Biaryl amino acid templates in place of D-Pro-l-Pro in cyclic ss-hairpin cationic antimicrobial peptidomimetics

Article abstract of DOI:10.1039/b706370a

The turn-forming d-Pro-l-Pro template has been frequently used to promote regular ss-hairpin conformations in cyclic protein epitope mimetics. Here the use of three isomeric biaryl templates has been studied as alternatives to d-Pro-l-Pro in the preparati

Full text of DOI:10.1039/b706370a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Biaryl amino acid templates in place of D-Pro-L-Pro in cyclic ß-hairpin
cationic antimicrobial peptidomimetics
Nityakalyani Srinivas,a Kerstin Moehle,a Khaled Abou-Hadeed,a Daniel Obrechtb and John A. Robinson*a
Received 27th April 2007, Accepted 11th July 2007
First published as an Advance Article on the web 17th August 2007
DOI: 10.1039/b706370a
The turn-forming D-Pro-L-Pro template has been frequently used to promote regular ß-hairpin
conformations in cyclic protein epitope mimetics. Here the use of three isomeric biaryl templates has
been studied as alternatives to D-Pro-L-Pro in the preparation of ß-hairpin peptidomimetics. The o,oꢀ-
o,mꢀ- and m,mꢀ-isomers of carboxymethyl- and aminomethyl-substituted biaryl templates have been
incorporated into novel macrocyclic mimics of the naturally occurring cationic antimicrobial peptide
protegrin I. The presence of the o-carboxymethyl-oꢀ-aminomethyl-biaryl template within the
macrocyclic peptide resulted in the appearance of slowly interconverting atropisomers. Although none
of the resulting mimetics adopted stable ß-hairpin structures in aqueous solution, they all nevertheless
retained a significant antimicrobial activity against Gram positive and Gram negative bacteria. These
mimetics provide interesting starting points for an optimization program in the search for potent and
novel antimicrobial compounds.
in solution and bound to lipid micelles.7–10 The ß-hairpin structure
of PG-I is important for its ability to interact with lipid membranes
Introduction
and its broad spectrum antimicrobial activity.11–15 Unfortunately,
PG-I is also highly hemolytic towards mammalian cells, reflecting
a poor selectivity in binding bacterial cell membranes rather than
eukaryotic cell membranes. The mechanism(s) underlying this
selectivity in membrane binding is (are) still incompletely un-
derstood. Macrocyclization is an effective strategy for restricting
peptide conformation, and the introduction of semi-rigid linkers
or templates provides a further opportunity for influencing the
conformation, as well as the biological activity and selectivity.
In previous work,16–18 we described PG-I peptidomimetics
comprising a backbone cyclic peptide attached to a ß-hairpin-
inducing D-Pro-L-Pro template (Fig. 1).19–21 The mimetic 1, for
example, retains much of the antimicrobial activity of PG-I,
but shows a much reduced hemolytic activity against human
red blood cells, reflecting an enhanced selectivity for bacterial
membranes.17 In this work, we set out to test whether templates
based on biaryl amino acids could be used in place of D-Pro-L-
Pro to generate new PG-I mimetics with antimicrobial activity.
The large class of naturally occurring cationic antimicrobial
peptides has attracted great interest recently in the search for new
antibiotics to combat the growing threat posed by multiple antibi-
otic resistant bacteria.1–4 The antimicrobial peptides discovered so
far display divergent sequences and secondary structures, although
for many, an underlying common feature is their ability to
adopt amphiphilic conformations, with hydrophobic and cationic
residues clustered in spatially distinct regions. This can endow on
the peptide an ability to selectively bind and disrupt bacterial cell
membranes.3
Protegrin-I (PG-I), for example, is an 18 amino acid disulfide
bridged cationic antimicrobial peptide from porcine leucocytes
(Fig. 1),5,6 which adopts an amphiphilic ß-hairpin structure both
aDepartment of Chemistry, University of Zurich, Winterthurerstrasse 190,
8057 Zurich, Switzerland. E-mail: robinson@oci.uzh.ch; Fax: (+41) 1-635-
6833
bPolyphor AG, Gewerbestrasse 14, 4123 Allschwil, Switzerland
Fig. 1 Protegrin-I sequence (dashed lines indicate disulfide bridges), template-bound mimetic 1 and the biaryl templates (2–4) studied here.
