Structure-activity relationship study based on autoinducing peptide (AIP) from dog pathogen S. schleiferi

Article abstract of DOI:10.1021/acs.orglett.7b02550

Herein, an effective protocol for solid-phase synthesis of peptide thiolactones by concomitant ring closure and cleavage from the solid support is reported. The strategy was applied for mapping the importance of the structural features in S. schleiferi AIP (5) by performing an alanine scan and truncation of this natural compound. This furnished some of the most potent inhibitors of accessory gene regulator (agr)-I in the human pathogen S. aureus reported to date.

Full text of DOI:10.1021/acs.orglett.7b02550

Letter
pubs.acs.org/OrgLett
Structure−Activity Relationship Study Based on Autoinducing
Peptide (AIP) from Dog Pathogen S. schleiferi
Bengt H. Gless,† Pai Peng,# Katja D. Pedersen,‡ Charlotte H. Gotfredsen,‡ Hanne Ingmer,#
and Christian A. Olsen*,†
†Center for Biopharmaceuticals and Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences,
University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen, Denmark
#Department of Veterinary and Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, DK-1870
Frederiksberg C, Denmark
‡Department of Chemistry, Technical University of Denmark, Kemitorvet 207, Kgs. Lyngby DK-2800, Denmark
S
* Supporting Information
ABSTRACT: Herein, an effective protocol for solid-phase
synthesis of peptide thiolactones by concomitant ring closure
and cleavage from the solid support is reported. The strategy
was applied for mapping the importance of the structural
features in S. schleiferi AIP (5) by performing an alanine scan
and truncation of this natural compound. This furnished some
of the most potent inhibitors of accessory gene regulator (agr)-
I in the human pathogen S. aureus reported to date.
he increased emergence of bacterial pathogens that are
resistant to so-called last-resort antibiotics, in combination
S. aureus strains may be divided into four groups (I−IV)
based on their secreted AIP (1−4; Figure 1B). The AIPs have a
length of 7−9 amino acids and are cyclized by thiolactone
formation between the C-terminal carboxylate and the cysteine
residue in the fifth position from the C-terminal.4 AIPs activate
their cognate AgrC receptor and competitively inhibit the agr-
dependent signaling circuit of other groups (Figure 1A).8 The
function of this cross-inhibitory mechanism remains unclear,
but niche competition between groups has been hypothesized.
The agr-dependent QS signaling is conserved for other
staphylococci, and interspecies communication between strains
through agr inhibition has been reported.9 Structure−activity
relationship (SAR) studies based on S. aureus AIP-I, -II, and -III
(1−3),3,4,10 as well as other staphylococcal AIPs, have been
investigated in the search for potent QS inhibitors and to study
the biology of agr signaling in staphylococci.11 In a recent study,
we demonstrated cross-talk between S. aureus and 14 other
staphylococcal strains and identified the dog pathogen S.
schleiferi as a strong suppresser of S. aureus virulence through S.
schleiferi AIP (5, Figure 1B).9d The exceptionally strong
interspecies communication between S. aureus and S. schleiferi
is intriguing and could be indicative of evolutionary niche
competition. In this Letter, we therefore investigate the
structural requirements of the activity of the S. schleiferi AIP
against S. aureus through chemical synthesis of AIPs and
analogues.
T
with the lack of novel antibiotics brought to market, represents
a major challenge for the public health sector and requires the
development of alternative treatments.1 An alternative to
conventional antibiotic treatment could be to target the
virulence of the pathogen rather than the viability, which may
circumvent evolutionary pressure and, in turn, potentially also
resistance development.2 In the pathogenic bacterium Staph-
ylococcus aureus, virulence gene expression is regulated through
a quorum sensing (QS) system, an intercellular communication
process that generates population-wide behavioral responses
depending on the cellular density.3 Inhibiting QS signaling in S.
