Studies on the Inhibition of the Desaturases by Cyclopropenoid Fatty Acids 1 R. JEFFCOAT, Basic Studies Unit, Biosciences Division, Unilever Research, Colworth House, Sharnbrook, Bedford, MK44 1LQ, England, and M.R. POLLARD, Department of Chemistry, The Purdie Building, The University, St. Andrews, KY16 9ST, Scotland ABSTRACT
ase, has led to three theories of cyclopropenoid inhibition:
Unwashed rat liver microsomes were used to study the inhibition of the A 6 and A9 desaturases by cyclopropenoid fatty acids with the ring structure about the 9,10 or 6,7 carbon atoms. The 9,10 cyclopropenoid acid (sterculic a c i d ) i s shown to be an effective inhibitor of only A9 desaturase and then only in the presence of MgC12 and coenzyme A (presumably due to the formation of sterculoyl-CoA). Two 6,7 cyclopropenoid acids of different chain lengths showed no marked inhibition of either the A6 or A9 desaturase. By the use of [3H]sterculic acid, it has been shown that under conditions of high inhibition of the A9 d e s a t u r a s e the inhibitor is not covalently attached to the enzyme at any point. This disproves older ideas on the mechanism of inhibition that assumed reaction between the cyclopropenoid ring and sulphydryl groups on the enzymes.
(a) The effect is nonspecific and is due to a detergent-like effect on the enzyme (6). (b) The effect is not specific for the desaturase enzyme that converts the form of "activated" stearic acid that the tissue forms from added stearic acid, but the effect takes place at some transfer stage of the activated stearic acid to the enzyme (5). (c) There are two mechanisms for the formation of oleic acid, one involving the direct desaturation of stearoylCoA that is inhibited by sterculic acid and the other an undefined mechanism that produces oleic acid by chain elongation of some shorter chain unsaturated acid (4).
The outstanding questions are therefore (a) in what form does sterculic acid act as inhibitor, (b) is this inhibition demonstrated at lower concentrations than other fatty acids, (c) INTRODUCTION does the cyclopropenoid group have to be at When sterculic acid [8-(2-octyl-l-cyclopro- the same position in the chain as the double penyl) octanoic acid] is included in the diet of bond to be introduced into the true substrate, hens (1), pigs (2), and cows (3), there is an (d) is the A 6 desaturase that normally converts accumulation of stearic acid and an apparent linoleic acid to ~/-linolenic acid inhibited by 6,7 loss of oleic acid both in tissues and in the or 9,10 cyclopropenoid acids, and (e) is the lipids of egg yolk and milk. This compositional inhibitory cyclopropenoid fatty acid covalently change is due to a marked inhibition of the A9 bonded to the desaturase enzyme? desaturase of the tissues that converts stearic It is the purpose of this paper to try to acid to oleic acid and utilizes the acyl-CoA answer these questions. thiolesters as substrate. Inhibition of this enzyme also occurs with the rat (4) and the MATERIALS AND METHODS single-celled alga, Chlorella vulgaris (5). There is one singular feature of the inhibition that has Substrates and Inhibitors never been satisfactorily explained and that is 1-[14C]-palmitic, stearic, linoleic, and t~that though sterculic acid inhibits the con- linolenic acids were obtained from the Radioversion of stearic to oleic acid where the start- c h e m i c a l Centre, Amersham, Bucks. ing material is labeled stearic acid, no inhibition 1-[ 14C].stearoyl_CoA and 1-[ 14C]_palmitoyl. is observed when labeled acetate is the pre- CoA were obtained from NEN Chemicals Ltd. cursor (4,5). This, together with the observa- (location, pls) Stearoyl-CoA and palmitoyl-CoA tion that high concentrations of many fatty were obtained from Sigma Chemicals, London. acids s h o w s o m e inhibition of the A9 desatur- Thus urea adduct of methyl sterculate was ICyclopropenoid fatty acids are n a m e d according to the number of carbon atoms in the chain; thus sterculic acid is a C18 cyclopropenoid fatty acid.
donated by A.R. Johnson, C.S.I.R.O. Division of Food Preservation, Ryde, N.S.W., Australia. Free sterculic acid which was released from the
480
FATTY ACYL-CoA DESATURASES
481
TABLE I The Effect of Added Sterculic Acid on the A9 Desaturation of Palmitic Acid, Stearic Acid, Palmitoyl-CoA and StearoyI-CoA in Presence and Absence of the Cofactors for Acyl-CoA F o r m a t i o n a Additions NADH, NADPH ATP, CoA, MgCI2 Sterculic acid (24 laM) NADH, NADPH Sterculic acid (24/~M)
Percentage desaturation of substrate (% inhibition in brackets) Palmitic acid Palmitoyi-CoA Stearic acid Stearoyl-CoA 10.4
7.9
9.6
10.0
(28.0) 3.1
(30.0) 10.5 (0)
(38.5) 1.0
(19.0) 11.3 (0)
aAll incubations were carried out for 15 rain at 30 C in a total volume of 1.25 ml phosphate buffer pH 7.4 containing 0.5 mg microsomal protein from rats fed a high carbohydrate diet. All cofactor concentrations are given in the Materials and Methods. A
