Diversity-oriented synthesis of cyclic acyldepsipeptides leads to the discovery of a potent antibacterial agent

Article abstract of DOI:10.1016/j.bmc.2010.08.032

A class of cyclic acyldepsipeptide antibiotics collectively known as the enopeptins has recently attracted much attention because of their activity against multidrug-resistant bacteria, including methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus faecalis. These antibiotics are further distinguished by their novel mechanism of action in which they bind and deregulate the tightly controlled activity of the cytoplasmic protease ClpP. Although the natural products have poor pharmacological properties, a synthetic derivative called acyldepsipeptide 4 (ADEP 4) showed remarkable antibacterial activity both in vitro and in mouse models of bacterial infections. A novel route to the ADEP 4 peptidolactone core structure, featuring the Joullié-Ugi three-component reaction, was developed. This multicomponent reaction and a related multicomponent reaction, the Ugi four-component reaction, were used to prepare analogs that were designed using the principles of conformational analysis. These cyclic acyldepsipeptides were tested for their activity against drug-resistant, clinical isolates of Staphylococci and Enterococci. One ADEP 4 analog in which the pipecolate was replaced by 4-methyl pipecolate exhibited in vitro antibacterial activity against Enterococci that was fourfold higher than the parent compound.

Full text of DOI:10.1016/j.bmc.2010.08.032

Bioorganic & Medicinal Chemistry 18 (2010) 7193–7202
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Diversity-oriented synthesis of cyclic acyldepsipeptides leads to the discovery
of a potent antibacterial agent
Aaron M. Socha a, Nicholas Y. Tan a, Kerry L. LaPlante b,c, Jason K. Sello a,
⇑
a Department of Chemistry, Brown University, Providence, RI 02912, USA
b Pharmacy Practice, University of Rhode Island, Kingston, RI 02881, USA
c Infectious Disease Research Laboratory, Veterans Affairs Medical Center, Providence, RI 02908, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 July 2010
Revised 11 August 2010
Accepted 16 August 2010
Available online 19 August 2010
A class of cyclic acyldepsipeptide antibiotics collectively known as the enopeptins has recently attracted
much attention because of their activity against multidrug-resistant bacteria, including methicillin-resis-
tant Staphylococcus aureus and vancomycin-resistant Enterococcus faecalis. These antibiotics are further
distinguished by their novel mechanism of action in which they bind and deregulate the tightly con-
trolled activity of the cytoplasmic protease ClpP. Although the natural products have poor pharmacolog-
ical properties,
a synthetic derivative called acyldepsipeptide 4 (ADEP 4) showed remarkable
Keywords:
antibacterial activity both in vitro and in mouse models of bacterial infections. A novel route to the ADEP
4 peptidolactone core structure, featuring the Joullié-Ugi three-component reaction, was developed. This
multicomponent reaction and a related multicomponent reaction, the Ugi four-component reaction, were
used to prepare analogs that were designed using the principles of conformational analysis. These cyclic
acyldepsipeptides were tested for their activity against drug-resistant, clinical isolates of Staphylococci
and Enterococci. One ADEP 4 analog in which the pipecolate was replaced by 4-methyl pipecolate exhib-
ited in vitro antibacterial activity against Enterococci that was fourfold higher than the parent compound.
Ó 2010 Elsevier Ltd. All rights reserved.
Antibacterial agent
Cyclic acyldepsipeptide
Clinical pathogens
MRSA
VRE
Multicomponent reactions
Ugi
ClpP
1. Introduction
means to enhance their biological activity and/or pharmacological
properties.6
Infectious diseases are the third leading cause of death in devel-
oped countries and the second leading cause of death world-
wide.1,2 The efficacy of many antibacterial drugs has been
compromised by the emergence of drug-resistant, pathogenic
bacteria.3 The Infectious Disease Society of America has recently
outlined the deadly implications of a growing number of drug-
resistant pathogens.4 Methicillin-resistant Staphylococcus aureus
(MRSA), vancomycin-resistant Enterococci (VRE), and penicillin-
resistant Streptococcus epidermis are especially worrisome in clini-
cal settings. Unfortunately, the prevalence of multidrug-resistant
bacteria has made antibiotics of last resort, like vancomycin, the
first-line of therapy. The capacity of bacteria to commonly develop
resistance to virtually any antibacterial agent necessitates a con-
tinuous search for new drugs. There is much evidence in the liter-
ature that natural products derived from microorganisms will
continue to be a source of novel antibacterial drugs.5 Although nat-
ural products often have chemical properties that are incompatible
with chemotherapy, it is possible to use medicinal chemistry as a
A recent case wherein medicinal chemistry was used to improve
the activity of a natural product was that of the enopeptins.7,8 The
parent compounds were isolated from the soil dwelling bacterium
Streptomyces sp. RK-1051 and are defined by a 16-membered pept-
idolactone consisting of five L-amino acids to which a lipophilic
polyene side chain is appended (Fig. 1).9 A group of closely related
compounds, called A54556 A and B, were isolated from Streptomy-
ces hawaiienesis by a research group at Eli Lilly (Fig. 1).10 The eno-
peptins attracted attention because of their potent activity
against drug-resistant bacterial pathogens, including MRSA and
VRE.11 The apparent lack of cross-resistance for all antibacterial
agents on the market or those in clinical development has been as-
cribed to a peculiar mechanism of action.11 The enopeptin antibiot-
ics inhibit cell division and cause cell death by binding and
deregulating the activity of the casein lytic protease (ClpP).11 Under
normal conditions, this fourteen-subunit protease selectively de-
grades proteins through a physical and functional association with
accessory ATPases that recognize and unfold its substrates.12–15 In
the presence of the enopeptins, ClpP indiscriminately degrades
folded cytoplasmic proteins, which ultimately causes cell death.11
Recent structural studies indicate that the enopeptins bind the ClpP
core structure and cause it to undergo a conformational change that
exposes the enzymatic active sites of its subunits.16
⇑
Corresponding author. Tel.: +1 401 863 1194.
E-mail address: jason_sello@brown.edu (J.K. Sello).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.08.032

7194
A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
known as acyldepsipeptide 4 (ADEP 4, Fig. 1), exhibited remarkable
R
activity both in vitro and in vivo.7 As was the case for the parent
enopeptins, no cross-resistance to ADEP 4 was observed and the
compound was active against multidrug-resistant pathogens. The
minimum inhibitory concentration (MIC) and in vivo antibacterial
activity of ADEP 4 equaled or even surpassed that of established
antibacterial drugs. Remarkably, mice infected with a lethal inocu-
lum of Enterococcus faecalis had a 100% survival rate when treated
with a single dose of ADEP 4 at 0.5 mg/kg.7 The development of
ADEP 4 clearly indicates the power of medicinal chemistry in
enhancing the properties of natural products.
O
Enopeptin A (R = CH3)
Enopeptin B (R = H)
N
O
N
O
O
O
HO
O
NH
O
N
H
O
HN
N
H
N
O
O
Given the promise of ADEP 4 as an antibacterial drug, we sought
to develop an efficient synthetic route to the peptidolactone core of
this molecule that also allowed for the preparation of other con-
formationally constricted enopeptin derivatives. In the interest of
expediency, we planned to capitalize on the convergent scheme
used in the syntheses of ADEP 4 (and of enopeptins A and B) in
which a tripeptide fragment and an ester fragment are coupled, cy-
clized and acylated (Scheme 1).7,17 Our objective was to develop an
efficient and diversity-oriented synthesis of the tripeptide precur-
sor of ADEP 4 (Ala-Pip-Pro). The published chemical synthesis of
this tripeptide fragment uses standard peptide chemistry and as
such is limited by three distinct steps for removal of protecting
groups, a challenging acylation of a secondary amide (pipecola-
mide), and a dependence on commercially-available amino acids.
An ideal synthesis of the tripeptide would be higher yielding,
greater in atom economy, and enable facile exchange of ADEP 4’s
pipecolate with other conformationally rigid amino acids for SAR
studies. We hypothesized that the tripeptide precursor of ADEP 4
and other conformationally restricted tripeptides could be pre-
pared in a single step via isocyanide-based multicomponent reac-
tions (Scheme 2).18,19 Pipecolate containing tripeptides
reminiscent of the ADEP 4 tripeptide precursor can be prepared
from six-membered cyclic imines, N-Boc-proline, and an isocya-
nide derived from alanine methyl ester in a Joullié-Ugi three com-
ponent reaction (Joullié-Ugi 3CR).20,21 Similarly, peptides with
O
F
F
N
ADEP4
O
N
O
O
O
O
NH
N
H
O
HN
N
O
Figure 1. Enopeptin natural products and the pharmacologically-enhanced deriv-
ative, ADEP 4.11
Although the enopeptin natural products have remarkable anti-
bacterial activity in vitro, their chemical lability and poor solubility
limit their efficacy in vivo.7 In an effort to improve the pharmaco-
logical properties of the enopeptins, a research group at Bayer Phar-
maceutical Research carried out an intensive structure–activity
relationship (SAR) study of the enopeptins.7,8 Their medicinal
chemistry program was guided by the observation that the confor-
mation of the peptidolactone core structure was constrained by two
transannular hydrogen bonds to the phenylalanine moiety of the
side chain. On this basis, it was anticipated that replacing the con-
stituents of the peptidolactone with conformationally restricted
amino acids would yield enopeptin derivatives with improved anti-
bacterial activity. Although most changes in the peptidolactone de-
creased antibacterial activity, replacement of N-methylalanine with
a rigid pipecolate moiety yielded a more potent analogue.7 Presum-
ably, rigidification of the peptidolactone lowers the entropic cost of
binding to the ClpP core.7 The stability and antibacterial activity of
these rigidified enopeptin analogs were further enhanced by
replacement of the exocyclic phenylalanine with 3,5-difluorophe-
nylalanine and by substitution of the polyunsaturated side chain
with a 2-heptenoyl moiety. The optimized enopeptin derivative,
conformationally restricted N-methyl,
a,a-disubstituted amino
acids can be prepared with an Ugi four-component reaction (Ugi
4CR)22 from methylamine, various ketones, N-Boc-proline, and an
isocyanide derived from alanine methyl ester.