3100 | Org. Biomol. Chem., 2007, 5, 3100–3105 This journal is The Royal Society of Chemistry 2007
©

View Article Online
Biaryl units represent interesting semi-rigid templates to control
peptide conformation. Biaryls occur in many peptide natural
products, such as the glycopeptide antibiotic vancomycin, the
proteosome inhibitor TMC-95A,22 and the peptide antibiotic WS-
43708A.23 Biaryl units have also been incorporated previously into
various cyclic loop mimetics.24–28 The biaryl units chosen for study
here are the positional isomers 2–4, which have been inserted
directly into the mimetic 1 in place of the D-Pro-L-Pro template.
same two peaks in a 1 : 1 ratio. This is due to the presence
of two atropisomeric forms, which interconvert slowly on the
HPLC timescale at room temperature. Heating the HPLC column
to 60 ◦C caused the two peaks to coalesce to a single peak.
Furthermore, when each step of the assembly of the linear peptide
precursor of 18 was followed by HPLC analysis of peptides
released from the resin with 95% TFA, all the linear peptides
obtained up to and including the coupling of the template showed
only a single HPLC peak, whereas linear peptides obtained after
coupling valine to the template each appeared as two peaks
by HPLC, again consistent with the slow interconversion of
diastereomers due to atropisomerism at the biphenyl template
within the peptide chain. Variable temperature 1H NMR spectra
of 18 in aqueous solution also revealed a coalescence of signals
over the range 300–330 K, consistent with a faster interconversion
of the biaryl atropisomers on the NMR timescale at the elevated
temperature.
Results and discussion
The o,oꢀ-biphenyl template 2 has been prepared previously from
commercially available diphenic anhydride by Brandmeier and
Feigel,24 whereas a related (but not identical) o,mꢀ-biphenyl
template has been prepared by Suich et al. using a Suzuki
coupling.25 Here the required Fmoc-protected building blocks
15–17 could all be prepared by the route shown in Scheme 1,
which is based on the Suzuki chemistry reported by Suich et al.25
The target peptidomimetics 18–20 were then synthesized by a
two-step strategy, using Fmoc solid-phase peptide chemistry to
assemble a linear precursor on the resin, followed by a solution-
phase macrocyclization and then deprotection (Scheme 2). These
mimetics contain the same hairpin sequence as in the mimetic 1.
The desired products were obtained in 50–80% yield after HPLC
purification, and each was fully characterized by NMR and MS.
The mimetics 19 and 20 produced in this way showed single
We also tested whether the rate of atropisomer interconversion
around the biaryl link could be slowed down by incorporating
the o,oꢀ-biphenyl template into the smaller macrocyclic pep-
tidomimetic 21 (Scheme 2). This cyclic molecule was prepared in
an analogous way, and again appeared upon analysis by HPLC as
two peaks. Now, however, the two components could be separated
by HPLC and were stable (i.e. did not interconvert rapidly) at room
temperature. NMR studies in aqueous solution (H2O–D2O 9 : 1,
pH 3, 300 K) were then pursued with these two cyclic mimetics
21a (elutes first from a C18 reverse phase HPLC column with a
MeCN–H2O gradient of 5–100% MeCN over 30 min) and 21b
1
peaks upon reverse phase HPLC analysis, and H NMR spectra
were consistent with the presence of one major average conformer
in solution. The mimetic 18, however, behaved differently, showing
two peaks in a 1 : 1 ratio by HPLC, which could be separated,
but upon re-analysis by HPLC each appeared again as the
1
(elutes second). Both samples gave H NMR spectra consistent
with a single average conformer present in solution, with all amide
bonds trans. However, they differed clearly in the pattern of NOE
Scheme 1 Synthetic route to template building blocks 15–17. Reagents: (a) Pd2(dba)3, P(o-Tol)3, 2 M Na2CO3, MeOH–toluene (1 : 4), 70–80 ◦C, 16 h; 9
(30%), 10 (85%), 11 (87%). (b) NH2OH·HCl, NaOMe, MeOH 50 ◦C, 16 h. (c) Pd/C (10%), H2 (1 bar), MeOH, rt, 16 h; 12 (72%), 13 (72%), 14 (85%).
(d) i. NaOH, MeOH; ii. FmocOSu, dioxane–H2O, rt, 15 h; 15 (56%), 16 (70%), 17 (70%).
Scheme 2 Synthesis of the template-bound protegrin mimetics 18–20, and the small loop mimetic 21.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3100–3105 | 3101
©

View Article Online
contacts observed in 2D NOESY plots. For both, most observed
NOEs were intra-residue or sequential, with very few being longer-
range NOEs. The resulting NOE-derived distance restraints were
used to calculate compatible average solution structures, using the
program DYANA.29 For each molecule, the resulting 20 lowest
energy structures are shown in Fig. 2, and statistics from the
calculations are given in Table 1.
