Intermediacy of eudesmane cation during catalysis by aristolochene synthase

Article abstract of DOI:10.1021/jo902397v

(Chemical Equation Presented) Aristolochene synthase from Penicillium roqueforti (PR-AS) catalyzes the formation of the bicyclic sesquiterpene (+)-aristolochene (5) from farnesyl diphosphate (1, FDP) in two mechanistically distinct cyclization reactions. The first reaction transforms farnesyl diphosphate to the uncharged intermediate (S)-(-)-germacrene A (3) through a macrocyclization process that links C1 and C10 upon magnesium ion-assisted diphosphate ester activation. In the second reaction mediated by PRAS, a protonation induced cyclization has been suggested to generate the highly reactive trans-fused eudesmane cation 4 as a consequence of the precise folding of the enzyme-bound germacrene A intermediate. This contribution describes the use of the transition state analogue inhibitor 4-aza-eudesm-11-ene to explore the intermediacy of cation 4 as an on-path intermediate in the biosynthesis of aristolochene. 4-Aza-eudesm-11-ene as the hydrochloride salt 6 was stereospecifically synthesized in seven steps and 37% overall yield starting from chiral enamine 9. The synthetic sequence featured a highly regio- and stereoselective deracemization reaction of 9 that gave rise to the correspondingMichael adduct in >95% diastereomeric excess as evidenced by optical rotation and NMR measurements. 6 acts as a potent competitive inhibitor of PR-AS (K_i = 0.35 ± 0.12 μM) independent of the presence of diphosphate (K_i = 0.24 ± 0.09 μM). The failure of exogenous PP_i to enhance the binding affinity of 6 for PR-AS could be interpreted against an eudesmyl cation/diphosphate anion pair mechanism as the enzymatic strategy to stabilize the highly reactive eudesmane cation 4. In addition, these observations seem to rule out simple favorable electrostatic and/or hydrogen bonding interactions between the active site anchored diphosphate ion and the ammonium ion 6 as the binding mode. Ammonium ion 6 seems to act as a genuine mimic of eudesmane cation (4) that most likely binds the active site of PR-AS in a productive conformation resembling that adapted by 4 during the PR-AS-catalyzed synthesis of 5.

Full text of DOI:10.1021/jo902397v

pubs.acs.org/joc
Intermediacy of Eudesmane Cation during Catalysis by
Aristolochene Synthase
Juan A. Faraldos, Benson Kariuki, and Rudolf K. Allemann*
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff, CF10 3AT, United Kingdom
allemannrk@cardiff.ac.uk
Received November 12, 2009
Aristolochene synthase from Penicillium roqueforti (PR-AS) catalyzes the formation of the bicyclic
sesquiterpene (þ)-aristolochene (5) from farnesyl diphosphate (1, FDP) in two mechanistically
distinct cyclization reactions. The first reaction transforms farnesyl diphosphate to the uncharged
intermediate (S)-(-)-germacrene A (3) through a macrocyclization process that links C1 and C10
upon magnesium ion-assisted diphosphate ester activation. In the second reaction mediated by PR-
AS, a protonation induced cyclization has been suggested to generate the highly reactive trans-fused
eudesmane cation 4 as a consequence of the precise folding of the enzyme-bound germacrene
A intermediate. This contribution describes the use of the transition state analogue inhibitor 4-
aza-eudesm-11-ene to explore the intermediacy of cation 4 as an on-path intermediate in the
biosynthesis of aristolochene. 4-Aza-eudesm-11-ene as the hydrochloride salt 6 was stereospecifically
synthesized in seven steps and 37% overall yield starting from chiral enamine 9. The synthetic
sequence featured a highly regio- and stereoselective deracemization reaction of 9 that gave rise to the
corresponding Michael adduct in >95% diastereomeric excess as evidenced by optical rotation and
NMR measurements. 6 acts as a potent competitive inhibitor of PR-AS (Ki = 0.35 ( 0.12 μM)
independent of the presence of diphosphate (Ki = 0.24 ( 0.09 μM). The failure of exogenous PPi to
enhance the binding affinity of 6 for PR-AS could be interpreted against an eudesmyl cation/
diphosphate anion pair mechanism as the enzymatic strategy to stabilize the highly reactive
eudesmane cation 4. In addition, these observations seem to rule out simple favorable electrostatic
and/or hydrogen bonding interactions between the active site anchored diphosphate ion and the
ammonium ion 6 as the binding mode. Ammonium ion 6 seems to act as a genuine mimic of
eudesmane cation (4) that most likely binds the active site of PR-AS in a productive conformation
resembling that adapted by 4 during the PR-AS-catalyzed synthesis of 5.
Introduction
precursor of several sesquiterpenoid mycotoxins including
the lethal PR toxin.1,2 The crystal structure of recombinant
Recombinant aristolochene synthase from Penicillium
roqueforti (PR-AS) is a 39 kDa monomeric fungal terpene
cyclase that catalyzes the Mg2þ-dependent cyclization and
rearrangement of farnesyl diphosphate (1, FDP) to the
bicyclic sesquiterpene (þ)-aristolochene (5) (Figure 1), the
(1) Hohn, T. M.; Plattner, R. D. Arch. Biochem. Biophys. 1989, 272,
137–43.
(2) Cane, D. E.; Wu, Z.; Proctor, R. H.; Hohn, T. M. Arch. Biochem.
Biophys. 1993, 304, 415–419.
DOI: 10.1021/jo902397v
r
Published on Web 01/22/2010
J. Org. Chem. 2010, 75, 1119–1125 1119
2010 American Chemical Society

Faraldos et al.
JOCArticle
FIGURE 1. Proposed mechanism for the cyclization and rearrangements of FDP to aristolochene (5) catalyzed by aristolochene synthase via
enzyme-bound germacrene A (3) and eudesmane cation (4).5 The structure of the aza-analogue 6 that mimics putative 4 is shown.
˚
PR-AS has been solved at 2.5 A resolution revealing a typical
R-helical class I terpene cyclase fold that contains the two
Mg2þ-binding motifs found in many class I enzymes at the
top of the active site cleft, namely the aspartate-rich DDXX-
(D/E) motif, D115D116VLE,119 and the highly conserved
flexible (E,E)-configured (S)-(-)-germacrene A as the reactive
UU-conformer (14-CH3 and 15-CH3 diaxial; Figure 1),9,23
which 92% of the time remains tightly bound to the active site
of the enzyme (∼8% of 3 is released by the enzyme as the major
side product).13 Protonation of germacrene A by an as of yet
unidentified active site acid8,20,24 and electron flow from the C2,
C3 double bond to C6 leads to the formation of the bicyclic
trans-fused eudesmane cation (4). Successive 1,2-hydride and
methyl migrations followed by loss of the pro-S hydrogen from
C87,8 generate bicyclic (þ)-aristolochene (5).
