Biosynthesis of a novel cyclic C35-terpene via the cyclisation of a Z-type C35-polyprenyl diphosphate obtained from a nonpathogenic Mycobacterium species

Article abstract of DOI:10.1039/b808513g

Lipid components from 12 nonpathogenic Mycobacterium species were analysed. A novel cyclic C₃₅-terpene, named heptaprenylcycline 1, was obtained from 3 species, while octahydroheptaprenol 2, which has 3 Z-double bonds, was obtained from 6 species. The amounts of 1 and 2 in the cultured cells increased after the 4- to 6-d stationary phase. The yield of 1 was considerably greater at a higher temperature of 37 °C than at an optimal temperature of 28 °C, while that of 2 remained unchanged at all temperatures. A feeding experiment with d-[1-¹³C]glucose revealed that 1 was produced via isopentenyl diphosphate, which is a metabolite of glycolysis and the methylerythritol phosphate pathway. The conversion of octahydroheptaprenyl diphosphate 2-PP to 1 was successful by using the cell-free extracts of M. chlorophenolicum, demonstrating that 2-PP is the biosynthetic intermediate of 1. This is the first example of the biosynthesis of a natural terpene via the cyclisation of a linear C₃₅-isoprenoid. The substrate 2-PP for C₃₅-terpene cyclase has Z-type prenyl moieties; however, terpene cyclases usually employ E-type isoprenoids. The gene encoding the terpene cyclase that cyclises prenyl diphosphate containing Z-double bonds as the natural substrate has not yet been detected. Despite a careful search using the FASTA3 program, we could not detect any gene that is homologous to the known diphosphate-triggered type of mono-, sesqui- and diterpene cyclases in the genome of M. vanbaalenii, the DNA sequence of which has recently been elucidated. This suggests that a novel type of terpene cyclase might exist in the nonpathogenic Mycobacterium species. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b808513g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Biosynthesis of a novel cyclic C35-terpene via the cyclisation of a Z-type
C35-polyprenyl diphosphate obtained from a nonpathogenic
Mycobacterium species†
Tsutomu Sato,* Atsushi Kigawa, Ryosuke Takagi, Tomomi Adachi and Tsutomu Hoshino
Received 20th May 2008, Accepted 20th June 2008
First published as an Advance Article on the web 19th August 2008
DOI: 10.1039/b808513g
Lipid components from 12 nonpathogenic Mycobacterium species were analysed. A novel cyclic
C35-terpene, named heptaprenylcycline 1, was obtained from 3 species, while octahydroheptaprenol 2,
which has 3 Z-double bonds, was obtained from 6 species. The amounts of 1 and 2 in the cultured cells
increased after the 4- to 6-d stationary phase. The yield of 1 was considerably greater at a higher
temperature of 37 ◦C than at an optimal temperature of 28 ◦C, while that of 2 remained unchanged at
all temperatures. A feeding experiment with D-[1-13C]glucose revealed that 1 was produced via
isopentenyl diphosphate, which is a metabolite of glycolysis and the methylerythritol phosphate
pathway. The conversion of octahydroheptaprenyl diphosphate 2-PP to 1 was successful by using the
cell-free extracts of M. chlorophenolicum, demonstrating that 2-PP is the biosynthetic intermediate of 1.
This is the first example of the biosynthesis of a natural terpene via the cyclisation of a linear
C35-isoprenoid. The substrate 2-PP for C35-terpene cyclase has Z-type prenyl moieties; however, terpene
cyclases usually employ E-type isoprenoids. The gene encoding the terpene cyclase that cyclises prenyl
diphosphate containing Z-double bonds as the natural substrate has not yet been detected. Despite a
careful search using the FASTA3 program, we could not detect any gene that is homologous to the
known diphosphate-triggered type of mono-, sesqui- and diterpene cyclases in the genome of M.
vanbaalenii, the DNA sequence of which has recently been elucidated. This suggests that a novel type of
terpene cyclase might exist in the nonpathogenic Mycobacterium species.
elucidated.4 Recently, we have demonstrated that the Rv3377c
gene in the M. tuberculosis H37Rv genome encodes a novel
diterpene cyclase that plays a role in the production of the halimane
skeleton of tuberculosinyl diphosphate (Fig. 1).5 The Rv3377c
Introduction
The genus Mycobacterium includes more than 90 species.1 Several
pathogens causing diseases such as tuberculosis and leprosy
have been extensively studied due to their medical importance.2
On the other hand, a few nonpathogens are likely to be used
for bioremediation because of their ability to degrade a wide
range of environmentally toxic chemicals such as polycyclic
aromatic hydrocarbons (PAHs) and pentachlorophenols (PCPs).3
The genomes of 10 pathogenic strains (M. tuberculosis H37Rv,
M. tuberculosis CDC1551, M. tuberculosis F11, M. tuberculosis
H37Ra, M. bovis AF2122/97, M. bovis BCG Pasteur 1173P2, M.
