Stabilisation of transition states prior to and following eudesmane cation in aristolochene synthase

Article abstract of DOI:10.1039/b604147g

The mechanistic details of the cyclisation of farnesylpyrophosphate (FPP) by aristolochene synthase (AS) from Penicillium roqueforti have only recently begun to emerge, mainly through the analysis of the reaction products generated by AS-mutants. The reaction proceeds through several intermediates including germacrene A and eudesmane cation. Previous work suggested that the side chain of phenylalanine 178 promoted the conversion of eudesmane cation to aristolochene. We now report that the catalytic function of this residue during the conversion of eudesmane cation to aristolochene is mainly due to the large size of its side chain, which facilitates the hydride shift from C2 to C3, rather than its aromatic character. In addition, F178 appears to control the regioselectivity of the final deprotonation step and, together with F112, helps stabilise the developing positive charge on C1 after the expulsion of pyrophosphate from the substrate. These results complete a screen of likely active-site aromatic residues and establish their respective roles in the conversion of FPP to aristolochene. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b604147g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Stabilisation of transition states prior to and following eudesmane cation in
aristolochene synthase†
Silvia Forcat and Rudolf K. Allemann*
Received 21st March 2006, Accepted 8th May 2006
First published as an Advance Article on the web 25th May 2006
DOI: 10.1039/b604147g
The mechanistic details of the cyclisation of farnesylpyrophosphate (FPP) by aristolochene synthase
(AS) from Penicillium roqueforti have only recently begun to emerge, mainly through the analysis of the
reaction products generated by AS-mutants. The reaction proceeds through several intermediates
including germacrene A and eudesmane cation. Previous work suggested that the side chain of
phenylalanine 178 promoted the conversion of eudesmane cation to aristolochene. We now report that
the catalytic function of this residue during the conversion of eudesmane cation to aristolochene is
mainly due to the large size of its side chain, which facilitates the hydride shift from C2 to C3, rather
than its aromatic character. In addition, F178 appears to control the regioselectivity of the final
deprotonation step and, together with F112, helps stabilise the developing positive charge on C1 after
the expulsion of pyrophosphate from the substrate. These results complete a screen of likely active-site
aromatic residues and establish their respective roles in the conversion of FPP to aristolochene.
Introduction
Sesquiterpene synthases catalyse the cyclisation of the universal
acyclic precursor farnesyl pyrophosphate (FPP, 1) in marine and
terrestrial plants, fungi, bacteria and insects, to generate more
than 300 distinct sesquiterpene products, many of which have
important antibiotic, antifungal or neurotoxic activities. Many
sesquiterpene cyclases share a common protein structure known as
the terpenoid fold.1,2 Plant sesquiterpene cyclases show significant
amounts of homology and may be the products of divergent
evolution from a common precursor.3 Many fungal enzymes on the
other hand show little sequence similarity among themselves and
with their plant counterparts, nevertheless, many appear to share
the same fold.1,4–7 All these enzymes serve as high fidelity templates
for their common substrate and subtly control conformation
and stereochemistry of the intermediates during the cyclisation
reactions.2,8 They represent a masterpiece of evolution in achieving
exquisite chemical and stereochemical selectivity within a common
protein fold for the conversion of a shared substrate to many
Scheme 1 Mechanism of aristolochene synthase.
different products.
Aristolochene synthase (AS) from Penicillium roqueforti is a
monomeric enzyme that catalyses the Mg2+-dependent cyclisation
molecular details of the AS catalysed reaction have only recently
of 1 to the bicyclic sesquiterpene (+)-aristolochene 6 (Scheme 1),
begun to emerge. The crystal structures of epi-aristolochene
the precursor of fungal toxins such as PR-toxin, sporogen-AO1,
synthase from N. tabacum6 and trichodiene synthase from F.
