Alanine Scan of the Peptide Antibiotic Feglymycin: Assessment of Amino Acid Side Chains Contributing to Antimicrobial Activity

Article abstract of DOI:10.1002/cbic.201300032

The antibiotic feglymycin is a linear 13-mer peptide synthesized by the bacterium Streptomyces sp. DSM 11171. It mainly consists of the nonproteinogenic amino acids 4-hydroxyphenylglycine and 3,5-dihydroxyphenylglycine. An alanine scan of feglymycin was performed by solution-phase peptide synthesis in order to assess the significance of individual amino acid side chains for biological activity. Hence, 13 peptides were synthesized from di- and tripeptide building blocks, and subsequently tested for antibacterial activity against Staphylococcus aureus strains. Furthermore we tested the inhibition of peptidoglycan biosynthesis enzymes MurA and MurC, which are inhibited by feglymycin. Whereas the antibacterial activity is significantly based on the three amino acids D-Hpg1, L-Hpg5, and L-Phe12, the inhibitory activity against MurA and MurC depends mainly on L-Asp13. The difference in the position dependence for antibacterial activity and enzyme inhibition suggests multiple molecular targets in the modes of action of feglymycin. à la mode scanning: An alanine scan of the peptide antibiotic feglymycin was performed by solution-phase peptide synthesis to assess the significance of individual amino acid side chains for its biological activity. Antibacterial activity against Staphylococcus aureus and the inhibition of enzymes MurA and MurC depend on different amino acids.

Full text of DOI:10.1002/cbic.201300032

CHEMBIOCHEM
FULL PAPERS
DOI: 10.1002/cbic.201300032
Alanine Scan of the Peptide Antibiotic Feglymycin:
Assessment of Amino Acid Side Chains Contributing to
Antimicrobial Activity
Anne Hꢀnchen,[a] Saskia Rausch,[a] Benjamin Landmann,[a] Luigi Toti,[b] Antje Nusser,[b] and
Roderich D. Sꢁssmuth*[a]
The antibiotic feglymycin is a linear 13-mer peptide synthe-
sized by the bacterium Streptomyces sp. DSM 11171. It mainly
consists of the nonproteinogenic amino acids 4-hydroxyphe-
nylglycine and 3,5-dihydroxyphenylglycine. An alanine scan of
feglymycin was performed by solution-phase peptide synthesis
in order to assess the significance of individual amino acid side
chains for biological activity. Hence, 13 peptides were synthe-
sized from di- and tripeptide building blocks, and subsequently
tested for antibacterial activity against Staphylococcus aureus
strains. Furthermore we tested the inhibition of peptidoglycan
biosynthesis enzymes MurA and MurC, which are inhibited by
feglymycin. Whereas the antibacterial activity is significantly
based on the three amino acids d-Hpg1, l-Hpg5, and l-Phe12,
the inhibitory activity against MurA and MurC depends mainly
on l-Asp13. The difference in the position dependence for
antibacterial activity and enzyme inhibition suggests multiple
molecular targets in the modes of action of feglymycin.
Introduction
Peptide antibiotics synthesized by bacteria and fungi show re-
markable structural variety and involve sophisticated mecha-
nisms of action.[1] In particular, nonribosomally synthesized
peptide antibiotics often contain nonproteinogenic amino
acids and display various unusual structural features, such as
N-methylation, b-hydroxylation, and biaryl crosslinking.[2]
Among these there exists a number of antibacterial peptides
that contain the nonproteinogenic amino acids 4-hydroxyphe-
nylglycine (Hpg) and 3,5-dihydroxyphenylglycine (Dpg), or
both. Examples are the type I–IV glycopeptide antibiotics (e.g.,
vancomycin, teicoplanin, and complestatin),[3] ramoplanin,[4]
and arylomycin,[5] which mostly display antibacterial activity
and address different targets of the bacterial cell wall or mem-
brane.
ray structure analysis; this is reminiscent of gramicidin A[8] and
polytheonamide B.[9] The first total synthesis of feglymycin was
recently developed by our group and is based on a [6+7] frag-
ment coupling strategy (Scheme 1).[10] One of the main chal-
lenges of the synthesis was establishing suitable coupling con-
ditions to suppress racemization and epimerization of phenyl-
glycines and phenylglycine-containing peptides. Furthermore,
phenolic protecting groups caused severe solubility problems
in the later stages of the peptide assembly, thus requiring an
appropriate condensation strategy and suitable protecting
groups for mild final deprotection.
