Effect of Dietary Fat and Cholesterol Supplements on Glucagon ...

Lipids

Effect of Dietary Fat and Cholesterol Supplements on Glucagon Receptor Binding and Adenylate Cyclase Activity of Rat Liver Plasma Membrane1 CHI-RÃœLEE2 ANDMICHAEL W. HAMM Department of Nutrition, Rutgers university, New Brunswick, NJ 08903

ABSTRACT We have examined the effect on rat liver glucagon receptors and adenylate cyclase activity of a high saturated fat diet (butter fat), a high n-6 polyunsaturated fat diet (com oil), and a high n-3 polyunsaturated fat diet (menhaden fish oil) with or without additional cholesterol. The number and affinity of the glucagon receptors were unaffected by diet. The glucagon-stimulated adenylate cy clase activity from fish oil-fed animals exhibited the great est stimulation, followed by corn oil-fed animals. Butter fat-fed and all cholesterol-supplemented groups showed a depression in stimulation. The pattern of adenylate cy clase activity with fluoride stimulation was similar to that observed with glucagon. The effect of dietary fat on forskolin stimulation was similar to glucagon and fluoride for the groups without added cholesterol. However, the cho lesterol-supplemented groups did not exhibit a decreased activity. It is suggested that the effect of dietary lipid on glucagon-stimulated adenylate cyclase is not due to changes in the glucagon receptor, but rather due to changes in signal transduction, the G5-protein or the catalytic unit. J. fiutr. 119:539-546, 1989.

INDEXING KEY WORDS: dietary fat receptor

dietary cholesterol adenylate cyclase

glucagon

Adenylate cyclase (EC 4.6.1.1.), the major site of ac tion of many hormones, catalyzes the conversion of adenosine triphosphate (ATP) to adenosine 3',5'monophosphate (cAMP), the "second messenger" for a variety of metabolic regulations. It has been dem onstrated that the hormone-sensitive adenylate cyclase complex resides within the plasma membrane and con sists of at least three distinct components: a) the hor mone receptor binding site, b) a guanine nucleotide reg ulatory protein, and c) the catalytic site (1). In recent years the subunits of the hormone-sensitive adenylate cyclase system, the mechanism by which hormones bind to membrane receptor sites, and subsequent ac tivation of adenylate cyclase have been studied inten 0022-3166/89 $3.00 Â©1989 American Institute of Nutrition.

'New Jersey Agricultural Experiment Station Publication No. D-14137-4-88. Supported in part by State funds, National Institutes of Health grant No. HL38223, American Heart Association (NJ Af filiate! grant No. 87-05, and BiomÃ©dicalResearch Support grant No. RR 07058-21. 2Present address: Department of Pathology, Duke University Med ical Center, P.O. Box 3432, Durham, NC 27710.

Received 22 June 1988. Accepted 18 November 1988.

539

Downloaded from jn.nutrition.org by guest on July 10, 2011

sively. In the 1960s, studies reported that partial solubilization or phospholipase treatment of adenylate cyclase caused a loss of hormone responsiveness (2, 3). Since that time, in vitro studies have investigated the relationship between membrane lipids and adenylate cyclase activity (4, 5). Previous work (6-13) has examined the effect of di etary fat (principally saturated vs. n-6 polyunsaturated) on adenylate cyclase activity in a variety of tissues. The data from different laboratories have often pre sented conflicting viewpoints. Recently, Alam, Ren and Alam (12) investigated the effect of fish oil feeding on heart adenylate cyclase and reported increased activity relative to coconut oil-fed animals. Studies in our laboratory (13) have examined the in fluence of dietary fat on ÃŸ-adrenergicreceptor binding and isoproterenol-stimulated adenylate cyclase in rat heart. Corn oil-fed rats exhibited a greater ÃŸ-adrenergic receptor number and higher isoproterenol-stimulated adenylate cyclase activity than butter fat-fed rats. The effect of these same dietary fats on isoproterenol-stim ulated and propranolol-inhibited heart rate in the rat was also investigated (14), with the corn oil-fed rats demonstrating greater sensitivity to ÃŸ-adrenergic ago nists/antagonists than butter fat-fed rats. Current research has suggested that diets rich in n-3 polyunsaturated fatty acids have beneficial vascular effects such as an inhibitory effect on platelet aggre gation, a lowering of plasma cholesterol levels and the prevention of thrombosis (15, 16). However, research on cellular membrane functional effects is more lim ited. Furthermore, the amount of cholesterol in fish oil

LEE AND HAMM

540

MATERIALS AND METHODS

Animals and diets. Weanling Sprague-Dawley male rats (Ace Animal, Boyertown, PA) were housed indi vidually in wire-bottomed cages in a temperature-con

