Stabilisation of eudesmane cation by tryptophan 334 during aristolochene synthase catalysis

Article abstract of DOI:10.1039/b306867f

Analysis of the hydrocarbons produced during catalysis by mutants of aristolochene synthase from Penicillium roqueforti indicated that Trp 334 had a pivotal function for the efficient production of aristoiochene from farnesylpyrophosphate most likely by stabilising the intermediate, eudesmane cation.

Full text of DOI:10.1039/b306867f

Stabilisation of eudesmane cation by tryptophan 334 during
aristolochene synthase catalysis†
Athina Deligeorgopoulou, Susan E. Taylor, Silvia Forcat and Rudolf K. Allemann*
School of Chemical Sciences, University of Birmingham, Edbaston, Birmingham, UK B15 2TT.
E-mail: r.k.allemann@bham.ac.uk; Fax: 44 121-414 4446; Tel: 44 121 414 4359
Received (in Cambridge, UK) 17th June 2003, Accepted 9th July 2003
First published as an Advance Article on the web 22nd July 2003
Analysis of the hydrocarbons produced during catalysis by
mutants of aristolochene synthase from Penicillium roque-
forti indicated that Trp 334 had a pivotal function for the
efficient production of aristolochene from farnesylpyr-
ophosphate most likely by stabilising the intermediate,
eudesmane cation.
germacrene A was followed by protonation at C6 and
cyclisation through electron flow from the C2–C3 double bond
to yield the bicyclic eudesmane cation (3). We have recently
shown that Tyr 92 acts as the active site acid responsible for the
formation of eudesmane cation.9 A hydride shift from C2 to C3
of eudesmane cation followed by a methyl shift from C7 to C2
and deprotonation at C8 results in the formation of 4.
Sesquiterpene cyclases catalyse the cyclisation of the universal
alicyclic precursor farnesyl pyrophosphate (1, FPP) to produce
more than 300 different hydrocarbon skeletons often with high
regio- and stereospecificity.1–3 The solution of the X-ray
structures of 5-epi-aristolochene synthase from Nicotiana
tabacum,4 pentalenene synthase from Streptomyces UC5319,5
trichodiene synthase from Fusarium sporotrichoides6 and
aristolochene synthase from Penicillium roqueforti7 revealed
that they share the same terpenoid fold despite the absence of
any significant sequence similarity.8 Sesquiterpene cyclases
must therefore serve as high fidelity templates for FPP, which
subtly channel conformation and stereochemistry during the
cyclisation reactions. Despite many mechanistic studies of FPP
cyclisation, the molecular details of catalysis by sesquiterpene
cyclases are only just beginning to emerge.6,9–13
Aristolochene synthase (AS) from P. roqueforti is a mono-
meric enzyme that catalyses the Mg2+-dependent cyclisation of
FPP to the bicyclic sesquiterpene aristolochene (4),14 the
precursor of fungal toxins such as PR-toxin, sporogen-AO1,
phaseolinone, gigantenone, phomenone and bipolaroxin.15,16
The correct folding of FPP within the active site of AS has been
shown to be a critical determinant of the reaction pathway.10,13
AS appears to bind the substrate in a conformation favouring
attack of C1 by the double bond at C10–C11 subsequent to
metal triggered expulsion of pyrophosphate (Scheme 1). Recent
work supported the proposal17 that AS catalyses the cyclisation
of FPP to aristolochene via the intermediate S-germacrene A
(2).9 X-Ray crystallography suggested that the formation of
The generation of eudesmane cation from germacrene A is an