In subsequent studies, and unlike in previous reports,[6,7] we
found that feglymycin displays significant inhibition of three
selected S. aureus strains, with minimal inhibitory concentra-
tions (MICs) of 0.25–0.5 mm, in addition to a broad-spectrum in-
hibitory activity against HIV replication, with an IC50 of 0.8–
3.2 mm.[11] The remarkable antibacterial activity initiated subse-
quent studies and a more detailed search for a molecular
target. The relatively high molecular mass of feglymycin impli-
cated peptidoglycan biosynthesis as a reasonable target.[12]
However, an intermediate of the late stages of this biosynthe-
sis pathway, that is, lipid I, was experimentally excluded.[13] Fur-
ther assays with the cytoplasmic enzymes MurA–F from E. coli
and MurA–D from S. aureus (assembly of muramyl pentapep-
tide) finally revealed noncompetitive inhibition of MurA (UDP-
N-acetylglucosamine enolpyruvyl transferase) and MurC (UDP-
N-acetylmuramyl-l-alanine ligase) of both S. aureus (IC50 =3.5
and 1.0 mm, respectively) and E. coli (IC50 =3.4 and 0.3 mm,
respectively).[13] Hence, feglymycin is the first natural product
inhibitor with MurC as a main target. Unlike other synthetic
compounds, such as substrate or transition-state analogues
Recently, the peptide antibiotic feglymycin, which is synthe-
sized by the bacterium Streptomyces sp. DSM 11171, attracted
our attention.[6] This 13-mer peptide (1, Scheme 1) mainly con-
sists of the amino acids Hpg and Dpg. Furthermore, structural
analysis reveals an alternation of d- and l-configurations for
the peptide stretch between d-Dpg2 and l-Hpg11. Sheldrick
and co-workers revealed an antiparallel b-helical dimer[7] by X-
[a] Dr. A. Hꢀnchen, Dr. S. Rausch, B. Landmann, Prof. Dr. R. D. Sꢁssmuth
Technische Universitꢀt Berlin, Fakultꢀt II, Institut fꢁr Chemie
Strasse des 17. Juni 124, 10623 Berlin (Germany)
E-mail: suessmuth@chem.tu-berlin.de
Homepage: http://www.biochemie.tu-berlin.de
[b] Dr. L. Toti, A. Nusser
Sanofi-Aventis Deutschland GmbH
R&D LGCR/Parallel Synthesis & Natural Products Frankfurt
Industriepark Hoechst, 65926 Frankfurt am Main (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201300032.
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 625

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
Scheme 1. Structures of feglymycin (1) and key peptide fragments. The flexible strategy for the total synthesis from four dipeptides and one tripeptide was
applied to the design of a straightforward synthesis concept for an alanine scan of feglymycin, by exchanging amino acids in these building blocks. Seven
new Ala-containing dipeptides (35, 36, 38, 39, 41, 43, 44) and three new tripeptides (31–33) were synthesized for assembly into 13 alanine scan peptides (2–
14) and one double-alanine derivative [d-Ala2,10]-feglymycin (45). a) N-terminal deprotection of hexapeptides; b) fragment condensation.
with competitive binding,[14] feglymycin represents an unusual
case of a noncompetitive inhibitor.
fluences of amino acid side chains with antibacterial activity
when using the 17-mer peptide ramoplanin.[15b,c] Therefore we
designed a strategy to successively replace each amino acid
position by Ala and thereby yielded 13 new feglymycin deriva-
tives. Their antibacterial properties against S. aureus and inhibi-
tion of the peptidoglycan biosynthesis enzymes MurA and
MurC were tested. This initial structure–activity relationship
We sought to establish further details on the mode of action
of feglymycin. In this context, the “alanine scan” technique
proved as a useful and valuable tool for identifying the impor-
tance of specific residues for antibacterial and other bioactivi-
ties.[15] Boger and co-workers were able to relate distinct in-
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 626

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
study on feglymycin gives insights into contributions of single
amino acid side chains to the biological activity.
ic hydrogenolysis at the dipeptide stage. For rapid removal of
the temporary Boc group from di- and tetrapeptides, 4n HCl
in dioxane was used.[18] To realize the alanine scan, seven N-ter-
minal heptapeptides and six C-terminal hexapeptides were
prepared in quantities that allowed us to perform multiple
fragment condensations (Figure 1).
Results
Synthesis of N-terminal alanine scan heptapeptides (d-
Hpg1-l-Hpg7) and of C-terminal alanine scan hexapeptides
(d-Dpg8-l-Asp13)
As examples, the overall synthesis strategy for the prepara-
tion of peptides Cbz-[d-Ala4]-hepta-OH (19) and Boc-[l-Ala4]-
hexa-OBn (27) are outlined in Schemes 2 and 3, respectively.
For enhanced purification, hexa- or heptapeptides were pre-
cipitated from the reaction mixtures with water, and purified
by silica gel column chromatography. Prior to assembly of the
final 13-mer peptide, the methylester protecting group of the
N-terminal heptapeptide was cleaved under mild basic condi-
tions with trimethyltin hydroxide in 1,2-dichloroethane at 858C
(Scheme 2).[19]
Conceptually, we were aiming to exchange every position of
the 13-mer peptide feglymycin, while retaining the wild-type
d/l configuration. Thus the direct influence of Ala-exchange
on the secondary structure of the peptide would be mini-
mized. The synthetic strategy, which was based on the total
synthesis published previously by our group,[10] was adapted
to the requirements of the alanine scan. Accordingly, the [6+7]
fragment coupling of the 13-mer peptide was further parti-
tioned into coupling of Ala-containing di- and tripeptides to
readily available building blocks. Hence, for each alanine scan
peptide, the preparation of only one new dipeptide was neces-
sary. Overall, the synthetic concept afforded the preparation of
ten new Ala-containing di- and tripeptides (Scheme 1). This al-
lowed a manageable level of synthetic effort, as similar proce-
dures could be applied for the assembly of these peptides. In
summary, we obtained a straightforward synthesis of feglymy-
cin Ala-analogues by using a convergent strategy with newly
synthesized building blocks and known fragments for feglymy-
cin total synthesis.
