells secrete thepolar metabolite of PGAl by a saturable mechanism (at 37 OC, K , = 0.6 &MyV,, = 0.5 pmol/min/107 cells) that lowered temperatures = 21 kcal/mol). Because uptake and metabinhibit (Eact ofismprogress with much greater ratesthan metabolite secretion, red cells transiently concentrate the polar compound intracellularly. Onset and reversal of inhibition of cyclic AMP export by PGAl coincide with accumulation and secretion of PGAl metabolite, suggesting that the polar metabolite acts at an intracellular site to inhibit cyclic AMP efflux. In the aceompanying Appendix, we present chromatographic and amino acid analyses demonstrating that the polar metabolite is a glutathione adduct of PGAI.
Intracellular cAMP escapes from many bacterial, acrasial, and metazoal cells, In avian erythrocytes and cultured mammalian cells, cAMP escapes by an energy-dependent mechanism that has many properties of active transport (1-3). One of the remarkable features of cAMP extrusion is its inhibition by certain prostaglandins (2, 3). We have characterized this inhibitory effect of prostaglandins in pigeon erythrocytes, a system in which PGAI’ is the most potent inhibitor (2). The effect of PGAl is rapid, irreversible to washing, enhanced by lower temperatures, and unrelatedto alterations of adenylate cyclase activity or cellular ATP. In addition, the order of potency of the prostaglandins as inhibitors of cAMP efflux (A >> E = €3 > F; PGI, and thromboxane BZinactive) is not * This work was supported by National Institutes of Health Grant GM 25819 and Research Career Development Award HL 00935 (to L . L. B.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18U.S.C. Section 1734 solelyto indicate this fact. f To whom correspondence and requests for reprints should be addressed. The abbreviation used is: PG, prostaglandin.
their order of lipophilicity (B > A > E > F), suggesting that the PG-red cell interaction is more specific and complex than a simple partitioning between aqueous and hydrophobic membrane phases. Using these characteristics as a guide, we have examined the interaction of radioactive PGAl with suspensions of pigeon red ceIls, looking for a component of interaction that accounts for the inhibition of cAMP export. In thispaper, we present data supporting the hypothesis that PGAl is rapidly metabo~izedto a polar compound within the red cell and that this polar metabolite, acting at theinner surface of the plasma membrane, is the actual inhibitor of cAMP efflux. In the accompanying Appendix, we present data identifying the polar metabolite as a glutathione adduct of PGA,. E X F E ~ I ~ E N T AFRO~EDURES L
Cell S ~ p e ~ i o ~ - B l o drawn od into a heparinized syringe from the wing vein of a male, white Carneux pigeon was diluted 10- to 25fold with 166 mM NaCI, 10 mM EDTA (pH 7 ) and centrifuged (5 min, 200 X g, 4 “C). The supernatant and buffy coat was aspirated and the cell pellet was suspended in 15 ml of Earle’s solution (containing in g/l: NaCI,6.8; KCI, 0.4; NaH2P04.HzO, 0.14; MgS04.7 H20,0.2; NaHC03,2.2; glucose, 2; equilibrated with 95% 02,5% C02,pH 7.4). The cells were again collected by centrifugation. This procedure was repeated t,wice more and the cells were finally suspended to a hematocrit of 10% (5 X lo8 cells/ml) in Earle’s solution (containing the phosphodiesterase inhibitor R020-1724 (50 p M ) when CAMP was measured) and allowed to equilibrate at 37 “C for 30 min prior to experiments. Assays of [3H]PGA1Accumulation and Metabolite Formation-Red cell suspensions were added to polypropylene tubes containing 5 nM l3HIPGA1and various concentrations of unlabeled PGAI. The tubes were flushed with OJCO,, capped, and incubated in a shaking water bath (usually at 37 “C). Incubations were terminated by placing duplicate 100-p1 aliquots of the suspension into 400-pi microfuge tubes containing adrop of an inert phthalate/sebecat,e mixture (density = 1.02 g/ml). Following