Carbon Hydroxylation of Alkyltetrahydropyranols: A Paradigm for Spiroacetal Biosynthesis in Bactrocera sp.

Article abstract of DOI:10.1021/ol0069047

(Equation Presented) In a number of Bactrocera species the penultimate step in the biosynthesis of spiroacetals is shown to be the hydroxylation of an alkyltetrahydropyranol followed by cyclization. The monooxygenases that catalyze this side chain hydroxy

Full text of DOI:10.1021/ol0069047

ORGANIC
LETTERS
2001
Vol. 3, No. 3
397-400
Carbon Hydroxylation of
Alkyltetrahydropyranols: A Paradigm for
Spiroacetal Biosynthesis in Bactrocera
sp.
Jeanette E. Stok, Cheng-Shan Lang, Brett D. Schwartz, Mary T. Fletcher,
William Kitching, and James J. De Voss*
Department of Chemistry, UniVersity of Queensland, Brisbane, 4072 Australia
deVoss@chemistry.uq.edu.au
Received November 21, 2000
ABSTRACT
In a number of Bactrocera species the penultimate step in the biosynthesis of spiroacetals is shown to be the hydroxylation of an
alkyltetrahydropyranol followed by cyclization. The monooxygenases that catalyze this side chain hydroxylation show a strong preference for
oxidation four carbons from the hemiketal center, to produce the spiroacetal. The hydroxy spiroacetals observed in Bactrocera appear to
derive from direct oxidation of the parent spiroacetals and not from alternate precursors.
Spiroacetals such as 1 are widely distributed across the insect
kingdom and are often found as volatile components of
various secretions.1 In some cases they exhibit pheromonal
activity, and their structure, stereochemistry, and biological
activity have been actively investigated.2 Until recently,
essentially nothing was known about the biosynthesis of these
compounds. However, the possibility of environmentally
friendly insect control strategies based on disruption of pher-
omone formation has aroused interest in their biosynthesis.
We have previously demonstrated3 that the penultimate
step in the biosynthesis of 1,7-dioxaspiro[5.5]undecane (1)
in the female olive fly, Bactrocera oleae, is the ω-hydroxy-
lation of 6-n-butyltetrahydropyran-2-ol (2) (Scheme 1).
This species is widely distributed throughout southern Europe
and northern Africa. B. cacuminata and B. cucumis, two
species of fruit fly prevalent in the coastal regions of eastern
Australia, also biosynthesize spiroacetals. The major rectal
glandular component of male B. cacuminata is 1,7-dioxaspiro-
[5.5]undecane (1), identical to that released by female B.
oleae.4 B. cucumis, however, produces mainly 2,8-dimethyl-
1,7-dioxaspiro[5.5]undecane (3) as an unequal mixture of
isomers that have been previously synthesized and character-
ized.5 This raised the question whether oxidation of alkyl-
tetrahydropyranols was a general step in the formation of
spiroacetals in fruit flies or whether different species utilized
different biosynthetic pathways.
To test the generality of this oxidative pathway (Scheme
1) we first examined B. cacuminata. The deuterium labeled
precursor, [2H3]-2-n-butyltetrahydropyran-2-ol ([2H3]-2),3 was
Scheme 1
(1) Francke, W.; Kitching, W. Curr. Org. Chem. 2000, 4, 411-431.
(2) Fletcher, M. T.; Kitching, W. Chem. ReV. 1995, 95, 789-828.
(3) Hungerford, N. L.; Mazomenos, B. E.; Konstantopoulou, M. A.;
Krokos, F. D.; Haniotakis, G. E.; Hubener, A.; Fletcher, M. T.; Moore, C.
J.; De Voss, J. J.; Kitching, W. Chem. Commun. 1998, 863-864.
(4) Kitching, W.; Lewis, J. A.; Fletcher, M. T.; Drew, R. A. I.; Moore,
C. J.; Franke, W. Chem. Commun. 1986, 853-854.
10.1021/ol0069047 CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/17/2001