(Table 1). No regular ß-turn-like conformers are seen within the
tetrapeptide motif LRRV. Rather, several residues exhibit ␾/w
values in non-allowed regions of Ramachandran space, suggesting
either that the structures are strained, or that they represent
averages of two or more rapidly interconverting backbone con-
formations. However, the structure calculations generated unique
configurations around the biaryl template, with 21a having the S
and 21b the R absolute configuration at the biaryl axis of chirality
(Fig. 2). That the biaryl template of type 2 does not stabilize a
regular ß-turn or ß-hairpin in the tetrapeptide motif within 21
is consistent with an earlier modelling study, which suggested
that the geometry of this o,oꢀ-biphenyl template (in at least one
configuration) is incompatible with such a ß-turn loop structure,30
at least when occupying a non-hydrogen-bonding position within
a ß-hairpin loop.
Further insights into the solution conformations of 21a and 21b
were obtained from CD spectra. In the aromatic region (238 nm),
21a displayed a positive CD peak ([h] = +3000 deg cm2 dmol−1),
whereas at this wavelength 21b showed a very weak negative CD
peak ([h] = −400 deg cm2 dmol−1). Earlier studies of the aromatic
adsorption bands in the CD spectra of optically active biphenyls
have shown that the sign and shape of the curves is dependent
upon the type of substituent in the biphenyl nucleus as well as on
the absolute configuration.31–33 For this reason, an unambiguous
correlation between the sign of the 238 nm band and the sense of
chirality in the template has not been attempted here. On the other
hand, in the amide region, larger mirror image-related CD peaks
are observed. Thus, 21a displayed a negative peak at 190 nm ([h] =
−31·500 deg cm2 dmol−1) and a positive peak at 212 nm ([h] =
+8400 deg cm2 dmol−1), whereas 21b displayed a positive peak
at 190 nm ([h] = +32·100 deg cm2 dmol−1) and a negative one
at 212 nm ([h] = −16·500 deg cm2 dmol−1). The shape of the
CD curve for 21b thus resembles those typically seen for ß-sheet
structures, whereas that of 21a resembles more the CD profile of
polyproline-II and unordered peptide conformations.34 These data
confirm the divergent chiroptical properties of 21a and 21b, and
support the proposal that they are related as atropisomers in the
biphenyl template.
1H NMR studies were also undertaken on the mimetics 18,
19, and 20. The appearance of two signal sets, for the two
interconverting biaryl rotamers of 18, made complete chemical
shift assignments impossible. However, no evidence could be found
in 2D NOESY plots of long range NOEs. This suggested that a
single stable folded structure is not populated under the conditions
used (H2O–D2O, 9 : 1, pH 6). In the case of 19 and 20, only a single
set of signals was observed for each. Again, no long range NOEs
were observed, suggesting strongly that a stable fold, such as a
ß-hairpin, is not significantly populated in solution. These studies
strongly suggest that 18, 19, and 20 have relatively flexible (i.e.
disordered) backbones in aqueous solution. Similar results were
found for the peptidomimetic 1 in earlier work.17
Fig. 2 Superposition of the final 20 NMR structures for 21a (left) and
21b (right), with the N-terminal phenyl ring of the biaryl template pale
green and the C-terminal phenyl ring dark green. A single representative
NMR structure is represented as a stick model (N atoms blue and O atoms
red).
It is notable that for each molecule, the 20 lowest energy
backbone conformers populate divergent regions of ␾/w-space
(Table 2), although there were no significant NOE violations
Table 1 Summary of the upper distance restraints and statistics for the
NMR structure calculations
21a
21b
NOE upper-distance limits
Intraresidue
Sequential
32
12
20
41
22
17
Medium- and longe-range
0
2
2
˚
Residual target function value/A
0.31 0.01
0.42 0.04
˚
Mean rmsd values/A
All backbone atoms
All heavy atoms
1.17 0.50
3.31 1.13
0.72 0.23
2.39 0.67
Residual NOE violation
˚
Number > 0.2 A
Maximum/A
0
0
2
0.22
˚
Residues in
- most favoured ␾/w regions
- additional allowed ␾/w regions
- generously allowed ␾/w regions
- disallowed regions
1%
10%
55%
30%
5%
55%
28%
16%
Table 2 Averaged backbone torsion angles of the 20 NMR structures
21a 21b
69.1 95.1 43.6 87.0
Leu1
Arg2
Arg3
Val4
U
w
U
w
U
w
U
w
na
83.7 52.9
25.3 94.2
58.7 29.6
156.5 70.6
−64.4 18.5
−133.3 28.1
58.5 18.4
130.7 3.2
134.9 46.4
35.1 49.4
68.1 74.8
62.2 80.2
169.0 75.0
76.8 94.2
18.5 2.5
Finally, the antimicrobial activities of the peptidomimetics 18,
19, and 20 were determined by the NCCLS broth microdilution
method,35 using representative gram positive and gram negative
test microorganisms. The hemolytic activity was also measured
on fresh human red blood cells, as described earlier.16,17 The
results, given in Table 3, show that all three peptidomimetics retain
significant antimicrobial activity against all the organisms tested.
