Regulation of the activity and synthesis of malic enzyme in 3T3-L1 cells

ARCHIVES OF BIOCHEMISTRY Vol. 228, No. 1, January,

Regulation ALAN

AND BIOPHYSICS pp. 54-63, 1984

of the Activity and Synthesis of Malic Enzyme in 3T3-Ll G. GOODRIDGE,l

JUDITH

Departments of Phm-macology and Biochemistry, Received

April

E. FISCH,

AND

MANUEL

Cells

J. GLYNIAS2

Case We&-m Reserve University, Cleveland, Ohio 64106

29, 1983, and in revised

form

August

10, 1983

Malic enzyme activity in differentiated 3T3-Ll cells was about 20-fold greater than activity in undifferentiated cells. A new steady-state level was achieved about 8 days after initiating differentiation of confluent cultures with a 2-day exposure to dexamethasone, isobutylmethylxanthine, and insulin. This increase in enzyme activity resulted from an increase in the mass of malic enzyme as detected by immunotitration of enzyme activity with goat antiserum directed against purified rat liver malic enzyme. Malic enzyme synthesis was undetectable in undifferentiated cells and increased to about 0.2% of soluble protein in differentiated cells, suggesting that the increase in enzyme mass was due primarily to an increase in enzyme synthesis. Thyroid hormone, a potent stimulator of malic enzyme activity in hepatocytes in culture and in liver and adipose tissue in intact animals, decreased or increased malic enzyme activity in differentiating 3T3-Ll cells by about 40% when it was removed or added to the medium, respectively. Insulin, another physiologically important regulator of malic enzyme activity in ti~o, had no effect on the initial rate of accumulation of malic enzyme activity in the differentiating cells and caused a 30 to 40% decrease in the final level of enzyme activity in the fully differentiated cells. Cyclic AMP, a potent inhibitor of malic enzyme synthesis in hepatocytes in culture, inhibited this process in 3T3-Ll cells by 30%. Malic enzyme is like several other enzymes in that the large increase in its concentration which accompanies differentiation of 3T3-Ll cells is due to increased synthesis of enzyme protein. However, the hormonal modulation of malic enzyme characteristic of liver and adipose tissue in intact animals does not appear to occur in differentiated 3T3-Ll cells, suggesting that differentiated 3T3-Ll cells may not be an appropriate model system in which to study the hormonal modulation of malic enzyme that occurs in liver and adipose tissue of intact animals.

Fatty acid biosynthesis is one of the major pathways to utilize the NADPH produced by cytoplasmic malic enzyme (EC 1.1.1.40) (1, 2). In liver and adipose tissue of rats and mice, malic enzyme activity correlates positively with the rate of fatty acid biosynthesis. Malic enzyme activity in these tissues is decreased in starved or diabetic animals and elevated in animals fed a high-carbohydrate diet or injected with

insulin (3-5). Therefore, increased malic enzyme activities in liver and adipose tissue are correlated with elevated insulin to glucagon ratios and, conversely, decreased enzyme activities with decreased hormone ratios (6-9). Thyroid hormone is also a major regulator of malic enzyme activity. In hyperthyroid rats, malic enzyme activities are elevated in both liver and adipose tissue; in hypothyroid animals, they are decreased (4,5,10,11). In liver, thyroid hormone regulates the relative synthesis of malic enzyme (12, 13) which, in turn, correlates positively with malic enzyme mRNA level

’ To whom correspondence should be addressed. ‘Supported as a predoctoral trainee by National Institutes of Health Grant GM 07382. 0003-9861/84 Copyright All rights

$3.00

0 1984 by Academic Press. Inc. of reproduction in any form reserved.

54

REGULATION

OF MALIC

(14,15). Thyroid hormone regulates malic enzyme activity in liver by a direct action on hepatocytes, rather than by an indirect effect of thyroid status on other hormones which, in turn, regulate the enzyme (16). Neither modulation of catalytic efficiency nor covalent modification has been reported as a means of regulating malic enzyme activity in avian or mammalian tissues. Regulation of malic enzyme activity in adipose tissue probably occurs by the same pathway which has been established for liver. Hormonal regulation of malic enzyme synthesis can be studied in maintenance cultures of hepatocytes from the embryonic chick (15-1’7). However, these hepatocytes do not grow in culture, restricting their potential for studies of hormone action. In addition, tissue-specific regulation may limit the usefulness of hepatocytes as a valid model for studying regulation of the synthesis of the “lipogenic” enzymes in mature adipocytes. In seeking a permanent cell line in which lipogenesis and its component enzymes were regulated by hormones, we considered the mouse-derived, 3T3-Ll cells. This cell line exhibits fibroblastic morphology when growing. Upon reaching confluence, lipid accumulates and, morphologically, the cells take on the appearance of mature adipocytes (18, 19). Concomitant with the accumulation of lipid, the activities of several lipogenic enzymes increase to levels characteristic of adipose tissue from fed rats and mice (20, 21). In this paper, we report that the large accumulation of malic enzyme which accompanies differentiation of 3T3-Ll cells is correlated with increased synthesis of the enzyme. In differentiated cells, we tested the effects of several agents which profoundly alter malic enzyme activity in hepatocytes and which, based on studies with intact animals, should have had similar effects on the adipose tissue enzyme. However, the effects of thyroid hormone, insulin, and cyclic AMP on malic enzyme activity in the 3T3-Ll cells were very small compared to known effects on hepatocytes in culture and liver and adipose tissue in intact animals.