TABLE 1 Diet composition and fatty acid profile Diet Ingredient1

C/CH

B/CH

F/CH

% dry wt Com oil Butter fat Menhaden oil Cholesterol2

10

10

10

10 10 1 Fatty acid composition

1

10 1


trolled room (22-23Â°C)on a 12-h light/dark cycle and fed a nonpurified diet (Ralston-Purina, St. Louis, MO) for 3 d upon delivery to acclimate them to our facilities. Rats were then weight matched, divided into six dietary groups and fed their respective diets for 4 wk. The di etary regimen was designed as a two by three factorial: 10% (by wt) com oil (C) (courtesy of Best Foods Re search, Union City, NJ 07083), 10% butter fat (B),10% menhaden fish oil (F) (courtesy of Zapata Haynie, Reedsville, VA 22539), and each of these with a total of 1% cholesterol (C/CH, B/CH, and F/CH). The 1% cholesterol in all diets was the final concentration, in cluding the amount present in the dietary fat. Choles terol concentrations in the dietary fat were determined prior to cholesterol addition to the diet (19). The com position and the fatty acid profiles of the six diets are given in Table 1. Fatty acid compositions of the diets were determined by gas-liquid chromatography follow ing lipid extraction (20) and methylation of the fatty acids (21).There were no significant differences in peroxidation levels between the fresh diet and the diet after 24 h in a food cup (22). Rats were fed fresh diet

(e.g., 526 mg cholesterol per 100 g of menhaden oil) is not negligible. Therefore, the purpose of the present study was to investigate the interactive effect of the type of fatty acid with or without additional cholesterol supplements on glucagon receptor binding and glucagon-stimulated adenylate cyclase activity in rat liver. To further elucidate the mechanism for any such dietinduced alterations in glucagon-stimulated adenylate cyclase, fluoride- and forskolin-stimulated activity were also measured. Fluoride has been found to stimulate adenylate cyclase via the regulatory protein while forskolin may either act on the catalytic unit (17) or in teract with the coupling mechanism of the catalytic unit and regulatory protein (18). Therefore, these two ligands were used to begin elucidating the mechanism for dietary fat alterations.

Area % 10:012:014:014:l|n-5)16:016:l(n-7)18:018:l(n-9)18:2(n-6)18:3(n-3)20:118:4|n-3)20:4|n-6)20:5(n-3)22:4(n-6)22:5jn-3)22:6(n-3)â€”â€”â€”â€”10.30.31.725.160.71.1â€”â€”â€

'All diets contained (g/kg): 200 casein; 2 DL-methionine; 20 vitamin mix; 40 mineral mix; 2 choline bitartrate; 50 cellulose; 450 corn starch; 132.8 sucrose; 0.2 butylated hydroxytoluene; 2 tocopheryl acetate (500 lu/g). Vitamin mix contained (g/kg diet): 0.012 thiamin HC1; 0.012 riboflavin; 0.014 pyridoxine HC1; 0.06 niacin; 0.032 calcium pantothenate; 0.004 folie acid; 0.0004 biotin; 0.02 vitamin B-12 (0.1%); 0.016 retinyl palmitate (500,000 u/g); 0.005 cholecalciferol (400,000 u/g); 0.2 tocopheryl acetate (500 iu/g); 0.016 menadione sodium bisulfate. Mineral mix contained (g/kg diet): 20 calcium phosphate, dibasic; 2.96 sodium chloride; 8.8 potassium citrate, monohydrate; 2.08 potassium sulfate; 0.96 magnesium oxide; 0.14 manganous carbonate; 0.24 ferric citrate, USP; 0.064 zinc carbonate; 0.012 cupric carbonate; 0.004 potassium iodate; 0.004 sodium selenite; 0.022 chromium potassium sulfate. ^Cholesterol was added to bring the total in each diet to 10 g per kg of diet. The content of the dietary fat was first subtracted such that the following amounts of exogenous cholesterol were added: C/CH, 10 g; B/CH, 9.77 g; F/CH, 9.47 g.

541

DIETARY LIPID AND LIVER ADENYLATE CYCLASE

Adenylate cyclase assay. Adenylate cyclase activity determination was based on the conversion of a-[32P]ATP to [32P]cAMP.It was performed essentially as described by Salomon, Londos and Rodbell (33). The final incu bation mixture contained 10-20 n-gmembrane protein, 25 mM Tris (pH 7.5), 0.5 mM ATP, 10 mM MgCl2, 1 mM dithiothreitol, 20 M.McAMP, 5 mM phosphocreatine, and 3 units of creatine phosphokinase in a final reaction volume of 100 jil [32P]ATP (0.5 jiCi) and [3H]cAMP(0.01 n-Ci)were added to each tube. Prelim inary experiments with membranes made permeable with digitonin (0.1 mg/ml) demonstrated no differences in activity between intact and permeablized mem branes (unpublished observations). Basal adenylate cy clase activity was measured in the absence of GTP. Glucagon stimulation was measured at various con centrations (1 x 10~15to 1 x 10~5 M)in the presence of GTP. Sodium fluoride-stimulated (10 mM) and forskolin-stimulated (0.1 mM) adenylate cyclase activity were also measured. Reactions were incubated for 10 min at 30Â°Cand stopped with 100 (xl of 2% sodium dodecyl sulfate, 1 HIMcAMP, followed by 0.8 ml H2O. [32P]cAMPwas isolated by sequential column chromatography. Samples were quantitated by counting, with sample recovery calculated from [3H]cAMP recovery. Recovery was generally greater than 80%. Data were calculated as pmol/(min-mg) protein and presented as the n-units/mg protein where 1 (Â¿unit= 1 pmol/min. Statistics. Statistical significance was analyzed by two-way analysis of variance (ANOVA) and Duncan's multiple range test with the SAS statistical package (SAS,Statistical Analysis Systems, Cary, NC). Data were considered significantly different when P < 0.05. RESULTS

