Initial efforts toward the optimization of arylomycins for antibiotic activity

Article abstract of DOI:10.1021/jm1016126

While most clinically used antibiotics were derived from natural products, the isolation of new broad-spectrum natural products has become increasingly rare and narrow-spectrum agents are typically deemed unsuitable for development because of intrinsic limitations of their scaffold or target. However, it is possible that the spectrum of a natural product antibiotic might be limited by specific resistance mechanisms in some bacteria, such as target mutations, and the spectra of such "latent" antibiotics might be reoptimized by derivatization, just as has been done with clinically deployed antibiotics. We recently showed that the spectrum of the arylomycin natural product antibiotics, which act via the novel mechanism of inhibiting type I signal peptidase, is broader than previously believed and that resistance in several key human pathogens is due to the presence of a specific Pro residue in the target peptidase that disrupts interactions with the lipopeptide tail of the antibiotic. To begin to test whether this natural resistance might be overcome by derivatization, we synthesized analogues with altered lipopeptide tails and identified several with an increased spectrum of activity against S. aureus. The data support the hypothesis that the arylomycins are latent antibiotics, suggest that their spectrum may be optimized by derivatization, and identify a promising scaffold upon which future optimization efforts might focus.

Full text of DOI:10.1021/jm1016126

ARTICLE
pubs.acs.org/jmc
Initial Efforts toward the Optimization of Arylomycins
for Antibiotic Activity
Tucker C. Roberts, Mark A. Schallenberger, Jian Liu, Peter A. Smith, and Floyd E. Romesberg*
Department of Chemistry, The Scripps Research Institute, La Jolla, California 92037, United States
S Supporting Information
b
ABSTRACT: While most clinically used antibiotics were derived from natural
products, the isolation of new broad-spectrum natural products has become increas-
ingly rare and narrow-spectrum agents are typically deemed unsuitable for development
because of intrinsic limitations of their scaffold or target. However, it is possible that the
spectrum of a natural product antibiotic might be limited by specific resistance
mechanisms in some bacteria, such as target mutations, and the spectra of such “latent”
antibiotics might be reoptimized by derivatization, just as has been done with clinically
deployed antibiotics. We recently showed that the spectrum of the arylomycin natural
product antibiotics, which act via the novel mechanism of inhibiting type I signal
peptidase, is broader than previously believed and that resistance in several key human
pathogens is due to the presence of a specific Pro residue in the target peptidase that disrupts interactions with the lipopeptide tail of
the antibiotic. To begin to test whether this natural resistance might be overcome by derivatization, we synthesized analogues with
altered lipopeptide tails and identified several with an increased spectrum of activity against S. aureus. The data support the
hypothesis that the arylomycins are latent antibiotics, suggest that their spectrum may be optimized by derivatization, and identify a
promising scaffold upon which future optimization efforts might focus.
’ INTRODUCTION
mechanism of action, there was originally much enthusiasm for
these compounds, but despite their ability to inhibit SPase in
vitro and their in vivo activity against the soil bacteria Rhodococcus
opacus and Brevibacillus brevis and the human pathogen Strepto-
coccus pneumoniae, they were found to have no activity against a
variety of other important human pathogens.15,16 This appar-
ently narrow spectrum is surprising considering that SPase is
located in an accessible, external site and appears to be present
and essential in all eubacteria.12,17ꢀ19 To explore the origins of
their narrow spectrum, we synthesized and evaluated arylomycin
A2, as well as several derivatives, including arylomycin C16
(Figure 1).20 Interestingly, we found that the arylomycins are
as active against Staphylococcus epidermidis as the antibiotics used
for its treatment, and importantly, we determined that S.
epidermidis evolves resistance by introducing a Pro residue into
SPase at position 29,21 which is predicted to interact with the P5
residue22 of a bound peptide.23ꢀ25 Remarkably, all bacteria that
had been shown to be resistant to the arylomycins have a Pro at
the corresponding position, and we identified a wide variety of
bacteria that lack this residue and showed that the majority of
them are sensitive to the arylomycins, including the Gram-
positive pathogens Streptococcus pyogenes and Staphylococcus
hemolyticus and the Gram-negative pathogens Helicobacter
pylori and Chlamydia trachomatis. Moreover, while the arylo-
mycins slow the growth of Staphylococcus aureus strain 8325,
they do not actually prevent growth,20 even at concentrations
The evolution of bacteria that are resistant to multiple
antibiotics poses a serious threat to human health.1,2 Although
much effort has been focused on the discovery and development
of fully synthetic antibiotics,3 most of the antibiotics used in the
clinic were derived from natural products, which evolved over
eons of time to penetrate bacteria, avoid efflux, and inhibit
essential biochemical processes.4 Unfortunately, broad-spectrum
natural product antibiotics have become increasingly difficult to
isolate and the more plentiful narrow-spectrum agents are
limited either by unknown factors or by factors that are intrinsic
to the compound, such as poor penetration or targeting proteins
that are not sufficiently conserved and that are viewed as
challenging to overcome by optimization.3 In contrast, there is
much precedent for reoptimizing antibiotics after their spectrum
has been compromised by specific resistance mechanisms ac-
quired during clinical use, as evidenced by the development of
many “next generation” antibiotics.5ꢀ8
Nature is replete with antimicrobial peptides that hold pro-
mise as therapeutics.9,10 The arylomycins are a class of non-
ribosomally synthesized lipopeptide antibiotics that inhibit
bacterial type I signal peptidase (SPase), an essential serineꢀ
lysine dyad protease that is anchored to the outer leaflet
of the cytoplasmic membrane and that removes N-terminal
signal peptides from proteins that are transported out of the
cytoplasm.11ꢀ13 Three related series of arylomycins have been
identified, the arylomycins A and B and the lipoglycopeptides,
which have similar core macrocycles but different substituents
and fatty acid tails (Figure 1).14,15 On the basis of their novel
Received: December 20, 2010
Published: June 01, 2011
r
2011 American Chemical Society
4954
dx.doi.org/10.1021/jm1016126 J. Med. Chem. 2011, 54, 4954–4963
|

Journal of Medicinal Chemistry
ARTICLE
a
Scheme 1. Retrosynthesis of Arylomycin C16
Figure 1. Arylomycin class of natural products and compound 1.
a Arylomycin derivatives were synthesized by modification of previously
reported protocols.20,34 Briefly, tripeptide macrocycle precursor was
synthesized by solution phase peptide couplings and then cyclized via
SuzukiꢀMiyaura macrocyclization. The various lipopeptide tails were
assembled via solution phase peptide couplings and then coupled to the
macrocyclic core or the glycine-homologated macrocycle core.
Despite not having activity against wild type E. coli, two crystal
structures of arylomycin A2 bound to a soluble fragment of E. coli
SPase have been reported (Figure 2).23,26 The arylomycin is seen
to bind in an extended β-sheet conformation that likely mimics
the binding of membrane bound pre-protein substrates.23 The
C-terminal macrocycle of the arylomycin binds in a deep
hydrophobic cleft and makes multiple H-bonds and hydrophobic
interactions with the protein, while the C-terminal carboxyl
group forms a critical salt bridge with the catalytic residues.