NSE triad, N244DIYS248YDKE252 3,4
.
The complex cationic reaction mechanism of the PR-AS-
catalyzed conversion of FDP (1) to aristolochene (5)
(Figure 1) has been probed previously with deuterium-
labeled derivatives and fluorinated analogues of FDP,6-10
alternative substrates,11 mechanism based inhibitors,12 and
site-directed mutagenesis experiments.13-19 These studies
provided evidence for two distinct cyclization reactions
involving two discrete intermediates, the neutral hydrocarbon
(S)-(-)-germacrene A (3) and the highly reactive eudesmane
cation (4). Upon binding of three magnesium ions,3,20,21
diphosphate expulsion from the C1 position of FDP is
triggered by the intramolecular attack of the distal C10,
C11 π-bond on C1 (Figure 1) in a concerted reaction22 that
occurs with inversion of configuration at C1.6 Deprotona-
tion of the resulting germacren-11-yl cation (2) leads to the
In contrast to the mechanism of the formation of germa-
crene A, which has been studied extensively, the intermedi-
acy of eudesmane cation (4) and the mechanism of its
formation have received less attention. The existence of the
intermediate 4 has been inferred from site-directed mutagen-
esis experiments, in which replacement of Trp 334, an active
site residue ideally placed to stabilize a positive charge on C4
of eudesmane cation, with the aliphatic amino acids Val and
Leu led to the accumulation of germacrene A.16 In addition,
the involvement of trans-fused 4 in PR-AS catalysis has also
been suggested based on stereochemical results obtained
from experiments with deuterated FDP’s.6,7
(3) Caruthers, J. M.; Kang, I.; Rynkiewicz, M. J.; Cane, D. E.; Chris-
tianson, D. W. J. Biol. Chem. 2000, 275, 25533–25539.
(4) Christianson, D. W. Chem. Rev. 2006, 106, 3412–3442.
(5) Allemann, R. K. Pure Appl. Chem. 2008, 80, 1791–1798.
(6) Cane, D. E.; Prabhakaran, P. C.; Salaski, E. J.; Harrison, P. H. M.;
Noguchi, H.; Rawlings, B. J. J. Am. Chem. Soc. 1989, 111, 8914–16.
(7) Cane, D. E.; Prabhakaran, P. C.; Oliver, J. S.; McIlwaine, D. B. J. Am.
Chem. Soc. 1990, 112, 3209–3210.
(8) Miller, D. J.; Gao, J. L.; Truhlar, D. G.; Young, N. J.; Gonzalez, V.;
Allemann, R. K. Org. Biomol. Chem. 2008, 6, 2346–2354.
(9) Miller, D. J.; Yu, F. L.; Allemann, R. K. ChemBioChem 2007, 8, 1819–
1825.
(10) Miller, D. J.; Yu, F. L.; Knight, D. W.; Allemann, R. K. Org. Biomol.
Chem. 2009, 7, 962–975.
(11) Cane, D. E.; Tsantrizos, Y. S. J. Am. Chem. Soc. 1996, 118, 10037–
10040.
(12) Cane, D. E.; Bryant, C. J. Am. Chem. Soc. 1994, 116, 12063–12064.
(13) Calvert, M. J.; Ashton, P. R.; Allemann, R. K. J. Am. Chem. Soc.
2002, 124, 11636–11641.
(14) Calvert, M. J.; Taylor, S. E.; Allemann, R. K. Chem. Commun. 2002,
2384–2385.
Several candidates have been proposed for the acid within the
active site that reprotonates germacrene A, including Tyr 92 or
a proton shuttle from the solvent to Tyr 92 by way of Arg 200,
Asp 203, and Lys 206,5,13,14 an unprecedented active site
oxonium ion,19 or the diphosphate ion itself.20 However, there
is currently little evidence for any of these proposals. In addi-
tion, an alternative mechanism involving protonation of the C6,
C7 double bond of germacrene A by intramolecular transfer of
a proton from C1224 has also been excluded experimentally.8
Hence, the absence of an identified general acid could be
interpreted against the existence of the intermediate 4.
An alternative approach to identify postulated carbo-
cationic intermediates in terpene synthase catalyzed reac-
tions has relied on the use of aza analogues of these reactive
and transient cationic species.25-29 Numerous aza analogues
(15) Deligeorgopoulou, A.; Allemann, R. K. Biochemistry 2003, 42,
7741–7747.
(16) Deligeorgopoulou, A.; Taylor, S. E.; Forcat, S.; Allemann, R. K.
Chem. Commun. 2003, 2162–2163.
(17) Forcat, S.; Allemann, R. K. Chem. Commun. 2004, 2094–2095.
(18) Forcat, S.; Allemann, R. K. Org. Biomol. Chem. 2006, 4, 2563–2567.
(19) Felicetti, B.; Cane, D. E. J. Am. Chem. Soc. 2004, 126, 7212–7221.
(20) Shishova, E. Y.; Di Costanzo, L.; Cane, D. E.; Christianson, D. W.
Biochemistry 2007, 46, 1941–1951.
(23) Faraldos, J. A.; Wu, S.; Chappell, J.; Coates, R. M. Tetrahedron
2007, 63, 7733–7742.
(24) Allemann, R. K.; Young, N. J.; Ma, S.; Truhlar, D. G.; Gao, J.
J. Am. Chem. Soc. 2007, 129, 13008–13013.
(25) Narula, A. S.; Rahier, A.; Benveniste, P.; Schuber, F. J. Am. Chem.
Soc. 1981, 103, 2408–2409.
(26) Rahier, A.; Taton, M.; Benveniste, P. Biochem. Soc. Trans. 1990, 18,
48–52.
(21) Shishova, E. Y.; Yu, F. L.; Miller, D. J.; Faraldos, J. A.; Zhao, Y. X.;
Coates, R. M.; Allemann, R. K.; Cane, D. E.; Christianson, D. W. J. Biol.
Chem. 2008, 283, 15431–15439.
(27) Poulter, C. D. Acc. Chem. Res. 1990, 23, 70–77.
(28) Abe, I.; Rohmer, M.; Prestwich, G. D. Chem. Rev. 1993, 93, 2189–
2206.
(22) Yu, F. L.; Miller, D. J.; Allemann, R. K. Chem. Commun. 2007,
4155–4157.
(29) Abe, I.; Tomesch, J. C.; Wattanasin, S.; Prestwich, G. D. Nat. Prod.
Rep. 1994, 11, 279–302.
1120 J. Org. Chem. Vol. 75, No. 4, 2010

Faraldos et al.
