Inhibition of the stearyl coenzyme A desaturase system by sterculate.

Abstract The mechanism of the inhibition in vitro of the microsomal stearyl-CoA desaturase system by sterculate has been investigated. Our data do not support the suggestions that this is mediated by the binding of thiol groups of enzyme proteins by the cyclopropene group and that this is competitive with substrate. Sterculate did inhibit the desaturase system in vitro but data suggest this to be nonspecific, stemming from the detergent nature of free fatty acid and hence of doubtful significance for metabolism in vitro. However, it was found that dietary (methyl) sterculate decreased the activity of the microsomal stearyl-CoA desaturase system. The mechanism of this action of sterculate remains obscure. A new procedure for the precise measurement of the microsomal stearyl-CoA desaturase system is described based on the permanganate-periodate oxidation of fatty acid methyl esters followed by the separation of monomethyl azelate-14C derived from oleate-1-14C from other fatty acid methyl esters with the use of a Florisil column. The utilization of rat liver microsomes for the enzymatic synthesis of stearyl-1-14C-CoA is also described.

for the precise measurement of the microsomal stearyl-CoA desaturase system is described based on the permanganate-periodate oxidation of fatty acid methyl esters followed by the separation of monomethyl azelate-14C derived from oleate-1-"C from other fatty acid methyl esters with the use of a Florisil column.
The utilization of rat liver microsomes for the enzymatic synthesis of stearyl-l-14C-CoA is also described.
One of the well known effects of dietary cyclopropene fatty acids is a relative decrease in the concentration of monounsaturated fatty acids with a concomitant increase in the amount of the corresponding saturated fatty acids in tissue lipids (l-3). Various investigators have reported that dietary cyclopropene acids inhibited the conversion in tivo of stearate-l-14C to oleate (3-6). Subsequently the conversion in vitro of stearate to oleate has also been found to be inhibited by sterculate (2,3, 5). Raju and Reiser (5) concluded that the mechanism of this inhibition involves the binding of enzyme sulfhydryl groups by the cyclopropene group. According to Allen et al. (3) this effect of * These studies were supported in part by Contract AT(04-l)-GEN-12 between the Atomic Energy Commission and the University of California, and by United-States Public Health Service Research Career Award   Palmityl-l-14C-CoA and oleic acid-l-14C were generous gifts from Dr. Armand J. Fulco. Stearic-1-'4C was a product of Nuclear-Chicago.
Sources of other chemicals have been described previously (7,8). The radioactive fatty acids were purified by thin layer chromatography.
Potassium or ammonium salts of fatty acids were prepared as described earlier (7). In view of the lability of sterculate, the ammonium sterculate used was frequently a freshly prepared product.
When such preparations were to be used later, they were kept frozen under nitrogen and were exposed to air for no more than a few minutes prior to use. The sterculate, thus stored, appeared to remain stable, as judged by the analysis of the isolated fatty acid (after methylation and hydrogenation) by gas-liquid chromatography. Rat liver microsomes were obtained as described previously (7). Stearyl-l-14C-CoA was synthesized basically according to the method of Kornberg and Pricer (9) but rat liver microsomesl were used as the enzyme source and the possibility of any concomitant desaturation of stearyl-CoA formed was eliminated by carrying out the synthesis under nitrogen in the presence of cyanide.2 The incubation system contained the following com-1 Since rat liver microsomes could be used without any further manipulations and the long chain fatty acid-activating enzyme remains stable, when microsomes are kept frozen (7), this pricedure may be simDler than those with IvoDhilized guinea Die liver  20.66 (0.5 mCi). Incubation was for 1 hour at 37". Stearyl-1-r4C-CoA was isolated and purified as described by Bloomfield and Bloch (10). The over-all yield, as estimated by hydroximate calorimetry (7), was 65% based on the amount of fatty acid added.3 Paper chromatography of the product in water-pyridine-isopropanol (9) showed that over 90% of the total radioactivity was associated with the area occupied by stearyl-CoA.
Assay of Stearyl-CoA Desaturase System-The usual assay system, in a final volume of 0.25 ml, contained: 50 pmoles of Tris-HCl buffer (pH 7.4), 0.2 pmole of NADH, 16 mhmoles of stearyl-l-*4C-CoA (40,000 to 60,000 cpm), and 200 to 300 pg of microsomal protein to start the reaction. Incubations were for 5 min at 37" in air.
The reaction was terminated by the addition of 0.25 ml of methanolic KOH (90% v/v methanol,0.73 N KOH) and the contents were saponified for either 30 min at 80" or overnight at 50". After acidification, lipids were extracted with ether-petroleum ether (1 :l, v/v) and fatty acids were converted to methyl esters by diazomethane in the presence of methanol.