phaseolinone, gigantenone, phomenone and bipolaroxin.9 The
sporotrichioides,1 which, unlike the structure of AS,7 were ob-
tained in the presence of substrate analogues, suggest that the
pyrophosphate group of 1 binds tightly to the enzyme through
two Mg2+ binding sites at the entrance to the active site,10 while
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff,
CF10 3AT, UK. E-mail: allemannrk@cf.ac.uk; Fax: (+44)-29-2087-4030
the aliphatic tail of the substrate binds in a deep hydrophobic
cleft of the enzyme. Recent work supported the proposal11 that
† Electronic supplementary information (ESI) available: total ion chro-
matograms from GC-MS analyses of the products of AS-F178C, AS-
AS catalyses the cyclisation of FPP to aristolochene via the
intermediate S-germacrene A 4 (Scheme 1).12 The conformation
of 1 bound in the enzyme’s active site is known to be a critical
determinant of the reaction pathway leading to attack of C1 of
farnesyl cation 2 by the double bond at C10–C11 subsequent to
F178V, AS-F178Y, AS-F112A and ASF112AF178V. Mass spectral analy-
sis of the sesquiterpene products (aristolochene, germacrene A, valencene,
a-farnesene, b-farnesene, a-selinene, b-selinene, selina-4,11-diene) and of
authentic reference samples. GC-traces for the determination of the
stereochemistry of germacrene A by heat induced Cope rearrangement
and comparison with authentic b-elemenes. See DOI: 10.1039/b604147g
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2563–2567 | 2563
©

View Article Online
the expulsion of pyrophosphate (Scheme 1).8,13 The intermediate
germacrene A then undergoes a further cyclisation to form the
bicyclic eudesmane cation 5 through protonation of the C6–C7
double bond by Y92.12 The positive charge on 5 is stabilised
through interaction with the p-system of the indole ring of W334.14
Successive 1,2-hydride shift and methyl migration followed by
loss of HSi on C8 finally results in the generation of (+)-
aristolochene.15
Previous work suggested that the conversion of eudesmane
cation 5 to aristolochene was facilitated by the bulky aromatic
side chain of F178.16 This residue is ideally placed to stabilise the
developing positive charge on C2/C3 of eudesmane cation (Fig. 1).
Its replacement with valine in AS-F178V led to the accumulation
of the intermediate germacrene A 4 and the production of selinenes
10, 11, and 12 as well as the linear a-8 and b-farnesene 9, which
are the result of erroneous deprotonation (Scheme 2).16 These
observations were interpreted to suggest that F178 fulfilled a dual
role during AS catalysis. The presence of F178 is required for
the conversion of farnesyl cation to germacrene A, most likely
by stabilising the developing positive charge on C11 of 3. In
addition, F178 promotes the conversion of eudesmane cation 5
to aristolochene through either the stabilisation of the cations
on C2 and C3, through interaction of the aromatic phenyl ring
with the positive charge, or through induction of the hydride shift
from C2 to C3 of 5. Here we report experiments that confirm the
function of F178 during AS catalysis and suggest that the large
size of F178 rather than its aromaticity promotes the formation of
aristolochene during AS catalysis.
Scheme 2 Sesquiterpene hydrocarbons produced by AS-mutants.
Results and discussion
In the previously characterised AS-F178V,16 which produced 50%
germacrene A 4 together with smaller amounts of sesquiterpenes
7–12 (Table 1), the smaller, non-aromatic isopropyl group replaced
the bulky benzyl group. To distinguish between the importance
of size and aromaticity of residue 178, cDNAs for AS-F178C,
AS-F178I, AS-F178Y, AS-F178W were generated by site directed
mutagenesis from a cDNA of wild type AS isolated from P.
