Towards structural determination of the ComX pheromone: Synthetic studies on peptides containing geranyltryptophan

Article abstract of DOI:10.1271/bbb.68.2374

Bacteria produce and respond to signal molecules depending on their cell density. This process is called "quorum sensing". The ComX pheromone, controlled by quorum sensing, activates natural genetic competence in Bacillus subtilis. ComX is an oligopeptide

Full text of DOI:10.1271/bbb.68.2374

This article was downloaded by: [SUNY Health Science Center]
On: 07 April 2015, At: 01:22
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Towards Structural Determination of the ComX
Pheromone: Synthetic Studies on Peptides
Containing Geranyltryptophan
Masahiro OKADAa, Isao SATOa, Soo Jeong CHOb, Yoshihiro SUZUKIa, Makoto OJIKAa, David
DUBNAUb & Youji SAKAGAMIa
a Graduate School of Bioagricultural Sciences, Nagoya University
b Public Health Research Institute
Published online: 22 May 2014.
To cite this article: Masahiro OKADA, Isao SATO, Soo Jeong CHO, Yoshihiro SUZUKI, Makoto OJIKA, David DUBNAU & Youji
SAKAGAMI (2004) Towards Structural Determination of the ComX Pheromone: Synthetic Studies on Peptides Containing
Geranyltryptophan, Bioscience, Biotechnology, and Biochemistry, 68:11, 2374-2387, DOI: 10.1271/bbb.68.2374
To link to this article: http://dx.doi.org/10.1271/bbb.68.2374
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Biosci. Biotechnol. Biochem., 68 (11), 2374–2387, 2004
Towards Structural Determination of the ComX Pheromone:
Synthetic Studies on Peptides Containing Geranyltryptophan
Masahiro OKADA,1 Isao SATO,1 Soo Jeong CHO,2 Yoshihiro SUZUKI,1
y
1;
Makoto OJIKA,1 David DUBNAU,2 and Youji SAKAGAMI
1Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa-ku, Nagoya 464-8601, Japan
2Public Health Research Institute, 225 Warren Street, Newark, NJ 07103, U.S.A.
Received August 2, 2004; Accepted September 8, 2004
Bacteria produce and respond to signal molecules
depending on their cell density. This process is called
‘‘quorum sensing’’. The ComX pheromone, controlled
by quorum sensing, activates natural genetic compe-
tence in Bacillus subtilis. ComX is an oligopeptide with a
posttranslational modification. It has been suggested
that ComX pheromone is modified with an isoprenoid at
its tryptophan residue, but the complete chemical
structure is unknown. We first determined the molecu-
lar formula of ComXRO-E-2, a competence factor for
B. subtilis strain RO-E-2. Then we synthesized putative
pheromones with 1-, 2-, 4-, 5-, 6-, or 7-geranyl sub-
stituted tryptophan residues. The regio- and stereo-
selective synthesis of the geranyl tryptophans was
successful, and we prepared the six peptides with
modified tryptophan residues. These peptides had the
same molecular formula and showed similar hydro-
phobicity to the natural ComXRO-E-2 in LC–MS analysis.
But, none of them showed the same retention time as the
natural pheromone and none exhibited its biological
activity. These results suggest that the isoprenoid
modification pattern of the tryptophan residue is more
complex than postulated.
tones with species-specific acyl chain lengths. In the
case of Gram-positive bacteria, the quorum sensing
pheromones are usually oligopeptides.
Bacillus subtilis is a Gram-positive bacterium that
secretes several quorum sensing pheromones, including
ComX, which stimulates natural genetic competence.
The ComX pheromone of B. subtilis 168, the commonly
used laboratory strain, is an oligopeptide possessing 10
amino acids, derived from the C-terminal codons of the
comX gene.2) The purified peptide has the same amino
acid sequence as that determined from the comX DNA
sequence except that a predicted tryptophan residue is
not recovered by Edman degradation. The molecular
weight of the pheromone is 206 Daltons higher than that
of the simple peptide, suggesting that it receives
posttranslational modification at the position of the
tryptophan residue. It has been shown that ComQ, the
enzyme responsible for modifying ComX, resembles
isoprenyl transferase in its primary sequence,3) and that
tritium labeled mevalonic acid was incorporated into the
ComX pheromone molecule.4) Hence the modification
of the strain 168 ComX pheromone was predicted to be
a farnesyl chain, in accordance with the molecular
weight of the modification. It appears likely that an
isoprenoid chain modifies the ComX pheromone, al-
though its complete structure is unknown.
Key words: Bacillus subtilis; ComX; posttranslational
modification; quorum sensing
To investigate the chemical structure of the ComX
pheromone, we first determined its molecular formula
by obtaining a high-resolution mass spectrum (HRMS).
From the results obtained, we estimate that the isopre-
noidal moiety is simply substituted for a proton in the
tryptophan residue. We have considered two approaches
for determining the structure. The first approach is to
obtain an NMR spectrum of a purified ComX pher-
omone and the other is to synthesize peptides with all
the possible simple structures of the natural ComX
pheromone. The preparation of enough natural ComX
pheromone for NMR analysis proved to be difficult
Bacteria monitor their environments, allowing adap-
tation to changing conditions. Among the sensor systems
employed for this purpose, quorum sensing is a
mechanism that determines cellular behavior in response
to cell density.1) To accomplish this, quorum sensing
detects the concentrations of pheromones secreted into
the external medium. Among the responses governed by
quorum sensing are biofilm formation, sporulation,
conjugation, bioluminescence, genetic competence, and
the acquisition of virulence. The pheromones of Gram-
negative bacteria are generally N-acylhomoserine lac-
y
To whom correspondence should be addressed. Tel: +81-52-789-4116; Fax: +81-52-789-4118; E-mail: ysaka@agr.nagoya-u.ac.jp
Abbreviations: Clt, 2-chlorotrityl; DIPEA, N,N0-diisopropylethylamine; dppf, diphenylphosphinoferrocene; Ger, geranyl; HATU, O-(7-
azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate; HBTU, O-benzotriazol-1-yl-N,N,N0,N0-tetramethyluronium hexafluorophos-
phate; LDA, lithium diisopropylamide; Phth, phthaloyl; Pp, 2-phenyl-2-propyl; Su, succinimide; TMEDA, N,N,N0,N0-tetramethylethylenediamine

Synthetic Studies on Putative ComX Pheromone
2375
because the concentration of secreted pheromone in the
medium is low. Consequently, the latter approach was
adopted, and we have elected to study the simplest
pheromone, produced by B. subtilis strain RO-E-2.
ComXRO-E-2 contains only six amino acid residues and
is predicted to have a geranyl moiety, based on
arguments analogous to those described above for the
ComX168 pheromone.4)
10 ꢁl). The mixture was introduced for ESI (positive)–
TOF mass spectroscopy with aqueous 50% CH3CN and
0.05% formic acid at a flow rate of 5 ꢁl/min and
recorded. Two polypropylene glycol fragment ions
(889.6458 and 947.6877 ½M þ NH4ꢁþ) were used as an
internal standard for calibration. The ComXRO-E-2 ion
½M þ Hꢁþ was found at m=z 915.4983, which was
assigned as C48H67N8O10 (Calcd. 915.4975) with an
error of þ0:83 ppm.
Materials and Methods
Synthesis of putative pheromones.
General. High performance liquid chromatography
was performed on an HPLC system equipped with Jasco
PU-980 pumps and a Jasco UV-980 UV detector or with
Shimadzu LC-6A pumps and a Shimadzu SPD-6A UV
detector. HRMS (ESI-TOF, positive) was recorded on a
Mariner system (Applied Biosystems) using either an
angiotensin/bradykinin/neurotensin mixture or a poly-
propylene glycol solution as a calibration standard. LC–
MS spectra (ESI-Q, positive) were recorded on an
API2000 mass spectrometer (Applied Biosystems) and
an HPLC (Hewlett-Packard Agilent LC system). The
optical density was measured with an AE-15F photo-
electric colorimeter (Erma).
General. All starting materials and reagents were
purchased from commercial sources (Aldrich, Sigma,
Wako Pure Chemicals, Watanabe Chemical, Nacalai
Tesque, Fluka, Tokyo Chemical Industry, Kanto, Pep-
tide Institute, Applied Biosystems) and used without
purification. Organic solvents were purchased as anhy-
drous grade except for the following solvents, which
were freshly distilled prior to use: tetrahydrofuran (THF)
and diethyl ether were dried by distillation from Na and
benzophenone ketyl, and dichloromethane was dried by
distillation from calcium hydride. Analytical TLC was
conducted with silica gel 60 F254 plates (Merck). Air-
sensitive reactions were carried out under a dry nitrogen
atmosphere in well-dried equipment with a tightly fitted
rubber septum. Open column chromatography was
performed using silica gel BW-300 (Fuji silysia) or
ODS Cosmosil 140C18-OPN (Nacalai Tesque). NMR
spectra were recorded on a Bruker ARX-400 (1H;
400 MHz, 13C; 100 MHz) or a Bruker AMX-600 (1H;
600 MHz, 13C; 150 MHz) instrument. NMR chemical
shifts in ꢂ (ppm) were referenced to the solvent peaks of
ꢂH 7.26 and ꢂC 77.0 for CDCl3, ꢂH 2.04 and ꢂC 29.8 for
CD3COCD3, ꢂH 3.30 and ꢂC 49.0 for CD3OD, and ꢂH
1.98 and ꢂC 1.3 for CD3CN. Optical rotations were
measured on a Jasco DIP-370 digital polarimeter. Solid-
phase peptide synthesis was performed with a model
433A peptide synthesizer (Applied Biosystems). Amino
acid sequencing was performed with a 490 Procise
protein sequencer (Applied Biosystems).
Bacterial strains, pheromone production, and bio-
logical activity. Escherichia coli ComXRO-E-2 producer
strain (ED413) and B. subtilis tester strain (BD3020)
were prepared as described previously.4,5) Pheromone
production was as performed previously.4) Biological
activity was measured by a ꢀ-galactosidase assay.2,5)
Pheromone purification. Partial purification was per-
formed as described previously.4) The supernatant
obtained after centrifugation of conditioned medium
was filtered through a membrane filter (0.22 ꢁm). The
filtrate was loaded onto a C-18 Sep-Pak cartridge
(Waters) in 20% aqueous CH3CN and 0.1% trifluoro-
acetic acid (TFA). ComX pheromone was eluted with a
step gradient of CH3CN (20, 50, and 80% in 0.1%
aqueous TFA). The 50% eluant mainly contained ComX
pheromone. The solution of ComX-RO-E-2 was neu-
tralized with aqueous ammonium, then concentrated and
dissolved in 50% aqueous CH3CN (about 1 ml). The
solution was purified by HPLC on an ODS column
(4:6 ꢀ 250 mm ID, Develosil ODS-HG-5, Nomura
Chemical) at a flow rate of 0.8 ml/min, with a gradient
of 20–60% CH3CN in 0.1% aqueous ammonium acetate
over 45 min. The most active fraction was collected and
purified by HPLC on a cyanopropyl column (4:6 ꢀ
250 mm ID, Develosil CN-UG-5, Nomura Chemical) at
a flow rate of 1.0 ml/min, with a gradient of 20–32%
CH3CN in 0.1% aqueous ammonium acetate for 24 min
to give the ComXRO-E-2 solution (about 0.1 ml; yield
about 1 ꢁg from 1 l broth).
N-Phthaloyl-1-geranyl-L-tryptophan methyl ester 2.
To a solution of N-phthaloyl-L-tryptophan methyl ester 1
(1.55 g, 4.45 mmol) in N,N-dimethylformamide (DMF)
(45 ml) at 0 ꢂC under nitrogen was added NaH (60% oil
suspension, 265 mg, 6.63 mmol). After stirring at 0 ꢂC
for 1 h, geranyl bromide (1.0 ml, 5.02 mmol) was added
to the mixture at 0 ꢂC. The reaction mixture was stirred
and warmed to room temperature for 4 h. The reaction
mixture was quenched with 5% aqueous KHSO4
(50 ml), and then the aqueous layer was extracted with
Et2O (4 ꢀ 50 ml). The organic extract was washed with
sat. brine, dried over Na2SO4, and evaporated. The
residue was purified by silica gel column chromatog-
raphy (hexane/EtOAc 10=1!5=1) to give N-phthaloyl-
1-geranyl-L-tryptophan methyl ester 2 (1.32 g, 2.72
27
mmol, 61%) as a yellow oil. ½ꢃꢁ
þ0:20ꢂ (c 1.0,
D
HRMS analysis. The solution of ComXRO-E-2 (10 ꢁl)
was added to a polypropylene glycol solution (20 ꢁM,
CHCl3). NMR (400 MHz) ꢂH (CDCl3) 1.60 (3H, s), 1.62
(3H, s), 1.68 (3H, s), 1.96–2.08 (4H, m), 3.75 (2H, d,