ENZYME

IN 3T3-Ll

55

CELLS

EXPERIMENTAL

PROCEDURES

Materials. 3T3-Ll cells were obtained from the American Type Culture Collection. ACTH: dexamethasone, dibutyryl cyclic AMP, 8-bromo cyclic AMP, theophyline, 3,3’,5’+triiodothyronine, calcium pantothenate, biotin, penicillin G, streptomycin sulfate, and trypsin were purchased from Sigma. Insulin was purchased from either Sigma or Elanco (Indianapolis, Ind). l-Methyl-3-isobutylxanthine (Aldrich), heat-denatured fetal bovine serum, calf serum, and Waymouth medium MD ‘705/l (Gibco); Dulbecco’s modified Eagle’s medium (M. A. Bioproducts); and [‘Hlleucine (Amersham) were provided by the indicated suppliers. Metaproterenol was obtained from Boehringer Ingelheim Ltd. (Elmsford, N. Y.). All other reagents were purchased from sources indicated previously (22) or were of the highest quality available commercially. Cell culture. Cells were grown in lOO-mm tissue culture plates in 8 ml Dulbecco’s modified Eagle’s medium containing glucose (25 mM) and supplemented with calf serum (lo%), penicillin (60 mg/liter), streptomycin sulfate (100 mg/liter), biotin (8 mg/liter), and calcium pantothenate (8 mg/liter). Incubation was at 37°C in an atmosphere of 10% COz and 90% air. The medium was changed every 2 to 3 days during exponential growth and every 2 days after confluence. The differentiation of confluent cultures was initiated as previously described (23) by changing the medium to one identical to that just described except that fetal bovine serum (10%) was used instead of calf serum and the medium contained insulin (10 pg/ml), dexamethasone (0.25 PM), and methylisobutylxanthine (0.5 mM). After 2 days, insulin, dexamethasone, and methylisobutylxanthine were removed from the medium. The incubations were continued in medium containing fetal bovine serum and the various additions indicated in the figures and tables. Thyroid hormone was removed from calf and fetal bovine serum by treatment with AG-1X-8 ion exchange resin (24). Serum thyroxine was measured with a radioimmunoassay procedure used in the Clinical Biochemistry Laboratory, University Hospitals of Cleveland. Preparation of cell extracts. The monolayers were rinsed two times with 5 ml of sodium phosphate (10 mM)-buffered saline, pH 7.4, scraped into 0.25 M sucrose, 5 mM Tris-HCl, 1 mM dithiotreitol, pH 7.4, 1 ml per plate, and homogenized in a Dounce homogenizer with 13 strokes of a tight-fitting pestle. A cytoplasmic fraction was prepared by centrifuging the extract at 40,OOOg for 30 min. All operations were carried out at 0-4°C. Malic enzyme (4) and isoeitrate dehydrogenase (25) activities and protein (26) were

’ Abbreviations used: ACTH, adrenocorticotropic hormone; SDS, sodium dodecyl sulfate.

GOODRIDGE,

56

FISCH, AND GLYNIAS

measured in the cytoplasmic fraction. DNA (27) was measured in the pellet. Purification of ma& enzyme and preparation of anti-mdic enzyme serum Malic enzyme was purified from the livers of rats which had been fed for 5 days with a high-carbohydrate, low-fat fat diet containing 1% thyroid powder (Nutritional Biochemical). This dietary manipulation causes a very large increase in hepatic malic enzyme activity (4). Purification of the enzyme was achieved by a combination of published methods (23). The purified enzyme migrated as a single protein band on SDS-polyacrylamide gel electrophoresis. A female goat was immunized with the purified enzyme, its blood was collected, and antiserum was prepared from the blood as described elsewhere (29). Measurement of relative qfnthesti of m&c enzyme. After the incubation medium was removed, each plate was washed twice with 5 ml of leucine-free Waymouth medium MD ‘705/l which had been warmed to 37°C. Three milliliters of the same medium containing 150 &i [‘Hlleucine (61 Ci/mmol) was then added to each plate and the plates were incubated as described above for 1 h. Experimental agents initially present in the culture medium were added back to their respective plates during the labeling period. A cytoplasmic extract was prepared from these cells as described above except that centrifugation was at 130,OOOgfor 60 min. Quantification of radioactivity in total soluble protein and immunoprecipitation of malic enzyme in the extracts was carried out as previously described (16). Partially purified mouse liver malic enzyme was used as carrier in the immunoprecipitations. The immunoprecipitates were subjected to SDS-polyacrylamide disc gel electrophoresis (30). The tube gels were sliced into 1.5-mm sections and assayed for radioactivity (16). The position of added malic enzyme was determined by staining the gels with Coomassie blue; only radioactivity which comigrated with the carrier enzyme was considered to be malic enzyme. RESULTS