We developed a diversity-oriented synthesis of the tripeptide
fragment featuring isocyanide-based multicomponent reactions
that enabled the preparation of eight enopeptin derivatives. The
antibacterial activity of these analogues against clinical isolates
of methicillin-sensitive S. aureus (MSSA), methicillin-resistant
O
F
N
O
O
O
O
O
NH
N
N
H
O
HN
N
O
OH
O
F
O
O
O
O
NH
N
Cbz
Boc
N
O
NH
HO
NH
HN
O
O
Phenacyl
Ester fragment
O
F-Phe-Heptenamide fragment
Tripeptide fragment
Scheme 1. Retrosynthesis of simplified ADEP 4 analog 10a.

A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
7195
R2
N
O
O
O
NH2
R5
R5
+
+
+
R1
R2
CN
N
H
R3
R1
OH
R4
R3
R4
O
O
R1
N
N
O
R5
+
+
CN
R1
OH
R5
O
N
H
Scheme 2. The Ugi (top) and Joullié-Ugi (bottom) multicomponent reactions.
S. aureus (MRSA), vancomycin-susceptible E. faecalis and vancomy-
cin-resistant E. faecium (VRE) was systematically assessed. Addi-
tionally, we report for the first time the minimum bactericidal
concentration (MBC) of any enopeptin derivatives. It is noteworthy
that a novel ADEP 4 derivative with 4-methyl pipecolate in place of
pipecolate has considerably enhanced activity against pathogenic
Enterococci.
revealed that no racemization of the amino acid occurred during
the formylation and dehydration.
We envisioned that the cyclic imine substrate of the Joullie-Ugi
reaction could be prepared by halogenation and dehydrohalogen-
ation of piperidine. N-chlorination of piperidine via in situ genera-
tion of tert-BuOCl gave the desired product in quantitative yield.24
Dehydrohalogenation of the N-chloropiperidine was effected by
sodium methoxide in methanol (Scheme 4).20 Due to its instabil-
ity,25,26 the crude dehydropiperidine was directly used in a Joul-
lié-Ugi 3CR with N-Boc-proline and the isocyanide derived from
alanine methyl ester (2). After stirring for five days at room tem-
perature, the expected tripeptide methyl ester, 3, was isolated in
76% yield (Scheme 4). Analytical HPLC analysis of saponified 3
(LiOH, 1:1 THF/H2O, quantitative yield) revealed that its diastereo-
meric ratio was 70:30. Based on the retention time of an authentic
standard [(S)-Boc-Pro-(S)-Pip-(S)-Ala-COOH], we determined that
the desired tripeptide was the minor diastereomer. The basis of
the diastereoselectivity of this reaction is not clear and is under
investigation. In spite of the undesirable stereoselectivity, the yield
and the atom economy of the one-pot Joullié-Ugi 3CR make it a via-
ble alternative to standard peptide chemistry for synthesis of the
ADEP 4 tripeptide precursor.
2. Results
2.1. Synthesis and elaboration of the ADEP 4 peptidolactone
core structure
Our initial objective was to prepare 10a, a more synthetically
accessible analogue of ADEP 4 in which 4-methylproline is re-
placed by proline and 3,5-difluorophenylalanine is replaced by 3-
fluorophenylalanine. In analogy to the published synthesis of ADEP
4, our plan was to prepare 10a in a convergent fashion. We antic-
ipated that the requisite tripeptide fragment could be synthesized
via a Joullié-Ugi 3CR. The substrates of the reaction were N-Boc
proline, dehydropiperidine, and an isocyanide derived from alanine
methyl ester. While N-Boc proline was commercially available, the
other two substrates had to be chemically synthesized. The isocy-
anide was prepared via the N-formylation and dehydration of ala-
nine methyl ester under racemization-free conditions described in
The diastereomeric mixture of tripeptide products were sapon-
ified and carried forward to prepare the desired peptidolactone
using the convergent synthetic route described in the literature
(Scheme 5).7,8 In contrast to the reported synthesis of ADEP 4,
the diastereomeric peptidolactones were chromatograpically sepa-
rated and acylated with N-2-heptenoyl, 3-fluorophenylalanine to
yield 10a and its diastereomer 10b (Scheme 5).
the literature.23 N-formyl-
from -alanine methyl ester hydrochloride according to a literature
L-alanine methyl ester was synthesized
L
procedure for the preparation of amino acid tert-butyl esters.24
Dicyclohexylcarbodiimide (DCC) activation of formic acid as the
O-formylisourea intermediate followed by the addition of the ami-
no acid ester provided the desired N-formyl product 1 in good
yield. Dehydration of 1 to form the isocyanide (2) was effected
by triphosgene in the presence of N-methylmorpholine (NMM).23
Synthesis of the isocyanide was realized in 69% yield over two
steps (Scheme 3). Chiral GC–MS analysis of the isocyanide product
2.2. Diversity-oriented synthesis of cyclic acyldepsipeptides
with conformationally constrained amino acids
In addition to the efficiency with which an isocyanide-based
multicomponent reaction could be used to synthesize a tripeptide,
O
Triphosgene (0.35 eq)
HCOOH, DCC,
NMM, DMAP
DCM, 0oC to RT, O/N
72%
O
O
NMM (2 eq)
O
CN
H
N
H2N
DCM, -78 to -30oC, 3hrs
95%
H
O
O
O
1
2
Scheme 3. Preparation of an enantiomerically-pure isocyanide from
L
-alanine methyl ester.
H
1. tert-BuOH, NaOCl, AcOH,
MTBE, 0oC, 30mins, Quantitative
Cmpd. 2 and N-Boc-
(S)- Pro, MeOH, 5d
N
N
N
N
Boc
76%
O
O
2. NaOMe, MeOH, 2.5 hrs
O
N
H
O
3
Scheme 4. Synthesis of dehydropiperidine and its use in the Joullié-Ugi three component reaction.

7196
A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
Cbz
O
HN
O
Cbz
HN
OH
O
N
O
O
O
a, b
N
O
O
O
O
N
O
Boc
N
+
O
OH
81%
Boc
N
H
N
O
HN
N
H
O
O
3
c-e
23%
O
F
N
N
O
O
N
O
O
O
O
f, g
CBz
O
O
NH
N
O
NH
O
34%
N
H
O
H
N
HN
N
N
O
10a
Scheme 5. Reagents and conditions: (a) TPTU, HOBt, DCM, DIEA, 0 °C to rt, 18 h; (b) Zn, 90% HOAc in H20, rt, 30 min; (c) pentafluorophenol, EDC, DCM, À78 °C to rt, 18 h; (d)
3 N HCl in EtOAc, 0 °C, 3.5 h; (e) DCM/10% NaHCO3 (3:1) 0.1 mM, rt, 18 h; (f) NH4COOH, Pd/C, isopropanol, 85 °C microwave, 8 min; (g) N-2-heptenoyl-3-fluorophenylalanine
(9), TPTU, HOBt, DIEA, DMF, rt, 18 h.
a major consideration in the selection of this reaction type was its
applicability in the preparation of structurally diverse peptides.
Rather than preparing a random collection of tripeptides, we
sought to synthesize tripeptides that were more conformationally
constrained than the tripeptide precursor of ADEP 4. Our utiliza-
tion of conformationally constrained amino acids was an extension
of the findings of the Bayer research group wherein replacement of
the N-methylalanine moiety of the natural enopeptins with pipe-
colate yielded an enopeptin derivative with enhanced antibacterial
activity.
2 days. Analytical HPLC analysis of the saponified product revealed
that its diastereomeric ratio was 70:30. As the retention time of the
minor diastereomer was very similar to that of the authentic ADEP
4 tripeptide precursor, we deduced that it was the desired diaste-
reomer containing (2S,4R)-4-methylpipecolamide. The mixture of
diastereomeric peptides was carried forward in the aforemen-
tioned scheme for convergent synthesis of the peptidolactone core
structure. The diastereomeric peptidolactones were chromatograp-
ically separated and acylated with N-2-heptenoyl, 3-fluorophenyl-
alanine to yield the desired compound, 10c (Fig. 2).