JOCArticle
SCHEME 1. Stereospecific Synthesis of Bicyclic Imine 13
have been shown to effectively mimic the topological and
electrostatic properties of putative carbocation intermedi-
ates, and this strategy has provided potent competitive
inhibitors of mono-,30 sesqui-,31 and diterpene synthases,32-35
as well as inhibitors of squalene and oxidosqualene
synthases.28,36-38 In addition, some aza-terpenoids have
served as crystallization aids of mono- and sesquiterpene
synthases39,40 to yield valuable structural information.4
Here, we report the stereoselective synthesis of the nitrogen-
containing eudesmane cation analogue 6, as well as the kinetic
evaluation of this compound as a possible transition state
inhibitor of PR-AS in both the absence and presence of exogen-
ous PPi. Our results support the existence of the on-path inter-
mediate 4 during PR-AS catalysis and shed light on the nature of
the stabilization of this important carbocationic intermediate.
The optical rotation [R]25D of -8.1° (c 2.0, CHCl3) mea-
sured for the Michael adduct 10 was in excellent agreement
with the value of þ9.0° reported previously for the known
enantiomer of 10, which had been obtained with a de of
>95%.42 The high degree of regio- and diastereoselectivity
observed during the asymmetric Michael addition involving
chiral enamines44 such as 9b is well precedented in several
synthetic schemes.45-49 The observed regio- and steroselec-
tive induction has been explained previously based on steric
and electronic arguments.50 Primary amide 12 was synthe-
sized in 90% yield over two steps by ketalization of 10 to
yield ester 1142 followed by reaction with NH4OH. LiAlH4
reduction of amide 12 to the corresponding primary ketal-
amine and subsequent exposure of the crude reaction pro-
duct to aqueous acid afforded the cyclic Schiff base 13 in
87% yield.46
The trans-fused decahydroquinoline 14 was obtained in
Results and Discussion
85% yield by reduction with sodium in ethanol of Δ1,9
-
octahydroquinoline 13 (Scheme 2).51,52 1H and 13C NMR
spectroscopic analysis of 14 revealed that it was produced as
a single trans-isomer with characteristic 1H NMR signals at
δ 3.11 (dd, J = 12.1, 4.6 Hz), 2.68 (td, J = 12.5, 3.1 Hz), 2.44
(dd, J = 12.5, 3.1 Hz), and 0.99 (s), assigned to protons H3eq,
H3ax, H5, and the angular CH3 at C10, respectively (Scheme 3).
These values were in good agreement with those previously
reported for the CH3 at C10 analogue of trans-decahydroquino-
line at δ 3.06 (d, J = 12.0 Hz, H3eq), 2.64 (td, J = 12.0, 3.5 Hz,
H3ax), 2.21 (H5), and 0.96 (s, 10-Me) ppm,51 which lacks the
isopropylidene substituent at C7. The resonance assigned to the
methine H5 of 10-CH3 trans-decahydroquinoline51 at 2.21 ppm
is relatively upfield of the corresponding resonance of H5 at
2.44 ppm in compound 14. This field shift could be explained by a
deshielding effect caused by the 1,3-diaxial interaction between
the methine H5 and the isopropylidene group at C7 of 14
(Scheme 3). In addition, the diagnostic 13C NMR signal observed
Synthesis of 4-Aza-eudesm-11-ene (6). The synthetic se-
quence leading to the bicyclic aza-analogue 6 is outlined in
Schemes 1 and 2. Enamine 9b (prepared in 91% yield from
(-)-dehydrocarvone and (R)-(þ)-1-phenylethylamine)41 was
alkylated with methyl acrylate (THF, room temperature, 7 days)
to produce ketone 1042,43 in 63% yield as the only regioisomer in
an approximately 94% diastereomeric excess (de) as evidenced by
1H and 13C NMR spectroscopy.
(30) McGeady, P.; Pyun, H. J.; Coates, R. M.; Croteau, R. Arch.
Biochem. Biophys. 1992, 299, 63–72.
(31) Cane, D. E.; Yang, G. H.; Coates, R. M.; Pyun, H. J.; Hohn, T. M.
J. Org. Chem. 1992, 57, 3454–3462.
(32) Peters, R. J.; Ravn, M. M.; Coates, R. M.; Croteau, R. B. J. Am.
Chem. Soc. 2001, 123, 8974–8978.
(33) Ravn, M. M.; Peters, R. J.; Coates, R. M.; Croteau, R. J. Am. Chem.
Soc. 2002, 124, 6998–7006.
(34) Roy, A.; Roberts, F. G.; Wilderman, P. R.; Zhou, K.; Peters, R. J.;
Coates, R. M. J. Am. Chem. Soc. 2007, 129, 12453–12460.
(35) Mann, F. M.; Prisic, S.; Hu, H.; Xu, M.; Coates, R. M.; Peters, R. J.
J. Chem. Biol. 2009, 284, 23574–23579.
(36) Sandifer, R. M.; Thompson, M. D.; Gaughan, R. G.; Poulter, C. D.
J. Am. Chem. Soc. 1982, 104, 7376–7378.
(37) Steiger, A.; Pyun, H. J.; Coates, R. M. J. Org. Chem. 1992, 57,
3444–3449.
(38) Koohang, A.; Coates, R. M.; Owen, D.; Poulter, C. D. J. Org. Chem.
1999, 64, 6–7.
(39) Whittington, D. A.; Wise, M. L.; Urbansky, M.; Coates, R. M.;
Croteau, R. B.; Christianson, D. W. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
15375–15380.
(44) Pfau, M.; Revial, G.; Guingant, A.; d’Angelo, J. J. Am. Chem. Soc.
1985, 107, 273–274.
(45) d’Angelo, J.; Desmaele, D.; Dumas, F.; Guingant, A. Tetrahedron:
Asymmetry 1992, 3, 459–505.
(46) Costa, P. R. R.; Castro, R. N.; Farias, F. M. C.; Antunes, O. A. C.;
Bergter, L. Tetrahedron: Asymmetry 1993, 4, 1499–1500.
(47) Tenius, B. S. M.; Rohde, A. R.; Victor, M. M.; Viegas, C. Synth.
Commun. 1996, 26, 197–203.
(40) Vedula, L. S.; Rynkiewicz, M. J.; Pyun, H.-J.; Coates, R. M.; Cane,
D. E.; Christianson, D. W. Biochemistry 2005, 44, 6153–6163.
(41) Zhang, Z.; Zhou, G.; Gao, X. L.; Li, Y. L.; Liao, R. N. Synth.
Commun. 2001, 31, 847–851.
(48) Zhabinskii, V. N.; Minnaard, A. J.; Wijnberg, J.; deGroot, A. J. Org.
Chem. 1996, 61, 4022–4027.