roqueforti. Mutants were expressed in E. coli BL21(DE3) cells
and purified as previously described.8 The steady-state kinetic
parameters of the mutants were measured by incubating them
with [1-3H]-FPP and determining the amount of the hexane
extractable, tritiated products. The assay was based on that
reported previously12 but optimised to ensure a linear dependence
of the rates on protein concentrations. AS is known to aggregate at
increased concentrations of protein and its activity is only a linear
function of protein concentration for a narrow concentration
range.10 Kinetic assays for AS and all mutant proteins were
carried out within the carefully established linear range. For the
measurements of the steady-state kinetic parameters reported here,
the optimal incubation times were carefully determined to ensure
linearity of turnover with reaction time. Under these optimised
conditions the Michaelis constant (KM = 8.7
1.67 lM) and
the turnover number (kcat = 0.55
0.03 s−1) were increased
Fig. 1 Sketch indicating the proximity of F178 to eudesmane cation 5,
which was modelled into the active site of the X-ray crystal structure of
the apo-enzyme of AS.7
compared to those previously reported (kcat = 0.03 0.01 s−1; KM
=
2.3 0.5 lM).12,17 Overall the catalytic efficiency was increased
Table 1 Relative amounts of hexane extractable products and kinetic parameters for wild-type and mutants of AS
Relative product distribution (%) Kinetic parameters
AS
4
6
91.5
8.8
10.8
43.0
86.4
88.3
7
8
9
10
11
12
KM (lM)
8.7 1.7
5.9 0.3
2.0 0.01
0.9 0.2
9.9 3.3
4.7 2.5
19.7 0.7
46.3 3.8
kcat (s−1 × 103)
k )
cat/KM (M−1 s−1
63,218 12,825
1,678 221
WT
7.5
0.4
20.2
5.2
3.0
2.7
2.1
550.0 30
9.9 1.2
6.3 0.3
4.1 0.5
52.0 1.0
79.0 8.0
2.3 0.3
1.4 0.01
F178C
F178Va
F178I
F178Y
F178W
F112A
F112A-178V
67.2
54.1
54.0
10.7
9.6
12.5
37.1
3.8
9.1
2.7
9.2
5.7
2.1
3,150 151
4,556 1,155
5,253 1,754
16,809 9,102
116.8 11.0
30.2 2.5
53.1
22.3
34.3
40.6
a Data from Forcat & Allemann.16
2564 | Org. Biomol. Chem., 2006, 4, 2563–2567
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
approximately 5-fold under the optimised conditions (kcat/KM
=
is worthy of note that AS-F178C produced increased proportions
of valencene 7 when compared to the wild type enzyme. Since 7
is most likely produced by loss of the proton on C6 of the final
carbocation, F178 must also be involved in preventing erroneous
deprotonation in the last step of the synthesis of aristolochene. No
residue has previously been reported to be involved in this step of
AS catalysis.
The X-ray structure of AS,7 which indicated the proximity of
F178 to C2/C3 of eudesmane cation (Fig. 1), together with the
results presented here suggests that this residue primarily promotes
the hydride shift from C2 to C3. Inspection of the X-ray structure
also indicated that F178 is ideally placed between C3 and C11
of germacrene A (Fig. 3) to first support conversion of farnesyl
cation to germacrene A and then to promote the hydride shift.
63,218 12,825 M−1s−1), relative to that reported previously.12 The
kinetic parameters for the mutant enzymes were also determined
using the same assay conditions. The Michaelis constants of the
mutants were generally slightly reduced when compared to the wild
type enzyme. Surprisingly, the bigger reductions were observed for
non-aromatic side chains. The reduction in the catalytic efficiencies
of the mutants was largely a consequence of reduced turnover
numbers (Table 1).
GC-MS analysis of the hexane extractable products generated
in incubations of the mutant enzymes with FPP indicated that
AS-F178Y and AS-F178W, in which the benzyl group of residue
178 was replaced with another aromatic side chain, generated
products very similar to those of the wild type enzyme in that
the main product was aristolochene (86.4 and 88.3%, respectively)
(Fig. 2 and Table 1). Germacrene A 4 and valencene 7 were formed
as side-products. These products were identified by comparison
of their GC-retention times and of their mass spectra with
those of the products formed by AS or with authentic samples
(ESI†). However, in good agreement with the postulated role of
residue 178 affecting a step after the formation of germacrene
A 4, AS-F178C and AS-F178V produced significantly reduced
amounts of the wild type product aristolochene, but increased
amounts of the intermediate germacrene A. Unlike AS-F178V,
AS-F178C produced only a small amount of b-selinene 11, but
no a-selinene 10 or selina-4,11-diene 12 (ESI†). This observation
together with the product distribution observed for AS-F178V
suggested that F178 also affected a step after the formation of
eudesmane cation 5. The absence of reaction products originating
from deprotonation of C1 may indicate that the hydride shift from
C2 to C3 and the methyl shift from C7 to C2 occur simultaneously,
thereby preventing the formation of an intermediate with a positive
charge on C3.
