the remaining 8% being the 9-hydroper- oxide. At both pH's small amounts of hydroxyoctadecadienoic acid accumu- lated during the reaction. This acid from.
Reprinted from
LIPIDS,
"3 '313
Vol. 8, No.5, Pages: 271-276 (1973)
"Purchased by U.S. Department of Agriculture for Official Use."
Dimorphotheca sinuata Lipoxygenase: Formation of 13-L-Hydroperoxy-cis-9,trans-11-0ctadecadienoic Acid from Linoleic Acid H.W. GARDNER. D.O. CHRISTIANSON and R. KLEIMAN, Northern Regional Research Laboratory,' Peoria, Illinois 61604
Lipoxygenase (EC 1.13.1.13) from the seed of Dimorphotheca sinuata oxidized linoleic acid to predominantly 13-L-hydroperoxy-cis-9,trans-ll-octadecadienoic acid. When the reaction proceeded at pH 6.9, the 13-hydroperoxide was the only isomer detected; but at pH 5.1, the 13-isomer was 92% of the total, the remaining 8% being the 9-hydroperoxide. At both pH's small amounts of hydroxy octadecadienoic acid accumulated during the reaction. This acid from the pH 6.9 reaction was analyzed as I 3 -hy d rox y -cis, tra ns-octadecadienoic. The postulate advanced by many workers that dimorphecolic acid, 9-D-hydroxytrans-I a,trans-12-octadecadienoic acid, is biosynthesized via a lipoxygenase product was not proved. Although the product specificity of D. sinuata lipoxygenase is like that of lipoxygenase type I from soybeans, its inactivity at pH 9 demonstrated that it is a novel enzyme.
acidic conditions but is accompanied by positional isomerization of the hydroxyl (6). That lipoxygenase may also be capable of producing a trans,trans isomer was suggested by the presence of 5% trans,trans isomer in the hydroperoxides derived from corn lipoxygenase (3,7). Whether this compound was an artifact or an actual enzyme product is open to speculation. In working wi th D. sinuata lipoxygenase we found, instead of 9-D-specificity, that the oxidation was at the 13-L-carbon. This oxidation specificity is like that of the predominant soybean lipoxygenase activity which yields 13-L-hydroperoxY-cis-9,trans-ll-octadecadienoic acid from linoleic acid (8,9). According to Christopher et aI. (1a,II), there are three lipoxygenase isoenzymes in soybeans of which at least one, lipoxygenase type I, is known to produce only the 13-hydroperoxide (12). The type I enzyme has its optimum activity at pH 9.5 (I a). We report here a lipoxygenase activity that is like soybean lipoxygenase type I in product specificity, but dissimilar in pH optimum.
INTRODUCTION
METHODS
ABSTRACT
Lipoxygenase (EC 1.13.1.13) from Dimorphotheca sinuata, formerly known as D. aurantiaca (I), has not been reported previously. Our research wi th Dimorphotheca was prompted to evaluate the postulate that dimorphecolic acid, the predominant fatty acid in the glycerides of the seed oil (I), is derived from a lipoxygenase product. Others (1-3) have already speculated on the possible involvement of lipoxygenase in the biosynthesis of dimorphecolic acid, 9-D-hydroxy-trans-1 a,trans-12-octadecadienoic acid. Both corn (3) and potatoes (4) contain lipoxygenases that oxidize linoleic acid to a poten tial dimorphecolic acid precursor, 9-D-hydroperoxy-trans-1 a,cis-12-octadecadienoic acid. Hydroperoxide can be converted to hydroxyl with reducing agents of low potential, such as have been proposed for lipoxygenase in wheat dough through oxidation of SH-groups (5). Isomerization of the cis, trans double bond to the more thermodynamically stable trans, trans configuration occurs under lARS, USDA.
Lipoxygenase Assay
A water extract was prepared by homogeniz-
E 0.02
---
.:
E ......... N
o
E 0.01 :::L
4
5
7
6
8
9
pH FIG. I. Effect of pH on Dimorphotheca sinuata lipoxygenase activity. Buffers: ... , acetate; 0, phosphate;., ammonium.