energetically demanding step. We have postulated that a
hydrogen bonding network from Arg 200, which is exposed to
the solvent at the top of the active site cleft, through Asp 203 and
Lys 206 to Tyr 92, which is buried deeply in the hydrophobic
environment of the active site, might enhance the acidity of the
phenolic hydroxy group sufficiently to allow protonation of the
C6–C7 double bond.9 However, protonation of this double bond
leads to the formation of an unstable carbocation. Inspection of
the X-ray structure of AS suggested that the p-system of Trp
334 could interact favourably with the positive charge at C3 of
the eudesmane cation.7 This is an ideal strategy for the
stabilisation of high-energy carbocationic intermediates3 since
it prevents the quenching of the positive charge, which might
result from interaction with a nucleophilic intermediate. In
order to address whether Trp 334 is indeed central for the
formation of eudesmane cation we have produced ASW334F,
ASW334L, and ASW334V, in which Trp334 was replaced with
phenylalanine, leucine, and valine, respectively.
cDNAs for ASW334F, ASW334L, and ASW334V were
constructed by site directed mutagenesis from a cDNA of wild-
type AS isolated from P. roqueforti, expressed to high levels in
E. coli BL21(DE3) and purified to homogeneity.13 The steady-
state kinetic parameters of the mutant proteins were measured
by incubating them with [1-3H]-FPP and determining the
amount of hexane-extractable, tritiated products formed.10 The
Michaelis constants for ASW334F, ASW334L, and ASW334V
were similar. Relative to the wild-type enzyme they were
increased between 15- and 33-fold (Table 1). The turnover
number kcat for ASW334F was 0.16 min21, which is an 11-fold
reduction relative to AS. The catalytic efficiency of ASW334F
was reduced approximately 350-fold. A further reduction in kcat
was observed when hydrophobic but non-aromatic residues
replaced Trp334 leading to an overall decrease in the catalytic
efficiencies of ASW334L and ASW334V of more than 4 orders
of magnitude.
When the hexane extractable materials produced by
ASW334F, ASW334L, and ASW334V were further analyzed
by GC-MS (Fig. 1),9,13 the importance of residue 334 of AS for
the production of aristolochene became apparent. AS had been
reported previously to produce ~ 92% aristolochene, ~ 8%
germacrene A and a small amount of valencene, which is
produced by abstraction of a proton from C6 rather than from
C8 in the final step of aristolochene production.9 When Trp 334
was replaced with phenylalanine, the relative production of 2
was increased to 9.4% and that of 4 reduced to 85.5% (Table 1).
2 was identified as germacrene A from its mass spectrum by
comparison with the spectrum in the Wiley library18 and with
that of an authentic sample.‡ Co-injection with an authentic
sample confirmed its identity. While the catalytic competence
of ASW334F was significantly diminished, it still produced
Scheme 1 Formation of 4 via eudesmane cation 3 during AS catalysis.
† Electronic Supplementary Information (ESI) available: GC profiles of co-
injections of germacrene A from ASW334F and ASW334V with authentic
germacrene A; mass spectra of germacrene A produced by ASW334V and
of an authentic sample; mass spectra of valencene produced by ASW334F
and of an authentic sample. See http://www.rsc.org/suppdata/cc/b3/
b306867f/
2162
CHEM. COMMUN., 2003, 2162–2163
This journal is © The Royal Society of Chemistry 2003

