Insulin antagonizes epinephrine activation of the membrane transport ...

“%E JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 261, No. 7, Issue of March 5, pp. 2999-3001,1988 0 1986 by The American Societv of BioIozical Chemists. Inc. Printed in LfS.4.

Communication

the suppression of epinephrine-activated triglyceride lipase. In thepresent study, we have tested whether insulin can also suppress the stimulation of FA transport by epinephrine.

Insulin Antagonizes Epinephrine Activation of the Membrane Transport of Fatty Acids

EXPERIMENTAL PROCEDURES

PO~ENTIALSITE FOR HORMONAL SUPPRESSION OF LIPID MOBILIZATION*

Methods Preparation of Fat CeZls-Adipocytes were prepared from the epi(Received for publication, November 1,1985) didymal fat of one or two fed 170-200 g Sprague-Dawleyrats (Harlan Industries). The tissue was digested for 45 min at 37 “C with collaNada A. Abumrad, Patricia R. Perry, and genase (1mg/1.2 ml of buffer/] pair of fat pads) in modified KrebsRichard R. Whiteseil Ringer medium (3) buffered to pH7.4 with 25 mM bicarbonate (KRB). The digestion medium also contained 20 mg/ml (2%)bovine serum From the Departmentof Molecular Physiology and albumin (BSA, Fraction V, Sigma) and 2 mM glucose. The cells were Biophysics, Vanderbilt University School of Medicine, washed twice in KRB containing 2% albumin and then twice with Nashville, Tennessee 37232 Krebs-Ringer solution buffered to pH 7.4 with Hepes (KRH) containing 30 p~ BSA 10.2%)and 2 mM glucose. They were resuspended Membrane transportof long chainfatty acids in the in this medium a t a density of 30% (v/v), to contain approximately isolated rat adipocyte can be strongly stimulated by 0.3 X lo7 cells/ml. epinephrine (Abumrad,N. A,, Perry, P. R., and WhiAssay of Fatty Acid Xransport-Transport was usually measured tesell, R. R. (1985) J. Biol. Chern. 260, 9969-9971). by the initial rate of uptake of p’Ij]oleate a t 23 “C as described in We now report that insulin at physiological concentra- detail previously (1, 2). Aliquots of the cell suspension (30 GI) were tions can completely block or reverse the epinephrine incubated with the isotope (20 pl) for 1-40 s. At the desired time, 5 effect. Insulin was optimally effective at a concentra- ml OF ice-cold stop solution (KRH containing 200 p M phloretin) was The cells were separated from medium on glass fiber filters tion of about 0.1 nM in inhibiting transportactivation added. (Gelman Type A/E 61630, Gelman) using less than 10 mm Hg by 0.3 and 3 PM epinephrine (0.1 and 1.0 pg/ml). High filtration pressure. The filters were washed twice with stop solution concentrations of insulin (above1 nM) were generally (1 ml) and transferred to scintillation vials containing 6 ml of scinless effective and this was particularly true at the tillation fluid (ACS, Amersham) for counting. highest dose of epinephrine (1.0 Ng/ml). The insulin Zero-time radioactivity was determined by adding cells to stop effect was shown to be on the transport processsince solution premixed with isotope. The cells were then processed as insulin inhibited epinephrine activationof transport in described above. The incubation medium was prepared byfirst adding with stirring both directions(influx and efflux). No effect of insulin on basal transport was observed over a wide range of a few microliters of ethanol containing labeled fatty acid, of approconcentrationand specific activity, to water (1-2 ml). BSA was concentrations (0.01-10 nM). Insulin’s antagonism of priate added From a 10% stock solution in water buffered with 10 mM transport activation by epinephrine appeared depend-then Hepes (pH 7.4). The medium was made isotonic by addition of an ent on ATP metabolismsince it was abolished byprein- equal volume of 2 X concentrated buffer. cubating thecells with dinitrophenol (1 mM), DinitroHormonal Treatment-CeUs (30% (v/v)) were incubated at 37 “C phenol, however, could not reverse the insulin effect with epinephrine (0.1 or 1 pg/ml) for 5-10 min as indicated in the when exposure to the hormone preceded that todini- figure legends. Insulin was added at the indicated concentration trophenol, consistent with an action of insulin at the together with epinephrine, or 5 min prior to or following exposure to The cells were exposed to insulin for 10 min in all cases transport step. The data indicate that regulation of epinephrine. the membrane transport of fatty acids is a potential site after which they were allowed to cool to room temperature (3 min) for insulin’s action to suppress lipid mobilization. before the assay was started.

