j. lipid res. 1970 de gomez dumm 96 101

7/30/2019 J. Lipid Res. 1970 de Gomez Dumm 96 101

1/6

Effect of diet on linoleic acid desaturation andon some enzymes of carbohydrate metabolism

I . NELVA T . DE GOMEZ DUMM, MARIA J .T . DE ALANIZ, and RODOLFO R. BRENNER*Citedra de Bioquhica, Instituto de Fisiologik, Facultad deCiencias Mhdicas, Calle60 y 120, La Plata, Argentina

ABSTRACT The effect of diet on the desaturation of linoleicacid to y-linolenic acid by liver microsomal preparations, onblood glucose and insulin levels, and on activities of gluco-kinase, hexokinase, pyruvate kinase, and a-glycerophosphatedehydrogenase have been studied. The female rats used in theseexperiments were maintained on one of the following dietaryregimes: ( a ) fasted, (6) fasted for 96 hr and refed glucose, (c)balanced diet, (d) carbohydrate-free diet, ( e) lipid-free diet,or ( f ) protein-free diet. Fasting for 96 hr caused a decrease ofboth linoleic acid desaturation and glucokinase and pyruvatekinase activity together with a slight decrease of the bloodinsulin level. a-Glycerophosphate dehydrogenase activity wasnot modified. Refeeding of glucose for 50 hr increased theconversion of linoleic acid to linolenic acid as well as theactivities of all the enzymes studied except a-glycerophosphatedehydrogenase. The increase in desaturation, however, wastransient. The feeding of a lipid-free diet did not modify thetested parameters. Feeding a carbohydrate-free diet for 96 hrresulted in increased linoleic acid desaturation but decreasedglucokinase and pyruvate kinase activity, thus apparentlyeliminating a putative correlation between the fatty aciddesaturating activity and glycolytic activity or blood insulinlevels under these experimental conditions. The findingssuggest that dietary proteins may play an important role indetermining the level of fatty acid desaturation.

SUPPLEMENTARY KEY WORDS pyruvate kinase *glucokinase . a-glycerophosphate dehydrogenase . insulinliver microsomes fasting * lipid-free diet . carbohydrate-freediet protein-free diet

A L T H O U G H THE EFFECT of insulin on lipid metabolismhas been studied for many years, only in the last 8 yrhas it been shown that insulin and dietary glucose*Member of the Carrera del Investigador Cientifico of theConsejo Nacional de Investigaciones Cientlficas y Tecnicas,Argentina.

modify the fatty acid desaturating capacity in liver andother tissues(1-7) of diabetic and fasted rats, respectively.

These modifications of the enzymic activity con-sequently produce changes in the fatty acid compositionof the tissues (7, 8). Rats made diabetic by administra-tion of alloxan have a decreased capacity to desaturatefatty acids. The administration of insulin restores thisdesaturating activity and thereby, causes conversionsofstearic acid to oleic acid (2, 9 ) , linoleic acid to y-linolenicacid, a-linolenic to octadeca- 6, 9, 12, 15- t et raenoi cacid,and oleic acid to octadeca-6,9-dienoic acid (5, 7). I thas been suggested that insulin probably regulates thesereactions by means of enzyme induction (4, 7). In thefasted rat there is also a decrease in fatty acid desatura-tion (7, 10).Administration of glucose for 12 hr practi-cally restores the desaturation to normal. This effect hasbeen attributed to an increase in the secretion of insulinevoked by the glucose administration (7, 10). However,recent findings have shown that thyroxine (11) may re-place insulin in the restoration of the fatty acid desaturat-ing capacity of the alloxan-diabetic rat. Besides, al-though the concentration of blood glucose may not bedirectly responsible for inducing the desaturating activityas the level of glucose is high in diabetic rats and low infasted animals, it is impossible to elucidate clearly fromthe aforementioned studies (7, 10) whether the effectof insulin is due to insulin per se or whether it is actuallymediated via the utilization of glucose at the cellularlevel. The following experiments constitute an attemptto study simultaneously the effects of different diets onfatty acid desaturation and on the activity of someenzymes involved in the metabolismof glucose.