The peptide tail extends down a shallow cleft in the surface of
SPase and forms two H-bonds with backbone residues of the
protein. The critical resistance-conferring residue, Pro84, inter-
acts with the N-terminal end of the peptidic tail and appears to
preclude the formation of a H-bond to a carbonyl oxygen of the
arylomycin and possibly to alter the trajectory of the lipid moiety
as it enters the membrane (Figure 2). While the crystal structures
are likely to reveal little information about the biologically
relevant structure of the lipid tail because of the use of a soluble
truncated fragment of SPase and the absence of a membrane
bilayer, it most likely adopts an extended conformation to
maximize packing within the outer leaflet of the cytoplasmic
membrane.
Figure 2. The crystal structure of arylomycin A2 bound to E. coli SPase
(PDB code 1T7D) reveals that Pro84 is located in a surface depression
where it appears to preclude formation of a stabilizing H-bond with the
fatty acid carbonyl oxygen (shown in red) and possibly to preclude
accommodation of the lipid tail.21,23 D-MeSer2 and D-Ala3 of the
arylomycin are labeled. The resistance-conferring Pro is also labeled
(using E. coli numbering). Note that because of a discrepancy in the
numbering system used, Pro84 in Paetzel et al.23 and in the structure
1T7D is denoted Pro83.
as high as 128 μg/mL; however, they do prevent the growth
of the epidemic MRSA isolate USA300 with an MIC of
16 μg/mL. While this might result from unique features
associated with methicillin resistance, it suggests that the
arylomycin scaffold has the potential for broader spectrum
S. aureus activity. Importantly, we showed that the Pro residue
imparts resistance by reducing the affinity with which the
arylomycin binds and that removing this residue is sufficient
to render resistant S. aureus, Escherichia coli, and Pseudomonas
aeruginosa highly sensitive. These data suggest that if the
arylomycins could be optimized to bind SPases regardless of
the resistance-conferring Pro, then they would have a remark-
ably broad spectrum of activity.
As with a variety of other antibiotics that interact with
membranes or that have membrane-associated targets,27ꢀ33 the
lipopeptide tail of the arylomycins has been shown to play an
important role in their activity.15,16,27ꢀ30 Here, we report the first
structureꢀactivity relationship study of synthetic arylomycins,
focusing on derivatives with altered lipopeptide tails. The
activities of the arylomycin derivatives were evaluated with
4955
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
S. epidermidis, S. aureus, E. coli, and P. aeruginosa. With each
pathogen, the derivatives were evaluated in the context of SPases
with and without the critical resistance-conferring Pro, to identify
the changes in activity that result from altered interactions with
this resistance-conferring residue and to identify the types of
modifications that might be pursued to overcome resistance and
thereby instill the arylomycin scaffold with broad-spectrum
antibacterial activity.
from the inability of a hydrophobic environment to accom-
modate the charge, suggesting that this portion of the tail is
embedded in the membrane or within the interface between the
membrane and SPase.
To explore the minimal tail length required for activity and to
determine whether there is a limit to the tail length that can be
accommodated within the cytoplasmic membranes of the differ-
ent bacteria, we synthesized and characterized derivatives 2ꢀ5
(Table 1). None of these derivatives gained activity against any of
the resistant bacteria relative to arylomycin C16, but significant
differences were apparent with S. epidermidis and the genetically
sensitized strains. With the sensitive strains, the C8 derivative
2 has no activity, but the C10 derivative 3 has activity against
S. epidermidis, S. aureus, and E. coli, while only 4 and 5 show
activity against P. aeruginosa, revealing that at minimum a C12 tail
is required. In each case, activity increased with increasing tail
length until it plateaued with the C16 fatty acid tail (i.e., aryl-
omycin C16), and activity decreased slightly with the C18
derivative 5 with all but P. aeruginosa.
To further explore the effects of increased hydrophobicity, we
synthesized and characterized derivatives with tails that contain
one or more aromatic rings (Table 1). We first examined the
series of napthyl and biphenyl derivatives 6ꢀ8. The napthyl
derivative 6 shows no activity against any of the bacteria tested,
while the biphenyl derivative 7 retains some activity against wild
type S. epidermidis. We found that compound 8, which lacks the
methylene spacer between the fatty acid carbonyl and the
biphenyl moiety, also retains some activity against S. epidermidis,
suggesting that flexibility of the biphenyl moiety is not essen-
tial. To further explore this biphenyl architecture, we synthe-
sized the p-alkyl substituted biphenyl derivatives 9ꢀ12. We
observed an increase in activity with increasing alkyl substi-
tuent length against wild type S. epidermidis that plateaued with
the C6 and C8 derivatives 11 and 12, which are also active
against resistant S. epidermidis. Interestingly, several of the
compounds in this series are also active against both sensitized
and wild type S. aureus, with relative activities similar to those
observed with S. epidermidis but with absolute activities that
were somewhat lower. None of the biphenyl derivatives have
activity against the wild type or sensitized strains of P. aeruginosa,
but they do maintain activity against sensitized E. coli, again
showing trends that are similar to those observed with S. epidermi-
dis and S. aureus.
We next examined the series of phenyl substituted tail
mimetics 13ꢀ15 (Table 1). With S. epidermidis and the geneti-
cally sensitized strains, we again observed an increase in activity
with increasing alkyl chain length. Moreover, the decylphenyl
derivative 15 has activity against wild type S. aureus. Because the
number of carbon atoms in this derivative is similar to that of
arylomycin C16, which has no activity against wild type S. aureus,
the data suggest that at least some of the activity is mediated by
the interaction of the polarizable aromatic moiety with the
membrane or with SPase.
To explore the effects of lipopeptide methylation and to begin
a more focused exploration of modifications that might over-
come the deleterious effects of the resistance-conferring Pro, we
synthesized and characterized derivatives with altered N-methy-
lation at D-MeSer2 and D-Ala3 (Table 2), arylomycin residues
that are proximal to this critical residue when bound to SPase
(Figure 2). The absence of the D-MeSer2 N-methyl group in 16
results in a slight decrease in activity against both the wild type
and resistant S. epidermidis strains and a more pronounced loss of
’ RESULTS
The arylomycin derivatives were synthesized by modification
of previously reported protocols20,34 as illustrated in Scheme 1
(see Experimental Section and Supporting Information for
details). Briefly, the tripeptide macrocycle precursor was as-
sembled from ortho-iodinated hydroxyphenylglycine, alanine,
and a tyrosine boronic ester. The tripeptide was cyclized via
SuzukiꢀMiyaura macrocyclization and then methylated via a
nosylated amine and either directly coupled to a lipopeptide tail
or first coupled to glycine and then to a lipopeptide tail. The
different lipopeptide tail derivatives were assembled using solu-
tion phase peptide couplings of the corresponding natural or
unnatural amino acids, followed by lipidation. Methylation of the
peptide backbone of the tail was accomplished prior to coupling
to the lipid in the case of D-Ser2 methylation or after, in the case
of the D-Ala3 methylation. Finally, all derivatives were globally
deprotected using AlBr3 and ethanethiol.