The glucagon receptor binding data is presented in Table 2. The dissociation constant (KD)and maximum binding (Bmax)were not different between any of the dietary treatments.


daily just prior to the start of the dark cycle and allowed food and water ad libitum. Food intake and body weights were recorded three times per week. There were no differences in food intake. Thus, the cholesterol intake of the B-fedgroup was 2.3% that of the B/CH-fed group, and that of the F-fed group was 5.3% that of the F/CHfed group. Final body weights in cholesterol-fed rats tended to be slightly greater but were only significantly different between C-fed and C/CH-fed rats (22). Liver plasma membrane preparation. Intact livers were removed and membranes prepared. Livers were significantly heavier in F-fed rats compared to C-fed and B-fed rats, while those from C/CH- and F/CH-fed rats were heavier than those from all other groups (22). Liver plasma membranes were purified according to the method of Touster et al. (23). Plasma membranes col lected from the 0.25 M-37.2% sucrose interface were washed with excess 5 mM Tris buffer (pH 7.5) and centrifuged at 85,000 x g for 30 min. Membranes were resuspended in 5 mM Tris buffer and frozen at - 70Â°C until used. The purity of the membrane fraction was determined by comparison of marker enzyme levels in both the initial homogenate and final membrane prep aration. Liver plasma membranes were monitored with 5'-nucleotidase (24), while microsomal and mitochondrial contamination were monitored by glucose-6phosphatase (25) and succinate-INT-reductase, respec tively (26). Protein was measured by the method of Lowry et al. (27) as modified by Markwell et al. (28). The purified liver membrane fraction was enriched in liver plasma membrane (16- to 20-fold) and depleted of mitochondria (0.1-fold) and microsomes (0.2-fold)with respect to the initial homogenate. Glucagon receptor binding. Glucagon receptor bind ing was performed according to Rojas and Birnbaumer (29) and Noda et al. (30). The assay medium contained 12sl-glucagon(5 ruvi)(New England Nuclear, Cat. # NEX207) in 20 mM N-2-hydroxyethylpiperazine-N'-2ethanesulfonic acid (HEPES)buffer, pH 7.6, 2% bovine serum albumin (BSA,fraction V, Sigma), 1 HIMEDTA, and 50 jig membrane protein with various concentra tions of unlabeled glucagon (2-200 nM) in a final vol ume of 0.25 ml. Nonspecific binding was measured in the presence of 200 nM (final concentration) unlabeled glucagon. The samples were incubated for 20 min at 30Â°C followed by the addition of 3 ml of 4Â°C wash buffer (20 mM HEPES, 0.2% BSA, pH 7.6). The samples were vortexed and vacuum filtered on presoaked (10% BSA overnight) 0.5n,m cellulose acetate filters, rinsed twice with 3 ml of wash buffer, and the filters quantitated in a gamma counter. The amount of specifically bound 125I-glucagonwas determined as the difference in the mean (duplicate determinations) of total and nonspe cific binding. Binding data were analyzed by the method of Scatchard (31) via EBDA/LIGAND program (32),us ing an IBM-PC. Binding affinity and receptor number are expressed as the KD (nM) and Bmax(pmol/mg pro tein).

2Glucagon

binding1KDnM3.993.734.514.394.204.660.350.550.620.960. receptor DietarygroupCBFC/CHB/CHF/CHTABLE protein5.10 0.574.73 0.495.23 0.405.75 0.765.31 0.874.74 1.00

ANOVA FAT CH FAT x CH

NS NS NS

'Data are means Â±SEMfor six animals in each group.

NS NS NS

LEE AND HAMM

542

The dose-response curve for adenylate cyclase stim ulation by glucagon (10~15 to 10 ~5 M) for each dietary group (n = 4) is presented in Figure 1. The y-axis rep resents the fold stimulation (specific activity of glucagon-stimulated adenylate cyclase divided by basal ac tivity). The various dietary treatments produced significant differences in the ability of glucagon to stimulate adenylate cyclase. At low glucagon concen trations ( < 10"10 M) there were no differences by twoway ANO VA. However, at 10~9 M glucagon there was

there was a significant cholesterol effect at lower glu cagon concentrations (5 x 10~9 M and 1 x 10~9 M), while there were no significant differences at the low est glucagon concentrations (1 x 10 ~10 M and 1 x 10-15M).