Table 1 Kinetic constants for and relative amounts of hydrocarbon products formed by AS, AS W334F, ASW334F, ASW334V and ASW334L
Kinetic dataa
Relative amounts of products
Enzyme
KM/mM
kcat/s21
kcat/KM/s21 M21
2
4
5
Others
AS9
W334F
W334V
W334L
2.3 ± 0.5
66.8 ± 5.8
33.6 ± 5.2
74.8 ± 17
0.03 ± 0.01
13043 ± 2989
37.1 ± 6.3
2.8 ± 0.5
7.5
9.4
95.3
91.5
85.5
4.7
0
0.4
4.7
0
0
0.4
0
(2.65 ± 0.7) 3 1023
(9.15 ± 1.0) 3 1025
(2.25 ± 0.2) 3 1025
0.3 ± 0.1
100
0
0
a Standard errors of mean were determined from a minimum of 3 measurements.
of aristolochene by stabilising the positive charge of eudesmane
cation. When this aromatic residue was replaced with an
aliphatic amino acid, only small amounts of aristolochene were
produced and the main reaction product was germacrene A.
While the possibility that germacrene A is a side product of AS
catalysis could not be excluded, these observations were in
agreement with the proposal that germacrene A is an inter-
mediate of AS catalysis. Together with the previous observa-
tions addressing the functional role of Tyr 92 and the results
obtained with mechanism based inhibitors,17,19 our findings
support strongly the reaction mechanism proposed above
(Scheme 1) for the production of aristolochene from FPP.
This work was financially supported by the BBSRC (student-
ship (SF) and research grant 6/B17177 (RKA & SET)).
Fig. 1 Total ion chromatograms from GC-MS analysis of the products of
catalysis by AS, ASW334F, ASW334V and ASW334L. (*: non-sesqui-
terpene contaminant.)
Notes and references
‡ We thank Dr. Larry Cool (Forest Products Laboratory, University of
California, Berkeley) for a generous gift of 2 and Mr T. Cannon of
DeMonchy Aromatics Ltd. for valencene.
aristolochene with only slightly reduced specificity. ASW334F
also produced an increased amount of valencene (5) (Fig. 1 and
Table 1), which was identified by comparison of its mass
spectrum with that of an authentic sample,‡ and 0.4% of an
unidentified hydrocarbon of mass 204 (Fig. 1).
While the smaller aromatic ring of Phe appeared not to
stabilise the developing positive charge in the transition state
preceding eudesmane cation as well as the indole ring of
tryptophan, stabilisation was nevertheless sufficient for the
production of significant amounts of aristolochene. However,
when Trp 334 was replaced with Val, 95.3% of the hexane
extractable products were germacrene A, while the remainder
was aristolochene (Table 1). When Leu replaced Trp, germa-
crene A was the only product. In the absence of the stabilising
interaction with the aromatic p-system, the formation of
eudesmane cation was prevented by an exceedingly high energy
barrier.
1 J. S. Glasby, Encyclopedia of Terpenoids, Wiley, Chichester, 1982.
2 D. E. Cane, Chem. Rev., 1990, 90, 1089–1103.
3 C. A. Lesburg, J. M. Caruthers, C. M. Paschall and D. W. Christianson,
Curr. Opin. Struct. Biol., 1998, 8, 695–703.
4 C. M. Starks, K. Back, J. Chappell and J. P. Noel, Science, 1997, 277,
1815–1820.
5 C. A. Lesburg, G. Zhai, D. E. Cane and D. W. Christianson, Science,
1997, 277, 1820–1824.
6 M. J. Rynkiewicz, D. E. Cane and D. W. Christianson, Proc. Natl. Acad.
Sci. USA, 2001, 98, 13543–13548.
7 J. M. Caruthers, I. Kang, M. J. Rynkiewicz, D. E. Cane and D. W.
Christianson, J. Biol. Chem., 2000, 275, 25533–25539.
8 B. Greenhagen and J. Chappell, Proc. Natl. Acad. Sci. USA, 2001, 98,
13479–13481.
The very low rate constants for ASW334L and ASW334V
suggest that Trp 334 also affected the production of germacrene
A from FPP in the first step of AS catalysis. If the overall rate
of germacrene A formation had only been controlled by product
release from the active sites of ASW334L and ASW334V, then
kcat would reflect the rate of this physical step. The rate of
germacrene A release from AS could be estimated, but would be
too low to account for the formation of 7.5% germacrene A by
the wild-type enzyme. The speed of germacrene A formation
from FPP must therefore be controlled at least in part by
Trp334. The p-system of the indole ring of Trp334 could
contribute to the stabilisation of the positive charge on C1.
Previous evidence has suggested that cyclisation of FPP to
germacrene A is not a concerted process, but rather proceeds in
a stepwise fashion via an allylic cation.13 Alternatively, the size
of the side chain of residue 334 might be important to orientate
the pyrophosphate leaving group in a way that allows optimal
orbital overlap with the p-orbitals of the neighbouring double
bond and the C2–C3 double bond.
9 M. J. Calvert, P. R. Aston and R. K. Allemann, J. Am. Chem. Soc., 2002,
124, 11636–11641.
10 M. J. Calvert, S. E. Taylor and R. K. Allemann, Chem. Commun., 2002,
2384–2385.
11 M. J. Rynkiewicz, D. E. Cane and D. W. Christianson, Biochemistry,
2002, 41, 1732–1741.
12 M. Seemann, G. Zhai, J.-W. de Kraker, C. M. Paschall, D. W.
Christianson and D. E. Cane, J. Am. Chem. Soc., 2002, 124,
7681–7689.
13 A. Deligeorgopoulou and R. K. Allemann, Biochemistry, 2003, 42,
7741–7747.
14 D. E. Cane, P. C. Prabhakaran, E. J. Salaski, P. H. M. Harrison, H.
Noguchi and B. J. Rawlings, J. Am. Chem. Soc., 1989, 111,
8914–8916.
15 T. M. Hohn and R. D. Plattner, Arch. Biochem. Biophys., 1989, 272,
137–143.
16 R. H. Proctor and T. M. Hohn, J. Biol. Chem., 1993, 268, 4543–4548.
17 D. E. Cane and Y. S. Tsantrizos, J. Am. Chem. Soc., 1996, 118,
10037–10040.
18 Wiley Mass Spectral Database, on-line version. Accession number for
the mass spectrum of 2: # 37724.
In summary, the work described here established that Trp 334
in aristolochene synthase played a central role in the production
19 D. E. Cane and C. Bryant, J. Am. Chem. Soc., 1994, 116,
12063–12064.
CHEM. COMMUN., 2003, 2162–2163
2163