The activity of malic enzyme increased rapidly after confluent cells were incubated for 2 days with insulin, dexamethasone, and methylisobutylxanthine (Fig. 1A). A new steady-state level was achieved at about 6 days after removing the initiating drugs and was about 20-fold higher than the level in untreated cells. The relative increase in malic enzyme activity and the absolute level of enzyme activity achieved at steady state were comparable to those previously reported for malic enzyme in differentiating 3T3-Ll (31, 32) or 3T3F442A cells (33). The presence of insulin in the medium after the initiation period

resulted in a steady-state level of malic enzyme activity which was about 30% lower than that achieved in the absence of this hormone (Fig. 1A). The activity of another NADP-linked dehydrogenase in the “soluble” compartment of the cell, isocitrate dehydrogenase (EC 1.1.1.42), is not correlated with fatty acid biosynthesis in liver or adipose tissue (5,34,35). Unlike malic enzyme, isocitrate dehydrogenase is not considered a lipogenic enzyme. During differentiation of 3T3-Ll cells the activity of this enzyme increased about three-fold (Fig. 1B). Insulin had no effect on the accumulation of isocitrate dehydrogenase. Protein content of the culture increased about five-fold when the 3T3-Ll cells were differentiated in the presence of insulin (Fig. 1C). The absolute increases in malic enzyme and isocitrate dehydrogenase activities during differentiation were, therefore, ‘75- and l&fold, respectively. If insulin was omitted from the medium after the removal of the initiating drugs, soluble protein content of terminally differentiated cultures was decreased by about 50%. Insulin, therefore, had small but opposite effects on soluble protein and specific activity of malic enzyme, increasing the former and decreasing the latter. The net result was no effect of insulin on the amount of malic enzyme accumulated per culture plate. An increase in the specific activity of an enzyme can be caused by an increase in either the concentration or the catalytic efficiency of that enzyme. The quantity of antiserum required to precipitate a given activity of malic enzyme was the same in extracts prepared from differentiated or undifferentiated cells.4 Thus, the 20-fold increase in malic enzyme activity observed during differentiation was due to a comparable increase in the concentration of malic enzyme protein. An increase in malic enzyme protein could be caused by increased enzyme synthesis or decreased enzyme degradation. We evaluated the role of enzyme synthesis by pulse-labeling all proteins in undiffer4A. G. Goodridge, J. E. Fisch, and T. D. Johnston, unpublished results.

REGULATION

0

2

t

t

4

DAYS

6 AFTER

8

OF

10

12

MALIC

ENZYME

IN

0

14

CONFLUENCE

.t

3T3-Ll

2

t

CELLS

4

DAYS

6 AFTER

57

8

10

12

14

CONFLUENCE

1. Changes in malic enzyme activity (A), isocitrate dehydrogenase (NADP) activity (B) and soluble protein (C) in 3T3-Ll cells differentiated in the presence (solid squares) or absence (solid circles) of insulin after removing the initiating drugs or in cells which were not exposed to the initiating drugs or insulin (open circles). Other incubation conditions were as described under Experimental Procedures. Enzyme activities are expressed as milliunits per milligram soluble protein. Soluble protein is expressed as milligrams per culture plate. Each point is the average of three to eight separate experiments k SE except for those at 2 days (all treatments) and 6 days (with insulin) which were the average of two experiments. The absence of error bars indicates that they were smaller than the symbol. The time during which the initiating drugs were present in the medium is defined by the positions of the arrows. FIG.

entiated or differentiated 3T3-Ll cells for 1 h with [3H]leucine. Electrophoresis of malic enzyme immunoprecipitates prepared from differentiated cells demonstrated the presence of a single major peak of radioactivity which comigrated with carrier mouse enzyme at a molecular weight of 62,000 (Fig. 2), attesting to the specificity of the antiserum. Completeness of the precipitation was determined by performing a second immunoprecipitation. After the first precipitate was removed by centrifugation, carrier mouse malic enzyme, in an amount equivalent to the activity in the first precipitation, was added to the reaction mixture. A volume of antiserum equivalent to that added to the first precipitation reaction also was added. The immunoprecipitate was recovered and

subjected to SDS-polyacrylamide gel electrophoresis. No peaks were observed, indicating that the first precipitation was quantitative (Fig. 2). Synthesis of malic enzyme was undetectable in undifferentiated 3T3-Ll cells incubated for 5 or 13 days after confluence (Fig. 2). The rate of incorporation of labeled amino acids into malic enzyme could be influenced by degradation if the enzyme’s t112were short relative to the labeling time. The approximate half-life for malic enzyme in differentiating 3T3-Ll cells is equivalent to the length of time required to achieve 50% of the difference between the two steady-state levels (36). In the absence of insulin this was achieved at about 5 days after confluence. The same half-life was calculated from results plotted on a “per