adduct by the method of James et al. (5) was determined directly from its weight. Ring labeled sterculic acid was prepared by the method of Van Tilborg (7). The free acid was esterified with diazomethane and stored in benzene at -25 C under nitrogen. The radiochemical purity of the methyl sterculate was checked by thin layer chromatography (TLC) on Silica Gel G in the solvent system, 10% diethyl ether in petroleum ether. 1-[14C] Linoleoyl-CoA was synthesized from the free acid via the N-hydroxysuccinimide ester ( 8 ) b y the method of Al-arif and Blecher (9). The radiochemical purity was determined, after hydrolysis and methylation, by TLC on Silica Gel G in the solvent system, 50% diethyl ether in petroleum ether. Its chemical purity, which was determined from its absorption at 232 and 260 nm, was in good agreement with the reported values of other coenzyme A esters (10). 5-(2-pentyl-l-cyclopropneyl)- and 5-(2undecyl-l-cyclopropenyl)-pentanoic a c i d s were synthesized from the corresponding methyl alk6-ynoates according to the method of Sgoutas and Williams (11,12). The structure of the methylcyclopropenoid esters was confirmed by i n f r a r e d , proton magnetic resonance, and carbon magnetic resonance spectroscopy and the purity of the methyl esters checked by TLC and by gas liquid chromatography of the silver nitrate-methanol adducts (23). Animals
Rats were obtained from the Colworth House stocks and, except where stated, were fed a balanced laboratory diet. In certain experiments, it was advantageous to raise the level of the z~9 desaturase activity by feeding the rats a high carbohydrate diet (13). Subcellular fractionation and protein determinations were carried out as described by
f
,
f
8
12
,
,
2
Percentage desaturation 0 0 & ~me {rain)
16
20
24
FIG. 1. The effect of sterculoyl-CoA on the time course of the A9 desaturase. Incubations contained 86 nmoles stearoyl-CoA, ATP, NADH, NADPH, MgCI2 and CoA (see Materials and Methods) and 2 mg microsomal protein in the absence (.) and presence (A) of 24 nmoles of sterculate. Jeffcoat et al. (13). Assays for A6 and A9 Desaturase
Desaturations using free fatty acids as substrates were, unless otherwise stated, carried out at 30 C using mitochondrial supernatants in t h e p r e s e n c e o f 86 n m o l e f a t t y acid, 0.16/~mole coenzyme A, 17.5/1mole ATP, 1.1 / a m o l e N A D H , 0 . 4 8 /~mole NADPH, 4 / l m o l e magnesium chloride, and 2 5 0 p m o l e potassium phosphate pH 7.4 in total volume of 1.0 ml. The amount of protein varied from 1 . 0 - 5 . 0 mg and the incubation times are given in the text. Incubations using the coenzyme A esters of the fatty acids were carried out in a similar way with NADH and NADPH as the only added cofactors. Determinations of the fatty acid conversions catalyzed by A6 and A9 desaturases were carried out as previously described (13). Gel Electrophoresis
The reversibility of binding of the acyl-CoA to microsomal protein was determined by gel electrophoresis in the presence of sodium dodecylsulphate according to the method of Weber and Osborn (14). The protein bands LIPIDS, VOL. 12, NO. 6
R. JEFFCOAT AND M.R. POLLARD
482 100~
I
I
8C 6C
2C % octivit y
I I 4 8 Stercu|ate concentration (,uM)
//
I 24
FIG. 2. The effect of sterculoyl-CoA concentration on the activity of the induced A9 desaturase. Incubations contained 0.4 mg microsornal protein, 86 nmoles stearoyl-CoA, cofactors as for Figure 1 and were carried out at 30 C for 15 min with a 6 min pre-incubation in the absence of microsomal protein. 100% activity represented 26.3% desaturation/0.4 g protein/ 15 min. TABLE II The Effect of Added Sterculic Acid a Addition
Percentage desaturation o f Linoleic acid LinoleoyI-CoA
NADH, NADPH ATP, CoA, MgCI2 Sterculic acid (24 #M) NADH, NADPH Sterculic acid (24/~M)
l 1.3 (0) 0
9.9 (0) 12.0 (0)
aFor conditions see Table I on the desaturation of linoleic acid and linoleoyl-CoA. bMagnitude of inhibition is shown in parentheses. were stained with Coomassie Brilliant Blue R (Sigma Chemical C o m p a n y , L o n d o n ) and the location of the radioactive material determined by a m o d i f i c a t i o n of the m e t h o d of Tishler and Epstein ( 1 5 ) . A p p r o x i m a t e l y 1 . 5 m m gel sections were digested at 50 C for 45-60 min in a sealed scintillation vial with 0.5 ml 30% hydrogen p e r o x i d e and I drop of 1 8 M ammonia. A f t e r the g e l was cooled, 0.3 ml 4N HC1 was added t o / t h e digested gel which was then c o u n t e d in 15 m l 2% (w/v) butyl-PBD in 2:1 (v/v) t o l u e n e - T r i t o n X-100. RESULTS A N D DISCUSSION
In Table I we show the effect of 24/aM s t e r c u l i c acid on the A9 desaturation of palmitic acid, stearic acid, p a l m i t o y l - C o A , and stearoyl-CoA w h e n incubated in the presence and absence of the: cofactors required for thiotester formation. The desaturation of palmitic acid and o f p a l m i t o y l - C o A is inhibited LIPIDS, VOL. 12, NO. 6
by a b o u t 30% when conditions would allow f o r m a t i o n of p a l m i t o y l - C o A and sterculoylCoA. The time course o f the inhibition is shown in Figure 1. When the c o n d i t i o n s prevent f o r m a t i o n of the acyl-CoA, then desaturation o f palmitic acid is low as expected and the desaturation of p a l m i t o y l - C o A is unaffected. The same is true for stearic acid and stearoylCoA. Clearly n o inhibition occurs when sterculoyol-CoA cannot be formed and hence it is a safe assumption that sterculoyl-CoA is the i n h i b i t o r and n o t sterculic acid. This is in agreem e n t with earlier indications by o t h e r groups (5,16-18). The inhibitory effect of a range of sterculic acid c o n c e n t r a t i o n s is shown in Figure 2. Maximal inhibition is shown at a c o n c e n t r a t i o n of a b o u t 5/aM at a stearoyt-CoA c o n c e n t r a t i o n of 69/aM. This represents an inhibitor-substrate c o n c e n t r a t i o n of 1:14. In our experience, it requires m u c h higher ratios of inhibitor to substrate (of the order of 50-100) w h e n o t h e r fatty acids are tested as inhibitors (e.g., see Table III). Here inhibition appears to be due to a nonspecific fatty acid effect. In any case, the t h e o r y of a detergent effect to a c c o u n t for the m a r k e d