Riochimica et Biophysics Acta, 424 (1976) 296-302 @ Elsevier Scientific Publishing Company, Amsterdam

-

Printed

in The Netherlands

BBA 56729

THE ISOMERIZATION OF’ ‘2,5- AND 9,123OCTADECADIENOIC AN EXTRACT OF RUTYRIVIRRI~ ~IRR~SUI~ VENS *

PILAR TERESA GARCIA a, W.W. CHRISTIE ANDERSON c and RALPH T. HOLMAN c

b> HOWARD

M. JENKIN

ACIDS BY

c, LLOYD

a Centro

de InLlestj~aciones en Ciencias Veterinarjas, fnsfituto National de T~crzofrgia A~r~~ecuaria, Villa Udaondo, Casteiar (Fcia, &.A.) (~~r~eniina), ” ~a~lnah Research Institute, Ayr, (U.K.) and c The fiormel Institute, University of Minnesota, 801 16th Avenue NJ., Austin, Minn. 5591Z (i:.S.A.) (Received

September

16th,

1975)

Summary A cell-free particulate preparation from Butyriuibrio fibrisolvens was used to study the relative rates of isomerization of all cis,cis-methylene-interrupted isomers of octadecdienoic acid. Only two isomers were found to be substrates, the 9,f.Sisomer was isomerized at 41 5 4 ~mol/min per mg protein, and the 2,Sisomer at 11 k 1 gmol/min per mg. The product of the isomerization of the 2,Sisomer had an ultraviolet absorption maximum at 233 nm indicating that it was the 3,Sisomer. The isomerization of the 2,5-isomer was studied in detail. Its rate of isomerization was linear with protein concentration up to 0.047 mg/ml, and was linear with substrate concentration up to 48 FM. The pH optimum was 6.8. Below pH 6, the substrate was also subject to spontaneous isomerization. The inhibition of isomerization of the 9,12-isomer by the other isomers was studied. Those isomers in which the double bonds are close to the carboxyl group were the most effective inhibitors. The preparation was also found capable of hydrogenating the conjugated diene product. from the 2,5isomer to a monoene after prolonged incubation.

Introduction

In our program of study of the effects of double bond position in fatty acids upon biochemic~ phenomena, each biochemical reaction or biological response has shown a unique pattern of response relative to position of double bond(s) [l-7]. Each biological system was unique in discriminating between locations * Presented Italy,

2-7

before

the

September

12th 1974

World as paper

Congress no.

231.

of

the

International

Research

conducted

Society

for

at the

Hormel

Fat

Research,

Institute.

Milan,

297

of cis double bonds, and in many cases the displacement of a double bond system one carbon atom along the chain caused drastic differences in acceptability of the fatty acid as a substrate. A cell-free, particulate preparation of linoleate isomerase (EC 5.2.1.5) has been prepared from Butyriuibrio fibrisoluens by Kepler et al. [8,9] which catalyzes the isomerization of cis,cis-9,12-octadecadienoic acid (linoleic acid) to cis-9,trans-ll-octadecadienoic acid. Of the few unsaturated substrates they tested, only free fatty acids having the cis-9,cis-12 diene configuration were found to be substrates for the linoleate isomerase [lo]. The importance of a double bond in binding the substrate to the enzyme was suggested by observations in which other unsaturated fatty acids inhibited the enzyme activity. We have now tested the substrate specificity of the above enzyme preparation with the complete series of cis,cis-methylene-interrupted octadecadienoic acids, and have found that the 2,5- as well as the 9,12-isomer is isomerized. The inhibitory effects of the other isomers upon both reactions, the structure of the product formed from the 2,5-isomer, and evidence for hydrogenation of the diene product to a trans-monoene by the extract are presented. Experimental