58

GOODRIDGE,

FISCH,

M, x 10-3

FIG. 2. SDS-polyacrylamide gel electrophoresis of malic enzyme immunoprecipitates from differentiated (A, upper tracing) and undifferentiated (B) 3T3-Ll cells incubated for 5 days after confluence. Differentiation was initiated as described under Experimental Procedures. There was no insulin in the medium after removal of the initiating drugs. The position of carrier mouse malic enzyme was located by staining the gel with Coomassie blue prior to slicing. The molecular weight scale was determined by electrophoresis of ‘%-methylated marker proteins on a separate tube gel. The marker proteins were phosphorylase b (9’7,000), bovine serum albumin (68,000), and ovalbumin (46,000). The immunoprecipitate in the lower tracing of (A) was precipitated from the same extract as that in the upper tracing but after the precipitate in the upper tracing had been removed. The immunoprecipitation reactions for the precipitates contained 1.7 X lo6 dpm (A) and 1.4 X lo6 dpm (B) in total soluble protein. The peak of radioactivity which migrated with the tracking dye was variable in quantity and probably represents nonspecific absorption of small peptides which were not fractionated in this gel system.

plate” basis. If it is assumed that accumulation did not begin until the initiating agents were removed, as suggested by the results in Fig. 1 (A and C), then the halftime for malic enzyme was about 3 days. Although the degradation rate may have been even slower (i.e., if accumulation started on Day 0), the l-h labeling period we used was too short to be significantly affected by enzyme degradation. A similar degradation rate constant can be calculated

AND

GLYNIAS

for the cultures incubated with insulin. In sum, these results indicate that the increase in malic enzyme protein was due to an increased rate of enzyme synthesis. The relative synthesis of malic enzyme in differentiated 3T3-F442A cells was reported to be 2% of soluble protein (37). In our experiments, synthesis rates for malic enzyme were 0.12 + 0.01 and 0.20 f 0.04% (means f SE, n = 5 and 4, respectively) of soluble protein at 5 and 13 days after the initiation of differentiation, respectively. The 3T3-F442A cells are different from, but closely related to, 3T3-Ll cells (38). In addition, differentiation of the 3T3-F442A cells was initiated in a manner significantly different from that used in our experiments. The difference in relative synthesis rates could have been due to the use of different cell types and culture conditions. However, the specific activity of malic enzyme in differentiated 3T3-F442A cells (39, 40) was nearly identical to that in our differentiated 3T3-Ll cells. Synthesis rates of malic enzyme in the 3T3-F442A cells were based on rates of incorporation of labeled methionine into a band on an SDSpolyacrylamide gel. The putative malic enzyme band had a proteolytic peptide map which was similar but not identical to authentic rat malic enzyme. Based on our immunoprecipitation experiments, we suggest that the “65,000-Da band” in differentiated 3T3-F442A cells may have been a mixture of malic enzyme and some other peptide( with the latter peptide constituting 90% of the radioactivity. We tested the effect of thyroid hormone on malic enzyme activity in the 3T3-Ll cells by incubating the cells in medium containing serum which had been depleted of thyroid hormone (24). Thyroxine was undetectable in depleted serum and was 13 pg/lOO ml in fetal bovine serum. In bovine serum, about 0.05% of total thyroxine is free, i.e., not bound to specific proteins in the serum (41). The absolute concentration of the unbound part of a tightly bound ligand remains constant when serum is diluted. Thus, the depleted medium contained no free thyroxine and the medium with normal serum about 6 ng/lOO ml. The concentration of the free hormone determines

REGULATION

OF

MALIC

biological activity and this free thyroxine concentration is twice that of normal mouse serum (41). The same considerations would apply to free triiodothyronine in the media with normal and depleted sera (24). Differentiation of the cells in the absence of thyroid hormone caused a small, but statistically significant, decrease in the accumulation of malic enzyme which was not statistically significant until 8 to 12 days after initiating differentiation (Table I). Addition of triiodothyronine to the cultures incubated with depleted sera restored the normal accumulation of malic enzyme. The effects of thyroid hormone on isocitrate dehydrogenase activity were qualitatively similar to those on malic enzyme. The presence of thyroid hormone in the medium had no effect on soluble protein content of the cultures (data not shown). The effects of thyroid hormone were the same even when the cells were grown for 7 days prior to confluence in media containing thyroid hormone-depleted calf serum (data not shown). The activities of malic enzyme and other lipogenic enzymes in both liver and adipose tissue are reduced dramatically by, starvation (4, 5, 34). Plasma glucagon; acting via changes in intracellular cyclic AMP, is

TABLE

I

THE EFFECT OF THYROID HORMONE ON MALIC ENZYME ACTIVITY IN DIFFERENTIATING 3T3-Ll CELLSa Days Culture

conditions

Depleted serum Depleted serum plus triiodothyronine Number of experiments

after

confluence

4

8 to 12

80 f 15

59 zk 9

118 f 26 4

99 f 8 6

a Fetal calf serum was depleted of thyroid hormone by treatment with an ion exchange resin (24). The triiodothyronine concentration was 1 fig/ml. The results are expressed as a percentage of the activity in cells incubated in untreated serum without added triiodothyronine k SE. The actual activity of malic enzyme can be ascertained from Fig. 1A. Insulin (10 rg/ml) was present in all of these experiments.

