Synthesis of novel unnatural amino acid as a building block and its incorporation into an antimicrobial peptide

Article abstract of DOI:10.1016/S0968-0896(99)00247-3

Considering the biological mechanism and in vivo stability of antimicrobial peptides, we designed and synthesized novel unnatural amino acids with more positively charged and bulky side chain group than lysine residue. The unusual amino acids, which were synthesized by either solution phase or solid phase, were incorporated into an antimicrobial peptide. Its effect on the stability, activity, and the structure of the peptide was studied to evaluate the potential of these novel unnatural amino acids as a building block for antimicrobial peptides. The incorporation of this unusual amino acid increased the resistance of the peptide against serum protease more than three times without a decrease in the activity. Circular dichroism spectra of the peptides indicated that all novel unnatural amino acids must have lower α helical forming propensities than lysine. Our results indicated that the unnatural amino acids synthesized in this study could be used not only as a novel building block for combinatorial libraries of antimicrobial peptides, but also for structure-activity relationship studies about antimicrobial peptides. Copyright (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00247-3

Bioorganic & Medicinal Chemistry 7 (1999) 2985±2990
Synthesis of Novel Unnatural Amino Acid as a Building Block and
Its Incorporation into an Antimicrobial Peptide
Jong Eun Oh and Keun Hyeung Lee*
Protein Chemistry Laboratory, Mogam Biotechnology Research Institute, 341 Pojung-Ri, Koosung-Myun, Yongin-City, Kyunggi-Do,
449-910, South Korea
Received 21 June 1999; accepted 14 August 1999
AbstractÐConsidering the biological mechanism and in vivo stability of antimicrobial peptides, we designed and synthesized novel
unnatural amino acids with more positively charged and bulky side chain group than lysine residue. The unusual amino acids,
which were synthesized by either solution phase or solid phase, were incorporated into an antimicrobial peptide. Its e€ect on the
stability, activity, and the structure of the peptide was studied to evaluate the potential of these novel unnatural amino acids as a
building block for antimicrobial peptides. The incorporation of this unusual amino acid increased the resistance of the peptide
against serum protease more than three times without a decrease in the activity. Circular dichroism spectra of the peptides indicated
that all novel unnatural amino acids must have lower a helical forming propensities than lysine. Our results indicated that the
unnatural amino acids synthesized in this study could be used not only as a novel building block for combinatorial libraries of
antimicrobial peptides, but also for structure±activity relationship studies about antimicrobial peptides. # 1999 Elsevier Science
Ltd. All rights reserved.
Introduction
The recent emergence of multidrug-resistant bacteria
has stimulated the development of novel antibacterial
molecules with unexploited mechanisms of action. A
4
Recent advances of biological techniques made it possi-
ble to identify novel biological active compounds from
large number of defense peptides produced in eukar-
yotic systems have been isolated and their functions
characterized.5 Some of them have selectivity between
prokaryotic and eukaryotic cells and a broad range of
the activity against bacteria and fungi. Although the
mode of the action of these peptides is not fully under-
stood, it is suggested that the peptides ful®l their biolo-
gical function by enhancing the permeability of lipid
membranes of pathogenic cells.8 In the ®rst step, the
positively charged peptides bind to the negatively
charged lipid membranes of the pathogen mainly by
charge interactions and then they adopt mostly a helical
structure or b sheet structure. In the next step, the pep-
tides increase the permeability of the lipid membranes
either by ion channel formation 1 or by the perturba-
tion of the structure of the bilayer,12 resulting in the
death of target cells. As the peptides have a di€erent
mode of action from classical antibiotics, many combi-
natorial libraries screening for peptides active against
1
various natural sources. For developing the novel
±7
compounds as therapeutic agents, rational design and
synthesis of novel candidate molecules, originated from
the natural form, were required. This process, however,
should overcome the following obstacles such as a
time-consuming optimization process, a nonlinear rela-
tionship between structure and activity, and a lack of
information on the structure of compounds in target
sites. As an alternative method to overcome these lim-
itations, combinatorial libraries have been developed
and proved to be an ecient method to screen novel
compounds as well as to optimize activity. The recent
increase of the use of the combinatorial library techni-
que has demanded the novel building block because
increasing the number of building blocks resulted in the
exponential increase of combination mixtures and the
increase of probability for identifying novel compounds.