(49) Xiong, Z. M.; Yang, J.; Li, Y. L. Tetrahedron: Asymmetry 1996, 7,
2607–2612.
(42) Fang, L. J.; Chen, J. C.; Zheng, G. J.; Zhang, C.; Li, Y. L. Chirality
2005, 17, 163–166.
(50) Tenius, B. S. M.; Deoliveira, E. R.; Ferraz, H. M. C. Tetrahedron:
Asymmetry 1993, 4, 633–636.
(43) Shono, T.; Kise, N.; Fujimoto, T.; Tominaga, N.; Morita, H. J. Org.
Chem. 1992, 57, 7175–7187.
(51) Vierhapper, F. W.; Eliel, E. L. J. Org. Chem. 1975, 40, 2734–2742.
(52) Eliel, E. L.; Vierhapper, F. W. J. Org. Chem. 1976, 41, 199–208.
J. Org. Chem. Vol. 75, No. 4, 2010 1121

Faraldos et al.
JOCArticle
SCHEME 2. Synthesis of Trans-Fused Bicyclic Amine 15,
Its Hydrochloride Salt 6, and the cis-Fused 4-Aza-decalin 16
FIGURE 2. ORTEP representation of the chloride salt of 4-
aza-eudesm-11-ene analogue 6.
SCHEME 3. syn-Diaxial Interactions (Double Arrow) in
10-Methyl trans-Decahydroquinoline 14 and Its Corresponding
N-Methyl Derivative (15)
sponding to the N-CH3 protons, led to NOE enhancements
of H3eq (7%) and H5ax (3%) at 2.87 and 2.09 ppm, respec-
tively, while no effect was seen on the protons of the methyl
group on C10. NOE enhancements were also observed for
protons H3ax (18%) and the N-CH3 group (6%), when the
broad doublet assigned to H3eq at δ 2.87 was irradiated. The
configuration of the methyl at the tertiary C10 position in 15
was therefore unambiguously assigned as anti (trans-fused)
relative to H5 based on the absence of NOE enhancement
with H5ax when the angular 10-methyl group at 1.00 ppm
was irradiated. On the basis of these considerations, it is clear
that protonation of amine 15 generates a single diastereo-
meric hydrochloride, in which the proton on nitrogen adopts
the axial position (see trans-aza-decalin in Scheme 2 and
Figure 2). The equatorial N-methyl and trans-fused structure
of the ammonium salt 6 displayed in Scheme 2 was further
corroborated by the solution of the X-ray crystal structure of
this protonated bicyclic amine (Figure 2).56
for the CH3 at C10 of amine 14 at δc 15.32 is essentially identical
with that previously recorded (15.60 ppm) for the analogous CH3
group at C10 of trans-decahydroquinoline.52
Reductive N-alkylation of compound 14 with formalde-
hyde in the presence NaBH3CN provided the N-methyl
trans-decahydroquinoline 15 in 88% yield.53 Exposure of
15 to hydrochloric acid in diethyl ether furnished the corre-
sponding ammonium salt 6 in crystalline form and in essen-
tially quantitative yield. Only the isomer with the methyl
Notably, the cis-fused decahydroquinoline 16 was ob-
tained exclusively58,59 from imine 9 by reduction with
NaBH4 (Scheme 2).60,61 The related enzymes tobacco 5-epi-
aristolochene synthase (TEAS)62 and henbane premnaspirodiene
synthase (HPS)63 generate their hydrocarbon reaction products
via the cis-fused eudesmane cation 17 (Figure 3).64,65 Hence, our
group in the equatorial position was detected in 1H and 13
C
NMR spectra of N-methyl amine 15 (Scheme 3), likely as a
consequence of the unfavorable interactions between the two
axial methyl groups and two CH3/H-1,3-syn-diaxial inter-
actions in the N-axial conformer of 15.54,55 Compound 15
displayed characteristic 1H NMR signals for H3eq, the
equatorial N-methyl, and the angular CH3 at C10 at δH
2.87, 2.11, and 1.00, respectively, in good accord with the 1H
NMR signals reported for the parent (N,10)-dimethyl trans-
decahydroquinoline (DMDHQ) at 2.87, 2.15, and 0.98 ppm.54,55
In addition, the bicyclic amine 15 displayed the diagnostic
13C NMR resonance assigned to its N-Me equatorial con-
former at 43.03 ppm (43.11 for DMDHQ)52,54,55 as well as
the signal associated with the angular 10-methyl group at
17.11 ppm (17.35 ppm for DMDHQ).52 The relative stereo-
chemistry of 15 was confirmed by NOE measurements.
Saturation of the well-separated signal at δH 2.11, corre-
˚
(56) C14H26NCl, fw = 243.81, T = 150 K, λ = 0.71073 A, orthorhombic,
˚
P212121, a = 6.7550(1) A, b = 8.4270(2) A, c = 25.7000(4) A, V = 1462.96(5)
˚
˚
3
3
˚
A , Z = 4, F(cal) = 1.107, size = 0.20 ꢀ 0.15 ꢀ 0.15 mm , independent
reflections = 3304, final R1= 0.042, wR2 = 0.097, for I > 2σ(I). Thedatawere
˚
collected with use of graphite-monochromated Mo KR (λ = 0.710 73 A)
radiation at 150 K. The structures were solved by direct methods, using
SHELXS-96, and refined with all data on F2 full-matrix least-squares, using
SHELXL-97.57 All non-hydrogen atoms were refined anisotropically. Hydro-
gen atom positions were located from difference Fourier maps and a riding
model with atomic displacement parameters 1.2 times (1.5 times for methyl
groups) those of the atom to which they are bonded was used for subsequent
refinements.
(57) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467–473.
(58) Vierhapper, F. W.; Eliel, E. L. J. Org. Chem. 1977, 42, 51–62.
(59) 1H (500 MHz) and 13C (125 MHz) NMR analysis revealed that this
sample was composed of 95% cis- (16) and 5% trans-decahydroquinoline
(14).
(60) Maiti, S. B.; Raychaudhuri, S. R.; Chatterjee, A.; Chakravarty, A. K.
J. Chem. Soc., Perkin Trans. 1 1988, 611–621.
(61) Bhattacharya, S.; Mandal, A. N.; Chaudhuri, S. R. R.; Chatterjee, A.
J. Chem. Soc., Perkin Trans. 1 1984, 5–13.
(62) Starks, C. M.; Back, K.; Chappell, J.; Noel, J. P. Science 1997, 277,
1815–1820.
(53) Lee, S. S.; Venkatesham, U.; Rao, C. P.; Lam, S. H.; Lin, J. H.
Bioorg. Med. Chem. 2007, 15, 1034–1043.
(54) Eliel, E. L.; Vierhapper, F. W. J. Am. Chem. Soc. 1975, 97,
2424–2430.
(55) Eliel, E. L.; Vierhapper, F. W. J. Am. Chem. Soc. 1974, 96, 2257–2258.
(63) Back, K.; Chappell, J. J. Biol. Chem. 1995, 270, 7375–7381.
(64) Schenk, D. J.; Starks, C. M.; Manna, K. R.; Chappell, J.; Noel, J. P.;
Coates, R. M. Arch. Biochem. Biophys. 2006, 448, 31–44.
(65) Faraldos, J. A.; Zhao, Y. X.; O’Maille, P. E.; Noel, J. P.; Coates,
R. M. ChemBioChem 2007, 8, 1826–1833.
1122 J. Org. Chem. Vol. 75, No. 4, 2010

Faraldos et al.
JOCArticle
It had previously been observed that the potency of
nitrogen-containing and sulfonium ion analogues of pre-
sumed carbocationic intermediates in the reactions catalyzed
by pinene synthase,30,67 bornyl diphosphate synthase,39,67
trichodiene synthase,31,40 abietadiene synthase,32,33 ent-kaurene
synthase,34 and squalene synthase36-38,68 is increased by the
addition of inorganic diphosphate (PPi). These observat-
ions have been interpreted to suggest that formation of
carbocation-diphosphate ion pair intermediates might be
a general strategy used by terpene synthases to stabilize
transiently generated carbocations during catalysis. In an
attempt to mimic the complex of PR-AS with 4-azaeudesm-
11-ene (6) in the presence of PPi, the inhibition constant of 6
for the PR-AS-catalyzed conversion of FDP was also deter-
mined in the presence of diphosphate at concentrations of 50,
250, and 500 μM.69 The Ki value of 0.24 ( 0.09 μM measured
in the presence of 250 μM of PPi was similar to the values
(0.35, 0.43, and 0.22 μM) obtained at diphosphate concen-
trations of 0, 50, and 500 μM, respectively. Clearly the
addition of exogenous PPi did not enhance the binding
affinity of this competitive inhibitor, indicating that the
stable binding of 4 to the active site of PR-AS was not
enhanced by ion pairing with the diphosphate ion. Structural
studies with terpene cyclases-substrate analogue cocrystals
have also shown that the enzyme-bound PPi counterions
occupy a relative invariant position,4 and in the case of
aristolochene synthase from A. terreus,20 the proximity of
the O3 oxygen atom of the [Mg2þ]3-PPi complex to the C6
and C8 atoms (farnesyl chain) was interpreted to involve the
active site anchored PPi ion as the possible general acid/base
responsible for the second cyclization reaction catalyzed by
aristolochene synthase.20
FIGURE 3. Structures of eudesmane cations 4 (trans-fused) and
17 (cis-fused) produced in the active sites of PR-AS and TEAS,
respectively.
approach for the synthesis of the aza-analogue 6 can also be used
to prepare the cis-aza-decalin of 17 from (þ)-dihydrocarvone
to probe TEAS and HPS mechanisms.
Inhibition of PR-AS by 4-Aza-eudesm-11-ene (6). Incuba-
tions of recombinant PR-AS with FDP (1) in the presence of
Mg2þ yielded the expected 92% aristolochene (4) accompa-
nied by ∼7% germacrene A and a small amount of valencene
as determined by GC-MS (see the Supporting Information).
Inhibition studies with compound 6 indicated that this
ammonium salt acted as a competitive inhibitor with a Ki
of 0.35 ( 0.12 μM, similar to the Michaelis constant (KM
=
0.53 ( 0.21 μM) determined for the PR-AS-catalyzed con-
version of FDP to aristolochene. The similar values for Ki
and KM indicate that the tetrahedral geometry of the ammo-
nium nitrogen relative to the presumably trigonal sp2 con-
figuration of C4 in eudesmane cation (4, Figure 3) did not
strongly affect the binding affinity of 6. The inhibition
constant was approximately two times smaller than that
measured for 12,13-difluoro FDP.22 To the best of our
knowledge 6 is the tightest binding inhibitor of PR-AS
described so far.
Structural studies with soaked cocrystals of the aza-analogues
synthesized as presumed mimics of bisabolyl, R-terpenyl,
and bornyl cations39,40 have revealed that these analogues as
their protonated amines are sufficiently strong competitive
inhibitors of trichodiene synthase and bornyl diphosphate
synthase. However, with the exception of 2-azabornane,
these protonated aza-terpenes bound to the corresponding
[Mg2þ]3-PPi enzyme complex in what appears to be nonpro-
ductive conformations, presumably as a result of favorable
electrostatic and hydrogen bonding interactions between the
positively charged ammonium nitrogen and the diphosphate
moiety.39,40 In contrast, 2-azabornane, a product-like mimic
of bornyl cation that matches the product-like contour of the
active site of bornyl diphosphate synthase in its competent
closed conformation, was shown to bind the enzyme in a
“catalytically productive conformation”. Hence, it is likely
thattheconformationadoptedby2-azabornane inthe enzyme
resembles that of the actual bornyl cation in solution.39,66
The conformational change from the open to the cataly-
tically competent closed conformation of PR-AS is believed
to be induced by the formation of a ternary complex invol-
ving the binding of three magnesium ions to the Asp-rich and
Asn-Ser-Glu (NSE) motifs of AS, as well as to the diphos-
phate group of the substrate. Hence the tight binding of 6 to
PR-AS, even in the absence of inorganic diphosphate, is
most likely a consequence of its close structural resemblance
to the putative on-pathway carbocationic intermediate 4
rather than a simple consequence of favorable electrostatic
interactions with PPi.39,40
If these observations hold true, it seems likely that the
spatial separation between the newly generated cation at C4
of 4 (Figure 3) and the diphosphate ion could be slightly
longer than the distance of a typical covalent C-O bond
˚
(1.42 A), therefore precluding the formation of a presumed
tertiary 4-eudesmyl carbocation-OPPi ion pair. Since the
binding affinity of 6 was clearly unenhanced by the addition
of exogenous PPi, this observation suggests that the imper-
fect “puckered” tetrahedral ammonium geometry of 6 does
not place the nitrogen atom closer to O3, thus resembling the
reactive conformation of a putative unpaired eudesmane
cation. The inhibition studies presented here for ammonium
ion 6, showing an insensitivity to PPi concentrations, indicate
therefore that counterion stabilization could be limited to
carbocations in close proximity to the original PPi position
within the active sites of these terpene cyclases.34
Conclusions
The aza-analogue of the putative eudesmane cation (4)
intermediate during the enzymatic conversion of FDP to
aristolochene is a potent competitive inhibitor of PR-AS
independent of exogenously added PPi. These observations
indicate that compound 6 acts as an effective transition state
inhibitor of PR-AS and further support the intermediacy of 4
(67) Croteau, R.; Wheeler, C. J.; Aksela, R.; Oehlschlager, A. C. J. Biol.