ENZYME

IN

3T3-Ll

CELLS

59

considered an important mediator of the effect of starvation. We therefore examined the effects of analogs of cyclic AMP on the activities of malic enzyme and isocitrate dehydrogenase in differentiating 3T3-Ll cells. If dibutyryl cyclic AMP was added to the cells on the fourth day after initiation of differentiation, malic enzyme continued to accumulate over the next 2 days but at only 65% of the rate of accumulation in the control cells (Fig. 3A; P < 0.01, n = 11). Neither isocitrate dehydrogenase activity nor soluble protein per plate was significantly affected by dibutyryl cyclic AMP added on Day 4 (Figs. 3B and C). If dibutyryl cyclic AMP was added to cells on Day 12, malic enzyme not only stopped accumulating, it decreased dramatically. By Day 14, malic enzyme activity in the cyclic AMP-treated cells was 46% of that in the control cells (Fig. 3A). This decrease in activity was due to a decrease in immunoreactive malic enzyme.4 In cells treated with cyclic AMP on Day 12 both isocitrate dehydrogenase activity and soluble protein per plate were inhibited to about the same extent as malic enzyme (Figs. 3B and C). In the experiments described above, insulin was removed from the medium after the 2-day initiation period. When insulin was present throughout the experiment, the inhibition of malic enzyme activity caused by cyclic AMP was smaller (data not shown). Thus, lack of insulin was not responsible for the inability of cyclic AMP to decrease malic enzyme activity. This apparent difference in the effect of cyclic AMP on accumulation suggested a differential effect on malic enzyme synthesis. However, addition of this agent to cells for 17 h had the same 30% inhibitory effect on malic enzyme synthesis on both Day 4 and Day 12 (Table II). One interpretation of this result is that cyclic AMP stimulated degradation of malic enzyme at Day 12 but not (or less so) at Day 4. The decrease in soluble protein caused by cyclic AMP at Day 12 suggested a nonselective increase in degradation at this stage of differentiation. Loss of enzyme was probably not due to differential loss of cells rich in malic enzyme because total DNA on the

60

GOODRIDGE,

3

160.

FISCH,

AND

GLYNIAS

160-

160-

T

60-

456

12 1314

:

121314

456 DAYS

I

I j’ 2: :/“1 ‘k-3

AFTER

CONFLUENCE

FIG. 3. The effect of dibutyryl cyclic AMP on malic enzyme activity (A), isocitrate dehydrogenase (NADP) activity (B), and soluble protein (C). Control cells were differentiated as described under Experimental Procedures. Insulin was omitted from the incubation medium after removal of the initiating drugs. The cells were incubated for 4 or 12 days in the control medium and then shifted to the same medium with (open circles) or without (closed circles) dibutyryl cyclic AMP (0.5 mM). The results are expressed as a percentage of the activity in cells incubated for 8 days after confluence in the control medium. Each point represents the average f SE of three (Days 5 and 6) or four (Days 13 and 14) separate experiments. The actual enzyme activities and protein levels can be calculated from the results shown in Fig. 1.

plates was not decreased by cyclic AMP (Table II and other data not shown). The apparent differential effect of cyclic AMP on malic enzyme specific activity may have a simple kinetic explanation. If cyclic AMP decreased the relative synthesis of malic enzyme by 30’S, then in the absence of a selective effect of the drug on degradation, the new steady state for enzyme activity should be 30% lower in cyclic AMPtreated cells than in the control cells. If the inhibitor is added when enzyme activity is less than the predicted steady-state level, then accumulation should continue, albeit at a slower rate, until that steady-state level is achieved. If the inhibitor is added when enzyme activity is already higher than the predicted steady-state level, then activity should fall until that level is achieved. Within the limits of error of experiments of this type, the results in Table II and Fig. 3A are consistent with this explanation. The concentration of dibutyryl cyclic AMP in these experiments was 0.5 mM; 0.1

dibutyryl cyclic AMP had no significant effect on any of these variables (data not shown). Butyrate at a concentration of 1 mM had no effect. A different analog of cyclic AMP, Sbromo cyclic AMP, produced the same effects as the dibutyryl derivative. ACTH and /3-adrenergic agonists activate adipocyte adenylate cyclase and raise intracellular CAMP levels (42,43). We therefore tested the effects of ACTH (2 PM) and the relatively stable P-agonist metaproterenol(4 PM), on malic enzyme activity in 3T3-Ll cells which had been differentiated for 12 days. Neither agent had any effect when added alone. Both agents inhibited malic enzyme activity by 20 to 30% (data not shown) when they were added in the presence of theophylline (0.5 mM), a phosphodiesterase inhibitor. Similar effects, barely distinguishable from the controls, were observed for isocitrate dehydrogenase and soluble protein. These results are consistent with the relatively high concentrations of dibutyryl cyclic AMP required to inhibit malic enzyme activity. mM