The minimal inhibitory concentration (MIC) of each deriva-
tive was determined using a standard broth dilution method. Test
strains included wild type S. epidermidis (strain RP62A), as well as
mutant strains of S. aureus (strain PAS8001), E. coli (strain
PAS0260), and P. aeruginosa (strain PAS2006) that were ren-
dered sensitive to the arylomycins by mutation of the resistance-
conferring Pro to a residue that does not confer resistance (P29S
in the S. aureus SPase, and P84L in the E. coli and P. aeruginosa
SPase).21 MICs were also determined with the isogenic wild type
strains of S. aureus (strain NCTC 8325), E. coli (strain MG1655),
and P. aeruginosa (strain PAO1), as well as an isogenic mutant
strain of S. epidermidis where the resistance-conferring Pro has
been introduced (strain PAS9001).21 The term “resistant” is used
to refer to the reduced sensitivity of a strain harboring the
resistance-conferring Pro relative to the isogenic strain without
the Pro and not to the absolute sensitivities of the different
bacteria (e.g., resistant S. epidermidis is inhibited with an MIC of
8 μg/mL and is significantly more sensitive to arylomycin C16
than the resistant mutants of the other pathogens, which are not
inhibited at concentrations as high as 128 μg/mL).
SPase and the N-terminal portion of its natural substrates are
embedded within the bacterial cell membrane,35ꢀ37 suggesting
that some part of the inhibitor’s lipopeptide tail must also be
accommodated within the membrane, although the absence of a
lipid membrane in the crystal structure makes determining
precisely what part of the tail is embedded difficult. To probe
this interaction and also to determine whether a positive charge
at the lipidꢀpeptide junction of arylomycin is capable of inter-
acting with the negatively charged headgroups of the phospho-
lipid bilayer, we synthesized derivative 1 (Figure 1), which
replaces the lipid tail amide with a charged tertiary amine. This
derivative has significantly reduced activity against S. epidermidis
and sensitized S. aureus (MICs of 32 and 64 μg/mL, respectively)
and no activity against any of the wild type or mutant Gram-
negative strains examined. The decreased activity likely results
4956
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
Table 1. Activity (MICs (μg/mL)) of Arylomycin Derivatives with Altered Fatty Acid Tails against Strains of S. epidermidis,
S. aureus, E. coli, and P. aeruginosa Harboring SPase without (Sensitive) and with (Resistant) the Arylomycin-Resistance
Conferring Pro Residue
a Sensitive strains include wild type S. epidermidis RP62A, lepB(P29S) S. aureus NCTC 8325, lepB(P84L) E. coli MG1655, and lepB(P84L) P. aeruginosa
PAO1. See text for details. b Resistant strains include spsB(S29P) S. epidermidis RP62A, and wild type strains, S. aureus NCTC 8325, E. coli MG1655, and
P. aeruginosa PAO1. See text for details. c Arylomycin C16, included for reference.
activity against each of the other strains that is most pronounced
with P. aeruginosa. The loss in activity is even more pronounced
with compound 17 where methylation of D-Ala3 ablates activity
against all organisms tested.
To explore the effects of lipopeptide tail rigidity and to further
explore modifications that might directly compensate for the
resistance conferred by Pro29/84 of SPase, we designed the
hydroxyproline derivative 18 (Table 2). In this compound, the
side chain of D-MeSer2, which interacts with the side chain and
backbone of the SPase residue at position 29/84,23 is homo-
logated by a methylene unit and fused with the methyl group of
the neighboring N-methyl amide bond. We found that this
modification results in a complete loss of activity against the
Gram-negative organisms but only little to moderate loss in
activity against the Gram-positive organisms. Interestingly, be-
cause 18 retained full activity against resistant S. epidermidis, the
disparity in activities against the wild type and resistant variants is
greatly decreased, suggesting that at least for this organism, 18
recognizes both the Ser- and the Pro-variant SPases similarly.
To further explore the effects of decreased rigidity, we synthe-
sized derivative 19 (Table 2), which lacks the peptide bond between
the serine and the fatty acid tail (and thus should impart the tail with
greater rotational freedom). Compared to arylomycin C16 this
molecule exhibited significantly less activity against all organisms
tested, with no observable activity against E. coli, S. aureus, or
P. aeruginosa and only moderate activity against S. epidermidis.
4957
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
Table 2. Activity (MICs (μg/mL)) of Arylomycin Derivatives with Altered Lipopeptide Tails against Strains of S. epidermidis,
S. aureus, E. coli, and P. aeruginosa Harboring SPase without (Sensitive) and with (Resistant) the Arylomycin-Resistance
Conferring Pro Residue
a Sensitive strains include wild type S. epidermidis RP62A, lepB(P29S) S. aureus 8325, lepB(P84L) E. coli MG1655, and lepB(P84L) P. aeruginosa PAO1.
See text for details. b Resistant strains include spsB(S29P) S. epidermidis RP62A, and wild type strains, S. aureus 8325, E. coli MG1655, and P. aeruginosa
PAO1. See text for details. c Arylomycin C16, included for reference.
To increase flexibility without introducing or deleting other
peptidic functionalities that might contribute to binding SPase,
we synthesized and evaluated derivatives with one or two
methylene units inserted immediately N-terminal or C-term-
inal to the amide bond linking D-MeSer2 to D-Ala3 (20ꢀ23,
Table 2). These compounds did not gain activity against either
of the wild type Gram-negative bacteria. With the sensitized
Gram-negative strains, activity was observed only with 20,
which relative to the parent compound is 16-fold less active
against E. coli but only 2-fold less active against P. aeruginosa.
The effects of methylene addition were significantly different
with the Gram-positive bacteria. Relative to arylomycin C16,
derivatives 20ꢀ23 lost 8- to 16-fold activity against sensitive
S. epidermidis but retained activity against the resistant strain.
This demonstrates that once the H-bond donor of the protein is
removed (by mutation to Pro), perturbing the H-bond acceptor
does not further decrease activity. The results were somewhat
more complicated with S. aureus. As expected, relative to
arylomycin C16, addition of the methylene units decreased
activity against the sensitized strain of S. aureus, 2- to 8-fold
for 21ꢀ23 and at least 64-fold for 20. In the case of wild type
S. aureus, however, no activity is observed with 20 or 21, but
interestingly, 22 and especially 23 gain activity.
Compounds 11, 12, 15, 22, and 23 have increased activity
against wild type S. aureus relative to arylomycin C16. Because
these compounds also have increased hydrophobicity and/or
aromaticity, the increased gains could result from nonspecific
effects such as membrane destabilization. While no mutants of
S. aureus are available that are resistant to the arylomycins via a
mechanism other than Pro29, S. epidermidis is closely related and
the S31P mutation confers high levels of arylomycin resistance.
Derivatives whose activity results from nonspecific effects should
4958
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
still be active against this mutant; however, no activity was
observed (MIC > 64 μg/mL). These results suggest that the
increased activity of the derivatives against S. aureus does indeed
result specifically from SPase inhibition.
the plasma membrane of P. aeruginosa, such as the presence of
phosphatidylcholine,56ꢀ60 or from different constituent fatty
acids.61ꢀ64 For example, P. aeruginosa appears to employ a higher
percentage of cis-vaccenic acid (a C18 fatty acid) relative to
palmitic and palmitoleic acids (which are C16 fatty acids),61ꢀ64
possibly resulting in a slightly thicker plasma membrane and
possibly accounting for the generally longer fatty acid tail lengths
that were observed to be required for activity against P. aerugi-
nosa. Overall, the data collected with the different tail derivatives
suggest that the phenyl-modified derivatives are likely better
scaffolds for arylomycin optimization than the natural, saturated
fatty acid chains. This is most clearly highlighted by compound
15, which retains all of the activities of the parent compound
arylomycin C16 but also gains activity against S. aureus.