±10
2
,3
1
bacteria and fungi have been used for developing novel
3±15
therapeutic agents.1
activity and rapid enzymatic degradation of anti-
microbial peptides, compared with classical antibiotics,
are still regarded as the main limitations.
However, the relatively low
Key words: Amino acids and derivatives; antimicrobial compounds;
depsipeptides; solid phase synthesis.
*
2
Corresponding author. Tel.: +82-331-260-9814; fax: +82-0331-
62-6622; e-mail: lkh@kgcc.co.kr
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00247-3

2
986
J. E. Oh, K. H. Lee / Bioorg. Med. Chem. 7 (1999) 2985±2990
In the present study, to improve activity and in vivo
stability of antimicrobial peptides, we designed and
synthesized novel unnatural amino acids that had more
positively charged and bulky side chain group than
that of lysine residue on the basis of the following
facts. The increase of the net positive charge of an
antimicrobial peptide enhanced its binding to lipid
membranes, resulting in the improvement of its activ-
In the second method, the peptide containing unnatural
amino acids was synthesized in solid phase peptide
synthesis as described in Scheme 2. Classical reductive
alkylation of aldehyde with the amino group of the
peptide has been used for the modi®cation of the side
2
1
chain in Lys of the peptide attached to the resin. The
reductive alkylation using Boc-glycinal and NaBH CN,
3
yielded a more bulky and positively charged unnatural
amino acid than a natural amino acid, Lys. When
Fmoc-Lys(Fmoc)-OH instead of Boc-Lys(Fmoc)-OH
was introduced at the N-terminal of the peptide
attached to the resin, an a-amino group as well as an e-
amino group of side chain in lysine was also modi®ed by
the same procedure. The completion of the reaction was
monitored by a ninhydrin test.22 All synthesized pep-
tides containing unnatural amino acid were puri®ed by
preparative RP-HPLC and characterized by analytical
HPLC and mass spectroscopy.
1
6,17
ity
and the peptide containing the unnatural amino
acid with the bulky side chain group should have more
resistance against enzymatic digestion than the peptide
consisting of natural amino acids. We incorporated
the novel unnatural amino acids into a well-char-
acterized antimicrobial peptide, MP,18 and studied the
activity, stability in the presence of serum, and the
secondary structure of the peptide to evaluate their
e€ectiveness as novel building blocks for the combi-
natorial libraries to screen for peptides active against
microorganisms.
The ®rst method can provide a variety of positively
charged unnatural amino acids with various side chains
by substitution reaction of intermediate 4 with various
nucleophiles. Primary or secondary amine was success-
fully introduced to the side chain of the amino acid by
substitution reaction. For example, tert-butyl N-(2-ami-
noethyl) carbamate, tert-butyl N-(3-aminopropyl) car-
bamate, and 4-piperidonopiperidine23 were reacted with
intermediate 4 yielding the corresponding unnatural
positively charged amino acids (data not shown). The
second method directly modi®ed the side chain in Lys of
the peptide in SPPS and, therefore, it was not necessary
to synthesize and purify the unnatural amino acid. In
addition, there is no risk of racemization. Also, the sec-
ond method can be easily applied to other antimicrobial
peptides containing Lys residue, in SPPS.
Chemistry
Antimicrobial peptides containing unnatural amino
acids were synthesized by two di€erent methods. In the
rst method, we synthesized unnatural amino acid(s) in
solution phase and then incorporated it into the model
peptide in solid phase peptide synthesis (SPPS). The
unnatural amino acid 1 was prepared from l-serine 2 as
described in Scheme 1. The amino group of l-serine 2
was protected with Boc group and then the carboxyl
group was converted to methyl ester 3. N-Boc serine
methyl ester 3 was converted to brominated compound
in 60% yield by treatment with CBr and PPh . The
4 3
treatment of activated intermediate 4 with Boc hydra-
zine and followed by hydrolysis provided protected
unnatural amino acid 1. No racemization was observed
in this step by using HPLC with a chiral column. The
incorporation of unnatural amino acid into the peptide
was performed by SPPS in the presence of PyBOP and
TEA in DMF.