REGULATION TABLE

OF

MALIC

II

THE EFFECT OF ANALOGS OF CYCLIC AMP THE SYNTHESIS OF MALIC ENZYME ANDSOLUBLE PROTEIN'

ON

Treatment for 16 hr beginning on Measurement

Day

Malic enzyme activity* Malic enzyme synthesis Soluble protein per plate Soluble protein synthesis DNA per plate No. of experiments

4

69f 5 70f 9 99f 6 124 + 18 108 f 18 5

Day

12

81f 9 692 5 62 + 16 130 + 25 126 f 14 3

a The results are expressed as a percentage of control incubated without cyclic AMP. There was no insulin present after removal of the initiating agents on Day 2. The actual level of malic enzyme activity and soluble protein per plate on Days 4,5,12, and 13 can be ascertained from Figs. 1A and 1C. Relative synthesis of malic enzyme was 0.12 f 0.01 (n = 5) and 0.20 f 0.04 (n = 4) dpm/lOO dpm in soluble protein on Days 5 and 13, respectively. Soluble protein synthesis was 6.7 + 0.6 (n = 5) and 6.9 f 1.1 (n = 4) X lo6 dpm/mg protein on Days 5 and 13, respectively. The DNA content was 94 f 1’7 (n = 4) and 47 k 8 (n = 6) pg per plate on Days 5 and 13, respectively. All data are expressed as means f SE. Since butyrate (1 mM) had no effect on these variables, the data from butyrate-treated cultures were averaged with the controls to obtain these results. Both dibutyryl cyclic AMP (0.5 mM) and 8-bromo cyclic AMP (0.5 mM) had similar effects and therefore the data from these two treatments were pooled to yield the averages recorded in this Table. *These changes in activity are less than those recorded in Fig. 3 because the cells were incubated with cyclic AMP for only 16 h.

DISCUSSION

The 20-fold increase in malic enzyme activity which occurred during the differentiation of 3T3-Ll cells was due to a comparable increase in enzyme protein concentration. The increase in protein concentration was due primarily, perhaps exclusively, to increased synthesis of the enzyme. In this respect, malic enzyme is similar to several enzymes, the activities of which increase several-fold during the conversion of preadipocytes to adipocytes. Differentiation-induced increases in the

ENZYME

IN

3T3-Ll

CELLS

61

activities of fatty acid synthase (44, 45), a-glycerophosphate dehydrogenase (46), pyruvate carboxylase (47, 48), and glutamine synthetase (49) correlate with increases in enzyme concentration and rate of enzyme synthesis. Cyclic AMP is the intracellular messenger which mediates the almost complete inhibition of malic enzyme synthesis caused by glucagon in hepatocytes (16). In rodents, the level of the adipose tissue enzyme changes in parallel with that of the liver enzyme under a variety of nutritional and hormonal conditions (1, 2, 4). Cyclic AMP and agents which modulate adipocyte cyclic AMP, e.g., ACTH and B-adrenergic agonists, were, therefore, expected to profoundly inhibit the activity and rate of synthesis of malic enzyme in differentiated 3T3-Ll cells. Cyclic AMP did inhibit malic enzyme activity in differentiated 3T3-Ll cells, and the decrease in activity was due, primarily, to a decrease in enzyme synthesis. The inhibition of enzyme synthesis was, however, very small compared to that which occurred in hepatocytes in culture (16). This lack of sensitivity to cyclic AMP was not due to a lack of cyclic AMP-dependent protein kinase because this protein is active in differentiated 3T3-Ll cells (31). Furthermore, the synthesis of fatty acid synthase was markedly inhibited by cyclic AMP and extracellular agents which elevate the intracellular cyclic AMP concentration (44). Synthesis of glutamine synthetase also was inhibited by cyclic AMP (49). Thus, the intracellular pathway(s) by which cyclic AMP regulates the synthesis of some specific proteins was intact in differentiated 3T3-Ll cells. Synthesis rates for fatty acid synthase and malic enzyme were coordinately regulated in hepatocytes in culture during treatment with glucagon (16, 50) and in liver during starvation (2,12,50). Despite coordinate regulation during differentiation (Fig. 2 and Refs. (44,45)), the synthesis rates of these two enzymes were regulated independently by cyclic AMP in differentiated 3T3-Ll cells. The low sensitivity of malic enzyme synthesis to cyclic AMP may reflect a significant difference in the mechanism by which cyclic AMP regulates the

