Synthesis of Gramicidin S and Its Analogues via an On-Resin Macrolactamization Assisted by a Predisposed Conformation of the Linear Precursors

Article abstract of DOI:10.1021/jo035712x

A simple and efficient preparation of gramicidin S and its analogues is described. It involves solid-phase peptide synthesis and on-resin macrolactamization without side chain protection, affording cyclic products in high yield and high purity. The high specificity of the cyclization reaction was shown to originate in the formation of a pre-organized conformation of the linear biosynthetic precursor of gramicidin S. This facile method will provide convenient access to the analogues of the natural product for functional optimization to counter microbial resistance.

Full text of DOI:10.1021/jo035712x

Syn th esis of Gr a m icid in S a n d Its An a logu es via a n On -Resin
Ma cr ola cta m iza tion Assisted by a P r ed isp osed Con for m a tion of
th e Lin ea r P r ecu r sor s
Xianzhang Bu, Xiaoming Wu, Na Lee J oyce Ng, Chun Kit Mak, Chuanguang Qin, and
Zhihong Guo*
Department of Chemistry and Biotechnology Research Institute, The Hong Kong University of Science and
Technology, Clear Water Bay, Kowloon, Hong Kong SAR, China
chguo@ust.hk
Received November 21, 2003
A simple and efficient preparation of gramicidin S and its analogues is described. It involves solid-
phase peptide synthesis and on-resin macrolactamization without side chain protection, affording
cyclic products in high yield and high purity. The high specificity of the cyclization reaction was
shown to originate in the formation of a pre-organized conformation of the linear biosynthetic
precursor of gramicidin S. This facile method will provide convenient access to the analogues of
the natural product for functional optimization to counter microbial resistance.
In tr od u ction
Gramicidin S (1a ) is a cyclic decapeptide antibiotic with
two repetitive pentapeptide units of D-Phe-Pro-Val-Orn-
Leu, which is produced by Bacillus brevis.1 In liquid
media, it exhibits potent antibiotic activities against a
broad spectrum of both Gram-positive and Gram-nega-
tive bacteria, as well as several pathogenic fungi.2
However, it also exhibits high hemolytic activity that
restricts the antibiotic to topical application.3 Under
physiological conditions, the antibiotic adopts a pre-
formed rigid antiparallel â-pleated sheet structure that
is amphipathic with the polar side chains on one side and
the hydrophobic side chains on the other (Figure 1).4 The
natural product’s activity is dependent on its interaction
with the cell membranes, resulting in disruption of
normal barrier functions of the cell walls.5 Due to this
unique mode of action, it is likely to have low risk of
provoking bacterial resistance because development of
such resistance requires significant alteration of the lipid
composition of the cell wall.6 Therefore, it has become a
focal point of interest to optimize this antibiotic’s thera-
F IGURE 1. Antiparallel â-pleated sheet structure of grami-
cidin S.
peutic index for treatment of microbial resistance that
has become an increasing threat to public health.7
Availability of gramicidin S analogues is a major issue
in the functional optimization of the natural product.
Traditional synthesis of the analogues involves ring
closure at high dilution after C-terminal activation by
N-hydroxysuccinimide ester (ONSu) or azide.1b,8 A num-
ber of other methods are also available for cyclic peptide
synthesis.9 Nonetheless, these methods are insufficient
to meet the need of providing economic access to large
(1) (a) Gause, G. G.; Brazhnikova, M. G. Nature 1944, 154, 703. (b)
Izumiya, N.; Kato, T.; Aoyagi, H.; Waki, M.; Kondo, M. In Synthetic
Aspects of Biologically Active Cyclic Peptides-Gramicidin and Tyro-
cidines; Kodansha Ltd. and Wiley: Tokyo, 1979.