Chem. 1986, 261, 7257–7263.
(68) Poulter, C. D.; Capson, T. L.; Thompson, M. D.; Bard, R. S. J. Am.
Chem. Soc. 1989, 111, 3734–3739.
(69) PPi on its own is not an inhibitor of PR-AS. For details see ref 1.
(66) Christianson, D. W. Curr. Opin. Chem. Biol. 2008, 12, 141–150.
J. Org. Chem. Vol. 75, No. 4, 2010 1123

Faraldos et al.
JOCArticle
as an on-pathway intermediate in PR-AS catalysis. The
high potency of this inhibitor in the absence of favorable
electrostatic interactions (i.e., in the absence of PPi) can be
ascribed to its actual resemblance to the eudesmane cation
and suggests that 6 is likely to bind the active site of PR-AS
in a conformation that closely matches that adopted by the
actual intermediate 4.39,66 Finally, the unusual lack of
synergism observed with this compound and inorganic
PPi can rule against the formation of a tightly bound
4-eudesmyl-pyrophosphate ion pair intermediate in
the cyclization catalyzed by PR-AS presumably due to
substantial spatial separation between cation 4 and the
PPi counterion. Moreover, this suggestion reinforces the
central role played by the aromatic active site residue
Trp334 in stabilizing the highly reactive eudesmane ca-
tion.16
2H), 1.69, (s, 3H), 1.64 (d, J = 12.5 Hz, 1H), 1.60 (dd, J = 4.5, 1.2
Hz, 1H), 1.57-1.46 (m, 3H), 1.27-1.25 (m, 1H), 0.83 (s, 3H); 13
C
NMR (100 MHz, CDCl3) δ 176.4, 149.2, 112.8, 108.6, 65.0, 64.8,
42.4, 40.6, 35.2, 34.2, 30.8, 29.6, 25.9, 20.9, 19.3.
(4aS,7S)-4a-Methyl-7-(prop-1-en-2-yl)-2,3,4,4a,5,6,7,8-octa-
hydroquinoline (13). The bicyclic imine 13 was prepared accord-
ing to a literature procedure with some modifications.46 To a
stirred solution of amide 12 (1.26 g, 4.72 mmol) in dry THF (100 mL)
was added LiAlH4 (1.1 g, 28.95 mmol) at room temperature under
N2. The resulting suspension was refluxed for 1 h and then cooled to
room temperature. A 5% solution of NaOH was added slowly to
precipitate the aluminum salt byproduct. The white salts were filtered
under reduced pressure and washed with additional THF (100 mL).
To the resulting THF filtrate was added 10% aqueous HCl (80 mL)
and the two-phase mixture was vigorously stirred at room tempera-
ture for 18 h. Hexane (150 mL) was added and the two phases were
separated. The aqueous layer was washed with hexane (2 ꢀ 50 mL),
basified with NaOH (pellets), and extracted with Et2O (2 ꢀ 50 mL).
The combined ethereal extracts were dried (K2CO3) and filtered, and
the solvent was removed under reduced pressure to give imine 13 (778
mg, 87%) as an oil. HR(EI)MS (Mþ) found 191.1668, C13H21N
requires 191.1674; 1H NMR (500 MHz, CDCl3) δ 4.93 (br s, 1H),
4.89 (br s, 1H), 3.67 (app ddd, Japp = 17.5, 5, 2 Hz, 1H), 3.46-3.38
(m, 1H), 2.66 (app d quintets, Japp = 14.5, 3.0 Hz, 1H), 2.60-2.50
(m, 2H), 1.96 (app tdd, Japp = 14.5, 5.5, 3.5 Hz, 1H), 1.79-1.66 (m,
2H), 1.72, (s, 3H), 1.61-1.54 (m, 1H), 1.50 (d, J = 4.0 Hz, 1H), 1.47
(br s, 1H), 1.39, (dt, J = 13.5, 4 Hz, 1H), 1.30 (dt, J = 13.5, 4.0 Hz,
1H), 1.19 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 175.2, 146.7,
112.8, 49.8, 41.5, 38.6, 36.4, 36.1, 36.0, 24.2, 23.4, 22.7, 29.2.
Experimental Section
Representative preparative procedures leading to and char-
acterization data for compounds 6, 10, 11, 12, 13, 14, 15, and 16
are given below.
Methyl 3-((1S,4S)-1-Methyl-2-oxo-4-(prop-1-en- 2-yl)cyclo-
hexyl)propanoate (10).42,43 Ketone 10 was synthesized following
the method of d’Angelo.44 Chiral imine (9),41 obtained by
condensation of (-)-dehydrocarvone and (R)-(þ)-a-methylben-
zylamine, was subjected to asymmetric Michael alkylation with
methyl acrylate, followed by removal of the chiral auxiliary to
(4aS,7S,8aR)-4a-Methyl-7-(prop-1-en-2-yl)decahydroquinoline
(14). The trans-decahydroquinoline 14 was prepared according
to the procedure of Vierhapper and Eliel.51,52 To a stirred
solution of imine 13 (268 mg, 1.40 mmol) in dry EtOH
(13 mL) was added Na (1.0 g, 43.5 mmol) at room temperature
under N2. After 7 h, additional Na (250 mg, 10.9 mmol) was
added, and stirring was continued overnight. After 29 h, water
(20 mL) was added, and EtOH was removed under reduced
pressure. The resulting aqueous layer was acidified with 10%
HCl (1 mL) and extracted with Et2O (3 ꢀ 10 mL). The aqueous
phase was basified with NaOH (pellets) and extracted with Et2O
(3 ꢀ 10 mL). The combined ethereal layers were dried (K2CO3)
and filtered, and the solvent was removed under reduced
pressure. The crude amine was dissolved in CH2Cl2 (5 mL)
afford ketone 10 in 63% overall yield.46 [R]25 -8.1 (c 2.0,
D
1
CHCl3) {lit.42 þ9° (for mirror image)}; H NMR (500 MHz,
CDCl3) δ 4.77 (br s, 1H), 4.71 (br s, 1H), 3.65 (s, 3H), 2.48 (dd,
J = 14.0, 12.5 Hz, 1H), 2.38-2.72 (m, 3H), 2.14 (ddd, J = 14.2,
11.0, 5.1 Hz, 1H), 2.04 (ddd, J = 16.2, 11.0, 5.1 Hz, 1H), 1.87 (dt,
J = 13.6, 3.9 Hz, 1H), 1.83-1.75 (m, 2H), 1.73 (s, 3H), 1.70
(ddd, J = 14.2, 11.2, 5.1 Hz, 1H,), 1.55 (ddd, J = 13.6, 11.0, 4.8
Hz, 1H), 1.02 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 214.5,
173.7, 147.3, 110.0, 51.7, 47.3, 46.1, 43.4, 38.3, 32.1, 28.8, 25.9,
21.8, 20.6. The 1H and 13C NMR data are in excellent agreement
with those previously reported for ent-10.42
Methyl 3-((1S,4S)-1-Methyl-2,2-ethylenedioxy-4-(prop-1-en-
2-yl)cyclohexyl)propanoate (11).42 To a stirred solution of ke-
tone 10 (1.25 g, 5.25 mmol) in dry benzene (100 mL) was added
ethylene glycol (0.5 mL, 9 mmol) and a catalytic amount of p-
TsOH. The mixture was refluxed for 4 h with continuous
removal of water, using a Dean-Stark apparatus. The solvents
were removed under reduced pressure, and the crude material
was purified by flash chromatography on silica gel (eluting with
5% EtOAc-hexane) to give ketal 11 as a clear oil (1.38 g, 93%).