MATERIAL AND METHODSAn,imalsFemale albino rats of the Institute strain weighing

96 JOURNALF L IP IDRESEARCH OLUME 1, 1970


2/6

90-140 g and maintained on a standard laboratory dietwere used.Treatmento AnimalsIn the first experiment the rats, in groups of five animals,were killed after 24, 48, 72, or 96 hr of fasting. Fivegroups of rats which were fasted for 96 hr were offered a5% glucose solution and were killed after 4, 8, 16, 32and 50 hr of glucose refeeding. A group of animals fed abalanced diet was used as control. The fasted rats weregiven water ad lib.

In the second experiment the rats, in groups of 10animals each, were kept in individual cages and feddifferent isocaloric diets (25 kca1/100 g body weight)for 96 hr. All the experimental diets contained the sameamount of salts and vitamins (12). Water was given adlib. Fat-free casein, sunflower seed oil, and dextrinwere used as sources of protein, fat, and carbohydrate,respectively. In the balanced diet (CD) 50, 23, and 27%of the calories were supplied as carbohydrate, protein,and fat, respectively. The distribution of calories in thevarious deficient diets was as follows: carbohydrate-freediet (CF), 73% protein, 27% fat; fat-free diet (LF),730/, protein, 270/, carbohydrate; protein-free diet(PF), 73% carbohydrate, 27% fat. An additional group(F) was fasted. The animals were killed after they hadbeen on the respective diets (or fasting) for 96 hr.I solation of MicrosomesThe rats were killed by decapitation without anesthesia.The blood was allowed to drain and was collected forglucose and insulin determinations.

The livers were excised, blotted on filter paper, andweighed. The whole liver was homogenized briefly in acold solution (3:1v/w) consistingof 0.15 M KC1, 0.005M MgC12, 0.004 M EDTA, 0.004 N-acetylcysteine, 0.05Mphosphate buffer (pH7), and 0.25M sucrose. Cell debrisand mitochondria were removed by sedimentation at12,000g for 10 min at 0C. The microsomal fractionwas prepared by centrifugation at140,000gfor 60 min ina Spinco Model L2 ultracentrifuge (13). The pelletswere resuspended in the homogenizing solution (1:5v/v>.Assay ProceduresThe desaturation of linoleic acid to y-linolenic acid bymicrosomes was measured by estimation of the per-centage conversion of linoleic acid-l-14C (51.9 mci/mmole; Radiochemical Centre, Amersham, England)to y-linolenic acid. 5 nmoles of labeled linoleic acidand 5 nmoles of unlabeled linoleic acid were incubatedwith 2 mg of microsomal protein in a Dubnoff shakerat 25C for 20 min in a total volume of 3 ml of a 0.15M KCl, 0.25 M sucrose solution containing 4 pmole of

ATP, 0.2 pmole of CoA, 2.5 pmole of NADH, 15 pmoleof MgC12, 4.5 pmole of glutathione, 125 pmole NaF, 1pmole nicotinamide, and 125 pmole of phosphate (pH7). At the end of the incubation, the fatty acids were re-covered by saponification of the incubation mixture andextraction with petroleum ether (bp 30C). The acidswere esterified with methanolic 3 M HCl (3 hr at 68"C),and the distribution of the radioactivity between linoleicand y-linolenic acid was determined by gas-liquidradiochromatography in a Pye apparatus with a pro-portional counter. No measurable radioactivity wasfound in other fatty acid peaks. The recovery of theradioactivity from individual incubation mixtures wasSO-lOO%. The methyl ester of y-linolenic acid wasidentified by comparing its retention time to that of anauthentic sample. The y-linolenic acid which wassynthetized by the microsomes, was also separated bypreparative gas-liquid radiochromatography on a col-umn of 10% diethylene glycol succinate on ChromosorbW (80-100 mesh) at 180C. The structure and positionof labeling was confirmed by reductive ozonolysis (14)followed by gas-liquid radiochromatography on thesame column. Radioactivity was found only in thealdehyde-ester fraction with 6 carbon atoms.

