2
W. Wacławczyk-Biedro n´ et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
racemic compound according to the published procedure3 and
attempted to obtain potential products of its enzymatic degrada-
tion. The most likely target for enzymes in this compound is the
primary alcohol, which could be oxidized to an aldehyde or car-
boxylic acid by an alcohol dehydrogenase and aldehyde dehydro-
genase/oxidase. An analogous inactivation pathway has been
determined for bombykol, the sex pheromone of the silkmoth,
Bombyx mori. No degradation involving the other functional
groups—the ketone and the tertiary alcohol has been described
so far. We have therefore subjected 1,3-dihydroxy-3,7-dimethyl-
isomer 3S with 98% de. Removal of the protecting group was per-
formed by transesterification with hydrogen chloride generated
from trimethylsilyl chloride in methanol, which gave the corre-
sponding methyl ester 4 with 63% yield. Deprotection of the diac-
etal by acid hydrolysis with trifluoroacetic acid in water was
unsuccessful. Such problems were previously also reported for
other butanediacetal derivatives of glycolic acid but their cause
1
4
22
has not yet been explained. The ester 4 was then reduced with
lithium aluminum hydride to the triol 5 in 57% yield. Next steps
followed the procedures described previously by Oliver and co-
3
6
-octen-2-one to Swern or Dess–Martin oxidation. However,
workers (Scheme 1) giving (S)-1,3-dihydroxy-3,7-dimethyl-6-
regardless of the oxidation method used the only product that
could be isolated from the reaction mixture was 6-methyl-5-hep-
ten-2-one 1. This ketone was previously also detected during the
GC analysis of the aggregation pheromone of the Colorado potato
octen-2-one 8 with an overall yield of 26%. Enantiomeric excess
of the TBDPS-protected pheromone 7 determined by chiral HPLC
was 98% (Fig. S2). The chemical purity of the final product deter-
mined by GC was >99% and its specific optical rotation was
1
,3
beetle, which implied thermal instability of the pheromone. It
3
+1.6 ± 2.1 (c 0.79, CHCl ).
was also formed when tert-butyldiphenylsilyl-protected precursor
of the pheromone was oxidized with pyridinium chlorochromate.
The pheromone (0.25–2.5 mM) was then incubated with pro-
tein extracts from the antennae (50–200 equiv) or legs (10–
20 equiv) of the Colorado potato beetle for 1–96 h at 22 or 30 °C
in 0.5–1.5 ml of 20 mM sodium phosphate buffer, pH 7.5 with or
3
Decomposition of the pheromone was also reported during
removal of the TBDPS group from the primary alcohol in the final
step of its synthesis by Tashiro and Mori.4 They attributed this
decomposition to the basicity of TBAF and compared this reaction
to the decomposition of fructose under basic conditions.
+
without NAD at equal concentration. Aliquots were extracted with
an equal volume of diethyl ether, the solvent volume was reduced
ca. 20 times and the samples were initially analyzed by TLC. These
experiments indicated that 6-methyl-5-hepten-2-one 1 was the
major degradation product in these reactions, although they usu-
ally required 96 h incubations of the 1 mM pheromone with 200
antenna-equivalents in 1.5 ml of buffer to proceed to near comple-
tion. Such reaction conditions, however, are usually required, when
unlabeled pheromones are used at relatively high concentrations
It should be noted that when 1,3-dihydroxy-3,7-dimethyl-6-
octen-2-one was first described as one of the oxidation products
of geraniol by Pseudomonas incognita, 6-methyl-5-hepten-2-one 1
was also detected as one of these products.15 A degradation path-
way of geraniol by this bacterium was then proposed, which actu-
ally involved several steps analogous to those employed in the
synthesis of 1,3-dihydroxy-3,7-dimethyl-6-octen-2-one from this
2
4
with antennal protein extracts. We then adjusted the conditions
to minimize the background amounts of 6-methyl-5-hepten-2-one
1 in GC analysis and were able to demonstrate the degradation of
(S)-1,3-dihydroxy-3,7-dimethyl-6-octen-2-one 8 also by this tech-
nique. The pheromone was stable during the incubation in a buffer
3
substrate, that is, regioselective epoxidation of the proximal dou-
ble bond, epoxide hydration, and oxidation of the secondary alco-
1
5
hol. 6-Methyl-5-hepten-2-one 1 has also been described as the
degradation product of monoterpene alcohols—geraniol and nerol,
and aldehydes—geranial and neral, in several fungal species:
+
with/without NAD , but was degraded to 6-methyl-5-hepten-2-
1
6
17
Botrytis cinerea,
tum.