leprae TN, M. avium 104, M. avium subsp. paratuberculosis K-
10 and M. ulcerans Agy99) have been analysed. In addition, the
genomes of 6 nonpathogens (M. vanbaalenii PYR-1, M. gilvum
PYR-GCK, Mycobacterium sp. JLS, Mycobacterium sp. KMS,
Mycobacterium sp. MCS and M. smegmatis MC2) have also been
Department of Applied Biological Chemistry, Faculty of Agriculture, and
Graduate School of Science and Technology, Niigata University, Ikarashi,
Niigata, 950-2181, Japan. E-mail: satot@agr.niigata-u.ac.jp; Fax: +81-25-
262-6854
† Electronic supplementary information (ESI) available: Amount of C35
terpenes in the M. chlorophenolicumcells cultured under different durations
and temperatures, comparison of 13C NMR spectrum of 13C-labelled 1
with natural compound 1 and structure of C35-terpenes found in nature,
search for the terpene cyclase gene from the M. vanbaalenii genome,
sesquiterpene cyclase utilising farnesyl diphosphate with Z-double bonds
as the natural substrate, NMR spectra of 1 and NMR spectra of 2. See
DOI: 10.1039/b808513g
Fig. 1 Structure of tuberculosinyl diphosphate, tuberculosinol, hep-
taprenylcycline (1) and octahydroheptaprenol (2).
3788 | Org. Biomol. Chem., 2008, 6, 3788–3794
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Nonpathogenic Mycobacterium species analysed in this study
and their type, which was determined on the basis of their yields of
compounds 1 and 2
Species
1
2
Type
I
M. chlorophenolicum
M. thermoresistibile
M. vanbaalenii
M. aichiense
᭺
᭺
᭺
–
–
–
–
–
–
–
᭺
᭺
᭺
᭺
᭺
᭺
–
–
–
–
–
II
M. smegmatis
M. parafortuitum
M. agri
III
M. aurum
M. diernhoferi
M. moriokaense
M. nonchromogenicum
M. pulveris
Fig. 2 GC-MS chromatogram displaying the total ions in nonsaponifiable
lipids. The data of M. chlorophenolicum are shown as a representative
example. The retention time of tuberculosinol was 11.2 min under the
same conditions.
–
–
–
᭺: detected and –: not detected.
chlorophenolicum cells for the isolation of 1 and 2. After a 50 L
cultivation of M. chlorophenolicum, the cells were saponified with
15% KOH–MeOH at 80 C for 30 min and were then extracted
homologous genes are limited to other Mycobacterium species
causing tuberculosis (M. tuberculosis CDC1551, M. tuberculosis
F11, M. tuberculosis H37Ra and M. bovis) and are not found in the
other species. On the basis of the findings of transposon-insertion
experiments, Pethe et al. reported that Rv3377c was indispensable
for survival in macrophages.6 Therefore, we have proposed that
the Rv3377c gene may be responsible for the pathogenicity of
tuberculosis.5
In order to confirm this hypothesis, lipid constituents from 12
nonpathogenic Mycobacterium species (Table 1) were analysed.
We detected a novel monocyclic C35-terpene 1 from 3 species
and octahydroheptaprenol 2 from 6 species (Fig. 1). Herein, we
demonstrate that 1 is biosynthesised by the cyclisation of a linear
octahydroheptaprenyl diphosphate containing Z-double bonds;
the cyclisation of this molecule has never been reported hitherto.
◦
with n-hexane. The nonsaponifiable lipids were subjected to silica
gel column chromatography, followed by reverse-phase high-
performance liquid chromatography (HPLC); this process yielded
5.5 mg and 5.1 mg of pure 1 and 2, respectively. The structure of
1 was determined by detailed nuclear magnetic resonance (NMR)
data analysis (DEPT, COSY, HOHAHA, NOESY, HMQC and
HMBC) and electron impact mass spectrometry (EIMS). The
complete NMR assignments are shown in Table 2. The molecular
formula of 1 was assigned to be C35H64 on the basis of high
resolution (HR)-EIMS. As shown in Fig. 3, a partial structure
(right side) of compound 1 (C-1–C-14 and C-28–C-30) was mainly
assigned according to the data obtained from HMBC and COSY
spectra. A clear NOE was observed between H-10 (d 5.41, t, J
H
6.8) and Me-30 (dH 1.87, s), indicating that H-10 and Me-30 were
arranged in the Z-configuration. The comparison of a chemical
shift of C-12 (32.42 ppm) with that of C-12 in ficaprenol7 (Z:
ca. 32.0 ppm, E: ca. 40.0 ppm), which was previously reported,
further supported the Z-geometry. The structure of the isobutyl
moiety on the left side of compound 1 (C-26, C-27, C-34 and C-35)
was determined by HMBC and COSY (Fig. 3) spectral analyses.