Synthesis and analytical characterization of Ala-exchange
peptides of feglymycin
In the preparation of the penultimate [6+7] fragment coupling
of Ala-substituted hepta- (16–22) or hexapeptides (24–29)
with the corresponding wild-type peptide fragments 15 and
23, we employed Boc deprotection of sensitive hexapeptides
with 25% trifluoroacetic acid (TFA) in CH2Cl2 with triethylsilane
(TES) as scavenger. The deprotections were quantitative after
30 min at ambient temperature, and the obtained peptide TFA
salts could be used in the subsequent coupling step after
simple precipitation with Et2O. Fragment condensations of
heptapeptides 16–22 with hexapeptide 23 and of hexapep-
tides 24–29 with heptapeptide 15 were performed by using
DEPBT as the coupling reagent with NaHCO3 as a mild base, in
DMF. The best results (in terms of coupling efficiency and
yield) were obtained after a short stirring time (1 h) at 08C, fol-
lowed by a reaction time of ~2 d at ambient temperature. In
a similar fashion, the peptide [d-Ala2,10]-feglymycin (45) bear-
ing a double exchange of d-Dpg against d-Ala was assembled
by coupling 17 to 26. The resulting target peptides were pre-
cipitated by addition of water, and separated from water-solu-
ble byproducts by centrifugation. As an intermediate purifica-
tion step, we used Sephadex LH20 size-exclusion chromatogra-
phy with MeOH. To ensure high purity of the final alanine scan
In general, all peptide couplings were performed with 3-(di-
ethoxyphosphoryloxy)-1,2,3-benzotriazin-4(3H)-one
(DEPBT)
and NaHCO3 in N,N-dimethylformamide (DMF). For small oligo-
peptides, the use of tetrahydrofuran (THF) was also feasi-
ble.[10,16] These conditions ensured the epimerization-free as-
sembly of all arylglycine-containing oligopeptide fragments,
with good to high overall yield (60–90%). Standard peptide
coupling protocols (e.g., TBTU, DIPEA) were applied to gener-
ate dipeptide building blocks that did not contain either Hpg
or Dpg.[17] The benzylic side-chain-protecting groups were gen-
erally maintained in the first coupling step for the synthesis of
dipeptides. However, as the use of DEPBT as a coupling re-
agent does not necessarily require the protection of phenolic
hydroxy groups, these were quantitatively removed by catalyt-
Figure 1. Series of alanine exchange peptides of the N-terminal heptapeptide fragment 15 and of the C-terminal hexapeptide fragment 23.
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 627

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
spectra (see the Supporting In-
formation), and therefore likely
assemble as similar antiparallel
b-helical dimers, as suggested
for wild-type feglymycin.[7]
Antibacterial activity of fegly-
mycin alanine scan peptides
The antibacterial testing was
performed with three S. aureus
standard reference strains: one
MRSA (ATCC33592, resistant to
gentamicin and methicillin) and
two MSSA (ATCC29213 and
ATCC13709, sensitive to methi-
cillin and oxacillin). The antibac-
terial activities (Figure 2A) were
determined as MICs in order to
assess the influence of Ala-ex-
changes on inhibition of bacte-
rial growth. Of the series of
peptides tested, feglymycin de-
rivatives with the mutations
Val3Ala (4), Hpg7Ala (8), Val9Ala
(10), Hpg11Ala (12), and
Asp13Ala (14) showed MICs be-
tween 0.25 and 1 mm (feglymy-
cin 1, 0.25–0.5 mm). Replace-
ments of l-Val by the sterically
less demanding but comparably
nonpolar l-Ala apparently had
no effect on activity. Likewise,
substitutions Hpg7Ala (8) and
Hpg11Ala (12) had an almost
negligible
effect,
although
Scheme 2. Synthesis of the N-terminal peptide fragment Cbz-[d-Ala4]-hepta-OH (19). a) DEPBT, NaHCO3; b) H2, Pd/
C; c) 4n HCl/dioxane; d) Cbz-d-Hpg-OH, DEPBT, NaHCO3; e) Me3SnOH.
a more pronounced contribu-
tion from aromatic side chains
and from hydrogen bonding ca-
peptides, the protected tridecapeptides were additionally puri-
fied by reversed-phase HPLC prior to final quantitative depro-
tection with H2 and Pd/C (10%) in MeOH. The peptides thus
obtained could be directly used for biological assays.
pabilities of the phenolic groups had been expected. Surpris-
ingly, the Asp13Ala mutant (14, lacking the acidic side chain
functionality) had no negative impact on antimicrobial activity
(MIC=0.5 mm).
The identities of all Ala-feglymycin derivatives (compounds
2–14, 45) were confirmed by high resolution mass spectrome-
try, ESI-MS/MS, and 2D NMR spectroscopy (see the Supporting
Information). All peptides showed solubility in water and
MeOH similar to that of feglymycin. An exception was [d-Ala1]-
feglymycin (2), which showed significantly lower solubility and
eluted only as a broad peak from the analytical reversed-phase
column. To examine whether the exchange of each individual
amino acid against Ala had an influence on the secondary
structure of the corresponding peptides, CD spectra were re-
corded in water and trifluoroethanol (TFE; 20% aqueous).
Except for [l-Ala13]-feglymycin (14, substitution of the C-termi-
nal l-Asp), all alanine scan analogues showed comparable CD
All feglymycin peptides with exchange of the amino acid d-
Dpg!d-Ala at positions 2, 4, 6, 8, and 10 (peptides 3, 5, 7, 9,
and 11) showed MICs (0.5–4 mm) higher than for the wild-type
peptide (four- to 16-fold reduction in antimicrobial activity).