N-Methylation is common with nonribosomally synthesized
peptides such as the arylomycins and is generally thought to
optimize hydrophobicity, H-bonding potential, conformation,
and/or resistance to proteases.65ꢀ68 The peptide portion of the
arylomycin lipopeptide tail is backbone methylated at D-MeSer2
and MeHpg5 but not at D-Ala3 or Gly4. Previously, we showed
that the methyl group at MeHpg5 preorganizes the biaryl ring
system for recognition of SPase.20 When we altered the backbone
methylation state of D-MeSer2 and D-Ala3, which are both
proximal to the critical resistance-conferring Pro in the E. coli
SPase-arylomycin A2 complex,23 activity was lost against both
Gram-positive and Gram-negative bacteria. The slight decrease
in activity that results from removal of the D-MeSer2 N-methyl
group is unlikely to result from specific deleterious interactions
with the lipid membrane, because of membrane fluidity, or from
any interactions with SPase, as the structure of the E. coli
SPaseꢀarylomycin A2 complex suggests that this region of the
lipopeptide tail is either disordered or oriented away from the
protein (although as discussed above, the N-terminally truncated
form of SPase used in the structural studies renders this conclusion
somewhat speculative).23 Thus, the observed decrease in activity
is likely the result of decreased hydrophobicity, outer membrane
penetration, or protease resistance. The more pronounced loss in
activity observed upon methylation of D-Ala3 likely results from
replacement of a stabilizing H-bond with a destabilizing steric
clash. Whatever the specific origins of the decreased activity at
the two sites examined, the data suggest that natural lipopeptide
tail methylation pattern is already optimized for activity.
’ DISCUSSION
It is widely accepted that most if not all of the broad-
spectrum antibiotic scaffolds produced by the bacteria cul-
tured to date have been discovered38 and that the reduced
spectrum of the more commonly identified narrow spectrum
agents is a limitation that is intrinsic to their structure or
mechanism of action and thus difficult to overcome via scaffold
derivatization. However, just asresistanceplays a roleinthe clinical
arms race between humans and pathogens, resistance might also
play a role in the natural arms race between microbes for which
many natural product antibiotics were evolved, via selection or
genetic drift.39ꢀ44 This suggests that the reduced spectrum of
some narrow spectrum natural products might not be an intrinsic
limitation of their scaffold or target but rather the result of specific
resistance mechanisms of the sort that medicinal chemists have
overcome via derivatization. We have termed these natural
products “latent” antibiotics. Previously, we showed that the
arylomycins have a broader spectrum of activity than previously
appreciated and that both innate and acquired resistance is
commonly caused by the presence of a specific Pro residue that
reduces the affinity of inhibitor binding.21 This specific mechanism
of resistance suggests that the arylomycins may be latent anti-
biotics, and thus that their spectrum of activity might be optimized
by derivatization, and also focused our efforts to optimize them on
increasing the affinity with which they bind SPase.
We began testing the hypothesis that the arylomycins are
latent antibiotics by exploring tail modifications, as interactions
with the lipopeptide tail are disrupted by the resistance-confer-
ring Pro. We first focused attention on the lipid portion of the
lipopeptide tail which likely embeds into the cytoplasmic mem-
brane and increases the effective concentration of the inhibitor in
the vicinity of SPase, as has been suggested with other antibiotics
that possess lipid tails and inhibit membrane-bound targets.45ꢀ55
However, how far submerged the arylomycin lipopeptide is
within the membrane is not known, nor is the optimal length
of the fatty acid alkyl chain that may be inserted into the
membranes of Gram-positive or Gram-negative bacteria. Our
data suggest that the lipopeptide tail enters the hydrophobic
environment of the membrane, or the membrane-SPase inter-
face, C-terminal to the fatty acid carbonyl, that among the
saturated fatty acid derivatives the C16 analogue is optimal for
activity, and that the inhibition of P. aeruginosa generally requires
slightly longer fatty acid tails.
In an effort to more directly compensate for the negative
interactions introduced by the resistance-conferring Pro, we
synthesized several derivatives with increased or decreased
flexibility around D-MeSer2 and D-Ala3. While none of these
derivatives gained activity against the wild type Gram-negative
bacteria or against S. epidermidis, 22 and especially 23 gain
significant activity against wild type S. aureus. While the precise
mechanism by which these derivatives gain activity against
S. aureus remains to be determined, the data nonetheless support
the possibility that the spectrum of the arylomycins may be
optimized by derivatization.
The phenyl- and biphenyl-fatty acid tail series showed similar
activities against S. epidermidis, S. aureus, and E. coli, with the
longer p-alkyl derivatives having activity against both sensitive
and resistant strains of S. aureus. Interestingly, P. aeruginosa again
shows unique behavior, as it is not inhibited by any of the
biphenyl-modified derivatives. This is particularly noteworthy
considering that it is inhibited by the C8- and C10-substituted
phenyl analogues, which in some cases are less hydrophobic.
While some of the differences may result from altered outer
membrane penetration or in vivo stability, the data likely reflect
suboptimal insertion of the arylomycin into the plasma mem-
brane of P. aeruginosa. It is interesting to speculate that this might
result from unique aspects of the phospholipids that comprise
’ CONCLUSION
From a practical perspective, the data reveal that both the
methylation state and the length of the straight chain fatty acid of
the lipopeptide tail of the natural arylomycins are already
optimized for activity but that the unnatural phenyl analogues
are more promising scaffolds. From a conceptual perspective, the
identification of derivatives with an expanded spectrum against
4959
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
wild type S. aureus strains, most notably 12 and 23, supports the
hypothesis that arylomycins are latent antibiotics and focuses
attention on the types of lipopeptide tail modifications that may
be most likely to optimize their spectrum. Finally, this study
identified derivatives that provide an improved scaffold for
further optimization efforts focused on other parts of the
molecule, for example, the macrocyclic ring core that interacts
with parts of SPase that are proximal to the S1 and S3 binding
pockets,13,22,24 which are more traditional targets for peptidase
inhibitor optimization. Such efforts are currently underway.
(90.3 mg, 3 equiv). The mixture was allowed to stir overnight, after
which water, saturated NaHCO3, and EtOAc were added, the aqueous
phase was extracted 3ꢁ with EtOAc, and the combined organic layers
were washed with 5% citric acid (pH 3) and brine. The organics were
dried over sodium sulfate and concentrated. The crude material was
purified by column chromatography (5.5% MeOH in DCM) to give the
protected arylomycin C16 (72.4 mg, 45% yield).