®
1
9
20
4
Results and Discussion
The model peptide, MP consisting of eleven amino acid
residues, was reported to have potent antimicrobial
Scheme 1. Synthesis of unusual amino acid in solution phase. (a) (Boc)
2
O, NaOH, H
PPh , CH Cl , 1 h, 66%; (d) BocNHNH , TEA, CH Cl , 4 h, 57%; (e) LiOH, THF/MeOH/ H O (3/1/1), 3 h, 37%.
2
O, 24 h, 95%; (b) DCC, HOBt, MeOH, 5 h, 72%; (c) CBr
4
,
3 2 2 2 2 2 2

J. E. Oh, K. H. Lee / Bioorg. Med. Chem. 7 (1999) 2985±2990
2987
Scheme 2. Synthesis of MPU analogues in solid phase.
activity without hemolytic activity.18 We synthesized
several MP analogues containing the unnatural amino
acid (Scheme 3) and investigated the unnatural amino
acid e€ect on the activity, stability, and secondary
structure of the peptide. To investigate the e€ect of the
unnatural amino acid on the stability, we measured the
half-life of the peptides in the presence of serum. Our
previous study indicated that exoprotease in the serum
played a major role in the degradation of the anti-
microbial peptide with a primary structure similar to
that of MP. In the present study, we incorporated the
unusual amino acid at the N-terminal and simulta-
neously C-terminal amino acid was replaced by d-
amino acid for easy monitoring the half-life in the
presence of serum. As shown in Figure 1, MPU had a
very short half-life (6 min), whereas MPU2 (21 min) and
peptide containing unnatural amino acid in membrane-
mimetic environment to elucidate the real net charge
e€ect on the activity. As shown in Figure 2, CD spectra
of MPU 1-3, measured in the presence of 50% TFE,
indicated that MPU had 54% a helicity whereas MPU1,
MPU2, and MPU3 had 29, 42, and 33% a helicity,
respectively. The secondary structure change of the
peptide containing the unnatural amino acid indicated
that the introduction of novel unnatural amino acid
resulted in the decrease of a helicity. As the other minor
factors for the activity such as hydrophobicity and
amphiphilicity were regarded to be similar, we could
®nd out the relationship between the activity and
structural parameters in terms of net positive charge
and a helical structure. All MP analogues (MPU1-3)
employed in this study retained antimicrobial activities
in spite of the decrease of a helicity, which suggested
that the increase of net positive charge must compensate
for the decrease of a helicity, resulting in the retention
of the activity. CD spectra also revealed that the novel
unnatural amino acid had a lower a helical forming
propensity than lysine. It is due to the fact that the
novel unnatural amino acid has a di€erent number and
location of amino groups, which can participate in
hydrogen bonding, and a di€erent bulkiness in the side
chain from those of lysine residue.
2
4
3
(24 min) had a much longer half-life than MPU. This
result suggests that the bulky side chain of the unnatural
amino acid must play a major role in the increase of the
resistance against exoprotease in the serum.
As shown in Table 1, the incorporation of unnatural
amino acids resulted in the increase of the net positive
charge of the peptide from 1 to 3. This incorporation,
however, did not signi®cantly a€ect antimicrobial
activity. As many structure±activity relationship (SAR)
studies indicated that net positive charge and a helical
structure played a crucial role in antimicrobial activ-
A positively charged amino acid, for example Lys resi-
due, had been added to the N- or C-terminal of
1
6,17
ity,
we investigated the secondary structure of each

2
988
J. E. Oh, K. H. Lee / Bioorg. Med. Chem. 7 (1999) 2985±2990
Figure 1. Stabilities of MPU and its analogues in the presence of
serum half-life was measured in the presence of 25% mouse serum (v/
v). MPU (!), MPU2 (*), MPU3 (*).
Figure 2. CD spectra of MPU and its analogues in 50% TFE (v/v).
