Eur. J. Biochem. 101,439-445 (1979)

Stearoyl-CoA Desaturase : A Control Enzyme in Hepatic Lipogenesis Roger JEFFCOAT, Peter A. ROBERTS, Jane ORMESHER, and Anthony T. JAMES Basic Studies Unit, Biosciences Division, Unilever Research, Colworth House, Sharnbrook (Received May 7/August 31, 1979)

1. Hepatocytes were isolated by perfusion of the liver with collagenase/salt solutions and incubated in culture after attachment to plastic culture dishes for periods up to 48 h. 2. The cells, when incubated in serum-free culture medium in the presence of insulin, showed enhanced stearoyl-CoA desaturase activity which was not observed when 50 pM cycloheximide was included. When insulin was omitted from the medium, the cells lost 80 % of their original desaturase activity. 3. Cells isolated from animals fed 20% (w/w) sucrose for two weeks prior to sacrifice, showed high levels of fatty acid synthesis, stearoyl-CoA desaturase activity and triacylglycerol synthesis when compared with cells isolated from animals fed a corn oil supplemented diet. 4. The observations are discussed in terms of the influence of stearoyl-CoA desaturase activity on hepatic lipogenesis.

It is now becoming increasingly apparent that hepatic lipogenesis is controlled by (a) dietary carbohydrate which stimulates enzyme synthesis [l, 21, (b) dietary saturated fat which inhibits the activities of acetyl-CoA carboxylase [l] and fatty acid synthetase [3] and (c) dietary polyunsaturated fat which controls the amount of fatty acid synthetase [4] and probably the amount of other lipogenic enzymes as well. In an earlier report, Jeffcoat and James [5] showed that polyunsaturated fatty acids in the form of corn oil supplemented diets were more effective than hydrogenated tallow supplemented diets containing predominantly saturated fat in repressing the activity of hepatic stearoyl-CoA desaturase. More recent studies by Jeffcoat and James [6] have further shown that in young animals (4- 6-week-old rats), stearoylCoA desaturase responds three to four times as quickly to changes in the linoleic acid content of the diet as does fatty acid synthetase. It was, therefore, suggested that as a control enzyme in the context of the diurnal feeding pattern of the rat, stearoyl-CoA desaturase was more effective than fatty acid synthetase. However, these studies only compared stearoyl-CoA desaturase and fatty acid synthetase by analysis of enzyme activity in vitro. By studying the incorporation of [2-I4C]acetate into saturated and monounsaturated fatty acids, we have been able to extend these studies to take account of acetyl-CoA

carboxylase, palmitoyl-CoAelongase and the NADPHgenerating enzymes. Furthermore, it has been possible to determine directly the rate-limiting step in the conversion of acetate into monounsaturated fatty acid within the liver cell and to investigate those factors which influence the induction of the desaturase enzyme protein. MATERlALS AND METHODS

Animals and Diets Male rats of the Colworth-Wistar strain, weighing approximately 200 g were used throughout the experiments described. These animals had been maintained from weaning on a balanced diet. Occasionally, rats were starved for 24 h and refed a carbohydrate rich/ low fat diet for 17 h. This diet consisted of the following components on a weight for weight basis: starch 10.7%, sucrose 54%, cellulose powder 6%, casein 25%, salts 4 % and vitamins 0.3%, resulting in an energy content of approximately 15 MJ/g. Isolation of Hepatocytes

Rats were killed by cervical fracture and the liver perfused via the portal vein with calcium-free Hanks [7] solution for 5-10 min. The livers were then per-