aureus was identified as a potential new drug target in the late
1990s4 and has been shown to attenuate infections.5
The communication from cell to cell is based on secreted
autoinducing peptides (AIPs), which are detected by the
cognate QS system and activate the system when sufficiently
high concentrations are present.4,6 Thus, these peptide
pheromones are extracellular signaling components, which are
expressed as part of the accessory gene regulator (agr), a
chromosomal locus consisting of two divergent operons
transcribed from the P2 and P3 promoters, respectively (Figure
1A).3b The RNAII transcript encodes AgrB and AgrD,
responsible for AIP synthesis, as well as a two component
signal transduction system with membrane bound AgrC
histidine kinase and the AgrA response regulator.3a The P3
promoter expresses RNAIII, which is the effector molecule of
agr and regulator of the expression of virulence factors.7 This
population-controlled induction of the QS system allows the
bacteria to grow to a certain density before engaging in a
synchronized attack on the host, turning virulent and invasive.3d
The first synthetic AIP molecules from S. epidermis were
prepared by thiolactonization using dicyclohexylcarbodiimide
Received: August 17, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b02550
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
optimization step resulted in significant reduction of by-
products and gave good yields for Ss-AIP (5), but for AIP-I (1),
small amounts of impurities were still present (Figure S1). We
therefore envisioned that the cyclization could be performed
on-resin to release only the cyclized product into solution while
leaving the byproducts attached to the solid support (Figure
2C). Thus, the MeDbz linker was coupled directly to the
ChemMatrix resin (10), enabling global deprotection of the
resin-bound linear peptide after Nbz-formation (11). After
extensive washing of the deprotected and MeNbz-activated
resin-bound peptide (11), swelling the resin in DMF−
phosphate buffer (0.1 M, pH = 6.8) (1:1, v/v) for 2 h at 50
°C furnished crude purities of 70−80% in LC−MS scale test
reactions. However, more consistent results were achieved with
an increased strength of the phosphate buffer (0.2 M, pH = 6.8)
in MeCN (1:1, v/v), presumably due to the remaining TFA.
Finally, we added 5% of an acid-labile MeDbz resin to the
MeDbz-ChemMatrix resin (11) during compound synthesis to
enable analysis of the efficiency of the SPPS by TFA cleavage.
This still leaves 95% of the material on the resin for the
cyclization−cleavage event. Independently, a preliminary
account of similar cyclization−cleavage of MeNbz-activated
resin-bound peptides was reported by Albericio and co-workers
while this manuscript was in preparation.16 However, this is, to
the best of our knowledge, the first demonstration of this
strategy for preparation of full AIPs.
Using our optimized protocol, we performed a SAR study on
Ss-AIP (5) and compared its inhibitory abilities against agr-I to
one of the most potent pan-group inhibitors of agr in S. aureus
reported, AIP-III D4A (3a).10e The specificity group I of S.
aureus represents one of the most common and was therefore
chosen for preliminary evaluation of the compound series. We
synthesized the control compounds AIPs I−IV (1−4) and AIP-
III D4A (3a) and performed an alanine scan of the S. schleiferi
AIP (12−18) to investigate the role of specific residues on the
inhibition of agr-I. Furthermore, a truncated and N-acetylated
analogue (19) was included to determine the importance of the
exocyclic tail, inspired by the potent global inhibitor, truncated
AIP-I D2A, reported by Muir and co-workers.10b The crude
cyclization products were obtained in purities of 56−87% , and
compounds were isolated by preparative HPLC in yields of 8−
45% based on the resin loading (Table 1).
Figure 1. Quorum sensing signaling in S. aureus and structures of
selected AIPs. (A) Membrane-bound peptidase AgrB converts the AIP
precursor peptide AgrD to the mature AIP and releases it into the
extracellular space. The AIP activates AgrC and thus initiates
phosphorylation of AgrA. AgrA binds to promoters P2 and P3 leading
to up-regulation of RNA-II−III transcription. (B) S. aureus AIPs I−IV
(1−4) and S. schleiferi AIP (5).