2376
M. OKADA et al.
J ¼ 7:6 Hz), 3.80 (3H, s), 4.55 (2H, d, J ¼ 6:8 Hz), 5.05
(1H, t, J ¼ 6:8 Hz), 5.20 (1H, t, J ¼ 6:8 Hz), 5.28 (1H, t,
J ¼ 7:2 Hz), 6.91 (1H, s), 7.05 (1H, t, J ¼ 7:2 Hz), 7.12
(1H, t, J ¼ 7:2 Hz), 7.22 (1H, d, J ¼ 7:2 Hz), 7.61 (1H,
d, J ¼ 7:2 Hz), 7.66 (2H, m), 7.76 (2H, m). NMR
(100 MHz) ꢂC (CDCl3) 16.2, 17.6, 24.7, 25.6, 26.2, 39.3,
43.8, 52.6, 52.8, 109.4, 118.5, 118.9, 119.5, 119.7,
121.4, 123.2, 123.2, 123.3, 123.6, 123.7, 125.8, 127.8,
131.7, 133.9, 136.2, 139.6, 167.4, 169.6. Anal. Found:
C, 74.38; H, 6.74; N, 5.88. Calcd. for C30H32N2O4: C,
74.36; H, 6.66; N, 5.78.
1-Triisopropylsilyl-4-geranylgramine 6. To a solution
of 1-(triisopropylsilyl)gramine 5 (4.98 g, 15.1 mmol) in
Et2O (78 ml) at ꢃ78 ꢂC under nitrogen was added t-
BuLi (1.7 M in heptane, 10 ml, 17 mmol). After stirring
at 0 ꢂC for 1 h, geranyl bromide (2.4 ml, 12.1 mmol) was
dropped at 0 ꢂC. The reaction mixture was stirred at 0 ꢂC
overnight. The reaction mixture was quenched with H2O
(50 ml), and then the aqueous layer was extracted with
Et2O (4 ꢀ 50 ml). The organic extract was washed with
sat. brine, dried over Na2SO4, and evaporated. The
residue was purified by silica gel column chromatog-
raphy (hexane/EtOAc 50=1!20=1) to give 1-(triiso-
propylsilyl)-4-geranylgramine 6 (3.07 g, 6.59 mmol,
44%) as a pale yellow oil. NMR (400 MHz) ꢂH (CDCl3)
1.15 (18H, d, J ¼ 7:6 Hz), 1.61 (3H, s), 1.69 (3H, s),
1.76 (3H, s), 2.05–2.17 (4H, m), 2.25 (6H, s), 3.59 (2H,
s), 3.94 (2H, d, J ¼ 6:8 Hz), 5.14 (1H, t, J ¼ 6:0 Hz),
5.46 (1H, t, J ¼ 6:4 Hz), 6.91 (1H, d, J ¼ 7:8 Hz), 7.04
(1H, t, J ¼ 7:8 Hz), 7.09 (1H, s), 7.32 (1H, d, J ¼
7:8 Hz). NMR (100 MHz) ꢂC (CDCl3) 12.8, 16.3, 17.7,
18.2, 25.7, 26.7, 31.6, 39.7, 44.9, 56.6, 111.7, 115.9,
119.8, 121.4, 124.4, 124.5, 129.0, 131.2, 131.4, 134.8,
135.1, 142.2. Anal. Found: C, 76.98; H, 11.24; N, 5.84.
Calcd. for C30H50N2Si: C, 77.19; H, 10.80; N, 6.00.
Compound 8. To a solution of 1-Boc-4-geranyl-
gramine 7 (685 mg, 1.67 mmol) in toluene (15 ml) at
room temperature under nitrogen was dropped ethyl
chloroformate (0.18 ml, 1.89 mmol). After the reaction
mixture had been stirred for 15 min, the solvent was
evaporated. The residue was dissolved in THF (15 ml)
under nitrogen and used without any further purification.
Fmoc-Trp(1-Ger) 3. To a solution of N-phthaloyl-1-
geranyl-L-tryptophan methyl ester 2 (820 mg, 1.69
mmol) in 1:1 CH2Cl2/CH3OH (20 ml) at 40 ꢂC under
nitrogen was added hydrazine hydrate (80%, 0.11 ml,
116 mg, 1.85 mmol). The reaction mixture was stirred at
40 ꢂC for 24 h. The reaction mixture was poured into
water (30 ml), and extracted with Et2O (4 ꢀ 30 ml). The
organic extract was washed with sat. brine, dried over
Na2SO4, and evaporated. The residue was purified by
silica gel column chromatography (hexane/acetone
8=1!5=1) to give 1-geranyl-L-tryptophan methyl ester
(494 mg, 1.39 mmol, 83%) as a colorless oil. To a
solution of 1-geranyl-L-tryptophan methyl ester (375
mg, 1.06 mmol) in THF (5 ml) was dropped 1 M aqueous
LiOH (5 ml) at room temperature. After the reaction
mixture was stirred for 1 h, it was neutralized with 5%
aqueous KHSO4, and H2O (10 ml) was added. After
THF was removed, the precipitate was filtered and
washed with H2O (2 ꢀ 5 ml). The precipitate was
dissolved in dioxane (20 ml) and 1 M aqueous Na2CO3
(10 ml). To the solution was added Fmoc-OSu (460 mg,
1.36 mmol) at room temperature. After the reaction
mixture had been stirred for 3 h, the reaction was
quenched with 5% aqueous KHSO4 (50 ml). The
reaction mixture was extracted with EtOAc (4 ꢀ
30 ml). The organic layer was washed with sat. brine,
dried over Na2SO4, and evaporated. The residue was
purified by silica gel column chromatography (CHCl3/
To a solution of Schollkopf chiral auxiliary 4 (414 mg,
¨
1.95 mmol) in THF (20 ml) was dropped n-BuLi (1.7 M
in hexane, 1.3 ml, 2.2 mmol) at ꢃ78 ꢂC under nitrogen.
After stirring at ꢃ78 ꢂC for 60 min, N,N,N0,N0-tetra-
methylethylenediamine (TMEDA) (0.33 ml, 2.19 mmol)
was added at ꢃ78 ꢂC. After stirring for 30 min at
ꢃ78 ꢂC, the solution prepared above was dropped in at
ꢃ78 ꢂC. The reaction mixture was stirred and warmed to
0 ꢂC for 18 h. It was poured into H2O. (20 ml), and then
the aqueous layer was extracted with Et2O (4 ꢀ 20 ml).
The organic extract was washed with sat. brine, dried
over Na2SO4, and evaporated. The residue was purified
by silica gel column chromatography (hexane/EtOAc
CH3OH 100=1!80=1) to give Fmoc-Trp(1-Ger) 3
22
(514 mg, 0.913 mmol, 87%) as a white powder. ½ꢃꢁ
D
þ1:6ꢂ (c 1.0, CH3OH). NMR (600 MHz) ꢂH (CD3OD)
1.47 (3H, s), 1.55 (3H, s), 1.70 (3H, s), 1.88–1.91
(2H, m), 1.97–2.01 (2H, m), 3.13 (1H, dd, J ¼ 8:4,
14.4 Hz), 3.28 (1H, dd, J ¼ 4:8, 14.4 Hz), 4.07 (1H, t,
J ¼ 7:2 Hz), 4.15 (1H, dd, J ¼ 7:2, 10.8 Hz), 4.24 (1H,
dd, J ¼ 7:2, 10.8 Hz), 4.51 (1H, q, J ¼ 4:8, 8.4 Hz), 4.56
(1H, d, J ¼ 6:6 Hz), 4.93 (1H, t, J ¼ 6:6 Hz), 5.19 (1H,
t, J ¼ 6:6 Hz), 7.00 (1H, s), 7.01 (1H, t, J ¼ 7:2 Hz),
7.10 (1H, t, J ¼ 7:2 Hz), 7.17 (1H, d, J ¼ 7:2 Hz), 7.22
(2H, m) 7.31 (2H, m), 7.51 (2H, d, J ¼ 7:2 Hz); 7.59
(1H, d, J ¼ 7:2 Hz) 7.71 (2H, d, J ¼ 7:2 Hz). NMR
(150 MHz) ꢂC (CD3OD) 16.4, 17.7, 25.8, 27.3, 28.7,
40.4, 44.8, 68.0, 110.7, 110.7, 119.6, 119.9, 120.9,
121.5, 122.4, 124.9, 126.2, 126.3, 127.5, 128.1, 128.7,
129.7, 132.5, 137.8, 140.3, 142.5, 145.2, 145.2, 158.3,
175.6. Anal. Found: C, 76.86; H, 6.98; N, 4.83. Calcd.
for C36H38N2O4: C, 76.84; H, 6.81; N, 4.98.
50/1) to give compound 8 (756 mg, 1.31 mmol, 78%) as
27
a pale yellow oil. ½ꢃꢁ
ꢃ11ꢂ (c 1.0, CHCl3). NMR
D
(400 MHz) ꢂH (CDCl3) 0.72 (3H, d, J ¼ 7:2 Hz), 1.03
(3H, d, J ¼ 7:2 Hz), 1.21 (3H, t, J ¼ 7:2 Hz), 1.30 (3H,
t, J ¼ 7:2 Hz), 1.60 (3H, s), 1.65 (9H, s), 1.69 (3H, s),
1.76 (3H, s), 2.07–2.14 (4H, m), 2.25 (1H, m), 3.12 (1H,
dd, J ¼ 7:2, 14.8 Hz), 3.55 (1H, dd, J ¼ 4:6, 14.8 Hz),
3.77 (1H, t, J ¼ 3:4 Hz), 3.84–4.28 (7H, m), 5.13 (1H, t,
J ¼ 7:0 Hz), 5.34 (1H, t, J ¼ 6:5 Hz), 7.01 (1H, d,
J ¼ 7:8 Hz), 7.19 (1H, t, J ¼ 7:8 Hz), 7.45 (1H, s), 8.08
(1H, d, J ¼ 7:8 Hz). NMR (100 MHz) ꢂC (CDCl3) 14.3,
16.3, 16.7, 17.7, 19.0, 25.7, 26.7, 28.2, 31.3, 31.8, 31.9,
39.7, 56.5, 60.5, 60.6, 60.7, 82.9, 113.0, 117.6, 123.1,
123.6, 123.9, 124.3, 124.6, 129.0, 131.4, 135.0, 135.8,