Fig. 3 Relative orientations of W334, F112 and F178 relative to
germacrene A 4 bound in active site of AS. 4 was docked into the active
site cavity in the X-ray crystal of the apo-enzyme.7
An additional function of F178 during formation of germacrene
A has been described previously.16 The results obtained in the anal-
ysis of AS-F178V suggested that residue 178 could be involved in
the stabilisation of carbocation 3, or the transition state preceding
it, through interactions between the cation and the p-system of the
ring. Alternatively, the size of the ring could be involved in folding
the substrate into a reactive conformation. The observation that
AS-F178I did not however produce a- or b-farnesene suggested
the involvement of other residues in the stabilisation of germacryl
cation 3. Analysis of the structure of AS7 indicated that the benzyl
group of F112 was placed between W334 and F178 to potentially
assist F178 during these early steps of catalysis and to stabilise
the developing charge on 3 (Fig. 3). In order to define the role of
F112 in the catalytic mechanism, AS-F112A was produced and its
products determined. GC-MS analysis revealed the production of
a small amount of germacrene A, in addition to the formation of
the linear products, 8 and 9 (Table 1, ESI†).
It had been previously proposed that F112 helped define
the stereochemistry on C10 of germacrene A 4.7 However,
co-injection of b-elemene generated by the heat-induced Cope
rearrangement12,18 of germacrene A 4 produced by AS-F112A with
a racemic sample of b-elemene, revealed that both wild-type AS
and AS-F112A produced the same stereoisomer of 4 (ESI†). The
presence of a- and b-farnesene suggested that intermediate 2 accu-
mulated during catalysis by AS-F112A, leading to deprotonation
from C4 and C15. Residue 112, therefore, most likely contributes
significantly to the stabilisation of the transition state following
farnesyl cation 2.
Fig. 2 Total ion chromatograms from GC-MS analyses of the hexane
extractable products of FPP utilisation by AS (purple), AS-F178W (blue),
As-F178Y (green) and AS-F178I (red). * is a non-terpene contaminant.
The formation of the selinenes, which result from deprotonation
of eudesmane cation 5 prior to the 1,2-hydride shift from C2
to C3, was not observed when the benzyl group of F178 was
replaced with the 2-butyl group of isoleucine (Fig. 2). These results
strongly suggest that the size of residue 178 rather than its aromatic
character is central to its mechanistic function in the events that
control the conversion of eudesmane cation to aristolochene. It
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2563–2567 | 2565
©

View Article Online
Once 4 is synthesised, W334 and F112 are involved in the
formation of 5. When F112 was replaced by alanine, the reaction
terminated at 4, indicating that W334 was not able to induce the
production of 5 in the absence of F112. Hence, the side products
10, 11 and 12 were not formed. The absence of selinenes in catalysis
by ASF112A suggested that 10, 11, and 12 were genuine products
formed by ASF178V through anomalous deprotonations of 5
rather than products of an acid-catalysed rearrangement of 4
during workup.10 To confirm the proposed roles of both F178
and F112, the double mutant AS-F112AF178V was produced
and its products determined by GC-MS analysis. The products
generated by the double mutant were the same as those produced
by AS-F112A supporting the proposed roles of residues 112 and
178, namely the stabilisation of transition states prior to and
following eudesmane cation 5. As had been observed for ASF178V,
germacrene A was also the major product generated by AS-
F112AF178V. The double mutants did however not produce any
selinenes since in the absence of an aromatic side chain on residue
112 eudesmane cation was not formed. These observations further
confirm that the selinenes produced by ASF178V and ASF178C
are genuine reaction products.10
Berkeley, USA. Valencene was a gift from De Monchy Aromatics
Ltd. Prof. Koenig, University of Hamburg, Germany, provided us
with a racemic sample of a- and b-elemene and standards of a-
and b-selinene and selina-4,11-diene. b-(E)-Farnesene was a gift
from John A. Pickett, FRS, and Lynda Ireland, BBSRC-Institute
for Arable Crops, Rothamsted, UK. (E,E)- and (E,Z)-a-farnesene
were obtained from Silvia Dorn, Institute of Plant Science ETH-
Zurich, Zurich, Switzerland, while aristolochene was generated by
incubation of FPP with WTAS.
Production of AS-mutants
The following mixture was prepared: 2 ll of a solution containing
10 lM of both forward and reverse primers in sterile water, 100 ng
of cDNA template, 1 ll of 10 mM dNTPs, 5 ll of Pfu Polymerase
buffer, 1 ll of Pfu polymerase and 40 ll of sterile water. The
PCR amplification reaction was carried out following the Promega
recommended thermal cycling conditions. The mutagenic primers
were as follows: 5ꢀ-CCGACTTTCGTGTGCATGCGCGCGCAG
and 5ꢀ-CTGCGCGCGCATGCACACGCCAGTCGG for AS-
F178C, 5ꢀ-CCGACTTTCGTGATTATGCGCGCGCAG and 5ꢀ-
CTGCGCGCGCATAATCACGCCAGTCGG for AS-F178I,
5ꢀ-CCGACTTTCGTGTGGATGCGCGCGCAG and 5ꢀ-CTG-
CGCGCGCATCCACACGCCAGTCGG for AS-F178W, and 5ꢀ-
GACTGCTGACCGTTCTCGCCCTTATCGATGATGTTCT-
TG and 5ꢀ-CAAGAACATCATCGATAAGGGCGAGAACG-
GTCAGCAGTC for AS-F112A (altered nucleotides in bold).