We have previously identified (1) and characterized (2) a membrane transport system with highaffinity for long chain fatty acids (FA’) in the ratadipocyte. More recently,we have shown that this system can be stimulated 5-10-fold by epinephrine (3). This finding indicated that changes in FA transport are physiologically coordinated with activation of lipase to mobilize fatty acids. The effect of insulin to inhibit lipid mobilizationhas been well documented and is one of its major metaboIic actions. This effect has been explained by

Materials o ninsulin , (Bovine Collagenase was obtained from W o ~ ~ i n ~and pancreas recrystallized) from SchwarzlMann (Lot P2 3695). [3H] Oleate was purchased from New England Nuclear. All reagents for buffer preparation except for Hepes were obtained from Fisher. Hepes was obtained from ICN Biomedicals Inc. RHC 80267 was a generous gift from Dr. Charles Sutherland of the Revlon Care Group, Tuckahoe, NY. RESULTS

Effect o ~ ~ ~onu~ lp ~~ n~ p ~ r ~ ~ FA e Transport- s ~ i ~ u ~ t e ~ Epinephrine (0.1 rglml) aspreviously reported (3) stimulated FA transport severalfold (Fig. I). Addition of insulin (1 nM) * This work was supported by National Institutes of Health Grant to the epinephrine-treated cells brought the stimulated transAM 33301. The costs of publication of this article were defrayed in port rate down to basal level. The order of addition of insulin part by the payment of page charges. This article must therefore be before, during, or after incubation with epinephrine did not hereby marked “advertisement” in accordance with 18 U.S.C. Section affect the results, indicating that insulin could either prevent 1734 solely to indicate this fact. The abbreviations used are: FA, long chain fatty acids; KRB, or reverse the epinephrine effect. No effect of insulin (0.01Krebs-Ringer bicarbonate buffer; KRH, Krebs-Ringer Hepes buffer; 10 nM) could be observed on basal FA transport (data not Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; BSA, bo- shown). vine serum albumin; DNP, dinitrophenol. In some experiments,the cells were first incubated with the

2999

Insulin Antagonism of FA Transport Activation

3000

Seconds FIG. 1. Antagonism by insulin of epinephrine stimulation of [%]oleate influx into adipocytes. Cells prepared as described under “Experimental Procedures” were exposed to epinephrine (A) or to epinephrine and insulin (A) for 10 min a t 37 “C. m, indicate that the cells were preincubated with the lipolytic inhibitor RHC 80267 (RHC) for 15 min a t 37 “C before addition of epinephrine and insulin. Preincubation with RHC 80267 did not significantly alter stimulation of FA transport by epinephrine (data notshown and Ref. 3). Oleate transport was assayed a t 23 “C as outlined under “Experimental Procedures.” In this and all other data shown fatty acid was 20 @a BSA, , 16 p ~ and , unbound FA, 0.12 PM. The times shown on the abscissa refer to seconds following addition of cells to isotope. Uptake of oleate is expressed per ml of packed cells or approximately 1 X lo7 cells. The curves shown are a composite of three experiments which were typical of a t least 10 more. EPI, epinephrine.