The methyl esters were freed of polar impuri-ties4 by passing the solution of lipids in 5% ether in petroleum ether through deactivated Florisil (12). To the solvent-freed methyl esters were added 0.6 ml of 0.5 M NazHP04.7HzO and 0.90 ml of a freshly prepared oxidation mixture obtained by dissolving 210 mg of NaI04 in 6 ml of water and adding 7 ml of t-butanol and 5 ml of 0.32% aqueous KMn04. After 1 hour at room temperature, the contents were acidified with 0.10 ml of 1.2 N HCl and the lipids were extracted twice with ether-petroleum ether (l:l, v/v). The lipid extract was applied to an approximately 500-mg Florisil column packed in, and containing, petroleum ether. The column was washed twice with 12 to 13 ml of ether and the Florisil was air-dried and transferred to a counting vial containing 2 ml of water. After adding 15 ml of naphthalene-dioxane counting fluid (13), the radioactivity was determined with the Packard Tri-Carb 574 scintillation spectrometer.
A correction was applied for the decreased counting efficiency that resulted from the presence of water and Florisil. Controls involving either heated microsomes or to which stearyll-YXoA was added following the addition of methanolic KOH were also included.

AND DISCUSSION
Principle and Validity of Assay Procedure-In the procedure used for measuring stearyl-CoA desaturation, advantage has been taken of the fact that under appropriate conditions per-3 An increase in the time of incubation or the amount of enzyme protein did not improve the yield because of the attainment of equilibrium.
The yield, with respect to added fatty acid, could be increased, however, by increasing CoA concentration.
4 Where applicable (such as under conditions of the usual assay procedure) the procedure may be simplified by omitting this step since kuown polar products derived from cy-, fl-, and w-oxidation or elongation reactions of fatty acids are not retained by Florisil under the conditions used and hence do not contribute to error. However, inclusion of this step in the analysis increases the specificity of the assay procedure by eliminating possible complications stemming from the formation of any sufficiently polar material which may be retained by Florisil.
For example, such material(s) was nonenzymatically formed under the usual assay conditions if cysteine was included at high levels. manganate-periodate oxidation completely cleaves the double bond of the fatty acid methyl esters (13) with the formation of earboxylic groups, while the saturated fatty acid methyl esters remain unaffected.
Thus oleate-l-14C derived from the desaturation of stearyl-1-14C-CoA gives rise to monomethyl 14C-azelate and pelargonic acid on oxidation. These products derived from methyl oleate-l-14C are strongly adsorbed by the Florisil because of the free carboxylic group while methyl esters of fatty acids are readily eluted.
Under the conditions described in the assay procedure, analysis of known stearate-l-14C and oleate-1-W mixtures showed that the oxidation of oleate was complete in 1 The complete reaction system contained 50 pmoles of Tris-HCI (pH 7.6)) 0.8 @mole of NADH, 294 sg of microsomal protein (from "refed" rat liver), and 16 mpmoles of stearyl-lJ4C-CoA, in a final volume of 0.25 ml.
Incubation for enzyme assay was for 7.5 min at 37'. When appropriate, addition of reduced glutathione was made after that of NADH.
When GSH and sterculate both were present, addition of GSH preceded that of fatty acid. After fatty acid addition, the contents were incubated for 7 min at room temperature before microsomes were added. After adding microsomes, incubation at room temperature was continued for 7 more min before the assay was initiated by the addition of stearyl-CoA. , and either 60 mltmoles of GSH or 588 pg of microsomal protein were previously incubated with or without 450 mpmoles of sterculate in 0.8 ml. After 5 min 0.75 pmole of 5,5'-dithiobis(2-nitrobenzoic acid) and water were added to bring the volume to 3.0 ml and absorbance was measured 3 to 5 min later against a blank treated in an identical manner except that 5,5'-dithiobis(2-nitrobenzoic acid) was not added. A correction was applied for the contribution of 5,5'-dithiobis(2-nitrobenzoic acid) to the absorbance measurement. Usual assay conditions were followed with either 213 pg (0-O) or 532 pg (X-X) microsomal protein from normal rat liver and 7.5-min incubation.
hour at room temperature and that the adsorption of mono-methylJ4C azelate on Florisil and the elution of methyl stearatel-14C from the column were essentially complete (98 to 99%). With this analytical procedure it was possible to measure stearate conversions to oleate of the order of 1 To with a high degree of accuracy and with excellent reproducibility.
This enabled assays to be carried out at nonlimiting levels of stearyl-CoA under linear kinetic conditions whereby activity of the desaturase system could be expressed precisely.