Plasmids were purified from overnight cultures using the Qiagen
miniprep kit as described by the manufacturer. All DNA sequences
were confirmed by sequence analysis using an Applied Biosystems
3700 automated DNA sequencer (Functional Genomics Labora-
tory, University of Birmingham).
For protein production, E. coli BL21 (DE3) cells were trans-
formed with the cDNA constructs, grown at 37 ◦C in LB medium
with 0.3 mM ampicillin until they reached an OD600 of 0.6
and protein production was induced with 0.5 mM isopropyl-
b-D-1-thiogalactopyranoside for 3.5 h. Cells were harvested by
centrifugation at 8000 g for 10 min and resuspended in 20 mM Tris,
pH 8, 5 mM EDTA and 5 mM 2-mercaptoethanol. Proteins were
then extracted from the inclusion bodies and purified following
the protocol described previously.12
Conclusions
The site directed mutagenesis study of aristolochene synthase
from P. roqueforti reported here provides further evidence that
germacrene A is an intermediate in the production of aristolochene
from FPP as suggested previously (Scheme 1).12 F178 is a critical
residue that controls the efficient formation of germacrene A and
its conversion to aristolochene through both its large size and the
aromaticity of its side chain. The 1,2-hydride shift that initiates
the conversion of eudesmane cation 5 to aristolochene appears to
require the presence of a bulky side chain on residue 178 implying
that steric effects initiate this reaction. Whether the subsequent
methyl shift from C7 to C2 is also promoted by steric bulk, for
instance of residue Y92, or whether it is induced by proton loss
from C8 is currently being actively investigated.
Experimental
Materials
Characterisation by GC-MS of sesquiterpene products
All chemicals were purchased from Fluka or Sigma. Oligonu-
cleotides were purchased from Alta Biosciences (University of
Birmingham). Pfu polymerase and restriction endonuclease DpnI
were obtained from Promega and New England Biolabs, re-
spectively. Competent E. coli BL21(DE3) and ultracompetent
500 ll incubations were set up containing 25 lM purified protein,
2 mM FPP, 10 mM Tris (pH 7.5), 5 mM MgCl2, 5 mM 2-
mercaptoethanol and 15% glycerol at 25 ◦C. Reactions were termi-
nated by addition of 200 ll of 100 mM EDTA (pH 7.25). Products
were extracted into n-hexane (3 × 3 ml) and passed through
1.5 g silica gel columns. Pooled fractions were concentrated to
10–25 ll by a rotary evaporator and analyzed by GC-MS using
a ThermoQuest Finnigan GC 8000 gas chromatograph equipped
with a 30 m ZB5 column and a MD 1000 mass spectrometer.
Splitless injections of 2 ll were performed at 110 ◦C. The column
te◦mperature was increased from 50 ◦C to 150 ◦C at a rate of
4 C min−1 and maintained for 15 min at 150 ◦C.
R
XL10-Goldꢀ were purchased from Stratagene. The miniprep
kit was purchased from Qiagen and used according to the
supplier’s instructions. HP-Q-sepharose for protein purification
was obtained from GE-Healthcare. Farnesyl pyrophophate was
synthesised according to a published protocol by Dr D. J. Miller
and Mr Fanglei Yu.19 [1-3H]-FPP (20 mC mmol−1) was purchased
either from Sigma or from Tocris and diluted with cold FPP to a
working specific activity of 50 lC lmol−1. Eco-scint scintillation
fluid was obtained from National Diagnostics.