100

+INSULIN, InM

adipocytes preloaded with 13H]oleate. The action of insulin to suppress both influx and efflux providesadditional evidence localizing the site of hormone action to thetransport process. This had been demonstrated previously with regard to the epinephrine effect (3). Dose Response of the Insulin Effect-Insulin was found to be effective onthe transportprocess at concentrations which were as low or lower than those reported to be antilipolytic (4,5). As shown in Fig.3A, insulin at 0.01 nM was about 40% effective and at 0.1 nM was optimally effective in opposing the effect of 0.1 Icgfml ~pinephrine.Fig. 3B shows that these insulin concentrations were equally effectivein counteracting a 10 times higher doseof epinephrine (1pg/ml), In thiscase, however, the dose response curvewas clearly biphasicand the effectiveness of insulin was lost as its concentration was increased above 1 nM. Insulin at 10 nM was completely ineffective and, in some cases, augmentedthe epinephrine effect. This biphasic pattern of the insulin dose response curvehas its antilipolytic action (4,5). been described previously for Effect ofDinitrophenol (DNP)--Preincubation of cells with DNP under conditions (1mM DNP for 5 min at 37 “C)which have been shown to markedly reduce cellular ATP (6) abolished the stimulation of FA transport by epinephrine2(data not shown). DNP treatment, however, couldnot reverse transport stimulat~onwhen exposare to the hormone preceded it (Fig. 4).The latterfinding allowed usto check whether DNP exposure would also abolish the insulin effect. As shown, FA transport in cells first treated with epinephrine (5 mid, then with DNP (5 rnin), and finally withinsulin (10 min) remained as fully stimulated as in cells treated only with epinephrine. These data suggestedthat theinsulin effect on transport was dependent on ATP, as has been observed for many of its other actions. Onthe other hand, similar to thecase with epinephrine, treatment with DNP was ineffective in reversing the insulin effect. FA transport in cells exposedto epinephrine (5 min), insulin (10 rnin), and then to DNP (5 min) was inhibited A

-

0 0

I

2

Minutes after dilution FIG. 2. Antagonism by insulin of epinephrine stimulation of 13H]oleateefflux from adipytes. 40 pl (30% (v/v)) of control, epinephrine, or epinephrine-plus-insulin-treated cells (10 min at 37 “C) were incubated for 12 s with 25 p1 of medium containing t3H] oleate (5 pCi/ml). Buffer (4 ml) at 23 “C containing 1%BSA was added to start efflux (0 time on abscissa), and the cell suspension,

8 04,..OI

0.1

1.0

10

Insulin conc.(nM)

diluted about 60-fold, was sampled at the times indicated. Aliquots (1 ml) were pipetted into 5 ml of chilled stop solution. Cellular radioacti~tywas determined as described under “ExperimentalProcedures.” The datashown are a composite of three experiments.

lipolytic inhibitor RHC 80267 (10 pM) for 15 min at 37 “C prior to exposure to both hormones. As previously reported, this treatment completely abolished lipolysis induced by epinephrine without affecting epinephrine stimulation of FA transport (3). Similarly, treatment with RHC 80267 did not s i ~ i f i c ~ talter l y theability of insulin to inhibit epinephrinestimulated rates. Thus, the insulin effect on transport, like that of epinephrine, was not secondary to effects on lipase activity. More relevant to the physiological action of insulin to suppress lipid mobilization is the finding that italso inhibited stimulation of fatty acid efflux by epinephrine. As shown in Fig. 2 addition of 1 nM insulin almost completely abolished the epinephrine-induced acceleration of isotope efflux from

FIG. 3. Dose response of the insulin effect on[‘Hjoleate transport. Cells were treated for 10 rnin at 37 “C with 0.1 pg/ml ( A ) or 1 pg/ml ( B ) epinephrine and the concentrations of insulin indicated on the abscissa. Transport rate is shown as multiples of basal transport which is given the arbitrary value of 1. The bars indicate standard errors n = 5-10.