The carbon signals, which were yet to be assigned, demonstrated
3 methyl groups at dC 24.94–25.26 (blank square), 3 methylene
groups at dC 24.94–25.26 (blank circle), 2 methylene groups at
dC 37.71–37.91 (filled circle) and 3 methyne groups at dC 33.12–
33.24 (filled square) (Fig. 3). HMBC data demonstrated that the
3 isopentyl moieties represented by dotted squares in Fig. 3 were
linked to each other. The complete structure of compound 1 could
be elucidated by connecting the 3 partial structures obtained from
the HMBC cross peaks between H-12 (dH 2.21, m) and C-14
(dC 37.28) and between H-26 (dH 1.64, m) and C-25 (dC 25.26).
The stereochemistries at C-6, C-15, C-19 and C-23 remain to
be elucidated. Compound 1 is novel and its proposed name is
heptaprenylcycline.
Results and discussion
Lipid analyses of nonpathogenic Mycobacterium spp
According to the method described in the experimental section,
the hexane extract of the nonsaponifiable lipids from the cells
and the ethyl acetate (EtOAc) extract from the broth filtrate were
obtained from the 12 mycobacteria cultures; both the extracts
were subjected to gas chromatography-mass spectrometry (GC-
MS). Tuberculosinol was not detected (Fig. 2); this may support
the theory that tuberculosinol is produced only by pathogenic
species and that the Rv3377c enzyme encoding diterpene cyclase
may be involved in the pathogenicity of these bacteria. In contrast,
compounds 1 and/or 2 were detected in the nonsaponifiable
fraction obtained from the cells of several species (Fig. 2 and
Table 1). The 12 species were categorised on the basis of the
production of 1 and 2 as shown in Table 1: (type I) both 1 and 2
were produced by M. chlorophenolicum, M. thermoresistibile and
M. vanbaalenii; (type II) only 2 was produced by M. aichiense,
M. smegmatis and M. parafortuitum and (type III) both 1 and
2 were not produced by M. agri, M. aurum, M. diernhoferi, M.
moriokaense, M. nonchromogenicum and M. pulveris.
The molecular formula of 2 was assigned to be C35H66O on the
basis of HR-EIMS data. The detailed 2D-NMR analysis revealed
that 2 was octahydroheptaprenol, as shown in Fig. 1. Compounds
3 and 4 were isolated from M. smegmatis8 (Fig. 4), and compound
2 was obtained by acid hydrolysis of 3 and 4. The NMR assignment
of 2 (Table 3), however, has not been reported. NOE correlation
Isolation and structural determination of 1 and 2
Among the mycobacteria examined, M. chlorophenolicum pro-
duced the largest amounts of 1 and 2. Thus, we employed M.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3788–3794 | 3789
©

View Article Online
Table 2 NMR assignments of 1 in C6D6 and abundance of 13C-labeled 1
III species. M. chlorophenolicum and M. vanbaalenii can degrade
PCPs and PAHs, respectively.3 M. thermoresisti◦bile can survive and
propagate at elevated temperatures of up to 55 C.9 It is interesting
to elucidate the functions of compound 1 in Mycobacterium cells.
As the first step towards the functional analysis of 1, we determined
the amounts of 1 and 2 produced in M. chlorophenolicum cells
under varied culture conditions—time and temperature. The
yields of 1 and 2 increased after the 4- to 6-d stationary phase.
Similar results have been observed for many secondary metabolites
produced by micro-organisms. Moreover, the yield of compound
1 was considerably greater at a higher temperature of 37 ◦C than
at an optimal growth temperature (28 ◦C), while the amount of 2
remained unchanged at all temperatures. A similar phenomenon
has been noted during the analysis of mycolic acid: in several
Mycobacterium species, the amount and composition of mycolic
acid varied at higher temperatures.9,10 The adaptive changes in
mycolic acids in response to temperature changes are assumed to
play a role in the maintenance of suitable membrane function.9,10
It was proposed that 3 may be involved in the transport of
mycolic acids through the plasma membrane in order to constitute
the cell-wall component of a covalently linked peptidoglycan–
arabinogalactan–mycolate framework.8 Since both 1 and 3 may be
biosynthesised via a common intermediate octahydroheptaprenyl
diphosphate 2-PP (Fig. 4), the function of 1 may be closely related
to that of mycolic acid. However, further experiments are required
to establish the functions of 1.
No.
dH (mult., J in Hz)
dC (mult.)