This finding points to a significant contribution of d-Dpg to
the antimicrobial effects. However, the most pronounced
impact was found for [d-Ala1]-feglymycin (2, MIC=8 mm), and
[l-Ala5]-feglymycin (6) and [l-Ala12]-feglymycin (13), which
showed MICs greater than 32 mm. Hence, residues d-Hpg1, l-
Hpg5, and l-Phe12 are crucial for antibacterial activity, as other
exchanges seem to be fairly well tolerated. The control
“double-mutant” peptide (dual d-Dpg!d-Ala mutations, [d-
Ala2,10]-feglymycin (45), MIC=4–8 mm) showed accumulating
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 628

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
Discussion and Conclusion
A full alanine scan of feglymycin was conducted to
provide insights into the contribution of each amino
acid side chain of feglymycin (1) to its antimicrobial
activity against S. aureus strains. The most important
residues were d-Hpg1, l-Hpg5, and l-Phe12. With
MICs of 8 mm (2) and >32 mm (6 and 13), Ala-ex-
change at these positions yielded the most signifi-
cant effects. This implies significant position-depen-
dent contributions from the aromatic side chains,
which might be attributable to p–p interactions. In
the case of Hpg, the additional possibility of H-bond
formation should be considered. However, only
a negligible influence on antibacterial activity was
found for the l-Hpg!l-Ala peptides [l-Ala7]- and
[l-Ala11]-feglymycin (8 and 12). This implies that the
importance of all four Hpg’s of feglymycin for anti-
microbial activity is not equal. The least effects on
antibacterial activity were found for the two l-Val!
l-Ala peptides, 4 and 10, possibly because of com-
parable polarity and less-pronounced steric effects
of the side chains. The MIC values of the d-Dpg!d-
Ala mutant peptides 2, 4, 6, 8, and 10 were mostly
between 2–4 mm (four- to 16-fold decrease in anti-
bacterial activity compared to feglymycin). Unlike d-
Hpg1, l-Hpg5, and l-Phe12, the aromatic amino acid
d-Dpg in general seems to exert a less significant in-
fluence on antibacterial activity. In addition, the in-
termediate MIC of the control “double mutant” pep-
tide [d-Ala2,10]-feglymycin (45, MIC 4–8 mm) sug-
gests, rather, an additive effect of Ala substitutions
of this amino acid on antimicrobial activity.
Scheme 3. Synthesis of the C-terminal peptide fragment Boc-[l-Ala4]-hexa-OBn (27).
a) DEPBT, NaHCO3; b) EDC, HOAt, NaHCO3; c) H2, Pd/C; d) 4n HCl/dioxane.
The results of the antibacterial assays contrast
with those from investigations of the Ala-exchange
peptides for their inhibitory activity against MurA and MurC
(enzymes of early peptidoglycan biosynthesis). The inhibition
assays against MurA and MurC consistently identified l-Asp13
as the only residue substitution to significantly contribute to
a drop in inhibition (seven- to 12-fold). The activity of MurC
was also reduced in the presence of the [d-Ala2,10] peptide
(45, eightfold), whereas MurA activity was not affected. In addi-
tion, for [l-Ala5]feglymycin (6), an inhibitory effect of l-Hpg5
for MurA was determined (greater than fivefold reduction in in-
hibition).
effects upon removal of natural d-Dpg amino acids, although
this was less pronounced than for 2, 6, and 13.
Inhibition of peptidoglycan biosynthesis enzymes MurA and
MurC
These Ala-exchange peptides were then tested for inhibitory
activity against enzymes MurA and MurC (early peptidoglycan
biosynthesis). These enzymes have been identified to have
been inhibited by feglymycin, whereas subsequent stages
(e.g., lipid I biosynthesis) remained unaffected.[13] The IC50
values were determined according to protocols established
previously for feglymycin.[13] Apart from [l-Ala5]-feglymycin (6,
IC50 =11.2 mm) and [l-Ala13]-feglymycin (14, IC50 =16.9 mm), all
Ala-exchange peptides (Figure 2B) inhibited MurA at concen-
trations similar to that for feglymycin (1, IC50 =2.5 mm). The
effect observed for [l-Ala13]-feglymycin (14) was paralleled in
the series testing MurC (feglymycin IC50 =0.3 mm): more than
tenfold reduction in inhibitory activity (IC50 =3.4 mm; Fig-
ure 2C).
Remarkably, comparison of the anti-staphylococcal assays
and the enzyme inhibition assays draws a picture of different
positional dependence between antibacterial activity in the
S. aureus system and MurA/MurC inhibition. Although three
residues, d-Hpg1, l-Hpg5, and l-Phe12, were identified as cru-
cial for antibacterial activity in vitro, only l-Asp13 contributed
to inhibition of MurA and MurC. In this context, the outcome
of the anti-HIV assays performed for the Ala-exchange peptides
deserves closer attention.[11] Previous investigations of feglymy-
cin by our groups allowed us to assign the HIV surface protein
gp120 as the molecular target. The data showed that the resi-
due dependence of the alanine scan in the anti-HIV assay
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 629

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
largely reflected that of the MurA and MurC inhibition testing.
Accordingly, [l-Ala13]-feglymycin (14) showed minor potency
in inhibition of HIV-1, whereas [l-Ala5]-feglymycin (6) might
suggest a more specific interaction by l-Hpg5 in the inhibition
of the target protein MurA (i.e., not required for MurC or
gp120).