Our first synthetic target was an analog of the ADEP 4 tripeptide
fragment with a substituent on the pipecolate moiety. Based on the
principles of conformational analysis, we predicted that strategic
placement of a substituent on the ring would increase the inver-
sion barrier of pipecolate.27 In particular, a substituent on the 4-po-
sition of pipecolate would be especially rigidifying since ring
inversion would result in a strongly disfavored 1,3-diaxial interac-
tion between ring substituents. To prepare such a compound, we
carried out a Joullié-Ugi 3CR with N-Boc-proline, the isocyanide de-
rived from alanine methyl ester (2), and dehydro 4-methyl piperi-
dine. The latter compound was prepared in an analogous manner
to dehydropiperidine. Interestingly, the dehydro-4-methylpiperi-
dine was a better substrate in the Joullié-Ugi 3CR, yielding the ex-
pected tripeptide methyl ester in 84% in a reaction time of only
Our other synthetic targets were tripeptides containing a,a-
disubstituted amino acids. These amino acids, while of limited
commercial availability, are known to limit the conformation of
peptides.28,29 We suspected that the desired tripeptides could be
prepared from methylamine, N-Boc-proline, the isocyanide derived
from L-alanine methyl ester (2), and various ketones in an Ugi 4CR.
To avoid mixtures of diastereomeric products, we used symmetric
ketones (i.e., 3-pentanone, cyclobutanone, cyclopentanone, and
cyclohexanone) as substrates (Scheme 6). The reactants were stir-
red in MeOH for a period of 1–5 days to optimize reaction condi-
tions. Products 5–8 were isolated in good to excellent yields. The
results are summarized in Table 1. Good yields of the tripeptides
were only observed with reaction times >4 days and reaction
concentrations P1.9 M (Increasing the reaction temperature of a
O
O
F
F
N
O
N
O
O
O
O
O
O
O
O
NH
O
NH
N
N
O
O
R
H
H
HN
HN
N
N
N
n( )
N
O
O
R
R
Pentapeptolide Macrocycle
Pentapeptolide Macrocycle
10e R = CH3, n = 0
10f R = H, n = 1
10g R = H, n = 2
10h R = H, n = 3
10a (2S) R = H
10b (2R) R =H
10c (2S) R =CH3
10d (2R) R= CH3
81%
78%
79%
77%
23%
25%
22%
26%
80%
78%
73%
81%
24%
21%
26%
25%
Figure 2. Chemical structures of cyclic acyldepsipeptides prepared and evaluated in this study. Shown below are the yields for reactions yielding the protected
pentapeptolide and the protected macrocycle precursors of compounds 10a–10h (see experimental Sections 5.2.6.1 and 5.2.6.2).

A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
7197
O
O
N
O
O
MeOH, 5d
75-93%
NH2
N
N
CN
OH
Boc
H
N
O
O
O
Boc
( )n
( )n
O
R
R
R
R
2
5-8
Scheme 6. Ugi reactions with symmetric ketones (R = H or CH3, n = 0–3).
Table 1
Optimized Ugi 4CR conditions for symmetric ketones
Table 3
Biological activity of synthetic acyldepsipeptides versus Enterococci (
l
g/mL)
VREd
Ugi product
R
n
Conc (M)
Temp (°C)
Time (d)
Yield (%)
EFa EFb VREc
5
6
7
7
8
CH3
H
H
H
H
0
1
2
2
3
1.90
1.90
0.65
1.90
1.90
25
25
25
25
25
5
5
5
5
5
75
93
62
78
90
MIC
MBC
MIC
MBC
MIC
MBC
MIC
MBC
10a
10c
0.16
0.04
1.25
0.08
0.04
0.02
0.04
0.04
0.08
0.04
0.08
0.08
0.04
0.04
0.04
0.04
a
b
c
Enterococcus faecalis (ATCC 29212).
Enterococcus faecalis (clinical isolate from rectal swab).
Vancomycin-resistant Enterococcus faecium (clinical isolate obtained from
tissue).
d
model reaction with cyclopentanone and methylamine to 50 °C did
not significantly improve the reaction yield). Using four different
ketones as substrates in the Ugi 4CR, we were able to prepare four
different tripeptide methyl esters (5–8). Those products were
saponified as previously described to yield the requisite tripep-
tides. The tripeptides prepared via the Ugi 4CR were used in the
aforementioned convergent synthesis to yield enopeptin deriva-
tives, 10e–h (Fig. 2).
Vancomycin-resistant Enterococcus faecalis (clinical isolate from rectal swab).
colate in place of pipecolate, had better activity than compound
10a against MSSA. Compound 10a was also evaluated against
Staphylococcus epidermis (ATCC 25984) and exhibited an MIC of
1.25 lg/mL and an MBC of 1.25 lg/mL.
Based on the activities of compounds 10a and 10c against the S.
aureus strains, they were selected for further evaluation against a
panel of pathogenic Enterococci (Table 3). Remarkably, compound
10c exhibited more potent activity against E. faecalis and vancomy-
cin-resistant E. faecium than compound 10a.
2.3. Antibacterial assays of the synthetic acyldepsipeptides
The naturally occurring enopeptins and the synthetic derivative
ADEP 4 are reported to have strong antibacterial activity.7 Broth
micro-dilution assays30,31 were used to determine the MICs of
the acyldepsipeptide derivatives 10a–h against both ATCC strains
and clinical isolates of S. epidermis, S. aureus, and Enterococci ob-
tained from patients of the Veterans Affairs Medical Center in Prov-
idence, Rhode Island (Table 2). Clinical isolates were included to
gain a better indication of the potential of these promising antibac-
terial agents. In the interest of assessing the cell-killing ability of
these compounds, the minimal bactericidal concentrations (MBC)
of the compounds was also determined against the aformentioned
strains. The MIC was defined as the lowest concentration of an
antimicrobial agent visually inhibiting more than 99% of the colo-
nies. The MBC was defined as the lowest antibiotic concentration
to show no growth (99.9% kill) after 24 h of incubation.
3. Discussion
The medicinal properties of natural products can be enhanced
by medicinal chemistry.6 Optimization of natural product struc-
tures can be facilitated through the design of diversity-oriented
chemical syntheses.32 In this regard, isocyanide-based multicom-
ponent reactions are ideally suited for the preparation of peptide
antibiotics and closely related analogs. Specifically, the Ugi 4CR
and the related Joullié-Ugi 3CR can be used to construct peptides
consisting of non-proteinogenic and other non-natural amino
acids.18–21 Here, we demonstrate the utility of these reactions in
synthesizing a promising class of acyldepsipeptides known as the
enopeptins. In terms of atom economy and versatility, multicom-
ponent reactions offer advantages over standard peptide
synthesis.18,19
Compound 10a, an analog of ADEP 4, exhibited the expected
antibacterial activity against drug-resistant, pathogenic S. aureus
strains. All synthetic acyldepsipeptides in which the pipecolate
A comparison of the in vitro activities of the acyldepsipeptide
derivatives affords two major structure–activity relationships.
moiety was replaced by a,a-disubstituted amino acids lacked anti-
First, placing an additional substituent on the
a-carbon of the N-
bacterial activity. In contrast, compound 10c, which 4-methylpipe-
methylalanine moiety of the enopeptin peptidolactone core struc-
ture abolishes antibacterial activity. Indeed, derivatives containing
a
,
a
-disubstituted amino acids in the macrocycle were completely
Table 2
inactive. Second, the S-configuration of the pipecolate -carbon is
a
Biological activity of synthetic acyldepsipeptides versus S. aureus (
lg/mL)
essential for the bioactivity of ADEP 4 analogues. Specifically,
the pipecolate derivatives containing (R)-pipecolic acid and
(R)-4-methylpipecolic acid were inactive. Taken together, these
observations suggest that the ClpP protease can only accommodate
MSSA MRSA
MIC
MBC
MIC
MBC
10a
10b
10c
10d
10e
10f
1.25
>312
0.6
>156
78.1
312
1.25
>312
0.6
1.25
>312
0.6
1.25
>312
2.4
.156
>312
>312
>312
>312
acyldepsipeptides with mono-substituted, L-amino acids in place
of the N-methyl alanine moiety in the parent compound. In
recently published crystal structures of acyldepsipeptides bound
to ClpP, either the N-methyl alanine of enopeptin B or the pipeco-
late moiety of ADEP 4 are in close proximity to Tyr112 and Leu189
of ClpP.16 It stands to reason that modification of the peptidolac-
tone with bulkier amino acids at these positions would result in
unfavorable steric interactions with the aforementioned residues
>156
>312
>312
>312
>312
>156
>312
>312
>312
>312
10g
10h
39.1
39.1
MSSA = Methicillin-susceptible S. aureus (ATCC 35556).
MRSA = Methicillin-resistant S. aureus (clinical isolate from blood).