Moreover, the hemolytic activities of all three are much lower than
Biaryl5
−119.8 5.9
a The torsion is defined as C2–C1–C1ꢀ–C2ꢀ from the NH2 to COOH
terminus.
3102 | Org. Biomol. Chem., 2007, 5, 3100–3105
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 3 Antimicrobial activities (minimal inhibitory concentrations (MIC) in lg ml−1) and hemolytic activity (% hemolysis of human red blood cells at
100 lg ml−1
)
E. coli
(ATCC 25922)
P. aeruginosa
(ATCC 27853)
S. aureus
(ATCC 29213)
S. aureus
(ATCC 25923)
Peptide
Hemolysis (%)
1
18
19
20
12
64
32
16
6
32
16
32
12
16
8
25
16
16
16
1.0
2.5
0
16
0.2
seen with PG-I. Indeed the o,mꢀ-biaryl templated mimetic 19 was
essentially non-hemolytic under the conditions tested, despite the
fact that the biaryl unit is quite hydrophobic in character and might
be expected to enhance association of the molecule with a lipid
bilayer. The retention of significant antimicrobial activity with all
three mimetics is interesting, given that the three templates used
have different geometries and will influence the preferred peptide
backbone conformation in different ways. Nevertheless, all three
mimetics may still be capable of adopting amphiphilic structures
in a membrane environment.
One of the most important conclusions of this work is that novel
macrocyclic mimetics of PG-I containing the biaryl templates
2–4 can be made, which retain significant antimicrobial activity
and show low to zero hemolytic activity. As such, the mimetics
18–20 represent interesting starting points for the discovery of
new antimicrobial compounds, whose properties might be further
optimized through cycles of combinatorial library synthesis and
screening.16
a pale yellow oil (0.2 g, 80%). Rf (silica, cyclohexane–EtOAc 3 :
1) 0.24. 1H-NMR (300 MHz, CDCl3): 7.89–7.08 (m, 10H, 8 arom.
=
H, HC N, OH), 3.47 (s, OMe), 3.42 (d, J = 16.0, PhCH), 3.31
(d, J = 16.0, PhCH). MS (CI): m/z 270.1 (100, [M + H]+). The
oxime (0.26 g, 0.96 mmol) and 10% palladium on charcoal (0.06 g,
10% by wt) in methanol (5 mL), was stirred under hydrogen gas.
Upon completion, the catalyst was removed by filtration through
celite and washed with MeOH. Saturated aq. Na2CO3 was added
and extracted with CH2Cl2. The organic layers were combined,
dried over Na2SO4, and the solvent evaporated to afford the
amine (12) Rf (silica, EtOAc–MeOH 3 : 1): 0.18. The amine in
EtOH (15 mL) and H2O (30 mL) was treated with NaOH (10 mL
10 N), refluxed for 16 h, to 25 ◦C, acidified to pH 2, and extracted
with EtOAc. The combined organic layers were dried (Na2SO4)
and concentrated to yield the amino acid. A suspension of the
amino acid (1.9 g, 7.8 mmol) in dioxane (20 mL) was treated with
10% Na2CO3 (55 mL), cooled to 0 ◦C, treated with a solution
of Fmoc-O-succinimide (FmocOSu) (2.4 g, 7.3 mmol) in dioxane
(20 mL) over 15 min and stirred at rt for 1 h. Water (pH 3) was
added, and extracted with EtOAc. The organic layer was dried and
concentrated. Flash chromatography (silica, cyclohexane–EtOAc–
AcOH 80 : 20 : 1) afforded the desired protected amino acid 15
(1.2 g, 56%). Rf (silica, cyclohexane–EtOAc–AcOH 60 : 40 : 1):
Experimental
2-Methyl carboxymethyl-2ꢀ-formyl-biphenyl (9)
◦
1
To a solution of methyl 2-bromophenylacetate (5) (4.8 g, 20 mmol)
in toluene (100 mL) with aqueous Na2CO3 (20 mL, 2 M) was
added Pd2(dba)3 (0.6 g, 3 mol%) and P(o-tolyl)3 (0.4 g, 6 mol%)
followed by 2-formylbenzeneboronic acid (7) (2.7 g, 18.2 mmol) in
MeOH (25 mL), and the reaction mixture was heated to 70–80 ◦C
under Ar for 16 h, then cooled to 25 ◦C, diluted with EtOAc,
and washed sequentially with aq. NaHCO3, 10% citric acid and
brine. The organic layer was dried (Na2SO4) and concentrated.