1H NMR (400 MHz, CDCl3) δ 4.77 (br s, 1H), 4.74 (br s, 1H),
3.96-3.87 (m, 4H), 3.64 (s, 3H), 2.55 (m, 1H), 2.37 (dd, J = 7.9,
˚
and then loaded onto a basic alumina (Aldrich, 150 mesh, 58 A)
column. Elution with CH2Cl2 (removes unreacted starting imine
13) and then Et2O gave the bicyclic amine 14 as a clear liquid (230
mg, 85%). HR(EI)MS (Mþ) found 193.1831, C13H23N requires
193.1830; 1H NMR (500 MHz, CDCl3) δ 4.91 (br s, 1H), 4.86 (br
s, 1H), 3.11 (dd, J = 12.1, 4.7 Hz, 1H), 2.68 (dt, J = 12.6, 3.3 Hz,
1H), 2.44 (dd, J= 12.6, 3.3 Hz, 1H), 2.40 (br s, 1H, NH), 1.86-
1.70 (m, 4H), 1.75 (s, 3H), 1.57 (dt, J = 13.0, 5.8 Hz, 1H), 1.42
(app br d, Japp =10.5 Hz, 2H), 1.34-1.24 (m, 2H), 1.20-1.11 (m,
2H), 0.99 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 146.9, 110.6,
58.8, 47.6, 40.0, 39.2, 35.7, 34.1, 30.1, 32.1, 22.9, 22.4, 15.3.
(4aS,7S,8aR)-1,4a-Dimethyl-7-(prop-1-en-2-yl)decahydroquino-
line (15). N-Methyl-trans-decahydroquinoline 15 was prepared
by using a modification of the procedure of Lee et al.53 To a
stirred solution of amine 14 (70 mg, 0.36 mmol) in MeOH
(10 mL) was added formalin solution (2 mL, 37%), acetic acid
(1.5 mL, 26.3 mmol), and NaCNBH3 (1.5 g, 23.8 mmol) at room
temperature under N2. After being stirred for 48 h, the reaction
mixture was poured into 5% aqueous NaOH (20 mL) and
extracted with Et2O (3 ꢀ 15 mL). The combined ethereal layers
were dried (MgSO4), filtered, and concentrated under reduced
pressure. The crude product was purified by flash-chromato-
graphy on silica gel (using a linear gradient from 25% Et2O in
CH2Cl2 to 100% Et2O) to give amine 15 as a liquid (66.3 mg,
88%). HR(EI)MS (Mþ) found 207.1902, C14H25N requires
3.2 Hz, 2H), 1.65 (s, 3H), 1.60-1.40 (m, 6H), 1.29 (s, 3H); 13
NMR (125 MHz, CDCl3) δ 174.7, 149.1, 112.5, 109.7, 65.0, 64.7,
51.5, 42.3, 40.6, 35.2, 34.0, 29.2, 29.0, 25.9, 20.9, 19.2. The 13
C
C
NMR data are in excellent agreement with those previously
reported for ent-11.42
3-((1S,4S)-1-Methyl-2,2-ethylenedioxy-4-(prop-1-en-2-yl)-
cyclohexyl)propanamide (12). To a stirred solution of 11 (1.63 g,
5.77 mmol) in MeOH (120 mL) was added a ca. 35% solution of
NH4OH (80 mL) at room temperature. After 24 h, the MeOH
was removed under pressure and the resulting aqueous solution
was poured into brine (300 mL), extracted with EtOAc (3 ꢀ
100 mL), and dried (MgSO4). Removal of the solvent under
reduced pressure gave amide 12 as a white solid (1.49 g, 97%).
HR(EI)MS(Mþ) found 267.1804, C15H25NO3 requires267.1834;
1H NMR (400 MHz, CDCl3) δ 5.83 (br s, 1H), 5.57 (br s, 1H), 4.67
(br s, 2H), 3.98-3.85 (m, 4H), 2.31-2.11 (m, 3H), 1.91-1.76 (m,
1124 J. Org. Chem. Vol. 75, No. 4, 2010

Faraldos et al.
JOCArticle
207.1987; 1H NMR (400 MHz, CDCl3) δ 4.90 (br s, 1H), 4.84 (br
s, 1H), 2.87 (app d quintets, Japp = 10.8, 1.8 Hz, 1H), 2.41
(br s, 1H), 2.18 (s, 3H), 2.09 (br d, J = 13.1 Hz, 1H), 1.98 (app
ddd, Japp = 13.3, 11.0, 3.0 Hz, 1H), 1.87 (qt, J = 13.8, 4.0 Hz,
1H), 1.79-1.73 (m, 2H), 1.74 (s, 3H), 1.64 (dd, J = 12.4, 2.9 Hz,
1H), 1.41 (dt, J = 13.1, 5.6 Hz, 2H), 1.37-1.27 (m, 2H), 1.13
(dt, J = 13.7, 3.4 Hz, 1H), 1.06 (dt, J = 12.8, 4.4 Hz, 1H), 1.00
(s, 3H); 13C NMR (125 MHz, CDCl3) δ 147.0, 66.6, 59.2, 43.1,
40.5, 39.3, 35.9, 34.4, 26.4, 22.8 (2C), 22.0, 17.1.