CD spectra were measured at the concentration of 100 mg/mL sample
in 10 mM sodium phosphate bu€er (pH 7.4) including 50% TFE (v/v).
charge on the antimicrobial activity. In the present
study, the single amino acid replacement by the unna-
tural amino acid resulted in the increase of net positive
charge (1±3) of the peptide with little change of size and
angle subtended by the charged helix face. This result
suggests that the unnatural amino acids described here
can be useful to study the relationship between the
activity and net positive charge of the novel anti-
microbial peptides.
In conclusion, our result indicated that the unnatural
amino acid as a novel building block for antimicrobial
peptide could improve the activity against bacteria and
fungi as well as the stability of the peptide in serum. We
also suggest that this unusual amino acid can be used to
investigate the structural parameter of the antimicrobial
peptides on the activity.
Scheme 3. Structure of MPU and its analogues.
antimicrobial peptides and various amino acids in
antimicrobial peptides had been replaced by Lys resi-
due to study structural parameters for the activity
such as net positive charge, a helicity, and hydro-
8
±10,16,17
phobicity.
However, the augmentation of Lys
residue and the multiple amino acid replacement by Lys
residue must result in the change of several structural
parameters such as hydrophobicity, size, and angle
subtended by the charged helix face. Therefore, it is
dicult to evaluate the independent e€ect of net positive
Experimental
Methyl N-(tert-butyloxycarbonyl)-L-serinate (3). To a
stirred solution of N-(terbutyloxycarbonyl)-l-serine in
MeOH, which was prepared from l-serine 2 using

J. E. Oh, K. H. Lee / Bioorg. Med. Chem. 7 (1999) 2985±2990
Table 1. Antimicrobial activity of MPU and its analogues
2989
a
Name
Net charge
Minimum inhibitory concentration (mg/mL)
S. aureus
ATCC6538
MRSAb
Sr1550
M. luteus
ATCC 9341
P. aeruginosa
ATCC 9027
C. albicans
ATCC 36232
MPU
MPU1
MPU2
MPU3
MagaininII
+7
+8
+9
+10
4
8
5
5
40
25
Ð
25
25
4
5
3
3
50
6
8
13
9
4
10
4
4
50
>100
20
a
Average MIC values were calculated from three independent experiments performed in duplicate, which provided a standard deviation below
0%.
MRSA is methcillin resistant S. aureus.29
3
b
(
BOC) O, were added DCC and HOBt under N . After
2
h, the mixture was ®ltered. The combined ®ltrate was
analyzed using linear gradient of 0±70% B in 70 min.
Yield: 37%. Oil. TLC R 0.5 (CH Cl :MeOH, 4:1). [a]
2
5
concentrated and the residue puri®ed by FC. Yield:
7
f
2
2
25
ꢀ
1
(� 19 , MeOH) H NMR (CDCl ): 0.85±0.88 (m, 2H);
3
1
2%. oil. Rf 0.37 (hexane:EtOAc, 7:3). H NMR
1.46 (s, 18H), 1.93±2.06 (m, 1H); 5.85 (s, 1H); 6.20 (s,
1H); 7.07 (s, 1H); 8.4 (s, 1H).
(
CDCl ): 1.43 (s 9H), 2.45 (m, 1H), 3.76 (s, 3H), 3.80±
3
3.94 (m, 2H), 4.36 (m, 1H), 5.42 (m, 1H)
Synthesis of peptides
Methyl-3-bromo-N-(tert-butyloxycarbonyl)-L-alaninate
4). To a stirred solution of 3 (0.98, 4.5 mmol) and
CBr (1.84 g, 5.5 mmol) in CH Cl (10 mL) was added
(
Peptides were prepared by stepwise solid-phase synth-
esis on an Applied Biosystems model 431A automatic
peptide synthesizer. The peptide chain was assembled
on PAL resin with a Fmoc/tert-butyl strategy.2
4
2
2
ꢀ
portionwise PPh (1.75 g, 6.6 mmol) at 0 C. The mixture
3
5
was stirred for 20±30 min, and then the solvent was
evaporated. The residue was triturated with Et O
2
(
20 mL), the mixture ®ltered, and the ®lter cake washed
MPU1 was synthesized as follows; unusual amino acid
(1), which was synthesized in solution phase, was incor-
porated into the peptide, MPU by the coupling of unu-
sual amino acid (1) with free amino groups of resin
bound peptides, in the presence of benzotriazole-1-yl-
with Et O (3Â10 mL). The combined ®ltrate was con-
2
centrated and the residue puri®ed by FC. Yield: 66%.