General Procedure B: Global Deprotection. Example: Ar-
ylomycin C16. The fully protected arylomycin20 (72.4 mg, 72 μmol,
1 equiv) was dissolved in ethanethiol (2 mL) under Ar and treated with 1.0 M
AlBr3 in CH2Br2 (1.79 mL, 25 equiv). The reaction vial was sealed and heated
to 50 °C and stirred for 4 h. The mixture was cooled to room temperature,
MeOH was added (0.5 mL), and the volatiles were evaporated under a stream
of nitrogen. MeOH was added again and was evaporated under a stream of
nitrogen, and the crude product was dried under vacuum. The crude product
was then dissolved in MeOH and purified by HPLC (linear gradient, 0.67%
B/min, product eluted at 80% B) to give arylomycin C16 (32.6 mg, 51%
yield). For 1H NMR and 13C NMR see Supporting Information. ESI HRMS
calcd for C46H69N6O11 [(M þ H)þ], 881.5019; found, 881.5021.
General Procedure C: Macrocycle and Tail Coupling. Ex-
ample: Compound 5. The procedure is based on the conditions
reported previously.69 The lipid tripeptide tail (23.5 mg, 52 μmol, 1
equiv) and the tripeptide macrocycle (70 mg, 2.2 equiv) were dissolved
in THF (2 mL) under Ar and treated with TEA (7 μL, 1 equiv) and
DEPBT (39 mg, 2.5 equiv). The mixture was allowed to stir overnight.
Then the volatiles were evaporated under a stream of nitrogen, the
residue was dried under vacuum, and EtOAc and saturated NaHCO3
were added. The aqueous layer was extracted. Then the organic layer was
washed with 0.1 N HCl, dried over sodium sulfate, and concentrated.
Compound 1 was synthesized using general procedures A and B. For
1H NMR see Supporting Information. ESI HRMS calcd for
C46H70N6O10 [(M þ H)þ], 867.5226; found, 867.5207.
’ EXPERIMENTAL SECTION
1
General Methods. H and 13C NMR spectra were recorded on
Bruker AMX 400, Bruker DRX 500, or Bruker DRX 600 spectrometers.
Chemical shifts are reported as δ values (parts per million, ppm) relative to
chloroform (δ 7.26), methanol (δ 3.31), or dimethylsulfoxide (DMSO)
(δ2.50) for 1H NMR and to chloroform (δ77.16), methanol (δ49.00), or
DMSO (δ 39.52) for 13C NMR. High-resolution time-of-flight mass
spectra (HRMS) were measured at the Scripps Center for Mass Spectro-
metry. ESI mass spectra were measured on either an HP series 1100 MSD
(accuracy of 0.1 amu) or a PESCIEX API/Plus (accuracy of 0.5 amu)
instrument. Yields refer to chromatographically and spectroscopically pure
compounds unless otherwise stated, with the purity of all compounds
determined to be >95% by HPLC. Mixtures were magnetically stirred, and
reactions were monitored by thin layer chromatography (TLC) with
0.25 mm Whatman precoated silica gel (with fluorescence indicator)
plates. Flash chromatography was performed with silica gel (particle size
40ꢀ63 μm, EMD Chemicals). Dry solvents were purchased from Acros.
Anhydrous 1-hydroxybenzotriazole (HOBt) was purchased from Chem-
Impex. All other chemicals were purchased from Fisher/Acros or Aldrich.
All preparative reverse phase chromatography was performed using two
Dynamax SD-200 pumps connected to a Dynamax UV-D II detector
(monitoring at 220 nm) on a Phenomenex Jupiter C18 column (10 μm,
2.12 cm ꢁ 25 cm, 300 Å pore size). All solvents contained 0.1% TFA: solvent
A, H2O; solvent B, 90% acetonitrile/10% H2O. All samples were loaded onto
the column at 0% B, and the column was allowed to equilibrate for ∼10 min
before a linear gradient was started. Retention times are reported according
to the linear gradient used and the %B at the time the sample eluted.
Compound 19 was synthesized racemically, and the two diastereo-
mers were separated and subjected to biological assay independently.
The stereochemistry of each of the compounds was unknown, but the
MICs for the most active compound are reported and discussed in the
main text. Compounds 20 and 21 were also synthesized racemically;
however, the diastereomers were not separated and the compounds
were assayed as racemic mixtures.
Biological Methods. Minimum inhibitory concentrations (MICs)
were performed in at least triplicate using a modified CLSI broth micro-
dilution method in cation adjusted MuellerꢀHinton broth for all strains.
Briefly, by use of cultures inoculated from a fresh overnight plate scrape, 96-
well flat bottom plates containing media and compounds at the appropriate
concentrations were inoculated to a final density of 5 ꢁ 105 cfu/mL. Plates
were sealed with oxygen permeable membranes and incubated for 24 h at
37 °C. Following incubation, seals were removed and optical density at
590 nm was measured using a Perkin-Elmer Envision 2103 multilabel
reader. MICs were then confirmed by measuring viable cell count by serial
dilution in phosphate buffered saline and plating on MuellerꢀHinton II
agar (BD Diagnostics). The MIC values given are the median of at least
three independent replicates with no more than 2-fold deviation.
Compound 2 was synthesized using general procedures A and B. For
1H NMR and 13C NMR see Supporting Information. ESI HRMS calcd
for C38H53N6O11 [(M þ H)þ], 769.3767; found, 769.3770.
Compound 3 was synthesized using general procedures A and B. For
1H NMR and 13C NMR see Supporting Information. ESI HRMS calcd
for C40H57N6O11 [(M þ H)þ], 797.408; found, 797.4070.
Compound 4 was synthesized using general procedures A and B. For
1H NMR and 13C NMR see Supporting Information. ESI HRMS calcd
for C42H61N6O11 [(M þ H)þ], 825.4393; found, 825.4386.
Compound 5 was synthesized as described in general procedure C and
then subjected to general procedure B to give the product (20.6 mg, 58%
yield). For 1H NMR and 13C NMR see Supporting Information. ESI HRMS
calcd for C48H72N6O11 [(M þ H)þ], 909.5332; found, 909.5328.
Compound 6 was synthesized using general procedures A and B. For 1H
NMR and 13C NMR see Supporting Information. ESI HRMS calcd for
C43H54N6O11 [(M þ H)þ], 811.3297; found, 811.3300.
Compound 7 was synthesized using general procedures A and B. For
1H NMR see Supporting Information. ESI HRMS calcd for
C44H48N6O11 [(M þ H)þ], 837.3454; found, 837.3443.
Compound 8 was synthesized using general procedures A and B. For
1H NMR and 13C NMR see Supporting Information. ESI HRMS calcd
for C43H46N6O11 [(M þ H)þ], 823.3297; found, 823.3296.
Compound 9 was synthesized using general procedures A and B. For
1H NMR see Supporting Information. ESI HRMS calcd for
C45H50N6O11 [(M þ H)þ], 851.361; found, 851.359.
Compound 10 was synthesized using general procedures A and B.
For 1H NMR and 13C NMR see Supporting Information. ESI
HRMS calcd for C47H54N6O11 [(M þ H)þ], 879.3923; found,
879.3924.
General Procedure A: Macrocycle and Tail Coupling. Ex-
ample: Arylomycin C16. The glycine homologated macrocycle
(Scheme 1)20 (80 mg, 0.16 mmol) was taken up in CH3CN (7.2 mL)
and DMF (3.2 mL) and treated sequentially with HOBt (64 mg, 3
equiv), the lipopeptide tail (Scheme 1)20 (81.3 mg, 1 equiv), and EDC
Compound 11 was synthesized using general procedures A and B.
For 1H NMR see Supporting Information. ESI HRMS calcd for
C49H58N6O11 [(M þ H)þ], 907.4236; found, 907.4246.