incubation of varying amounts of [1-3H]FDP (specific acti-
vity 75 mCi/mmol) at pH 7.5 with a fixed concentration of
purified AS (100 nM) in 20 mM Tris buffer containing 5 mM
MgCl2, 5 mM β-mercaptoethanol, and 15% glycerol at pH
7.5. The reactions mixtures containing buffer, FDP, and
protein were prepared on ice in a total volume of 250 μL,
and were overlaid with 1 mL of HPLC-grade hexane prior to
incubation. The assay mixtures were incubated at room
temperature (22 °C) for 12 min. The reactions were immedi-
ately ice-cooled and quenched by addition of 200 μL of
100 mM EDTA (pH 8.5) and brief vortexing. The hexane
overlay and two additional 1 mL hexane extracts were passed
through a short pipet column containing silica gel and
MgSO4. The column was washed with additional hexane
(1 mL) and the combined filtrates were analyzed by liquid
scintillation counting with use of 15 mL of scintillation
cocktail. Steady state kinetic parameters for wild-type AS
were obtained by direct fitting of the data to the Michaelis-
Menten equation by nonlinear least-squares regression in
conjunction with the graphical procedures developed by
(4aS,7S,8aR)-1,4a-Dimethyl-7-(prop-1-en-2-yl)decahydroqui-
nolinium Chloride (6). To a stirred solution of amine 15 (66.0 mg,
0.32 mmol) in anhydrous Et2O (1.2 mL) at 0 °C was added an
ethereal solution of HCl (2 M, 160 μL, 0.32 mmol) under N2.
After standing for 1 h, the mixture was filtered to give the
hydrochloride salt 6 (77.1 mg, 99%). Mp 215-217 °C; HR(ES)-
MS (Mþ) found 208.2064, C14H26Nþ requires 208.2065; 1H
NMR (500 MHz, D2O) δ 5.01 (br s, 1H), 4.86 (br s, 1H), 3.44
(app dd. Japp = 12.7, 4.6 Hz, 1H), 3.03 (dt, J = 13.2, 3.7 Hz,
1H), 2.96 (dd, J = 13.1, 3.1 Hz, 1H), 2.79 (s, 3H), 2.60 (br s, 1H),
2.29 (br d, J = 13.5 Hz, 1H), 1.99 (qt, J = 13.8, 4.4 Hz, 1H),
1.84-1.77 (m, 2H), 1.74 (s, 3H), 1.71 (dt, J = 13.3, 5.5 Hz, 2H),
Lineweaver-Burk.71 The calculated values for KM, kcat
,
1.51 (br d, J = 13.5 Hz, 1H), 1.46-1.30 (m, 3H), 1.07 (s, 3H); 13
C
and kcat/KM of 0.53 ( 0.21 μM, 0.084 s-1, and 11.8 ꢀ 104 M-1
s-1, respectively, are in good agreement with those parameters
obtained earlier for aristolochene synthase.13,19
NMR (125 MHz, D2O) δ 146.5, 111.2, 67.9, 57.5, 40.6, 38.8,
36.7, 35.4, 34.6, 24.0, 21.8, 21.6, 19.3, 15.1.
(4aS,7S,8aS)-4a-Methyl-7-(prop-1-en-2-yl)decahydroquino-
line (16). The bicyclic cis-fused amine 16 was prepared by using a
modification of the procedure of Maiti et al.60 To a stirred
solution of imine 13 (18.8 mg, 0.10 mmol) in MeOH (1 mL) at
0 °C was added 1 drop of water followed by NaBH4 (19.0 mg,
0.5 mmol). The reaction mixture was stirred at 0 °C for 5 h and
then allowed to warm to room temperature overnight (16 h).
Water (3 mL) was added and the resulting solution was stirred at
room temperature for 30 min. The solution was then extracted
with Et2O (2 ꢀ 5 mL), the combined ethereal extracts were dried
(MgSO4) and filtered, and the solvent was evaporated with a
stream of nitrogen. The crude amine was then purified by flash
chromatography on silica gel with Et2O and then MeOH as
eluting solvents. Fractions having the basic product (MeOH)
were combined and the solvent was evaporated with a stream
of nitrogen to give a 95:5 mixture of the cis57 and trans51,52
decahydroquinolines 16 and 14 (14.3 mg, 75%). HR(EI)MS
Inhibition Studies with PR-AS. The procedure described
above to determine the kinetic parameters of PR-AS was
followed. Assay mixtures containing buffer, inhibitor, tritiated
FDP, and PR-AS (100 nM) were prepared on ice with use of a
total volume of 250 μL overlaid with HPLC-grade hexane
(1 mL) and incubated at room temperature (22 °C). In some
instances, fixed amounts of inorganic PPi were added to
evaluate the effects of this reaction product (PPi) in the
presence of the inhibitor. Kinetic data in the presence or
absence of inhibitor with or without PPi were evaluated as
mentioned above. The graphic method of Lineweaver-Burk
was used to evaluate the level of inhibition. KM and Ki were
determined by varying [1-3H]FDP concentrations in the pre-
sence or absence of fixed amounts of inhibitor with or without
fixed concentrations of PPi.
1
(Mþ) found 193.1822, C13H23N requires 193.1830; H NMR
Acknowledgment. We thank Dr Veronica Gonzalez-
Gonzalez for helping with the expression and purification
of the PR-AS gene, and Dr. Rob Jenkins for assistance with
NMR spectroscopy and mass spectrometry. This work was
supported by BBSRC grant BB/G003572/1 (R.K.A.) and
Cardiff University.
(500 MHz, CDCl3) δ 4.70 (app br d, Japp = 4.6 Hz, 2H), 2.81
(app dt, Japp = 13.0, 3.5 Hz, 1H), 2.74 (dd, J = 13, 5.0 Hz, 1H),
2.54 (dd, J = 11.3, 4.1 Hz, 1H), 2.31 (br s, 1H, NH), 1.95-1.83
(m, 3H), 1.78-1.66 (m, 1H), 1.74 (s, 3H), 1.54-1.40 (m, 4H),
1.10 (s, 3H), 0.99 (app br d, Japp = 12.4 Hz, 1H); 13C NMR (125
MHz, CDCl3) δ: 150.2, 108.2, 59.5, 45.0, 40.4, 39.6, 32.3, 32.0,
28.6, 27.3, 26.7, 22.7, 21.2.
Supporting Information Available: Materials, general meth-
ods, and instrumentation; protein production; GC-MS spectra;
kinetic plots; reproduction of NMR spectra; and X-ray crystallo-
graphic data for compound 6. This material is available free of
charge via the Internet at http://pubs.acs.org.
Aristolochene Synthase Kinetics. Kinetics assays were carried
out according to the standard, linear range, microassay pro
cedure developed for limonene and bornyl diphosphate
synthases with modifications.70 This protocol involved the
(70) Karp, F.; Zhao, Y.; Santhamma, B.; Assink, B.; Coates, R. M.;
Croteau, R. B. Arch. Biochem. Biophys. 2007, 468, 140–146.
(71) Lineweaver, H.; Burk, D. J. Am. Chem. Soc. 1934, 56, 658–666.
J. Org. Chem. Vol. 75, No. 4, 2010 1125