62

GOODRIDGE,

FISCH,

synthesis of “lipogenic” enzymes in liver as opposed to adipose tissue. Alternatively, a component(s) in the pathway by which cyclic AMP regulates malic enzyme synthesis in adipose tissue may be deficient in differentiated 3T3-Ll cells. The activity of malic enzyme in liver and adipose tissue of rodents is elevated in animals fed high-carbohydrate diets, decreased in animals which are diabetic, and increased in diabetic animals which are injected with insulin (1,4). In addition, insulin is known to stimulate malic enzyme activity in hepatocytes maintained in culture (16, 51, 52). In differentiating 3T3-Ll cells, however, insulin had either no effect or a small negative effect. This lack of response to insulin was not due to a lack of functional insulin receptors on differentiated 3T3-Ll cells. These cells have fiveto eight-fold more insulin receptors per cell than mature rat fat cells (23,53). Secretion of lipoprotein lipase (54), the activity of lipoprotein lipase (54), and uptake of glucose (55) were stimulated markedly by insulin in differentiated 3T3-Ll cells. Despite an abundance of insulin receptors and sensitivity of other processes to insulin, accumulation of neither malic enzyme (Fig. 1) nor fatty acid synthase (44,56) required addition of exogenous insulin. Thyroid hormone caused a 40-fold increase in malic enzyme activity in the liver of thyroidectomized rats (57) and a lOOfold increase in avian hepatocytes in culture (16). In rat adipose tissue, thyroid hormone caused at least a fivefold increase in malic enzyme activity (4). The small changes in malic enzyme activity caused by thyroid hormone in differentiated 3T3Ll cells contrast markedly with the enormous effects observed in viva and in hepatocytes in culture. The point in the pathway at which 3T3-Ll cells are deficient in their response to thyroid hormone is unknown. To our knowledge, there are no reports on thyroid hormone receptors in these cells. The activity of malic enzyme in differentiated cells (Fig. 1; Refs. (31, 32)) was comparable to that in adipose tissue of rats fed normal chow diets (4). High-carbohydrate, low-fat diets, starvation-reali-

AND

GLYNIAS

mentation regimens, and treatment with thyroid hormone stimulate accumulation of much higher levels of enzyme (4). Thus, the lack of stimulation by insulin and thyroid hormone was not due to “maximal” rates of expression in the differentiated 3T3-Ll cells. The relatively small effects of insulin, thyroid hormone, and cyclic AMP on the activity and/or synthesis of malic enzyme in differentiated 3T3-Ll cells indicate that hormonal modulation of malic enzyme in hepatocytes is significantly different from that observed in the 3T3-Ll cells. Our results also raise the possibility that differentiated 3T3-Ll cells may not be an appropriate model for the hormonal modulation of malic enzyme activity in adipose tissue. ACKNOWLEDGMENTS This work was supported by a grant from the Kroc Foundation. We are grateful to Timothy Uyeki for technical assistance with some of these experiments. We also are indebted to Dr. L. T. Webster, Jr. and Dr. J. H. Nilson for reading and criticizing this manuscript.

REFERENCES 1. FRENKEL, R. (1975) Curr. Top. Cell. Regal 9,157181. 2. VOLPE, J. J., AND VAGELOS, P. R. (1976) Physiol Rev. 66,339-417. 3. TEPPERMAN, H. M., AND TEPPERMAN, J. (1964) Amer. J. Physiol. 206, 357-361. 4. WISE, E. M., AND BALL, E. G. (1964) Proc. Natl Acad Sci USA 62, 1255-1263. 5. PANDE, S. V., KAHN, R. P., AND VENKITASUBRAMANIAN, T. A. (1964) Biochim. Biophys. Acta 84.239-250. 6. GOETZ, F. C., MANEY, J. W., AND GREENBERG, B. Z. (1967) J. Lab. Clin. Med 69, 537-557. 7. MALAISSE, W. J., MALAISSE-LAGAE, F., AND WRIGHT, P. H. (1967) Amer. J. Physiol 213, 343-348. 8. BLACKARD, W. G., NELSON, N. C., AND ANDREW% S. S. (1974) Diabetes 23, 199-202. 9. OHNEDA, A., AGUILAR-PARADA, E., EISENTRAUT, A. M., AND UNGER, R. M. (1969) Diabetes 18, l10. 10. TARENTINO, A. L., RICHERT, D. A., AND WESTERFELD, W. W. (1966) Biochim Biophys. Acta 124, 295-309.