procedures

Preparation of bat terial extract B. fibrisoluens, strain 38, was obtained

from Dr. M. Bryant of the University of Illinois. The pure culture was grown under anaerobic conditions for 16-20 h at 37°C in the non-selective medium of Bryant and Robinson [ll] , harvested and stored at -20°C. The enzyme extract was prepared by a modified procedure of Kepler and Tove [ 121. The frozen cells were thawed and diluted 1 : 8 with phosphate buffer (0.1 M, pH 7). A maximum volume of 5 ml of the cell suspension was disrupted at a time by sonic oscillation for 30 min with a Branson Sonifier Cell Disruptor Model W 140D (20 kHz) equipped with a microtip. A study of the effect of time of sonication revealed that the activity of the extract from fresh cells increased linearly with sonication time up to 30 min, and remained essentially constant to 48 min. After centrifugation at 500 X g to remove cell debris, the supernatant fluid was used as a crude enzyme preparation and was found to be stable for several months when kept at -20” C. Protein determinations were made using the biuret method [ 131. One typical preparation contained 3.0 mg of protein per ml. Substrates

The isomeric octadecadienoic acids or esters were synthesized in this laboratory [14], and the monoenoic acid (or ester) isomers were obtained from F.D. Gunstone [15]. All the fatty acids used as substrates or as inhibitors were tested for purity using thin-layer chromatography on silica gel G or H employing a solvent system of light-petroleum (b.p. 30-60”C)/ethyl ether/acetic acid (85 : 15 : 2, v/v), and all were found uniform. Gas-liquid chromatography was performed at 190°C using a 6 ft X VI inch column packed with 20% ethylene glycol succinate plus 2% phosphoric acid on Gas Chrom P, 80-160 mesh, used in a Beckman GCBA instrument equipped with a hydrogen flame detector.

298

Infrared spectra of liquid films of isolated products were measured using a Perkin-Elmer Model 21 spectrometer. Methyl esters of the isomeric 18 : 2 acids were saponified at room temperature and purity and yield of the free fatty acids were determined by thin-layer chromatography [16]. A stock solution (2 mg/ml) of each fatty acid was prepared in benzene and stored at -20°C. Immediately prior to assay, the solvent was removed from 0.5 ml of the stock solution by evaporation under N2 and the fatty acids were suspended in 5 ml of 1,3-propanediol (0.7 mM final concentration). Method of assay Appropriate amounts of these fatty acid solutions were mixed in 10% 1,3propanediol in potassium phosphate buffer (0.1 M, pH 7.0) in a final volume of 3.0 ml. After preincubation of the fatty acid at 35°C for 5 min, the isomerization was initiated by the addition of the crude enzyme solution. The reaction was observed for 10 min at 35°C. The increasing absorbance at 233 nm characteristic of a conjugated diene was recorded using a Beckman DU spectrophotometer. Rates of reactions were calculated from the slopes of the tracings and compared to the activity upon linoleic acid which was run as a control with each group of samples. Each isomer of 18 : 2 was tested as a possible substrate for linoleate isomerase at two concentrations of enzyme (0.01 and 0.025 ml enzyme extract per tube), and at final concentrations of substrate of 6, 12, 18, 24, 36 and 48 PM. At least five replicates at each concentration of each isomer were measured. Results and Discussion When the assay was performed as described above, the reaction was found to proceed as reported by Kepler et al. [lo]. Under our conditions of assay, the isomerization of linoleic acid took place at a rate of 41 rf- 4 pmol/min per mg protein (n = 20). In addition, we have observed that, under the same conditions, 2,5-octadecadienoic acid was isomerized at the rate of 11 + 1 pmol/min per mg protein (n = 31). None of the other isomers was isomerized to a degree greater than the experimental error of the method. Thus, the crude extract not only isomerizes a cis-9,cis-12 diene system as previously reported [lo], but can also catalyze the movement of a double bond from carboxyl conjugation into diene conjugation. Inhibition of linoleate isomerase by isomers of linoleic acid To test the inhibitory effects of the fatty acid isomers upon the isomerization of linoleic acid, 24 PM cis-9,cis-12-18 : 2 was incubated with 6, 12, 18, 24, 36 and 48 ,uM of each inhibitor. The results at equimolar concentrations of substrate and inhibitor shown in Fig. 1 clearly indicate that the positions of the double bonds in an octadecadienoic acid affect the degree of inhibition of the isomerization of linoleic acid. The 13,16- and 14,17-isomers were poor inhibitors, suggesting that their double bonds probably do not interact with the binding sites to which linoleic acid is bound. The 7,10-, 8,11-, 10,13-, 11,14and 12,15-octadecadienoic acids were almost equally inhibitory, indicating