Synthetic Studies on Putative ComX Pheromone
2377
135.9, 149.6, 162.8, 163.2. Anal. Found: C, 72.76; H,
9.13; N, 7.26. Calcd. for C35H51N3O4: C, 72.75; H, 8.90;
N, 7.27.
Compound 12. To a solution of 1-Boc-gramine 9
(9.35 g, 34.1 mmol) in toluene (200 ml) at room temper-
ature under nitrogen was dropped ethyl chloroformate
(3.5 ml, 36.9 mmol). After the reaction mixture had been
stirred for 15 min, the solvent was evaporated. The
residue was recrystallized from hexane/benzene to give
1-Boc-3-(chloromethyl)indole (7.80 g, 29.4 mmol, 86%)
1.74 (3H, s), 1.95–2.04 (4H, m), 2.17–2.23 (1H, m), 2.94
(1H, dd, J ¼ 7:8, 14.4 Hz), 3.30 (1H, dd, J ¼ 4:2,
14.4 Hz), 3.61 (1H, t, J ¼ 6:6 Hz), 3.78 (1H, d,
J ¼ 6:0 Hz), 4.02–4.10 (3H, m), 4.19–4.25 (2H, m),
5.06 (1H, t, J ¼ 6:0 Hz), 5.20 (1H, t, J ¼ 6:0 Hz), 7.16
(1H, t, J ¼ 7:2 Hz), 7.20 (1H, t, J ¼ 7:2 Hz), 7.54 (1H,
d, J ¼ 7:2 Hz), 8.06 (1H, d, J ¼ 7:2 Hz). NMR
(150 MHz) ꢂC (CDCl3) 14.4, 16.5, 16.5, 17.6, 19.1,
25.6, 26.0, 26.6, 28.1, 29.4, 31.4, 39.6, 56.9, 60.4, 60.4,
60.7, 83.3, 115.0, 115.4, 119.0, 121.8, 122.1, 123.2,
124.2, 130.6, 131.4, 135.2, 136.0, 137.9, 150.5, 162.8,
163.2. Anal. Found: C, 72.75; H, 8.95; N, 7.30. Calcd.
for C35H51N3O4: C, 72.75; H, 8.90; N, 7.27.
Intermediate 18. To a solution of diethylbutanoyl
derivative 17 (419 mg, 0.896 mmol) and TMEDA
(0.27 ml, 1.79 mmol) in Et2O (10 ml) was dropped s-
BuLi (1.3 M in Et2O, 0.49 ml, 0.637 mmol) at ꢃ78 ꢂC
under nitrogen. After stirring at ꢃ78 ꢂC for 1 h, geranyl
bromide (0.09 ml, 98.1 mg) was added to the mixture.
The reaction mixture was stirred at ꢃ78 ꢂC for 5 h. It
was quenched with H2O (5 ml), and the aqueous layer
was extracted with Et2O (3 ꢀ 20 ml). The organic
extract was washed with sat. brine, dried over Na2SO4,
and evaporated. The residue was purified by silica gel
column chromatography (hexane/EtOAc 100=1!50=1)
and ODS column chromatography (CH3CN/i-PrOH
as a white crystalline solid. To a solution of Schollkopf
¨
chiral auxiliary 4 (1.82 g, 8.57 mmol) in THF (80 ml)
was dropped n-BuLi (1.7 M in hexane, 5.5 ml, 9.35
mmol) at ꢃ78 ꢂC under nitrogen. After stirring at
ꢃ78 ꢂC for 60 min, TMEDA (1.7 ml, 11.3 mmol) was
added to the mixture at ꢃ78 ꢂC. After stirring for 60 min
at ꢃ78 ꢂC, 1-Boc-3-(chloromethyl)gramine (2.95 g,
11.1 mmol) in THF (20 ml) was dropped into the
mixture at ꢃ78 ꢂC. The reaction mixture was stirred
and warmed to 0 ꢂC for 18 h. It was poured into H2O
(50 ml), and the aqueous layer was extracted with Et2O
(4 ꢀ 100 ml). The organic extract was washed with sat.
brine, dried over Na2SO4, and evaporated. The residue
was purified by silica gel column chromatography
(hexane/EtOAc 100=1!20=1) to give compound 12
27
(3.35 g, 7.56 mmol, 88%) as a white powder. ½ꢃꢁ
D
þ52ꢂ (c 1.0, CH3OH). NMR (600 MHz) ꢂH (CDCl3)
0.67 (3H, d, J ¼ 6:8 Hz), 0.95 (3H, d, J ¼ 6:8 Hz), 1.24
(3H, t, J ¼ 7:1 Hz), 1.33 (3H, t, J ¼ 7:1 Hz), 1.65 (9H,
s), 2.18 (1H, m), 3.17 (1H, dd, J ¼ 5:8, 14.3 Hz), 3.21
(dd, 1H, J ¼ 3:4, 14.3 Hz), 3.56 (t, 1H, J ¼ 3:3 Hz),
3.78 (d, 1H, J ¼ 6:0 Hz), 4.02 (1H, m), 4.09–4.18 (3H,
m), 4.29 (1H, m), 7.18 (1H, t, J ¼ 7:8 Hz), 7.25 (1H, t,
J ¼ 7:8 Hz), 7.37 (1H, s), 7.59 (1H, d, J ¼ 7:8 Hz), 8.09
(1H, br). NMR (150 MHz) ꢂC (CDCl3) 14.4, 16.6, 19.0,
28.2, 29.2, 31.7, 56.0, 60.4, 60.5, 60.6, 83.1, 114.9,
116.7, 119.6, 122.0, 123.9, 124.2, 131.4, 135.1, 149.7,
162.3, 163.5. Anal. Found: C, 59.61; H, 6.85; N, 8.77.
Calcd. for C25H35N3O4: C, 59.61; H, 6.88; N, 8.69.
Intermediate 15. To a solution of compound 12
(910 mg, 2.06 mmol) in Et2O (15 ml) was dropped
lithium diisopropylamide (LDA) (0.5 M in Et2O, 10 ml,
5.00 mmol) at ꢃ78 ꢂC under nitrogen. After stirring at
ꢃ78 ꢂC for 90 min, CuCN (185 mg, 2.06 mmol) was
added to the mixture at ꢃ78 ꢂC. After stirring for 90 min
at ꢃ78 ꢂC, geranyl bromide (547 mg, 2.52 mmol) was
added to the mixture at ꢃ78 ꢂC. The reaction mixture
was stirred overnight. The reaction mixture was
quenched with H2O (10 ml), and the aqueous layer
was extracted with Et2O (4 ꢀ 50 ml). The organic
extract was washed with sat. brine, dried over Na2SO4,
and evaporated. The residue was purified by silica gel
column chromatography (hexane/EtOAc 100=1!50=1)
100=0!9=1) to give intermediate 18 (171 mg,
24
0.366 mmol, 41%) as a yellow oil. ½ꢃꢁ
þ43ꢂ (c
D
0.10, CHCl3). NMR (600 MHz) ꢂH (CDCl3) 0.67 (3H, d,
J ¼ 7:2 Hz), 0.88 (9H, t, J ¼ 7:2 Hz), 0.94 (3H, d,
J ¼ 7:2 Hz), 1.23 (3H, t, J ¼ 7:2 Hz), 1.30 (3H, t,
J ¼ 7:2 Hz), 1.61 (3H, s), 1.69 (3H, s), 1.70 (3H, s), 1.74
(6H, q, J ¼ 7:2 Hz), 2.05–2.22 (5H, m), 3.17–3.25 (2H,
m), 3.33–3.42 (2H, m), 3.51 (1H, t, J ¼ 3:6 Hz), 3.93–
4.04 (1H, m), 4.05–4.10 (1H, m), 4.11–4.20 (2H, m),
4.34 (1H, m), 5.14 (1H, t, J ¼ 6:6 Hz), 5.31 (1H, t,
J ¼ 6:6 Hz), 7.16 (1H, t, J ¼ 7:2 Hz), 7.19 (1H, t,
J ¼ 7:2 Hz), 7.37 (1H, s), 7.44 (1H, d, J ¼ 7:2 Hz).
NMR (150 MHz) ꢂC (CDCl3) 8.7, 14.3, 14.4, 15.2, 16.1,
16.6, 17.7, 19.0, 25.7, 26.6, 27.9, 29.1, 31.6, 32.9, 39.8,
52.4, 55.8, 60.5, 60.6, 116.7, 116.8, 122.5, 122.8, 123.3,
124.4, 125.3, 129.1, 131.4, 131.8, 135.6, 136.8, 162.2,
163.7, 176.0. Anal. Found: C, 75.58; H, 9.45; N. 6.77.
Calcd. for C38H57N3O3: C, 75.58; H, 9.51; N, 6.96.
Pinacol borate 19. To a suspension of compound
.
13 (1.97 g, 3.78 mmol), PdCl2(dppf) CH2Cl2 (155 mg,
0.190 mmol) and triethylamine (1.6 ml, 1.16 g, 11.5
mmol) in toluene (38 ml) at 90 ꢂC under nitrogen was
dropped pinacolborane (0.84 ml, 741 mg, 5.79 mmol).
The reaction mixture was stirred for 9 h at 90 ꢂC. The
reaction mixture was filtered through a glass filter and
the residue was washed with EtOAc (3 ꢀ 20 ml). The
combined filtrate was washed with H2O and sat. brine,
dried over Na2SO4, and evaporated. The residue was
purified by silica gel column chromatography (hexane/
to give intermediate 15 (798 mg, 1.38 mmol, 67%) as
22
a yellow oil. ½ꢃꢁ
þ29ꢂ (c 1.0, CHCl3). NMR
D
(600 MHz) ꢂH (CDCl3) 0.64 (3H, d, J ¼ 6:6 Hz), 0.96
(3H, d, J ¼ 6:6 Hz), 1.20 (3H, t, J ¼ 7:2 Hz), 1.33 (3H,
t, J ¼ 7:2 Hz), 1.56 (3H, s), 1.64 (3H, s), 1.65 (9H, s),
EtOAc 20=1!12=1) to give pinacol borate 19 (1.91 g,
25
3.36 mmol, 89%) as a white powder. ½ꢃꢁ
þ23ꢂ (c 1.0,
D
CHCl3). NMR (400 MHz) ꢂH (CDCl3) 0.66 (d, 3H,