White solid. R 0.2 (hexane:EtOAc, 9:1). Mp 50±52 C
ꢀ
H NMR (CDCl ): 1.42 (s, 9H); 3.65±4.93 (m, 5H);
f
1
3
4
.70±4.80 (m, 1H); 5.40 (m, 1H).
oxytris(pyrrolidino)phosphonium hexa¯uorophosphate
ꢀ
(
PyBOP) and triethylamine (TEA) for 5 h at 25 C. The
0
Methyl-3-[2 -(tert-butyloxycarbonyl)-hydrazino]-L-alani-
nate (5). To a stirred solution of 4 (0.4 g, 1.5 mmol) in
CH Cl (10 mL) was added t-butyl carbazate (0.23 g,
1.8 mmol) and triethylamine (0.31 mL, 2.25 mmol).
After 4 h, the mixture was diluted with CH Cl (30 mL).
The organic layer was washed with 5% HCl and brine,
dried (Na SO ), and evaporated. The residue was pur-
i®ed by FC. Yield: 57%. oil. R 0.29 (hexane:Et O,
reaction was repeated until no color change was
observed in the ninhydrin test. MPU2 and MPU3 were
synthesized in solid phase by three steps. First, N-Boc-
Lys(Fmoc)-OH in MPU2 or N-Fmoc-Lys(Fmoc)-OH
in MPU3 was incorporated into the peptide. Second,
the Fmoc-group was selectively deprotected in the pre-
sence of 30% piperidine in N,N-dimethylformamide
(DMF). Third, the side chain of Lys was modi®ed by
reductive alkylation of tert-butoxycarbonyl (Boc)-Glyc-
inal with free amino group of resin bound peptides, in
2
2
2
2
2
4
f
2
ꢀ
1
9
5:5). [a]25 (� 11 , CHCl ) H NMR (CDCl ): 0.82±0.88
3
3
(
m, 2H); 1.25 (s, 9H), 1.49 (m, 10H); 3.83 (s, 3H); 5.729±
5.734 (m, 1H); 6.16 (s, 1H); 7.02 (s, 1H).
the presence of NaBH CN (1 M solution in tetra-
3
hydrofuran) in 1% acetic acid in DMF.
0
To a stirred solution of 5 in mixed solution (8 mL) of
3
-[2 -(tert-Butyloxycarbonyl)-hydrazino]-L-alanine (1).
Deprotection was achieved by treatment with a mixture
of tri¯uoroacetic acid (TFA):water:thioanisole (9:0.5:
0.5, v/v/v) at room temperature for 2±4 h. After ®ltra-
tion of the resin and washing with TFA, a gentle stream
of nitrogen was used to remove the excess TFA. The
crude peptide was triturated with diethyl ether chilled at
THF and MeOH (3:1) was added LiOH in H O (2 mL)
2
ꢀ
temperature for 2 h. THF and MeOH were removed by
at 0 C. After 1.5 h, the mixture was stirred at room
evaporation and the residue in H O was acidi®ed with
2
10% citric acid and extracted with EtOAc. The com-
bined organic layer was washed with H O and brine,
ꢀ
� 20 C and was centrifuged at 3000Âg for 10 min. Die-
2
dried (Na SO ), and concentrated. The compound 1
4
thyl ether was decanted and crude peptide was dried
under nitrogen. The peptide was puri®ed by high per-
formance liquid chromatography with a Phenomenex
C18 column (21.2Â250 mm; Phenomenex, Torrance,
CA, USA). The homogeneity of the peptides (>95%)
was checked by analytical HPLC with a Waters Delta
Pak C18 column (3.9Â150 mm; Waters, Milford, MA,
USA). Mass spectrometry on Platform II (Fisons
2
was dried on vacuum and puri®ed by FC. The extent of
racemization of the unusual amino acid (>95%) was
con®rmed by HPLC with Chirex (D) penicillamine col-
umn (10Â250 mm, phenomenex, Torrance, CA, USA).
