LETTERS TO THE EDITOR
12,13,16-Trihydroxy-9(Z)-octadecenoic Acid,
A Possible Intermediate in the Bioconversion of Linoleic
Acid to Tetrahydrofuranyl Fatty Acids by
Clavibacter sp. ALA2
Sir:
the method described earlier (7) to isolate pure material for
structure identification. The reaction products were analyzed by
Microbial systems convert unsaturated fatty acids to monohy- GC and MS as described previously (5,6). GC/MS of methyl
doxy-, dihydroxy-, and trihydroxy fatty acids (1). Some of these esters or methyl esters/trimethylsilyloxy ethers (OTMSi) was
hydroxy fatty acid products were reported to have antimicrobial completed as described previously (6).
activities (2–4). In our previous work (5), biotransformation of
We found a small amount of new product, 12,13,16-TOA,
linoleic acid by Clavibacter sp. ALA2 produced a number of existed in the reaction mixture. In our GC analysis of methyl
products, among which 12,13,17-trihydroxy-9(Z)-octadecenoic esters of products, this compound was overlapped with the
acid (THOA) was the predominant metabolite. Two tetrahydro- main product, THOA, and could not be separated. Recently,
furanyl fatty acids (THFA), 12-hydroxy-13,16-epoxy-9(Z)- we analyzed GC/MS of the methyl ester and trimethylsilane
octadecenoic acid (12-hydroxy-THFA) and 7,12-dihydroxy- ether derivatives of the products and found a small peak be-
13,16-epoxy-9(Z)-octadecenoic acid (7,12-dihydroxy-THFA), fore the main product, THOA. Earlier, we dismissed this small
were tentatively identified by gas chromatography/mass spec- peak as a stereoisomer of THOA. Now, we have found that this
trometry (GC/MS) and microchemical techniques (6) and later is a new compound, the missing link of the bioconversion
confirmed by nuclear magnetic resonance (NMR) analyses (7). pathway between linoleic acid and THFA. The mass spectrum
It is interesting to note that the structure of THOA resembles (Fig. 1) with a molecular ion of 560 was interpreted as fol-
those of the plant self-defense substance (4). THFA were also lows; electron ionization mass spectrometry m/z (relative in-
known to have anticancer activity (8). We also identified 12,13- tensity): 529 [M − OCH3]+ (1), 441 [M − CH2CH3 − TMSOH]+
dihydroxy-9(Z)-octadecenoic acid (12,13-DOA) in the reaction (1), 363 [M − C11H18O2CH3] (5), 299 [M − C6H112(OTMS)]+
mixture (7). For the biosynthetic pathway of linoleic acid to (15), 273 [363 − TMSOH]+ (10), 270 (7), 261 [C6H112-
THFA, it appears that we have missed an intermediate between (OTMS)]+ (9), 217 (5), 171 [261 − TMSOH]+ (100), 159 (4),
12,13-DOA and THFA. Now, we have found this missing inter- 147 (18), 131 [C3H6 OTMS]+ (25), 129 (28), 117 (5), 73 (100),
mediate in the reaction mixture. We report here the identifica- 55 (9). The mass spectrum was largely identical to the spec-
tion by GC/MS of this missing intermediate as 12,13,16-trihy- trum of the main product, THOA, except the 131 fragment,
droxy-9(Z)-octadecenoic acid (12,13,16-TOA). We also report which represents a hydroxy group at C16. Therefore the struc-
the optimal conditions for the production of THFA and discuss ture of the new product is identified as 12,13,16-trihydroxy-
a possible biotransformation pathway.
9(Z)-octadecenoic acid.
Strain Clavibacter sp. ALA2 (NRRL B-21660) was grown
This study and previous work (5,6) showed that Clavibac-
at 30°C aerobically in a 125-mL Erlenmeyer flask (shaker at ter ALA2 is especially efficient at oxidizing C-12, -13, and
200 rpm) containing 50 mL of medium as previously reported -17 with hydroxyl groups. To a lesser extent, hydroxyls also
(5). Bioconversions were carried out by adding 0.35 mL occurred at C-7 and -16. Although not proved, it would ap-
linoleic acid to a 24-h-old culture, and the flasks were shaken pear that linoleic acid is converted by strain ALA2 into 12,13-
again at 200 rpm at 30°C for 2–3 d. At the end of this time, DOA first, and then further oxidized to 12,13,16-TOA.
the culture broth was acidified to pH 2 with 6 N hydrochloric (Scheme 1). In a subsequent step, 12,13,16-TOA is cyclized
acid. The culture broth was then extracted with an equal vol- into THFA. In an analogous mechanism, cyclization of the
ume of ethyl acetate and then diethyl ether. The solvent was main product THOA leads to the formation of diepoxy bi-
removed from the combined extracts with a rotary evapora- cyclic fatty acid (7). It remains interesting to resolve the
tor. Crude extracts containing reaction products were sub- linoleic acid biotransformation pathway by strain ALA2.
jected to high-performance liquid chromatography following
In order to find an optimal condition to produce these
THFA, the following variables of the bioconversion of linoleic
acid by strain ALA2 were studied. The effect of pH on the pro-
duction of THFA and THOA was studied using 0.1 M buffer
Paper no. J9902 in JAOCS 78, 1093–1095 (November 2001).
Copyright © 2001 by AOCS Press
1167
JAOCS, Vol. 78, no. 11 (2001)