^ ^ c= O iâ€” en Iâ€”I

LO

15

10

(O

u

a zi o11 â€¢â€” en S J

10-15

10

io-5

-13

[GLUCAGON](M) FIGURE 1 Dose-response curve of fold stimulation (glu cagon stimulated/basal) of adenylate cyclase activity. Curve is the log of glucagon concentration versus fold stimulation of adenylate cyclase activity. Data are means of four deter minations in each group: (O) F; (D) C; (A) B; (â€¢)F/CH; {â€¢) C/CH; (^) B/CH. Statistical analysis by two-way ANOVA demonstrated no differences at glucagon concentrations s 10~10M. At 10~9M glucagon there was a significant effect on type of fat and cholesterol while at 10 "8 to 10~5M glucagon there was a significant effect on type of fat, cholesterol, as well as an interaction (P < 0.05). Pair-wise comparisons are described in Results.


an effect of dietary fat, and the liver plasma membranes from F-fed animals exhibited a greater (P < 0.05) fold stimulation than all groups except C-fed. At 10 ~5 M glucagon the liver membranes from F-fed animals showed a 19.3-fold stimulation (Â±3.5) (P < 0.05 com pared to all groups except C-fed), followed by the C-fed animals with a maximum fold stimulation of 14.3 Â± 2.1 (P < 0.05 compared to all groups except C/CH- and F-fed). The B-fed rats and the three cholesterol-supple mented groups demonstrated the lowest maximum fold stimulation (B, 8.1 Â±1.2; C/CH, 8.8 Â±2.0; B/CH, 6.9 Â±1.0; F/CH, 7.2 Â±0.8). These were not different from one another by pair-wise comparison. The two-way ANO VA demonstrated that there were significant dif ferences due to the type of fat, the addition of choles terol, as well as an interactive effect at glucagon con centrations from 1 x 10~8Mtol x 10~5M. Furthermore,

The specific activity of basal and glucagon-stimulated (l M.M)adenylate cyclase is presented in Figure 2. Two-way ANOVA showed a significant effect (P < 0.05) of dietary fat and of cholesterol on basal activity (Fig. la}. However, pair-wise comparison demonstrated no significant differences between C-fed (4.7 Â±0.5), B-fed (4.4 Â±1.0), and F-fed (6.1 Â±1.1) rats. Similarly, there were no significant differences in livers from C/CH-fed (6.9 Â±1.3), B/CH-fed (6.8 Â±0.2) and F/CH-fed (9.7 Â± 0.9) rats. The F- and F/CH-fed basal activities were sig nificantly different. In the case of 1 (JLMglucagon stimulation (Fig. 1b], there were marked dietary effects with differences due to dietary fat (P < 0.01), cholesterol (P < 0.05) and an interactive effect (P < 0.01) evident by two-way AN OVA. Pair-wise comparisons revealed that liver plasma membranes from F-fed animals demonstrated a signif icantly (P < 0.05) higher glucagon-stimulated adenylate cyclase activity (113.9 Â±11.7) relative to all other groups. C-fed animals exhibited the next highest activity (70.4 Â±1.8), and livers from B-fed animals were the lowest (37.6 Â±7.6). These were all different from one another (P < 0.05). In response to cholesterol, both the F/CHand C/CH-fed rats showed a suppression of adenylate cyclase activity relative to their non-supplemented counterparts (74.6 Â±9.4 and 44.4 Â±5.2, respectively). There were no differences in livers from B-fed animals with (54.5 Â±4.5) or without cholesterol. The specific activity of fluoride-stimulated adenylate cyclase is presented in Figure 3a. The results demon strated the same trends as the glucagon-stimulated ac tivity, except that by two-way ANOVA there was not a distinct effect of cholesterol. F-fed rats had the great est activity (105.9 Â±7.1) with livers of C-fed rats being significantly lower (86.5 Â±5.6) and B-fed rats demon strating the lowest activity (45.1 Â±8.3). These were all significantly different (P < 0.05) from one another by pair-wise comparison. Among the cholesterol-sup plemented groups the F/CH and C/CH groups were significantly lower than their non-supplemented coun terparts (84.7 Â± 10.9 and 57.1 Â±6.7, respectively). Livers from the B/CH-fed group did not exhibit sig nificantly altered activity (66.0 Â±4.7) compared to those from the B-fed group. The activity in the F/CH-fed group was significantly different from the C/CH-fed group. The pattern of activity with forskolin stimulation was somewhat different from both glucagon and fluor ide stimulation (Fig. 3b}. There were no significant dif ferences in forskolin-stimulated adenylate cyclase either between the F-fed (98.7 Â±13.7) and C-fed (74.5 Â±7.7) rats or between the C-fed and B-fed (42.2 Â±12.1) rats. However, the F-fed rats exhibited a significantly higher forskolin-stimulated activity than the B-fed rats (P < 0.05). Adenylate cyclase activity with cholesterol sup plementation was increased (B/CH, 114.1 Â± 10.9) or not changed (C/CH, 94.6 Â±12.0; F/CH, 117.1 Â±15.3) compared to their non-supplemented counterparts. None