inhibition of the A9 desaturase c a n n o t explain the specific effect of sterculoyl C o A shown above when sterculic acid is ineffective. A l t h o u g h it could be argued that the lower c o n c e n t r a t i o n s of sterculoyl-CoA required to inhibit the stearoyl-CoA desaturase reflect a specificity o f this inhibitor, it might also be argued that it reflects the greater denaturating effect of fatty acids c o e n z y m e esters c o m p a r e d with the free fatty acids. This is not likely to be the explanation in this case since J e f f c o a t et al. (24) have shown that c o n c e n t r a t i o n s of fatty acyl-CoA as high as 100/aM do n o t inhibit stearoyl-CoA desaturase. The crude microsomes used to test the inhibition of the A9 desaturase were also used to test the inhibition of the A6 desaturase. In Table II we d e m o n s t r a t e the presence of a A6 desaturase in the preparation by showing conversion of b o t h linoleic acid (provided the cofactors for f o r m a t i o n of linoleoyl-CoA are present) and synthetic linoleoyl-CoA. The results are consistent with the previously r e p o r t e d observations of Brenner (25) that the linoleic acid must first be converted i n t o its c o e n z y m e A ester b e f o r e it can be desaturated. This implies that the substrate for the A6 desaturase is the c o e n z y m e A or ACP derivative of linoleic acid or some specific linoleic acid containing lipid. In Table II we d e m o n s t r a t e the effect o f added sterculic acid (at the level at which it shows a clear inhibition of the A9 desaturase)
FATTY ACYL~oA DESATURASES on the desaturation of linoleate and linoleoylCoA under similar conditions to those in Table I. The time course of the reaction is shown in Figure 3. No inhibition whatsoever can be found under conditions where both linoleoylCoA and sterculoyl-CoA are formed. In Table II1 we show the effect of adding the C12 cyclopropenoid acid with the cyclopropene ring across carbon atoms 6 and 7, on the conversion of ct-linolenic acid (9c, 12c, 1 5 c - 1 8 : 3 ) to the C l 8 tetraenoic acid (6c, 9c, 12c, 1 5 c - 1 8 : 4 ) at a range of inhibitor-substrate molar ratios. No inhibition could be demonstrated except at such high levels of inhibitor to substrate that one would expect a nonspecific fatty acid effect. In Table III we demonstrate that the same is true for the C18 cyclopropenoid fatty acid. Clearly none of the cyclopropenoid fatty acids are capable of exerting any specific inhibition of the A6 desaturase. The possible effect of the C18 6,7 cyclopropenoid fatty acid on other desaturation enzymes is shown in Table IV and again it can be seen that effects are observable only at very high inhibitor-substrate ratios unlike the effect of sterculic acid on the A9 desaturase. Of the limited number of cyclopropenoid fatty acids studied to date, only a few are capable of inhibiting a desaturase enzyme and then only the A9 desaturases. Fogerty et al. (19) and earlier work of the Johnson group (17) have shown that the only effective inhibitors are the following: the C 17 8,9-, C18 9 , 1 0 - a n d C19 10,11-cyclopropenoid fatty acids. Assuming that the active center of t h e A9 desaturase falls against the 9,10 methylene groups of the true substrate, then it would follow that a cyclopropenoid fatty acid having the ring involving either or both the C-9 and C-10 atoms could also interact. It has been demonstrated by Kircher (20) and others (21,22) that cyclopropenoid fatty
483
B !
I
I
I
I
A
6
I. Percentage des~orcl|ion
0
I
8 Time (rain)
I
I
I
I
16
2/.
32
/~0
FIG. 3. The effect of sterculoyl-CoA on the time course of the A6 desaturase. Incubations contained 86 nmoles linoleoyl-CoA, the cofactors as in Figure 1 and 10 mg microsomal protein in the absence (o) and presence (•) of 24 nmoles sterculate. All incubations were carried out at 30 C with a pre-incubation time of 6 min in the absence of microsomal protein. acids, exemplified by sterculic acid, react with thiol groups to form thio-ethers. It has, therefore, been assumed that since the position of the cyclopropene ring in the fatty acid chain is important (19) t h e inhibition by sterculate is by an interaction with an essential sulphydryl group. We t h e r e f o r e utilized 3H-sterculic acid (synthesized by Dr. Koch of our Vlaardingen laboratories) as an inhibitor in order to investigate the nature of the enzyme-inhibitor binding. We demonstrate first (Table V) that the [3H]-sterculic acid is as effective an inhibitor as sterculic acid itself. Under conditions which would give ca. 70% inhibition of the stearoylCoA desaturase, 750 nmoles [3H]-sterculic acid (7.2 nCi/nmole) were incubated with 100 mg of rat liver microsomal protein obtained from rats fed the high carbohydrate diet. The specific activity of the stearoyt-CoA desaturase was 3.35 nmoles oleic acid produced per min/mg microsomal proteim After a 15 min incubation at 30 C, the activity of the enzyme had been reduced to a specific activity of 1.32. The 25-45% (NH4) 2 SO 4 fraction was prepared as
TABLE III The Effect of 5-(2-Pentyl-l-cyclopropenyl)- and 5-(2-Undecyl-l-cyclopropenyi)-Pentanoic Acid on the A6 Desaturation of c~-LinolenicAcida 5-(2-Pentyl-l-cyclopropenyl)of cyclopropenoic acid to substrate (mole/mole)
C o n c e n t r a t i o n ratio
0 0.01 0.10 1.0 10.0 100.0
p e n t a n o i c acid
%Conversion
% Inhibition
35 38.5 40.0 33.5 34.0 16.5
0 0 0 4 2 53
5-(2-Undecyl-l-cyclopropenyl)p e n t a n o i c acid
% Conversion 35 37 42.5 32.5 19.0 4.5
% Inhibition 0 0 0 7.5 41.5 87
a F o r c o n d i t i o n s see e x p e r i m e n t a l s e c t i o n .