(2) (a) Kondejewski, L. H.; Farmer, S. W.; Wishart, D. S.; Hancock,
R. E. W.; Hodges, R. S. Int. J . Pept. Protein Res. 1996, 47, 460-466.
(b) Kondejewski, L. H.; Farmer, S. W.; Wishart, D.; Kay, C. M.;
Hancock, R. E. W.; Hodges, R. S. J . Biol. Chem. 1996, 271, 25261-
25268.
(7) (a) Neu, H. C. Science 1992, 257, 1064-1073. (b) Cohen, M. L.
Science 1992, 257, 1050-1055.
(3) Lambert, H. P.; O’Grady, F. W. In Antibiotic and Chemotherapy,
6th ed.; Churchill Livingstone: Edinburgh, U.K., 1992; pp 232-233.
(4) (a) Stern, A.; Gibbons, W. A.; Craig, L. C. Proc. Natl. Acad. Sci.
U.S.A. 1968, 61, 734-741. (b) Ohnishi, M.; Urry, D. W. Biochim.
Biophys. Res. Commun. 1969, 36, 194-202.
(8) (a) Kondejewski, L. H.; Farmer, S. W.; Wishart, D. S.; Kay, C.
M.; Hancock, R. E. W.; Hodges, R. S. J . Biol. Chem. 1996, 271, 25261-
25268. (b) Kondejewski, L. H.; J elokhani-Niaraki, M.; Farmer, S. W.;
Bruce, L.; Kay, C. M.; Sykes, B. D.; Hancock, R. E. W.; Hodges, R. S.
J . Biol. Chem. 1999, 274, 13181-13192. (c) McInnes, C.; Kondejewski,
L. H.; Hodges, R. S.; Sykes, B. D. J . Biol. Chem. 2000, 275, 14287-
14294. (d) Lee, D. L.; Hodges, R. S. Biopolymers 2003, 71, 28-48.
(9) For reviews, see: (a) Lambert, J . N.; Mitchell, J . P.; Roberts, K.
D. J . Chem. Soc., Perkin Trans. 1 2001, 471-484. (b) Humphrey, J .
M.; Chamberlin, A. R. Chem. Rev. 1997, 97, 2243-2266. (c) Wipf, P.
Chem. Rev. 1995, 95, 2115-2134.
(5) Prenner, E. J .; Lewis, R. N. A. H.; McElhaney, R. N. Biochim.
Biophys. Acta 1999, 1462, 201-221.
(6) (a) Hancock, R. E. W. Lancet 1997, 349, 418-422. (b) Konde-
jewski, L. H.; J elokhani-Niaraki, M.; Farmer, S. W.; Bruce, L.; Kay,
C. M.; Sykes, B. D.; Hancock, R. E. W.; Hodges, R. S. J . Biol. Chem.
1999, 274, 13181-13192.
10.1021/jo035712x CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/17/2004
J . Org. Chem. 2004, 69, 2681-2685
2681

Bu et al.
SCHEME 1. Syn th esis of th e Lin ea r Biosyn th etic P r ecu r sor of Gr a m icid in S (2)a
a
Reagents and conditions: (a) PyBOP, DIPEA, CHCl3, Fmoc-Leu-OH, -20 °C, 8 h, repeat once; (b) standard Fmoc solid-phase peptide
synthesis; deprotection: 20% piperidine in DMF; coupling: Fmoc-amino acid (Boc-DPhe-OH for the last residue), DIC, HOBt, 2 h; (c)
ICH2CN, NMP, DIPEA, 24 h; (d) NAC-SH/THF; (e) CF3COOH/phenol/i-Pr3SiH/H2O ) 88:5:5:2, 1 h.
amount of natural product analogues due to the poor
tendency of the linear peptide precursors to cyclize.
Recently, we showed that the isolated thioesterase
domain in gramicidin synthase can be used to cyclize
linear precursors on solid support as a general method
for combinatorial synthesis of gramicidin S analogues.10
However, this method is limited by the enzyme’s sub-
strate specificity, and the enzymatic cyclization of the
linear precursors is incomplete. For the optimization of
gramicidin S therapeutic index, a less restrictive and
more convenient synthetic method that is adaptable to
combinatorial synthesis is highly desirable.