7198
A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
of ClpP. Alternatively, modification of the peptidolactone core
structure could result in conformational constraints that preclude
the formation of key hydrogen bonds to Ser60 and Tyr62 of ClpP.16
It is not clear which of these two possibilities explain the decreased
13C). All spectra were referenced to residual solvent signals in
CDCl3 (7.24 ppm for 1H, 77.0 ppm for 13C.)
5.2. Synthesis
activity of our analogs (10e–h) with a,a-disubstituted amino acids.
In any case, the derivative of 10a with 4-methylpipecolate in place
of pipecolate was highly active. This compound, 10c, had fourfold
lower MIC values against E. faecalis than 10a. This observation indi-
cates that increasing the conformational rigidity of the macrocycle
does indeed lead to improved antibacterial activity in vitro. It is
highly likely that the replacement of the pipecolate of ADEP 4 with
4-methyl pipecolate will enhance its antibacterial activity. It re-
mains to be seen if such changes will improve pharmacological
activity in vivo.
The molecular mechanism of action of the enopeptins and
derivatives thereof has been described in great detail.7,11,16 From
studies of Bacillus subtilis (a non-pathogenic model bacterium), it
was evident that an early effect of these antibacterial compounds
was inhibition of cell division. The nominal differences between
the MICs and MBCs that we observed are indicative of bactericidal
activity.33 This is especially meaningful with regard to VRE, since
current treatments involve mostly bacteriostatic drugs, and few
compounds have been found that are bactericidal against VRE.34
Given this observation, the enopeptin analogs hold great promise
for use in the treatment of VRE infections.
5.2.1. Preparation of substrates for multicomponent reactions
5.2.1.1. N-Formyl-L-alanine methyl ester (1). Dicyclohexylcar-
bodiimide (DCC) (10 g, 48.5 mmol) was dissolved in 50 mL of
DCM and cooled to 0 °C for 5 min. 96% formic acid (2.6 mL,
69 mmol) was added slowly. In a separate flask, L-alanine methyl
ester HCl salt (5 g, 36 mmol) was dissolved in 25 mL dichlorometh-
ane (DCM) with 7 mL N-methylmorpholine (NMM) and 1 g of
DMAP. The contents of this flask were added to the DCC-formic
acid mixture at 0 °C and the reaction was allowed to warm to room
temperature overnight with stirring. The majority of the solvent
was removed in vacuo until a viscous slurry was obtained. This
slurry was vacuum filtered through silica gel and the residue was
washed with 25 mL DCM to ensure complete elution of product.
The filtrate was concentrated in vacuo and loaded on
a
75 Â 50 mm Si gel resin bed and eluted with 4:1 EtOAc/hexanes.
Yield: 3.4 g (72.1%), clear oil, Rf = 0.6 (4:1 EtOAc/hexanes), stains
white with p-anisaldehyde. 1H NMR (CDCl3) 8.14 (1H, s), 6.43
(1H, br s), 4.54 (1H, m), 3.72 (3H, s), 1.39 (3H, d, J = 7 Hz). 13C
NMR (CDCl3) 173.0, 160.5, 52.6, 46.7, 18.4. MS (EI, m/z): 131.0
[M]+ for C4H9NO2.
5.2.1.2. Isocyanide derived from alanine methyl ester (2).
N-formyl-L-alanine methyl ester (1) (2.2 g, 16.8 mmol) was dis-
4. Conclusion
solved in 50 mL anhydrous DCM and the solution was cooled to
À78 °C. In a separate, dry, pear shaped flask was added triphosgene
(1.7 g, 5.7 mmol) and 7 mL dry DCM. The triphosgene solution was
added slowly to the N-formyl amino acid ester. Then, 3.6 mL
(33 mmol) NMM added slowly over 10 min. The reaction was stir-
red for 1.5 h at À78 °C then quenched with 22 mL deionized water
with a vent needle in the septum. The stirring solution was allowed
to warm to room temperature. The resulting mixture was parti-
tioned and the water layer washed with an additional 50 mL DCM.
The organic fraction was collected, dried over Na2SO4, and solvent
removed in vacuo. Crude product was loaded on a 75 Â 50 mm Si
gel resin bed and eluted with 4:1 hexanes/EtOAc (Rf = 0.65, Iodine/
Si stain). Pale yellow oil, Yield: 1.8 g (95%). 1H NMR (CDCl3) 4.32
(1H, q), 3.81 (3H, s), 1.63 (3H, d, J = 8 Hz). 13C NMR (CDCl3) 167.3,
159.0, 60.0, 51.3, 19.3. MS (EI, m/z): 113.0 [M]+ for C5H7NO2.
A number of highly conformationally restricted enopeptin ana-
logs were synthesized via isocyanide-based multicomponent reac-
tions. Two strategies were explored: one involving the
replacement of the N-methyl alanine moiety in the peptidolactone
with
a,a-disubstituted amino acids, and the other involving the
replacement of this residue with a substituted pipecolic acid. The
latter strategy was found to be successful, yielding a derivative that
has better in vitro activity than the compound with pipecolic acid.
This highlights the utility of conformational analysis in the medic-
inal chemistry optimization of the enopeptin antibiotics. The opti-
mized compound shows especially promising activity against VRE
in vitro, and additional assays are being carried out to further
investigate this observation.
5. Experimental
5.1. General
5.2.1.3. N-Chloropiperidine. 5.8 mmol (0.55 mL) of tert-BuOH,
11.7 mmol (1.39 mL) piperidine were added to 30 mL MTBE and
stirred at 0 °C. 11.8 mmol (23.5 mL) cleaning bleach and 11.8 mmol
(0.68 mL) of acetic acid were added together slowly and reaction
mixture was stirred at 0 °C for 30 min. Reaction mixture was then
quenched with 15 mL of water, extracted with 20 mL MTBE, and
the organic layer was washed with brine and dried with Na2SO4.
Evaporation of solvent under reduced pressure gave the desired
product 1.4 g (10.5 mmol, 89% yield) as a pale yellow oil. Rf = 0.9
(3:1 EtOAc/hexanes) MS (EI, m/z): 119.1 [M]+ for C5H10NCl.
All amino acids and peptide coupling reagents were purchased
from NovaBiochem. Additional chemicals were purchased from
Sigma-Aldrich unless otherwise stated. A Biotage Initiator micro-
wave reactor was used. GC–MS analysis was performed on a Hew-
lett Packard 5971A GC–MS system using an HP5-MS column and
helium as the carrier gas. Splitless injections of 1 lL were made
using an initial temperature of 60 °C, holding 2 min, then 20 °C/
min ramp to 280 °C, and holding 2 min. Chiral GC–MS was per-
formed on a Hewlett Packard G1800C instrument equipped with
a Varian CP-Chirasil-DEX 25 m  0.32 mm column. Low-resolution
analytical LC–MS was performed in the positive ion mode on a
Thermo LCQ Deca XP MAX high sensitivity MSn ion trap mass spec-
trometer with a Shimadzu HPLC system using a Waters X-Terra MS
5.2.1.4. Dehydropiperidine. 10.5 mmol (1.4 g) of N-chloropiperi-
dine was dissolved in 3.5 mL (10.91 mmol) of 25% by wt NaOMe in
MeOH. The reaction was stirred for 10 min under N2, after which a
white precipitate of NaCl was formed and the reaction was
complete.
C
18 column (2.5
l
m, 2.1 Â 50 mm). The method consisted of 5–95%
MeCN in H2O + 0.1% TFA over 15 min. The flow rate was 0.2 mL/
min. High-resolution mass analyses were performed on a JOEL
JMS-600H double focusing magnetic sector mass spectrometer
using FAB ionization. NMR analyses were performed on a Bruker
Avance Ultrashield spectrometer (400 MHz for 1H, 100 MHz for
5.2.2. General procedure for the Joullié-Ugi 3CR
The crude cyclic imines, 4 mL dry MeOH and N-Boc-L-proline
(1.95 g, 8.5 mmol) were added directly to the reaction mixture,
and were stirred for 15 min. The isocyanide 2 (780 mg, 6.9 mmol)
was added, and the reaction stirred for 2–5 days. Solvent was

A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
7199
removed under vacuum and crude mixture was purified on a Si gel
column with 7:3 EtOAc/hexanes eluent.
3.45 (1H, m), 3.00 (3H, s), 2.52 (1H, m), 2.11–1.52 (12H, m) 1.43
(9H, s), 1.38 (3H, d). 13C NMR (CDCl3, mixture of rotamers) 175.4,
173.6, 171.0, 154.4, 153.5, 79.4, 73.4, 73.0, 60.2, 58.0, 57.9, 52.0,
50.5, 48.2, 46.9, 36.2, 35.5, 34.8, 33.0, 29.2, 28.3 (3C), 24.2, 24.0,
Diastereomeric ratio determinations of the saponified Joullié-
Ugi 3CR products were determined using a Haisil-100 C18 column
(5 micron, 150 Â 4.6 mm). The flow rate was 1.5 mL/min. The
method consisted of 20–65% MeCN in H2O + 0.1% TFA over
20 min. The detector was set to 214 nm. The retention times of
the diastereomeric products were 7.63 and 8.58 min. The retention
23.1, 22.4, 17.6, 14.1. HRMS (FAB): calcd for
C21H35N3O6Na
[M+Na]+ 448.2424, obsd 448.2432 (
D
= 1.8 ppm).