LIPIDS, VOL. 12, NO. 6
R. JEFFCOAT AND M.R. POLLARD
484 Of
"z
r
N
..=
G
< g
O
,-k r
o
?
t 'i 00
._= "~
~
g
g,
~o
O
"el
;O [..,
r
O
O
8 t"~
t,
r
g 0o
LIPIDS, VOL. 12, NO. 6
o
t--
described by Shimakata et al. (26) and the dialyzed material subjected to electrophoresis as described by Weber and Osborn (14) or extracted with chloroform-methanol (27). After three extractions with the organic solvent, the pellet contained 62 mg protein and 6 nCi [3H]-sterculate. If it is assumed that 0.8% of the protein is desaturase enzyme of which 60% is inhibited by the covalent binding of 1 mole of sterculic acid per mole of enzyme of mol wt 53,000 (28), then the pellet should have contained at least 40 nCi. From similar calculations based on the specific activity of the sterculic acid and the knowledge that 400 pg of protein were applied to the SDS gels, it can be calculated that the desaturase protein band should contain 0.45 nCi or 1000 dpm which in our system is approximately five times background. This apparent lack of label associated with detergent or chloroform-methanol treated protein cannot be reconciled with a covalent binding of inhibitor to substrate. The only tenable conclusion is that the noncovalent b i n d i n g energy of sterculoyl-CoA to the enzyme is much greater than that of stearoyICoA and the clue to the effectiveness must be either in better fit or in some polarization interaction dependent on the cyclopropenoid ring in the vicinity of the C-9 and C-10 atoms of the chain. That the binding is very strong is indicated by the failure of our own group and the Johnson group to show any reversibility of the inhibition by addition of stearic acid (19). The results all point toward a highly specific and very strong noncovalent attachment of either 8,9-, 9,10-, or 10,11-cyclopropenoid fatty acids that is qualitatively and quantitatively different from the nonspecific fatty acid inhibition postulated by Pande and Meade (6). In conclusion, therefore, we have been able to establish that whereas sterculic acid is not t h e i n h i b i t o r of stearoyl-CoA desaturase, another derivative, probably its coenzyme A ester, is. Furthermore, in the presence of cofactors required to synthesize the coenzyme A ester, inhibition of A9 fatty acyl-CoA desaturase by this C18 cyclopropenoid fatty acid is effected at much lower concentrations than for other fatty acids (see Fig. 2 and Table III). Further, specificity for the inhibition has also been demonstrated by the fact that the "activated" form of sterculic acid is not an inhibitor of the A6 fatty acyl-CoA desaturase. Finally, although it has been suggested that the inhibition by cyclopropenoid fatty acids may occur by covalent interaction with sulphydryl groups on the enzyme, our data provides good evidence that this is not the mode of inhibition o f s t e a r o y l - C o A desaturase by the C18
FATTY ACYL-CoA DESATURASES
485
TABLE V Effect of [ 3H ]-Sterculate on A 9 Desaturase Activity a % Desaturation mg Mierosomal protein
Without sterculate
0.18 0.36 0.54
11.9 23.1 32.6
With sterculate 3.1 7.6 9.6
% Inhibition 74 67 71
aAssays were carried out under the standard conditions described in Materials and Methods using stearoyl-CoA as substrate.
9 , 1 0 - c y c l o p r o p e n o i d fatty acid. The one remaining question still to be answered is one concerning the identity of the target protein. Is this the cyanide sensitive desaturase protein or some other transferase involved in transporting exogenously supplied stearoyl-CoA to the site of desaturation? The answer to this question can only come from binding studies using purified desaturase and sterculoyl-CoA. AC KNOWLEDGMENTS M.R. Pollard is the recipient of an SRC Case Studentship and we are indebted to Professor F. Gunstone and Drs. A.T. James and L.J. Morris for many helpful discussions concerning the work and the preparation of the manuscript. The skilled technical assistance of Mr. P.R. Brawn is also acknowledged. REFERENCES 1. Evans, R.J., J.A. Davidson, and S.L. Brandemer, J. Nutr. 73:282 (1961). 2. Ellis, N.R., C.S. Rothwell, and W.P. Pool, J. Biol. Chem. 92:385 (1931). 3. Brown, W.H., J.W. Stull, and G.H. Stott, J. Dairy Sci. 45:191 (1963). 4. Raju, P.K, and R. Reiser, Biochim. Biophys. Acta 176:48 (1969). 5. James, A.T., P. Harris, and J. Bezard, Eur. J. Biochem. 3:318 (1968). 6. Pande, S.V., and J.F. Meade, J. Biol. Chem. 245:1856 (1970). 7. Van Tilborg, H., J. Labelled Compd. XI:281 (1975). 8. Lapidot, Y., S. Rappoport, and Y. Wolman, J. Lipid Res. 8:142 (1967). 9. Al-arif, A., and M. Blecher, Ibid. 10:344 (1969).
10. Seubert, W., and D. Pappajohn, Biochem. Prep. 7:80 (1960). 11. Williams, J.L., and D.S. Sgoutas, Chem. Phys. Lipids 9:295 (1972). 12. Williams, J.L., and D.S. Sgoutas, J. Org. Chem. 36:3064 (1971). 13. J e f f c o a t , R., P.R. Brawn, and A.T. James, Biochim. Biophys. Acta 431:33 (1976). 14. Weber, K., and M. Osborn, J. Biol. Chem. 244:4406 (1969). 15. Tishler, P.V., and C.J. Epstein, Anal. Biochem. 22:89 (1968). 16. Allen, E., A.R. Johnson, A.C. Fogerty, J.A. P e a r s o n , and F.S. Shenstone, Lipids 2:419 (1967). 17. Johnson, A.R., J.A. Pearson, F.S. Shenstone, and A.C. Fogerty, Nature 214:1244 (1967). 18. Raju, P.K., and R. Reiser, J. Biol. Chem. 242:379 (1967). 19. Fogerty, A.C., A.R. Johnson, and J.A. Pearson, Lipids 7:335 (1972). 20. Kircher, H.W., JAOCS 41:4 (1964). 21. Ory, R.L., and A.M. Altschul, Biochem. Biophys. Res. Commun. 17:12 (1964). 22. Wallenfeis, K., and H. Sund, Biochem. Z. 329:17 (1957). 23. Sneider, E.L., S.P. Loke, and D.T. Hopkins, JAOCS 45:585 (1968). 24. Jeffcoat, R., P.R. Brawn, R. Safford, and A.T. James, Biochem. J. 161:431 (1977). 25. Brenner, R.R., Lipids 6:567 (1971). 26. Shimakata, T., K. Mihara, and R. Sato, J. Biochem. 72:1163 (1972). 27. Folch, J., and M. Lees, J. Biol. Chem. 191:807 (1951). 28. Strittmatter, P., L. Spatz, D. Corcoran, M.J. Rogers, B. Setlow, and R. Redline, Proc. Natl. Acad. Sci. 71:4565 (1974).