It has been implicated in the literature that the linear
biosynthetic precursor of gramicidin S adopts a pre-
organized conformation that may be utilized for develop-
F IGURE 2. Circular dichroism spectra of gramicidin S in 50
ment of novel facile synthesis of the natural product and
mM Tris‚HCl buffer (pH 7.0, red line) and its linear precursor
analogues.10,11 In our investigation of enzymatic cycliza-
tion of gramicidin S precursor, we found that the carboxy-
activated on-resin precursor and its analogues cyclize
in THF (blue line). The inset is the projected structure of the
linear precursor (2).
spontaneously into head-to-tail products in ammonia
synthetic purposes. Here, we explore the probable con-
solution, without protection of active side-chain amine
formational propensity of the linear precursors to develop
groups of the two ornithine residues.10 The absence of
a convenient synthetic method for the natural product
the hydrolytic or aminolytic or side chain-to-tail cycliza-
and its analogues that is readily adaptable to combina-
tion products indicates that the linear precursor adopts
torial synthesis.
a predisposed conformation highly favorably for the head-
to-tail cyclization reaction. Similar conformational pro-
pensity is also implicated in another report of highly
Resu lts a n d Discu ssion s
specific and high-yielding synthesis of the natural peptide
product.11 In addition, the linear biosynthetic precursor
of tyrocidine A, a distinct but similar decapeptide anti-
biotic of gramicidin S, has also been demonstrated to
possess a similar favorable conformation that has been
utilized to develop a facile synthesis of cyclic peptides
on the basis of the structural framework of the natural
product.12 Since both tyrocidine A and gramicidin S form
antiparallel â-pleated sheet structures,4,13 we reasoned
that the precursor for gramicidin S most likely has a
similar conformational preference that can be used for
First, we examined whether the linear biosynthetic
precursor of gramicidin S exists in an antiparallel â-sheet
structure as the cyclic natural product. The N-acetylcys-
teaminyl (NAC) thioester of the linear precursor (2) was
synthesized with 4-sulfamylbutyryl AM resin (Scheme
1) and subjected to circular dichroism (CD) analysis. As
shown in Figure 2, analysis of the CD spectrum indicated
as expected that the precursor possesses a typical â-sheet
structure in THF attested by the curvature at around
220 nm. This conformation closely resembles the anti-
parallel â-pleated sheet structure of the aqueous solution
(10) Wu, X.; Bu, X.; Wong, K. M.; Yan, W.; Guo, Z. Org. Lett. 2003,
5, 1749-1752.
(11) Wishart, D. S.; Kondjewski, L. H.; Semchuk, P. D.; Sykes, B.
D.; Hodges, R. S. Lett. Pept. Sci. 1996, 3, 53-60.
(12) (a) Bu, X.; Wu, X.; Xie, G.; Guo, Z. Org. Lett. 2002, 4, 2893-
2895. (b) Qin, C.; Bu, X.; Wu, X.; Guo, Z. J . Comb. Chem. 2003, 5,
353-355.
(13) (a) Kuo, M.-C.; Gibbons, W. A. Biochemistry 1979, 18, 5855-
5867. (b) Kuo, M.-C.; Gibbons, W. A. J . Biol. Chem. 1979, 254, 6278-
6287. (c) Kuo, M.-C.; Drakenberg, T.; Gibbons, W. A. J . Am. Chem.
Soc. 1980, 102, 520-524. (d) Kuo, M.-C.; Gibbons, W. A. Biophys. J .
1980, 32, 807-836. (e) Zhou, N.; Mascagni, P.; Gibbons, W. A.; Niccolai,
N.; Rossi, C.; Wyssbrod, H. J . Chem. Soc., Perkin Trans. 2 1985, 4,
581-587.