a 20-s centrifugation in a Beckman microfuge B, the supernatants (above oil layer) were removed and the bottoms of the tubes containingthe packed cells were cut off and placed in scintillation vials. The cellular material was bleached with 200 pl of 30% hydrogen peroxide, 60% perchloric acid (2:1), then 100 pl of 15% ascorbic acid was added to each vial to reduce subsequent chemiluminescence. After addition of scintillant, samples were counted in a scintillation spectrometer with an efficiency of 20%. For separation of PGAl and polar metabolites, supernatant fractions (90 PI) of cell suspensions were acidified wit.h 4 volumes of 2% formic acid and extractedtwice with 2 volumes (1 ml) of ethylacetate. Cellular PGA, was extracted similarly following lysis of the cell pellet in 10 volumes (100 pl) of water. This procedure extracted more than 97% of radiolabeled PGA, from Earle’s buffer and separated PGAl from polar metabolites that remained in the aqueous phase. Miscellaneous Procedures-Cyclic AMP extrusion was analyzed as previously described (2) using the protein binding assay of Gilman (4) to quantify CAMP. ATP was assayed in neutralized perchloric acid extracts of cell suspensions by a coupled enzyme fluorescent assay (5). Cellular glutathione was assayed in perchloric acid extracts according to Beutler (6). Cell counts were made with a hemocytometer.
11514
P r o s t u g ~ n d ~AIn ~ e t a b o and ~ ~~nhibi~~on s ~ of Cyclic AMP Efflux
11515
Materials-['H]PGA1 (52 Ci/mmol; New England Nuclear), ['HI PGF2, (80 Ci/mmol; Amemham), ['*C]PGE, (56 mCi/mmol; Arnersham, gift of Dr. Morton Printz, UCSD). [3H]PGBl (52 Ci/mmol), prepared from [3H]PGA, by methanolic KOH treatment (71, COchromatographed on silica gel thin layer plates with authentic PGBI. ["H]PGA2, prepared by acid-catalyzed dehydration of [3H]PGE~ (160 Ci/mmol) (7), was purified by thin layer chromato~aphy(RF= 0.80 on silica gel plates developed in chloroform/methanol/acetic acid (90:5:5), where RF of PGE2 = 0.33. Unlabeled prostaglandins (generously supplied by the Upjohn Co.) were stored as 10 mM stocks in ethanol. R020-1724, (+)-~-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone, was a gift of Hoffman-La Roche. RESULTS
Znhibit~o~ of cAMP Export by PGA,-Having previously established ( 2 ) that PGAl is the most potent inhibitor of cAMP efflux among the common prostaglandins, we characterized the effect of PGA, in detail (Fig. 1).At 37 "C, PGAl inhibits extrusion of cAMP in aconcentration-dependent manner with an EC6, = 41 nM. Lowering the temperature increases the apparent potency of PGA,; at 30 "C, its EC," = 7 nM. Interuction of [3H]PGA, with Celts-With these characteristics in mind, we examined the interaction of red cells with ['HHJPGA,. We exposed red cells (10% suspension) to 5 nM 3H-ligand in the absence and presence of an excess (30 PM) of unlabeled ligand to judge whether a saturable component of interactionexisted over that concentration range. The results (Fig. 2 A ) were surprising. Cell suspensions incubated with f3H]PGA1alone yielded a low level of cell-associated ligand that appeared to reachasteady state by 10 min. Curiously, incubation of cells with [3H]PGAIplus 30 p M unlabeled,iigand produced not the expected isotope dilution but rather anenhanced association of 3H-ligandwith the cell pellet. Thinking that this anomalous effect might be due to the log,,
CPGAJ
FIG. 2. PGAl-stimulated accumulation oftritiumderived from [3H]PGAI.A, red cell suspensions (lo%, 37 "C) were exposed to 5 nM [3H]PGA1in the absence or presence of excess PGA, (30 PM) and, a t the indicated times, aliquots were centrifuged and cell pelletassociated radioactivity wasdet.ermined. B, red cell suspensions (37 "C) were exposed to 5 nM [3H]PGA1and various concentrations of excess PGAt for 30 min and were then sampled for cell-associated radioactivity. These data are represent.ative of four replicate experiments.