299

9.12 25

I1 I I I i I POSlTlONS OF DOUBLE BONDS OF THE INHIEITOR

I

I

14.17 h

Fig. 1. The effect of positions of double bonds of cis,cis-methylene-intenupted octadecadienoic acids upon isomerization of linoleic acid. Substrate and inhibitor were present in equimolar amounts (24 PM). The degree of isomerization of linoleic acid in the inhibited samples was compared to linoleic acid alone set at 100%. The effect of concentration of 7.10-octadecadienoic acid inhibitor is shown at concentrations of 6. 12, 18, 24, 36 and 48 pM shown from top to bottom, respectively. The concentration of isomers of linoleic acid used in this experiment was 24 IJM. Standard deviations of values are shown.

mid-chain double bond positions are not discriminated by the enzyme. Among the remaining isomers, the closer the double bonds were to the carboxyl group, the greater was the inhibition of activity of linoleate isomerase. The inhibition of linoleate isomerase by dienoic acids seems to be different from the inhibition by monoenoic acids reported by Kepler et al. [lo], who reported that the 4- and the 12-octadecenoic acids were especially inhibitory. The inhibition by dienoic acids in our experiment was greatest for isomers in which the bonds were near the carboxyl group. In all the cases studied, the inhibition was dependent upon the inhibitor concentration, and this effect is illustrated in Fig. 1 for the 7,10-isomer. As the inhibitor concentration increased, isomerization of the substrate decreased. This was likewise true for all other isomers, data for which are not included in the figure. Isomerization

of 2,5-octadecadienoic

acid

The isomerization of the 2,5-isomer could produce either the 2,4-isomer, which is carboxyl conjugated, or the 3,5-isomer, which has diene conjugation isolated from the carboxyl group. When 24 PM cis-2,cis-5-octadecadienoic acid was incubated with the enzyme extract (0.01 ml), the chromophore produced had an ultraviolet absorption maximum at 233 nm, measured directly in the incubation medium. Its maximum was indistinguishable from known 9,11- and 10,12-octadecadienoic acid isomers, and from the maximum produced when linoleic acid was the substrate, corrections being made for the enzyme and substrate absorptions. The absorption maxima of the geometric isomers of 2,4-decadienoic acid reported by Crombie [17] are all in the region of 257260 nm. Thus, the isomerization of 2,5-octadecadienoic acid yields a 3,5-

300

isomer and not a 2,4-isomer, although energetically the 2,4-isomer might have been expected. To isolate enough product for further characterization, 4 mg of the 2,5isomer was incubated in 270 ml of buffer at pH 7.0 with 2.5 ~1 enzyme solution at 35°C. After 3 min of incubation, the entire mixture was lyophilized and the dry powder was extracted with ether. The total lipid, largely free fatty acids from substrate and product, was examined by ultraviolet spectrometry and found to have the spectrum of a conjugated diene, not conjugated with the cat-boxy1 group. After preparation of methyl esters using BF3 /methanol, gas-liquid chromatography of this product with ethylene glycol succinate/phosphoric acid packing revealed numerous substances which emerged very early and which were traced to the enzyme preparation. The enzyme extract alone yielded no detectable fatty acid methyl esters emerging later than 18 : 2. The crude product of enzymatic isomerization contained 7% of the residual substrate, 65% of a substance with an equivalent chain length value 20.5, 9% with equivalent chain length 20.8, 6% with equivalent chain length 21.05 and 7% with equivalent chain length 21.5. The latter four substances had strong ultraviolet absorption at 233 nm indicating diene conjugation, but their structures were not elucidated because of limited samples. The rate of isomerization of 2,5-octadecadienoic acid was nearly linear with enzyme concentration, up to 0.14 mg protein (Fig. 2A) when the substrate was 24 FM. At 10 and at 25 ~1 of enzyme, the rate of reaction was linear with substrate concentration from 6 to 48 I_IM (Fig. 2B). These studies were conducted at pH 7.0 and measurements were made directly upon the aqueous system. At the pH of the routine assay used, the spontaneous isomerization was negligible. A study of the effect of pH upon the isomerization of 2,5-18 : 2 revealed that acid conditions also induce isomerization (Fig. 2C). Substrate incubated with or without enzyme and then subjected to acid extraction and to methylation in HCl/methanol was nearly completely isomerized to three products, the