440

fused at a flow rate of 30-35 ml/min with buffer containing collagenase (Worthington CLS 11) at a concentration of 0.25 mg/ml and the cells isolated by the method of Bonney et al. [8] as described by Geelen and Gibson [9]. The suspension of cells was filtered through 250 pm and 99 pm nylon mesh, washed twice with calcium-free Hanks solution and once with Ham's F-12 culture medium [lo]. Approximately lo8 cells per liver were obtained and these were plated out on 60 x 15 mm Falconised plastic tissue culture dishes at a cell density of 4-5 x lo6 cells in a total volume of 4 ml Ham's F-12 medium supplemented with 15 % (v/v) foetal calf serum and incubated at 37 "C for 4 h. (Goldfarb et al. [ l l ] have shown that after 4-5 h in culture, hepatocytes show a marked improvement both in morphology and in their ability to synthesise lipids.) After this time the medium containing unattached cells was poured off and replaced by serum-free Ham's F-12 medium. The plating efficiency which was usually about 50% and agreed well with the values of other workers [8], was determined by dissolving the attached cells in 1.5 ml 0.5 M sodium hydroxide at 37 "C for 30 min. 2 0 0 4 aliquots were then assayed for protein by the method of Lowry et al. [12]. The number of cells was determined from a calibration curve (Fig. 1) constructed from the protein determination of cell suspensions, the cell densities of which were determined in the Coulter Counter (model Z B ) . Except where mentioned in the text, all Ham's culture media were supplemented with insulin (0.5 pg/ ml, 11.9 m units/ml) and were buffered at pH 7.2 with the addition of 12.5 mM [2-(N-Morpholino)ethanesulphonic acid] and 12.5mM N-{ [tris(hydroxymethyl)methyl]-2-aminosulphonic acid}. All solutions were protected by the addition of penicillin (100 units/ml) and streptomycin (100 pg/ml) and equilibrated before use with 95 % O2/5 % COZ. Enzyme Assays

Total fatty acid synthesis and sterol synthesis were carried out by incubating the cells with 4 ml Ham's F-12 medium supplemented with 2 pCi [2-14C]acetate for various times at 37'C in an atmosphere of 95% 0 2 5 % CO2 as described by Alberts et al. [13j. The distribution of radioactivity between saturated and monounsaturated fatty acids was determined in a similar way except that after saponification and acidification, fatty acids were methylated and resolved by chromatography on Kieselgel H F impregnated with 10 % (w/w) silver nitrate and developed in the solvent system 8 % (v/v) diethyl ether in 40 - 60 "C light petroleum as described by Jeffcoat et al. [14]. Triacylglycerol synthesis and secretion were analysed by incubating attached cells with 4 ml Ham's F-12 culture medium supplemented with insulin containing either 5 pCi [2-l4C]acetate or 0.5 pCi (l-I4C)-

Stearoyl-CoA Desaturase

il

76-

0 0

1

2 3 4 106.Nurnber of cells

*5

6

Fig. 1. Calibration curve relating amount of protein to number of hepatocytev

labelled fatty acid and 250 pg bovine serum albumin per ml. The specific activity of the acetate was 5560 Ci/mol. The radioactive fatty acids were diluted with 100 nmol non-radioactive fatty acid to 4- 5 Ci/ mol in the final incubations. After the incubation times stated in the text, the medium was removed and centrifuged to sediment any unattached cells and then extracted overnight with 3.75 ml chloroform/ methanol (1 : 2) for every ml of medium. 400 pl rat serum were added to provide carrier lipid. The cells were removed from the culture dishes with a rubber policeman and washed with 4 ml 0.9 % (w/v) sodium chloride in 10 mM sodium phosphate pH 7.4. The cells were then centrifuged and resuspended in the same salt solution before extraction of the lipids as described above. To the chloroform/methanol extracts were added 1.25 ml chloroform and 1.25 ml 2 M potassium chloride in 0.5 M potassium phosphate pH 7.4 for every ml of initial extract. The chloroform layer was evaporated to dryness under a stream of nitrogen, redissolved in a small volume of ether and spotted on a 0.25-mm plate of Kieselgel G. The plate was developed in the solvent system light petroleum (40-60 "C)/diethyl ether/ammonia (85 : 15:2) and the lipid fractions visualised by spraying with 0.01 % (w/v) phloxin in methanol and viewing in ultraviolet light or by scanning the plate for radioactivity using a Panax Thin-Layer ScannerRTLS-I. All experiments were carried out at least twice and duplicate or triplicate determinations were made. The results show the data from a single typical experiment. [2-14C]Acetate (sodium salt), 55 - 60 Ci/mol and [l-14C]stearicacid, 55 - 60 Ci/mol were obtained from the Radiochemical Centre (Amersham, Bucks, England).