The 140,000g supernatant solution was used for theother enzyme assays. Glucokinase and hexokinase weremeasured by the methods of Walker and Parry (15).Pyruvate kinase was assayed according to the procedureof Bucher and Pfleiderer (16). a-Glycerophosphatedehydrogenase activity was measured by the method ofFitch and Chaikoff (17). The specific activities of theseenzymes were expressed as pmoles of substrate trans-formed per minute. The values were expressed as unitsper mg of DNA in the total homogenate so that theymight be related more directly to the number of cells.

Proteins in the microsomal and supernatant frac-tions were determined by the biuret method of Gornall,Bardawill, and David (18) using crystalline bovineserum albumin as standard. DNA in the homogenate wasmeasured by the method of Burton (19). Immuno-reactive insulin was determined by a radioimmunoassayprocedure using beef insulin standards (20). Glucosewas measured by the glucose oxidase method (21).

RESULTS AND DISCUSSIONAs it was previously postulated and discussed in theintroduction, insulin may induce the synthesis of thedesaturating enzyme (2-10). The evidence for this wasobtained by measuring the fatty acid desaturase activityand messenger RNA levels in tissues of diabetic orfasted animals previously injected with actinomycin Dor puromycin before the administration of insulin orglucose, respectively. However the effect of insulin may

DE GOMEZ m T AL. Linoleate Desaturation and GlucoseMetabolism 97


3/6

also be mediated by its effect on the utilization of glucoseat the cellular level. Moreover, this hormone causes anonspecific increase in protein synthesis (22). In this re-spect, there is evidence which supports the suggestionthat insulin may also increase the generalized synthesisof proteins by a direct effect on the ribosomes (22, 23).This concept does not conflict with the hypothesis thatthe action of glucose in increasing the fatty acid desatura-tion activity in the fasted rat is mediated by an enhancedsecretion of insulin.Effect of F asting and Subsequent Refeeding o GlucoseonHepatic Microsomal Desaturation of Linoleic AcidThe level of desaturation activity was studied in relationto blood glucose and insulin levels and hepatic activitiesof phosphorylation of glucose to glucose-6-phosphate(glucokinase), pyruvate kinase, and a-glycerophos-phate dehydrogenase. Glucokinase (24) and pyruvatekinase are two insulin-dependent enzymes that may beresponsible for the regulation of glycolysis (24). Thelevel of a-glycerophosphate dehydrogenase activity maybe related to glycerophosphate synthesis (17). Al-though the total glucose phosphorylation to glucose-6-phosphate includes both hexokinase and glucokinaseactivity, the hexokinase enzyme is not dependent oninsulin. Thus, a measurement of total glucose phosphory-lation can be substituted (25) for a direct assay ofglucokinase activity in order to measure insulin-de-pendent variations of glucose metabolism.

Fig. 1 illustrates how the conversion of l in~leic- l-~Cto 7-linolenic acid decreases during a 96 hr period offasting. This decrease is coincident with significant de-clines in both glucokinase and pyruvate kinase activities.This latter effect has already been reported by severalother laboratories (26-28). The decrease in glucokinaseactivity suggests a diminished utilization of glucose viathe Emden-Meyerhoff pathway. The concomitant de-crease in pyruvate kinase activity may also indicate areduction in the flow of glucose metabolites into thetricarboxylic acid cycle. Fitch and Chaikoff (17) havefound that the feeding of a carbohydrate-rich diet maymodify the activity of a-glycerophosphate dehydro-genase. However, the level of this enzyme was not sig-nificantly changed in our experiment. If its activity is ameasurement of the extent of the diversion of glucosemetabolites toward lipid synthesis, the present resultswould imply that the fall of linoleic acid desaturatingactivity could not be ascribed to an increase of thedirect incorporation of glucose metabolites into lipids.This observation is relevant since an increase in phospho-lipid synthesis could modify the microsomal fatty aciddesaturating activity (29, 30). A 96 hr fast caused only aslight decrease in blood insulin levels (Fig. 2). However,the decrease in the activities of the two insulin-dependent

enzymes, glucokinase and pyruvate kinase, was on theorder of the decline of desaturase activity; this might beexplained by the aforementioned postulate that allthree of these enzymes are insulin-dependent. Moreover,the immunological estimation of insulin may not neces-sarily show real variations of biologically active insulin,since other forms of insulin such as proinsulin may be inthe blood (31, 32). On the other hand, the level ofdesaturase activity may be related to the rate of glucosemetabolism.