With the reference compounds and S. schleiferi AIP analogues
in hand, we determined their ability to inhibit AgrC-I using a
previously reported β-lactamase reporter assay (Table 1).5a As
expected, the cognate AIP-I (1) and agr-I activator AIP-IV (4)
did not affect the β-lactamase activity.10b The determined IC50
values for native agr-I inhibitors AIP-II (2) at 12 nM, AIP-III
(3) at 8 nM, and AIP-III D4A (3a) at 0.16 nM were
comparable to previously reported 1.6, 5.1, and 0.5 nM,
respectively.10e
The Ss-AIP (5) has several interesting structural features,
including a flexible glycine residue in the macrocycle and a
proline residue in the exocyclic tail, which are not present in
AIP-I−IV (1−4). Nevertheless, our alanine scan based on Ss-
AIP (5) recapitulated several requirements for AgrC−AIP
interaction established through previous SAR studies of AIP-I−
IV (1−4)3b,d (Table 1).
and 4-dimethylaminopyridine.12 The following year, Muir and
co-workers conducted the first extensive SAR study on S. aureus
AIPs using an elegant, chemoselective on-resin trans-thioester-
ification.4 Stemming from these general approaches, sophisti-
cation of the synthesis of macrocyclic thiodepsipeptides has
since resulted in methods involving Fmoc-based latent
thioesters,10d metal-mediated thioesterification,10e and solid-
supported carbodiimide reagents.11a Recently, Blackwell and
co-workers applied an N-acyl-benzimidazolinone (Nbz) (7)
approach, using the N3-Fmoc-3,4-diaminobenzoic acid (Dbz)13
linker (6) developed by Dawson and co-workers (Figure 2A).14
We adopted this strategy for the synthesis of Ss-AIP (5);9d
however, in our hands, this protocol provided insufficient
purities when attempting to prepare AIP-I (1) as the control
compound. Cyclization of the crude linear peptide resulted in
mixtures that were difficult to separate due to acylation of the
second aniline of the Dbz linker (6) (Figure 2A and Figure S1).
To avoid these challenging HPLC purifications, we then
applied the second generation linker, which is N-methylated on
the 4-positioned aniline of the linker (MeDbz, 8).15 This
(i) Compounds Ss-AIP Y7A (17) and Ss-AIP F8A (18)
exhibited loss in affinity of 10-fold and 4000-fold, respectively,
showing that hydrophobic residues at the C-terminal are
important for tight binding. (ii) The Ss-AIP analogues 12−15
were only slightly less potent than the natural compound. Thus,
B
DOI: 10.1021/acs.orglett.7b02550
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 2. Preparation of thiolactone-containing peptides with chromatograms of crude AIP-I (1) inserted. (A) Previously reported strategy. (B)
Initial attempt of optimization. (C) On-resin cyclization−cleavage strategy.
that Ss-AIP (5) and G6A (16) inhibited agr-I−III, while no
inhibition was observed for agr-IV. As expected, control
compound 3a exhibited pan-group inhibition (Figure S5).
Based on previous experiments using S. schleiferi supernatant,9d
it was surprising that 5 did not exhibit pan-group inhibition,
which warrants further investigation.