administered to male flies through their diet, and the rectal
glandular contents were extracted and examined by GC-
MS. The parent spiroacetal (1) was significantly [2H]-
enriched with an estimated incorporation level of 30%. This
confirmed that the final step in the biosynthesis of 1 in both
male B. cacuminata and female B. oleae is the oxidation of
2.
or from 2S-4 by C4′ R-hydroxylation, followed in both cases
by cyclization. As the [2H4]-4 synthesized was racemic, the
amount of deuterium incorporation, after administration to
male B. cucumis, should reveal the importance of the
stereochemistry at C-2. If the S configuration at C-2 of 4 is
essential for recognition by the oxidative enzyme that
transforms it into the spiroacetal, selective incorporation of
only the S isomer would produce the same proportional level
of deuterium in all isomers of 3. However, if both 2R and
2S 4 can be oxidized to yield a spiroacetal the amount of
deuterium incorporation into EZ and ZE-3 would be increased
relative to the amount in the EE and ZZ forms, as 2R-4 can
only act as a precursor for the EZ and ZE isomers. We
observed approximately 50% deuterium enrichment of the
EZ and ZE isomers, compared with 10-15% incorporation
into the EE and ZZ isomers. Thus, 2R-4 can be hydroxylated
in B. cucumis and serve as a precursor of EZ and ZE 3.
However, further experiments are necessary to demonstrate
whether this route is the natural pathway to these compounds.
The hydroxy spiroacetals 5 and 7 also occur naturally in
both B. cacuminata and B. oleae (6 is also found in B. oleae).
Plausible biosynthetic routes to these compounds include
direct hydroxylation of 1 or epoxidation of an alkene such
as 8 followed by cyclization (for 5 and 6; Scheme 2). We
To explore whether B. cucumis produces the bis-homo-
logue, 2,8-dimethyl-1,7-dioxaspiro[5.5]undecane (3), via the
same pathway, [2H4]-labeled 2-methyl-6-pentyl-tetrahydro-
pyran-2-ol6 ([2H4]-4) was required. Simple Grignard addition
of n-pentylmagnesium bromide to δ-hexalactone gave race-
mic, unlabeled 4, and deuterium incorporation was ac-
complished via acid-catalyzed exchange of the R-protons,
presumably via a ring-opened form.
[2H4]-4 was administered to male B. cucumis in their diet,
and the rectal gland contents were subsequently extracted
and analyzed by GC-MS. Spiroacetal 3 was significantly
deuterium enriched, thus confirming that the final step in its
formation is the hydroxylation of 4. Therefore, carbon
hydroxylation followed by cyclization appears to be a general
paradigm for spiroacetal biosynthesis in, at least, Bactrocera
sp.
Scheme 2
One enantiomer of each of the EE (2S,6R,8S) and ZZ
(2S,6S,8S) forms and the EZ (2S,6S,8R), ZE (2S,6R,8R)
diastereomers are seen naturally in B. cucumis (Figure 1).5
have previously demonstrated that the most likely biosyn-
thetic route to these hydroxy spiroacetals was hydroxylation
of 1.3 To determine whether B. cacuminata also produced
the hydroxy spiroacetals via hydroxylation of 1 or epoxida-
tion of an appropriate alkene, [2H3]-87 and [2H4]-1 were
administered to B. cacuminata.
For such flies fed [2H3]-8 the level of hydroxy spiroacetals
5 and 6 was observed to increase dramatically and both were
deuterium enriched. As expected, no deuterium was detected
in the peak corresponding to 7. Enantioselective GC-MS
analysis of the labeled diastereomers of 5 and 6 demonstrated
that they were significantly different from those found
naturally. Indeed, 6 is not usually seen in B. cacuminata but
was observed in these extracts and was derived entirely from
deuterated precursor. In contrast, when the hydroxy spiroac-
Figure 1.
The EE and the ZZ-diastereomers differ only in the stereo-
chemistry at the anomeric carbon, both having the S
configuration at the C2 and C8 positions. They should thus
arise from 2S-4 by hydroxylation at the 4′ side chain position
to yield an S alcohol followed by cyclization. The EZ and
ZE isomers again differ only in the stereochemistry of the
anomeric center but have a 2S,8R configuration. Therefore
they may derive either from 2R-4 by C4′ S-hydroxylation
(5) Kitching, W.; Lewis, J. A.; Perkins, M. V.; Drew, R.; Moore, C. J.;
Schurig, V.; Konig, W. A.; Francke, W. J. Org. Chem. 1989, 54, 3893-
3902.
(6) Fletcher, M. T.; Wells, J. A.; Jacobs, M. F.; Krohn, S.; Kitching,
W.; Drew, R. A. I.; Moore, C.; Francke, W. J. Chem. Soc., Perkin. Trans.
1992, 2827-2831.
(7) Compounds 1, 3, 5-8, 10, 12, and 18 were obtained from the Kitching
research group collection.
398
Org. Lett., Vol. 3, No. 3, 2001