13C Enrichmentb
1
2
3
4
5
6
7
8
9
2.11 (m); 2.29 (m)
5.56 (1H, br t)
—
1.97 (2H, m)
1.63 (m)
2.23 (2H, m)
—
2.26 (2H, m)
2.41 (2H, m)
5.41 (1H, t, 6.8)
—
2.21 (2H, m)
1.57 (m); 1.62 (m)
1.29 (m); 1.51 (m)
1.61 (m)
1.29 (m); 1.51 (m)
1.42 (m); 1.54 (m)
1.29 (m); 1.51 (m)
1.61 (m)
1.29 (m); 1.51 (m)
1.42 (m); 1.54 (m)
1.29 (m); 1.51 (m)
1.61 (m)
1.29 (m); 1.51 (m)
1.42 (m); 1.54 (m)
1.64 (m)
31.83 (t)
121.26 (d)
133.43 (s)
30.96 (t)
4.5
1.1
1.1
1.0
3.1
1.6
1.5
1.0
4.0
1.4
1.7
1.0
2.8
1.1
1.0
1.0
4.1
1.6
1.5
1.6
4.1
1.6
1.1
1.7
4.5
1.7
1.2
4.5
3.0
2.9
3.0
3.1
3.4
2.6
1.0
28.68 (t)
40.15 (d)
154.06 (s)
35.63 (t)
27.18 (t)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
125.37 (d)
135.59 (s)
32.42 (t)
25.84 (t)
37.28 (t)
33.12 (d)a
37.71 (t)a
24.95 (t)a
37.84 (t)a
33.22 (d)a
37.85 (t)a
24.97 (t)a
37.90 (t)a
33.24 (d)a
37.91 (t)a
25.26 (t)a
28.32 (t)
1.31 (m)
1.75 (3H, s)
39.73 (d)
23.62 (q)a
107.80 (t)
23.62 (q)a
19.91 (q)a
19.98 (q)a
20.03 (q)a
22.789 (q)a
22.89 (q)a
5.03 (1H, s); 5.05 (1H, s)
1.87 (3H, s)
1.06 (3H, d, 6.8)
1.07 (3H, d, 6.8)
1.08 (3H, d, 6.7)
1.04 (3H, d, 6.6)
1.04 (3H, d, 6.6)
Biosynthesis of 1
As shown in Fig. 4, we propose that the pathway for the biosyn-
thesis of 1 is as follows. The carbon framework of 1 is produced
by the head-to-tail condensations of 7 isopentenyl diphosphate
(IPP) units, followed by hydrogenation reactions that occur at 4
double bonds to give rise to 2-PP and a diphosphate-triggered type
of cyclisation to form the complete structure of compound 1. The
detection of 2 strongly supports the existence of the intermediate
2-PP, because 2 is produced by the dephosphorylation of 2-
PP. It has been reported that mannosyl-b-1-phosphoisoprenoid 5
bearing a C32 polyprenyl-like moiety with a structure analogous
to those of 3 and 4, is found in M. tuberculosis, and that the C32
moiety may be produced by polyketide synthase.11 The proposed
elongation reaction is primed by a fatty acyl group and proceeds
by alternate condensations of methylmalonate and malonate in
repeated cycles to yield a carboxylic acid product.11 In order to
identify the biosynthetic pathway involved in the production of
1, we carried out a feeding experiment using D-[1-13C]glucose. M.
chlorophenolicum was cultured in a 10 L medium with 10% isotopic
abundance of D-[1-13C]glucose, and 0.2 mg of 13C-labelled 1 was
isolated. 13C NMR analysis revealed that the isotopic abundant
signals (Table 2) of 1 were in agreement with the position of the
filled circles shown in Fig. 4; this unambiguously demonstrated
a The assignments may be exchangeable. b The values for enrichments
were determined by comparison of the relative peak intensities of the
corresponding carbons in labelled and unlabelled spectra. The bold type
shows the 13C-enriched position (>2.5-fold).
and the chemical shifts of C-4, C-8 and C-12 (32.42, 32.50 and
32.57 ppm)7 suggested that all the 3 prenyl residues in 2 have
Z-configurations. The stereochemistries at C-15, C-19 and C-23
could not be determined. Compound 2 was not detected in the
nonsaponified (no alkaline-treatment) samples obtained from all
the 12 Mycobacterium species. The octahydroheptaprenyl moieties
of compounds 3 and 4 are stable under alkaline conditions,8
strongly indicating that an unknown compound possessing 2 as
the moiety exists in type I and II species.
Yields of 1 and 2 under different culture conditions
Species that produce 1 are categorised as type I species. These
exhibited special abilities that were not found in type II and
Fig. 3 Significant correlations in the NMR spectra of compound 1.
3790 | Org. Biomol. Chem., 2008, 6, 3788–3794 This journal is The Royal Society of Chemistry 2008
©

View Article Online
Fig. 4 The proposed pathway for the biosynthesis of C35-terpenes from Mycobacteium species. The filled circle indicates the 13C-enriched positions at
which compound 1 could be produced via IPP by glycolysis of D-[1-13C]glucose and the MEP pathway. The blank square of 1 indicates the 13C-enriched
positions at which 1 could be produced using methylmalonate and malonate by polyketide synthase. The blank triangle of IPP indicates the 13C-enriched
positions at which IPP could be produced by glycolysis of D-[1-13C]glucose and the mevalonate pathway.11 Compounds 3 and 4 were obtained from M.
smegmatis.4 It has been proposed that compound 5 may be biosynthesised by polyketide synthase in M. tuberculosis.8
Table 3 NMR assignments of 2 in C6D6
that compound 1 was produced from IPP, which is produced by
glycolysis and the methylerythritol phosphate (MEP) pathway.12
If 1 was biosynthesised by polyketide synthase or the mevalonate
pathway, a different incorporation pattern should be observed as
shown in Fig. 4.