5.2.3.4. Cyclohexanone Ugi tripeptide (8). Yield: 3.75 g
(8.53 mmol, 90%). Yellow oil 1H NMR (CDCl3, mixture of rotamers)
6.72 (1H, m), 4.51 (1H, m), 4.37 (1H, m) 3.54 (3H, s), 3.45 (1H, m),
3.35 (1H, m), 2.98 (3H, s), 2.11–1.52 (14H, m) 1.43 (9H, s), 1.38 (3H,
d). 13C NMR (CDCl3, mixture of rotamers) 175.6, 173.7, 171.1,
154.5, 79.4, 66.6, 65.8, 60.2, 58.6, 52.2, 48.1, 46.7, 32.9, 32.4,
31.8, 30.4, 29.5, 28.5 (3C), 24.2, 22.8, 21.9, 20.9, 18.1, 17.7, 14.1.
HRMS (FAB): calcd for C22H37N3O6Na [M+Na]+ 462.2580, obsd
time of authentic N-Boc-
L-Pro-L-Pip-L-Ala-COOH, prepared accord-
ing to the literature,8 was 7.65 min.
5.2.2.1.
N-Boc-Pro-pipecolate-Ala-OMe
(3). Yield:
2.16 g
(5.2 mmol, 76%). Rf = 0.3 (7:3 EtOAc/hexanes) 1H NMR (CDCl3, mix-
ture of diasteromers) 8.45 (d, 0.5H), 7.38, (d, 0.5 H), 4.52 (m, 1H),
4.48 (m, 1H), 4.42 (m, 1H), 3.85 (d, 0.5H), 3.67 (s, 3H), 3.44 (1H,
m), 3.40 (1H, m), 2.45 (1H, m), 2.08 (2H, m), 1.87 (3H, m), 1.64
(2H, m) 1.35 (12H, s), 0.99 (m, 2H). 13C NMR (CDCl3): 173.3,
172.1, 170.0, 154.4, 79.6, 55.4, 52.0, 49.9, 48.1, 46.7, 43.4, 39.4,
34.4, 29.3, 28.0 (3C), 25.7, 24.5, 20.3, 16.6. HRMS (FAB): calcd for
462.2562 (D = 3.9 ppm).
5.2.4. General procedure for the saponification of tripeptides
The tripeptide methyl esters (1.55 g, 3.62 mmol) were dissolved
in 10.4 mL of 1:1 THF/H2O solution and LiOHÁH2O (356 mg,
8.47 mmol) was added. The reaction was stirred for 2.5 h then
quenched with 10 mL 1 N HCl. THF was removed under vacuum
and aqueous layer was extracted with 2 Â 20 mL EtOAc. The organ-
ic layer was dried over Na2SO4 and concentrated under vacuum.
C
20H33N3O6Na [M+Na]+ 434.2267, obsd 434.2267 (
D = 5.8 ppm).
5.2.2.2. N-Boc-Pro-4-methylpipecolate-Ala-OMe (4). Yield:
2.47 g (5.8 mmol, 84%). Rf = 0.3 (7:3 EtOAc/hexanes) 1H NMR
(CDCl3, mixture of diasteromers): 8.39 (d, 0.5H), 7.28, (d, 0.5 H),
4.55 (m, 1H), 4.48 (m, 1H), 4.40 (m, 1H), 3.85 (d, 0.5H), 3.65 (s,
3H), 3.49 (1H, m), 3.41 (1H, m), 2.43 (1H, m), 2.05 (2H, m), 1.83
(3H, m), 1.64 (2H, m) 1.38 (12H, s), 0.99 (m, 2H), 0.85 (3H, d).
13C NMR (CDCl3): 173.4, 172.1, 170.1, 154.6, 79.8, 56.4, 52.2,
48.5, 47.0, 43.4, 34.4, 33.1, 29.5, 29.4, 28.4 (3C), 27.1, 24.9, 21.8,
16.9. HRMS (FAB): calcd for C21H35N3O6Na [M+Na]+ 448.2424, obsd
5.2.5. Synthesis of N-2-heptenoyl-3-fluorophenylalanine-COOH
(9)
N-Boc-L-3-fluorophenyl-L-alanine (500 mg, 1.76 mmol) and
NaHCO3 (300 mg, 3.6 mmol) were dissolved in 10 mL DMF. MeI
(1 mL, 16 mmol) was added dropwise and the mixture stirred for
24 h. The reaction mixture was then partitioned in 25 mL EtOAc
25 mL 0.1 N HCl and 25 mL brine, then dried over Na2SO4 and con-
448.2406 (D = 4.1 ppm).
5.2.3. General procedure for the Ugi 4CR
Each ketone (9.48 mmol) was added to a solution of 40%
methylamine in MeOH (4.75 mL, 9.48 mmol) and allowed to pre-
centrated in vacuo. Yield: 500 mg (1.68 mmol, 96%). N-Boc-
L-3-
fluorophenyl- -alanine methyl ester (500 mg, 1.7 mmol) was dis-
L
solved in 6.75 mL DCM and 4.3 mL 9:1 TFA/H2O was added at
0 °C. The reaction was allowed to warm up to room temperature
and stirred for 2 h, then 2 Â 100 mL toluene successively added
condense for 1.5 h. N-Boc-
and the resulting mixture stirred for 15 min, then isocyanide de-
rived from -alanine methyl ester (1.08 g, 9.48 mmol) was added.
L-proline (2.04 g, 9.48 mmol) was added
L
and removed in vacuo. Yield: 510 mg of TFA salt of L-3-fluoro-
The reaction was stirred 5 days at room temperature, then concen-
trated in vacuo. Crude product was loaded onto a 80 Â 50 mm Si
gel resin bed and eluted with 7:3 EtOAc/hexanes. By TLC, all Ugi
products had an Rf of approximately 0.2 (7:3 EtOAc/hexanes) and
stained white with a p-anisaldehyde reagent.
phenyl-L-alanine methyl ester (1.63 mmol, 93%). The product
(650 mg, 2.09 mmol) was dissolved in 25 mL of DCM and then 3
mL of DIEA was added. The mixture was transferred to a separate
flask containing HATU (1.14 g, 3 mmol) and trans-2-heptenoic acid
(350 lL, 2.6 mmol) in 15 mL DCM. The reaction was stirred over-
night. The reaction mixture was then washed with 30 mL 0.1 N
HCl, dried over Na2SO4 and concentrated under vacuum, then puri-
fied by flash Si gel chromatography with 1:1 EtOAc/hexanes as elu-
ent. Rf = 0.95 (1:1 EtOAc/hexanes).
5.2.3.1. 3-Pentanone Ugi tripeptide (5). Yield 3.03g (7.11 mmol,
75%). Yellow oil 1H NMR (CDCl3, mixture of rotamers) 6.35 (1H, d),
4.59 (1H, m), 3.70 (3H, s), 3.50 (1H, m), 3.45 (1H, m), 3.08 (3H, s),
2.23 (2H, m), 2.10 (2H, m), 1.85 (4H, m), 1.48 (3H, d), 1.38 (9H, s),
0.87 (3H, m) 0.78 (3H, m). 13C NMR (CDCl3, mixture of rotamers)
173.9, 173.6, 172.9, 157.1, 154.7, 79.6, 68.7, 68.2, 58.0, 57.7, 52.0,
50.5, 48.5, 46.9, 32.5, 29.2, 28.5 (3C), 26.1, 24.2, 24.0, 17.7. HRMS
(FAB): calcd for C21H37N3O6Na [M+Na]+ 450.2580, obsd 450.2566
N-2-heptenoyl-3-fluorophenylalanine methyl ester (400 mg,
1.3 mmol) was dissolved in 4 mL of 1:1 THF/H2O, then LiOHÁH2O
(149 mg, 3.55 mmol) was added and the reaction stirred for
30 min. 1 N HCl was added to adjust the pH to 1, THF was removed
under vacuum and the resulting aqueous layer was extracted with
2 Â 20 mL EtOAc. The organic layer was then dried with Na2SO4
and concentrated under vacuum. Yield: 400 mg (100%) of 9. 1H
NMR (CDCl3) 7.18 (1H, m), 6.88 (4H, m), 6.37 (1H, d, J = 8 Hz),
5.77 (1H, d, J = 16 Hz), 4.91 (1 H, d, J = 8 Hz), 3.22 (1H, m) 3.12,
(1H, m), 2.13 (2H, m), 1.33 (4H, m), 0.86, (3H, m). 13C NMR (CDCl3,
mixture of rotamers) 174.2, 167.0, 164.1, 161.7, 147.5, 138.5,
130.3, 125.4, 122.7, 116.6, 114.4, 53.7, 37.1, 32.0, 30.2, 22.3, 13.9.
HRMS (FAB): calcd for C16H20NO3FNa [M+Na]+ 316.1325, obsd
(D = 3.1 ppm).