2378
M. OKADA et al.
J ¼ 6:8 Hz), 0.93 (d, 3H, J ¼ 6:8 Hz), 1.22 (t, 3H,
J ¼ 7:1 Hz), 1.30–1.35 (m, 15H), 1.64 (s, 9H), 2.16 (m,
1H), 3.15–3.29 (m, 2H), 3.50 (t, 1H, J ¼ 3:5 Hz), 4.02
(m, 1H), 4.07–4.18 (m, 3H), 4.31 (m, 1H), 7.33 (s, 1H),
7.71 (dd, 1H, J ¼ 1:1, 8.3 Hz), 8.06–8.09 (m, 2H). NMR
(100 MHz) ꢂC (CDCl3) 14.4, 14.4, 16.6, 19.0, 24.8, 25.0,
28.1, 29.1, 31.7, 56.0, 60.4, 60.5, 60.6, 83.2, 83.5, 114.1,
117.0, 124.2, 127.1, 130.4, 131.0, 137.3, 149.6, 162.2,
163.5. Anal. Found: C, 65.54; H, 8.23; N, 7.51. Calcd.
16.6, 17.7, 19.0, 25.7, 26.7, 28.2, 29.3, 31.7, 34.7, 39.8,
56.0, 60.4, 60.5, 60.6, 82.9, 114.6, 116.7, 119.3, 122.7,
123.7, 123.7, 124.3, 129.5, 131.4, 135.6, 138.0, 149.8,
162.3, 163.4. Anal. Found: C, 72.61; H, 8.99; N. 7.15.
Calcd. for C35H51N3O4: C, 72.75; H, 8.90; N, 7.27.
Compound 23. According to the procedure described
for compound 16, intermediate 15 (47.0 mg, 0.0813
mmol) in THF (0.18 ml), and NaOMe (28% in CH3OH,
0.08 ml, 22.4 mg, 0.415 mmol) gave compound 23
22
for C36H38N2O4: C, 65.61; H, 8.17; N, 7.40.
26
(33.5 mg, 0.0701 mmol, 86%) as a colorless oil. ½ꢃꢁ
D
Pinacol borate 20. ½ꢃꢁ
þ24ꢂ (c 1.0, CHCl3). NMR
þ35ꢂ (c 1.0, CHCl3). NMR (600 MHz) ꢂH (CDCl3) 0.61
(3H, d, J ¼ 6:6 Hz), 0.91 (3H, d, J ¼ 6:6 Hz), 1.20 (3H,
t, J ¼ 7:2 Hz), 1.34 (3H, t, J ¼ 7:2 Hz), 1.63 (3H, s),
1.74 (6H, s), 2.09–2.18 (5H, m), 3.15 (1H, dd, J ¼ 6:0,
14.4 Hz), 3.22 (1H, t, J ¼ 3:6 Hz), 3.29 (1H, dd,
J ¼ 4:2, 14.4 Hz), 3.48 (1H, d, J ¼ 6:6 Hz), 3.99–4.05
(1H, m), 4.08–4.12 (2H, m), 4.17–4.24 (1H, m), 4.28–
4.30 (1H, m), 5.12 (1H, t, J ¼ 6:6 Hz), 5.31 (1H, t,
J ¼ 6:6 Hz), 7.01 (1H, t, J ¼ 7:2 Hz), 7.06 (1H, t,
J ¼ 7:2 Hz), 7.21 (1H, d, J ¼ 7:2 Hz), 7.56 (1H, d,
J ¼ 7:2 Hz) 7.74 (1H, s). NMR (150 MHz) ꢂC (CDCl3)
14.3, 14.4, 16.1, 16.4, 17.7, 19.1, 25.0, 25.8, 26.4, 29.0,
30.9, 39.6, 57.3, 60.1, 60.3, 60.5, 107.2, 109.9, 118.7,
119.0, 120.6, 120.7, 124.1, 129.7, 131.8, 135.0, 135.5,
138.1, 162.9, 163.4. Anal. Found: C, 75.44; H, 8.95; N,
D
(600 MHz) ꢂH (CDCl3) 0.66 (3H, d, J ¼ 7:2 Hz), 0.95
(3H, d, J ¼ 7:2 Hz), 1.24 (3H, t, J ¼ 7:2 Hz), 1.32 (3H,
t, J ¼ 7:2 Hz), 1.64 (9H, s), 2.17 (1H, m), 3.16 (2H, d,
J ¼ 4:8 Hz), 3.56 (1H, t, J ¼ 3:6 Hz), 3.98–4.18 (4H,
m), 4.27 (1H, m), 7.30 (2H, m), 7.44 (1H, d,
J ¼ 8:4 Hz), 8.30 (1H, br). NMR (100 MHz) ꢂC (CDCl3)
14.4, 16.6, 19.0, 28.1, 29.1, 31.9, 55.9, 60.4, 60.5, 60.7,
83.7, 116.6, 117.7, 118.1, 120.8, 125.0, 125.2, 130.3,
135.8, 149.3, 162.0, 163.6. Anal. Found: C, 57.76; H,
6.62; N, 7.96. Calcd. for C25H34BrN3O4: C, 57.69; H,
6.58; N, 8.07.
Intermediate 21. To a suspension of pinacol borate 19
.
(155 mg, 0.273 mmol) and PdCl2(dppf) CH2Cl2 (23.0
mg, 0.0282 mmol) in THF (6.8 ml) and 2 M aqueous
KOH (1.37 ml) at 60 ꢂC under nitrogen was dropped
geranyl bromide (0.16 ml, 174 mg, 0.803 mmol). The
reaction mixture was stirred for 48 h at 60 ꢂC. It was
filtered through a glass filter and the residue was washed
with Et2O (3 ꢀ 20 ml). The combined filtrate was
washed with H2O and sat. brine, dried over Na2SO4,
and evaporated. The residue was purified by silica gel
column chromatography (hexane/EtOAc 100=1!30=1)
8.55. Calcd. for C30H43N3O2: C, 75.43; H, 9.07; N, 8.80.
26
Compound 24. ½ꢃꢁ
þ5:2ꢂ (c 1.0, CHCl3). NMR
D
(400 MHz) ꢂH (CDCl3) 0.66 (3H, d, J ¼ 6:8 Hz), 0.97
(3H, d, J ¼ 6:9 Hz), 1.16 (3H, t, J ¼ 7:1 Hz), 1.26 (3H,
t, J ¼ 7:1 Hz), 1.58 (3H, s), 1.67 (3H, s), 1.74 (3H, s),
2.05–2.12 (4H, m), 2.19 (1H, m), 3.18 (1H, dd, J ¼ 6:6,
14.8 Hz), 3.55–3.61 (2H, m), 3.84 (1H, m), 3.91–4.08
(4H, m), 4.20 (1H, m), 4.27 (1H, m), 5.12 (1H, m), 5.38
(1H, m), 6.85 (1H, d, J ¼ 7:6 Hz), 6.97 (1H, d,
J ¼ 2:3 Hz), 7.03 (1H, t, J ¼ 7:6), 7.15 (1H, d,
J ¼ 7:6 Hz), 7.99 (1H, br). NMR (100 MHz) ꢂC (CDCl3)
14.3, 16.3, 16.6, 17.7, 19.1, 25.7, 26.7, 31.4, 31.6, 32.1,
39.8, 57.3, 60.5, 60.6, 109.0, 112.8, 119.6, 121.6, 123.0,
123.9, 124.4, 126.0, 131.3, 134.8, 135.4, 136.5, 163.0,
163.3. Anal. Found: C, 75.44; H, 9.31; N, 8.77. Calcd.
for C30H43N3O2: C, 75.43; H, 9.07; N, 8.80.
to give intermediate 21 (93.0 mg, 0.161 mmol, 59%) as a
25
pale yellow oil. ½ꢃꢁ
þ15ꢂ (c 1.0, CHCl3). NMR
D
(400 MHz) ꢂH (CDCl3) 0.67 (3H, d, J ¼ 6:8 Hz), 0.95
(3H, d, J ¼ 6:8 Hz), 1.23 (3H, t, J ¼ 7:1 Hz), 1.32 (3H,
t, J ¼ 7:1 Hz), 1.61 (3H, s), 1.64 (9H, s), 1.69 (3H, s),
1.74 (3H, s), 2.03–2.21 (5H, m), 3.12–3.20 (2H, m), 3.43
(2H, d, J ¼ 7:1 Hz), 3.56 (1H, t, J ¼ 3:4 Hz), 3.98–4.20
(4H, m), 4.29 (1H, m), 5.11 (1H, m), 5.37 (1H, m), 7.09
(1H, d, J ¼ 8:4 Hz), 7.32 (1H, s), 7.38 (1H, s), 7.93 (1H,
d, J ¼ 8:4 Hz). NMR (100 MHz) ꢂC (CDCl3) 14.4, 16.2,
16.7, 17.7, 19.0, 25.7, 26.7, 28.2, 29.2, 31.7, 34.3, 39.8,
56.1, 60.5, 60.6, 82.9, 114.7, 116.6, 118.9, 123.9, 124.3,
124.6, 131.4, 131.7, 135.5, 149.7, 162.3, 163.5. Anal.
Found: C, 72.65; H, 9.13; N. 7.23. Calcd. for
Ethyl ester 28. To a solution of compound 23
(32.1 mg, 0.0672 mmol) in THF (0.65 ml) at 0 ꢂC was
dropped 1.5 M aqueous HCl (0.44 ml). The reaction
mixture was stirred at 0 ꢂC for 2 h. It was quenched with
sat. aqueous NaHCO3 (5 ml), and then the aqueous layer
was extracted with EtOAc (4 ꢀ 10 ml). The organic
extract was washed with sat. brine, dried over Na2SO4,
and evaporated. The residue was purified by silica gel
column chromatography (CHCl3/CH3OH 50/1) to give
C35H51N3O4: C, 72.75; H, 8.90; N, 7.27.
27
Intermediate 22. ½ꢃꢁ
þ16ꢂ (c 1.0, CHCl3). NMR
D
(400 MHz) ꢂH (CDCl3) 0.66 (3H, d, J ¼ 6:8 Hz), 0.95
(3H, d, J ¼ 6:9 Hz), 1.23 (3H, t, J ¼ 7:0 Hz), 1.32 (3H,
t, J ¼ 7:1 Hz), 1.60 (3H, s), 1.64 (9H, s), 1.68 (3H, s),
1.74 (3H, s), 2.02–2.14 (4H, m), 2.18 (1H, m), 3.12–3.20
(2H, m), 3.46 (2H, d, J ¼ 7:2 Hz), 3.57 (1H, t,
J ¼ 3:4 Hz), 3.98–4.20 (4H, m), 4.28 (1H, m), 5.10
(1H, m), 5.39 (1H, t, J ¼ 7:2 Hz), 7.03 (1H, d,
J ¼ 8:0 Hz), 7.29 (1H, s), 7.47 (1H, d, J ¼ 8:0 Hz),
7.93 (1H, br). NMR (100 MHz) ꢂC (CDCl3) 14.4, 16.2,
ethyl ester 28 (20.0 mg, 0.0543 mmol, 81%) as a yellow
25
oil. ½ꢃꢁ
þ2:2ꢂ (c 0.50, CHCl3). NMR (400 MHz) ꢂH
D
(CDCl3) 1.23 (1H, t, J ¼ 7:2 Hz), 1.63 (3H, s), 1.73 (3H,
s), 1.75 (3H, s), 2.08–2.20 (4H, m), 2.97 (1H, dd,
J ¼ 8:3, 14.3 Hz), 3.28 (1H, dd, J ¼ 5:1, 14.3 Hz), 3.51
(2H, d, J ¼ 7:2 Hz), 3.79 (1H, m), 4.15 (2H, m), 5.11
(1H, m), 5.32 (1H, t, J ¼ 7:3 Hz), 6.98 (1H, s), 7.08 (1H,
t, J ¼ 7:2 Hz), 7.12 (1H, t, J ¼ 7:2 Hz), 7.27 (1H, d,