The bioconversion of 2-PP to compound 1 is anticipated.
Compound 2 was diphosphorylated using the method described
by Davisson et al.13 The product 2-PP was then incubated
with the cell-free extracts of M. chlorophenolicum. As shown
in Fig. 5, the amount of 1 significantly increased by the ad-
dition of 2-PP (>7-fold) compared with that of the control
(heat denaturation of the cell-free extracts or no addition of
2-PP); thus, 2-PP would be the biosynthetic intermediate of
1, as illustrated in Fig. 4. To the best of our knowledge, the
known C35-terpenes include 8 cyclic members: plagiospirolides
No.
dH (mult., J in Hz)
dC (mult.)
1
2
3
4.14 (2H, br dd)
5.53 (1H, t, 6.8)
—
59.03 (t)
125.97 (d)
139.14 (s)
32.42 (t)a
26.79 (t)a
125.23 (d)
135.77 (s)
32.50 (t)a
26.86 (t)a
125.23 (d)
135.77 (s)
32.57 (t)a
25.94 (t)
4
5
2.17 (2H, m)
2.24 (m)
6
7
5.29 (1H, t, 6.8)
—
8
9
2.24 (m)
2.32 (2H, m)
5.39 (1H, t, 6.8)
—
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
2.24 (m)
1.57 (m)
1.31 (m); 1.48 (m)
1.60 (m)
1.31 (m); 1.48 (m)
1.41 (m); 1.54 (m)
1.31 (m); 1.48 (m)
1.60 (m)
1.31 (m); 1.48 (m)
1.41 (m); 1.54 (m)
1.31 (m); 1.48 (m)
1.60 (m)
1.31 (m); 1.48 (m)
1.41 (m); 1.54 (m)
1.35 (m)
37.36 (t)a
33.25 (d)a
37.71 (t)a
24.96 (t)a
37.85 (t)a
33.21 (d)a
37.85 (t)a
24.96 (t)a
37.85 (t)a
33.25 (d)a
37.90 (t)a
25.25 (t)a
39.73 (t)
28.32 (d)
23.48 (q)
23.55 (q)
23.62 (q)
19.93 (q)a
19.97 (q)a
20.02 (q)a
22.78 (q)a
22.88 (q)a
1.65 (m)
1.77 (3H, s)
1.85 (3H, s)
1.87 (3H, s)
1.07 (3H, d, 6.6)
1.07 (3H, d, 6.6)
1.09 (3H, d, 6.7)
1.04 (3H, d, 6.6)
1.04 (3H, d, 6.6)
Fig. 5 The amount of 1 produced by the incubation of 2-PP with the
cell-free extracts of M. chlorophenolicum. The relative amount of 1 was
estimated by the abundance of the molecular ion m/z 484. A: 1 mg of 2-PP
was incubated with the cell-free extracts. B: 1 mg of 2-PP was incubated
with heat-denatured cell-free extracts. C: no substrate was incubated with
the cell-free extracts.
a The assignments may be exchangeable.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3788–3794 | 3791
©

View Article Online
A–D,14 ferrugicadinol,15 ferrugieudesmol,15 tetraprenylated-a-
curcumene16 and tetraprenylated-b-curcumene16 and 5 acyclic
members.8,17–19 Since the known cyclic C35-terpenes are hybrids
of sesquiterpene and diterpene,14–16 compound 1 would be the first
example of a terpene that can be biosynthesised via the cyclisation
of a linear C35-isoprenoid such as 2-PP.20
prenyltransferases,25 a chimera diterpene synthase having both
prenyltransferases and terpene cyclase activity,26 MEP pathway12
and a new biosynthetic pathway for menaquinone production.27
Therefore, a careful analysis of both terpene metabolites and
genome sequences may be useful in detecting a new biosynthetic
enzyme.
The reaction of head-to-tail condensation, hydrogenation,
and cyclisation in the biosynthesis of 1 (Fig. 4) would be
catalysed by Z-prenyltransferase, prenylreductase and terpene
cyclase, respectively. Since a complete genome sequence of M.
vanbaalenii, which has been categorised as a type I species
in this paper, was released on the website http://genome.jgi-
psf.org by the Joint Genome Institute in 2006, we attempted
to search a candidate gene for the biosynthesis of compound
1 in the M. vanbaalenii genome using the FASTA3 program
of Genome Information Broker (http://gib.genes.nig.ac.jp/). The
homologs of Z-prenyltransferase and prenylreductase having a
very low E value were detected (e.g. Mvan_3822 vs. undecaprenyl
Experimental
Analytical method
NMR spectra were recorded on a Bruker DMX 600 spectrometer
at 600 MHz for proton and at 125 MHz for carbon. GC was
performed on a Shimadzu GC-8A instrument equipped with a
flame-ionisation detector and a DB-1 capillary column (30 m ¥
0.25 mm; J & W Scientific. Inc.). GC-MS was performed on an
SX 102 spectrometer (JEOL) by electronic impact at 70 eV with a
DB-1 capillary column or a JMS-Q1000 GC K9 (JEOL) by
electronic impact at 70 eV with a ZB-5 ms capillary column (30 m ¥
0.25 mm; Zebron). HRMS was performed using a spectrometer
equipped with a direct inlet system. The specific rotation was
measured at 25 ◦C by using a Horiba SEPA-300 polarimeter.