In the same experiment (Fig. l ), the administrationof glucose in the drinking water caused an enhancementof linoleic acid desaturation activity, glucokinase andpyruvate kinase activity, and a significant increase inblood insulin levels (Fig. 2). These results would agreewith the aforementioned hypothesis of insulin action.However after 16 hr, although the activities of the twoglycolytic enzymes and the blood insulin levels had con-tinued to increase, the fatty acid desaturating activitydecreased abruptly (Fig. 1).The effect of insulin on fattyacid desaturation activity has already been thoroughlydemonstrated in a number of experiments (1-9). Theobserved lack of correlation between circulating insulinlevels and the desaturase activity suggests that anotherfactor(s) may influence the desaturation.EjTect o Different Dietary Components on LinoleicAcid Desaturat:onI n the second experiment the effects of various dietaryregimes on the desaturation of linoleic acid-1 4C toy-linolenic acid by liver microsomes, and on the gluco-kinase, pyruvate kinase, and a-glycerophosphate dehy-drogenase activities of liver supernatant fractions wereexamined. The results are shown in Figs. 3-6. In thefirst place, these results confirm that fasting for 96 hr(F) causes significant decreases in linoleic acid desatura-tion (Fig. 3) and in activities of glucokinase (Fig. 4)and pyruvate kinase (Fig. 5). The hexokinase activity(Fig. 4) was not affected, and blood insulin levels (Table1) and a-glycerophosphate dehydrogenase activity(Fig. 6) decreased slightly although these values werenot significantly different from values obtained foranimals on the balanced diet (CD).

A lipid-free diet (LF), containing only proteins andcarbohydrates, did not cause a modification of thedesaturation activity of the microsomes (Fig. 3). How-ever, carbohydrate-free diet (CF) caused an unexpectedand significant increase in the desaturation of linoleicto y-linolenic acid compared with the balanced diet. I ncontrast, the protein-free diet (PF), produced a smalldecrease in desaturation activity without affectingglucokinase activity (Fig. 4) or blood insulin levels(Table 1). Consequently, these results suggest thatdietary lipids do not modify the desaturase activity of

98 J OURNALF LIPIDRESEARCHVOLUME1, 1970


4/6

z0L>0I-

Y

L6

4z54zD34

600.

-EeE0 200 .

400 .\mulv)V

30t

bI1111I 1I I1 11 1I ,s / 50 ,xI I /I I ,\.I1 : - - - 3 25s 20+uI-d 0

CD F CF LF PFF IG .3. Effect of different diets on the conversion of linoleic-lJ4C to 7-linolenic acid. Rats were fed 25kca1/100gbody weight for 96hr.CD,balanced diet;F,96hr fast; CF, carbohydrate free; PF , protein free.The results are the meansof analysis of 10 animals (each analysis was performed in triplicate). The verticallines represent SEM.

DE GOMEZUMMT AL . Linoleate Desaturatwn and GlucoseMetabol i sm 99


5/6

A 6l

azm

0

25

I

CD F C F LF PFFIG.4.details. Effect of different diets on hexokinase (A ) and glucokinase(B) activity: Seelegend of Fig. 3 for

25UznE-\3

0 CD F CF LF PFFIG.5. Effect of different diets on pyruvate kinase activity. See legend of Fig. 3 for details.

2 5

6z0F\

0CD F CF LF PF

FIG.6.details.