543


B

F C/CHB/CHF/CH

B

F C/CHB/CHF/CH

FIGURE 2 Basal (left graph, Fig. 2a) and glucagon-stimulated (right graph, Fig. Ib] adenylate cyclase activity. Glucagon stimulation was at 1 (JLM.1 (Â¿unit= 1 pmol cAMP produced/min. All data are mean Â±SEMfor the number of animals indicated in parentheses. Diet groups with different superscripts are significantly different at P < 0.05. Basal: C(9)a, B(7)a, F(9)a, C/CH(6)ab, B/CH(6)ab, F/CH(6)b. Glucagon-stimulated: C(9)a, B(7)b, F(9)c, C/CH(6)bd, B/CH(6)ab; F/CH|6)a.

of the cholesterol-fed groups was different from an other. DISCUSSION The glucagon/adenylate cyclase system is a complex of three proteinsâ€”the hormone receptor, the guanine

C

B

F C/CH B/CH F/CH

nucleotide (Gs)protein and the catalytic unit (34). In vestigating the effect of dietary lipid on this system becomes a complex problem due to the number of pos sible points of interaction. An alteration in the mem brane lipid composition may affect any of the individ ual proteins and cause changes in the ultimate resultâ€” catalytic activity. Alternatively, there is evidence to

B

F

C/CH B/CH F/CH

FIGURE 3 Fluoride- (left graph, 3u) and forskolin-stimulated (right graph, 3b) adenylate cyclase activity. Fluoride (10 ITIM) and forskolin (0.1 HLM)concentrations were as indicated. 1 n.unit = 1 pmol cAMP produced/min. All data are means Â±SEM for the number of animals indicated in parentheses. Fluoride-stimulated: C(9)a, B(7)b, F(9)c, C/CH(6]M, B/CH(6)ab, F/CH(6)a. Forskolin-stimulated: C(9)ab, B(7)a, F(9)bc,C/CH(6)bc, B/CH(6)C, F/CH(6)C.


C

544

LEE AND HAMM

acids is downstream from the guanine nucleotide and, hence, at the catalytic unit. This would also tend to indicate that these fatty acid changes are functionally affecting, in this case, the inner hemileaflet of the plasma membrane. This is in agreement with the work of Alam, Ren and Alam (12) in the heart. In contrast with our data and others' (12) of higher hormone-stimulated adenylate cyclase activity in fish oil- and corn oil-fed groups, it has been reported that dietarily induced high levels of linoleic acid (10, 11) and linolenic acid (11) in liver plasma membrane de creases glucagon-stimulated adenylate cyclase activity. These discrepancies may be due to the differences in the quantity and/or the type of fat used. Their studies utilized 20% (by weight) fat diets and would be con sidered high fat diets compared to the 10% fat used in the present study. High fat diets might have altered the physiological state of the animals and thus modified the levels of circulating hormones. In addition, the fish oil used in our study is rich in C20:5(n-3) (EPA) and C22:6(n-3) (DHA), but not C18:3(n-3) (linolenate). We observed an increased incorporation of EPA and DHA in liver plasma membrane (22). These two fatty acids may have specific effects on the catalytic unit of the adenylate cyclase complex. Colard, Kervabon and Roy (37) have studied the effect in vitro of linoleate on aden ylate cyclase of rat liver plasma membrane. They found an increase in basal, fluoride-, and glucagon-stimulated adenylate cyclase activity in rat liver plasma membrane with increasing membrane linoleate. Prostaglandin E! (PGEJ-stimulated adenylate cyclase activity was also found to increase in mouse LM cells grown in a high 18:2(n-6) medium (38). These findings are consistent with our observations. Given the fact that we observed no changes in cholesterol/phospholipid molar ratio or membrane fluidity by fluorescence polarization (22) with these three treatments, it can be hypothesized that these effects are due to fatty acid changes. However, the exact rationale for these effects remains unclear. Since the major change in the membranes of C-fed versus B-fed rats is the (total n-6)/[(16:l(n-7) + 18: l(n-9)] ratio, while the liver plasma membranes from F-fed rats showed changes in longer-chain, more unsaturated fatty acids, it remains to be determined whether the effect is due to chain length, degree of unsaturation, or exact fatty acid positioning in particular phospholipids. The effect of dietary cholesterol is more difficult to interpret than that of dietary fatty acids. Dietary cho lesterol depressed activity in livers from C/CH- and F/ CH-fed rats relative to their non-supplemented coun terparts for glucagon fold stimulation (Fig. 1), glucagon specific activity (Fig. 2b], and fluoride specific activity (Fig. 3u). The activity in livers from B-fed animals was already at what may be considered a minimal level and, if anything, increased slightly with cholesterol feeding. It would appear that there is an impairment of either Gs-protein or the coupling process with cholesterol feeding. However, besides increases in membrane cho-