diphosphate synthase from Micrococcus luteus [E = 2.1 ¥ 10-39
]
and Mvan_2055 vs. geranylgeranyl reductase from Synechocystis
sp. PCC6803 [E = 3.6 ¥ 10-9]), while any enzyme homologous
to the known diphosphate-triggered type of mono-, sesqui- and
diterpene cyclases could not be detected (E value > 0.01).21 This
result suggested that a novel type of terpene cyclase might exist
in type I nonpathogenic Mycobacterium species. The C35-terpene
cyclases utilise 2-PP, which contains Z-type prenyl moieties, while
the standard cyclases utilise E-polyprenyl diphosphate. To the
best of our knowledge, there is only one report on the use of
Z-prenyl diphosphate as a natural substrate by terpene cyclases:
Nabeta et al. reported that sesquiterpene cyclases producing
(-)-g-cadinene and germacrene D from Heteroscyphus planus
utilised (2Z,6E)-farnesyl diphosphate.22 However, the gene coding
for Z-terpene cyclase has not yet been detected. It has been
reported that not only the primary but also the tertiary structure
of Z-prenyltransferase is considerably different from that of the
E- prenyltransferase.23 Therefore, the C35-terpene cyclase that is
situated downstream of the Z-prenyltransferase in the biosynthetic
pathway might differ structurally from the known E-terpene
cyclase.
Standard culture conditions
The following 12 nonpathogenic Mycobacterium species were em-
ployed in the study: M. agri JCM 6377, M. aichiense JCM 6376, M.
aurum JCM 6366, M. chlorophenolicum JCM 7439, M. diernhoferi
JCM 6371, M. moriokaense JCM 6375, M. nonchromogenicum
JCM 6364, M. parafortuitum JCM 6367, M. pulveris JCM 6370,
M. smegmatis JCM 6386, M. thermoresistibile JCM 6362 and
M. vanbaalenii JCM 13017. All species were reciprocally shake-
cultured at an optimal temperature (28 ◦C for M. chlorophenolicum
and M. vanbaalenii and 37 ◦C for the others) until the stationary
growth phase was achieved (M. chlorophenolicum, 4 d; remaining
species, 3–14 d); they were shake-cultured in 3000 mL Sakaguchi
flasks, each containing 1000 mL of medium containing 1%
polypeptone (Nihon Seiyaku Co.), 0.5% yeast extract (Oxoid
Co.), 0.5% malt extract (Difco Co.), 0.5% casamino acids (Nihon
Seiyaku Co.), 0.2% glycerol (Wako Co.), 0.005% Tween 80 (Wako
Co.) and 0.1% MgSO4·7H2O (Kanto Chemical Co.).
Conclusion
Preparation and GC-MS analysis of the lipid fraction obtained
from the cells and of the broth filtrate
The known cyclic C35-terpenes can be regarded as the derivatives
of sesqui- and diterpenes. In contrast, compound 1 cannot belong
to the well-known terpene family including mono-, sesqui-, di-,
sester-, tri- and tetraterpenes because it is biosynthesised from the
linear C35-isoprenoid. The function of the rare C35-terpene 1 in
the Mycobacterium cell may be interesting, because mycobacteria
adjust the amount of compound 1 produced according to their
growth phase or surrounding temperature. In addition, the
diterpene of (-)-axinyssene is known to be a terpene hydrocarbon
having a cyclic skeleton similar to that of compound 1, which
shows mild toxicity against acute promyelocytic leukemia HL-60
cell line.24 Thus, the bioactivity of 1 in other organisms might
also be interesting. Furthermore, the genome analysis of M.
vanbaalenii revealed that the C35-terpene cyclase producing 1 may
be novel. Recently, unusual enzymes and pathways for terpene
biosynthesis have been discovered. These are represented by a new
class of prenyltransferases having no similarity with the known
The cells and broth filtrate of each Mycobacterium sp. obtained
from the 1 L culture medium were separated by centrifugation
(6000 ¥ g at 4 ◦C for 10 min). After lyophilisation, the cells
were saponified with 15% methanolic KOH (600 mL) at 80 ◦C
for 30 min. The nonsaponifiable lipids were extracted with n-
hexane (600 mL ¥ 3); the broth filtrate was extracted with EtOAc
(1 L ¥ 3). GC-MS (JEOL SX 102) was performed ◦under the
following conditions: an injection temperature of 290 C and an
oven temperature of 180–270 ◦C at an increment of 3 ◦C min-1.