271
H.W. GARDNER, D.O. CHRISTIANSON AND R. KLEIMAN
272
A
B
A1200
B-
e-
1000
800
Freq. [cm.- 1j FIG. 3. IR absorption of cis,trans double bonds of hydroperoxyoctadecadienoic acid (A) and hydroxyoctadecadienoic acid (B), which were isolated from lipoxygenase product mixture. A, 10% in CS 2 ; B, 6% in CS2 using 0.1 mm aCl cell. using an O 2 electrode apparatus, Gilson Oxygraph K-l C. The rate of O 2 uptake in the appropriate endogenous control was subtracted from the maximum rate observed with linoleic acid added. Protein was determined by the method of Lowry et al. (14). Electrophoresis
FIG. 2. Thin layer chromatogram of linoleic acid oxidation products. Solvent: isooctane-ether-acetic acid 50:50:1. A, unreacted linoleic acid; B, hydroperoxyoctadecadienoic acid; C, hydroxyoctadecadienoic acid. B was positive to ferrous thiocyanate spray. ing hexane-defatted D. sinuata seed meal in the proportion of 7 gJ 100 rnl. After centrifugation at 8000 g for 10 min, the resultant supernatant was fractionated with (NH 4 hS04 at 0 C between 33 and 64% of saturation (13). The reaction mixture used to measure activity was 0.28% Tween-20, 8.4 mM linoleic acid (introduced into the reaction vessel as the K salt), 0.096 M buffer, and a quantity of (NH 4 hS04 fraction equivalent to 33 mg seed meal per milliliter. The buffers employed were Na acetate, acetic acid; KH 2 P0 4 , Na2HP04; and NH4 0H, H4 Cl. Activity was measured at 25 C by O 2 uptake LIPIDS, YOLo 8, NO.5
Lipoxygenase was separated on polyacrylamide gel by the disc electrophoretic method (15), with the exception that the separating gel was 5.25% polyacrylamide modified to contain starch for iodometry (16). Before electrophoresis, the following aqueous extracts were prepared: 2 g D. sinuata meal, 1 g soybeans and 180 mg lyophilized corn germ extract per 6 ml water. After centrifugation, the following quantities of supernatant were mixed with an equal volume of 40% sucrose: 30 J.11 D. sinuata, 7 J.11 soy (diluted five-fold with water) and 20 J.11 corn. The samples were then layered over large-pore stacking gels. All other conditions were the same as reported by Guss et al. (16). After electrophoresis, the gels were immersed in a K linoleate (32 mM)-1 % Tween-20 solution at pH 9.2 for 30 min at room temperature to impregnate the gels with linoleate and permit any lipoxygenase with a pH 9 optimum to react. Next the gels were submerged in 0.2 M phosphate buffer (pH 6.4) for 30 min to ensure that isoenzymes having a
DIMORPHOTHECA SINUA TA LIPOXYGENASE
lower pH optimum would react. After incubation the gels were stained with KI according to the method of Guss et al. (16).
C>
-
C> ......
0
•
¢:
.,
"0
§
•
*\,
... ... ..... ...... ... ...=
:z:
C>
.....0 ~
u
C>
::l "0 0
co
en ~
...
.-
en
CD----~.-~
C)
C>~
5. ....
C> CD
I....
0
c:
.§
C)
.....
:z:
c.:>
"0
'u
'"
u
C)
:z:
......
c.:>
I
"0
13 u 0
>,
......
II
:z: c.:>
., .,u'" c:
:a
co
c.:>
en
'0
C>
......
>
CD ......
..
C)~
• ...... ---~"'1
......
0
C>
¢:
......
"0
on
C>
......
......
.,
on on
.~
"' .~.,'"
~
"0
C> C)
!:l
C>
:a
...... co
.,'"c: u .,'" 0
"0 C> CD
Chromatography
..