mq Protam 0.3

[SlaIO’ 30

s

6

PH



Fig. 2. (A) Relationship of enzyme concentration to the rate of isomerization of 2.5~octadecadienoic acid (24 PM). (B) Relationship of substrate concentration to rate of isomerization of 2,5-octadecadienoic acid using 10 ~1 of enzyme preparation. (C) Effect of pH on isomerase activity with 24 IJM 2.5-18 : 2 as substrate and 10 ~1 enzyme preparation incubated at 35°C.

301

-100

2

3

A POSITION

4 IN

5

6

8

18.1 INHIBITOR

Fig. 3. The effects of position of the double bond in some cis-octadecenoic tion of cis.cis-2,5_octadecadienoic acid by linoleate isomerase. Concentration was 36 PM, and concentration of inhibitor was 12 ,uM.

isomers upon the isomerizaof 2.5~octadecadienoic acid

major two of which yielded 90% tridecanoic acid after von Rudloff oxidation [ 181 with periodate-permanganate. The ultraviolet absorption maximum was again found to be at 233 nm. Celmer and Salomons [19] showed that trans3,cis-5-dienes absorb at 233 nm, whereas trans-3,trans-5-dienes absorb maximally at 228-229 nm. Thus, the principal product observed in the aqueous incubation or isolated via acid conditions probably was the former isomer. The infrared spectrum of the major product isolated by lyophilization and gas-liquid chromatography and having an equivalent chain length of 20.5, had a doublet at 10.22 and 10.58 pm which also indicates a cis, trans- or trans,cis-conjugated diene system [20]. Thus, by physical criteria, the products formed via acid catalysis and via enzymatic action are identical. Inhibition of the isomerization of the 2,Cisomer The isomers of cis-18 : 1 from A2 to A’ were tested as inhibitors of the isomerization of the 2,5-octadecadienoic acid at 12 I.IM concentration of inhibitor and 36 MM substrate concentration. The results shown in Fig. 3 reveal that the maximum inhibition by monoenes was exerted by the AS isomer, and the A* isomer was also strongly inhibitory. Thus, the isomerase probably is capable of attachment at or near the A* or A’ double bonds of 2,5-18 : 2 in a fashion similar to the attachment of the enzyme at or near the A’ * double bond of linoleic acid. The product of the isomerization of 2,5-18 : 2 was again isolated using lyophilization and extraction without exposure to acid conditions, and was tested as an inhibitor of the isomerization of 9,12-18 : 2. This preparation, : 2, proved to be a very effective inhibitor, for in principally trans-3,cis-5-18 the presence of 24 FM 9,12-18 : 2 substrate, 12 /.IM reduced the isomerization to 38%, 24 PM to 23%, 36 PM to 18% and 48 /_IMto 12%. By comparison, in the study shown in Fig. 1, cis,cis-3,6-18 : 2 was employed as inhibitor at the same concentrations as the substrate and it suppressed the activity to 33%. The effects of 3,6-, 4,7- and 5,8-18 : 2 upon the isomerization of 2,5-18 : 2 were tested. At 24 PM concentrations of both substrate and inhibitor, 3,6-18 :

302

2 suppressed the isomerization of 2,518 : 2 to 4070, 4,7-18 : 2 to 37% and 5,8-18 : 2 to 32%. By comparison, the isomerization of 9,12-18 : 2 was suppressed to 43, 48 and 5870, respectively. Hydrogenation of the 3,5-octadecadienoic acid to a monoene 8 mg of 2,518 : 2 were incubated with 5.0 ml of enzyme preparation in 540 ml of buffer at pH 7.0 and 35°C and equal portions of the reaction mixture were taken at 0, 3 and 30 min. They were immediately frozen, lyophilized and extracted with ether. The ultraviolet and infrared spectra of the three preparations of free fatty acids were examined. The maximum at 230 nm decreased with time of incubation. The infrared spectra indicated the progressive development of an absorption maximum at 10.8 ym characteristic of an isolated trans double bond. The extended reaction time permitted a significant reduction of the 3,5-conjugated double bond system to an isolated axons-monoene indicating that the extract contains a third enzymatic activity related to the hydrogenation of polyunsaturated acids. The structure of the product has not been identified but by analogy with the product from the 9,11-isomer, it might be expected to be largely the trans-3-isomer. Acknowledgment This research was supported in part by grants from the National Institutes Health (HE 10489 and HE 08214), by the AID Program of the Department State, U.S.A., and The Hormel Foundation.