support the hypothesis that the process of activation of the catalytic unit involves a collision coupling be tween the protein units (35). This may also be affected by membrane lipid changes. The present data demonstrate that rats fed different dietary fats and/or cholesterol supplements do not ex hibit altered binding of glucagon for its receptor (Table 2). It is of interest that other receptors can be affected differently with a similar dietary protocol. In the rat heart we have observed a two- to threefold increase in the density of ÃŸ-receptorsin response to corn oil feeding relative to butter fat feeding (13). Similar data by Alam, Alam and Ren (8) have been reported with essential fatty acid deficiency. In liver it appears that changes in adenylate cyclase are not due to changes in the glu cagon receptor, whereas in the heart it appears that changes may be related to receptor density (8, 13). An other study by Alam, Ren and Alam (12) with fish oil feeding reported a lack of ÃŸ-receptor changes in the heart. Since the alterations in the glucagon receptor do not explain the observed changes, then changes in either adenylate cyclase, Gs-protein or the coupling of the system must be affected. Furthermore, that the proteins in question span different regions of membrane and membrane asymmetries with respect to lipid compo sitional changes may prove to be important (35). There are no differences in cholesterol/phospholipid molar ratio due to the type of dietary lipid but it is increased with cholesterol feeding in these membranes (22). Thus, it will be more instructive to consider the changes due to fatty acid alteration distinct from those due to cho lesterol. Investigators have related in vitro changes in membrane fluidity to adenylate cyclase changes (36), while others have reported no such effect in a dietary model (12). We have not been impressed that changes in this system are due to membrane fluidity alterations as we see no major changes due to these diets (22). Rather, we hypothesize the changes are due to the lipids themselves. We utilized fluoride and forskolin stimulation, along with glucagon, to help delineate the locus at which dietary manipulation induced changes in glucagonstimulated adenylate cyclase. Fluoride appears to ac tivate adenylate cyclase by acting directly on the reg ulatory protein. Most studies suggest that the action of forskolin on adenylate cyclase is at the catalytic unit (17), while others suggest that the forskolin activation may be via either a guanine nucleotide-dependent or a guanine nucleotide-independent component (18). Comparing the data for just the corn oil-, butter fat-, and fish oil-fed rats with the various adenylate cyclase activators (Fig. 2, 3) it is clear that the pattern of activity was identical for glucagon, fluoride, and for skolin activation. That is, in all three cases liver mem branes from F-fed rats had a greater activity than from C-fed rats, which, in turn, was greater than that from B-fed rats. This implies that the effect of altered fatty


ACKNOWLEDGMENT The authors would like to thank Barbara Hannon for typing the manuscript.