Flash chromatography (silica, cyclohexane–EtOAc 9 : 1) afforded
the desired product 9 as a yellow oil (1.6 g, 30%). Rf (silica,
0.2. Mp: 80–82 C. IR (CCl4): 3449 m, 3068 s, 1716 s. H-NMR
(300 MHz, CDCl3): 7.95–7.28 (m, 16 arom. H), 5.31 (br. s, NH),
4.53 (d, J = 6.5, OCH2), 4.30 (m, 3H, PhCH, PhCH2), 3.68 (d,
J = 15.9, PhCH), 3.56 (d, J = 15.9, PhCH). 13C-NMR (75 MHz,
CDCl3): 172.54, 156.10, 143.74, 133.16, 132.40, 119.94, 65.19,
46.67, 41.38, 38.42. MS (ESI): m/z 486.1 (100, [M + Na]+). HR-
MS (CI): m/z 464.1864 ([M + H]+; C30H26NO4 , calc. 464.1856).
+
3-Methyl carboxymethyl-2ꢀ-formyl-biphenyl (10)
1
Was prepared by the same method as for 9 (see above) as a yellow
oil. Rf (silica, cyclohexane–EtOAc 3 : 1): 0.38. IR (CHCl3): 1736 s,
cyclohexane–EtOAc: 3 : 1): 0.4. IR (CHCl3): 1735 s, 1693 s. H-
NMR (300 MHz, CDCl3): 9.65 (s, CHO), 7.98–7.15 (m, 8 arom.
H), 3.45 (d, J = 15.9, PhCH), 3.44 (s, OMe), 3.35 (d, J = 15.9,
PhCH). 13C-NMR (75 MHz, CDCl3): 191.83, 171.31, 144.27,
137.86, 134.08, 133.34, 132.63, 130.64, 126.92, 51.77, 38.78. MS
(CI): m/z 272.1 (100, [M + NH4]+).
1
1690 s. H-NMR (300 MHz, CDCl3): 10.22 (s, CHO), 8.27–7.51
(m, 8 arom. H), 3.95 (s, OMe), 3.93 (s, 2H, PhCH2). 13C-NMR
(75 MHz, CDCl3): 192.20, 171.53, 145.46, 137.96, 134.21, 133.66,
133.42, 127.49, 52.02, 40.88. MS (CI): m/z 272.1 (100, [M +
NH4]+).
2-(N-9-Fluorenylmethyloxycarbonyl)aminomethyl-2ꢀ-
carboxymethyl-biphenyl (15)
3-(N-9-Fluorenylmethyloxycarbonyl)aminomethyl-2ꢀ-
carboxymethyl-biphenyl (16)
To NH2OH·HCl (0.12 g, 1.47 mmol) and NaOMe (0.079 g, 1.47 m
mol) in H2O–MeOH (6 mL, 1 : 4), was added 9 (0.25 g, 0.98 mmol),
heated at 50–55 ◦C for 16 h. The solvent was then removed and the
residue partitioned between 1 N NaHSO4 and EtOAc. The organic
layer was dried (MgSO4) and evaporated. Flash chromatography
(silica, cyclohexane–EtOAc, 85 : 15) afforded the desired oxime as
Was prepared by the same method as for 15 (see above). Rf
(silica, cyclohexane–EtOAc–AcOH 60 : 40 : 1): 0.22. Mp: 145–
◦
1
146 C. IR (KBr): 3409 m, 1813 m, 1790 m, 1739 s, 1713 s. H-
NMR (300 MHz, DMSO): 7.68–7.13 (m, 16 arom. H), 4.88 (br. s,
NH), 4.29 (d, J = 6.8, OCH2), 4.23 (d, J = 5.7, PhCH2), 4.09 (t,
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3100–3105 | 3103
©

View Article Online
J = 6.7, PhCH), 3.59 (s, PhCH2). 13C-NMR (75 MHz, CDCl3):
172.51, 156.20, 143.74, 135.06, 130.04, 119.95, 65.20, 46.68, 41.41,
40.60. MS (ESI): m/z 486.1 (100, [M + Na]+). HR-MS (CI): m/z
Synthesis of peptidomimetics 18, 19, 20 and 21
The synthesis of mimetic 18 is a typical procedure: Fmoc-
Arg(Pbf)-OH (1.2 equiv.) was coupled to 2-chlorotrityl chlo-
ride resin (1.08 mmol g−1) in the presence of diisopropylethy-
lamine (4 equiv.) in CH2Cl2. Following removal of the Fmoc