[Received November 18, 1976]
LIPIDS, V O L 12, NO. 6
ase, has led to three theories of cyclopropenoid inhibition:
Unwashed rat liver microsomes were used to study the inhibition of the A 6 and A9 desaturases by cyclopropenoid fatty acids with the ring structure about the 9,10 or 6,7 carbon atoms. The 9,10 cyclopropenoid acid (sterculic a c i d ) i s shown to be an effective inhibitor of only A9 desaturase and then only in the presence of MgC12 and coenzyme A (presumably due to the formation of sterculoyl-CoA). Two 6,7 cyclopropenoid acids of different chain lengths showed no marked inhibition of either the A6 or A9 desaturase. By the use of [3H]sterculic acid, it has been shown that under conditions of high inhibition of the A9 d e s a t u r a s e the inhibitor is not covalently attached to the enzyme at any point. This disproves older ideas on the mechanism of inhibition that assumed reaction between the cyclopropenoid ring and sulphydryl groups on the enzymes.
(a) The effect is nonspecific and is due to a detergent-like effect on the enzyme (6). (b) The effect is not specific for the desaturase enzyme that converts the form of "activated" stearic acid that the tissue forms from added stearic acid, but the effect takes place at some transfer stage of the activated stearic acid to the enzyme (5). (c) There are two mechanisms for the formation of oleic acid, one involving the direct desaturation of stearoylCoA that is inhibited by sterculic acid and the other an undefined mechanism that produces oleic acid by chain elongation of some shorter chain unsaturated acid (4).
The outstanding questions are therefore (a) in what form does sterculic acid act as inhibitor, (b) is this inhibition demonstrated at lower concentrations than other fatty acids, (c) INTRODUCTION does the cyclopropenoid group have to be at When sterculic acid [8-(2-octyl-l-cyclopro- the same position in the chain as the double penyl) octanoic acid] is included in the diet of bond to be introduced into the true substrate, hens (1), pigs (2), and cows (3), there is an (d) is the A 6 desaturase that normally converts accumulation of stearic acid and an apparent linoleic acid to ~/-linolenic acid inhibited by 6,7 loss of oleic acid both in tissues and in the or 9,10 cyclopropenoid acids, and (e) is the lipids of egg yolk and milk. This compositional inhibitory cyclopropenoid fatty acid covalently change is due to a marked inhibition of the A9 bonded to the desaturase enzyme? desaturase of the tissues that converts stearic It is the purpose of this paper to try to acid to oleic acid and utilizes the acyl-CoA answer these questions. thiolesters as substrate. Inhibition of this enzyme also occurs with the rat (4) and the MATERIALS AND METHODS single-celled alga, Chlorella vulgaris (5). There is one singular feature of the inhibition that has Substrates and Inhibitors never been satisfactorily explained and that is 1-[14C]-palmitic, stearic, linoleic, and t~that though sterculic acid inhibits the con- linolenic acids were obtained from the Radioversion of stearic to oleic acid where the start- c h e m i c a l Centre, Amersham, Bucks. ing material is labeled stearic acid, no inhibition 1-[ 14C].stearoyl_CoA and 1-[ 14C]_palmitoyl. is observed when labeled acetate is the pre- CoA were obtained from NEN Chemicals Ltd. cursor (4,5). This, together with the observa- (location, pls) Stearoyl-CoA and palmitoyl-CoA tion that high concentrations of many fatty were obtained from Sigma Chemicals, London. acids s h o w s o m e inhibition of the A9 desatur- Thus urea adduct of methyl sterculate was ICyclopropenoid fatty acids are n a m e d according to the number of carbon atoms in the chain; thus sterculic acid is a C18 cyclopropenoid fatty acid.
donated by A.R. Johnson, C.S.I.R.O. Division of Food Preservation, Ryde, N.S.W., Australia. Free sterculic acid which was released from the
480
FATTY ACYL-CoA DESATURASES
481
TABLE I The Effect of Added Sterculic Acid on the A9 Desaturation of Palmitic Acid, Stearic Acid, Palmitoyl-CoA and StearoyI-CoA in Presence and Absence of the Cofactors for Acyl-CoA F o r m a t i o n a Additions NADH, NADPH ATP, CoA, MgCI2 Sterculic acid (24 laM) NADH, NADPH Sterculic acid (24/~M)
Percentage desaturation of substrate (% inhibition in brackets) Palmitic acid Palmitoyi-CoA Stearic acid Stearoyl-CoA 10.4
7.9
9.6
10.0
(28.0) 3.1
(30.0) 10.5 (0)
(38.5) 1.0
(19.0) 11.3 (0)
aAll incubations were carried out for 15 rain at 30 C in a total volume of 1.25 ml phosphate buffer pH 7.4 containing 0.5 mg microsomal protein from rats fed a high carbohydrate diet. All cofactor concentrations are given in the Materials and Methods. A
adduct by the method of James et al. (5) was determined directly from its weight. Ring labeled sterculic acid was prepared by the method of Van Tilborg (7). The free acid was esterified with diazomethane and stored in benzene at -25 C under nitrogen. The radiochemical purity of the methyl sterculate was checked by thin layer chromatography (TLC) on Silica Gel G in the solvent system, 10% diethyl ether in petroleum ether. 1-[14C] Linoleoyl-CoA was synthesized from the free acid via the N-hydroxysuccinimide ester ( 8 ) b y the method of Al-arif and Blecher (9). The radiochemical purity was determined, after hydrolysis and methylation, by TLC on Silica Gel G in the solvent system, 50% diethyl ether in petroleum ether. Its chemical purity, which was determined from its absorption at 232 and 260 nm, was in good agreement with the reported values of other coenzyme A esters (10). 5-(2-pentyl-l-cyclopropneyl)- and 5-(2undecyl-l-cyclopropenyl)-pentanoic a c i d s were synthesized from the corresponding methyl alk6-ynoates according to the method of Sgoutas and Williams (11,12). The structure of the methylcyclopropenoid esters was confirmed by i n f r a r e d , proton magnetic resonance, and carbon magnetic resonance spectroscopy and the purity of the methyl esters checked by TLC and by gas liquid chromatography of the silver nitrate-methanol adducts (23). Animals