Thin layer chromatography (TLC) on Silica Gel G plates either 250 IJ. or I mm thick was used to isolate products. The solvent employed to separate nonesterified compounds was redistilled hexane-ether-acetic acid 60:40: I. TLC scrapings of isolated products were slurried with chloroform-methanol 2: I and filtered through a few centimeters of Mallinckrodt 100 mesh silicic acid. The ratio of methyl 9- and 13-hydroxystearates in mixtures was determined by TLC densitometry (3).
C>
...= ---~-- ......= C)
= C>
co C)
CD
~ U
0
>,
>
.
-5 .,
E
'S
~
>.
..c: 0)
E
'5
.....0 E .,; ~
Spectroscopy
Optical rotation was measured by a Bendix Model 1169 automatic polarimeter at 546 nm in a 0.2 dm cell. Gas chromatography-mass spectrometry (GC-MS) was applied as described by Kleiman and Spencer (17). The GC peaks were sampled five times during elution to obtain mass spectra representative of the entire peak. All other spectra were determined as
-0 ~ E
C) C)
Deriva tives
Derivatives were synthesized by methods reported earlier (3). In certain experiments the hydroperoxyoctadecadienoic acid was hydrogenated directly to hydroxystearic acid in methanol with a 10% palladium catalyst on charcoal. Veldink (7) found that hydrogenation without prior use of agents to reduce the hydroperoxide was more reliable in retaining the original position of the oxygenated carbon. Trimethylsiloxy (TMS) derivatives of methyl hydroxystearates and hydroxyoctadecadienoic acid were prepared (17).
••
.."
Product Formation
An ( H4 h S04 fraction of a seed meal extract was prepared as described in the Lipoxygenase Assay Section and then taken up in a volume of either acetate or phosphate buffer. The buffered extract was transferred to a vessel of sufficient size to ensure good surface exposure, and the vessel was flushed with pure oxygen. A volume of K linoleate-Tween-20 solution was added with vigorous stirring which continued during the reaction time of 34 min at 23 C. The final reaction mixture was 7.7 mM linoleic acid, 0.044% Tween-20, 0.1 M buffer and equivalen t to 72 mg of the original seed meal per milliliter. The reaction was terminated by extraction of the fatty acids with triple the volume of chloroform-methanol 2: I.
273 u '
•
C)
......
.... .... '"
~ u~
~~ "'~
'"
'" ~ ::>0.5 '"
'"
~~
=
C>
~.!SUa.UI
•
Paz!I!WJON
S·S ~"O
'u
'"
LIPIDS, YOLo 8, NO.5
274
H.W. GARDNER, D.O. CHRISTIANSON AND R. KLEIMAN
Hydroperoxyoctadecadienoic acid ITlC isolate I
~
~ "z,'"
Hydroxystearic acid ITlC isolate I
!CH 1 N1 Methyl hydroxystearate 4 H Pd Bisltrimethylsilyl] 2, ~ trifluoracetamide
I
Methyl trimethylsiloxystearate
~
~~ ~ Infrared spectroscopy Hydroxyoctadecadienoic acid ITlC isolate I
! !
CH N 1 l
Methyl hydroxyoctadecadienoate
Nuclear magnetic resonance Optical rotation
Gas chromatography·mass spectroscopy Scheme 1. Derivatives from hydroperoxyoctadecadienoic acid.