group with 20% piperidine in N-methylpyrrolidine (NMP), chain
elongation was performed sequentially with Fmoc-Lys(Boc)-
OH (2×), Fmoc-Leu-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Leu-OH,
15, Fmoc-Val-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Tyr(tBu)-OH,
Fmoc-Lys(Boc)-OH, Fmoc-Trp(Boc)-OH, and Fmoc-Arg(Pbf)-
OH (each 4 equiv.), using 20% piperidine–NMP for Fmoc
deprotection, HOBt–HBTU for activation, diisopropylethylamine
as a base and NMP as solvent. After completion of the syn-
thesis, the linear peptide was cleaved from the resin with 0.6%
CF3COOH in CH2Cl2, and then cyclized overnight at RT in
1% diisopropylethylamine–dimethylformamide with HOBt and
HBTU (3 equiv.). Deprotection with CF3COOH–iPr3SiH–H2O
(95 : 2.5 : 2.5), precipitation with diethyl ether, followed by
purification by preparative HPLC-MS (C18 column, gradient 20–
50% MeCN in H2O in 5 column volumes) gave 18 in 10% overall
yield, and >95% purity by HPLC.
464.1870 ([M + H]+; C30H26NO4 , calc. 464.1856).
+
3-Methyl carboxymethyl-3ꢀ-formyl-biphenyl (11)
Was prepared by the same method as for 9 (see above) as a yellow
oil. Rf (silica, cyclohexane–EtOAc 3 : 1): 0.42. IR (neat): 1732 s,
1
1694 s, 1600 m. H-NMR (300 MHz, CDCl3): 10.19 (s, CHO),
8.21–7.37 (m, 8 arom. H), 3.824 (s, OMe), 3.819 (s, CH2). 13C-
NMR (75 MHz, CDCl3): 192.12, 171.70, 141.75, 139.97, 136.84,
134.69, 132.98, 125.87, 52.00, 41.04. MS (CI): m/z 272.2 (100,
[M + NH4]+).
3-(N-9-Fluorenylmethyloxycarbonyl)aminomethyl-3ꢀ-
carboxymethyl-biphenyl (17)
Was prepared by the same method as for 15 (see above). Rf
(silica, cyclohexane–EtOAc–AcOH 60 : 40 : 1): 0.19. Mp: 187–
◦
1
189 . IR (KBr): 3345 m, 1734 m, 1698 s. H-NMR (300 MHz,
DMSO): 7.69–7.19 (m, 16 arom. H), 5.58 (br. s, NH), 4.39 (s,
2H, OCH2), 4.37 (s, 2H, PhCH2), 4.17 (t, J = 6.9, PhCH), 3.59 (s,
2H, PhCH2). 13C-NMR (75 MHz, CDCl3): 172.51, 156.22, 143.72,
135.66, 128.75, 119.93, 65.26, 46.63, 43.72, 40.67. MS (ESI): m/z
486.1 (100, [M + Na]). HR-MS (CI): m/z 464.1867 ([M + H]+;
The retention times (tR) on analytical C18 reverse phase HPLC
−1
˚
(Vydac C18 218TP104, 10 lm, 125 A, 4 × 100 mm, 1 ml min )
using a gradient 5–50% MeCN in H2O (+0.1% TFA) in 3 column
volumes, and the mass by ESI-MS are reported below:
+
C30H26NO4 , calc. 464.1856).
Antimicrobial assays
Table 4
Innocula of each microorganism were diluted into Mueller–
Hinton (MH) broth to give approximately 106 colony form-
ing units (CFU) per mL. Aliquots (50 lL) of the innocula were
added to MH broth (50 lL) containing the peptide in serial
twofold dilutions. Antimicrobial activities are expressed as the
minimal inhibitory concentration (MIC) in lg mL−1 at which
100% inhibition of growth was observed after 18–20 h incubation
at 37 ◦C.