Effect of different diets on a-glycerophosphate dehydrogenase activity. See legend of Fig. 3 for

the microsomes but that the level of dietary proteinexerts an effect on the desaturation reaction.

We arrive at this conclusion as a result of the follow-ing reasoning. Previous experiments (7) have demon-strated that there is a nearly complete recovery of thelinoleic acid desaturation activity in 48-hr fasted ratsgiven glucose for 12hr, but that this effect is blocked byprior administration of actinomycin D. Similar resultsare also illustrated in Fig. 1. The carbohydrate-free,protein-rich diet which was fed to the animals caused anincrease, rather than a decrease, in the desaturationactivity. Hence, since dietary lipids appear to have no

effect, the excessively high intake of proteins in these ex-periments is probably the factor that causes the in-crease in desaturation activity. The small decrease indesaturation seen in the rats fed the protein-free diet,which contained carbohydrates and lipids, would alsoagree with this hypothesis. The activities of the twohepatic glycolytic enzymes, glucokinase (Fig. 4) andpyruvate kinase (Fig. 5), were not modified by thelipid-free diet (LF), but were significantly and similarlydecreased in fasted animals and in animals fed thecarbohydrate-free diet (CF). Hence, the glucokinase andpyruvate kinase activities showed no correlation with

TABLE 1 EFFECT F DIFFERENTIETS N THE LEVELSF BLOODGLUCOSEND IMMUNOREACTIVENSULINCD F CF L F PF

85 +0.03 140 =k 0.04 133 f 0.15 117 f 0.03lucose (mg/100 ml) 136 =t .54Insulin (aU/ml) 36.9 f 3. 4 29.1 f 7.4 30.5 =t 8 . 2 21.3=t .3 38.5 =t 5 . 5~

Diets: CD, balanced diet;F, fasted; CF,carbohydrate-free; LF, lipid-free; PF, protein-free.Analysis were performed in duplicateoneachof the10animals in each dietary group. The values are the means of f SEM.100 J OURNALF LIPIDRESEARCHVOLUME1, 1970


6/6

the fatty acid desaturating activity when a carbo-hydrate-free diet, rich in proteins and lipids, was ad-ministered. Therefore, these results appear to eliminatethe possibility that the increase in linoleic acid desaturat-ing activity under these experimental conditions is re-lated to an increase in glucose metabolism. I n fact, thedesaturation activity is apparently not related to theblood insulin levels (Table I ), since when a carbo-hydrate-free diet was fed, desaturation activity in-creased, whereas the blood insulin levels were not sig-nificantly modified and the activity of the two insulin-dependent enzymes were reduced. The desaturationactivity also appears to be independent of lipid syn-thesis since the a-glycerophosphate dehydrogenaseactivity is not significantly altered by diet (Fig. 6). Thus,in light of the aforementioned results, it becomes neces-sary to acknowledge the existence of a factor related toprotein metabolism that may cause an increase in themicrosomal desaturation of linoleic acid to y-linolenicacid under certain experimental circumstances. Thisfactor may play a special role in the biosynthesis ofpolyunsaturated fatty acids. In support of this is the re-cent work of Inkpen, Harris, and Quackenbush (33)who have published the results of a series of experimentswhich show that dietary proteins may modify thedesaturation of a-linolenic acid.This work was supported by the Consejo Nacional de Inves-tigaciones Cientificas y Ttcnicas, A rgentina, grantNo. 2800/66and the Comisibn de Investigaciones Cientificas of the Univer-sity of L a Plata, grantNo. 9.T he authors are indebted to Dr.R. 0.Peluffo for his fruitfuldiscussions and to M iss N. E . Cristalli for her excellent assis-tance.Manuscript receved 28 May 7969 and i n revised form 75September7969; accepted79November 7969.

1.2.3.4.5.6.