The divergence in the outcome of the assays with S. aureus
and MurA/MurC enzymes appears indicative of two independ-
ent pharmacophoric regions in feglymycin, addressing molecu-
lar targets at different interaction sites. Naturally, one would
expect similar patterns of inhibition for in vitro and enzyme in-
hibitory data. Therefore, it is reasonable to assume that al-
though feglymycin exhibits pronounced effects on the pepti-
doglycan biosynthesis enzymes MurA and MurC, additional or
other molecules are antibacterial targets. As we exclude sur-
face-exposed late peptidoglycan biosynthesis steps as feglymy-
cin targets, other targets or mechanisms of action have to be
considered. This is corroborated by the physical properties of
feglymycin (e.g., MW ꢀ1.9 kDa), which are not characteristic of
cell-permeable molecules. Therefore, apart from molecular tar-
gets on the surface of the bacterial cell, dedicated uptake sys-
tems might be involved in the mode of action. Our observa-
tion is reminiscent of previous studies performed on the anti-
bacterial compound ramoplanin (MW ꢀ2.5 kDa). Initially, inhibi-
tion of intracellular MurG[20] and components of the surface-lo-
cated trans-glycosylation step[21] were suggested as molecular
targets, but more recently it appears that the trans-glycosyla-
tion step of lipid II synthesis is inhibited.[22]
The differences between in vitro and enzyme inhibition re-
sults for the alanine scan peptides clearly require further work
on the conformation and on target-dependent conformational
changes of feglymycin. Our future attempts will be directed at
elucidating the mode of action and identification of the dedi-
cated molecular targets of feglymycin. In combination with
new data, this study could be useful for the design of simpli-
fied analogues of feglymycin that require less-demanding ef-
forts for their synthesis. Therefore, this study constitutes an im-
portant step towards a more stable and structurally less com-
plex peptidomimetic with improved biological activity.
Experimental Section
Reagents and analytical methods: Chemicals were obtained from
the following commercial sources and used without further purifi-
cation: ABCR (Karlsruhe, Germany), Alfa Aesar (Karlsruhe, Germany),
Bachem, Fisher Scientific, Iris Biotech (Marktredwitz, Germany),
Merck, Orpegen Pharma (Heidelberg, Germany), and Sigma–Al-
drich. Deuterated solvents for NMR spectroscopy were obtained
from Euriso-Top (Saint-Aubin, France). Flash chromatography was
carried out with Davisil silica gel (40–63 mm; Grace, Deerfield, IL) or
the CombiFlash Rf system with prepacked silica gel columns (Tele-
dyne Isco, Lincoln, NE). Analytical thin layer chromatography was
performed with precoated aluminium sheets (silica gel 60 F254;
Merck). An analytical HPLC system 1100 (Agilent Technologies) with
DAD detector and a Luna C18(2) 100ꢃ column (5 mm, 4.6ꢄ
100 mm; Phenomenex, Torrence, CA) was used for reaction and
purity control. Preparative RP-HPLC was performed with an 1100
HPLC system with DAD detector (Agilent Technologies) and an
Figure 2. A) Antibacterial activity of feglymycin alanine scan peptides (MIC,
minimum inhibitory concentration: no visible growth after incubation at
378C for 20 h). IC50 values for the inhibition of B) MurA and C) MurC (IC50,
inhibitor concentration that decreases enzyme activity by 50%).
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 630

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
ODS-5 ST RP-C18 column (Grom-Sil 120, 10 mm, 20ꢄ250 mm; All-
tech Grom, Worms, Germany). Size exclusion chromatography was
performed with Sephadex LH20 (GE Healthcare) packed into a
glass column (ECO25 999 VOE, 25ꢄ1000 mm; Kronlab/YMC
Europe, Dinslaken, Germany). Separations were carried out on an
dissolved in dry DMF (10 mL per mmol amino acid or peptide) to-
gether with the N-terminal-protected coupling partner (1.0 equiv
amino acid or peptide). After cooling to 08C, NaHCO3 (2.0 equiv)
and DEPBT (2.0 equiv) were added. The mixture was stirred for 1 h
at this temperature and for a further 24 h at RT. The turbid reaction
mixture was diluted with water (50 mL per mmol amino acid or
peptide) and extracted with EtOAc (4ꢄ30 mL per mmol amino
acid or peptide). The combined organic layers were successively
washed with water (2ꢄ), saturated aqueous NaHCO3 (3ꢄ), and
brine (1ꢄ), dried over Na2SO4, and filtered. After removal of vola-
tiles under reduced pressure the residue was purified by flash chro-
matography on silica gel. Because of the low solubility of protect-
ed hexa- and heptapeptides in EtOAc, a modified procedure was
use for work-up of the corresponding reaction mixtures. After addi-
tion of an excess of water, the precipitate, which contained most
of the desired peptide, was centrifuged. Then, the aqueous super-
natant was extracted with EtOAc as described above. After the
washing step the organic layer was dried over Na2SO4 and com-
bined with the pellet of the centrifugation step dissolved in MeOH.
The solvent was subsequently removed by rotary evaporation and
the resulting crude product was purified by flash chromatography
on silica gel.