Rats were obtained from the Colworth House stocks and, except where stated, were fed a balanced laboratory diet. In certain experiments, it was advantageous to raise the level of the z~9 desaturase activity by feeding the rats a high carbohydrate diet (13). Subcellular fractionation and protein determinations were carried out as described by
f
,
f
8
12
,
,
2
Percentage desaturation 0 0 & ~me {rain)
16
20
24
FIG. 1. The effect of sterculoyl-CoA on the time course of the A9 desaturase. Incubations contained 86 nmoles stearoyl-CoA, ATP, NADH, NADPH, MgCI2 and CoA (see Materials and Methods) and 2 mg microsomal protein in the absence (.) and presence (A) of 24 nmoles of sterculate. Jeffcoat et al. (13). Assays for A6 and A9 Desaturase
Desaturations using free fatty acids as substrates were, unless otherwise stated, carried out at 30 C using mitochondrial supernatants in t h e p r e s e n c e o f 86 n m o l e f a t t y acid, 0.16/~mole coenzyme A, 17.5/1mole ATP, 1.1 / a m o l e N A D H , 0 . 4 8 /~mole NADPH, 4 / l m o l e magnesium chloride, and 2 5 0 p m o l e potassium phosphate pH 7.4 in total volume of 1.0 ml. The amount of protein varied from 1 . 0 - 5 . 0 mg and the incubation times are given in the text. Incubations using the coenzyme A esters of the fatty acids were carried out in a similar way with NADH and NADPH as the only added cofactors. Determinations of the fatty acid conversions catalyzed by A6 and A9 desaturases were carried out as previously described (13). Gel Electrophoresis
The reversibility of binding of the acyl-CoA to microsomal protein was determined by gel electrophoresis in the presence of sodium dodecylsulphate according to the method of Weber and Osborn (14). The protein bands LIPIDS, VOL. 12, NO. 6
R. JEFFCOAT AND M.R. POLLARD
482 100~
I
I
8C 6C
2C % octivit y
I I 4 8 Stercu|ate concentration (,uM)
//
I 24
FIG. 2. The effect of sterculoyl-CoA concentration on the activity of the induced A9 desaturase. Incubations contained 0.4 mg microsornal protein, 86 nmoles stearoyl-CoA, cofactors as for Figure 1 and were carried out at 30 C for 15 min with a 6 min pre-incubation in the absence of microsomal protein. 100% activity represented 26.3% desaturation/0.4 g protein/ 15 min. TABLE II The Effect of Added Sterculic Acid a Addition
Percentage desaturation o f Linoleic acid LinoleoyI-CoA
NADH, NADPH ATP, CoA, MgCI2 Sterculic acid (24 #M) NADH, NADPH Sterculic acid (24/~M)
l 1.3 (0) 0
9.9 (0) 12.0 (0)
aFor conditions see Table I on the desaturation of linoleic acid and linoleoyl-CoA. bMagnitude of inhibition is shown in parentheses. were stained with Coomassie Brilliant Blue R (Sigma Chemical C o m p a n y , L o n d o n ) and the location of the radioactive material determined by a m o d i f i c a t i o n of the m e t h o d of Tishler and Epstein ( 1 5 ) . A p p r o x i m a t e l y 1 . 5 m m gel sections were digested at 50 C for 45-60 min in a sealed scintillation vial with 0.5 ml 30% hydrogen p e r o x i d e and I drop of 1 8 M ammonia. A f t e r the g e l was cooled, 0.3 ml 4N HC1 was added t o / t h e digested gel which was then c o u n t e d in 15 m l 2% (w/v) butyl-PBD in 2:1 (v/v) t o l u e n e - T r i t o n X-100. RESULTS A N D DISCUSSION
In Table I we show the effect of 24/aM s t e r c u l i c acid on the A9 desaturation of palmitic acid, stearic acid, p a l m i t o y l - C o A , and stearoyl-CoA w h e n incubated in the presence and absence of the: cofactors required for thiotester formation. The desaturation of palmitic acid and o f p a l m i t o y l - C o A is inhibited LIPIDS, VOL. 12, NO. 6
by a b o u t 30% when conditions would allow f o r m a t i o n of p a l m i t o y l - C o A and sterculoylCoA. The time course o f the inhibition is shown in Figure 1. When the c o n d i t i o n s prevent f o r m a t i o n of the acyl-CoA, then desaturation o f palmitic acid is low as expected and the desaturation of p a l m i t o y l - C o A is unaffected. The same is true for stearic acid and stearoylCoA. Clearly n o inhibition occurs when sterculoyol-CoA cannot be formed and hence it is a safe assumption that sterculoyl-CoA is the i n h i b i t o r and n o t sterculic acid. This is in agreem e n t with earlier indications by o t h e r groups (5,16-18). The inhibitory effect of a range of sterculic acid c o n c e n t r a t i o n s is shown in Figure 2. Maximal inhibition is shown at a c o n c e n t r a t i o n of a b o u t 5/aM at a stearoyt-CoA c o n c e n t r a t i o n of 69/aM. This represents an inhibitor-substrate c o n c e n t r a t i o n of 1:14. In our experience, it requires m u c h higher ratios of inhibitor to substrate (of the order of 50-100) w h e n o t h e r fatty acids are tested as inhibitors (e.g., see Table III). Here inhibition appears to be due to a nonspecific fatty acid effect. In any case, the t h e o r y of a detergent effect to a c c o u n t for the m a r k e d inhibition of the A9 desaturase c a n n o t explain the specific effect of sterculoyl C o A