Table 1. Compound Yields and IC50 Values for Inhibition of
AgrC-I
a
b
compound
AIP-I (1)
sequence
yield (%)
IC50 (nM)
YST-[CDFIM]
GVNA-[CSSLF]
IN-[CDFLL]
IN-[CAFLL]
15
8
AIP-II (2)
12 2.9
AIP-III (3)
14
33
21
37
37
21
45
27
25
38
39
8
1.1
The scaled syntheses allowed full characterization of Ss-AIP
(5) by NMR spectroscopy, which revealed two spin systems in
a ratio of 1:0.4 in intensity. Because the chromatography
(HPLC and LC−MS) indicated the presence of a single
compound and all proline-containing analogues in the alanine
scan exhibited the same behavior, we presumed that this arose
from cis−trans isomerism at Pro2. When performing NMR
spectroscopy at varying temperatures up to 60 °C, we did not
AIP-III D4A (3a)
AIP-IV (4)
0.16 0.01
YST-[CYFIM]
YPF-[CIGYF]
APF-[CIGYF]
YAF-[CIGYF]
YPA-[CIGYF]
YPF-[CAGYF]
YPF-[CIAYF]
YPF-[CIGAF]
YPF-[CIGYA]
Ac-[CIGYF]
Ss-AIP (5)
0.31 0.08
1.0 0.2
Ss-AIP Y1A (12)
Ss-AIP P2A (13)
Ss-AIP F3A (14)
Ss-AIP I5A (15)
Ss-AIP G6A (16)
Ss-AIP Y7A (17)
Ss-AIP F8A (18)
trSs-AIP (19)
0.8 0.2
1.2 0.2
1.1 0.2
0.20 0.02
3.1 0.9
1
observe coalescence (Figure S2A). However, H NMR spectra
were recorded in DMSO-d6, CD3CN-d3, and CD3OH-d3,
respectively, and these revealed different ratios of the two spin
systems, strongly suggesting the presence of rotamers (Figure
S2B). The signal overlap of these two rotamers precluded
determination of a high-resolution structure of either
conformation based on NOE and J-coupling constants. To
gather some structural insight related to the SAR, we therefore
compared the changes in chemical shifts of all residues in the
most abundant conformation of Ss-AIP (5) in response to each
alanine mutation (Figures 3 and S3). Not surprisingly, the most
affected residue was Tyr1 when its neighbor Pro2 was mutated
(purple bar), and the same was the case for neighboring
residues in other mutants. However, glycine to alanine
substitution affected all residues in the macrocycle (blue
bars). This strongly indicates that the presence of glycine
provides a unique conformational space compared to previously
investigated AIPs, and it will therefore be interesting to
investigate these conformations further. Finally, removal of the
exocyclic tail only affected the connecting cysteine residue
significantly, indicating that the tail does not have strong
interactions with the macrocycle in DMSO.
1200 240
16.8 0.6
27
a
b
Isolated yield based on the resin loading. Mean values standard
error of the mean based on at least three individual assays performed
in triplicate.
the Ss-AIP I5A (15) mutant did not result in significant
improvement of potency as observed for AIP-I D5A10b and
AIP-III D4A (3a).10e (iii) Removal of the exocyclic tail, to give
truncated Ss-AIP 19, resulted in a 56-fold decrease in inhibition.
(iv) Perhaps most interestingly, equipotency was recorded for
the Ss-AIP G6A (16) mutant, which is the only analogue that
gains structural complexity. Since AIP-I, -III, and -IV (1, 3, and
4) have important hydrophobic residues in this position, it
appears to be an obvious site for further modification.
Furthermore, double mutants will be of interest for extended
SAR studies to search for high potency compounds.
Finally, we tested Ss-AIP (5), its potent G6A mutant (16),
and reported pan-group inhibitor 3a for their ability to inhibit
all specificity groups (agr-I−IV). Using fluorescence reporter
strains and flow cytometry as previously described,9d we found
C
DOI: 10.1021/acs.orglett.7b02550
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Spellberg, B.; Bartlett, J. Clin. Infect. Dis. 2009, 48, 1. (c) David, M. Z.;
Daum, R. S. Clin. Microbiol. Rev. 2010, 23, 616. (d) DeLeo, F. R.; Otto,
M.; Kreiswirth, B. N.; Chambers, H. F. Lancet 2010, 375, 1557.
(2) (a) Rasko, D. A.; Sperandio, V. Nat. Rev. Drug Discovery 2010, 9,
117. (b) Khan, B. A.; Yeh, A. J.; Cheung, G. Y.; Otto, M. Expert Opin.
Invest. