Cloning and characterization of Pfl-1841, a 2-methylenebornane synthase in Pseudomonas fluorescens PfO-1

Article abstract of DOI:10.1016/j.tet.2011.05.084

The pfl-1841 gene from Pseudomonas fluorescens PfO-1 is the only gene in any of the three sequenced genomes of the Gram-negative bacterium P. fluorescens, that is, annotated as a putative terpene synthase. The predicted Pfl-1841 protein, which harbors the two strictly conserved divalent metal binding domains found in all terpene cyclases, is closely related to several known or presumed 2-methylisoborneol synthases, with the closest match being to the MOL protein of Micromonaspora olivasterospora KY11048 that has been implicated as a 2-methylenebornane synthase. A synthetic gene encoding P. fluorescens Pfl-1841 and optimized for expression in Escherichia coli was expressed and purified as an N-terminal His₆-tagged protein. Incubation of recombinant Pfl-1841 with 2-methylgeranyl diphosphate produced 2-methylenebornane as the major product accompanied by 1-methylcamphene as well as other minor, monomethyl-homomonoterpene hydrocarbons and alcohols. The steady-state kinetic parameters for the Pfl-1841-catalyzed reaction were K _M=110±13 nM and k_cat=2.4±0.1×10 ^-2 s^-1. Attempts to identify the P. fluorescens SAM-dependent 2-methylgeranyl diphosphate synthase have so far been unsuccessful.

Full text of DOI:10.1016/j.tet.2011.05.084

Tetrahedron 67 (2011) 6627e6632
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Cloning and characterization of Pfl_1841, a 2-methylenebornane
synthase in Pseudomonas fluorescens PfO-1
,
Wayne K.W. Chou a, Haruo Ikeda b, David E. Cane a
*
a Department of Chemistry, 324 Brook Street, Box H, Brown University, Providence, RI 02912-9108, USA
b Laboratory of Microbial Engineering, Kitasato Institute for Life Sciences, Kitasato University, 1-15-1 Kitasato, Sagamihara, Kanagawa 252-0373, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The pfl_1841 gene from Pseudomonas fluorescens PfO-1 is the only gene in any of the three sequenced
genomes of the Gram-negative bacterium P. fluorescens, that is, annotated as a putative terpene synthase.
The predicted Pfl_1841 protein, which harbors the two strictly conserved divalent metal binding domains
found in all terpene cyclases, is closely related to several known or presumed 2-methylisoborneol
synthases, with the closest match being to the MOL protein of Micromonaspora olivasterospora
KY11048 that has been implicated as a 2-methylenebornane synthase. A synthetic gene encoding P. flu-
orescens Pfl_1841 and optimized for expression in Escherichia coli was expressed and purified as an
N-terminal His6-tagged protein. Incubation of recombinant Pfl_1841 with 2-methylgeranyl diphosphate
produced 2-methylenebornane as the major product accompanied by 1-methylcamphene as well as
other minor, monomethyl-homomonoterpene hydrocarbons and alcohols. The steady-state kinetic pa-
rameters for the Pfl_1841-catalyzed reaction were KM¼110ꢀ13 nM and kcat¼2.4ꢀ0.1ꢁ10ꢂ2 sꢂ1. Attempts
to identify the P. fluorescens SAM-dependent 2-methylgeranyl diphosphate synthase have so far been
unsuccessful.
Received 23 March 2011
Received in revised form 16 May 2011
Accepted 18 May 2011
Available online 26 May 2011
Keywords:
Terpene biosynthesis
Monoterpenes
Volatile organics
Genome mining
Pseudomonas fluorescens
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Pfl_1841 is a variant of those found in most other terpene cyclases,
which usually contain a ‘DDXXD’ motif (where the X denotes any
Pseudomonas fluorescens is a Gram-negative, flagellated bacte-
rium, that is, found colonizing plants, soil, and water.1e3 Strains of P.
fluorescens, which are known for their production of fluorescent
siderophores, or pyoverdines, that are produced during nutrient
deprivation, also produce a variety of secondary natural products
thought to be useful for bioremediation applications. Several non-
ribosomal peptide synthase- and polyketide synthase-derived
metabolites have been reported in Pseudomonads, although no
terpenes or terpene cyclases have been identified in this genus to
date.
The complete genome sequences of three strains of P. fluo-
rescens, Pf-5, PfO-1 and SBW25, have recently been determined.2
Comparison shows that these three bacterial strains are genomi-
cally diverse, sharing only 61% of their gene cohort. A single gene,
pfl_1841, which is found only in the PfO-1 strain, has been anno-
tated as a putative terpene synthase. The predicted Pfl_1841 protein
contains the aspartate-rich motif, 95DDHYCDD, and the ‘NSE’ metal
binding motif, 245NDLYSAYKE, variations of which are universally
conserved in all terpene synthases.4 The aspartate-rich motif of
residue), but corresponds closely to that found in the sequences of
a variety of confirmed and likely 2-methylisoborneol synthases
(MIBS).5
2-Methylisoborneol (2-MIB) is a volatile homomonoterpene,
known for its characteristic earthy odor, that is, produced by a va-
riety of actinomycete bacteria, cyanobacteria, and myxobacteria.6,7
There is considerable interest in the detection and remediation of
2-MIB and 2-MIB-producing organisms, due to the unpleasant off
odor and taste that this metabolite imparts to water and food
products. Recently, the two-step biosynthesis of 2-MIB from ger-
anyl diphosphate (GPP) has been elucidated in several species of
Streptomyces, in Myxobacteria, and in a cyanobacterium (Fig. 1,
lower pathway).5,6,8 In the first step of this pathway, 2-
methylgeranyl diphosphate synthase (MGPPS) catalyzes the
transfer of a methyl group from S-adenosy-L-methionine (SAM) to
the C-2 position of GPP to form 2-methylgeranyl diphosphate (2-
MGPP). MIB synthase then catalyzes the cyclization of 2-MGPP
with capture of water to give 2-MIB.
In a 2008 study of the biosynthesis of 2-MIB, Komatsu et al.
utilized a bioinformatics-based strategy to identify the genes re-
sponsible for the production of 2-MIB in actinomycetes.8 Using
a hidden Markov model (HMM) based on the terpene synthase
family metal binding domain (Pfam entry PF03906), the databases
* Corresponding author. Tel.: þ1 401 863 3588; e-mail address: David_Cane@
brown.edu (D.E. Cane).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.05.084

6628
W.K.W. Chou et al. / Tetrahedron 67 (2011) 6627e6632
two genes surrounding pfl_1841 in P. fluorescens PfO-1 are annotated
as a ‘putative dehydrogenase’ (pfl_1840), and a ‘putative transport-
related, membrane protein’ (pfl_1842).
We describe below the expression and biochemical character-
ization of the recombinant Pfl_1841 protein of P. fluorescens PfO-1
and the demonstration that this enzyme is in fact
a 2-
methylenebornane synthase. This is the first terpene synthase to
be identified in any Pseudomonad.
Fig. 1. Biosynthesis of 2-methylenebornane (upper pathway) and 2-methylisoborneol
(lower pathway).
2. Results and discussion
2.1. Cloning, expression and biochemical characterization
of Pfl_1841
of all bacterial genome sequences, supplemented by several other
Actinomycete genome sequences, were searched in order to har-
vest first the presumptive genes encoding predicted terpene syn-
thases. More than 40 such sequences were identified and then
classified into one of three distinct groups by phylogenetic analysis,
with all the presumed monoterpene synthases falling into a single
class, Group I, predicted to consist predominantly of MIB synthases.
In the majority of organisms the predicted 2-MIB genes were lo-
cated within a three-gene operon (Fig. 2). The first gene in this
operon typically encodes a predicted nucleotide binding protein of
unknown function, with the second and third genes encoding
a MIBS and MGPPS, respectively. The P. fluorescens Pfl_1841 protein
was also found to be closely related to the general MIBS class, with
the closest match being to the predicted monoterpene cyclase,
MOL, from Micromonaspora olivasterospora KY11048. The 3-gene
operon containing the MOL protein also harbors conserved genes
encoding the upstream nucleotide binding protein and the down-
stream MGPP synthase. Although heterologous expression of the
MOL operon in an engineered strain of Streptomyces avermitilis did
not give any detectable product, cultures of M. olivasterospora were
found to accumulate the monomethyl-monoterpene hydrocarbon
2-methylenebornane (2-MB) but no methylisoborneol (MIB) (Fig. 1,
upper pathway). A truncated recombinant form of the M. olivas-
terospora 2-MB synthase lacking a membrane anchor peptide cat-
alyzed the formation of 2-MB from 2-MGPP (H. Ikeda,
unpublished).
A synthetic pfl_1841 gene, optimized for expression in Escher-
ichia coli, was subcloned into the pET28a expression vector. The
recombinant plasmid was introduced into E. coli BL21 (DE3) and the
resultant transformants were used to produce N-terminal His6-
tagged Pfl_1841 protein. NieNTA purification afforded pure
Pfl_1841 protein (>95%).
Since the natural substrate for Pfl_1841 was initially unknown,
we tested a number of acyclic allylic diphosphates as potential
substrates. Thus farnesyl diphosphate (FPP), geranyl diphosphate
(GPP) and 2-methylgeranyl diphosphate (2-MGPP) were in-
dividually incubated with purified Pfl_1841 protein. The enzymatic
reaction mixtures were each extracted with pentane and analyzed
for terpene products by GCeMS. Incubation of Pfl_1841 with FPP
produced only trace amounts of the corresponding FPP elimination
products, E,E-a-farnesene (1%) and E-b-farnesene (1%), along with
the hydrolysis products, E-nerolidol (4%) and E,E-farnesol (1%)
(Fig. S13, SD). The very minor amounts of sesquiterpenes detected
indicated that FPP was unlikely to be the natural substrate for
Pfl_1841. Incubation of Pfl_1841 with GPP produced in low yield
a complex mixture of 14 monoterpenes (C10H16, m/z 136) and
monoterpene alcohols (C10H18O, m/z 154) (Fig. S12, SD) including
tricyclene (8) (1%),
a
-pinene (9) (4%),
a
-fenchene (7) (52%),
a-
phellandrene (10) (1%),
b-phellandrene (11) (3%),
g-terpinene (12)
(1%), terpinen-4-ol (13) (5%) and geraniol (6%), as well as minor
amounts of as yet unidentified constituents. Geraniol is most likely
generated by non-enzymatic, Mg2þ-catalyzed solvolysis of GPP
(Fig. 3). The complexity of this mixture as well as the low yield of
cyclization products suggested that GPP was also not the natural
substrate of Pfl_1841, despite the production of a-fenchene as the
predominant product. Notably, the Streptomyces coelicolor MIBS
protein, SCO7701, which utilizes 2-MGPP as its natural substrate,
can also catalyze the cyclization of GPP to a complex mixture of
monoterpenes, but at
a
very low catalytic rate (kcat¼3.0
ꢀ0.2ꢁ10ꢂ5
s
ꢂ1).5 A similar result was also observed in the cycli-
zation of GPP by the recombinant MIBS protein of Streptomyces
lasaliensis NRRL 3382R (H. Ikeda, unpublished).
Fig. 2. Operon for the 2-MIB (S. coelicolor A3(2)) and 2-MB (M. olivasterospora) bio-
synthetic gene clusters. Grey, black and shaded arrows represent genes encoding
a cyclic AMP nucleotide binding protein, a terpene synthase and a 2-methylgeranyl
diphosphate synthase, respectively. Clear arrows represent the biosynthetically un-
related P. fluorescens PfO-1 hypothetical proteins Pfl_1840 (putative dehydrogenase)
and Pfl_1842 (putative transport-related, membrane protein).
Although variations on the 2-MIB operon have also been found in
which the order of the three genes is reversed or with additional
inserted genes encoding proteins of unrelated function, the pfl_1841
gene locus is unique in that it lacks the organization found in either
the usual 2-MIB operons or that for 2-methylenebornane in M. oli-
vasterospora.7 Thus P. fluorescens PfO-1 has no apparent genes
encoding an MGPP synthase or homologues of the cyclic nucleotide
binding protein either flanking the pfl_1841 gene or clearly evident
anywhereelse in the genome. None of these three genes is present in
either of the two other genome-sequenced P. fluorescens strains. The
Fig. 3. Cyclization of GPP catalyzed by Pfl_1841.
Incubation of Pfl_1841 with 2-MGPP produced one major
homomonoterpene product (C11H18, m/z 150) as well as several
minor monomethyl-monoterpenes (Fig. 4). The major product was
identified by GCeMS as 2-methylenebornane (71%), with two other
products shown to be 1-methylcamphene (1-MC, 3%) and 2-

W.K.W. Chou et al. / Tetrahedron 67 (2011) 6627e6632
6629
2-MGPP, with a calculated kcat/KM at low substrate concentrations
of 0.8ꢁ103 Mꢂ1 sꢂ1 4,10e12
.
By comparison, the steady-state kinetic
parameters for incubation of Pfl_1841 with [1-3H]GPP were
KM(GPP) 143ꢀ12 nM and kcat 4.0ꢀ0.1ꢁ10ꢂ3
s
ꢂ1, kcat/KM (GPP)
1.9ꢁ104 Mꢂ1 sꢂ1 (Fig. S17). The observed product distributions
combined with the 10-fold greater kcat/KM for the cyclization of 2-
MGPP compared with GPP are consistent with the conclusion that
2-MGPP is the natural substrate for Pfl_1841, which is therefore
conclusively shown to be a 2-methylenebornane synthase.
Fig. 4. GCeMS trace of pentane extracts from the incubation of Pfl_1841 with 2-
MeGPP. The major identified peak corresponds to 2-methylenebornane (3, 71%),
with minor identified peaks corresponding to 1-methylcamphene (5, 3%) and
2-methylgeraniol (6, 2%). Unidentified minor products corresponding to monomethyl-
monoterpenes (C11H18, m/z 150) are peaks A (7%) and B (10%) and to monomethyl-
monoterpene alcohols (C11H20O, m/z 168) are C (1%), D (3%), and E (2%).
2.2. Search for the MGPPS in P. fluorescens PfO-1
The discovery that the Pfl_1841 protein is a 2-MB synthase
prompted the search for the corresponding MGPPS protein that
typically is encoded by a flanking gene in the MIBS biosynthetic
operons as well as in the closely related cluster that harbors the MOL
protein that also mediates the formation of 2-methylenebornane. As
mentioned earlier, no gene encoding the MGPPS protein in P. fluo-
rescens was found flanking pfl_1841. BLASTP searches of the pre-
dicted P. fluorescens proteome also did not reveal candidate proteins
with significant sequence identity to any known MGPPS proteins.
We therefore searched for candidate genes that might encode the
MGPPS in P. fluorescens PfO-1 based on the presence of two motifs
that are found in all of the actinomycete MGPP synthases, a meth-
yltransferase motif PF08241 (methyltransf_11) and a cyclopropyl
fatty acyl synthase motif PF02353. Two candidates in P. fluorescens
PfO-1 that contained both motifs were encoded by the pfl_4741 and
pfl_5666 genes. Although the corresponding predicted protein se-
quences both had very low (w20%) identity to the established
MGPPS protein from S. coelicolor (SCO7701), no other candidates
were as promising.
Synthetic genes encoding Pfl_4741 and Pfl_5666 optimized for
expression in E. coli were subcloned into pET26b (Pfl_4741) and
pET28a (Pfl_5666) vectors, respectively. Expression in E. coli
BL21(DE3) followed by purification by ion-exchange (Pfl_4741) or
Ni2þeNTA affinity chromatography (Pfl_5666) afforded the corre-
sponding recombinant proteins. Individual overnight incubation of
the Pfl_4741 or Pfl_5666 proteins with SAM and GPP was followed
by treatment either with a mixture of acid phosphatase and apyrase
to hydrolyze any 2-MGPP or with Pfl_1841 protein to convert
eventual 2-MGPP to 2-MB. The resulting treated enzymatic reaction
mixtures were extracted with pentane and analyzed by GCeMS to
assay for the production of either 2-methylgeraniol or 2-MB. Un-
fortunately, the phosphatase-treated reaction mixtures produced
only geraniol resulting from hydrolysis of the unreacted substrate
GPP, while treatment with Pfl_1841 protein gave only GCeMS
traces similar to those resulting from direct incubation of Pfl_1841
with GPP (Supplementary data). It is therefore established that
neither the pfl_4741 gene nor the pfl_5666 gene encodes the req-
uisite MGPPS protein necessary for synthesis of the 2-MGPP sub-
strate for the 2-MB synthase Pfl_1841, nor are any other candidate
genes encoding an MGPPS protein evident in the P. fluorescens
PfO-1 genome.
methylgeraniol (3%). The peaks for 2-MB and 1-MC were each
identical in both retention time and mass spectrum to authentic
reference standards in a mixture synthesized by the previously
described acid-catalyzed dehydration of 2-MIB.9 None of the
compounds in the enzymatic reaction mixture co-eluted with au-
thentic 2-methyl-2-bornene, also formed by acid-catalyzed de-
hydration of 2-MIB, or with 2-methyllimonene, prepared by
treatment of dihydrocarvone with methyllithium followed by de-
hydration with POCl3/Py. The minor product, 2-methylgeraniol, was
likely formed by non-enzymatic, Mg2þ-catalyzed solvolysis of 2-
MeGPP. Other minor components in the enzymatic mixture were
unidentified monomethyl-monoterpenes A (7%) and B (10%) and
unknown homomonoterpene alcohols C (1%), D (3%), and E (2%).
The mixture of compounds produced by the incubation of 2-MGPP
with Pfl_1841 was less complex than that for the incubation of
Pfl_1841 with GPP. Peaks AeE may correspond to monomethyl
analogs of some of the minor products formed by the Pfl_1841-
catalyzed cyclization of GPP.
The steady-state kinetic parameters for the cyclization of 2-
MGPP by Pfl_1841 were KM (2-MGPP) 110ꢀ13 nM and kcat
2.4ꢀ0.1ꢁ10ꢂ2 sꢂ1, kcat/KM (2-MGPP) 2.14ꢁ105 Mꢂ1 sꢂ1, based on the
cyclization of [1-3H]-2-MGPP (Fig. 5).5 Both the KM and kcat de-
termined for the Pfl_1841-catalyzed reaction were comparable to
those reported for most other terpene cyclases, but contrast with
the behavior of the S. coelicolor MIBS, which could not be saturated
with 2-MGPP at concentrations up to the 100 mM solubility limit of
2.3. Screening of P. fluorescens PfO-1 for terpene production
Detection of 2-MB from cell extracts of P. fluorescens PfO-1
would indicate that there is a still cryptic MGPPS that can provide
endogeneous 2-MGPP substrate for the Pfl_1841 protein. P. fluo-
rescens PfO-1 was cultured on soy flour mannitol (SFM) agar plates,
under conditions that produce monomethyl-monoterpenes in Ac-
tinomycetes.8 The SFM cultures were incubated for 1e7 days at
30 ꢃC and individual plates were extracted daily with pentane and
analyzed for volatile secondary metabolites by GCeMS. Under
these conditions no terpenoid compounds were observed in the
extracts of in vivo surface cultures of P. fluorescens. It is still possible
Fig. 5. MichaeliseMenten plot of the reaction velocity for the Pfl_1841-catalyzed for-
mation of 2-methylenebornane as a function of the concentration of 2-MGPP.

6630
W.K.W. Chou et al. / Tetrahedron 67 (2011) 6627e6632
that the production of terpene compounds in P. fluorescens might
require alternative culture conditions, as has been observed for
expression of some terpenoid biosynthetic pathways in Strepto-
myces, but this remains to be demonstrated.12
standards, was performed according to the method of Schumann
and Pendelton.9 2-Methylgeranyl diphosphate, E-2-methylgeraniol
and [1-3H]-2-methylgeranyl diphosphate were synthesized accord-
ing to the method of Wang and Cane.5 P. fluorescens PfO-1 was a gift
from Prof. Stuart B. Levy and was cultured according to previously
published procedures.3
3. Conclusion
We have established that the P. fluorescens PfO-1 pfl_1841 gene
encodes the first terpene synthase (terpene cyclase) to be identified
4.2. Methods
in any Pseudomonad species. The Pfl_1841 protein is
a
2-
All DNA manipulations were performed following standard
procedures.13 DNA sequencing was carried out at the U. C. Davis
Sequencing Facility, Davis, CA. All proteins were handled at 4 ꢃC
unless otherwise stated. Protein concentrations were determined
according to the method of Bradford, using a HewlettePackard
8452A Diode Array UV/Vis spectrophotometer with bovine serum
albumin as the standard.14 Protein purity was estimated using SDS
PAGE gel electrophoresis and visualized using Coomassie Blue stain
according to the method of Laemmli.15
GCeMS analyses of in vitro-generated compounds were per-
formed using a HewlettePackard Series 2 GCeMSD instrument
(70 eV, Electron Impact, positive ion mode) and a 30 mꢁ0.25 mm
HP5MS capillary column. The instrument method used was an in-
methylenebornane synthase that catalyzes the cyclization of 2-
MGPP to 2-MB, with steady-state kinetic parameters comparable
to those of other previously characterized bacterial, fungal, and
plant terpene synthases. Curiously, however, the associated 2-
methylgeranyl diphosphate synthase (MGPPS) gene required for
provision of the required substrate for 2-methylenebornane syn-
thase, which would normally be expected to flank the 2-MB syn-
thase gene, has not yet been identified among any of the predicted
C-methyltransferases of P. fluorescens PfO-1, nor could any terpene
products be detected in the organic extracts of surface cultures of
this organism. At the moment it is a puzzle why P. fluorescens PfO-1
would harbor a fully functional homoterpene synthase, yet not
appear to have any endogenous source of the requisite substrate for
this 2-MB synthase. It is conceivable that the endogenous MGPPS
gene differs significantly from known forms of this enzyme, or even
that P. fluorescens might have an exogenous source of 2-MGPP or 2-
methylgeraniol, such as from another bacterium in the same soil
habitat. Alternatively, the absence of both the normally conserved
flanking nucleotide binding protein and MGPPS genes may indicate
that the standalone Pfl_1841 gene may be the artifactual residue of
an earlier gene deletion, especially since there are no orthologs of
any of the genes of the usual 3-gene homoterpene synthase operon
in any of the three P. fluorescens genome sequences determined to
date. Notably these three strains of nominally the same species
share only 60% of their common gene complement.
jection volume of 1 mL, a solvent delay of 3 min and temperature
program of 60 ꢃC for 2 min, followed by a temperature gradient of
60e280 ꢃC for 11 min (20 ꢃC minꢂ1) and ending with a 2 min hold at
280 ꢃC. Comparisons to GCeMS detected compounds were done
using the Mass Finder 4.0 Database (http://www.massfinder.com).
LC-MS analyses of proteins were performed using a Thermo LXQ
linear ion trap mass spectrometer and a Phenomenex C-4 column
(2.1 mmꢁ150 mm, 5
mm pore size). The instrument method used
was a linear gradient of 5% acetonitrile in water to 95% acetonitrile
in water over 15 min at a flow rate of 200
4.3. Cloning of Pfl genes
mL/min.
Additional insights into the biosynthesis of 2-MB should be
possible by studying the functional 2-MB biosynthetic genes that
are found in other bacteria. A likely ortholog of both Pfl_1841 and
the Micromonaspora olivasteospora KY11048 proteins is the
MCAG_05469 gene product of Micromonospora sp. ATCC39149.
The predicted terpene synthase shows 38% sequence identity with
the Pfl_1841 protein and the aspartate-rich binding motif of this
protein (149DDYMVDE) is the variant found in both the 2-MB and
MIB synthases. Genes encoding a predicted cyclic nucleotide
binding protein and MGPPS protein are also present in the same
operon.
Synthetic Pfl gene constructs were codon-optimized for ex-
pression in E. coli and synthesized by DNA2.0 into separate pJ201
vectors. The Pfl genes were flanked by 50-NdeI and 30-XhoI re-
striction sites, which were used in sub-cloning the genes into
pET28a (Pfl_1841 and Pfl_5666) or pET26b (Pfl_4741) vectors.
4.4. Over-expression and purification of Pfl proteins
Recombinant Pfl plasmids were transformed into chemically
competent E. coli BL21(DE3) cells. The transformed cells were used
to inoculate 5 mL of Terrific Broth (TB) media (with 50 mg/mL of
kanamycin sulfate), and incubated overnight at 37 ꢃC with shaking
at 225 rpm. The 5 mL culture was then used to inoculate a 500 mL
4. Experimental
4.1. Materials
culture of TB media (with 50
mg/mL of kanamycin sulfate),
which was grown to an OD600 of 0.6e0.9 at 37 ꢃC and 225 rpm.
Over-expression was induced by the addition of 0.5 mM IPTG, and
the cultures were allowed to grow at 28 ꢃC for an additional 12 h.
The cells were harvested via centrifugation at 6000ꢁg for 30 min
and the resulting cell pellet flash frozen in liquid nitrogen and
stored at ꢂ80 ꢃC.
Geranyl diphosphate, S-adenosyl-L-methionine and 2-methyl-
isoborneol were purchased from SigmaeAldrich. Other reagents
and solvents were purchasedfrom SigmaeAldrichorFisherScientific,
were of the highest quality available, and were used without further
purification. Restriction enzymes and T4 DNA ligase were purchased
from New England Biolabs and used according to manufacturer
specifications. The pET28aþ and pET26b vectors were purchased
from Novagen. E. coli BL21 (DE3) and E. coli XL-10 Gold cells and the
QuikChangeÒ II XL Site-Directed Mutagenesis Kit were obtained from
Protein purification of Pfl_1841 and Pfl_5666 involved thawing
and resuspending a 500-mL cell pellet in 30 mL of lysis buffer
(50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8.0) con-
taining pepstatin A (1 mg/L) and phenylmethylsulfonyl fluoride
(1 mM). The cells were lysed with two passes through an ice-chilled
French pressure cell (10,000 psi) and the resulting cell lysate was
clarified by centrifugation at 13,000ꢁg for 1 h before loading onto
a 5 mL NieNTA column, pre-equilibrated with lysis buffer, at a flow
rate of 2 mL/min. The column was washed with 10 column volumes
of lysis buffer, followed by 10 column volumes of wash buffer
(50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, pH 8.0) and the
Stratagene. Isopropylthio
b-D-galactopyranoside (IPTG) was pur-
chased from Invitrogen. NieNTA affinity resin was purchased from
Qiagen. Amicon Ultra Centrifugal Filter Units (Amicon Ultra-15,
10,000 MWCO) were from Millipore. The acid-catalyzed de-
hydration/rearrangement of 2-methylisoborneol, to produce the 2-
methylenebornane, 1-methylcamphene, and 2-methyl-2-bornene

W.K.W. Chou et al. / Tetrahedron 67 (2011) 6627e6632
6631
Pfl_1841 proteinwas eluted off with elution buffer (50 mM NaH2PO4,
300 mM NaCl, 500 mM imidazole, pH 8.0), all at a flow rate of 2 mL/
min. All fractions containing protein were buffer exchanged into
assay buffer (50 mM PIPES, 15 mM MgCl2, 100 mM NaCl, 5 mM
containing 7 mL of Opti-Fluor. The enzymatic reaction was
extracted 3 more times with 1 mL of ether, and all organic extracts
were passed through the silica plug and collected. The combined
extracts were counted via liquid scintillation counting using
a BeckmaneCoulter LS6500 scintillation counter. Kinetic constants
were calculated using the program Kaleidagraph 4.0 and were fit-
ted to the MichaeliseMenten equation, giving values of kcat
2.4ꢀ0.1ꢁ10ꢂ2 sꢂ1 and KM (MGPP) 110ꢀ13 nM. Reported standard
deviations in the steady-state kinetic parameters represent the
calculated statistical errors in the non-linear, least squares re-
gression analysis. To determine the steady-state kinetic parameters
with GPP, incubations were carried out under the same conditions
as above using varying concentrations of GPP (125e2500 nM) and
[1-3H]-GPP (170 Ci/mol). The calculated kinetic constants were kcat
4.0ꢀ0.1ꢁ10ꢂ3 sꢂ1 and KM(GPP) 143ꢀ13 nM.
b-mercaptoethanol, 20% glycerol, pH 6.7) using Amicon Ultra Cen-
trifugal Filter units and used directly in experiments. LCeESI-MS:
His6-tag-Pfl_1841 39,595 Da (predicted MD [M-Met] 39,595 Da);
His6-tag-Pfl_5666 46,143 Da (predicted MD [M-Met] 46,142 Da).
Protein purification of Pfl_4741 involved thawing a 500 mL cell
pellet in 30 mL of lysis buffer (50 mM NaH2PO4, pH 7.0) containing
pepstatin A (1 mg/L) and phenylmethylsulfonyl fluoride (1 mM).
The cells were lysed with two passes through an ice-chilled French
pressure cell and the cell lysate was clarified by centrifugation at
13,000ꢁg for 1 h and the supernatant was loaded onto a 50 mL
DEAE Sepharose fast flow column (GE healthcare) pre-equilibrated
with lysis buffer, at a flow rate of 2 mL/min. The column was
washed with 5 column volumes of lysis buffer, before eluting the
protein with a linear gradient of 0e1 M NaCl in 50 mM of NaH2PO4
buffer, pH 7.0 (300 mL total volume). Fractions containing the
Pfl_4741 protein were collected, buffer exchanged into assay buffer
using Amicon Ultra centrifugal filter units and used directly in
further experiments.
4.8. Growth of P. fluorescens PfO-1 and analysis of volatile
organic extracts
P. fluorescens PfO-1 was grown in soy flour mannitol (SFM)
media on 90 mm agar plates at 30 ꢃC. The production of metabolites
from P. fluorescens was assessed from 1 to 7 days of growth by
adding 5 mL of methanol to a single culture, letting the culture sit
for 30 min at room temperature and extracting the methanol layer
with 3ꢁ5 mL of pentane. The pentane extracts were dried over
4.5. In vitro incubations of Pfl_1841 with FPP, GPP, and 2-
MGPP
Na2SO4, concentrated in vacuo at 0 ꢃC to 200
mL and analyzed by
To 10 mL of assay buffer (50 mM PIPES, 100 mM NaCl, 15 mM
GCeMS.
MgCl2, 5 mM
b-mercaptoethanol and 20% glycerol, pH 6.7) and
60 M of FPP, GPP or 2-MeGPP, was added 10
m
m
M of purified
4.9. Synthesis of 2-methyllimonene
Pfl_1841 protein. The enzymatic reaction was overlaid with 10 mL
of pentane and incubated at 30 ꢃC for 12 h. Following the in-
cubation, the enzymatic products were extracted with 3ꢁ10 mL of
pentane and the organic extracts were combined, dried over
To a stirred solution of 5 mL of (þ)-dihydrocarvone (30.1 mmol)
in 30 mL of THF at ꢂ78 ꢃC, was added dropwise 28 mL of a 1.6 M
solution of methyllithium in diethyl ether (45 mmol). The reaction
mixture was allowed to warm to room temperature and stirred for
1 h before the addition of 100 mL of an ice-cold solution of satu-
rated NH4Cl in water. The crude reaction mixture was extracted
with 3ꢁ50 mL of diethyl ether, dried over Na2SO4, concentrated in
vacuo and used directly in the next step.
Na2SO4 and concentrated in vacuo to 200 mL for GCeMS analysis.
4.6. In vitro incubations of Pfl_4741 and Pfl_5666 with SAM
and GPP
To 10 mL of assay buffer (50 mM PIPES, 100 mM NaCl, 15 mM
To an ice-cold solution of 1.5 mL of pyridine and 200 mL
(1.2 mmol) of the crude reaction mixture from above, was added
400 mL of POCl3 (4.4 mmol). The reaction mixture was warmed to
room temperature and stirred for 1 h before the addition of 10 mL
of ice-cold water. The reaction mixture was extracted with 3ꢁ10 mL
of pentane, dried over Na2SO4, and concentrated in vacuo. The
crude reaction mixture was loaded onto a silica column pre-
equilibrated with pentane (1 cmꢁ5 cm) and eluted with pentane
to yield 127 mg of 2-methyllimonene as a colorless oil in 71% yield.
MgCl2, 5 mM
b
-mercaptoethanol and 20% glycerol, pH 6.7) con-
taining 120 M of SAM and 60
m
m
M of GPP were added 10 M of
m
either Pfl_4741 or Pfl_5666 protein. The enzymatic reactions were
overlaid with 10 mL of pentane and incubated at 30 ꢃC for 12 h.
Following the incubation period, the enzyme reactions were split
into two equal aliquots. To one aliquot was added 5 mL of phos-
phatase solution (containing 3 units of wheat germ acid phospha-
tase and 2 units of potato apyrase per 1.0 mL of 0.1 M sodium
acetate, pH 5.0) and the reaction incubated at 30 ꢃC for 2 h. To the
1H NMR (400 MHz, CDCl3)
d 4.71e4.66 (2H, m, CH2-8), 2.18e2.09
other aliquot was added 10 mM of purified Pfl_1841 and the reaction
(1H, m, CH-4), 2.09e1.91 (2H, m, CH2-6), 1.96e1.85 (2H, m, CH2-3),
1.77e1.71 (1H, m, CHa-5), 1.72 (3H, s, CH3-9), 1.60 (3H, s, CH3-11),
incubated at 30 ꢃC for 12 h. At the end of the incubation period,
each of the reaction mixtures was extracted with 3ꢁ5 mL of pen-
tane and the organic layers from each were combined, dried over
1.59 (3H, s, CH3-10),1.40 (1H, ddd, J 5.60, 11.50,12.50 Hz, CHb-5); 13
C
NMR (100 MHz, CDCl3) 150.4 (C-7), 125.3 (C-2), 125.1 (C-1), 108.3
(C-8), 42.0 (C-4), 37.2 (C-3), 32.3 (C-6), 28.2 (C-5), 20.8 (C-9),19.1 (C-
11), 18.8 (C-10) ppm; HRMS (GCeMS, EI): Mþ, found 150.1416
(150.1409 theoretical).