R. Jeffcoat, P. A. Roberts, J. Ormesher, and A . T. James

Ham’s F-12 culture medium and Falconised plastic petri dishes were obtained from Gibco-Bio-Cult (88 Staines Road, Hounslow, Middlesex, England) and Scientific Supplies (Scientific House, Vine Hill, London EC3 R 5EB), respectively. Nylon mesh was purchased from Henry Simon (Special Products Division, Cheadle Heath, Stockport, Cheshire, England). Insulin and antibiotics, 2-(N-morpholino)ethanesulphonic acid (MES) and N- { [tris(hydroxymethyl)methyl]-2-aminoethanesulphonic acid} (TES), were obtained from Sigma (Poole, Dorset, BH17 7NH, England).

441 600

-.al c

c

eQ 450

-E f

g U

3

300

U 3

g

.-m U

U

.-u

150

RESULTS AND DISCUSSION Induction of Stearoyl-CoA Desaturase It is well established that levels of hepatic stearoylCoA desaturase are increased in response to diets of high carbohydrate content. The mechanism by which this occurs is not yet known and it seemed likely that isolated hepatocytes could provide as useful system for studying this inductive process. Hepatocytes were isolated from rats which had been maintained from weaning for four weeks on a balanced laboratory diet, known as spital [ 5 ] . This diet is a high carbohydrate/low fat diet, but with a high proportion of the fatty acid as linoleic acid (34 % 18 : 2). This results in low levels of hepatic fatty acid synthetase and stearoyl-CoA desaturase activity [5]. Hepatocytes were plated out in triplicate on plastic culture dishes as before and incubated for 48 h in the presence of Ham’s F-12 medium containing 10 mM glucose. Enzyme activity was measured at zero time, i.e. after the initial 4-h incubation period, during which the cells were allowed to adhere to the culture dishes. During this period, cells which had been isolated in insulin-free buffers, were incubated in the presence or absence of 0.5 pg insulin/ml. The results shown in Fig. 2A, B clearly indicate that the presence or absence of hormone has no effect on the enzyme activity at this stage. Cells which were incubated for 48 h were replenished with fresh medium after 18 h and 24 h and were incubated without insulin, with insulin orwith insulin and 50pM cycloheximide (Fig.2C-E). The results shown in Fig.2 indicate three main findings. (a) Insulin is required to maintain the activity of stearoyl-CoA desaturase, since in its absence the enzyme activity decreases by approximately 80 %, after 48 h in culture (cf. Fig. 2A, C). (b) In the presence of insulin, there is fourfold increase in enzyme activity (cf. Fig.2B,D). (c) In the presence of insulin and cycloheximide the activity of the enzyme is maintained with no apparent net synthesis or degradation. These results suggest that insulin is required for the stabilisation of desaturase protein although it

A

0

C

Fig. 2. The ejfect o j ’ insuliir and cyclohexirnide on the induction of stearoyl-CoA desaturase. Hepatocytes were incubated with 4 ml Ham’s F-12 culture medium and assayed for (‘4C)-labelled monounsaturated fatty acids as in Fig.4 and 5 at zero time (A, B) and after 48 h (C, D, E). Sample A and C were plated in the absence of insulin and B, D, E in the presence of 0.5 pg insulin/ml. 50 pM cycloheximide was added to E