shown above when sterculic acid is ineffective. A l t h o u g h it could be argued that the lower c o n c e n t r a t i o n s of sterculoyl-CoA required to inhibit the stearoyl-CoA desaturase reflect a specificity o f this inhibitor, it might also be argued that it reflects the greater denaturating effect of fatty acids c o e n z y m e esters c o m p a r e d with the free fatty acids. This is not likely to be the explanation in this case since J e f f c o a t et al. (24) have shown that c o n c e n t r a t i o n s of fatty acyl-CoA as high as 100/aM do n o t inhibit stearoyl-CoA desaturase. The crude microsomes used to test the inhibition of the A9 desaturase were also used to test the inhibition of the A6 desaturase. In Table II we d e m o n s t r a t e the presence of a A6 desaturase in the preparation by showing conversion of b o t h linoleic acid (provided the cofactors for f o r m a t i o n of linoleoyl-CoA are present) and synthetic linoleoyl-CoA. The results are consistent with the previously r e p o r t e d observations of Brenner (25) that the linoleic acid must first be converted i n t o its c o e n z y m e A ester b e f o r e it can be desaturated. This implies that the substrate for the A6 desaturase is the c o e n z y m e A or ACP derivative of linoleic acid or some specific linoleic acid containing lipid. In Table II we d e m o n s t r a t e the effect o f added sterculic acid (at the level at which it shows a clear inhibition of the A9 desaturase)
FATTY ACYL~oA DESATURASES on the desaturation of linoleate and linoleoylCoA under similar conditions to those in Table I. The time course of the reaction is shown in Figure 3. No inhibition whatsoever can be found under conditions where both linoleoylCoA and sterculoyl-CoA are formed. In Table II1 we show the effect of adding the C12 cyclopropenoid acid with the cyclopropene ring across carbon atoms 6 and 7, on the conversion of ct-linolenic acid (9c, 12c, 1 5 c - 1 8 : 3 ) to the C l 8 tetraenoic acid (6c, 9c, 12c, 1 5 c - 1 8 : 4 ) at a range of inhibitor-substrate molar ratios. No inhibition could be demonstrated except at such high levels of inhibitor to substrate that one would expect a nonspecific fatty acid effect. In Table III we demonstrate that the same is true for the C18 cyclopropenoid fatty acid. Clearly none of the cyclopropenoid fatty acids are capable of exerting any specific inhibition of the A6 desaturase. The possible effect of the C18 6,7 cyclopropenoid fatty acid on other desaturation enzymes is shown in Table IV and again it can be seen that effects are observable only at very high inhibitor-substrate ratios unlike the effect of sterculic acid on the A9 desaturase. Of the limited number of cyclopropenoid fatty acids studied to date, only a few are capable of inhibiting a desaturase enzyme and then only the A9 desaturases. Fogerty et al. (19) and earlier work of the Johnson group (17) have shown that the only effective inhibitors are the following: the C 17 8,9-, C18 9 , 1 0 - a n d C19 10,11-cyclopropenoid fatty acids. Assuming that the active center of t h e A9 desaturase falls against the 9,10 methylene groups of the true substrate, then it would follow that a cyclopropenoid fatty acid having the ring involving either or both the C-9 and C-10 atoms could also interact. It has been demonstrated by Kircher (20) and others (21,22) that cyclopropenoid fatty
483
B !
I
I
I
I
A
6
I. Percentage des~orcl|ion
0
I
8 Time (rain)
I
I
I
I
16
2/.
32
/~0
FIG. 3. The effect of sterculoyl-CoA on the time course of the A6 desaturase. Incubations contained 86 nmoles linoleoyl-CoA, the cofactors as in Figure 1 and 10 mg microsomal protein in the absence (o) and presence (•) of 24 nmoles sterculate. All incubations were carried out at 30 C with a pre-incubation time of 6 min in the absence of microsomal protein. acids, exemplified by sterculic acid, react with thiol groups to form thio-ethers. It has, therefore, been assumed that since the position of the cyclopropene ring in the fatty acid chain is important (19) t h e inhibition by sterculate is by an interaction with an essential sulphydryl group. We t h e r e f o r e utilized 3H-sterculic acid (synthesized by Dr. Koch of our Vlaardingen laboratories) as an inhibitor in order to investigate the nature of the enzyme-inhibitor binding. We demonstrate first (Table V) that the [3H]-sterculic acid is as effective an inhibitor as sterculic acid itself. Under conditions which would give ca. 70% inhibition of the stearoylCoA desaturase, 750 nmoles [3H]-sterculic acid (7.2 nCi/nmole) were incubated with 100 mg of rat liver microsomal protein obtained from rats fed the high carbohydrate diet. The specific activity of the stearoyt-CoA desaturase was 3.35 nmoles oleic acid produced per min/mg microsomal proteim After a 15 min incubation at 30 C, the activity of the enzyme had been reduced to a specific activity of 1.32. The 25-45% (NH4) 2 SO 4 fraction was prepared as
TABLE III The Effect of 5-(2-Pentyl-l-cyclopropenyl)- and 5-(2-Undecyl-l-cyclopropenyi)-Pentanoic Acid on the A6 Desaturation of c~-LinolenicAcida 5-(2-Pentyl-l-cyclopropenyl)of cyclopropenoic acid to substrate (mole/mole)
C o n c e n t r a t i o n ratio
0 0.01 0.10 1.0 10.0 100.0
p e n t a n o i c acid
%Conversion
% Inhibition
35 38.5 40.0 33.5 34.0 16.5
0 0 0 4 2 53
5-(2-Undecyl-l-cyclopropenyl)p e n t a n o i c acid
% Conversion 35 37 42.5 32.5 19.0 4.5
% Inhibition 0 0 0 7.5 41.5 87
a F o r c o n d i t i o n s see e x p e r i m e n t a l s e c t i o n .