Na2SO4 and reduced in vacuo at 0 ꢃC to 200
mL for GCeMS analysis.
4.7. Pfl_1841 kinetic assays with GPP and 2-MGPP
Kinetic parameters were measured in 1 mL of assay buffer
(50 mM PIPES, 15 mM MgCl2, 100 mM NaCl, 5 mM
b
-mercaptoe-
Acknowledgements
thanol, 20% glycerol, pH 6.7), with varying amounts of 2-methylGPP
substrate (25 nM-2500 nM) and [1-3H]-2-methylGPP (5.5 Ci/mol).
The reactions were initiated by the addition of 10.5 nM of Pfl_1841
protein, overlaid with 1 mL of pentane and incubated at 30 ꢃC for
5 min. The reactions were quenched by the addition of 75 mL of
500 mM EDTA (pH 8.0) and vortexing for 30 s. The pentane layer
was loaded onto a silica plug (2 cm) in a Pasteur pipette and forced
We thank Dr. Matthew Mattana and Prof. Stuart R. Levy for their
generous gift of Pseudomonas fluorescens PfO-1. We also thank
Dr. Tun-Li Shen for assistance with mass spectrometry. This work
was supported by National Institutes of Health Grant GM30301 to
D.E.C. and by a Grant-in-Aid for Scientific Research on Innovative
Areas from MEXT Japan, from JSPS 20310122 and from the Institute
for Fermentation, Osaka, Japan (H.I.).
through with
a stream of nitrogen into a scintillation vial

6632
W.K.W. Chou et al. / Tetrahedron 67 (2011) 6627e6632
3. Compeau, G.; Al-Achi, B. J.; Platsouka, E.; Levy, S. B. Appl. Environ. Microbiol.
Supplementary data
1988, 54, 2432.
4. Christianson, D. W. Chem. Rev. 2006, 106, 3412.
5. Wang, C. M.; Cane, D. E. J. Am. Chem. Soc. 2008, 130, 8908.
6. Schulz, S.; Dickschat, J. S. Nat. Prod. Rep. 2007, 24, 814.
7. Giglio, S.; Chou, W. K.; Ikeda, H.; Cane, D. E.; Monis, P. T. Environ. Sci. Technol.
2011, 45, 992.
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2011.05.084. These data in-
clude MOL files and InChIKeys of the most important compounds
described in this article.
8. Komatsu, M.; Tsuda, M.; Omura, S.; Oikawa, H.; Ikeda, H. Proc. Natl. Acad. Sci. U.S.
A. 2008, 105, 7422.
9. Schumann, R.; Pendleton, P. Water Res. 1997, 31, 1243.
10. Tetzlaff, C. N.; You, Z.; Cane, D. E.; Takamatsu, S.; Omura, S.; Ikeda, H. Bio-
chemistry 2006, 45, 6179.
References and notes
11. Cane, D. E.; He, X.; Kobayashi, S.; Omura, S.; Ikeda, H. J. Antibiot. (Tokyo) 2006,
59, 471.
12. Takamatsu, S.; Lin, X.; Nara, A.; Komatsu, M.; Cane, D. E.; Ikeda, H. Microb. Bi-
otechnol. 2011, 4, 184.
13. Sambrook, J. F. E. F.; Maniatis, T. Molecular Cloning, a Laboratory Manual, 2nd
ed.; Cold Spring Harbor: NT, 1989.
1. Gross, H.; Loper, J. E. Nat. Prod. Rep. 2009, 26, 1408.
2. Silby, M. W.; Cerdeno-Tarraga, A. M.; Vernikos, G. S.; Giddens, S. R.; Jackson, R.
W.; Preston, G. M.; Zhang, X. X.; Moon, C. D.; Gehrig, S. M.; Godfrey, S. A.;
Knight, C. G.; Malone, J. G.; Robinson, Z.; Spiers, A. J.; Harris, S.; Challis, G. L.;
Yaxley, A. M.; Harris, D.; Seeger, K.; Murphy, L.; Rutter, S.; Squares, R.; Quail, M.
A.; Saunders, E.; Mavromatis, K.; Brettin, T. S.; Bentley, S. D.; Hothersall, J.;
Stephens, E.; Thomas, C. M.; Parkhill, J.; Levy, S. B.; Rainey, P. B.; Thomson, N. R.
Genome Biol. 2009, 10, R51.
14. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
15. Laemmli, U. K. Nature 1970, 227, 680.