LITERATURE CITED 1. RODBELL, M., BIRNBAUMER, L. & POHL,S. L. (1971| Hormones, receptors, and adenyl cyclase activity in mammalian cells. In: The Role of Adenylate Cyclase and Cyclic 3',5'-AMP in Biolog ical System (Rail, T. W., Rodbell, M., Condliffe, P., eds.], pp. 59103, U.S. Govt. Printing Office, Washington, DC. 2. SUTHERLAND, E. W., RALL,T. W. & MENON,T. I. (1962] Adenyl cyclase I. Distribution, preparation, and properties. /. Biol. Chem. 237: 1220-1227. 3. POHL, S. L., KRANS,H. M. I., KOZYREFF,V., BIRNBAUMER, L. & RODBELL, M. (1971] The glucagon-sensitive adenylate cyclase system in plasma membranes of rat liver. /. Biol. Chem. 246: 4447-4454. 4. ORLY,J. & SCHRAMM,M. (1975] Fatty acid as modulators of membrane functions: Catecholamine-activated adenylate cy clase of the turkey erythrocyte. Proc. Nat. Acad. Sci. USA 72(9): 3433-3437. 5. DIPPLE,I. & HOUSLAY,M. D. (1978] The activity of glucagonstimulated adenylate cyclase from rat liver plasma membrane is modulated by the fluidity of its lipid environment. Biochem. ]. 174: 179-190. 6. ALAM,S. Q. & ALAM,B. S. (1986) Effect of essential fatty acid deficiency on acyl group composition, membrane fluidity and adenylate cyclase activity in plasma membranes of rat submandibular salivary glands. /. Ã‘utÃ-.116: 1620-1630. 7. MCMURCHIE,E. J., PATTEN,G. S., CHARNOCK,J. S. &.MCLENNAN, P. L. (1987) The interaction of dietary fatty acid and choles terol on catecholamine-stimulated adenylate cyclase activity in the rat heart. Biochim. Biochem. Acta 898: 137-153. 8. ALAM,S. Q., ALAM,B. S. & REN,Y.-F. (1987) Adenylate cyclase activity, membrane fluidity and fatty acid composition of rat heart in essential fatty acid deficiency. /. Mol. Cell Cardio!. 19: 465-475. 9. NEELANDS,P.}. & CLANDININ,M. T. (1983) Diet fat influences liver plasma-membrane lipid composition and glucagon-stimu lated adenylate cyclase activity. Biochem. Â¡.212: 573-583. 10. MORSON,L. A. & CLANDININ,M. T. (1985) Dietary linoleic acid modulates liver plasma membrane unsaturated fatty acid composition, phosphatidylcholine and cholesterol content, as well as glucagon stimulated adenylate cyclase activity. Nutr. Res. 5: 1113-1120. 11. MORSON,L. A. & CLANDININ,M. T. (1986) Diets varying in linoleic and linolenic acid content alter liver plasma membrane lipid composition and glucagon-stimulated adenylate cyclase ac tivity. /. Nutr. 116: 2355-2362. 12. ALAM,S. Q., REN, Y.-F., & ALAM,B. S. (1988) [3H]Forskolinand [3H]dihydroalprenolol binding sites and adenylate cyclase activity in heart of rats fed diets containing different oils. Lipids 23: 207-213. 13. HAMM,M. W. & SHEI,G.-J. (1988) Dietary lipid, adenyl cyclase activity, and ÃŸ-adrenergicbinding in rat hearts. FASEB ]. 1: A639. 14. RULKA,C. A. & HAMM,M. W. (1988) Dietary fat and the ÃŸadrenergic mediated chronotropic response in the rat. /. Nutr. 118: 1304-1310. 15. GOODNIGHT,S. H., WILLIAMS, A. S., IR., CONNOR,W. E. & ILLINGWORTH,D. R. (1982] Polyunsaturated fatty acids, hyperlipidemia, and thrombosis. Arteriosclerosis 2: 87â€”113. 16. HAY, C. R. M., DURBER,A. P. & SAYNOR,R. (1982) Effect of fish oil on platelet kinetics in patients with ischemie heart dis ease. Lancet 1: 1269-1272. 17. SEAMON,K. B. (1985] Activation of hormone-sensitive aden ylate cyclase by forskolin. Drug Dev. Res. 6: 181-192. 18. WONG, S. K.-F. & MARTIN,B. R. (1983] The role of a guanine nucleotide-binding protein in the activation of rat liver plasmamembrane adenylate cyclase by forskolin. Biochem. /. 216: 753759. 19. CHIAMORI,N. & HENRY,R.J. (1959) Study of the ferric chloride method of determination of total cholesterol and cholesterol es ters. Am. I. Clin. Pathol. 31: 305-309.


lesterol content produced by this regimen, there are also alterations in fatty acid composition (22). The pri mary effect is a large increase in the 18:l(n-9)/18:2(n6) ratio in CH-supplemented versus non-supplemented animals. Thus, the exact membrane lipid constituent inducing this alteration in G8-protein or the coupling process cannot be determined as of yet. Surprisingly, forskolin stimulation of adenylate cyclase (Fig. 3b] with cholesterol feeding has the opposite effect. It increases the activity to what may be considered a maximal level in all cases. Since we also observe a generalized increase in basal activity (by two-way ANOVA) with cholesterol feeding, the increased cholesterol may play a role in stabilizing the catalytic unit such that it is more sus ceptible to direct activation. This could also serve to decrease its ability to undergo collision coupling with the receptor/Gs complex and further explain the results with fluoride and glucagon stimulation. It has been demonstrated in human platelets (39) and Chinese hamster ovary cells (40) that an increase in membrane cholesterol content elevates basal activity but decreases the PGEj and NaF-stimulated activity. Another study with ÃŸ-adrenergic-stimulated adenyl ate cyclase of rat lung membrane suggested that mem brane cholesterol incorporation decreased catalytic unit activity without modifying transduction of the hor mone signal (41). McMurchie et al. (7) fed rats with diets supplemented with high levels of saturated or unsaturated fatty acid (sheep kidney fat and sunflower seed oil, respectively) in the presence of dietary cho lesterol. They reported that the catecholamine-stimulated adenylate cyclase activity was significantly in creased with dietary cholesterol supplements and was positively correlated with the value of the membrane cholesterol/phospholipid ratio. The discrepancies between the work reported here and that of McMurchie et al. (7) most probably are due to the systems employed. In the heart, we do not see the effects of cholesterol on fatty acid composition ob served in the liver (22). There are also differences in quantity of fat and length of feeding. In conclusion, the present data suggest that dietary lipid-induced alterations in liver glucagon-stimulated adenylate cyclase are not due to changes in glucagon receptor density. These alterations, with cholesterol supplementation, appear to be due to changes in signal transduction either in the regulatory protein per se or an interaction of hormone receptor/Gs protein with the catalytic unit. In the case of dietary lipid without ad ditional cholesterol, the effect appears to be at the cat alytic unit.