N. A. Abumrad, P. R. Perry, and R. R. Whitesell, unpublished observations.

v) 0)

u

E 6

1 0 E c

* a 4 a LL Q)

t

2

6

IO

seconds FIG. 4. Effect of dinitrophenol. Cells (30% (v/v)) were exposed to nitrop phenol (I mpul) for 5 min a t 37 “C after treatment with epinephrine for 5 min and before treatment with insulin (10 min). Transport of [3H]oleate was determined at 23 “C as described under “Experimental Procedures.”

and similar to that in cells treated with e p i n e p h r ~ eand insulin (data notshown). This also provided further evidence that insulin’s action was not mediated by a metabolic effect. DZSCUSSION

We have recent evidence which documents ATP requirement and involvement of the cyclic AMP-dependent protein kinase in the stimulation ofFA transport by epinephrine.’ Thus, protein phosphorylation appears to play a role in transport activation. The present study would indicate that deactivation of transport by insulin may also involve phosphorylation. A likely site of phosphorylation could be one of the steps leading to activation of the cyclic AMP-degradingphosphodiesterase. ATP requirement for insulin activation of the phosphodiesterase has been established (6). Furthermore, our preliminary evidence’ indicates that insulin can only inhibit the transport stimulatory effect of those cyclic AMP ana-

logues whichare good substrates for the phosphodiesterase. Membrane transport of FA might constitute a regulatory site for the effects of both epinephrine and insulin on FA mobilization. It is well known that high concentrations of intracellular FA inhibit adenylate cyclase and consequently lipolysis (7). Thus, in the case of epinephrine, stimulation of FA transport should promote efflux ofFA releasedfrom triglyceride hydrolysis. This would prevent a build-up of intracellular FA which might beessential to trigger and sustain a high lipolyticrate. Conversely,antagonism by insulin of the transport-stimulating action of epinephrine would increase intracellular FA which then could act as an early signal to shut off triglyceride hydrolysis.In support of this interpretation are the striking parallels between the action of insulin on FA transport and that on lipolysis. The dose response of insulin for the two effects is similar andhas a biphasic character3 in both cases (Fig. 3 and Refs. 4 and 5). Also, as documented for insulin’s inhibition of lipolysis (8), its effect on FA transport is dependent on ATP andseems mediatedat least in part by an acceleration of cyclic AMP hydrolysis. On a general level our fiidings suggest an i m p o ~ n role t of fatty acid transport inoverall metabolism.It would appear that substrate utilization in many situations like fasting and diabetes is controlled through a coordinate regulation of the transport systems for fatty acids and glucose since both are modulated by insulin. REFERENCES 1. Abumrad, N. A., Perkins, R.C., Park, J. H., and Park, C. R. (1981) J. Bioi. Chem. 256,9183-9191 2. Abmrad, N. A., Park, J. N., and Park, C. R.(1984) J. Bioi. Chem. 259,8945-8953 3. Abumrad, N. A., Perry, P. R., and Whitesell, R. R. (1985) J. Biol. Chem. 260,9969-9971 4. Kono, T., and Barham, F. W. (1973) J. BioL Chem. 248,74177426 5 . Goodman, H. M. (1969) Proc. Soc. Exp. Biol. Med. 130, 97-100 6. Kono, T., Robinson, F. W., Sarver, J. A., Vega, F. V., and Pointer, R. H. (1977) J. Biol. Chem. 252,2226-2233 7. Rodbell, M. (1965) Ann. N. Y. Acnd. Sei. 131,302-307 8. Beebe, S. J., Redmon, J. B., Blackmore, P. F., and Corbin, J. D. (1985) J. Bid Chem. 260, 15781-15788

~The biphasic action of insulin on FA transport is probably not related to changes in cyclic AMP levels and might reflect a positive effect of the ~traceIluIar FA on its own membrane transport. In previous studies of lipolysis it has been shown that while the effect of insulin on epinephrine-stimulated triglyceride hydrolysis was biphasic that onCAMPproduction was monophasic (Ref. 4).