4960
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
Compound 12 was synthesized using general procedures A and B.
For 1H NMR see Supporting Information. ESI HRMS calcd for
C51H62N6O11 [(M þ H)þ], 935.4549; found, 935.4548.
N-dimethylformamide; EDC, 1-(3-dimethylaminopropyl)-3-
ethylcarbodiimide hydrochloride; EtOAc, ethyl acetate; Hex,
hexanes; Ar, argon; TFA, trifluoroacetic acid
Compound 13 was synthesized using general procedures A and B.
For 1H NMR see Supporting Information. ESI HRMS calcd for
C43H54N6O11 [(M þ H)þ], 831.3923; found, 831.3917.
’ REFERENCES
(1) Miller, L. G.; Kaplan, S. L. Staphylococcus aureus: a community
pathogen. Infect. Dis. Clin. North Am. 2009, 23, 35–52.
(2) Pearson, A. Historical and changing epidemiology of healthcare-
associated infections. J. Hosp. Infect. 2009, 73, 296–304.
(3) Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Pompliano, D. L.
Drugs for bad bugs: confronting the challenges of antibacterial discov-
ery. Nat. Rev. Drug Discovery 2007, 6, 29–40.
(4) Clardy, J.; Fischbach, M. A.; Walsh, C. T. New antibiotics from
bacterial natural products. Nat. Biotechnol. 2006, 24, 1541–1550.
(5) Gringauz, A. Introduction to Medicinal Chemistry: How Drugs Act
and Why; Wiley-VCH: Weinheim, Germany, 1996.
(6) Schneider, P.; Hawser, S.; Islam, K. Iclaprim, a novel diamino-
pyrimidine with potent activity on trimethoprim sensitive and resistant
bacteria. Bioorg. Med. Chem. Lett. 2003, 13, 4217–4221.
(7) von Nussbaum, F.; Brands, M.; Hinzen, B.; Weigand, S.; Habich,
D. Antibacterial natural products in medicinal chemistry: exodus or
revival? Angew. Chem., Int. Ed. 2006, 45, 5072–5129.
(8) Sharma, P. C.; Jain, A.; Jain, S. Fluoroquinolone antibacterials: a
review on chemistry, microbiology and therapeutic prospects. Acta Pol.
Pharm. 2009, 66, 587–604.
(9) Durr, U. H.; Sudheendra, U. S.; Ramamoorthy, A. LL-37, the
only human member of the cathelicidin family of antimicrobial peptides.
Biochim. Biophys. Acta 2006, 1758, 1408–1425.
(10) Gottler, L. M.; Ramamoorthy, A. Structure, membrane orienta-
tion, mechanism, and function of pexiganan—a highly potent antimi-
crobial peptide designed from magainin. Biochim. Biophys. Acta 2009,
1788, 1680–1686.
(11) Carlos, J. L.; Paetzel, M.; Brubaker, G.; Karla, A.; Ashwell,
C. M.; Lively, M. O.; Cao, G.; Bullinger, P.; Dalbey, R. E. The role of the
membrane-spanning domain of type I signal peptidases in substrate
cleavage site selection. J. Biol. Chem. 2000, 275, 38813–38822.
(12) Paetzel, M.; Dalbey, R. E.; Strynadka, N. C. The structure and
mechanism of bacterial type I signal peptidases. A novel antibiotic target.
Pharmacol. Ther. 2000, 87, 27–49.
Compound 14 was synthesized using general procedures A and B.
1
For H NMR and 13C NMR see Supporting Information. ESI HRMS
calcd for C45H58N6O11 [(M þ H)þ], 859.4236; found, 859.4231.
Compound 15 was synthesized using general procedures A and B.
1
For H NMR and 13C NMR see Supporting Information. ESI HRMS
calcd for C47H62N6O11 [(M þ H)þ], 887.4549; found, 887.4539.
Compound 16 was synthesized using general procedures A
1
and B. For H NMR and 13C NMR see Supporting Information. ESI
HRMS calcd for C45H66N6O11 [(M þ H)þ], 867.4862; found, 867.4873.
Compound 17 was synthesized using general procedures A and B.
For 1H NMR see Supporting Information. ESI HRMS calcd for
C47H70N6O10 [(M þ H)þ], 895.5175; found, 895.5190.
Compound 18 was synthesized using general procedures A and B. For
1H NMR and 13C NMR see Supporting Information. ESI HRMS calcd for
C47H68N6O11 [(M þ H)þ], 893.5019; found, 893.5014.
Diastereomer A of compound 19 was synthesized using general
procedures A and B. For 1H NMR and 13C NMR see Supporting
Information. ESI HRMS calcd for C43H63N5O9 [(M þ H)þ], 794.4698;
found, 794.4705. Diastereomer B of compound 19 was synthesized
1
using general procedures A and B. For H NMR see Supporting In-
formation. ESI HRMS calcd for C43H63N5O9 [(M þ H)þ], 794.4698;
found, 794.4689.
Compound 20 was synthesized racemically using general procedures
C and B. For 1H NMR see Supporting Information. ESI HRMS calcd for
C47H70N6O11 [(M þ H)þ], 895.5175; found, 895.5180.
Compound 21 was synthesized racemically using general procedures
C and B. For 1H NMR see Supporting Information. ESI HRMS calcd for
C48H72N6O11 [(M þ H)þ], 909.5332; found, 909.5334.
Compound 22 was synthesized using general procedures A
1
and B. For H NMR and 13C NMR see Supporting Information.
ESI HRMS calcd for C47H70N6O11 [(M þ H)þ], 895.5175; found,
895.5178.
Compound 23 was synthesized using general procedures A and B.
1
For H NMR and 13C NMR see Supporting Information. ESI HRMS
(13) Paetzel, M.; Karla, A.; Strynadka, N. C.; Dalbey, R. E. Signal
peptidases. Chem. Rev. 2002, 102, 4549–4580.
calcd for C48H72N6O11 [(M þ H)þ], 909.5332; found, 909.5305.
’ ASSOCIATED CONTENT
(14) Holtzel, A.; Schmid, D. G.; Nicholson, G. J.; Stevanovic, S.;
Schimana, J.; Gebhardt, K.; Fiedler, H. P.; Jung, G. Arylomycins A and B,
new biaryl-bridged lipopeptide antibiotics produced by Streptomyces sp.
T€u 6075. II. Structure elucidation. J. Antibiot. 2002, 55, 571–577.
(15) Kulanthaivel, P.; Kreuzman, A. J.; Strege, M. A.; Belvo, M. D.;
Smitka, T. A.; Clemens, M.; Swartling, J. R.; Minton, K. L.; Zheng, F.;
Angleton, E. L.; Mullen, D.; Jungheim, L. N.; Klimkowski, V. J.; Nicas,
T. I.; Thompson, R. C.; Peng, S. B. Novel lipoglycopeptides as inhibitors
of bacterial signal peptidase I. J. Biol. Chem. 2004, 279, 36250–36258.
(16) Schimana, J.; Gebhardt, K.; Holtzel, A.; Schmid, D. G.;
Sussmuth, R.; Muller, J.; Pukall, R.; Fiedler, H. P. Arylomycins A and B,
new biaryl-bridged lipopeptide antibiotics produced by Streptomyces
sp. T€u 6075. I. Taxonomy, fermentation, isolation and biological activities.