of of

References 1

Holman,

2

Reitz,

3

Jenkin,

R.T., R.C.,

Egwin. Lands,

P.O.

and

W.E.M.,

Christie,

Christie,

W.W. W.W.

(1969)

and

J. Biol.

Holman,

Chem.

R.T.

H.M.,

Anderson,

L.E..

Holman,

R.T..

Ismail,

LA.

H.M.,

Anderson,

L.E.,

Holman,

R.T.,

Ismail,

I.A.

244.

(1968)

and

1149-1151

J. Biol.

Gunstone,

Chem.

F.D.

243,

(1969)

2241-2246

J. Bacterial.

98,

1026-1029 4

Jenkin, 59.

and

Gunstone,

F.D.

(1970)

Exp.

Cell

Res.

l-5

5

Jensen,

R.G..

6

Chang.

H.C.,

7

Heimermann.

Gordon, Janke,

D.T.,

Heimermann,

J., Hirsch,

W.H.,

F. and

Holman,

W. and

R.T..

Holman,

R.T.

R.T. (1973)

Holman, Gordon,

D.T..

(1972)

Biochim.

Kowalyshyn,

Lipids

Biophys.

D.E.

and

7, 738-741 Acta

Jensen,

306,

21-25

R.G.

(1973)

Lipids

8, 45-47 8

Kepler,

C.R.,

9 Kepler, 10 Kepler,

C.R. C.R.,

11

Bryant,

M.P.

12

Kepler,

C.R.

105-109, 13

Kepler.

C.R.

Christie,

15

Gunstone,

16

Stahl,

and

W.P.

Tove,

17

Crombie,

18

van

19

Celmer,

W.D.

20

Jackson,

J..

IsmaII,

Thin-Layer

S.B.

(1961)

(1969)

New

S.B.

Holman,

Tove,

S.B.

(1966)

J. Biol.

Chem. 242. 5686-5692 (1970)

3. Biol,

J. Dairy

Methods

Sci.

Chem.

44,

245,

Chem. 241, 1350-1354 3612-3620

1446-1456

in Enzymology

(Lowenstein,

J.M.,

ed.),

Vol.

XIV.

pp.

York

(1957)

New

Methods

in

Enzymology

(Lowenstein,

J.M.,

ed.),

Vol.

III,

pp.

York

R.T. LA.

(1967) (1967)

Chem. Chem.

Phys. Phys.

Chromatography

Lipids Lipids

(Stahl.

1.407-423 1, 209-224

E..

ed.),

2nd

edn.,

PP.

371-372,

Springer-

York

L. (1955)

Rudloff,

Tove,

I.M.

Press,

and

(1969)

New

and

S.B.

Tove,

J.J. and

(1967) J. Bioi.

Press,

and

F.D.

McNeiIl.

S.B.

Robinson,

and

W.W. E.

29,

Tucker, and

Academic

14

Sot.

K.P.,

Tove,

Academic

45&-451,

Verlag,

Hirons. and

J. Chem.

E. (1956) and

Salomons.

Paschke,

229-234

Can. R.F.,

Sot.

1007-1025

J. Chem. LA.

(1953)

Tolberg.

34.1413-1418 J. Am. W.E.,

Chem.

Boyd,

H.M.

Sot. and

75,

3430-3435

Wheeler,

D.H.

(1952)

J. Am.

Oil

Chem.

The isomerization of 2,5- and 9,12-octadecadienoic acids by an extract of Butyrivibrio fibrisolvens.

A cell-free particulate preparation from Butyrivibrio fibrisolvens was used to study the relative rates of isomerization of all cis,cis-methylene-inte...
548KB Sizes 0 Downloads 0 Views