Synthetic Studies on Putative ComX Pheromone
2379
J ¼ 7:2 Hz), 7.54 (1H, d, J ¼ 7:2 Hz), 7.84 (1H, br).
NMR (100 MHz) ꢂC (CDCl3) 14.1, 16.2, 17.8, 25.0,
25.8, 26.4, 30.2, 39.6, 55.4, 60.9, 106.5, 110.4, 118.2,
119.4, 120.0, 121.2, 124.0, 128.9, 131.9, 135.2, 135.7,
138.6, 175.4. Anal. Found: C, 74.96; H, 8.75; N, 7.75.
zol-1-yl-N,N,N0,N0-tetramethyluronium hexafluorophos-
phate (HBTU) reagent, except that geranyl substituted
for Fmoc-tryptophan, which was conducted manually as
follows: To a suspension of Glu(Pp)-Gln-Clt resin
(0.10 mmol), which was synthesized with peptide syn-
thesizer, in N-methyl-2-pyrrolidinone (NMP) (1 ml)
under nitrogen was added Fmoc-Trp(n-Ger) (85 mg,
0.15 mmol) in NMP (2 ml), diisopropylethylamine (DI-
PEA) (0.40 mmol) in DMF (2 ml), and O-(7-azabenzo-
Calcd. for C23H32N2O2: C, 74.96; H, 8.75; N, 7.60.
24
Ethyl ester 29. ½ꢃꢁ
þ17ꢂ (c 1.0, CHCl3). NMR
D
(400 MHz) ꢂH (CDCl3) 1.23 (3H, t, J ¼ 7:1 Hz), 1.59
(3H, s), 1.68 (3H, s), 1.74 (3H, s), 2.04–2.13 (4H, m),
2.97 (1H, dd, J ¼ 8:9, 14.6 Hz), 3.50 (1H, dd, J ¼ 4:5,
14.6 Hz), 3.71–3.82 (3H, m), 4.18 (2H, q, J ¼ 7:1 Hz),
5.12 (1H, t, J ¼ 6:2 Hz), 5.38 (1H, t, J ¼ 6:2 Hz), 6.91
(1H, d, J ¼ 7:6 Hz), 6.99 (1H, s), 7.08 (1H, t,
J ¼ 7:6 Hz), 7.17 (1H, d, J ¼ 7:6 Hz), 8.47 (1H, br).
NMR (100 MHz) ꢂC (CDCl3) 14.1, 14.2, 16.3, 17.6,
25.6, 26.6, 31.9, 33.0, 39.6, 55.7, 60.8, 109.3, 111.8,
119.8, 122.1, 123.4, 123.5, 124.3, 125.1, 131.3, 134.5,
135.9, 137.1, 175.2. Anal. Found: C, 74.97; H, 8.89; N,
7.56. Calcd. for C23H31N2O2: C, 74.96; H, 8.89; N, 7.60.
Fmoc-Trp(2-Ger) 33. To a solution of ethyl ester 28
(267 mg, 0.725 mmol) in THF (5 ml) was dropped 1 M
aqueous LiOH (5 ml) at room temperature. After the
reaction mixture had been stirred for 1 h, it was
neutralized with 5% aqueous KHSO4, and H2O (10 ml)
was added. The reaction mixture was extracted with
EtOAc (5 ꢀ 20 ml). The organic solvent was removed
by evaporation. The residue was dissolved in dioxane
(10 ml) and 1 M aqueous Na2CO3 (10 ml). To the
solution was added Fmoc-OSu (337 mg, 1.00 mmol) at
room temperature. After the reaction mixture had been
stirred for 3 h, the reaction was quenched with 5%
aqueous KHSO4 (10 ml). The reaction mixture was
extracted with EtOAc (4 ꢀ 30 ml). The organic layer
was washed with sat. brine, dried over Na2SO4, and
evaporated. The residue was purified by silica gel
column chromatography (CHCl3/CH3OH 100=1!
triazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluoro-
phosphate (HATU) (0.15 mmol) in DMF (2 ml) at
room temperature under nitrogen. The reaction mixture
was shaken overnight. It was filtered with membrane
filter (PTFE, 0.45 ꢁm), washed with CH3CN (2 ꢀ 5 ml)
and dried in vacuo to give Fmoc-Trp(n-Ger)-Glu(Pp)-
Gln-Clt resin. The resin was treated with peptide
synthesizer again to give resin attached to the full
length amino acid residues.
Cleavage, deprotection, and purification.
1-Geranyl 38. To a resin (67 ꢁmol), 4% TFA in
CH2Cl2 (0.44 ml) was dropped. The reaction mixture
was stirred at room temperature for 1 h. It was filtered
with plug of cottons, and washed with 50% aqueous
CH3CN (4 ꢀ 10 ml). The filtrate was neutralized with
1% aqueous ammonium, concentrated in vacuo to
remove organic solvent, and the aqueous residue was
freeze-dried. The residue was dissolved with 50%
aqueous CH3CN and filtered with membrane filter
(PTFE, 0.45 ꢁm). The filtrate was purified by HPLC
on an ODS column (10 ꢀ 250 mm ID, Develosil ODS-
HG-5, Nomura Chemical) at a flow rate of 3.0 ml/min,
with a gradient of 20–60% CH3CN in 0.1% aqueous
ammonium acetate over 45 min. After freeze-drying, the
putative peptide 38 (2.4 mg, 2.6 mmol, 4.0%) was
obtained as a white powder. NMR (600 MHz) ꢂH
(CD3CN, D2O) 0.55 (3H, d, J ¼ 6:6 Hz), 0.70 (3H, t,
J ¼ 7:2 Hz), 0.91 (1H, m), 1.15 (1H, m), 1.50 (3H, s),
1.56 (3H, s), 1.59 (1H, m), 1.79 (3H, s), 1.79–1.85 (3H,
m), 1.94–2.20 (9H, m), 2.71 (1H, dd, J ¼ 8:4, 14.4 Hz),
2.89 (1H, dd, J ¼ 4:8, 14.4 Hz), 3.08 (1H, dd, J ¼ 8:4,
14.4 Hz), 3.19 (2H, s), 3.27 (1H, dd, J ¼ 4:8, 14.4 Hz),
3.98 (1H, d, J ¼ 7:2 Hz), 4.08 (1H, dd, J ¼ 4:8, 8.4 Hz),
4.22 (1H, dd, J ¼ 4:8, 9.0 Hz), 4.44 (1H, dd, J ¼ 4:8,
9.6 Hz), 4.59 (1H, dd, J ¼ 4:8, 9.6 Hz), 4.69 (2H, d,
J ¼ 6:6 Hz), 4.98 (1H, m), 5.25 (1H, m), 7.02–7.03 (2H,
m), 7.03–7.09 (2H, m), 7.14–7.18 (4H, m), 7.29 (1H, d,
J ¼ 7:8 Hz), 7.56 (1H, d, J ¼ 8:4 Hz). HRMS m=z
(½M þ Hꢁþ): Calcd. for C48H67N8O10 915.4975, found
915.4971.
50=1) to give Fmoc-Trp(2-Ger) 33 (246 mg, 0.437
23
mmol, 60%) as a yellow oil. ½ꢃꢁ
ꢃ0:62ꢂ (c 1.0,
D
CH3OH). NMR (600 MHz) ꢂH (CD3COCD3) 1.54 (3H,
s), 1.61 (3H, s), 1.76 (3H, s), 2.01–2.14 (4H, m), 3.26
(1H, dd, J ¼ 7:7, 14.6 Hz), 3.39 (1H, dd, J ¼ 5:6,
14.6 Hz), 3.55–3.62 (2H, m), 4.16 (1H, t, J ¼ 7:2 Hz),
4.24 (1H, d, J ¼ 7:4 Hz), 5.08 (1H, m), 5.41 (1H, t,
J ¼ 6:8 Hz), 6.52 (1H, d, J ¼ 8:5 Hz), 6.99 (1H, t,
J ¼ 7:0 Hz), 7.01 (1H, t, J ¼ 7:0 Hz), 7.25–7.30 (3H,
m), 7.38 (2H, t, J ¼ 7:4 Hz) 7.61–7.65 (3H, m) 7.82
(2H, d, J ¼ 7:6 Hz). NMR (150 MHz) ꢂC (CD3COCD3)
16.2, 17.5, 25.6, 25.6, 27.0, 27.5, 29.2, 40.2, 47.7, 55.6,
67.0, 106.4, 111.2, 118.5, 119.4, 120.5, 121.1, 121.7,
124.8, 125.9, 127.7, 129.6, 131.6, 136.6, 137.2, 137.3,
141.8, 144.8, 144.8, 156.6, 173.6. Anal. Found: C,
76.67; H, 6.93; N, 5.20. Calcd. for C36H38N2O4: C,
76.84; H, 6.81; N, 4.98.
2-Geranyl 39. NMR (600 MHz) ꢂH (CD3CN, D2O)
0.60 (3H, d, J ¼ 6:6 Hz), 0.73 (3H, t, J ¼ 7:2 Hz), 0.96
(1H, m), 1.23 (1H, m), 1.51 (3H, s), 1.57 (3H, s), 1.59
(1H, m), 1.82 (3H, s), 1.82–1.85 (3H, m), 1.97–2.18 (9H,
m), 2.79 (1H, dd, J ¼ 8:4, 14.4 Hz), 2.90 (1H, dd,
J ¼ 4:8, 14.4 Hz), 3.06 (1H, dd, J ¼ 8:4, 14.4 Hz), 3.15
(1H, dd, J ¼ 8:4, 14.4 Hz), 3.23 (2H, s), 3.38 (2H, d,
J ¼ 7:4 Hz), 3.97–3.99 (2H, m), 4.16 (1H, dd, J ¼ 4:8,
Solid-phase peptide synthesis.
Peptide bond formation. Solid-phase peptide syn-
thesis was performed on a peptide synthesizer using
Fmoc amino acid synthesis protocol with O-benzotria-