J. Antibiot. 2002, 55, 565–570.
S
Supporting Information. Experimental procedures and
b
1H, 13C, MS, and HRMS data. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: þ1 (858) 784-7290. Fax: þ1 (858) 784-7472. E-mail:
floyd@scripps.edu.
(17) Cragg, G. M.; Newman, D. J.; Snader, K. M. Natural products in
drug discovery and development. J. Nat. Prod. 1997, 60, 52–60.
(18) Dalbey, R. E.; Lively, M. O.; Bron, S.; van Dijl, J. M. The
chemistry and enzymology of the type I signal peptidases. Protein Sci.
1997, 6, 1129–1138.
(19) Date, T. Demonstration by a novel genetic technique that
leader peptidase is an essential enzyme of Escherichia coli. J. Bacteriol.
1983, 154, 76–83.
’ ACKNOWLEDGMENT
This work was supported by the Office of Naval Research
(Awards N000140310126 and N000140810478) and the
National Institutes of Health (Grant AI081126).
’ ABBREVIATIONS USED
(20) Roberts, T. C.; Smith, P. A.; Cirz, R. T.; Romesberg, F. E.
Structural and initial biological analysis of synthetic arylomycin A(2).
J. Am. Chem. Soc. 2007, 129, 15830–15838.
SPase, signal peptidase I; MIC, minimal inhibitory concentra-
tion; THF, tetrahydrofuran; DCM, dichloromethane; DMF, N,
4961
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
(21) Smith, P. A.; Roberts, T. C.; Romesberg, F. E. Broad spectrum
antibiotic activity of the arylomycin natural products is masked by
natural target mutations. Chem. Biol. 2010, 17, 1223–1231.
(22) Schechter, I.; Berger, A. On the size of the active site in
proteases. I. Papain. Biochem. Biophys. Res. Commun. 1967, 27, 157–162.
(23) Paetzel, M.; Goodall, J. J.; Kania, M.; Dalbey, R. E.; Page,
M. G. P. Crystallographic and biophysical analysis of a bacterial signal
peptidase in complex with a lipopeptide-based inhibitor. J. Biol. Chem.
2004, 279, 30781–30790.
(24) Paetzel, M.; Dalbey, R. E.; Strynadka, N. C. Crystal structure
of a bacterial signal peptidase apoenzyme. J. Biol. Chem. 2002,277, 9512–9519.
(25) Paetzel, M.; Dalbey, R. E.; Strynadka, N. C. Crystal structure of
a bacterial signal peptidase in complex with a beta-lactam inhibitor.
Nature 1998, 396, 186–190.
(43) Czaran, T. L.; Hoekstra, R. F.; Pagie, L. Chemical warfare
between microbes promotes biodiversity. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 786–790.
(44) Lynch, M. The frailty of adaptive hypotheses for the origins of
organismal complexity. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (Suppl. 1),
8597–8604.
(45) Allen, N. E.; LeTourneau, D. L.; Hobbs, J. N., Jr. The role of
hydrophobic side chains as determinants of antibacterial activity of
semisynthetic glycopeptide antibiotics. J. Antibiot. (Tokyo) 1997,
50, 677–684.
(46) Chen, L.; Yuan, Y.; Helm, J. S.; Hu, Y.; Rew, Y.; Shin, D.; Boger,
D. L.; Walker, S. Dissecting ramoplanin: mechanistic analysis of
synthetic ramoplanin analogues as a guide to the design of improved
antibiotics. J. Am. Chem. Soc. 2004, 126, 7462–7463.
(26) Luo, C.; Roussel, P.; Dreier, J.; Page, M. G.; Paetzel, M.
Crystallographic analysis of bacterial signal peptidase in ternary complex
with arylomycin A2 and a beta-sultam inhibitor. Biochemistry 2009,
48, 8976–8984.
(27) Beauregard, D. A.; Williams, D. H.; Gwynn, M. N.; Knowles, D. J.
Dimerization and membrane anchors in extracellular targeting of vanco-
mycin group antibiotics. Antimicrob. AgentsChemother. 1995, 39, 781–785.
(28) Breukink, E.; de Kruijff, B. Lipid II as a target for antibiotics.
Nat. Rev. Drug Discovery 2006, 5, 321–332.
(29) Kim, S. J.; Schaefer, J. Hydrophobic side-chain length deter-
mines activity and conformational heterogeneity of a vancomycin
derivative bound to the cell wall of Staphylococcus aureus. Biochemistry
2008, 47, 10155–10161.
(30) Nagarajan, R. Structureꢀactivity relationships of vancomycin-type
glycopeptide antibiotics. J. Antibiot. (Tokyo) 1993, 46, 1181–1195.
(31) Thennarasu, S.; Lee, D. K.; Tan, A.; Prasad Kari, U.;
Ramamoorthy, A. Antimicrobial activity and membrane selective inter-
actions of a synthetic lipopeptide MSI-843. Biochim. Biophys. Acta 2005,
1711, 49–58.
(32) Majerle, A.; Kidric, J.; Jerala, R. Enhancement of antibacterial
and lipopolysaccharide binding activities of a human lactoferrin peptide
fragment by the addition of acyl chain. J. Antimicrob. Chemother. 2003,
51, 1159–1165.
(33) Rustici, A.; Velucchi, M.; Faggioni, R.; Sironi, M.; Ghezzi, P.;
Quataert, S.; Green, B.; Porro, M. Molecular mapping and detoxification of
the lipid A binding site by synthetic peptides. Science 1993, 259, 361–365.
(34) Dufour, J.; Neuville, L.; Zhu, J. P. Total synthesis of arylomycin
A(2), a signal peptidase I (SPase I) inhibitor. Synlett 2008, 15, 2355–2359.
(35) Heller, H.; Schaefer, M.; Schulten, K. Molecular dynamics
simulation of a bilayer of 200 lipids in the gel and in the liquid crystal
phase. J. Phys. Chem. 1993, 97, 8343–8360.
(36) Wang, Y.; Bruckner, R.; Stein, R. L. Regulation of signal
peptidase by phospholipids in membrane: characterization of phospho-
lipid bilayer incorporated Escherichia coli signal peptidase. Biochemistry
2004, 43, 265–270.
(37) van Klompenburg, W.; Paetzel, M.; de Jong, J. M.; Dalbey, R. E.;
Demel, R. A.; von Heijne, G.; de Kruijff, B. Phosphatidylethanolamine
mediates insertion of the catalytic domain of leader peptidase in
membranes. FEBS Lett. 1998, 431, 75–79.
(47) Cooper, M. A.; Williams, D. H. Binding of glycopeptide
antibiotics to a model of a vancomycin-resistant bacterium. Chem. Biol.
1999, 6, 891–899.
(48) Dong, S. D.; Oberthur, M.; Losey, H. C.; Anderson, J. W.; Eggert,
U. S.; Peczuh, M. W.; Walsh, C. T.; Kahne, D. The structural basis for
induction of VanB resistance. J. Am. Chem. Soc. 2002, 124, 9064–9065.