5.2.3.2. Cyclobutanone Ugi tripeptide (6). Yield: 3.62 g
(8.81 mmol, 93%). Yellow oil 1H NMR (CDCl3, mixture of rotamers)
7.65 (1H, d), 4.42 (2H, m), 3.52 (3H, s), 3.45 (1H, m), 3.40 (1H, m),
2.87 (3H, s), 2.54 (4H, m), 2.03 (3H, m), 1.65 (2H, m), 1.54 (1H, m),
1.38 (3H, d), 1.28 (9H, s). 13C NMR (CDCl3, mixture of rotamers)
173.7, 173.3, 173.2, 154.7, 153.6, 79.7, 65.6, 60.5, 56.7, 53.6.0,
52.2, 48.1, 46.9, 33.2, 31.8, 28.5 (3C), 24.5, 23.4, 17.5, 14.5. HRMS
(FAB): calcd for C20H33N3O6Na [M+Na]+ 434.2267, obsd 434.2282
316.1306 (D = 6.0 ppm).
(
D = 3.4 ppm).
5.2.6. General procedure for synthesis of acyldepsipeptides
(10a–h)
5.2.6.1. Linear pentapeptolide formation and deprotec-
tion. Each tripeptide (3.9 mmol), TPTU (1.45 g, 4.88 mmol), HOBt
5.2.3.3. Cyclopentanone Ugi tripeptide (7). Yield: 3.12 g
(7.33 mmol, 78%). Yellow oil 1H NMR (CDCl3, mixture of rotamers)
7.65 (1H, d), 4.54 (1H, m), 4.40 (1H, m), 3.65 (3H, s), 3.60 (1H, m),

7200
A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
(770 mg, 5.7 mmol) and DIEA (2 mL, 11.5 mmol) were dissolved in
20 mL dry DCM and allowed to stir under N2, at rt for 10 min. Dep-
sipeptide (NH2-Pro-N-CBz-Ser-phenacylester)8 (2.2 g, 3.9 mmol)
was dissolved in 10 mL dry DCM and added dropwise over 5 min.
After 15 h, the reaction mixture was concentrated in vacuo and
then redissolved in 50 mL EtOAc. The EtOAc solution was extracted
sequentially with 20 mL each of 0.2 N HCl, saturated NaHCO3 and
brine. The organic layer was dried over Na2SO4 and concentrated
in vacuo. The product was purified over a 110x 55 mm bed of silica
gel using 20:0.5 EtOAc/MeOH. By TLC, all pentapeptolide products
had an Rf of approximately 0.3 (20:0.5 EtOAc/methanol) and
stained blue with ceric ammonium molybdenate.
then filtered through celite, the residue was washed with MeOH
and the filtrate was concentrated under vacuum. The CBz depro-
tected macrocycles (0.1 mmol) were coupled to
0.15 mmol) in 800 L DMF containing TPTU (50 mg, 0.17 mmol),
HOBt (20 mg, 0.15 mmol), and 60 L diisopropylethyamine. The
9 (44 mg,
l
l
reaction was stirred overnight at room temperature.
Purification of the acyldepsipeptides (10a–h) was accomplished
using a Phenomenex Gemini C18 (10 micron, 150 Â 10 mm, 110 Å).
The method consisted of 20–80% MeCN in H2O + 0.1% TFA over
20 min. The detector was set to 214 nm. The flow rate was
10 mL/min.
Phenacyl pentapeptolides (1.3 mmol) were deprotected by dis-
solving in 1.45 mL 90% AcOH in H2O, then Zn dust (580 mg,
9.07 mmol) added and reaction stirred for 30 min. Zn was filtered
through cotton and residue washed with 10 mL EtOAc and 10 mL
1 N HCl. 30 mL EtOAc was added to the filtrate and it was washed
with 3 Â 20 mL 1 N HCl. Organic layer was dried with Na2SO4 and
concentrated under vacuum. Yield: 945 mg (100%). Rf = 0.2 (20:1
EtOAc/MeOH).
5.2.6.4. Acyldpesipeptide 10a. Isolated yield from deprotected
peptidolactone: 25 mg (0.034 mmol, 34%) white amorphous pow-
der. 1H NMR (CDCl3) 8.55 (1H, d, J = 8 Hz), 7.26 (0.5H, m), 7.22
(0.5H, m), 6.99 (1H, m), 6.89 (3H, m), 6.52 (1H, d, J = 12 Hz), 6.17
(1H, d, J = 16 Hz), 5.12 (1H, m), 4.98 (1H, m), 4.75 (1H, d,
J = 12 Hz), 4.67 (2H, m), 4.55 (1H, m), 4.48 (2H, m), 3.73 (1H, m),
3.61 (1H, m), 3.51 (2H, m), 3.30 (1H, m), 2.94 (2H, m), 2.71 (1H,
m), 2.62 (1H, m), 2.34 (1H, m), 2.38 (4H, m), 1.92 (6H, m), 1.75
(1H, m), 1.64 (1H, m), 1.40 (1H, m), 1.38 (3H, m), 1.32 (4H, m),
0.88 (3H, m). 13C NMR (CDCl3) 172.5, 171.2, 170.9, 169.4, 166.4,
165.0, 164.0, 161.5, 146.2, 138.6, 130.0, 125.2, 123.1, 116.5,
116.4, 113.8, 113.6, 65.0, 59.0, 57.1, 56.8, 54.7, 51.2, 47.8, 47.0,
46.4, 41.3, 38.4, 31.8, 30.7, 30.4, 28.0, 24.9, 23.1, 22.2, 21.3, 18.0,
13.8. HRMS (FAB): calcd for C38H51N6O8FNa [M+Na]+ 761.3650,
5.2.6.2. Macrocyclization. Pentafluorophenol esters were pre-
pared from phenacyl deprotected linear peptides (1.37 mmol), by
dissolving in 5.25 mL dry DCM and adding pentafluorophenol
(1 g, 5.43 mmol) and EDC (285 mg, 1.64 mmol) at À78 °C. The reac-
tion was allowed to warm to room temperature overnight with
stirring, then concentrated under vacuum. Crude pentafluorophe-
nol esters (approximately 2 g) were Boc deprotected by treatment
with 8.35 mL 3 N HCl in EtOAc at 0 °C and stirred at 0 °C for 3.5 h.
For cyclization, the crude reaction mixture was diluted into 417 mL
DCM and added dropwise at a rate of ꢀ1 drop/s to a vigorously stir-
ring solution of 324 mL satd NaHCO3 and 557 mL DCM. The reac-
tion was left to stir overnight. The organic layer was collected
and dried over Na2SO4 and concentrated in vacuo, then purified
over a 100 Â 25 mm Si gel column with 20:1 EtOAc/MeOH eluent.
In the case of diastereomers derived from 3 and 4, the compound
with an Rf = 0.3 is (2S)-pipecolate and the compound with an
Rf = 0.2 is (2R)-pipecolate.
obsd 761.3625 (D = 3.3 ppm).
5.2.6.5. Acyldpesipeptide 10b. Isolated yield from deprotected
peptidolactone: 26 mg (0.035 mmol, 35%) white amorphous pow-
der. 1H NMR (CDCl3, mixture of rotamers) 8.03 (1H, m), 7.45 (1H,
m), 7.18 (1H, m), 6.99 (4H, m), 6.85 (1H, m), 6.75 (1H, m), 6.56
(0.6H, m), 5.90 (1H, d), 5.80 (1H, d), 5.38 (1H, d), 5.30 (2H, m),
4.92 (3H, m), 4.76 (2H, m), 4.37 (1H, m), 4.33 (1 H, m), 3.97 (1H,
m), 3.80 (1H, m), 3.75 (2H, m), 3.63 (3H, m), 3.25 (3H, m), 3.10
(1H, m), 2.92 (1H, m), 2.65 (1H, m), 2.10 (4H, m), 2.01 (10H, m),
1.52 (2H, m), 1.45 (4H, m), 1.38 (3H, m), 0.85 (3H, m). 13C NMR
(CDCl3, mixture of rotamers) 172.2, 170.9, 170.3, 167.7, 167.3,
166.0, 163.8, 161.1, 129.7, 129.6, 125.0, 123.3, 116.3, 113.3, 61.2,
60.0, 56.1, 53.1, 48.5, 47.5, 47.2, 46.3, 44.7, 31.7, 30.1, 28.8, 28.7,
26.2, 25.6, 25.5, 25.0, 22.1, 20.6, 17.3, 13.4. HRMS (FAB): calcd for
Spectral data for S-Pip-CBz-macrocycle derived from 3: 1H NMR
(CDCl3) 8.30 (1H,d, J = 8 Hz), 7.28 (4H, m), 5.68 (1H, d, J = 8 Hz),
5.15 (1H, m), 5.03 (2H, m), 4.99 (1H, m), 4.74 (1H, d, J = 8 Hz),
4.66 (1H, s), 4.49 (2H, m), 4.17 (1H, t), 4.06 (1H, dd), 3.72 (2H,
m), 3.65 (1H, m), 3.50 (2H, m), 2.68 (1H, d, J = 12 Hz), 2.52 (1H,
m), 2.31 (1H, m), 2.11 (3H, m), 1.91 (7H, m), 1.68 (1H, m), 1.58
(1H, m), 1.36 (5H, m). 13C NMR (CDCl3) 173.3, 171.5, 169.6,
168.5, 166.2, 156.2, 135.9, 128.5, 128.3 (3H), 128.0, 67.9, 65.8,
60.3, 58.9, 57.1, 56.6, 53.9, 53.4, 48.0, 46.8, 46.3, 41.0, 30.9, 30.6,
28.2, 24.8, 23.1, 21.3, 18.0, 14.1. HRMS (FAB): calcd for C30H39N5O8-
C
38H51N6O8FNa [M+Na]+ 761.3650, obsd 761.3638 (
D = 1.6 ppm).