Peptide
tR/min
MS (m/z) [M + H]+
18
18.2, 18.5
14.6
14.5
13.2
13.6
1906.8
1906.7
1906.7
748.9
19
20
21a
21b
749.0
Table 5 The 1H NMR assignments of peptides 21a and 21b are given below (H2O–D2O 9 : 1, pH 3, 300 K):
Residue
3J(NHCaH)
NH
H-C(a)
H-C(b)
Others
21a
Leu1
Arg2
Arg3
Val4
5.2
5.6
7.7
8.7
—
7.72
8.55
8.15
7.60
8.23
3.67
3.96
4.17
4.02
1.42, 1.42
1.86, 1.86
1.78, 1.93
2.01
CH(c) 1.42; CH3(d) 0.86, 0.86
CH2(c) 1.58, 1.58; CH2(d) 3.19, 3.19; NH(e) 7.20
CH2(c) 1.54, 1.54; CH2(d) 3.18, 3.18; NH(e) 7.19
CH3(c) 0.77, 0.87
Biaryl5
(NH)CH2 4.15, 4.37
Phenyl ring (N-term.): H(d1) 7.35, H(e1) 7.45, H(f) 7.40, H(e2) 7.22
Phenyl ring (C-term.): H(d1) 7.41, H(e1) 7.22, H(f) 7.44, H(e2) 7.28
CH2 (CO) 3.64, 3.76
21b
Leu1
Arg2
Arg3
Val4
5.1
5.6
6.3
8.0
—
7.67
8.35
8.17
7.84
8.16
4.16
4.03
4.01
3.91
1.47, 1.47
1.77, 1.85
1.74, 1.89
2.16
CH(c) 1.55; CH3(d) 0.87, 0.91
CH2(c) 1.56, 1.60; CH2(d) 3.18, 3.18; NH(e) 7.20
CH2(c) 1.50, 1.50; CH2(d) 3.14, 3.14; NH(e) 7.16
CH3(c) 0.91, 0.97
Biaryl5
(NH)CH2 3.88, 4.57
Phenyl ring (N-term.): H(d1) 7.43, H(e1) 7.22, H(f) 7.42, H(e2) 7.22
Phenyl ring (C-term.): H(d1) 7.46, H(e1) 7.44, H(f) 7.40, H(e2) 7.23
CH2 (CO) 3.52, 3.60
3104 | Org. Biomol. Chem., 2007, 5, 3100–3105
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Hemolytic activity
10 J. J. Buffy, A. J. Waring and M. Hong, J. Am. Chem. Soc., 2005, 127,
4477.
11 J. Chen, T. J. Falla, H. Liu, M. A. Hurst, C. A. Fujii, D. A. Mosca, J. R.
Embree, D. J. Loury, P. A. Radel, C. C. Chang, L. Gu and J. C. Fiddes,
Biopolymers, 2000, 55, 88.
12 S. S. L. Harwig, A. Waring, H. J. Tang, Y. Cho, L. Tan and R. I. Lehrer,
Eur. J. Biochem., 1996, 240, 352.
13 W. G. Jing, E. J. Prenner, H. J. Vogel, A. J. Waring, R. I. Lehrer and K.
Lohner, J. Pept. Sci., 2005, 11, 735.
14 R. Mani, J. J. Buffy, A. J. Waring, R. I. Lehrer and M. Hong,
Biochemistry, 2004, 43, 13839.
Fresh human red blood cells were washed three times with
phosphate buffered saline (PBS), and then incubated with peptide
at a concentration of 100 lg mL−1 for 1 h at 37 ◦C. The final
erythrocyte concentration was ca. 0.9 × 109 per mL. The values
of 0% and 100% lysis were determined by incubation of cells with
PBS or 0.1% Triton X-100 in water, respectively. The samples
were centrifuged, the supernatant diluted 20-fold in PBS, and the
optical density was measured at 540 nm.
15 R. Mani, A. J. Waring, R. I. Lehrer and M. Hong, Biochim. Biophys.
Acta: Biomembranes, 2005, 1716, 11.
16 J. A. Robinson, S. C. Shankaramma, P. Jetter, U. Kienzl, R. A.
Schwenderner, J. W. Vrijbloed and D. Obrecht, Bioorg. Med. Chem.,
2005, 13, 2055.
17 S. C. Shankaramma, Z. Athanassiou, O. Zerbe, K. Moehle, C. Mouton,
F. Bernardini, J. W. Vrijbloed, D. Obrecht and J. A. Robinson,
ChemBioChem, 2002, 3, 1126.
Abbreviations
Boc, t-butoxycarbonyl; DIEA, N,Nꢀ-diisopropylethylamine;
DMF,
N,N -dimethylformamide;
Fmoc,
[(9H-fluorenyl)-
methoxy]carbonyl; HBTU, 2-[1H-benzotriazole-1-yl]-1,1,3,3-
tetramethyluronium hexafluorophosphate; HOBt, N-hydroxy-
benzotriazole; NCCLS, National Committee for Clinical
Laboratory Standards (now, Clinical and Laboratory Standards
Institute, CLSI); Pbf, 2,2,4,6,7-pentamethyldihydrobenzofuran-
5-sulfonyl; NMP, N-methylpyrrolidone; TFA, trifluoroacetic
acid.