(49) Kerns, R.; Dong, S. D.; Fukuzawa, S.; Carbeck, J.; Kohler, J.;
Silver, L.; Kahne, D. The role of hydrophobic substituents in the
biological activity of glycopeptide antibiotics. J. Am. Chem. Soc. 2000,
122, 12608–12609.
(50) Mackay, J. P.; Gerhard, U.; Beauregard, D. A.; Maplestone,
R. A.; Williams, D. H. Dissection of the contributions toward dimeriza-
tion of glycopeptide antibiotics. J. Am. Chem. Soc. 1994, 116, 4573–4580.
(51) Maffioli, S. I.; Ciabatti, R.; Romano, G.; Marzorati, E.;
Preobrazhenskaya, M.; Pavlov, A. Synthesis and antibacterial activity
of alkyl derivatives of the glycopeptide antibiotic A40926 and their
amides. Bioorg. Med. Chem. Lett. 2005, 15, 3801–3805.
(52) Nagarajan, R.; Schabel, A. A.; Occolowitz, J. L.; Counter, F. T.;
Ott, J. L. Synthesis and antibacterial activity of N-acyl vancomycins.
J. Antibiot. (Tokyo) 1988, 41, 1430–1438.
(53) Nagarajan, R.; Schabel, A. A.; Occolowitz, J. L.; Counter, F. T.;
Ott, J. L.; Felty-Duckworth, A. M. Synthesis and antibacterial evaluation
of N-alkyl vancomycins. J. Antibiot. (Tokyo) 1989, 42, 63–72.
(54) Rodriguez, M. J.; Snyder, N. J.; Zweifel, M. J.; Wilkie, S. C.;
Stack, D. R.; Cooper, R. D.; Nicas, T. I.; Mullen, D. L.; Butler, T. F.;
Thompson, R. C. Novel glycopeptide antibiotics: N-alkylated deriva-
tives active against vancomycin-resistant enterococci. J. Antibiot. (Tokyo)
1998, 51, 560–569.
(55) Sharman, G. J.; Try, A. C.; Dancer, R. J.; Cho, Y. R.; Staroske,
T.; Bardsley, B.; Maguide, A.; Cooper, M. A.; O’Brien, D. P.; Williams,
D. H. The roles of dimerization and membrane anchoring in activity of
glycopeptide antibiotics against vancomycin-resistant bacteria. J. Am.
Chem. Soc. 1997, 1997, 12041–12047.
(56) Albelo, S. T.; Domenech, C. E. Carbons from choline present in
the phospholipids of Pseudomonas aeruginosa. FEMS Microbiol. Lett.
1997, 156, 271–274.
(57) Wilderman, P. J.; Vasil, A. I.; Martin, W. E.; Murphy, R. C.;
Vasil, M. L. Pseudomonas aeruginosa synthesizes phosphatidylcholine by
use of the phosphatidylcholine synthase pathway. J. Bacteriol. 2002,
184, 4792–4799.
(38) Baltz, R. H. Marcel Faber Roundtable: Is our antibiotic pipeline
unproductive because of starvation, constipation or lack of inspiration?
J. Ind. Microbiol. Biotechnol. 2006, 33, 507–513.
(39) Martinez, J. L. The role of natural environments in the
evolution of resistance traits in pathogenic bacteria. Proc. Biol. Sci.
2009, 276, 2521–2530.
(58) Sohlenkamp, C.; Lopez-Lara, I. M.; Geiger, O. Biosynthesis of
phosphatidylcholine in bacteria. Prog. Lipid Res. 2003, 42, 115–162.
(59) Cronan, J. E.; Vagelos, P. R. Metabolism and function of the
membrane phospholipids of Escherichia coli. Biochim. Biophys. Acta 1972,
265, 25–60.
(40) D’Costa, V. M.; Griffiths, E.; Wright, G. D. Expanding the soil
antibiotic resistome: exploring environmental diversity. Curr. Opin.
Microbiol. 2007, 10, 481–489.
(41) Allen, H. K.; Donato, J.; Wang, H. H.; Cloud-Hansen, K. A.;
Davies, J.; Handelsman, J. Call of the wild: antibiotic resistance genes in
natural environments. Nat. Rev. Microbiol. 2010, 8, 251–259.
(42) Laskaris, P.; Tolba, S.; Calvo-Bado, L.; Wellington, L. Coevolu-
tion of antibiotic production and counter-resistance in soil bacteria.
Environ. Microbiol. 2010, 12, 783–796.
(60) Brundish, D. E.; Shaw, N.; Baddiley, J. The phospholipids of
Pneumococcus I-192R, A.T.C.C. 12213. Some structural rearrange-
ments occurring under mild conditions. Biochem. J. 1967, 104, 205–211.
(61) Mechin, L.; Dubois-Brissonnet, F.; Heyd, B.; Leveau, J. Y.
Adaptation of Pseudomonas aeruginosa ATCC 15442 to didecyldimethyl-
ammonium bromide induces changes in membrane fatty acid composi-
tion and in resistance of cells. J. Appl. Microbiol. 1999, 86, 859–866.
(62) Oliver, J. D.; Colwell, R. R. Extractable lipids of Gram-negative
marine bacteria: phospholipid composition. J. Bacteriol. 1973, 114, 897–908.
4962
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963

Journal of Medicinal Chemistry
ARTICLE
(63) Nielsen, L. E.; Kadavy, D. R.; Rajagopal, S.; Drijber, R.;
Nickerson, K. W. Survey of extreme solvent tolerance in Gram-positive
cocci: membrane fatty acid changes in Staphylococcus haemolyticus grown
in toluene. Appl. Environ. Microbiol. 2005, 71, 5171–5176.
(64) De Siervo, A. J. Alterations in the phospholipid composition of
Escherichia coli B during growth at different temperatures. J. Bacteriol.
1969, 100, 1342–1349.
(65) Chatterjee, J.; Gilon, C.; Hoffman, A.; Kessler, H. N-Methyla-
tion of peptides: a new perspective in medicinal chemistry. Acc. Chem.
Res. 2008, 41, 1331–1342.
(66) Conradi, R. A.; Hilgers, A. R.; Ho, N. F.; Burton, P. S. The
influence of peptide structure on transport across Caco-2 cells. II.
Peptide bond modification which results in improved permeability.
Pharm. Res. 1992, 9, 435–439.
(67) De Zotti, M.; Biondi, B.; Formaggio, F.; Toniolo, C.; Stella, L.;
Park, Y.; Hahm, K. S. Trichogin GA IV: an antibacterial and protease-
resistant peptide. J. Pept. Sci. 2009, 15, 615–619.
(68) Walsh, C. T.; Chen, H.; Keating, T. A.; Hubbard, B. K.; Losey,
H. C.; Luo, L.; Marshall, C. G.; Miller, D. A.; Patel, H. M. Tailoring
enzymes that modify nonribosomal peptides during and after chain
elongation on NRPS assembly lines. Curr. Opin. Chem. Biol. 2001,
5, 525–534.
(69) Dufour, J.; Neuville, L.; Zhu, J. Intramolecular SuzukiꢀMiyaura
reaction for the total synthesis of signal peptidase inhibitors, arylomycins
A(2) and B(2). Chem.—Eur. J. 2010, 16, 10523–10534.
4963
dx.doi.org/10.1021/jm1016126 |J. Med. Chem. 2011, 54, 4954–4963