5.2.6.6. Acyldpesipeptide 10c. Isolated yield from deprotected
peptidolactone: 23.4 mg (0.034 mmol, 31%) white amorphous
powder. 1H NMR (CDCl3) 8.89 (1H, br s), 8.51 (1H, d, J = 12 Hz),
8.42 (0.5H, br s) 7.93 (0.5H, br s), 7.16 (1H, m), 6.88 (4H, m),
6.20 (1H, d, J = 16 Hz), 5.14 (1H, m), 4.98 (1H, m), 4.73 (3H, m),
4.67 (1H, m), 4.50 (2H, m), 3.75 (1H, m), 3.55 (3H, m), 3.30 (1H,
m), 2.94 (2H, m), 2.70 (2H, m), 2.31 (1H, m), 2.14 (3H, m), 1.88
(5H, m), 1.62 (2H, m), 1.38 (3H, m), 1.30 (5H, m), 0.94 (3H, m),
0.86 (3H, m). 13C NMR (CDCl3) 172.5, 171.1, 170.8, 169.6, 167.1,
165.1, 164.0, 161.6, 147.0, 138.4, 130.2, 125.2, 122.9, 116.5,
114.0, 65.0, 59.1, 57.1, 57.2, 57.0, 55.2, 51.1, 47.9, 47.1, 46.6,
41.0, 38.4, 36.2, 33.3, 31.9, 30.8, 30.4 (2C), 28.0, 23.1, 22.2, 21.9,
Na [M+Na]+ 620.2696, obsd 620.2681 (
D = 2.4 ppm).
Spectral data for S-4-Me-Pip-CBz-macrocycle derived from 4:
1H NMR (CDCl3) 8.31 (1H,d, J = 8 Hz), 7.28 (4H, m), 5.63 (1H, d,
J = 8 Hz), 5.17 (1H, m), 5.03 (2H, m), 4.99 (1H, m), 4.76 (1H, d,
J = 8 Hz), 4.67 (1H, s), 4.50 (2H, m), 4.19 (1H, t), 4.08 (1H, dd),
3.72 (2H, m), 3.66 (1H, m), 3.49 (2H, m), 2.68 (1H, d, J = 12 Hz),
2.53 (1H, m), 2.31 (1H, m), 2.12 (3H, m), 1.94 (7H, m), 1.56 (2H,
m), 1.36 (3H, m), 1.02 (2H, m), 0.96 (3H, m). 13C NMR (CDCl3)
173.2, 171.5, 169.7, 168.6, 166.2, 156.2, 135.8, 128.4, 128.3 (3H),
128.0, 67.9, 65.8, 60.3, 58.8, 57.1, 56.7, 53.9, 48.0, 46.9, 46.3,
40.7, 36.3, 33.1, 30.8, 30.5, 27.9, 23.0, 21.7, 21.2, 17.9. HRMS
(FAB): calcd for C30H39N5O8Na [M+Na]+ 634.2853, obsd 634.2868
18.0, 13.8. HRMS (FAB): calcd for
775.3807, obsd 775.3820 ( = 1.7 ppm).
C
39H53N6O8FNa [M+Na]+
D
5.2.6.7. Acyldpesipeptide 10d. Isolated yield from deprotected
peptidolactone: 22.1 mg (0.03 mmol, 30%) white amorphous pow-
der. 1H NMR (CDCl3, mixture of rotamers) 8.02 (1H, m), 7.74 (1H, br
s), 7.51 (2H, m), 7.16 (1H, m), 6.96 (4H, m), 6.74 (1H, m), 5.96 (1H,
m), 5.35 (0.5H, m), 5.30 (1H, m), 5.20 (0.5H, m), 4.82 (3H, m), 4.75
(1H, m), 4.40 (1H, m), 4.22 (1H, m), 4.01 (1H, m), 3.56 (3H, m), 3.25
(2H, m), 2.94 (1H, m), 2.62 (1H, m), 2.25 (3H, m), 2.02 (6H, m), 1.86
(5H, m), 1.62 (1H, m), 1.38 (3H, m), 1.30 (4H, m), 0.93 (3H, m), 0.85
(D = 2.4 ppm).
5.2.6.3. Deprotection of macrocycles. Cbz-protected macrocy-
cles (0.16 mmol) were deprotected using 75 mg 10% Pd/C and
100 mg NH4COOH dissolved in 4 mL isopropanol heated in a
microwave reactor for 8 min at 85 °C.35 The reaction mixture was

A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
7201
(3H, m). 13C NMR (CDCl3, mixture of rotamers) 172.1, 171.8, 171.3,
170.5, 170.2, 168.8, 161.5, 145.4, 140.2, 129.7, 124.8, 123.4, 116.4,
116.1, 113.4, 61.0, 60.3, 57.6, 56.6, 54.2, 53.3, 50.9, 47.5, 47.2, 46.4,
44.4, 40.9, 38.0, 33.9, 33.2, 32.8, 31.9, 30.4, 29.7, 28.9, 28.2, 27.3,
26.2, 25.7, 22.2, 21.7, 18.1, 15.1, 13.9. MS (ESI, m/z): 775.8
[M+Na]+ for C39H53N6O8F.
cation-adjusted Mueller–Hinton broth (SMHB; Difco Laboratories,
Sparks, MD, USA; 25 g/mL calcium and 12.5 g/mL magnesium)
was used to dilute antimicrobial agents in a serial twofold sche-
dule. Colony counts for the inoculum verification were determined
using Tryptic Soy Agar plates (TSA, Difco, Becton Dickinson Co.,
l
l
Sparks, MD, USA). Following incubation, 100 lL of broth was sub-
cultured onto TSA for MBC determination. MIC and MBC determi-
nation can be found in Table 2 and 3. Control isolates were
obtained from the American Type Culture Collection (ATCC) and
were as follows: Methicillin-resistant S. epidermidis (MRSE ATCC
35984); MSSA (ATCC 35556) and, VRE (ATCC 2030).
5.2.6.8. Acyldpesipeptide 10e. Isolated yield from deprotected
peptidolactone: 22.0 mg (0.037 mmol, 37%) white amorphous
powder. 1H NMR (CDCl3, mixture of rotamers) 7.69 (1H, d), 6.99
(1H, d), 7.18 (1H, m), 6.94 (4H, m), 6.75 (1H, m), 5.94 (1H, d),
5.75 (1H, m), 5.58 (1H, m), 5.02 (1H, m), 4.90 (1H, m), 4.80 (2H,
m), 4.62 (2H, m), 4.00 (1H, m), 3.70 (3H, m), 3.60 (2H, m), 3.28
(2H, m), 2.92 (1H, m), 2.60 (1H, m), 2.27 (3H, m), 2.02 (6H, m),
1.86 (5H, m), 1.72 (1H, m), 1.38 (3H, m), 1.30 (4H, m), 0.93 (3H,
m), 0.85 (3H, m). MS (ESI, m/z): 777.4 [M+Na]+ for C39H53N6O8F.
Acknowledgments
Brown University is gratefully acknowledged for financial sup-
port to J.K.S. and A.M.S. K.L.L. was supported by the Rhode Island
VA Hospital. The authors thank Suzanne Woodmansee for perform-
ing the biological assays.
5.2.6.9. Acyldpesipeptide 10f. Isolated yield from deprotected
peptidolactone: 25 mg (0.034 mmol, 34%) white amorphous pow-
der. 1H NMR (CDCl3, mixture of rotamers) 7.52 (1H, d), 7.47 (1H,
d), 7.15 (1H, m), 6.82 (4H, m), 5.90 (1H, d, J = 16 Hz), 5.30 (1H,
d), 4.92 (2H, m), 4.54 (1H, m), 4.40 (1H, m), 4.23, 1H, m), 3.85
(1H, d, J = 12 Hz), 3.58 (3H, m), 3.34 (1H, m), 2.91 (3H, s), 2.85
(1H, m), 2.62 (2H, m), 2.06 (4H, m), 1.98 (4H, m), 1.81 (2H, d),
1.23 (4H, m), 0.85 (3H, m). 13C NMR (CDCl3, mixture of rotamers)
172.4, 172.2, 171.0, 166.6, 161.6, 145.5, 140.2, 129.9, 125.3,
123.3, 116.5, 113.6, 66.8, 61.1, 60.2, 58.0, 56.9, 54.3, 51.2, 47.5,
38.0, 34.2, 32.9, 31.9, 30.3, 29.7, 29.4, 28.6, 28.1, 26.4, 26.2, 25.7,
23.3, 22.7, 22.2, 15.6, 15.3, 14.1, 13.9. MS (ESI, m/z): 761.3
[M+Na]+ for C38H51N6O8F.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.08.032.