2682 J . Org. Chem., Vol. 69, No. 8, 2004

Synthesis of Gramicidin S and Its Analogues
4-sulfamylbutyryl AM resin (0.8 mmol/g) for each com-
pound, the alanine-substituted products were obtained
in good yields and characterized without purification as
summarized in Table 1.
As shown in Figure 3, no product was formed when
Pro-2 in the linear precursor was substituted. For the
other nine substitutions, HPLC chromatograms showed
that the unpurified cleavage products predominantly
contained one peak with a purity of >80%. It is interest-
ing to note the identical retention time (tR) for the
cleavage products of precursor pairs 4b/4g, 4d /4i, 4e/4j,
and 4f/4k , which strongly suggests that the correspond-
ing cleavage products are the expected head-to-tail cy-
clization product since the symmetric tR distribution is
consistent with the C2 symmetry of the gramicidin S
template. Moreover, mass spectroscopic data showed that
each cleavage reaction gave rise to a dominant molecular
ion consistent with the cyclic product of the commensu-
rate linear peptide precursor, free from any linear
hydrolytic or aminolytic products. Furthermore, after
HPLC purification, the alanine-substituted analogues
were subjected to 1H NMR structural determination and
the cleavage products of the precursor pairs 4b/4g, 4d /
4i, 4e/4j, and 4f/4k were indeed found to be identical. In
addition, spectrum comparison revealed that the alanine
analogues differed from the wild-type gramicidin S only
for the substituted residue, indicating that the cyclic
products possessed a ring structure closely resembling
the antiparallel natural product and that the cyclization
of the linear precursors was dominantly head-to-tail with
little interference from the unprotected amino groups.
These results showed that the individual side chain of
the composing amino acid residues of the gramicidin S
scaffold (except for that of Pro-2) has little effect on the
formation of the pre-organized favorable conformation
and thus does not affect the strong tendency of the linear
precursors to self-cyclize.
It is interesting to note that Pro-2 is irreplaceable
whereas its equivalent at a symmetrical position in the
gramicidin S structure, proline-7, can be substituted with
an alanine. This difference can be understood on the basis
of the projected â-sheet conformation for the linear
precursor (2) in which Pro-2 facilitates the â-turn forma-
tion, a function well-known for proline residues,15 to bring
the two reactive groups in close proximity and thus is
indispensable in the ring closure. On the other hand,
although Pro-7 is also involved in the â-sheet conforma-
tion (2), its turn-facilitating capability is much less
significant due to the strong driving force of the four
inter-strand hydrogen bonds. The insignificance of this
proline residue in the â-turn formation is also evidenced
by comparing to the similar conformation-assisted syn-
thesis of tyrocidine A analogues in which no proline
residue at the corresponding position is required in the
sheet formation of the linear precursors.12b However, the
only proline involved in the ring closure of tyrocidine A
precursor, which occupies an equivalent position as the
Pro-2 in gramicidin S precursor, is replaceable by an
alanine.12b This discrepancy might result from subtle
difference in the orientation of the N- and C-termini of
F IGURE 3. HPLC chromatograms of the unpurified cleavage
products (1a -k ) of the linear precursors 4a -k . The analyses
were performed with a Waters 600E system with a reversed-
phase semipreparative XTerra RP18 column, 7 µm, 7.8 × 300
mm. Separation conditions were as follows: 2.0 mL min-1 flow
rate, a linear gradient of 80 to 20% A in 25 min, 20 to 0% A in
another 10 min, washed with 100% B for 10 min, and then
calibrated at 80% A for 15 min. Solution A was 0.1% TFA in
double-deionized H2O, and solution B was 0.1% TFA in
acetonitrile.
structure of gramicidin S, which is maintained by four
interstrand hydrogen bonds,4 shown by the near-identical
shape of CD spectra for the precursor (2) and the natural
product.