In the methods based on gas-liquid chromatographic separations of oleate from stearate, this has not been attained (23,5). Further, the presently described procedure is simpler, quicker, and more precise than the separation of oleate and stearate by argentation thin layer chromatography (5,15,16) and any cr,@-unsaturated fatty acid formed would not contribute to error in the present method as compared to the other chromatographic methods considered above.
To confirm that the radioactivity retained by the Florisil in the usual assay with stearyl-l-14C-CoA and microsomes as outlined in the procedure was due to monomethyl-14C azelate, in one case the ether-washed Florisil with adsorbed radioactive material was treated with BFI in methanol (16). The lipids were extracted and thin layer chromatography with 10% ether in pentane as developing solvent showed that at least 90% of the radioactivity was associated with the dimethyl azelate (Rp 0.40) fraction.
Under the assay conditions used, the rate of desaturation was linear with varying microsomal protein within the range of 200 to 480 pg of protein.
The rate of desaturation was linear for somewhat more than 10 min of incubation only. All assays were carried out within these limits and exact assay details have been described. suggested that the mechanism of inhibition of monodesaturation by cyclopropene fatty acids was due to the irreversible binding of enzyme sulfhydryl groups by the cyclopropene group. We attempted to re-examine this possibility with a more direct assay procedure starting with stearyl-1-14C-CoA as substrate and found that sterculate was indeed inhibitory for the microsomal desaturase system (Table I).
To see whether the cause of this inhibition was an irreversible reaction of cyclopropene groups with the sulfhydryl groups of the enzyme, sterculate was previously incubated with an excess of reduced glutathione before it was allowed to react with the enzyme system and the effect of such preliminary incubation of the effectiveness of sterculate as an inhibitor of the desaturase system was determined.
As is evident from the results obtained (Table I), the preliminary incubation of sterculate with a lOO-to 300-fold excess of reduced glutathione did not affect the inhibitory effectiveness of sterculate at all. This finding casts doubt on the suggested reaction between thiol groups and the cyclopropene groups as the cause of the sterculate inhibition of the desaturase system.
Direct experiments were then performed to see whether sterculate readily reacts with the thiol groups of reduced glutathione and of the desaturase system, as claimed by Raju and Reiser (5). Ellman's reagent (5,5'-dithiobis(2-nitrobenzoic acid)), readily reacts irreversibly with aliphatic thiols, giving rise to mercaptide ion with intense absorption at 412 rnp, and this conveniently permits the measurement of changes in the concentration of free thiol groups (20). Hence, experiments were initiated making use of 5,5'-dithiobis(2-nitrobenzoic acid), to see whether there was any evidence of disappearance of the thiol groups of either the reduced glutathione or the microsomal enzyme system on treatment with sterculate as is to be expected if the cyclopropene groups reacted with the thiol group(s). The conditions under which sterculate was allowed to react with reduced glutathione or microsomal enzyme preparation were the same as those under which sterculate inhibited the desaturase system.
As may be seen from the data obtained (Table II)  3. The effect of palmityl-CoA concentration on the activity of the microsomal desaturase system. Usual assay conditions were used with 294 pg of microsomal protein from refed rat liver and with 3-min incubation and varying palmityl-l-lPC-CoA concentration.
saturase system is not the result of the reaction of the thiol groups of the enzyme system with the cyclopropene groups.

Effect of Substrate Concentration on Desaturase System and Its
Reaction to Stcrculate Inhibition-Allen et al. (3) have suggested that sterculate acts as a competitive inhibitor for stearate desaturase.
Determination of such an effect in our initial experiments was complicated by the anomalous results obtained when the effect of stearyl-CoA concentration on the activity of the enzyme system was determined.
As may be seen from Fig. 1, at lower concentrations of stearyl-CoA the enzyme activity curve was noticeably S-shaped while at higher concentrations, stearyl-CoA inhibited the desaturase system. It can be seen (Fig. 1) from the comparison of results obtained at two different levels of microsomes that the range of concentration of stearyl-CoA giving optimum activity as well as showing an inhibitory effect was related to the concentration of microsomal material, being higher at the higher microsomal amount. This is similar to the effects of linolenate concentration on the microsomal activation of linolenate described earlier (7) and shows that microsomes readily bind stearyl-CoA as well as free fatty acids. The possibility that the contaminating acyl-CoA hydrolase of the microsomal fraction competes with the desaturase system for stearyl-CoA at lower levels and that this results in the S-shaped activity curve for the desaturation was tested. For this, microsomes from refed rat liver were used since these were found to have increased activity of the desaturase system while the acyl-CoA hydrolase activity was not increased by refeeding. To minimize the interference from the acyl-CoA hydrolase, the incubation period was reduced to 3 min only. However, as is evident from Fig. 2, an S-shaped activity curve was still seen, thus complicating the determination of possible competitive inhibition by sterculate. This type of enzyme activity curve was also seen with palmityl-CoA as a substrate (Fig. 3). However, with this substrate, the sigmoidity of the activity curve was less pronounced than that with stearyl-CoA.6 The determination 6 Zahler, Barden, and Cleland (21) also observed anomalous ensvmatic kinetics with palmityl-CoA as a substrate while studying"the enzymatic acylation of n-glycerol-3-P and this was found to be related to the formation of micelles of nalmitvl-CoA at low levels (22). It is likely that the presently observed S-shaped enzyme activity curve resulted from the preference of the enzyme system for the micellar form of long chain acyl-CoA esters assub- However, in such experiments, no inhibition by sterculate was noticeable.