1
ꢅKTApurifier 10 FPLC system (GE Healthcare). H and 13C NMR spec-
tra were recorded with an Avance 400 MHz or a DRX 500 MHz
NMR spectrometer (Bruker). 2D NMR experiments were performed
on a DRX 500 MHz NMR spectrometer. EI-MS and EI-HRMS spectra
were recorded on a Finnigan MAT 95 S (Thermo Scientific). HPLC
ESI-MS and ESI-HRMS measurements were performed with an Orbi-
trap LTQ XL (Thermo Scientific) in combination with a 1200 HPLC
system (Agilent Technologies) and Hypersil-100 C18 column (5 mm,
3ꢄ50 mm; Thermo Scientific). HPLC ESI-MS/MS spectra were ob-
tained with a Qtrap 2000 mass spectrometer (Applied Biosystems)
in combination with an 1100 HPLC system (Agilent Technologies)
with a Luna 3u C18(2) column (100 ꢃ, 3 mm, 1ꢄ50 mm; Phenom-
enex). HPLC ESI-MS/MS spectra for all deprotected feglymycin de-
rivatives were recorded on an ESI-Triple-Quadrupole mass spec-
trometer 6460 (Agilent Technologies) in combination with a 1290
Infinity LC system (column: Eclipse Plus C18 1.8 mm, 2.1ꢄ50 mm;
Agilent Technologies). IR spectra were recorded on a Nicolet
Magna-IR 750 FTIR spectrometer (Thermo Scientific) or a Nicolet
Avatar 360 E.S.P. FTIR spectrometer (Thermo Scientific). CD spectra
were recorded with a J715 CD spectrometer (JASCO Research). Op-
tical rotations were determined with a P-2000 digital polarimeter
(JASCO Research).
General procedure for fragment couplings with DEBPT for the
synthesis of N- and C-terminal-protected tridecapeptides: Direct-
ly before the fragment condensation step, the C-terminal coupling
partner (hexapeptide) was N-Boc-deprotected. Therefore, the hexa-
peptide was suspended in CH2Cl2 (16 mL per mmol hexapeptide)
and treated with Et3SiH (15.0 equiv) and TFA (4 mL per mmol hexa-
peptide) under Ar. After 30 min at RT, the deprotected peptide tri-
fluoroacetate was precipitated by addition of Et2O (200 mL per
mmol hexapeptide), centrifuged and dried under vacuum for 1 h.
The resulting colorless solid was dissolved in dry DMF (10 mL per
mmol hexapeptide) at 08C under Ar together with the N-terminal-
protected heptapeptide (0.9 equiv), NaHCO3 (5.0 equiv), and DEPBT
(2.5 equiv). After 1 h at 08C and 40 h at RT, the reaction mixture
was diluted with water. The precipitated product was centrifuged,
washed once with water and dried under reduced pressure. The
residue was dissolved in MeOH (1 mL per 100 mg of crude prod-
uct) and purified by Sephadex LH20 size-exclusion chromatogra-
phy. The purest fractions were pooled and concentrated by rotary
evaporation. The obtained protected tridecapeptide was further
purified by reversed-phase preparative HPLC.
General procedure for peptide couplings with EDC/HOAt:
NaHCO3 (3.0 equiv), HOAt (1.1 equiv) and EDC·HCl (1.1 equiv) were
added to a solution of the N- and C-terminal-protected amino acid
(ratio 1:1) in dry DMF (5 mL per mmol amino acid) at 08C. The mix-
ture was stirred at this temperature for 2 h. After a further 24 h at
RT, the reaction mixture was diluted with water (20 mL per mmol
amino acid) and extracted with EtOAc (3ꢄ20 mL per mmol amino
acid). Afterwards, the combined organic phases were washed with
saturated aqueous NaHCO3 (3ꢄ), aqueous KHSO4 (5%, 3ꢄ) and
brine (1ꢄ), dried over Na2SO4, and filtered. The solvent was re-
moved in vacuum, and the residue was purified by flash chroma-
tography on silica gel.
General procedure for peptide couplings with TBTU: DIPEA
(3.0 equiv) was added to a solution of the N- and C-terminal-pro-
tected amino acid (ratio 1:1) in dry CH2Cl2 (10 mL per mmol amino
acid). After stirring for 10 min, TBTU (2.0 equiv) was added to the
solution. After 24 h at RT, the mixture was reduced to one fourth
of the original volume under reduced pressure. The residue was di-
luted with water and extracted with EtOAc (3ꢄ20 mL per mmol
amino acid). Then, the combined organic phases were washed
with saturated aqueous NaHCO3 (3ꢄ), HCl (1n, 3ꢄ) and brine (1ꢄ),
dried over Na2SO4, and filtered. The volatiles were removed by
rotary evaporation, and the crude product was purified by flash
chromatography on silica gel.
General procedure for cleavage of benzylic protecting groups of
dipeptides by catalytical hydrogenolysis: Pd/C (10%, 200 mg per
mmol peptide) was added to a solution of the protected peptide
in THF (10 mL per mmol peptide), and the flask was successively
flushed with Ar and H2. The reaction progress was monitored with
analytical TLC. After complete consumption of the starting materi-
al, the reaction mixture was filtered (0.45 mm Rotilabo PTFE syringe
filter; Carl Roth), and the filter was washed thoroughly with MeOH.
After removing the solvent under reduced pressure, the resulting
solid was dissolved in a small amount of MeOH, treated with
water, and lyophilized. Unless otherwise stated, the corresponding
deprotected peptide was obtained in quantitative yield.