545

546

LEE AND HAMM

20. FoLCH,I., LEES,M. & STANLEY, G. H. S. (1957) A simple method for the isolation and purification of total lipids from animal tis sues. /. Biol. Chem. 226: 497-509. 21. MORRISON,W. R. & SMITH,L. M. |1964) Preparation of fatty acid methylesters and dimethylacetals from lipids with boron fluoride-methanol. /. Lipid Res. 5: 600-608. 22. LEE,C.-R., BEATTIE,O. & HAMM,M. W. (1989) Effect of dietary fat and cholesterol supplements on lipid characteristics of liver and heart plasma membrane. Nutr. Res. (in press). 23. TOUSTER,O., ARONSON,N. N., DULANEY,J. J. & HENDRICKSON, H. (1970) Isolation of rat liver plasma membranes: use of nucleotide pyrophosphatase and phosphodiesterase I as marker en zymes. /. Cell Biol. 47: 604-618. 24. NAITO, Y. & LOWENSTEIN, I. M. (1981) 5'-Nucleotidase from rat heart. Biochem. 20: 5188-5194. 25. DEDUVE, C., PRESSMAN,B. C, GIANETTO,R., WATTIAUX,R. & APPELMANS, F. (1955) Intracellular distribution patterns of en zymes in rat liver tissue. Biochem. J. 60: 604-617. 26.

27.

29. 30.

31.

40. SlNENSKY, M., MlNNEMAN & MOLINOFF, P. B.

(1979)

Increased

membrane acyl chain ordering activates adenylate cyclase. /. Biol. Chem. 254: 9135-9141. 41. SCARPACE,P. J., O'CONNOR, S. W. &. AfiRASS,I. B. (1985) Cholesterol modulation of ÃŸ-adrenergicreceptor characteristics. Biochim. Biophys. Acta 845: 520-525.


28.

SOTTACASA, G. L., KUYLENSTIERNA, B., ERNSTER, L. & BERGSTRAND,

A. (1967) An electron-transport system associated with the outer membrane of liver mitochondria. /. Cell Biol. 32: 415-438. LOWRY,O. H., ROSEBROUGH, N. J., FARR,A. L. & RANDALL,R. J. (1951) Protein measurement with the folin phenol reagent. /. Biol Chem. 193: 265-275. MARKWELL, M. A. K., HASS,S. M., BIEBER,L. L., TOLBERT,N. E. (1978) A modification of the Lowry procedure to simplify pro tein determination in membrane and lipoprotein samples. Anal. Biochem. 87: 206-210. ROÃ•AS, F. f. & BIRNBAUMER, L. (1985) Assays for the glucagon receptor binding. Methods in Enzymol. 109: 3-12. NODA, C., SHINJYO,F., TOMOMURA,A., KATO,S., NAKAMURA,T. & ICHIHARA,A. (1984) Mechanism of heterologous desensitization of the adenylate cyclase system by glucagon in primary cultures of adult rat hepatocytes. /. Biol. Chem. 259(12): 77477754. SCATCHARD,G. (1979) Attractions of proteins for small mol

ecules and ions. Ann. NY Acad. Sci. 51: 660-672. 32. MUNSON,P. J. & RODBARD,D. (1980) Ligand: A versatile com puterized approach for characterization of ligand-binding sys tems. Anal. Biochem. 107: 220-230. 33. SALOMON,Y., LONDOS,C. & RODBELL, M. (1974) A highly sen sitive adenylate cyclase assay. Anal. Biochem. 58: 541-548. 34. LEFKOWITZ, R. J., STADEL, J. M. & CARÃ“N, M. G. (1983) Adenylate cyclase-coupled beta-adrenergic receptors: structure and mech anisms of activation and desensitization. Ann. Rev. Biochem. 521: 159-186. 35. ROTHMAN,J. E. & LENARD,J. (1977) Membrane asymmetry. Science 195: 743-753. 36. HOUSLEY, M. D. & GORDON,L. M. (1983) The activity of aden ylate cyclase is regulated by the nature of its lipid environment. In: Current Topics in Membrane and Transport, Vol 18 (Kleinzeller, A. & Martin, B. R., eds.), pp. 179-231, Academic Press, NY. 37. COLARD,O., KERVABON, A. & ROY,C. (1980) Effects on aden ylate cyclase activities of unsaturated fatty acid incorporation into rat liver plasma membrane phospholipids: Specific modu lation by linoleate. Biochem. Biophys. Res. Commun. 95: 97102. 38. ENGELHARD,V. H., ESKO.,J. D., STORM,D. R. & GLASER,M. (1976) Modification of adenylate cyclase activity in LM cells by manipulation of membrane phospholipid composition in vivo. Proc. Nati. Acad. Sci. USA 73: 4482-4486. 39. SINHA,A. K., SAHTTIL,S. }. & COLMAN,R. W. (1977) Cyclic AMP metabolism in cholesterol-rich platelets. /. Biol. Chem. 252: 3310-3314.