When the effect of sterculate on the desaturase system was determined with concentrations of palmityl-CoA similar to those of stearyl-CoA used in the earlier experiments in which sterculate inhibition of stearyl-CoA desaturation was seen, it was found that, under these conditions, the desaturation of palmityl-CoA was also inhibited by sterculate (Table III).6 This suggested that the sterculate inhibition observed was related to the concentration of acyl-CoA ester and a subsequent experiment with stearyl-CoA also confirmed that the sterculate inhibition was not evident at lower levels of stearyl-CoA.
Thus, under the conditions of these experiments, the results obtained failed to confirm the contention of Allen et al. (3) that sterculate is a competitive inhibitor of the desaturase system. Inhibition of Desaturase System by Free Fatty Acids-The question of the specificity of sterculate on the desaturase system has been previously examined. Raju and Reiser (5) reported that, under conditions in which sterculate was inhibitory, its methyl mercaptan addition product and oleate were not. Allen et al. strate.
Since the precise micellar concentration of the substrates under the experimental conditions used was not known and the enzyme activity curve appeared kinetically complex, the determination of K, for acyl-CoA esters was not possible. Moreover, the effect of concentration of acyl-CoA esters on any enzymatic reaction is likely to be related to the amount of enzyme preparation wherever acyl-CoA esters become bound to the enzyme protein, as shown by the present experiments with stearyl-CoA at two different microsomal levels (Fig. 1). Consequently, the concentration of acyl-CoA esters required for half-maximal velocity will be variable and related to the amount of such enzyme preparation.
This may be one of the reasons for the variability of the K, value of stearate in the stearate desaturase system described by Allen et al. (3).
6 This experiment showed that the rate of stearyl-CoA desaturation by rat liver microsomes considerably exceeded the rates of palmityl-CoA desaturation (contrast Nakagawa and Uchiyama (23)). This agrees with the findings of Allman, Hubbard, and Gibson (24) which suggest that stearate may be desaturated to a greater extent in the liver than is palmitate and also is in agreement with experiments with rat liver slices in which stearate-W-desaturation was found to exceed that of palmitate-W (Dr. M. Uchiyama, personal communication    Procedure" under linear kinetic conditions. The "refed" group consisted of rats that had been fasted for 2 days and then allowed access to the standard laboratory diet for the 2 subsequent days. In the fat-free dietary group, rats were maintained on a fat-free diet from the age of 14 days to 12 weeks. In the methyl sterculate-fed group, 20 mg of this lipid were fed daily. (3) also concluded that the desaturase system was specifically inhibited by cyclopropene fatty acids since sterculate and malvalate both were inhibitory in vitro. Sterculate was found to be more inhibitory than malvalate while dihydrosterculate, a diene acid mixture (consisting of 9-methyleneoctadec-lo-enoic acid and isomeric IO-methyleneoctadec-S-enoic acid), and linoleic acids were only inhibitory at much higher levels.
On the other hand, Uchiyama,Nakagawa,and Okui (16) reported that unsaturated fatty acids' inhibited the desaturase system in vitro; 7 The possibility that the inhibition was due to the CoA esters of the fatty acids rather than free fatty acids, under the conditions of their experiments, is not eliminated. polyunsaturated fatty acids were found to be more effective in this regard than oleate.
It was implied that this could have a regulatory role for fatty acid desaturation. We re-examined this aspect and confirmed that dihydrosterculate was much less inhibitory than sterculate (Table IV). However, oleate and linoleate were also found to be effective inhibitors, as was phytanate (3,7,11,15-tetramethyl hexadecanoate). These experiments revealed that the inhibitions in vitro observed with sterculate were not specific.
Recently, from an investigation of the effects of free fatty acids on several enzyme activities, we were led to the conclusion that the inhibitory effects of free fatty acids are generally nonspecific and are due to the detergent properties of these inhibitory substances (8). The finding