In an attempt to determine whether the â-sheet
conformation predisposed the linear precursor of grami-
cidin S to self-cyclize, 3a was activated by cyanometh-
ylation, deprotected to remove Boc and t-Bu groups, and
treated in a basic solution to deprotonate the N-terminal
ammonium group for cyclization (Scheme 2), similar to
the tyrocidine A synthesis.12b After the reaction, the
product was precipitated with cold ether, dried to obtain
a white solid with a crude yield of 25%, and dissolved in
DMSO for determination of purity and structure. HPLC
analysis showed that only a product (purity >87%, Figure
3) with the same retention time as the authentic grami-
cidin S was present in the unpurified product mixture.
1
Mass spectroscopic and H NMR data indicated that the
only product was the expected head-to-tail cyclization
product. Moreover, the purified product showed consis-
tent antibacterial activity with the authentic gramicidin
S.2a Indeed, these results showed that the linear precur-
sor specifically cyclized head-to-tail without interference
from the δ-NH2 of the two ornithine residues, most likely
due to the predisposed product-like conformation.
To explore the usefulness of this conformation-depend-
ent method for gramicidin analogue synthesis, we deter-
mined the influence of the side chains of the constituent
amino acids of the gramicidin S template on the cycliza-
tion efficiency by an “alanine-scanning” method used for
the tyrocidine A scaffold.12b The component amino acids
of the gramicidin S precursor were sequentially substi-
tuted by alanine. Configuration of individual amino acids
was maintained in the side chain swapping since it has
been known to affect the conformation of the linear
peptides.12a The substituted linear precursors were syn-
thesized and cyclized in parallel using Irori’s AccuTag-
100 Combinatorial Chemistry System,14 according to the
method shown in Scheme 2. Starting from 30 mg of
(14) Nicolaou, K. C.; Xiao, X.-Y.; Parandoosh, Z.; Senyei, A.; Nova,
M. P. Angew. Chem., Int. Ed. Engl. 1995, 34, 2289-2291.
(15) Hutchinson, E. G.; Thornton, J . M. Protein Sci. 1994, 3, 2207-
2216.
J . Org. Chem, Vol. 69, No. 8, 2004 2683

Bu et al.
SCHEME 2. Solid -P h a se Syn th esis of Gr a m icid in S a n d Its Ala n in e-Su bstitu ted An a logu esa
a
Reagents and conditions: (a) standard Fmoc solid-phase peptide synthesis, see Scheme 1; (b) ICH2CN/DIPEA/NMP, 24 h; (c) CF3COOH/
phenol/i-Pr3SiH/H2O ) 88:5:5:2, 1 h; (d) DIPEA/THF.
TABLE 1. Ch a r a cter iza tion of th e Cycliza tion
structure without affecting the conformational propensity
in a previous investigation.10
Rea ction s of th e Lin ea r Deca p ep tid e P r ecu r sor s 4a -h
a
substituted
residue
tR
calcd
yieldb purityc
product
(min) mass [M + H]+ (%)
(%)
Con clu sion s
1a (wt)
1b
1c
1d
1e
1f
1g
1h
1i
none
D-Phe1
Pro2
29.7 1140.7
26.5 1064.7
1141.7
1065.8
25
53
0
87
92
In summary, we have demonstrated that the predis-
posed conformation of linear gramicidin S precursor and
its analogues can be utilized to design a simple and
efficient method for analogue synthesis. Apparently, this
facile synthetic method can be readily adapted to com-
binatorial synthesis. In comparison with the reported
chemoenzymatic cyclization method,10 this chemical ap-
proach allows more flexibility in the precursor sequence
and thus a much larger structural space to be explored.
This method should provide a convenient access to a large
amount of gramicidin S analogues to discover new leads
for containment of microbial resistance.
Val3
31.1 1113.4
27.7 1069.7
26.6 1098.6
26.6 1064.7
28.0 1114.7
31.1 1113.4
27.7 1069.7
26.6 1098.6
1113.9
1070.5
1099.9
1065.8
1116.0
1114.7
1070.8
1099.6
29
32
52
49
45
34
32
34
82
85
89
82
81
83
82
85
Orn4
Leu5
D-Phe6
Pro7
Val8
Orn9
Leu10
1j
1k
a
Retention time of the major product peak from HPLC chro-
matograms shown in Figure 2. b Yield of crude products after ether
precipitation. c Ratio of the area of the cyclic product peak to the
total integration over 15-35 min on the HPLC chromatograms in
Figure 2.