Determination of the absolute configuration of germacrene A
A reference sample of germacrene A from soldier cephalic secre-
tion of a subterranean termite species was provided by Larry Cool,
Forest Products Laboratory, University of Berkeley, California,
The absolute configuration of germacrene A 4 produced by
aristolochene synthase and its mutants was determined using a
2566 | Org. Biomol. Chem., 2006, 4, 2563–2567
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
GC equipped with a 30 m (0.2 mm) heptakis(O-TBDMS-2,3-di-
O-methyl)-b-cyclodextrin (50% in OV17) chiral column. Splitless
injections with an injector temperature of 250 ◦C induced the
Cope rearrangement of the enzymatically produced germacrene
A. The stereochemistry of the resulting b-elemenes was confirmed
by comparison with authentic samples.
References
1 M. J. Rynkiewicz, D. E. Cane and D. W. Christianson, Proc. Natl. Acad.
Sci. USA, 2001, 98, 13543–13548.
2 C. A. Lesburg, J. M. Caruthers, C. M. Paschall and D. W. Christianson,
Curr. Opin. Struct. Biol., 1998, 8, 695–703.
3 J. Bohlmann, G. Meyer-Gauen and R. Croteau, Proc. Natl. Acad. Sci.
USA, 1998, 95, 4126–4133.
4 M. J. Rynkiewicz, D. E. Cane and D. W. Christianson, Biochemistry,
2002, 41, 1732–1741.
5 C. A. Lesburg, G. Zhai, D. E. Cane and D. W. Christianson, Science,
1997, 277, 1820–1824.
6 C. M. Starks, K. Back, J. Chappell and J. P. Noel, Science, 1997, 277,
1815–1820.
7 J. M. Caruthers, I. Kang, M. J. Rynkiewicz, D. E. Cane and D. W.
Christianson, J. Biol. Chem., 2000, 275, 25533–25539.
8 A. Deligeorgopoulou and R. K. Allemann, Biochemistry, 2003, 42,
7741–7747.
9 R. H. Proctor and T. M. Hohn, J. Biol. Chem., 1993, 268, 4543–4548.
10 B. Felicetti and D. E. Cane, J. Am. Chem. Soc., 2004, 126, 7212–7221.
11 D. E. Cane and Y. S. Tsantrizos, J. Am. Chem. Soc., 1996, 118, 10037–
10040.
Steady-state kinetics
Optimal enzyme concentration for the kinetic assays was de-
termined for each mutant, incubating 100 lM FPP (200 lC
lmol−1) with increasing concentrations of protein (from 20 nM
to 3 lM). For the determination of the appropriate incubation
time, a solution of 100 lM FPP (200 lC lmol−1) was assayed
with the maximum concentration of protein that, following the
procedure described above, was found to be in linear relationship
to the formation of products. Kinetic reactions were set up by
incubating increasing concentrations of FPP (200 lC lmol−1) with
the appropriate concentration of each mutant enzyme. All kinetic
studies were carried out at 30 ◦C using the buffer described for
the characterization of products in a final volume of 250 ll. The
reactions were terminated by the addition of 200 ll of 100 mM
EDTA (pH 7.25).
12 M. J. Calvert, P. R. Ashton and R. K. Allemann, J. Am. Chem. Soc.,
2002, 124, 11636–11641.
13 M. J. Calvert, S. E. Taylor and R. K. Allemann, Chem. Commun., 2002,
2384–2385.
14 A. Deligeorgopoulou, S. E. Taylor, S. Forcat and R. K. Allemann,
Chem. Commun., 2003, 2162–2163.
15 D. E. Cane, P. C. Prabhakaran, J. S. Oliver and D. B. McIlwaine, J. Am.
Chem. Soc., 1990, 112, 3209–3210.
16 S. Forcat and R. K. Allemann, Chem. Commun., 2004, 2094–2095.
17 D. E. Cane, P. C. Prabhakaran, J. S. Oliver and D. B. Mc Ilwaine, J. Am.
Chem. Soc., 1990, 112, 3209–3210.
Acknowledgements
18 J.-W. d. Kraker, M. C. R. Franssen, A. d. Groot, W. A. Koenig and
H. J. Bowmeester, Plant Physiol., 1998, 117, 1381–1392.
19 V. J. Davisson, A. B. Woodside, T. R. Neal, K. E. Stremler, M.
Muehlbacher and C. D. Poulter, J. Org. Chem., 1986, 51, 4768–4779.
This work was supported by the BBSRC through a studentship
to SF and grant 6/B17177 (RKA). We thank Dr Miller and Mr
Fanglei Yu for the synthesis of FPP.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2563–2567 | 2567
©