Isolation of compounds 1 and 2
M. chlorophenolicum was cultured in a 50 L medium at 28 ◦C for
4 d. The nonsaponifiable lipids from lyophilised cells (dry weight,
195 g) were extracted as described above and concentrated by
3792 | Org. Biomol. Chem., 2008, 6, 3788–3794
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
vacuum drying to obtain 5.1 g of the residue. The n-hexane extract
was partially purified by silica gel (500 g) column chromatography
with n-hexane and EtOAc. The fraction eluted with n-hexane
(144 mg) was separated by subjecting it to silica gel (40 g) column
chromatography with n-hexane in order to obtain 5.5 mg of pure
compound 1. The fraction eluted with EtOAc (4.5 g) was subjected
to silica gel (300 g) column chromatography via gradient elution
using a mixed solvent comprising n-hexane and EtOAc (100 : 1
→ 100 : 5). The complete purification of compound 2 (5.1 mg)
was achieved by reverse-phase HPLC (Shiseido, CAPCELL PAK
C18) with MeOH–H2O (100 : 5).
added to the solution (0.3 mL) containing 1 mg substrate (2-PP),
10 mM MgCl2 and 10 mM MnCl2. 2-PP was synthesised from
the isolated compound 2 according to the method described by
Davisson et al.13 After incubation at 28 ◦C for 12 h, the reaction
was quenched by adding 5 mL of MeOH. The reaction mixture
was extracted with n-hexane (10 mL ¥ 3) and analysed by GC-MS
(JMS-Q1000 GC K9) under the following conditions: an injection
temperature of 290 ◦C and an oven temperature of 180–270 ◦C
◦
at an increment of 3 C min-1. The abundance of the molecular
ion m/z 484 was measured for estimating the relative amount of
product 1. These incubation experiments were carried out twice
and similar results were obtained on both occasions.
Instrumental data of compound 1
Colourless oil; [a]2D5 +80.5 (c 0.35, CHCl3); NMR assignment is
shown in Table 2. EIMS: m/z 484 (66), 387 (59), 189 (31), 161 (86),
135 (39), 121 (29), 119 (55), 107 (28), 94 (82), 81 (64), 71 (65), 67
(74) and 57 (100). HR-EIMS: m/z 484.5014 [M]+ (calculated for
C35H64, 484.5008).
Acknowledgements
This work was supported by a Grant-in-Aid for Young Scientists
(B) from Japan Society for the Promotion of Science (no.
19780085) and a grant for the promotion of the Niigata University
Research Project.
Instrumental data of compound 2
Colourless oil; [a]2D5 +134.09 (c 0.022, CHCl3); NMR assignment
is shown in Table 3. EIMS: m/z 502 (14), 484 (28), 416 (15), 189
(23), 161 (36), 135 (29), 122 (47), 109 (52), 95 (75), 81 (100), 71 (76),
69 (80) and 57 (94). HR-EIMS: m/z 502.5122 [M]+ (calculated for
C35H66O, 502.5114).
Notes and references
1 G. Devulder, M. Pe’ rouse de Montclos and J. P. Flandrois,
Int. J. Syst. Evol. Microbiol., 2005, 55, 293; J. P. Euze¢by’s List of
Bacterial Names with Standing in Nomenclature (http://www.bacterio.
cict.fr/m/mycobacterium.html).
2 T. Hussain, Crit. Rev. Microbiol., 2007, 33, 15.
3 S. Brandt, A. P. Zeng and W. D. Deckwer, Biotechnol. Bioeng., 1999,
65, 93; J. Moody, J. P. Freeman, P. P. Peter and C. E. Cerniglia, Appl.
Environ. Microbiol., 2004, 70, 340.
Analysis of the yield of C35-terpenes produced
under different culture conditions
4 P. R. Marri, J. P. Bannantine and G. B. Golding, FEMS Microbiol. Rev.,
2006, 30, 906; Genome Information Broker (http://gib.genes.nig.ac.jp).
5 C. Nakano, T. Okamura, T. Sato, T. Dairi and T. Hoshino, Chem.
Commun., 2005, 1016.
6 K. Pethe, D. L. Swenson, S. Alonso, J. Anderson, C. Wang and D. G.
Russell, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 13642.
7 Y. Tanaka and M. Takagi, Biochem. J., 1979, 183, 163.
8 K. Takayama, H. K. Schnoes and E. J. Semmler, Biochim. Biophys.
Acta, 1973, 316, 212; G. S. Besra, T. Sievert, R. E. Lee, R. A. Slayden
and P. J. Brennan, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 12735; B. A.
Wolucka and E. Hoffmann, Glycobiology, 1998, 8, 955.
9 L. Kremer, Y. Guerardel, S. S. Gurcha, C. Locht and G. S. Besra,
Microbiology, 2002, 148, 3145.
10 S. Toriyama, I. Yano, M. Masui, E. Kusunose, M. Kusunose and N.
Akimori, J. Biochem., 1980, 88, 211; T. Baba, K. Kaneda, E. Kusunose,
M. Kusunose and I. Yano, J. Biochem., 1989, 106, 81.
11 I. Matsunaga, A. Bhatt, D. C. Young, T. Cheng, S. J. Eyler, G. S. Besra,
V. Briken, S. A. Porcelli, C. E. Costello, W. R. Jacobs, Jr and D. B.
Moody, J. Exp. Med., 2004, 200, 1559.