The amino acid was eluted using solvent A consisting of
water and solvent B consisting of methanol and mon-
itored by absorbance at 214 nm. The amino acid was

2
990
J. E. Oh, K. H. Lee / Bioorg. Med. Chem. 7 (1999) 2985±2990
instruments, Manchester, United Kingdom) was used to
measure the mass of the puri®ed peptide. MPU (ES-
independent experiments performed in duplicate and all
pseudo-®rst plots were linear showing correlation coef-
®cient greater than 0.96.
+
MS:1376.43 [M+H] , calculated mass 1377.92), MPU1
+
(
MPU2 (ES-MS:1462.25 [M+H] , calculated mass
ES-MS:1349.14 [M+H] , calculated mass 1350.92),
+
+
1463.92), MPU3 (ES-MS:1504.92 [M+H] , calculated
mass 1505.92)
References and Notes
1
2
. Pirrung, M. C. Chemtracts-org. Chem. 1995, 8, 5.
. Terrett, N. K.; Gardner, M.; Gordon, D. W.; Kobylecki, R.
Antifungal and antibacterial assay
J.; Steele, J. Tetrahedron 1995, 51, 8135.
. Youngquisi, R. S.; Fuentes, G. R.; Lacey, M. P.; Keough,
T. J. Am. Chem. Soc. 1995, 117, 3900.
. Maloy, W. L.; Kari, P. U. Biopolymer 1995, 37, 105.
3
In vitro antimicrobial assays were performed by the
modi®ed checkboard microdilution method by follow-
ing the recommendation of the National Committee for
4
2
6
5. Steiner, H.; Hulmark, D.; Engstrom, A.; Bennich, H.;
Boman, H. G. Nature 1981, 292, 246.
Clinical Laboratory Standards. Antibiotic medium 3
ꢀ
(
M3; pH 7 at 25 C, Difco) was used as the antibacterial
6
. Kusuhara, T.; Nakajuma, Y.; Natsuyama, K.; Natori, S. J.
Biochem. 1990, 107, 514.
. Yamada, K.; Natori, S. J. Biochem. 1993, 291, 275.
assay media. Bacteria cells freshly grown on antibiotic
medium 3 agar plate were suspended in physiological
7
4
saline to 10 cells per 1 mL and used as the inoculum.
The test solution was added (100 mL per well) and seri-
ally diluted by twofold. After inoculation (100 mL per
8. Juvvadi, P.; Vummam, S.; Merri®eld, R. B. J. Am. Chem.
Soc. 1996, 118, 8989.
9. Matsuzaki, K.; Sugishit, K.; Fujii, N.; Miyajima, K. Bio-
chemistry 1995, 34, 3423.
10. Dathe, M.; Schumann, M.; Wieprecht, T.; Winkler, A.;
Beyermann, M.; Krause, E.; Matsuzaki, K.; Murase, O.;
Bienert, M. Biochemistry 1996, 35, 12612.
3
well, 5Â10 cells per 1 ml), plates were incubated at
ꢀ
3
7 C for 24 h and the absorbance was measured at
20 nm using an ELISA reader (Spectra, SLT, Salzburg,
6
Austria) to assess cell growth. Antifungal assay was
1
1
1. Sansom, M. S. P. Prog, Biophys. Mol. Biol. 1991, 55, 139.
2. Matsuzaki, K.; Murase, O.; Fujii, N.; Miyajima, K. Bio-
ꢀ
were incubated at 30 C for 24 h. Minimal inhibition con-
done in RPMI 1640 media (pH 7 at 25 C) and the plates
ꢀ
chemistry 1996, 35, 11361.
3. Blondelle, S. E.; Houghten, R. A. TIBTECH 1996, 14, 60.
centration (MIC) was de®ned as the lowest concentration
exhibiting no visible growth of the test organism. All
MICs were measured from three independent experi-
ments performed in duplicate.