LIPIDS, VOL. 12, NO. 6
R. JEFFCOAT AND M.R. POLLARD
484 Of
"z
r
N
..=
G
< g
O
,-k r
o
?
t 'i 00
._= "~
~
g
g,
~o
O
"el
;O [..,
r
O
O
8 t"~
t,
r
g 0o
LIPIDS, VOL. 12, NO. 6
o
t--
described by Shimakata et al. (26) and the dialyzed material subjected to electrophoresis as described by Weber and Osborn (14) or extracted with chloroform-methanol (27). After three extractions with the organic solvent, the pellet contained 62 mg protein and 6 nCi [3H]-sterculate. If it is assumed that 0.8% of the protein is desaturase enzyme of which 60% is inhibited by the covalent binding of 1 mole of sterculic acid per mole of enzyme of mol wt 53,000 (28), then the pellet should have contained at least 40 nCi. From similar calculations based on the specific activity of the sterculic acid and the knowledge that 400 pg of protein were applied to the SDS gels, it can be calculated that the desaturase protein band should contain 0.45 nCi or 1000 dpm which in our system is approximately five times background. This apparent lack of label associated with detergent or chloroform-methanol treated protein cannot be reconciled with a covalent binding of inhibitor to substrate. The only tenable conclusion is that the noncovalent b i n d i n g energy of sterculoyl-CoA to the enzyme is much greater than that of stearoyICoA and the clue to the effectiveness must be either in better fit or in some polarization interaction dependent on the cyclopropenoid ring in the vicinity of the C-9 and C-10 atoms of the chain. That the binding is very strong is indicated by the failure of our own group and the Johnson group to show any reversibility of the inhibition by addition of stearic acid (19). The results all point toward a highly specific and very strong noncovalent attachment of either 8,9-, 9,10-, or 10,11-cyclopropenoid fatty acids that is qualitatively and quantitatively different from the nonspecific fatty acid inhibition postulated by Pande and Meade (6). In conclusion, therefore, we have been able to establish that whereas sterculic acid is not t h e i n h i b i t o r of stearoyl-CoA desaturase, another derivative, probably its coenzyme A ester, is. Furthermore, in the presence of cofactors required to synthesize the coenzyme A ester, inhibition of A9 fatty acyl-CoA desaturase by this C18 cyclopropenoid fatty acid is effected at much lower concentrations than for other fatty acids (see Fig. 2 and Table III). Further, specificity for the inhibition has also been demonstrated by the fact that the "activated" form of sterculic acid is not an inhibitor of the A6 fatty acyl-CoA desaturase. Finally, although it has been suggested that the inhibition by cyclopropenoid fatty acids may occur by covalent interaction with sulphydryl groups on the enzyme, our data provides good evidence that this is not the mode of inhibition o f s t e a r o y l - C o A desaturase by the C18
FATTY ACYL-CoA DESATURASES
485
TABLE V Effect of [ 3H ]-Sterculate on A 9 Desaturase Activity a % Desaturation mg Mierosomal protein
Without sterculate
0.18 0.36 0.54
11.9 23.1 32.6
With sterculate 3.1 7.6 9.6
% Inhibition 74 67 71
aAssays were carried out under the standard conditions described in Materials and Methods using stearoyl-CoA as substrate.
9 , 1 0 - c y c l o p r o p e n o i d fatty acid. The one remaining question still to be answered is one concerning the identity of the target protein. Is this the cyanide sensitive desaturase protein or some other transferase involved in transporting exogenously supplied stearoyl-CoA to the site of desaturation? The answer to this question can only come from binding studies using purified desaturase and sterculoyl-CoA. AC KNOWLEDGMENTS M.R. Pollard is the recipient of an SRC Case Studentship and we are indebted to Professor F. Gunstone and Drs. A.T. James and L.J. Morris for many helpful discussions concerning the work and the preparation of the manuscript. The skilled technical assistance of Mr. P.R. Brawn is also acknowledged. REFERENCES 1. Evans, R.J., J.A. Davidson, and S.L. Brandemer, J. Nutr. 73:282 (1961). 2. Ellis, N.R., C.S. Rothwell, and W.P. Pool, J. Biol. Chem. 92:385 (1931). 3. Brown, W.H., J.W. Stull, and G.H. Stott, J. Dairy Sci. 45:191 (1963). 4. Raju, P.K, and R. Reiser, Biochim. Biophys. Acta 176:48 (1969). 5. James, A.T., P. Harris, and J. Bezard, Eur. J. Biochem. 3:318 (1968). 6. Pande, S.V., and J.F. Meade, J. Biol. Chem. 245:1856 (1970). 7. Van Tilborg, H., J. Labelled Compd. XI:281 (1975). 8. Lapidot, Y., S. Rappoport, and Y. Wolman, J. Lipid Res. 8:142 (1967). 9. Al-arif, A., and M. Blecher, Ibid. 10:344 (1969).
10. Seubert, W., and D. Pappajohn, Biochem. Prep. 7:80 (1960). 11. Williams, J.L., and D.S. Sgoutas, Chem. Phys. Lipids 9:295 (1972). 12. Williams, J.L., and D.S. Sgoutas, J. Org. Chem. 36:3064 (1971). 13. J e f f c o a t , R., P.R. Brawn, and A.T. James, Biochim. Biophys. Acta 431:33 (1976). 14. Weber, K., and M. Osborn, J. Biol. Chem. 244:4406 (1969). 15. Tishler, P.V., and C.J. Epstein, Anal. Biochem. 22:89 (1968). 16. Allen, E., A.R. Johnson, A.C. Fogerty, J.A. P e a r s o n , and F.S. Shenstone, Lipids 2:419 (1967). 17. Johnson, A.R., J.A. Pearson, F.S. Shenstone, and A.C. Fogerty, Nature 214:1244 (1967). 18. Raju, P.K., and R. Reiser, J. Biol. Chem. 242:379 (1967). 19. Fogerty, A.C., A.R. Johnson, and J.A. Pearson, Lipids 7:335 (1972). 20. Kircher, H.W., JAOCS 41:4 (1964). 21. Ory, R.L., and A.M. Altschul, Biochem. Biophys. Res. Commun. 17:12 (1964). 22. Wallenfeis, K., and H. Sund, Biochem. Z. 329:17 (1957). 23. Sneider, E.L., S.P. Loke, and D.T. Hopkins, JAOCS 45:585 (1968). 24. Jeffcoat, R., P.R. Brawn, R. Safford, and A.T. James, Biochem. J. 161:431 (1977). 25. Brenner, R.R., Lipids 6:567 (1971). 26. Shimakata, T., K. Mihara, and R. Sato, J. Biochem. 72:1163 (1972). 27. Folch, J., and M. Lees, J. Biol. Chem. 191:807 (1951). 28. Strittmatter, P., L. Spatz, D. Corcoran, M.J. Rogers, B. Setlow, and R. Redline, Proc. Natl. Acad. Sci. 71:4565 (1974).
[Received November 18, 1976]
LIPIDS, V O L 12, NO. 6