M. chlorophenolicum was cultured in a 1 L medium under varied
culture conditions—duration (1–8 d at 28 ◦C) or temperature
◦
◦
◦
(20 C, 28 C or 37 C for 8 d). The abundance of m/z 484 was
measured by GC-MS for estimating the yield of products 1 and 2.
Feeding experiment with D-[1-13C]glucose
D-[1-13C]Glucose (99% isotopic abundance) was purchased from
Cambridge Isotope Laboratories, Inc. (USA). M. chloropheno-
licum was grown at 28 ◦C for 3 d. A medium containing 0.5% yeast
extract (Oxoid Co.), 0.5% malt extract (Difco Co.), 1% CaCO3
(Wako Co.), 0.9% D-glucose (Wako Co.) and 0.1% D-[1-13C]glucose
was used. 13C-labelled 1 was isolated using a similar method as that
employed for natural compound 1 as described above, yielding
0.2 mg of pure compound 1. The 13C NMR spectra of 13C-labelled
1 and natural compound 1 were compared as shown in the ESI.†
12 M. Rohmer, Pure Appl. Chem., 2003, 75, 375.
13 V. J. Davisson, A. B. Woodside and C. D. Poulter, Methods Enzymol.,
1985, 110, 130.
14 J. Sporle, H. Becker, M. P. Gupta, M. Veith and V. Huch, Tetrahedron,
1989, 45, 5003; J. Sporle, H. Becker, N. S. Allen and M. P. Gupta,
Phytochemistry, 1991, 30, 3043.
Preparation of cell-free extracts and incubation
with the 2-PP substrate
15 C. Hseih, M. Tseng, Y. Shao, J. Chang, C. Kuo, C. Chang and Y. Kuo,
J. Nat. Prod., 2006, 69, 1611.
16 K. Boroczky, H. Laatsch, I. Wagner-Dobler, K. Stritzke and S. Schulz,
Chem. Biodiversity, 2006, 3, 622.
17 M. J. L. Hoefs, J. S. Sinninghe Damste and J. W. de Leeuw, Org.
Geochem., 1995, 23, 263.
18 B. P. Manquin, J. A. Morgan, J. Ju, T. Muller-Spath and D. S. Clark,
Extremophiles, 2004, 8, 13.
19 V. Costantino, E. Fattorusso, A. Mangoni, M. D. Rosa and A. Ianaro,
Tetrahedron, 2000, 56, 1393.
20 J. D. Connolly and R. A. Hill, Dictionary of terpenoids, Chapman &
Hall, New York, 1992; S. D. Barton, K. Nakanishi and O. Meth-Cohn,
M. chlorophenolicum was cultured at 28 ◦C for 60 h in a 6 L
medium and centrifuged at 6000 ¥ g at 4 ◦C for 20 min. After
washing the cells with 50 mM of methyl-3-aminopropanesulfonic
acid (MOPS; pH 7.9), the washed cells (wet weight, 41.4 g) were
suspended in 100 mL of 50 mM MOPS (pH 7.9) containing
5 mM dithiothreitol, 0.3% Triton X-100, 10 mM KF and 0.25 M
sucrose, followed by sonication at 4 ◦C for 30 min to prepare the
cell-free extract; the supernatant thus obtained was used as the
enzyme source. The prepared cell-free homogenate (5 mL) was
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3788–3794 | 3793
©

View Article Online
Comprehensive Natural Products Chemistry, Elsevier, Oxford, United
Kingdom, 1999; P. M. Dewick, Nat. Prod. Rep., 2002, 19, 181.
21 W. R. Pearson, Methods Mol. Biol., 2000, 132, 185.
22 K. Nabeta, M. Fujita, K. Katayama and T. Takasawa, J. Chem. Soc.,
Perkin Trans. 1, 1997, 2065.
23 S. Takahashi and T. Koyama, Chem. Rec., 2006, 6, 194.
24 K. Kodama, R. Higuchi, T. Miyamoto and R. W. Van Soest, Org. Lett.,
2003, 5, 169.
25 F. Pojer, E. Wemakor, B. Kammerer, H. Chen, C. T. Walsh, S.
Li and L. Heide, Proc. Natl. Acad. Sci. U. S. A., 2003, 100,
2316.
26 T. Toyomasu, M. Tsukahara, A. Kaneko, R. Niida, W. Mitushashi, T.
Dairi, N. Kato and T. Sassa, Proc. Natl. Acad. Sci. U. S. A., 2007, 104,
3084.
27 H. Seto, Y. Jinnai, T. Hiratsuka, M. Fukawa, K. Furihata, N. Itoh and
T. Dairi, J. Am. Chem. Soc., 2008, 130, 5614.
3794 | Org. Biomol. Chem., 2008, 6, 3788–3794
This journal is
The Royal Society of Chemistry 2008
©