Exp er im en ta l Section
Gen er a l Meth od s. 1H NMR spectra were recorded on a
spectrometer and calibrated with the residual undeuterated
solvent or TMS. Mass spectra were recorded on a triple-stage
quadruple mass spectrometer using fast atom bombardment
(FAB) ionization mode. The products were unpurified for the
mass spectroscopy, whereas the NMR spectra were recorded
with samples purified by HPLC.
Solid -P h a se Syn th esis of th e Lin ea r Biosyn th etic
P r ecu r sor of Gr a m icid in S (2). Synthesis of the linear
biosynthetic precursor of gramicidin S (2) was accomplished
according to Scheme 1. The synthesis started from 100 mg of
4-sulfamyl butyryl AM resin (0.8 mmol/g) and strictly followed
a published procedure up to the cyanomethylation.12b Subse-
quently, the activated resin was suspended in 4 mL of 10%
N-acetylcysteamine in THF and stirred under N2 atmosphere
at room temperature for 20 h. Most THF in the resulting
solution was removed by aspiration, and the product was
precipitated by cold ether. The Boc protecting groups were
removed by dissolving the linear precursor in 2 mL of the
deblocking agent (CF3COOH/phenol/i-Pr3SiH/H2O ) 88:5:5:2)
and stirring for 3 h at room temperature. The crude product
was then precipitated by cold ether, dissolved in 0.3 mL of
methanol, and purified by HPLC to obtain 21.6 mg of white
the predisposed â-sheet conformations of tyrocidine A and
gramicidin S precursors. In tyrocidine A precursor, the
termini might be brought closer together for the reactive
terminal groups to effectively react with each other
without the help of the proline, whereas a proline is
needed for the gramicidin S precursor to facilitate the
turn formation to bring the two more separated terminal
groups close enough together to achieve the ring closure.
Despite the indispensability of Pro-2, the readily
exchangeability of the other nine residues of the grami-
cidin S scaffold still allows a very large structural space
to be explored. Although the exact scope of the allowable
variation in the gramicidin S scaffold remains to be
defined, the pre-organized conformation of the linear
precursors should be tolerant to variations in the side
chains since the favorable conformation is maintained
by four strong backbone hydrogen bonds. Indeed, simul-
taneous variations at three composing amino acid resi-
dues have been successfully incorporated into the cyclic
2684 J . Org. Chem., Vol. 69, No. 8, 2004

Synthesis of Gramicidin S and Its Analogues
solid (21.4% yield). FAB-MS: m/z 1260.9 ([M + H]+, calcd m/z
Ala 4-gr a m icid in S (1e ) 1j). FAB-MS: m/z 1070.5 ([M +
H]+, calcd mass 1069.7). 1H NMR (500 MHz, DMSO-d6, 25 °C,
TMS): δ 9.05 (s, 1H), 8.98 (s, 1H), 8.75 (d, J ) 10 Hz, 1H),
8.69 (d, J ) 10 Hz, 1H), 8.47 (d, J ) 10 Hz, 1H), 8.32 (d, J )
10 Hz, 1H), 7.82 (s, 14H), 7.39-7.33 (m, 13H), 7.26 (d, J ) 10
Hz, 1H), 5.00-4.98 (m, 1H), 4.78-4.73 (m, 1H), 4.68-4.55 (m,
3H), 4.49-4.40 (m, 4H), 3.08-2.83 (m, 6H), 2.18-2.15 (m, 3H),
2.04 (s, 2H), 1.91-1.29 (m, 20H), 0.92-0.87 (m, 22H).