should be remembered that the stabilisation of ‘enzyme’ proteins by cycloheximide (i.e. prevention of degradation) has been reported [15]. However, Salmon [16] has reported that stearoyl-CoA desaturase activity was halved during a 3-h mouse liver perfusion if insulin was left out of the perfusing medium. This provides good evidence for the stabilisation of the enzyme activity (and may be enzyme protein). Furthermore, studies in vivo by Oshino and Sat0 [17] have shown that cycloheximide reduces the activity of stearoyl-CoA desaturase by 50 % in 3 -4 h. The addition of 15% (v/v) serum derived from either spitdl-fed rats or rats which had been starved and refed the high carbohydrate diet (see Table 1 for lipid composition), resulted in a marked change in cellular morphology (Fig. 3) from the rounded cell to a more flattened cell showing a very characteristic spreading and a complete lack of induction of stearoylCoA desaturase. These results are very similar to those observed by Alberts et al. [13] who showed that only after the removal of 10% calf serum was there a measurable increase in fatty acid synthetase. It would therefore appear that insulin is required not only for the synthesis of stearoyl-CoA desaturase enzyme (cycloheximide-sensitive induction) but also for protecting the enzyme against loss of activity or protein degradation. However, although Alberts et al. [ 131 have shown that insulin stimulates the synthesis of fatty acid synthetase in cultured cells when incubated in a serum-free medium containing 25 pM glucose, Volpe and Vagelos [ l ] have reported data which put a different emphasis on the hormonal effect.

442

Stearoyl-CoA Desaturase

Table 1 TheJatty acid coniposltron of cerum from spital and carbohydratefed rats Diet

Fatty acid composition ~~~~

~

12 0

14 0

14 1

~~~

16 0

16 1

~

16 2

18 0

~

18 1

18 2

~

~~

18 3

20 1

20 4

~~

22 6

~

total mg/100 ml serum

Carbohydrate Spital

1.6 1.6

1.0 1.7

0.8 0.6

26.3 26.8

8.8 6.9

0.3 0.3

6.5 6.6

33.7 25.8

10.8 17.7

~

0.3

-

1.0

8.8 8.3

-

1.0

27 27

Fig. 3. Thr ef/ecrs of .serum on the cellular morphology uf the isolated hepatocjte. (A) Interference contrast of hepatocytes after the initial 4 h in culture. The cells arc typically rounded and a bi-nucleate cell often observed can be seen. (B) Interference contrast ot' hepatocytes after 48 11 in culture without serum. (C) As (B) but incubated with 15?" (v/v) rat serum. (D) Phase contrast of hepatocytes after 48 h in culture with serum showing extensive spreading

Whereas diabetic animals show a decreased synthesis of fatty acid synthetase, the administration of insulin or fructose feeding without insulin restores the synthesis of the enzyme to a normal level, but not the elevated levels observed when normal animals are fed a high carbohydrate diet [18]. These data could thus be explained, in the light of our own observations with stearoyl-CoA desaturase which behaves in many respects in a similar way to fatty acid synthetase [5,6]. The induction of lipogenic enzymes may require insulin when glucose, but not fructose, is the carbon source. However, in the presence of fructose, insulin enhances the net synthesis by stabilising the newly formed enzyme. This could be achieved by a direct effect on the enzyme, or indirectly by inactivating a protease.

Lipid Synthesis in the Isolated Hepatocyie A 200 g male rat was starved and refed as described in Materials and Methods. Hepatocytes were isolated and allowed to become attached to culture dishes. To the attached cells were added 4 ml of serum-free Ham's F-12 medium plus 2 pCi [2-I4C]acetate and the cells incubated at 37-C in an atmosphere of 95%) 0 2 5 C 0 2 for various times up to 90 min. The incorporation of [14C]acetate into fally acids and sterols was determined and the results shown in Fig.4. As can be seen from the graph, there was a linear incorporation of acetate into both fatty acids and sterols over the first 60 min, although the rate of the former was approximately 5.5 times that of the latter. These