etals were analyzed from B. cacuminata fed [2H4]-1, there
was no significant change in the stereoisomers from those
usually seen.8 However, both 5 and 7 were observed to be
deuterium enriched, indicating that in both B. oleae and B.
cacuminata the hydroxy spiroacetals are formed by oxidation
of 1.
The results from feeding [2H3]-8 are consistent with our
hypothesis that a monooxygenase is responsible for the
hydroxylation of the alkyltetrahydropyranol 2 to produce the
spiroacetal 1. The alkene 8 is an analogue of 2 and would
be oxidized by the monooxygenase to yield an epoxide 9,
which could cyclize to the observed hydroxy spiroacetals
(Scheme 2). However, the differences in enantiomeric
excesses and isomer distribution between those derived from
8 and the ones seen in vivo8 agree with the natural hydroxy
spiroacetals arising from direct oxidation of 1.
Having established the intermediacy of alkyltetrahydro-
pyranols in spiroacetal biosynthesis, we wished to investigate
the specificity of the postulated monooxygenase involved
in their hydroxylation. Two possibilities arise as to the
specificity of the oxidation to produce 1 in B. cacuminata.
The first is that the enzyme involved could hydroxylate the
terminal carbon of the alkyl substituent (Scheme 3a), acting
Treatment of the extract with acid equilibrated the two
compounds so that only one remained. We therefore hy-
pothesized that both the E and Z isomers of 13 were being
produced. Similar results were obtained for the spiroacetal
12 produced from 10. As these examples would constitute
the first time that the less stable Z isomer of a simple
monoalkylspiroacetal had been isolated from a natural source,
we investigated this further by synthesis of both isomers of
13. In addition, as both 12 and 13 are generated with a new
stereogenic center we wished to investigate the stereoselec-
tivity of this process and this required racemic and enantio-
merically enriched samples of 13.
(E)-2-Ethyl-1,7-dioxaspiro[5.5]undecane 13 has been found
previously as a component of the cephalic secretions of a
cleptoparasitic bee, but only mass spectral data were
reported.9 The key compound in our synthesis of 13 (Scheme
4) was the TBDPS protected 1-iodopentan-3-ol 14. This (14)
Scheme 4a
Scheme 3
a (i) (a) MeMgBr/CuI, (b)TBDPS-Cl, (c) H2/Pd, (d) Ph3P/I2; (ii)
(a) NaBH4, (b) LiAlH4, (c) TosCl, (d) TBDPS-Cl, (e) NaI; (iii) (a)
LDA, (b) 14, -78 °C; (iv) (a) H+/methanol/H2O, (b) F-; (v)
benzene/CH3COOH.
independently of the length of the carbon chain present. The
second possibility is that the enzyme is specifically designed
to produce a 1,7-dioxaspiro[5.5]undecane and hence hy-
droxylates four carbons from the hemiketal center, regardless
of chain length (Scheme 3b).
To test these hypotheses, tetrahydropyranol precursors 107
and 11 were administered to male B. cacuminata. GC-MS
analysis of the excised gland extracts revealed that a small
amount of the spiroacetals 12 and 13, respectively, were
present. As 10 and 11 were non-natural substrates, the
amount of conversion to the corresponding spiroacetals was
low, approximately 4% of the parent spiroacetal 1, but easily
detectable by GC-MS. The production of 12 and 13 suggests
that the monooxygenase acts four carbons from the hemiketal
center rather than being an ω-hydroxylase that only oxidizes
the terminal methyl group of an alkyl chain.
was simply derived from ethylacetoacetate 15 for the racemic
series and from (R)-(2-benzyloxyethyl)oxirane 16,10 readily
available from (S)-aspartic acid, for the enantioselective
synthesis. The iodide 14 was used to alkylate a known
hydrazone11 (Scheme 4), and acid-catalyzed removal of
protecting groups yielded the lactol 17. Cyclization of 17 in
benzene/acetic acid proceeded slowly to give a mixture of
the E and Z spiroacetals 13, which were characterized by
1H and 13C NMR and GC-MS. Treatment of the mixture
with TFA resulted in equilibration to yield only the more
stable E isomer. The synthetic samples of E and Z spiroac-
etals had GC retention times and MS fragmentation patterns
identical with those of the compounds generated by B.
cacuminata, to which 11 had been administered.
Interestingly, GC-MS analysis of the gland extracts after
administration of 11 revealed two compounds with the mass
spectral characteristics expected of the 2-ethylspiroacetal 13.
(9) Tengo, J.; Bergstrom, G.; Borg-Karlson, A. K.; Groth, I.; Francke,
W. Z. Naturforsch. C. 1982, 37, 376-380.
(10) Frick, J. A.; Klassen, J. B.; Bathe, A.; Abramson, J. M.; Rappoport,
H. Synthesis 1992, 621.
(8) Fletcher, M. T.; Jacobs, M. F.; Kitching, W.; Krohn, S.; Drew, R.
A. I.; Haniotakis, G. E.; Francke, W. Chem. Comm. 1992, 1457-1459.
(11) Mitra, R. B.; Reddy, G. B. Synthesis 1989, 694-698.
Org. Lett., Vol. 3, No. 3, 2001
399