Ala 5-gr a m icid in S (1f ) 1k ). FAB-MS: m/z 1099.9 ([M +
H]+, calcd mass 1098.6). 1H NMR (500 MHz, DMSO-d6, 25 °C,
TMS): δ 9.14 (s, 1H), 9.05 (s, 1H), 8.81 (d, J ) 10 Hz, 2H),
8.45 (m, 2H), 7.79 (s, 8H), 7.41-7.08 (m, 18H), 4.82 (s, 2H),
4.68 (s, 2H), 4.51-4.33 (m, 8H), 3.07-2.86 (m, 6H), 2.21-2.06
(m, 2H), 1.86-1.34 (m, 18H), 1.38-1.29 (m, 2H), 0.97-0.89
(m, 14H).
1
1259.8). H NMR (500 MHz, DMSO-d6, 25 °C, TMS): δ 8.70
(t, J ) 9.0 Hz, 1H), 8.59 (d, J ) 7.5 Hz, 1H), 8.54 (d, J ) 8.0
Hz, 1H), 8.34 (s, 1H), 8.24 (t, J ) 9.0 Hz, 1H), 8.17 (m, 1H),
8.09 (m, 1H), 7.97-8.03 (m, 2H), 7.77-7.84 (br, 5H), 7.23-
7.45 (m, 10H), 5.07 (d, J ) 8.5 Hz, 1H), 4.84 (m, 1H), 4.17-
4.53 (m, 12H), 3.66 (m, 1H), 3.58 (m, 1H), 3.46 (m, 1H), 3.25
(m, 2H), 3.15 (m, 1H), 3.06 (m, 1H), 2.95 (m, 2H), 2.85 (m,
5H), 2.29 (m, 1H), 2.18 (m, 1H), 2.06 (m, 2H), 1.89 (s, 3H),
1.43-1.88 (m, 16H), 1.24 (m, 1H), 1.11 (m, 1H), 0.72-1.03 (m,
24H).
Solid -P h a se Syn th esis, An a lysis, a n d P u r ifica tion of
Cyclic Gr a m icid in
S a n d Its An a logu es. Solid-phase
synthesis of the linear peptide precursors and their cyclization
into gramicidin S and analogues strictly followed a published
procedure.12b Reactions were carried out in MicroKan reactors
with 30 mg of 4-sulfamylbutyryl AM resin (0.8 mmol/g) for
each compound. The crude product was precipitated with cold
ether, dissolved in 100-300 µL of methanol, and subjected to
HPLC analysis which was performed with a reversed-phase
semipreparative XTerra RP18 column, 7 µm, 7.8 × 300 mm.
Separation conditions were: 2.0 mL min-1 flow rate, a linear
gradient of 80% to 20% A in 25 min, 20 to 0%A in another 10
min, washed with 100% B for 10 min, and then calibrated at
80% A for 15 min. Solution A was 0.1% TFA in double-
deionized H2O and solution B was 0.1% TFA in acetonitrile.
For 1H NMR spectroscopic analysis, crude products were
purified by HPLC using above gradient and obtained as white
solid after evaporating solvent from the collected product
fractions.
Ala 6-gr a m icid in S (1g ) 1b). FAB-MS: m/z 1065.8 ([M +
H]+, calcd mass 1064.7).
Ala 7-gr a m icid in S (1h ). FAB-MS: m/z 1116.0 ([M + H]+,
calcd mass 1114.7). 1H NMR (500 MHz, DMSO-d6, 25 °C,
TMS): δ 9.16 (s, 1H), 8.99 (d, J ) 10 Hz, 1H), 8.92 (d, J ) 10
Hz, 1H), 8.79 (d, J ) 10 Hz, 1H), 8.75 (d, J ) 10 Hz, 1H), 8.42
(d, J ) 10 Hz, 1H), 8.38 (d, J ) 10 Hz, 1H), 7.78 (s, 6H), 7.61
(d, J ) 10 Hz, 1H), 7.41-7.31 (m, 14H), 4.86 (m, 2H), 4.69-
4.64 (m, 2H), 4.50-4.37 (m, 6H), 4.20-4.18 (m, 1H), 3.07-
2.83 (m, 8H), 2.22-2.18 (m, 2H), 2.06 (m, 1H), 1.86-1.34 (m,
20H), 1.15-1.14 (m, 4H), 0.98-0.90 (m, 16H).