2380
M. OKADA et al.
9.0 Hz), 5.07 (1H, m), 5.28 (1H, m), 7.02–7.09 (4H, m),
7.21–7.25 (3H, m), 7.30 (1H, d, J ¼ 7:9 Hz), 7.38 (1H,
d, J ¼ 7:6 Hz). HRMS m=z (½M þ Hꢁþ): Calcd. for
C48H67N8O10 915.4975, found 915.4966.
LC–MS analysis. A cyanopropyl column (2:0 ꢀ
250 mm ID, Develosil CN-UG-5, Nomura Chemical)
was attached to a HPLC system and equilibrated with
35% CH3CN in 0.05% aqueous formic acid over 10 min
before injection. The peptides (synthetic peptides; 10 ꢁM
in 10 ꢁl, isolated peptide was 5 ml broth in 10 ꢁl) were
eluted at a flow rate of 0.2 ml/min with a gradient of 35–
45% CH3CN in 0.05% aqueous formic acid for 20 min.
The elution was recorded into mass spectroscopy with
Q1MI scan mode (m=z value 915.5).
4-Geranyl 40. NMR (600 MHz) ꢂH (CD3CN, D2O)
0.62 (3H, d, J ¼ 6:6 Hz), 0.72 (3H, t, J ¼ 7:2 Hz), 0.95
(1H, m), 1.21 (1H, m), 1.55 (3H, s), 1.62 (3H, s), 1.63
(1H, m), 1.68 (3H, s), 1.78–1.81 (3H, m), 1.96–2.16 (9H,
m), 2.82 (1H, dd, J ¼ 9:6, 14.4 Hz), 3.02 (1H, dd,
J ¼ 8:4, 14.4 Hz), 3.17–3.21 (3H, m), 3.38 (1H, dd,
J ¼ 7:2, 14.4 Hz), 3.71 (2H, t, J ¼ 6:0 Hz), 4.01–4.05
(2H, m), 4.21 (1H, dd, J ¼ 4:8, 8.4 Hz), 4.55–4.60 (2H,
m), 5.09 (1H, m), 5.30 (1H, m), 6.81 (1H, d,
J ¼ 7:2 Hz), 7.00–7.03 (2H, m), 7.12–7.14 (2H, m),
7.18–7.24 (4H, m). HRMS m=z (½M þ Hꢁþ): Calcd. for
C48H67N8O10 915.4975, found 915.5000.
5-Geranyl 41. NMR (600 MHz) ꢂH (CD3CN, D2O)
0.53 (3H, d, J ¼ 6:7 Hz), 0.72 (3H, t, J ¼ 7:3 Hz), 0.91
(1H, m), 1.14 (1H, m), 1.46 (1H, m), 1.56 (3H, s), 1.62
(3H, s), 1.71 (3H, s), 1.78–1.87 (3H, m), 1.91–2.22 (9H,
m), 2.82 (1H, dd, J ¼ 9:4, 14.0 Hz), 3.02 (1H, dd,
J ¼ 5:4, 14.0 Hz), 3.08 (1H, dd, J ¼ 7:1, 14.4 Hz), 3.07–
3.10 (3H, m), 3.41 (2H, d, J ¼ 7:1 Hz), 3.97 (1H, dd,
J ¼ 7:3, 12.6 Hz), 4.07 (1H, d, J ¼ 8:3 Hz), 4.48 (1H,
dd, J ¼ 5:4, 9.4 Hz), 4.58 (1H, dd, J ¼ 6:2, 7.1 Hz), 5.08
(1H, m), 5.35 (1H, m), 6.95 (1H, d, J ¼ 8:3 Hz), 7.07–
7.09 (3H, m), 7.17–7.22 (3H, m), 7.30 (1H, d,
J ¼ 8:3 Hz), 7.36 (1H, s). HRMS m=z (½M þ Hꢁþ):
Calcd. for C48H67N8O10 915.4975, found 915.4952.
6-Geranyl 42. NMR (600 MHz) ꢂH (CD3CN, D2O)
0.47 (3H, d, J ¼ 6:6 Hz), 0.69 (3H, t, J ¼ 7:3 Hz), 0.89
(1H, m), 1.13 (1H, m), 1.38 (1H, m), 1.54 (3H, s), 1.63
(3H, s), 1.71 (3H, s), 1.82–1.88 (3H, m), 1.95–2.22 (9H,
m), 2.73 (1H, dd, J ¼ 9:0, 14.4 Hz), 2.91 (1H, dd,
J ¼ 5:4, 14.4 Hz), 3.06 (1H, dd, J ¼ 8:4, 15.0 Hz), 3.21–
3.27 (3H, m), 3.40 (2H, d, J ¼ 7:8 Hz), 3.92 (1H, d,
J ¼ 7:8 Hz), 4.08 (1H, dd, J ¼ 9:4, 13.8 Hz), 4.48 (1H,
dd, J ¼ 5:4, 9.0 Hz), 4.59 (1H, dd, J ¼ 5:4, 8.4 Hz), 5.09
(1H, m), 5.36 (1H, m), 6.89 (1H, d, J ¼ 7:8 Hz), 7.04
(1H, s), 7.07 (2H, d, J ¼ 6:6 Hz), 7.16 (1H, s), 7.16–7.23
(3H, m), 7.36 (1H, d, J ¼ 7:8 Hz). HRMS m=z
(½M þ Hꢁþ): Calcd. for C48H67N8O10 915.4975, found
915.4981.
Results
The molecular formula of ComXRO-E-2
The partial purification of ComXRO-E-2 was carried
out as described previously.4) Further purification was
performed by ODS HPLC with gradient elution of
CH3CN in 0.1% aqueous ammonium acetate. The most
active fraction, determined by the ꢀ-galactosidase assay,
which uses a srfA–lacZ reporter construct,2,5) was
applied to a cyanopropyl column, and the peak detected
by UV absorbance at 220 nm was collected. This
fraction was directly injected into an ESI–TOF mass
spectrometer to obtain a HRMS. The mass of the
observed ½M þ Hꢁþ ion was 915.4983, assigned to the
composition C48H67N8O10, and the molecular formula
of ComXRO-E-2 was therefore estimated to be
C48H66N8O10. The amino acid sequence of ComXRO-E-2
determined from DNA and peptide sequencing was
GIFWEQ, calculated to be C38H50N8O10.4) The remain-
ing C10H16 indicated an isoprenoidal group. Based on
this result, we estimated that the modified tryptophan in
the peptide was simply the substitution for a trypto-
phanyl proton by a geranyl group.
Geranyl tryptophan synthesis
In order to determine the complete structure of
ComXRO-E-2, we attempted to synthesize peptides that
possess the amino acid sequence of ComXRO-E-2 with a
geranyl tryptophan (TrpGer) residue. We synthesized six
geranyl substituted tryptophan derivatives: 1-geranyl
tryptophan [Trp(1-Ger)], 2-geranyl tryptophan [Trp(2-
Ger)], 4-geranyl tryptophan [Trp(4-Ger)], 5-geranyl
tryptophan [Trp(5-Ger)], 6-geranyl tryptophan [Trp(6-
Ger)], and 7-geranyl tryptophan [Trp(7-Ger)]. We then
synthesized each geranyl substituted Fmoc-tryptophan
[Fmoc-Trp(n-Ger)]. Finally, peptide bond formation was
carried out either with a peptide synthesizer or by
manual synthesis, with an Fmoc solid-phase peptide
synthesis strategy. Protecting groups and resin were
cleaved, and HPLC purification was used to isolate the
desired products.
7-Geranyl 43. NMR (600 MHz) ꢂH (CD3CN, D2O)
0.46 (3H, d, J ¼ 6:6 Hz), 0.84 (3H, t, J ¼ 7:3 Hz), 0.89
(1H, m), 1.19 (1H, m), 1.51 (1H, m), 1.53 (3H, s), 1.64
(3H, s), 1.69 (3H, s), 1.79–1.85 (3H, m), 1.95–2.22 (9H,
m), 2.70 (1H, dd, J ¼ 8:4, 13.6 Hz), 2.88 (1H, dd,
J ¼ 5:4, 13.6 Hz), 3.06 (1H, dd, J ¼ 8:4, 15.0 Hz), 3.21–
3.27 (3H, m), 3.51 (2H, d, J ¼ 6:6 Hz), 3.93 (1H, d,
J ¼ 7:8 Hz), 4.09 (1H, dd, J ¼ 4:8, 13.2 Hz), 4.47 (1H,
dd, J ¼ 5:4, 8.4 Hz), 4.61 (1H, dd, J ¼ 5:4, 8.4 Hz), 5.09
(1H, m), 5.39 (1H, m), 6.93 (1H, d, J ¼ 7:8 Hz), 7.00
(1H, d, J ¼ 7:8 Hz), 7.05 (2H, d, J ¼ 6:6 Hz), 7.12 (1H,
s), 7.16–7.22 (3H, m), 7.31 (1H, d, J ¼ 7:8 Hz). HRMS
m=z (½M þ Hꢁþ): Calcd. for C48H67N8O10 915.4975,
found 915.4994.
For the synthesis of Trp(1-Ger), an appropriate
protected tryptophan was used as starting material with
the synthetic route shown in Fig. 1. Treatment of Nꢃ-
phthaloyl tryptophan methyl ester 1 with geranyl
bromide and NaH as base in DMF gave the intermediate
2. The phthaloyl group was cleaved with hydrazine, the

Synthetic Studies on Putative ComX Pheromone
2381
Fig. 1. Synthesis of the Fmoc-Trp(1-Ger) 3.
Reaction conditions: (a) NaH, geranyl bromide, DMF, 0 ꢂC, 61%. (b) NH2NH2, CH2Cl2/CH3OH, 40 ꢂC, 83%. (c) 1 M aqueous LiOH, THF.
(d) Fmoc-OSu, 1 M aqueous Na2CO3, CH3CN, 87% in 2 steps.
Fig. 2. Synthetic Strategy of Putative Peptides.
methyl ester was hydrolyzed under basic conditions, and
an Fmoc group was introduced by a conventional
method to give Fmoc-Trp(1-Ger) 3.
The intermediate 8 of Trp(4-Ger) was prepared by
regioselective geranyl substitution of 1-(triisopropyl-
silyl)gramine 5 on the 4-position, as reported by
For the other five compounds, we chose convergent
and short step synthesis of TrpGer due to its efficiency
and versatility. An outline of the synthetic route is
illustrated in Fig. 2. We selected geranyl bromide, the
Shinohara et al.8) before Schollkopf’s amino acid
¨
synthesis, as shown in Fig. 3. Treatment of 1-(triisopro-
pylsilyl)gramine 5 with t-BuLi in ether at 0 ꢂC for 1 h
was mediated directly by lithiation of gramine at the 4-
position assisted by its amino functionality. The result-
ing species was reacted with geranyl bromide to give the
compound 6. Replacement of the protecting group in 2
steps from triisopropylsilyl (TIPS) to a Boc group was
followed by treatment with ethyl chloroformate to give
Schollkopf chiral auxiliary 4,6,7) and suitable gramine
¨
derivatives as starting materials. TrpGer derivatives
were synthesized with Schollkopf’s asymmetric amino
¨
acid synthesis and lithium-mediated or palladium-cata-
lyzed geranyl substitution.