REGULATION

OF

MALIC

11. DIAMANT, S., GORIN, E., AND SHAFRIR, E. (1972) Eur. J. B&hem. 26, 553-559. 12. SILPANANTA, P., AND GOODRIDGE, A. G. (1971) .J. BioL Chem. 250, 4134-4138. 13. LI, J. J., Ross, C. R., TEPPERMAN, H. M., AND TEPPERMAN, J. (1975) J. BioL C&m. 250, 141-148. 14. TOWLE, H. C., MARIASH, C. N., AND OPPENHEIMER, J. H. (1980) Biochemistry 19, 579-585. 15. WINBERRY, L. K., MORRIS, S. M., JR., FISCH, J. E., GLYNIAS, M. J., JENIK, R. A., AND GOODRIDGE, A. G. (1983) J. Biol Chem. 258, 1337-1342. 16. GOODRIDGE, A. G., AND ADELMAN, T. G. (1976) J. BioL Chem. 251, 3027-3032. 17. GOODRIDGE, A. G. (1978) Mol. Cell. Endocr. 11,1929. 18. GREEN, H., AND MEUTH, M. (1974) Cell 3,127-133. 19. GREEN, H., AND KEHINDE, 0. (1975) Cell 5,19-27. 20. MACKALL, J. C., STUDENT, A. K., POLAKIS, S. E., AND LANE, M. D. (1976) J. BioL Chem 2.51,64626464. 21. MACKALL, J. C., AND LANE, M. D. (1977) Biochem. Biophys. Res. Commun 79, 720-725. 22. SIDDIQUI, U. A., GOLDFLAM, T., AND GOODRIDGE, A. G. (1981) J. BioL Chem 256, 4544-4550. 23. RUBIN, C. S., HIRSCH, A., FUNG, C., AND ROSEN, 0. M. (1978) J. BioL Chem. 253, 7570-7578. 24. SAMUELS, H. H., STANLEY, F., AND CASANOVA, J. (1979) Endocrinology 105, 80-85. 25. ABRAHAM, S., MIGLIORINI, R. H., BORTZ, W., AND CHAIKOFF, I. L. (1962) B&him. Biophgs. Acta 62,27-34. 26. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. BioL Chem. 193, 2t35-37.5 37 BURTON, K. (1968) in Methods in Enzymology (Grossman, L., and Moldave, K., eds.), Vol. 12, pp. 163-169, Academic Press, New York. 28. CHANG, J., AND CHANG, G. (1982) Anal. B&hem. 121, 366-369. 29. CLAUSEN, J. (1971) Immunochemical Techniques for the Identification and Estimation of Macromolecules, American Elsevier, New York. 30. LAEMMLI, U. K. (1970) Nature (London) 227,680685. 31. LIU, A. Y.-C. (1982) .I. BioL Chem 257, 298-306. 32. KURI-HARCUCH, W., AND GREEN, H. (1977) J. BioL Chem 252, 2158-2160. 33. KASTURI, R., AND JOSHI, V. C. (1982) .I BioL Chem, 257, 12224-12230. --I

-..

_.“.

ENZYME

IN

3T3-Ll

CELLS

63

34. YOUNG, J. W., SHRAGO, E., AND LARDY, H. A. (1964) Biochemistry 3, 1687-1692. 35. LEVEILLE, G. A., AND HANSON, R. W. (1965) J. Lipid Res. 7, 46-55. 36. BERLIN, C. M., AND SCHIMKE, R. T. (1965) MoL PharmacoL 1, 149-156. 37. SPIEGELMAN, B. M., AND GREEN, H. (1981) Cell 24, 503-510. 38. GREEN, H., AND KEHINDE, 0. (1976) Cell 7, 105113. 39. KURI-HARCUCH, W., AND GREEN, H. (1978) Proc. NatL Acad Sci. USA 75, 6107-6109. 40. KURI-HARCUCH, W., WISE, L. S., AND GREEN, H. (1978) Cell 14, 53-59. 41. REFETOFF, S., ROBIN, N. I., AND FANG, V. S. (1970) Endocrinology 86, 793-805. 42. BUTCHER, R. W., Ho, R. J., MENG, H. C., AND SUTHERLAND, E. W. (1965) J. BioL Chem. 240, 4515-4523. 43. RODBELL, M. (1967) J. BioL Chem. 242,5751-5756. 44. WEISS, G. H., ROSEN, 0. M., AND RUBIN, C. S. (1980) J. BioL Chem. 255.4751-4757. 45. STUDENT, A. K., Hsu, R. Y., AND LANE, M. D. (1980) J. BioL Chem 255, 4745-4750. 46. SPIEGELMAN, B. M., AND GREEN, H. (1980) J. BioL Chem. 255,8811-8818. 47. FREYTAG, S. O., AND UTTER, M. F. (1983) J. BioL Chem 258, 6307-6312. 48. ANGUS, C. W., LANE, M. D., ROSENFELD, P. J., AND KELLY, T. J. (1981) Biochem. Biophys. Res. Commun. 103, 1216-1222. 49. MILLER, R. E., AND CARRINO, D. A. (1980) J BioL Chem 255, 5490-5500. 50. FISCHER, P. W. F., AND GOODRIDGE, A. G. (1978) Arch. Biochem. Biophys. 190, 332-344. 51. WILSON, E. J., AND MCMURRAY, W. C. (1981) J. BioL Chem. 256, 11657-11662. 52. MARIASH, C. N., MCSWIGAN, C. R., TOWLE, H. C., SCHWARTZ, H. L., AND OPPENHEIMER, J. H. (1981) J. Clin Invest. 68, 1485-1490. 53. REED, B. C., KAUFMAN, S. H., MACKALL, J. C., STUDENT, A. K., AND LANE, M. D. (1977) PWC. NatL Acad Sci. USA 74, 4876-4880. 54. SPOONER, P. M., CHERNICK, S. S., GARRISON, M. M., AND Scow, R. 0. (1979) J. BioL Chem. 254, 10021-10029. 55. ROSEN, 0. M., SMITH, C. J., FUNG, C., AND RUBIN, C. S. (1978) J. BioL Chem. 253. 7579-7583. 56. MARIASH, C. N., KAISER, F. E., AND OPPENHEIMER, J. H. (1980) Endocrinology 106, 22-27.