18 S. C. Shankaramma, K. Moehle, S. James, J. W. Vrijbloed, D. Obrecht
and J. A. Robinson, Chem. Commun., 2003, 1842.
19 C. M. Nair, M. Vijayan, Y. V. Venkatachalapathi and P. Balaram,
J. Chem. Soc., Chem. Commun., 1979, 1183.
20 J. Spa¨th, F. Stuart, L. Jiang and J. A. Robinson, Helv. Chim. Acta, 1998,
81, 1726.
21 J. A. Robinson, Synlett, 2000, 429.
22 M. Kaiser, M. Groll, C. Siciliano, I. AssfalgMachleidt, E. Weyher,
J. Kohno, A. G. Milbradt, C. Renner, R. Huber and L. Moroder,
ChemBioChem, 2004, 5, 1256.
23 R. Kannan and D. H. Williams, J. Org. Chem., 1987, 52, 5435.
24 V. Brandmeier and M. Feigel, Tetrahedron, 1989, 45, 1365.
25 D. J. Suich, S. A. Mousa, G. Singh, G. Liapakis, T. Reisine and W. F.
DeGrado, Bioorg. Med. Chem., 2000, 8, 2229.
26 R. B. Hopkins and A. D. Hamilton, J. Chem. Soc., Chem. Commun.,
1987, 171.
Acknowledgements
The authors thank the Swiss National Science Foundation for
financial support.
27 J. P. Schneider and J. W. Kelly, Chem. Rev., 1995, 95, 2169.
28 R. P. Cheng, D. J. Suich, H. Cheng, H. Roder and W. F. DeGrado,
J. Am. Chem. Soc., 2001, 123, 12710.
29 P. Guntert, C. Mumenthaler and K. Wu¨thrich, J. Mol. Biol., 1997, 273,
283.
References
1 K. A. Brogden, Nat. Rev. Microbiol., 2005, 3, 238.
2 J. B. McPhee and R. E. W. Hancock, J. Pept. Sci., 2005, 11, 677.
3 M. Zasloff, Nature, 2002, 415, 389.
4 D. Hopwood, S. Levy, R. P. Wenzel, N. Georgopapadakou, R. H. Baltz,
S. Bhavnani and E. Cox, Nat. Rev. Drug Discovery, 2007, 6, 8.
5 V. N. Kokryakov, S. S. L. Harwig, E. A. Panyutich, A. A. Shevchenko,
G. M. Aleshina, O. V. Shamova, H. A. Korneva and R. I. Lehrer, FEBS
Lett., 1993, 327, 231.
30 G. Mu¨ller, G. Hessler and H. Y. Decornez, Angew. Chem., Int. Ed.,
2000, 39, 894.
31 E. Bunnenberg, C. Djerassi, K. Mislow and A. Moscowitz, J. Am.
Chem. Soc., 1962, 84, 2823.
32 K. Mislow, E. Bunnenberg, R. Records, K. Wellman and C. Djerassi,
J. Am. Chem. Soc., 1963, 85, 1342.
6 C. Q. Zhao, T. Ganz and R. I. Lehrer, FEBS Lett., 1995, 368, 197.
7 A. Aumelas, M. Mangoni, C. Roumestand, L. Chiche, E. Despaux, G.
Grassy, B. Calas and A. Chavanieu, Eur. J. Biochem., 1996, 237, 575.
8 R. L. Fahrner, T. Dieckmann, S. S. L. Harwig, R. I. Lehrer, D. Eisenberg
and J. Feigon, Chem. Biol., 1996, 3, 543.
9 C. Roumestand, V. Louis, A. Aumelas, G. Grassy, B. Calas and A.
Chavanieu, FEBS Lett., 1998, 421, 263.
33 K. Mislow, M. A. W. Glass, R. E. O’Brien, P. Rutkin, D. H. Steinberg,
J. Weiss and C. Djerassi, J. Am. Chem. Soc., 1962, 84, 1455.
34 R. W. Woody, Adv. Biophys. Chem., 1992, 2, 37.
35 Methods for dilution antimicrobial susceptibility tests for bacteria
that grow aerobically, National Committee for Clinical Laboratory
Standards, Wayne, PA, USA, 2007, approved standard 7th edn, M07-
A7.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 3100–3105 | 3105
©