2382
M. OKADA et al.
Fig. 3. Synthesis of the Intermediate 8.
Reaction conditions: (a) t-BuLi, Et2O, DMF, ꢃ78 ! 0 ꢂC, 1 h, then geranyl bromide, 44%. (b) TBAF, THF, 77%. (c) (Boc)2O, Et3N, DMAP,
THF, 93%; (d) ethyl chloroformate, toluene. (e) n-BuLi, TMEDA, 4, THF, ꢃ78 ! 0 ꢂC, 78% from 7.
the chloride, which was used in the next step without
purification because of its instability. Metallation of the
The intermediate 15 of Trp(2-Ger) was synthesized
from 12 by treatment with LDA, geranyl bromide, and
CuCN in moderate yield (67%), as shown in Fig. 5. Due
to a steric effect and acidity, LDA was deprotonated
only at the 2-position of the indole ring,9) and the
formation of an organocopper complex mediated by
CuCN improved the yield of 15. On the other hand,
changing the protecting group from Boc to a diethylbu-
tanoyl group enabled deprotonation at the 7-position of
the indole ring, as discussed by Fukuda et al.10) The Boc
group was cleaved with NaOMe/CH3OH to give
compound 16, followed by treatment with NaH and
diethylbutanoyl chloride to give the diethylbutanoyl
derivative 17. Since the 2-position of the indole ring was
sterically hindered and the neighboring diethylbutanoyl
group participated in the reaction, the treatment with s-
BuLi and TMEDA to the diethylbutanoyl derivative 17
gave only the 7-position deprotonated compound, which
was reacted with geranyl bromide to give the inter-
mediate 18. The intermediates 21 of Trp(5-Ger) and 22
of Trp(6-Ger) were synthesized with Suzuki coupling, as
shown in Fig. 5, because transmetallation of the halogen
mediated by various alkyllithum reagents failed.11) The
bromide 13 or 14 was treated with catalytic PdCl2(dppf),
triethylamine, and pinacolborane to give the borate 19 or
20 respectively.12) The borate 19 or 20 was reacted with
geranyl bromide under Suzuki coupling conditions using
2 M aqueous KOH in THF to give the intermediate 21 or
22 respectively. Reverse geranyl substitution was ob-
served as a byproduct from the borate 19 or 20, but the
desired intermediate 21 or 22 was obtained in moderate
yield after the byproduct was separated carefully by
column chromatography. Employment of Pd(PPh3)4 and
Pd(Pt-Bu3)2 as a palladium catalyst gave undesired
reactions such as reductive elimination, and under
anhydrous conditions using K2CO3 or triethylamine as
base in DMF, the starting materials were recovered.13)
Schollkopf chiral auxiliary 4 with n-BuLi in THF was
¨
followed by alkylation of the chloride in the presence of
TMEDA to provide the intermediate 8 in moderate yield
(78%). The addition of TMEDA improved the diaster-
eomeric ratio up to 30:1 (trans:cis) detected by 1H-NMR
and NOESY analysis.
The precursors 12, 13, or 14 of regio-selective geranyl
substitution at the 2, 5, 6, and 7-position were also
synthesized along with Schollkopf’s amino acid syn-
¨
thesis, as shown in Fig. 4. Treatment of 1-Boc-gramine
9, 1-Boc-5-bromogramine 10, or 1-Boc-6-bromogra-
mine 11 with ethyl chloroformate gave the respective
chlorides. These chlorides, except for the 6-bromo
compound, were relatively stable, and could be purified
by recrystalization. Alkylation of Schollkopf chiral
¨
auxiliary 4 with these chlorides furnished compounds
12, 13, or 14 in higher yield (88ꢄ97%) compared to that
of 8 and with a similar trans diastereoselectivity.
Fig. 4. Synthesis of the Intermediates 12, 13, and 14.
Reaction conditions: (a) ethyl chloroformate, toluene. (b) n-BuLi,
TMEDA, 4, THF, ꢃ78 ! 0 ꢂC.

Synthetic Studies on Putative ComX Pheromone
2383
Fig. 5. Synthesis of the Intermediates 18, 21, and 22.
Reaction conditions: (a) LDA, CuCN, geranyl bromide, Et2O, ꢃ78 ꢂC, 67%. (b) NaOMe/CH3OH, THF, 97%. (c) NaH, diethylbutanoyl
chloride, THF, 0 ꢂC, 63%. (d) s-BuLi, TMEDA, geranyl bromide, Et2O, ꢃ78 ꢂC, 41%. (e) PdCl2(dppf), Et3N, pinacolborane, toluene, 90 ꢂC. (f)
PdCl2(dppf), geranyl bromide, 2 M aqueous KOH, THF, 60 ꢂC.
All of the geranyl substituted intermediates 8, 15, 18,
21, 22 were treated under similar conditions to give
geranyl substituted Fmoc-tryptophans 33ꢄ37, as shown
in Fig. 6. Boc and diethylbutanoyl groups were cleaved
with NaOMe/CH3OH in THF. Cleavage of the diethyl-
butanoyl group required a much longer time (2 d) than
the Boc group (3 h). When t-BuOK/H2O was employed
in place of NaOMe/CH3OH, removal of the diethyl-
butanoyl group occurred quickly, but racemization
occurred. Acidic hydrolysis was carefully operated
at 0 ꢂC to give ethyl esters 28ꢄ32, followed by
basic hydrolysis to give free Trp(n-Ger). Finally
Fmoc protection was achieved under general conditions
to give the desired geranyl substituted tryptophans
33ꢄ37.
Peptide synthesis
For solid-phase peptide synthesis as shown in Fig. 7,
we chose a 2-chlorotrityl (Clt) resin, Gln without a
protecting group, and Glu protected by a 2-phenyl-2-
propyl (Pp) group on the side chain because all of the
geranyl substituted tryptophans were acid labile. These
resin and protecting groups can be cleaved under
extremely mild acidic conditions. After the first peptide
bond formation in the peptide synthesizer, the resin was
removed from the machine. The resin was manually
treated with each Fmoc-Trp(Ger) prepared, using HATU
and diisopropylethylamine (DIPEA) in NMP. Then the
manually treated resin was remounted in the automatic
peptide synthesizer and the remaining three amino acid
residues were coupled. Crude peptides were obtained

2384
M. OKADA et al.
Fig. 6. Synthesis of Fmoc-Trp(n-Ger) 33ꢄ37.
Reaction conditions: (a) NaOMe/CH3OH, THF. (b) 1.5 M aqueous HCl, THF, 0 ꢂC. (c) 1 M aqueous LiOH, THF. (d) Fmoc-OSu, 1 M aqueous
Na2CO3, CH3CN.
Fig. 7. Synthesis of the Putative Peptides 38ꢄ43.
Reaction conditions: (a) HATU, DIPEA, NMP, Fmoc-Trp(n-Ger) 3, 33ꢄ37. (b) 4% TFA, 2.5% mercaptoethanol, CH2Cl2.

Synthetic Studies on Putative ComX Pheromone
2385
although none of the synthetic peptides showed activity
up to 3 ꢁM. Thus the synthetic peptides were different
from the natural pheromone with respect to both
biological activity and physicochemical properties.
These results suggest that the manner of modification
of the tryptophan residue in ComX pheromone is more
complex than expected.
Discussion
From previous reports, we believed that the ComX
pheromone is isoprenylated, but the possibility remained
that the tryptophan residue was not only isoprenylated
but also reduced.2) We have now investigated the HRMS
and the molecular formula of ComXRO-E-2, and con-
firmed that ComXRO-E-2 possesses a non-reduced ger-
anylated tryptophan residue.
Based on the hypothesis that ComXRO-E-2 is simply
geranylated at a position on the indole ring, we
synthesized regio- and stereo-selective geranyl trypto-
phans. Our method should be widely applicable to
synthesis of regio- and stereo-selective functionalized
tryptophans such as farnesyl and geranylgeranyl trypto-
phans. The 5- and 6-substituted tryptophans were
prepared from 5- and 6-bromoindole, because direct
geranylation as used for formation of other geranyl
tryptophans was impossible. After several trials, we
finally found that Suzuki coupling gave a satisfactory
result. This procedure should be useful for the intro-
duction of alkyl chains to the indole 5- or 6-positions.
Using geranyltryptophan, six putative ComXRO-E-2
candidates were synthesized by solid-phase peptide
synthesis. We found that the natural pheromone is acid
labile, because its biological activity decreased after
concentrating the HPLC eluant containing 0.1% TFA
and CH3CN at room temperature under vacuum. Hence,
we chose the Clt resin, which is cleaved under mild
acidic conditions. We also used Gln without a protecting
group and Glu with Pp protection. Peptide bond
formation between Trp(Ger) and Glu was carried out
manually to save the synthesized Fmoc-Trp(Ger) and to
obtain a high reaction yield. We used 1.5 equivalents of
Fmoc-Trp(Ger) and HATU as the coupling reagent in
the manual reaction, in contrast to 10 equivalents of
Fmoc-amino acids and HBTU reagent in the automatic
peptide synthesizer. The cleavage reaction was carried
out carefully under mild acidic conditions and for a short
time. The isolated yield of purified peptides was
approximately 5%, which is very low compared with
that of the simple peptide having the ComXRO-E-2
sequence without geranyl modification.
Fig. 8. LC–MS Analysis of Natural Product and Synthetic Peptides.
after the cleavage of protecting groups and resin with
4% TFA in CH2Cl2 for 1 h. Then each crude peptide was
purified by HPLC, and we finally obtained the desired
six peptides 38ꢄ43. We confirmed that all peptides had
1
the proper structures by H-NMR and HRMS analysis
and by protein sequencing.
Comparison of synthetic peptides and ComXRO-E-2
Using LC–MS, we compared the synthetic peptides
and the natural pheromone. When an HPLC system
was equipped with a cyanopropyl column, all the
synthetic peptides showed distinct retention times, as
shown in Fig. 8. During the analysis of synthetic
peptides by LC on a cyanopropyl column with CH3CN
and formic acid, the intensity of the ion at m=z 915.5 was
monitored by MS. Compared to ComXRO-E-2, all six
peptides prepared showed similar hydrophobicity, and
the 6-geranyl substituted peptide 42 had almost the same
retention time. However, using a different solvent
system (CH3CN–0.1% NH4OAc) for HPLC, the natural
peptide (Rt ¼ 22:80) and 42 (Rt ¼ 26:31) were com-
pletely distinguishable. Thus, none of the peptides
synthesized was identical to the natural ComX pher-
omone.
After purification of the crude peptides by HPLC, the
purity was high enough for NMR and MS analysis. We
could identify the substitution pattern of Trp in each
peptide, although only a small amount of these peptides
could be dissolved in NMR solvent. The molecular
weights of these six peptides and the natural one were
exactly the same as measured by HRMS.
We investigated the biological activity of the synthe-
sized peptides by a ꢀ-galactosidase assay.2,5) The natural
pheromone showed biological activity at about 1 nM,