previously described (3). RESULTS Effect of pH
The pH profile of D. sinuata lipoxygenase (Fig. 1) indicates an optimum at pH 6 and virtually no activity at pH 8-9. Specific Activity
Apparently the lipoxygenase activity in extracts we prepared from the sample of D. sinuata seed was low compared to activities from other plan t species. The highest activity observed at pH 6.9 with the (NH4hS04 isolate was 3.9 x 10- 3 units (Jlmol 02/min)/mg protein. By comparison, our unpublished results with crude corn germ extracts at pH 6.9 yielded specific activities much higher, 0.10-0.75 units per mg protein. Other plant materials have even greater lipoxygenase concentrations, such as potato (1.2 units per mg protein) (4) and soybeans (1.37 units per mg dry matter) (18). Oxidation Products
When linoleic acid was incubated with D. sinuata lipoxygenase, two detectable acids were formed, hydroperoxyoctadecadienoic and hydroxyoctadecadienoic. These acids were separated by TLC (Fig. 2). Hydroperoxyoctadecadienoic acid: When the hydroperoxyoctadecadienoic acid, isolated in 15% yield from a lipoxygenase reaction at pH 6.9, was analyzed by IR spectroscopy (IR), a cis, trans conjugated diene was revealed (Fig. 3A). The isolate was further characterized after LIPIDS, VOL. 8, NO.5
it was derivatized according to Scheme 1. A NMR spectrum of the methyl hydroxyoctadecadienoate derivative was identical to a spectrum of methyl coriolate (19); this similarity indicated the geometry of the dienol was aj3-transI,D-cis. The optical rotation of the methyl hydroxyoctadecadienoate derivative was [a]2s~·g = +7.0° (C, 0.84% hexane), which is the same direction and approximate magnitude of rotation reported for methyl L-coriolate (2). To discover the position of the oxygenated carbon, the hydroperoxyoctadecadienoic acid was hydrogenated either directly or after reduction with NaBH4 (Scheme 1). The methyl hydroxystearate prepared by either method was discovered to be oxygenated 100% at the 13-carbon as determined by GC-MS and TLC. Products were isolated from an oxidation mixture reacted at pH 5.1 to test the acidic side of the pH curve for specificity of the oxidation. A TLC-densitometry analysis of the fully deriva tized hydroperoxyoctadecadienoic acid (Scheme 1) showed that carbon-13 was oxygenated 92 ± 0.5%, a percentage almost identical to the results obtained on the pH 6.9 side of the optimum. The remaining 8% was oxygenated at carbon-9. GC-MS confirmed TLC results but was not quantitated. Hydroxyoctadecadienoic acid: Compared to the hydroperoxide, hydroxyoctadecadienoic acid was a minor product of the lipoxygenasecatalyzed oxidation. A TLC isolate was determined to be a conjugated cis,trans-dienol by IR (Fig. 3B). The TMS derivative of the isolate, trim ethyl sily 1 trim e thy Isiloxyoctadecadi-
275
DIMORPHOTHECA SINUATA LIPOXYGENASE
enoate, was analyzed by GC-MS (Fig. 4). The position of the hydroxyl was determined by GC-MS after hydrogenation, as was done with the hydroperoxide. The 13-hydroxyl was the only isomer detected. Thus this product was 13- hydroxy-cis, trans-oc tadeca-9, ll-dienoic acid and, presumably, would have geometry like the hydroperoxide, i.e., cis-9,trans-11.
5.250/0 Gels
Electrophoresis
A water extract of D. sinuata seed meal was applied to 5.25% polyacrylamide gel for electrophoresis (Fig. 5). The mobility of D. sinuata lipoxygenase compared with that of corn lipoxygenase, a predominantly 9-oxygenating lipoxygenase, and with the most mobile of the soybean isoenzymes, lipoxygenase type I (10). In our hands, the specific stain for lipoxygenase failed to develop in a high percentage of trials with corn and D. sinuata lipoxygenases. The reason for this failure is not apparent but may be due to a requirement for a threshold amount of hydroperoxide, which must be achieved to obtain release of 12 in the gel. By limiting the amount of soybean extract applied to the gel, we have been able to reduce several isoenzyme bands to only one. DISCUSSION
As a result of our experimentation with D. sinuata, we concluded that lipoxygenase may not be responsible for the biosynthesis of dimorphecolic acid, which makes up two-thirds of the fatty acids in the seed oil. This conclusion contradicts the hypothesis of numerous workers (1,3,7,20). However the possibility remains that a different lipoxygenase activity exists in the formative stage of seed ripening when the fatty acids are being biosynthesized. A unique feature of D. sinuata lipoxygenase activity is the production of the 13-L-isomer in high yield at low pH values. In oxidation specificity, it is like soybean lipoxygenase type I (12). Conversely, the D. sinuata enzyme has a pH curve more like lipoxygenases specific for oxidation at the 9-D-