Ala 8-gr a m icid in S (1i ) 1d ). FAB-MS: m/z 1114.7 ([M +
H]+, calcd mass 1113.4).
Ala 9-gr a m icid in S (1j ) 1e). FAB-MS: m/z 1070.8 ([M +
H]+, calcd mass 1069.7).
Wild -typ e Gr a m icid in S (1a ). FAB-MS: m/z 1141.4 ([M
1
+ H]+, calcd mass 1140.7). H NMR (500 MHz, DMSO-d6, 25
Ala 10-gr a m icid in S (1k ) 1f). FAB-MS: m/z 1099.6 ([M
°C, TMS): δ 9.06 (s, 1H, NH), 8.70 (d, J ) 9.2 Hz, 1H, NH),
8.33 (d, J ) 9.2 Hz, 1H, NH), 7.85 (d, J ) 7.6 Hz, 2H, NH),
7.74-7.68 (m, 8H), 7.4-7.2 (m, 18H), 7.1 (s, 1H), 6.98 (s, 1H),
4.74-4.04 (m, 10H), 2.95-2.66 (m, 10H), 2.07-2.03 (m, 2H),
1.91-1.19 (m, 24H), 0.81-0.73 (m, 24H).
+ H]+, calcd mass 1098.6).
Ack n ow led gm en t. This work was supported in part
by the Innovation and Technology Fund (ITS/119/00)
and a grant from the Research Grants Council (CERG
HKUST601503) of the Government of the Hong Kong
Special Administrative Region, China. We also thank
the Department of Chemistry at the HKUST for provid-
ing postgraduate studentships to X.W. and N.L.J .N.
Ala 1-gr a m icid in S (1b ) 1g). FAB-MS: m/z 1065.8 ([M +
H]+, calcd mass 1064.7). 1H NMR (500 MHz, DMSO-d6, 25 °C,
TMS): δ 9.17 (s, 1H), 8.97 (s, 1H), 8.85-8.80 (m, 2H), 8.43 (d,
J ) 10 Hz, 2H), 7.79-7.73 (m, 6H), 7.49-7.35 (m, 8H), 4.88
(m, 2H), 4.73-4.64 (m, 2H), 4.52-4.41 (m, 6H), 4.03-3.99 (m,
1H), 3.12-3.09 (m, 1H), 2.96-2.81 (m, 6H), 2.27-2.07 (m, 6H),
1.88-1.86 (m, 4H), 1.72-1.31 (m, 24H), 0.98-0.87 (m, 16H).
Ala 3-gr a m icid in S (1d ) 1i). FAB-MS: m/z 1113.9 ([M +
H]+, calcd mass 1113.4). 1H NMR (500 MHz, DMSO-d6, 25 °C,
TMS): δ 9.07 (s, 2H), 8.72 (d, J ) 10 Hz, 2H), 8.47 (d, J ) 10
Hz, 2H), 8.18 (d, J ) 10 Hz, 2H), 7.76 (s, 2H), 7.55-7.08 (m,
16H), 5.01-4.98 (m, 1H), 4.79 (m, 1H), 4.68-4.38 (m, 8H),
3.07-2.93 (m, 8H), 2.17-2.00 (m, 4H), 1.87 (m, 1H), 1.74-
1.30 (m, 24H), 0.92-0.89 (m, 16H).
Su p p or tin g In for m a tion Ava ila ble: FAB-MS and 1H
NMR spectra of the cyclic peptide products 1a -k . This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O035712X
J . Org. Chem, Vol. 69, No. 8, 2004 2685