2386
M. OKADA et al.
We then established the LC–MS analysis conditions
believe that the chemical structure of ComX pheromone
will be determined in the near future.
to separate all six peptides synthesized. Based on this
LC–MS analysis, only peptide 42, having a 6-geranyl
tryptophan residue, showed the same retention time as
the natural pheromone. However, this peptide showed a
completely different retention time relative to the natural
pheromone by HPLC using a different solvent system.
Thus all synthetic peptides had a different structure from
the natural ComX peptide. Since the synthetic peptides
were acid labile and exhibited hydrophobicity similar to
the natural pheromone, we expected that some of these
peptides would show weak biological activity. We
investigated the biological activities of the synthetic
peptides, but none of them showed activity, even at very
high concentrations. These results indicate that the
manner of modification of the tryptophan residue is not
as simple as we expected.
Acknowledgment
Work performed in Newark was supported by NIH
Grant GM57720. Soo Jeong Cho was supported by the
Post-doctoral Fellowship Program of the Korea Science
and Engineering Foundation (KOSEF).
References
1) Williams, P., Baldwin, T. J., and Downie, J. A., Bacterial
crosstalk: communication between bacteria, plant and
animal cells. In ‘‘Microbial Signalling and Communica-
tion’’, eds. England, R. R., Hobbs, G., Bainton, N. J., and
Roberts, D. McL., Cambridge University Press, Cam-
bridge, pp. 1–32 (1999).
There are four examples of naturally occurring
posttranslational modification on a tryptophan residue.
The first is a cross-link between two tryptophan
residues, in which one is modified to indole-6,7-dione
and attached at its 4-position to the 2-position of the
other.14) The second is C2-mannopyranosylated trypto-
phan.15,16) The third is 6-bromotryptophan,17) and the
last is kynurenine [2-amino-4-(2-aminophenyl)-4-oxo-
butanoic acid], formed by oxidative modification.18,19)
These four examples, however, do not appear to help in
the structural determination of the modified tryptophan
residue in ComX. Natural alkaloids substituted by
isoprenylation derived from tryptophan have been
reported.20) Most of them, such as tryprostatin, are
simply prenylated on their indole rings, as are the
compounds we synthesized in this study.
2) Magnuson, R., Solomon, J., and Grossman, A. D.,
Biochemical and genetic characterization of a compe-
tence pheromone from B. subtilis. Cell, 77, 207–216
(1994).
3) Schneider, K. B., Palmer, T. M., and Grossman, A. D.,
Characterization of comQ and comX, two genes required
for production of ComX pheromone in Bacillus subtilis.
J. Bacteriol., 184, 410–419 (2002).
4) Ansaldi, M., Marolt, D., Stebe, T., Mandic-Mulec, I.,
and Dubnau, D., Specific activation of the bacillus
quorum-sensing systems by isoprenylated pheromone
variants. Mol. Microbiol., 44, 1561–1573 (2002).
5) Tortosa, P., Logsdon, L., Kraigher, B., Itoh, Y., Mandic-
Mulec, I., and Dubnau, D., Specifity and genetic poly-
morphism of the Bacillus competence quorum-sensing
system. J. Bacteriol., 183, 451–460 (2001).
6) Groth, U., Schmeck, C., and Schollkopf, U., Asymmetric
¨
Isoprenoidal peptides were first discovered as sex
pheromones of basidiomycetous yeasts.21,22) These pep-
tides have cysteine residues at their carboxy termini with
modification of their sulfhydryl groups by farnesylation.
This type of farnesylation and geranylgeranylation of
thiols is now well studied and is known in many
proteins.23) The posttranslational farnesyl or geranyl-
geranyl modification is important for the function of
these proteins. The geranyl or farnesyl groups in ComX
are also essential for the function of this pheromone,
because the unmodified peptide shows no pheromone
activity,2,4) and the similar peptides described in the
present study also lacked biological activity. Determi-
nation of the ComX structure is essential to clarify its
biological role.
In this study, we synthesized six peptides for which
expected a high possibility for the chemical structure of
ComX pheromone. The possible structure remaining is a
tryptophan residue geranylated at the alpha carbon or
nitrogen, or substituted by two isoprenes. We cannot
exclude the possibility that the ComX pheromone has D-
amino acid(s), although the peptide is coded on DNA.
Now we are attempting to synthesize peptides having
possible ComX pheromone structure and to purify a
natural pheromone from microbial fermentation. We
synthesis of a amino acid benzyl esters via the bisbenzyl
bislactim ether of cyclo(-L-Val-Gly-). Liebigs Ann.
Chem., 321–323 (1993).
7) Ma, C., Liu, X., Li, X., Flippen-Anderson, J., Yu, S., and
Cook, J. M., Efficient asymmetric synthesis of bio-
logically important tryptophan analogues via a palla-
dium-mediated heteroannulation reaction. J. Org. Chem.,
66, 4525–4542 (2001).
8) Shinohara, H., Fukuda, T., and Iwao, M., A formal
synthesis of optically active clavitropic acids, unusual
azepinoindole-type ergot alkaloids. Tetrahedron, 55,
10989–11000 (1999).
9) Zhao, S., Gan, T., Yu, P., and Cook, J. M., Total
synthesis of tryprostatin A and B as well as their
enantiomers. Tetrahedron Lett., 39, 7009–7012 (1998).
10) Fukuda, T., Maeda, R., and Iwao, M., Directed C-7
lithiation of 1-(2,2-diethylbutanoyl)indoles. Tetrahe-
dron, 55, 9151–9162 (1999).
11) Suzuki, A., Recent advances in the cross-coupling
reactions of organoboron derivatives with organic
electrophiles, 1995–1998. J. Organomet. Chem., 576,
147–168 (1999).
12) Murata, M., Oyama, T., Watanabe, S., and Masuda, Y.,
Palladium-catalyzed borylation of aryl halides or triflates
with dialkoxyborane: A novel and facile synthetic route
to arylboronates. J. Org. Chem., 65, 164–168 (2000).
13) Dai, C., and Fu, G. C., The first general method for

Synthetic Studies on Putative ComX Pheromone
2387
palladium-catalyzed Negishi cross-coupling of aryl
and vinyl chlorides: use of commercially available
Pd(P(t-Bu)3)2 as a catalyst. J. Am. Chem. Soc., 123,
2719–2724 (2001).
M. J., Lim, H., Yates, J. R. III, and Barry, B. A.,
Posttranslational modifications in the CP43 subunit of
photosystem II. Proc. Natl. Acad. Sci. U.S.A., 99,
14676–14681 (2002).
14) McIntire, W. S., Wemmer, D. E., Chistoserdov, A., and
Lindstrom, M. E., A new cofactor in a prokaryotic
enzyme: tryptophan tryptophylquinone as the redox
prosthetic group in methylamine dehydrogenase. Sci-
ence, 252, 817–824 (1991).
19) Taylor, S. W., Fahy, E., Murray, J., Capaldi, R. A., and
Ghosh, S. S., Oxidative post-translational modification
of tryptophan residues in cardiac mitochondorial pro-
teins. J. Biol. Chem., 278, 19587–19590 (2003).
20) Cui, C. B., Kakeya, H., Okada, G., Onose, R., Ubukata,
M., Takahashi, I., Isono, K., and Osada, H., Tryprosta-
tins A and B, novel mammalian cell cycle inhibitors
produced by Aspergillus fumigatus. J. Antibiotics, 48,
1382–1384 (1995).
21) Kamiya, Y., Sakurai, A., Tamura, S., Takahashi, N.,
Tsuchiya, T., Abe, K., and Fukui, S., Structure of
rhodotoruicine A, a peptidyl factor inducing mating tube
formation in Rhodosporidium toruloides. Agric. Biol.
Chem., 43, 363–369 (1979).
22) Sakagami, Y., Yoshida, M., Isogai, A., and Suzuki, A.,
Peptidal sex hormones inducing conjugation tube for-
mation in compatible mating-type cells of Tremella
mesenterica. Science, 212, 1525–1527 (1981).
23) Clarke, S., Protein isoprenylation and methylation at
carbonyl-terminal cysteine residues. Ann. Rev. Biochem.,
61, 355–386 (1992).
15) Hofsteenge, J., Muller, D. R., de Beer, T., Loffler, A.,
¨
¨
Richter, W. J., and Vliegenthart, J. F. G., New type of
linkage between a carbohydrate and a protein: C-
glycosylation of a specific tryptophan residue in human
RNase Us. Biochemistry, 33, 13524–13530 (1994).
16) de Beer, T., Vliegenthart, J. F. G., Loffler, A., and
¨
Hofsteenge, J., The hexopyranosyl residue that is C-
glycosidically linked to the side chain of tryptophan-7
in human Us is a-mannopyranose. Biochemistry, 34,
11785–11789 (1995).
17) Jimenez, E. C., Craig, A. G., Watkins, M., Hillyard,
D. R., Gray, W. R., Gulyas, J., Rivier, J. E., Cruz, L. J.,
and Olivera, B. M., Bromocontryphan: post-translational
bromination of tryptophan. Biochemistry, 36, 989–994
(1997).
18) Anderson, L. B., Maderia, M. M., Ouellette, A. J. A.,
Putnam-Evans, C., Higgins, L., Krick, T., MacCoss,