REFERENCESImai, Y . 1961.J .Biochem. (Tokyo).49; 642.Gellhorn, A., and W. Benjamin. 1964. Biochim. Biophys.Acta. 84: 167.Benjamin, W., and A. Gellhorn. 1964. J . Biol. Chem. 239:64.Gellhorn, A ., and W. Benjamin. 1964.Science (Washington).146: 1166.M ercuri, O., R . 0. Peluffo, and R. R. Brenner. 1966.Biochim. Biophys. Acta. 116: 409.M ercuri, O., R. 0. Peluffo, and R. R. Brenner. 1967.Lipids. 2: 284.

7.8.9.

10.11.12.13.14.15.

16.

17.18.19.20.21.22.23.24.

25.

26.27.28.29.30.31.32.33.

Brenner, R. R., R. 0.Peluffo, 0.M ercuri, and M. A.Restelli. 1968.Amer.J .Physiol.215: 63.de A laniz, M . J . T.,and R. R. Brenner. 1969. Acta Physiol.Latino Amer. 19: 1.Gellhorn, A ., and W. Benjamin. 1966. Biochim. Biophys.Acta. 116: 460.Benjamin, W., and A . Gellhorn. 1966. J . Lipid Res. 7:285.Gompertz, D., and A. L . Greenbaum. 1966. Biochim.Biophys. Acta. 116: 441.Peluffo, R. O., R . R. Brenner, and 0.M ercuri. 1963.J . Nutr. 81: 110.Brenner, R . R., and R. 0.Peluffo. 1966. J . Biol. Chem.241: 5213.Stein, R . A ., and N. Nicolaides. 1962. J . Lipid Res. 3:476.Walker, D. C., and M. J . Parry. 1966. In M ethods inEnzymology. S. P. Colowick and N.0.K aplan, editors.Academic Press, Inc., New York. 9: 381.Bucher, T ., and G. Pf leiderer. 1955. M ethods n Enzymol-ogy. S. P. Colowick and N. 0.K aplan, editors. AcademicPress, Inc., New Y ork. 1: 435.Fitch, W. M ., and I . L . Chaikoff. 1960. J . Biol. Chem.235: 554.Gornall, A. G., C. J . Bardawill, and M . M . David. 1949.J .Biol. Chem. 177: 751.Burton, K. 1956.Biochem. J .62. 315.Herbert, V., K . L au, C. W . Gottlieb, and S. J . Bleicher.1965. J .Clin. Endocrinol. Metab.25 . 1375.Beach, E . F., and J . J .Turner. 1958.Clin. C h . : 462.L ukem, F. D . W . 1964.Diabetes. 13: 451.Wool, I . G., 0.R . Rampersad, and A. N . M oyer. 1966.Amer. J .Med. 40: 716.Weber, G., R . L . Singhal, N. B. Stamm, M . A. Lea, andE. A . Fisher. 1966. In Advances in Enzyme Regulation.M . G. Weber, editor. Pergamon Press L td., Oxford, Eng-land. 4: 59.Walker, D. G. 1967. In Advances in Enzyme Regulation.George Weber, editor. Pergamon Press L td., Oxford,England. 3: 163.Di Pietro, D. L ., and S. Weinhouse. 1960. J . Biol.Chem.235; 2542.Perez, N., L . C lark-T urri, E. Rabajil le, and H. Niemeyer.1964.J .Biol. Chem.239: 2420.Niemeyer, H., L. Clark-Turri, E. Garcts, and F. E.V ergara. 1962.Arch. Biochem. Biophys.98: 77.Nervi, A. M ., R. R. Brenner, and R. 0.Peluffo. 1968.Biochim. Biophys. Acta. 152: 539.Brenner, R. R ., and R. 0.Peluffo. 1969. Btochim. Biophys.Acta. 176: 471.Rubenstein, A. H., S. Cho, and D. F . Steiner. 1968.Lancet. 1353.Roth, J ., P. Gorden, and I . Pasan. 1968. Proc. Nat. Acad.Sci.U.S.A.61: 138.Inkpen, C. A., R. A . Harris, and F. UT. uackenbush.1969. J .LipidRes. 10 : 277.

DE GOMEZ UMMT AI.. Linoleate Desaturation and GlucoseMetabolism 101

j. lipid res. 1970 de gomez dumm 96 101

Documents