In addition, the enantiomers of both the E and Z isomers
of 13 were well resolved by enantioselective GC (â-
cyclodextrin column) and 2S-13, synthesized from S-aspartic
acid had an ee of 96%. Similarly, we demonstrated that the
predominant isomers of 13 produced by the fly had the 2S
configuration (approximately 84% ee).
2S-13 eluted before the 2R isomer during enantioselective
GC, and this appears to be a general phenomenon for alkyl-
substituted spiroacetals.12 Racemic 127 was similarly re-
solved, and the stereochemistry of the major insect product
once again was 2S (the first eluting isomer). However the
ee (56%) was significantly lower than that for 13 isolated
from B. cacuminata.
present that acts four carbons from a hemiketal center, as
was seen in B. cacuminata. However, feeding 2 to B. cucumis
resulted not in the formation of 1 but in the production of
spiroacetal 18, where oxidation has occurred three carbons
from the hemiketal. This result perhaps reflects the known
preference for monooxygenases to oxidize methylene rather
than methyl moieties because of the differences in CH bond
strengths.13 The stereochemistry of 18 was again determined
by comparison of synthetic standards7 with the isolated
material, using enantioselective GC. This revealed that both
enantiomers of the E and Z diasteromers were formed,
although the 2S isomers greatly predominated (18E:18Z, 1:1;
2S18:2R18, >95:5).
To our knowledge, these are the first disclosures of the
identification of a Z isomer of a monoalkylspiroacetal from
a natural source, and its presence is informative about the
environment in which it is formed. In acidic regions, such
as sections of the digestive tract, the Z isomer would be
anticipated to isomerize to the more stable E form. Thus the
environment in which the spiroacetal is formed by cyclization
of a hydroxyhemiketal is concluded to be only mildly acidic.
The stereoselectivity of the monooxygenase involved in
the oxidation of 4 to produce 3 in B. cucumis was also
investigated. The results of GC-MS analysis of gland
extracts from B. cucumis fed the hemiketals 2, 10, and 11,
are shown in Scheme 5.
The predominant formation of the 2S isomers of the
methyl-substituted spiroacetals 12 and 18 is not surprising
as the naturally occurring 3 has the same configuration.
However, the formation of a small amount of the 2R isomers
of both 12 and 18 suggest that the monooxygenase respon-
sible for this ω-1 hydroxylation is not stereospecific and may
explain the natural occurrence of the EZ and ZE isomers of
3. Alternatively, as discussed above, 4′S hydroxylation of
2R-4 may also give rise to EZ and ZE 3. Further experiments
with labeled enantiopure 4 would be necessary to completely
resolve the biosynthetic route(s) to EZ and ZE 3.
Overall, this work has revealed that the biosynthetic route
we originally demonstrated for 1 in B. oleae presents a
general paradigm for the formation of spiroacetals in, at least,
Bactrocera sp. Thus, an alkyltetrahydropyranol is hydroxy-
lated, and subsequent cyclization yields the spiroacetal.
Hydroxy spiroacetals appear to derive from oxidation of the
parent spiroacetal. The specificity of the monooxygenases
involved in spiroacetal formation seems to conform with the
hypothesis that oxidation occurs four carbons from the
hemiketal center of the precursor in the species investigated,
although the nature of the group oxidized (methylene versus
methyl) may also be important. Future work will focus on
further characterizing the oxidative enzymes implicated in
spiroacetal biosynthesis.
Scheme 5
Acknowledgment. This work was supported by a re-
search grant from the Australian Research Council. The
authors thank Annice Lloyd and Thelma Peek of the
Department of Primary Industries, Queensland and Renee
Denham for her preliminary work on the synthesis of 13.
The oxidation of 10 unsurprisingly resulted in the produc-
tion of 12, as this reaction is essentially the same hydroxy-
lation performed in vivo to form the spiroacetal 3. With 11,
the spiroacetal 13 was again produced. As observed with B.
cacuminata, both 12 and 13 were isolated as a mixture of E
and Z isomers, with predominantly the 2S configuration (12E:
12Z, 56:44; 2S12:2R12, >95:5; 13E:13Z, 70:30; 2S13:2R13,
75:25). These results again suggest a monooxygenase is
Supporting Information Available: Experimental pro-
cedures including administration and analysis of fly glands,
synthetic schemes, and characterization of compounds [2H4]-
4, 11 and 13. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL0069047
(12) Zhang, H.; Fletcher, M. T.; Avery, J. W.; Kitching, W. Tetrahedron
Lett. 1997, 19, 3477-3478.
(13) Frommer, U.; Ullrich, V.; Staudinger, H.; Orrenius, S. Biochim.
Biophys. Acta 1972, 280, 487-494.
400
Org. Lett., Vol. 3, No. 3, 2001