FIG. 1. Inhibition of cyclic AMP efflux by PGAI. Red cells loaded with cAMP by a 15-min exposure (37 "C)to 1PM isoproterenol were suspended (10% hematocrit) in fresh buffer containing 1 PM isoproterenol and 50 g~ R020-1724 at either 30 "C or 37 "C with varying [PGA,]. Samples (150 pl) were withdrawn after 10 min and analyzed for extracellular and total cAMP as previously described (2). Extrusion is expressed as a percentage of that from control cells (without PGAI) during the same 10-min period at 30 "C, 56 pmol/ aliquot and at 37 "C, 110 pmol/aliquot. The data shown are representative of two (30 "C) and five (37 "C) replicate experiments.
very high concentration of excess PGA, (30 p ~ )we , analyzed the concentration dependence of the interaction in the presence of unlabeled PGAl in the range of 1 nM to 30 pM and at a time, 30 min, sufficient for achieving a steady state accordrevealed two phases ing to Fig. 2 A . These isotope dilutio~ data of interaction of PGA, with the red cells (Fig. 2B). At concen, competed with a trations of PGA, less than 0.3 p ~ PGA, moderately high apparent affinity (KO= 20 nM), an affinity appropriate for the inhibition of cAMP efflux (Fig. 1). Concentrations of PGA, in excess of 0.3 M M caused the anomalous accumulation of 3H-ligand in the cell pellet, such that up to 80% of available [3H]PGA1became associated with the cells. Simple cell fractionation studies indicated that most of the 3H-ligand was actually in the cytosolic fraction of the cell. Since the red cells comprised only 10% of the volume of the suspension, the association of 80% of the ligand with the cells suggested active accumulation. Assuming isotopic equilibrium, 80% of 30 p~ PGA, concentrated into the entire intracellular volume implied an intracellular concentration approaching 0.24 mM, with 6 p M PGA, remaining outside the cells. By the simple criterion of resistance to metabolic poi-
P r o ~ t ~ ~AuI M~ i en t a and ~ Inhibition ~ ~ ~ of Cyclic AMP Efflux
11516
sons, however, the interaction was not active: concentrations metabolism, and secretion of radiolabeled material derived of rotenone (1pM), KCN (300 p ~ ) and , dinitrophenol (500 from i3H]PGAl. p ~ that ) reduced cellular ATP by 79%, 74%, and 97%, reMetabolism of PGAl by Red Cell Suspenswns-To test the spectively, had no effect onthis apparent accumulation. possibility of cellular metabolism of [3H]PGA,, we exposed We also tested the possibility that athigher concentrations, suspensions of pigeon red cells to PGA, (1 p ~6 ,X lo6dpm/ PGA, formed aggregates or micelles that interacted with the nmol), withdrawing samples over time. Each sample was red cell and led to bulk uptake. As a simple means of deter- acidified and extracted with ethylacetate such that born fide mining whether PGA, molecules form aggregatesin solution, PGA, (extractable into ethylacetate) could be distinguished we chromatographed 10 nM [3H]PGAlwith and without ex- from polar metabolites of PGA, (water-soluble,not extracted cess (100 p ~ PGA, ) over a gel exclusion column (Bio-Gel P- into ethylacetate; see "Experimental Procedures"). By this 2,0.9 X 57cm; using 0.4-ml fractions, 24 "C). I3H]PGA1 simple separation, we found that pigeon red cellsconvert f3H] (detected as radioactivity) and unlabeledPGA, (detected PGAl into a polar compound (Fig.4). The conversion is rapid spectrophotometrically at 279 nm after base-catalyzed iso- and extensive; indeed, we were not able to distinguish entry merization to PGB,, Ref. 7) co-eluted in fraction 50 with 5- from metabolism.Within 10 min, the cells metabolizegreater than 80% of added ligand by a process that the addition of fold dilution during passage over the column. The elution excessPGA, (upto 100 PM) fails to inhibit significantly. profile of [3H]PGA1 was unchanged by the presence of excess Thus, themetabolism