References and notes
1. World Health Organization (WHO) Geneva, World Health Report, 2002.
2. Nathan, C. Nature 2004, 431, 899.
3. National Nosocomial Infections Surveillance (NNIS) system Am. J. Infect. Control.
2004, 32, 470.
4. Boucher, H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice, L. B.;
Scheld, M.; Spellberg, B.; Bartlett, J. Clin. Infect. Dis. 2009, 48, 1.
5. Clardy, J. M. Nat. Biotechnol. 2006, 24, 1541.
5.2.6.10. Acyldpesipeptide 10g. Isolated yield from deprotected
peptidolactone: 23 mg (0.030 mmol, 30%) white amorphous pow-
der. 1H NMR (CDCl3, mixture of rotamers) 7.70 (1H, m), 7.49 (1H,
m), 7. 20 (1H, m), 6.97 (2H, m), 6.88 (1H, m), 6.72 (1H, m), 5.92
(1H, d, J = 16 Hz), 5.35 (1H, d, J = 12 Hz), 4.90 (2H, m), 4.83 (1H,
m), 4.61 (1H, d), 4.37 (1H, m), 3.86 (1H, d, J = 12 Hz), 3.69 (2H,
m), 3.58 (1H, m), 3.49 (1H, m), 3.40 (1H, m), 3.19 (3H, s), 2.91
(1H, m), 2.60 (1H, m), 2.21 (2H, m), 2.10 (2H, m), 2.00 (2H, m),
1.92 (2H, m), 1.33 (3H, m), 0.85 (3H, m). 13C NMR (CDCl3, mixture
of rotamers) 174.0, 173.2, 172.6, 171.8, 171.2, 170.2, 167.3, 166.7,
163.9, 161.5, 155.6, 140.5, 129.6, 125.1, 116.4, 113.2, 79.5, 74.6,
61.0, 60.8, 60.0, 59.4, 59.0, 56.2, 51.1, 47.5, 46.1, 38.5, 37.9, 36.9,
35.5, 35.1, 33.4, 31.9, 29.4, 28.2, 27.9, 26.4, 25.8, 25.0, 24.6, 23.5,
22.6, 18.2, 15.6, 14.1. MS (ESI, m/z): 775.6 [M+Na]+ for
6. von Nussbaum, F.; Brands, M.; Hinzen, B.; Weigand, S.; Häbich, D. Angew.
Chem., Int. Ed. 2006, 45, 5072.
7. Hinzen, B.; Raddatz, S.; Paulsen, H.; Lampe, T.; Schumacher, A.; Häbich, D.;
Hellwig, V.; Benet-Buchholz, J.; Endermann, R.; Labischinski, H.; Brötz-
Oesterhelt, H. ChemMedChem 2006, 1, 689.
8. Hinzen, B.; Brötz-Oesterhelt, H.; Endermann, R.; Henninger, K.; Paulsen, H.;
Raddatz, S.; Lampe, T.; Hellwig, V.; Schumacher, A. U.S. Patent 0107288 A1,
2005.
9. Osada, H.; Yano, T.; Koshino, H.; Isono, K. J. Antibiot. 1991, 44, 1463.
10. Michel, K. H.; Kastner, R. E. U.S. Patent 4492650, 1985.
11. Brötz-Oesterhelt, H.; Beyer, D.; Hein-Peter, K.; Endermann, R.; Ladel, C.;
Schroeder, W.; Hinzen, B.; Raddatz, S.; Paulsen, H.; Henninger, K.; Bandow, J. E.;
Hans-Georg, S.; Labischinski, H. Nat. Med. 2005, 11, 1082.
12. Maurizi, M. R.; Singh, S. K.; Thompson, M. W.; Kessel, M.; Ginsburg, A.
Biochemistry 1999, 37, 7778.
13. Singh, S. K.; Grimaud, R.; Hoskins, J. R.; Wickner, S.; Maurizi, M. R. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 8898.
14. Baker, T. A.; Sauer, R. T. Trends Biochem. Sci. 2006, 31, 647.
15. Hsiung, A. Y.; Houry, A. FEBS Lett. 2007, 581, 3749.
16. Lee, B. G.; Park, E. Y.; Lee, K. E.; Jeon, H.; Sung, K. H.; Paulsen, H.; Rübsamen-
Schaeff, H.; Brötz-Oesterhelt, H.; Song, H. K. Nat. Struct. Mol. Biol. 2010, 17, 471.
17. Schmidt, U.; Neumann, K.; Schumacher, A.; Weinbrenner, S. Angew. Chem., Int.
Ed. 1997, 36, 1110.
C39H55N6O8FNa.
5.2.6.11. Acyldpesipeptide 10h. Isolated yield from deprotected
peptidolactone: 24 mg (0.032 mmol, 32%) white amorphous pow-
der. 1H NMR (CDCl3, mixture of rotamers) 7.76 (1H, d, J = 8 Hz),
7.48 (1H, d, J = 4 Hz), 7.20 (1H, m), 6.82 (4H, m), 5.90 (1H, d,
J = 16 Hz), 5.36 (1H, d, J = 8 Hz), 4.92 (2H, m), 4.72 (1H, m), 4.37
(1H, m), 3.70 (1H, m), 3.58 (2H, m), 3.39 (2H, m), 3.18 (3H, s),
2.24 (2H, m), 2.05 (6H, m), 1.83 (1H, m), 1.63 (1H, m), 1.38 (3H,
m), 0.85 (3H, m). 13C NMR (CDCl3, mixture of rotamers) 174.7,
171.4, 170.7, 170.3, 167.2, 166.4, 145.9, 140.3, 129.7, 124.9, 123.3
(2C), 116.3, 113.4, 69.9, 61.1, 60.2, 60.0, 59.2, 54.4, 51.0, 47.4,
46.4, 40.3, 38.1, 34.5, 33.8, 33.4, 31.9, 29.4, 26.3, 25.7, 23.3, 22.7,
22.2, 18.5, 18.1, 14.2, 13.8. MS (ESI, m/z): 789.6 [M+Na]+ for
C40H55N6O8FNa.
18. Dömling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 3168.
19. Dömling, A. Chem. Rev. 2006, 106, 17.
20. Nutt, R. F.; Joullié, M. M. J. Am. Chem. Soc. 1982, 104, 5852.
21. Nenajdenko, V. G.; Gulevich, A.; Balenkova, E. S. Tetrahedron 2006, 62, 5922.
22. Simila, S. T.; Martin, S. F. Tetrahedron Lett. 2008, 49, 4501.
23. Zhu, J.; Wu, X.; Danishefsky, S. J. Tetrahedron Lett. 2008, 50, 577.
24. Waki, M.; Meienhofer, J. J. Org. Chem. 1977, 42, 2019.
25. Zhong, Y.; Zhou, H.; Gauthier, D. R.; Lee, J.; Askin, D.; Dolling, U. H.; Volante, R.
P. Tetrahedron Lett. 2005, 46, 1099.
26. Davis, B. G.; Maughan, M. A. T.; Chapman, T. M.; Villard, R.; Courtney, S. Org.
Lett. 2002, 4, 103.
27. Eliel, E.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley and
Sons: New York, 1994. Chapter 11.
28. Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6747.
29. Marshall, G. R.; Hodgkin, E. E.; Langs, D. A.; Smith, G. D.; Zabrocki, J.; Leplawy,
M. T. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 487.
30. Clinical and Laboratory Standard Institute (CLSI) Methods for dilution
antimicrobial susceptibility tests for bacteria that grow aerobically M07-A8; CLSI
8th ed., Wayne, PA, 2009.
31. Clinical Laboratory Standards Institute (CLSI) Performance standards for
antimicrobial susceptibility testing M100-S19. CLSI 19th informational suppl.,
Wayne, PA, 2009.
5.3. Biological assays
Minimum inhibitory concentrations (MIC) were determined in
duplicate using methods outlined by The Clinical and Laboratory
Standards Institute. Compounds were dissolved in DMSO and fresh

7202
A. M. Socha et al. / Bioorg. Med. Chem. 18 (2010) 7193–7202
32. Schreiber, S. L. Science 2000, 287, 1964.
34. National Committee for Clinical Laboratory Standards (NCCLS). Methodology for
the Serum Bactericidal Test M21-A. NCCLS Approved Standard M21-A. Villanova,
PA, USA, 1999.
33. National Committee for Clinical Laboratory Standards (NCCLS). Methods for
Determining Bactericidal Activity of Antimicrobial Agents M26-A. NCCLS
Approved Standard Villanova, PA, USA, 1999.
35. Daga, M. C.; Taddei, M.; Varchi, G. Tetrahedron Lett. 2001, 42, 5191.