General procedure for peptide couplings with DEPBT: If the C-
terminal amino acid or peptide was still N-Boc-protected, the Boc-
group was removed directly before peptide coupling. Therefore,
the amino acid or peptide was treated with 4n HCl/dioxane
(10 mL per mmol starting material) at RT under Ar, and stirred until
complete conversion was detected by analytical TLC. Afterwards,
the solvent was removed under vacuum, and the residue was
taken up in Et2O and evaporated again. This procedure was repeat-
ed twice and the resulting solid was dried under reduced pressure
for a further 30 min. The corresponding amino acid or peptide hy-
drochloride was obtained in quantitative yield, and subsequently
General procedure for cleavage of benzylic protecting groups of
tridecapeptides by catalytic hydrogenolysis: Water (1–2 drops)
and Pd/C (10%; 0.5 mg per mg of peptide) were added to a solu-
tion of the N- and C-terminal-protected tridecapeptide in MeOH
(0.1 mL per mg of peptide). The flask was flushed with, successive-
ly, Ar and H2. After 8 h at RT, the reaction mixture was filtered
(0.45 mm Rotilabo PTFE syringe filter; Carl Roth) and the filter was
washed with MeOH. The volatiles were removed by rotary evapora-
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 631

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
[4] a) B. Cavalleri, H. Pagani, G. Volpe, E. Selva, F. Parenti, J. Antibiot. 1984,
37, 309–317; b) R. Pallanza, M. Berti, R. Scotti, E. Randisi, V. Arioli, J. Anti-
biot. 1984, 37, 318–324; c) D. G. McCafferty, P. Cudic, B. A. Frankel, S.
Barkallah, R. G. Kruger, W. Li, Pept. Sci. 2002, 66, 261–284; d) S. Walker,
L. Chen, Y. Hu, Y. Rew, D. Shin, D. L. Boger, Chem. Rev. 2005, 105, 449–
475.
[5] a) J. Schimana, K. Gebhardt, A. Hçltzel, D. G. Schmid, R. Sꢁssmuth, J.
Mꢁller, R. Pukall, H.-P. Fiedler, J. Antibiot. 2002, 55, 565–570; b) A. Hçlt-
zel, D. G. Schmid, G. J. Nicholson, S. Stevanovic, J. Schimana, K. Geb-
hardt, H.-P. Fiedler, G. Jung, J. Antibiot. 2002, 55, 571–577; c) P. Kulan-
thaivel, A. J. Kreuzman, M. A. Strege, M. D. Belvo, T. A. Smitka, M. Clem-
ens, J. R. Swartling, K. L. Minton, F. Zheng, E. L. Angleton, D. Mullen,
L. N. Jungheim, V. J. Klimkowski, T. I. Nicas, R. C. Thompson, S.-B. Peng, J.
Biol. Chem. 2004, 279, 36250–36258; d) P. A. Smith, T. C. Roberts, F. E.
Romesberg, Chem. Biol. 2010, 17, 1223–1231; e) J. Liu, C. Luo, P. A.
Smith, J. K. Chin, M. G. P. Page, M. Paetzel, F. E. Romesberg, J. Am. Chem.
Soc. 2011, 133, 17869–17877.
tion. The obtained product was dissolved in a minimum quantity
of MeOH, treated with water, and lyophilized. Unless otherwise
stated, the corresponding deprotected tridecapeptide was ob-
tained in quantitative yield.
General procedure for methyl ester cleavage with trimethyltin
hydroxide: Peptide methyl ester and Me3SnOH (20.0 equiv) were
suspended in 1,2-dichloroethane (90 mL per mmol peptide) under
Ar and heated to reflux (858C). After a reaction time of 8 h, the sol-
vent was removed under reduced pressure, and the residue was
suspended in saturated aqueous NaHCO3 (100 mL per mmol pep-
tide). Then, the slightly turbid solution was acidified to pH 4–5
with solid citric acid. After extraction with EtOAc (4ꢄ50 mL per
mmol peptide), the combined organic layers were washed with
aqueous citric acid (10%, 1ꢄ), water (2ꢄ), and brine (1ꢄ), dried
over Na2SO4, and filtered. The volatiles were removed by rotary
evaporation. The obtained product was dissolved in a small quanti-
ty of MeOH, treated with water, and lyophilized.
[6] a) J. Aszodi, D. Isert, M. Knauf, D. Le Beller, G. Riess, W. Stahl, L. Vꢆrtesy,
J. Wink (Hoechst AG), EP 0 848 064 A1, 1999; b) L. Vꢆrtesy, W. Aretz, M.
Knauf, A. Markus, M. Vogel, J. Wink, J. Antibiot. 1999, 52, 374–382.
[7] G. Bunkꢇczi, L. Vꢆrtesy, G. M. Sheldrick, Angew. Chem. 2005, 117, 1364–
1366; Angew. Chem. Int. Ed. 2005, 44, 1340–1342.
Antimicrobial testing: All synthesized feglymycin derivatives were
tested for their antimicrobial activity in collaboration with Sanofi–
Aventis. All experiments were performed according to the guide-
lines of the Clinical and Laboratory Standards Institute (CLSI).[23]
[8] B. M. Burkhart, N. Li, D. A. Langs, W. A. Pangborn, W. L. Duax, Proc. Natl.
Acad. Sci. USA 1998, 95, 12950–12955.
[9] T. Hamada, S. Matsunaga, M. Fujiwara, K. Fujita, H. Hirota, R. Schmucki,
P. Gꢁntert, N. Fusetani, J. Am. Chem. Soc. 2010, 132, 12941–12945.
[10] F. Dettner, A. Hꢀnchen, D. Schols, L. Toti, A. Nußer, R. D. Sꢁssmuth,
Angew. Chem. 2009, 121, 1888–1893; Angew. Chem. Int. Ed. 2009, 48,
1856–1861.