1
14. Blondelle, S. E.; Takahashi, E.; Weber, P. A.; Houghten,
R. A. Antimicrob. Agents Chemother. 1994, 38, 2280.
15. Hong, S. Y.; Oh, J. E.; Kwon, M. Y.; Choi, M. J.; Lee, J.
H.; Lee, B. L.; Moon, H. M.; Lee, K. H. Antimicrob. Agent
Chemother. 1998, 42, 2534.
CD measurement
16. Kiyota, T.; Lee, S.; Sugihara, G. Biochemistry 1996, 35,
1
1
3196.
7. Dathe, M.; Wieprecht, T.; Nikolenko, H.; Handel, L.;
Circular dichroism (CD) spectra were recorded on a J-
715 spectropolarimeter (Jasco, Tokyo, Japan) using a
quartz cell of 1 mm path length, at wavelengths ranging
from 190 to 245 nm. CD spectra were obtained with a
Maloy, L. W.; MacDonald, D. L.; Beyermann, M.; Bienert,
M. FEBS Letter 1997, 403, 208.
18. Hong, S. Y.; Oh, J. E.; Lee, K. H. Antimicrob. Agent Che-
mother. 1999, 43, 1704.
0.5 nm bandwidth and a scan speed of 10 nm/min at
room temperature. Two scans were averaged to improve
the signal to noise ratio. CD spectra were expressed as
the mean residue ellipticity and the a helicity was cal-
19. Tarbell, D. S.; Yamamoto, Y.; Pope, B. M. Proc. Nat.
Acad. Sci. USA 1972, 69, 730.
2
0. Savithri, D.; Leumann, D.; Sche€old, R. Hel. Chim. Acta
1996, 79, 288.
1. Sasaki, Y.; Murphy, W. A.; Heiman, M. L.; Lance, V. A.;
27
culated from the mean residue ellipticity [y] at 222 nm.
2
Coy, D. H. J. Med. Chem. 1987, 30, 1162.
22. Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I.
Anal. Biochem. 1970, 34, 595.
23. N-(tert-butoxycarbonyl)-3-[4-piperodopiperidyl] alanine
+
ES/MS 356.22 [M+H] .
Half-lives measurement in the presence of serum
1 mL of 25% mouse serum/RPMI media (v/v) in 1.5 mL
ꢀ
Eppendorf tube was temperature-equilibrated at 37 C
for 15 min before adding 10 mL of peptide stock solution
24. Oh, J. E.; Hong, S. Y.; Lee, K. H. J. Peptide Res. 1999, 54,
129.
25. Fields, G. G.; Noble, R. L. Int. J. Pept. Protein Res. 1990,
(
10 mg/mL) to make the ®nal peptide concentration
100 mg/mL. The initial time was recorded and 100 mL of
3
5, 161.
6. Nakajima, R.; Kitamura, A.; Someya, K.; Tanaka, M.;
reaction solution was removed at known time intervals
and added into 100 mL of 10% aqueous trichloroacteric
acid (TCA) solution. The cloudy reaction sample was
2
Sato, K. Antimicrob. Agents Chemother. 1995, 39, 1517.
ꢀ
27. Chen, Y. H.; Yang, J. T.; Chau, K. H. Biochemistry 1974,
1
cooled at 4 C for 15 min and spun at 13,000 g for 15
2
3, 3350.
8. Powell, M. F.; Stewart, T.; Otvos, L. Jr.; Urge, L.; Gaeta,
8
min to precipitate serum protein. Peptide analysis was
carried out by reverse phase HPLC with Waters C18
2
F. C.; Sette, A.; Arrhenius, T.; Thomson, D.; Soda, K.; Colon,
S. M. Pharm. Res. 1993, 10, 1268.
29. Murakami, K.; Nomura, K.; Doi, M.; Yosida, T. Anti-
microb. Agents Chemother. 1987, 31, 1307.
column. Kinetic analysis was carried out by a linear
least square analysis of the logarithm of the peak area
versus time. Each half-life was determined from two