isnot saturable over the range of PGA, unlabeled PGA,, a finding that argues against the formation that we have studied. Using the greatest rate of metabolism of PGA, micelles and their participation in PGA, accumula- achieved at the highest [PGA,], we calculate that therate for tion by erythrocytes. entryand metabolism must exceed 1 nmol/min/107cells. We found what we believe is the proper explanation for Following uptakeand metabolism of [3H]PGAt,redcells PGA,-enhanced accumulation of PGA, whenwe did a detailed release the metabolite into the extracellular space (Fig. 4). analysis of the time course of the interaction of 13H]PGA1 This process appears to continue until the estimated intraand varying concentrations of unlabeled PGA, with redcells cellular and extracellular concentrations are approximately (Fig. 3). Tritium accumulated in the cell pellet at early times equal (assuming volume of distribution of polar metabolite and later reappeared in the extracellular medium. Unlabeled equals entire cell volume). To characterize the kinetics of metabolite release distinct PGA, did not reduce the initial uptake but noticeably slowed the later efflux of radioactive material. These data demon- from uptake and metabolism, we exposed cells briefly(0.5 to strate that theanomalous accumulative component (Fig. 2 A ) 5 min) to various concentrations of I3H]PGA1 to load them resulted from our choosing an incubation time that did not with metabolite, then washed the cells free of extracellular represent a steady state over the entire concentration range. ligand at 4 "C (at which temperature metabolite efflux is Indeed, as the inset to Fig. 3 demonstrates, one gets vastly inhibited; see Fig. 6). Following suspension in fresh medium, different results by sampling at early (1 min), intermediate the cells were warmed to 37 "C and sampled periodically to (30 min), and late (240 min) times. A t 240 min, there is an obtain initial rates of metabolite efflux. Assuming that the polar metabolite is a singlecompound and is distributed overall isotope dilution, as originally expected, with a KO in within the entirecell volume,we derive a K, of 0.6 p~ and a the range of 0.3 FM. This result suggests that 3H-liganddoes maximal rate of metabolite efflux of 0.5 pmol/min/107 cells interact with a saturable component (that is probably involved (Fig. 5). Thus, efflux of polar metabolite is readily saturable with production or removal of a polar metabolite of PGAI; see in terms of available intracellular metabolite (0.24 mM under below). These data suggested to us that our measurements of the conditions ofFig. 1) and with respect to the rate of cell-ligand interactions represented a composite of uptake,
OL
b
601 8 0
120
240
J
TIME (minutas)
FIG. 3. Time course of the interaction of ['HIPGAI with red cells as a function of [PGA,]. Cell suspensions (lo%, 37 "C) were exposed to 5 nM [3H]PGAland several concentrations of excess PGAI. Cell-associated radioactivity was assessed at the indicated times; the earliest time point represented is 1 min. Inset, data from the main figure presented to show the effect of incubation time on the anomalous accumulative component. The data arerepresentative of duplicate experiments.
FIG.4. Metabolism of PGAl by red cell suspensions. Red cell suspensions (10%) were exposed to t3H]PGA1 (1 p ~ 6 ,X lo6 dpmf nmol), then, periodically, aliquots (100 &I)were withdrawn, acidified, and extracted with ethylacetate (see "Experimental Procedures"). Individual data points are from two experiments, representative of five replicates. The figure displays the cellular and extracellular distribution of the total ra~oactivity (-58,000 dpm) in a 100-pl aliquot. When diamide (3 mM) was included, it was added 5 min prior to PGAl addition. This treatmentrapidly reduced cellular GSH from 3.4 nmol/107 cells to