(
Penicillium italicum,
In this degradation pathway oxidation of the alcohols
geraniol, nerol) to aldehydes was followed by their decomposi-
tion, which was attributed to a specific enzyme named citral
and Penicillium digita-
one 1 in the presence of antennal protein extracts. The degradation
1
8,19
+
was always faster when NAD was added (Fig. 1). Similar degrada-
tion was also observed with leg extracts. No degradation was
detected when the protein extracts were thermally inactivated
by boiling for 5 min. Very little conversion to 6-methyl-5-hepten-
2-one 1 was seen in reactions with male antennal extracts from
the silkmoth (25 antenna-equivalents) or the nun moth
(Lymantria monacha, 20 antenna-equivalents) with or without
1
9,20
lyase.
Biotransformation of farnesol by several fungal species
(Fusarium culmorum, B. cinerea, Rhodotorula rubra, Rhodotorula
marina) produced an analogous ketone—6,10-dimethyl-5,9-un-
decadien-2-one (geranylacetone).21 It seems therefore that such
ketones are common degradation products of terpenes in
microorganisms.
+
NAD (Table S1).
Enzymes, whose participation in pheromone inactivation has
been established, are quite diverse, but they generally attack par-
ticular functional groups: esters, aldehydes, alcohols, epoxides,
Performing a retrosynthetic analysis we came up with an idea
of using a reverse reaction to synthesize the pheromone from 6-
methyl-5-hepten-2-one 1 by attaching it to the butanediacetal
derivative of glycolic acid 2. Stereoselective ketone aldol reaction
with butanediacetals of glycolic acid was first studied by Ley and
9
sometimes alkyl groups attached to heteroatoms. However, none
of the reactions identified so far involves the breakdown of the car-
bon–carbon backbone, as we have demonstrated for the aggrega-
tion pheromone of the Colorado potato beetle. Moderate
2
2
co-workers. In this reaction the desired configuration at the gen-
erated chiral center can usually be obtained when the two chains
attached to the carbonyl group differ substantially in size and
+
enhancement of the reaction rate by addition of NAD to the crude
protein extracts suggests the participation of a dehydrogenase. We
suspected that the reaction may involve oxidation of the primary
alcohol to an aldehyde. It has been demonstrated previously that
citral decomposes to 6-methyl-5-hepten-2-one 1 and acetaldehyde
via 3-hydroxycitronellal in the presence of high concentration of
6
-methyl-5-hepten-2-one 1 fulfills this requirement. The butane-
diacetal of glycolic acid, (5R,6R)-5,6-dimethoxy-5,6-dimethyl-1,
-dioxanyl-2-one 2, was prepared according to published proce-
4
2
3
dures. 6-Methyl-5-hepten-2-one 1 was then attached to its
enolate generated with lithium bis(trimethylsilyl)amide. The reac-
tion proceeded with good yield (88%) and gave a mixture of two
diastereoisomers with the S and R configuration at the newly
generated chiral center ((3S,5R,6R)-5,6-dimethoxy-5,6-dimethyl-
2
5
amino acids,
which suggested that b-hydroxyaldehydes are
unstable and decompose by retro-aldol reactions to carbonyl com-
pounds. However, it was difficult to envisage a plausible mecha-
nism explaining such
a decomposition in the case of the
3
-[(1S)-1,5-dimethyl-1-hydroxyhex-4-enyl]-1,4-dioxan-2-one 3S
and (3S,5R,6R)-5,6-dimethoxy-5,6-dimethyl-3-[(1R)-1,5-dimethyl-
-hydroxyhex-4-enyl]-1,4-dioxan-2-one 3R) in 1:0.21 ratio (deter-
Colorado potato beetle pheromone, which contains a ketone at
C2. Nevertheless we have incubated (S)-1,3-dihydroxy-3,
7-dimethyl-6-octen-2-one 8 with recombinant equine alcohol
dehydrogenase to see whether enzymatic oxidation of the primary
alcohol may lead to its decomposition. 6-Methyl-5-hepten-2-one 1
1
1
mined by H NMR, Fig. S1). These diastereoisomers were easily
separated by column chromatography on silica gel giving the S