Inhibition of the enzymes MurA and MurC: The inhibition of the
enzymes MurA and MurC from E. coli was determined in coupling
assays as described previously.[13]
[11] G. Fꢆrir, A. Hꢀnchen, K. O. FranÅois, B. Hoorelbeke, D. Huskens, F. Dett-
ner, R. D. Sꢁssmuth, D. Schols, Virology 2012, 433, 308–319.
[12] T. Schneider, H.-G. Sahl, Int. J. Med. Microbiol. 2010, 300, 161–169.
[13] S. Rausch, A. Hꢀnchen, A. Denisiuk, M. Lçhken, T. Schneider, R. D. Sꢁss-
muth, ChemBioChem 2011, 12, 1171–1173.
Acknowledgements
This research was supported by grants from the Deutsche For-
schungsgemeinschaft (DFG SU239/9–1) and the Cluster of Excel-
lence “Unifying Concepts in Catalysis”, coordinated by the Techni-
sche Universitꢀt Berlin, as well as a fellowship to A.H. by the Stif-
tung Stipendienfonds of the VCI.
[14] a) F. Reck, S. Marmor, S. Fisher, M. A. Wuonola, Bioorg. Med. Chem. Lett.
ˇ
ˇ
2001, 11, 1451–1454; b) R. Frlan, A. Kovac, D. Blanot, S. Gobec, S. Pecar,
A. Obreza, Molecules 2008, 13, 11–30.
[15] a) C. J. Leitheiser, K. L. Smith, M. J. Rishel, S. Hashimoto, K. Konishi, C. J.
Thomas, C. Li, M. M. McCormick, S. M. Hecht, J. Am. Chem. Soc. 2003,
125, 8218–8227; b) J. Nam, D. Shin, Y. Rew, D. L. Boger, J. Am. Chem.
Soc. 2007, 129, 8747–8755; c) X. Fang, J. Nam, D. Shin, Y. Rew, D. L.
Boger, S. Walker, Bioorg. Med. Chem. Lett. 2009, 19, 6189–6191.
[16] Y.-h. Ye, H. Li, X. Jiang, Pept. Sci. 2005, 80, 172–178.
[17] R. Kaul, S. Surprenant, W. D. Lubell, J. Org. Chem. 2005, 70, 3838–3844.
[18] G. Han, M. Tamaki, V. J. Hruby, J. Peptide Res. 2001, 58, 338–341.
[19] K. C. Nicolaou, A. A. Estrada, M. Zak, S. H. Lee, B. S. Safina, Angew. Chem.
2005, 117, 1402–1406; Angew. Chem. Int. Ed. 2005, 44, 1378–1382.
[20] P. E. Reynolds, E. A. Somner, Drugs Exp. Clin. Res. 1990, 16, 385–389.
[21] a) M.-C. Lo, H. Men, A. Branstrom, J. Helm, N. Yao, R. Goldman, S.
Walker, J. Am. Chem. Soc. 2000, 122, 3540–3541; b) Y. Hu, J. S. Helm, L.
Chen, X.-Y. Ye, S. Walker, J. Am. Chem. Soc. 2003, 125, 8736–8737.
[22] X. Fang, K. Tiyanont, Y. Zhang, J. Wanner, D. Boger, S. Walker, Mol. Bio-
Syst. 2006, 2, 69–76.
Keywords: alanine scan · antibiotics · enzyme inhibitors ·
peptides · structure–activity relationships
[1] a) R. E. W. Hancock, D. S. Chapple, Antimicrob. Agents Chemother. 1999,
43, 1317–1323; b) F. von Nussbaum, M. Brands, B. Hinzen, S. Weigand,
D. Hꢀbich, Angew. Chem. 2006, 118, 5194–5254; Angew. Chem. Int. Ed.
2006, 45, 5072–5129; c) K. C. Nicolaou, J. S. Chen, D. J. Edmonds, A. A.
Estrada, Angew. Chem. 2009, 121, 670–732; Angew. Chem. Int. Ed. 2009,
48, 660–719; d) E. M. Nolan, C. T. Walsh, ChemBioChem 2009, 10, 34–
53.
[2] a) D. Schwarzer, R. Finking, M. A. Marahiel, Nat. Prod. Rep. 2003, 20,
275–287; b) S. A. Sieber, M. A. Marahiel, Chem. Rev. 2005, 105, 715–738;
c) D. Bischoff, S. Pelzer, B. Bister, G. J. Nicholson, M. Schirle, S. Stockert,
W. Wohlleben, G. Jung, R. D. Sꢁssmuth, Angew. Chem. 2001, 113, 4824–
4827; Angew. Chem. Int. Ed. 2001, 40, 4688–4691.
[23] a) “Methods for dilution antimicrobial susceptibility tests for bacteria
that grow aerobically”; approved standard 7th ed., M07-A7 Vol. 26
No. 2, Clinical and Laboratory Standards Institute (CLSI); b) I. Wiegand,
K. Hilpert, R. E. W. Hancock, Nat. Protoc. 2008, 3, 163–175.
[3] a) K. C. Nicolaou, C. N. C. Boddy, S. Brꢀse, N. Winssinger, Angew. Chem.
1999, 111, 2230–2287; Angew. Chem. Int. Ed. 1999, 38, 2096–2152; b) F.
Wolter, S. Schoof, R. D. Sꢁssmuth, Top. Curr. Chem. 2007, 267, 143–185.
Received: January 21, 2013
Published online on February 27, 2013
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2013, 14, 625 – 632 632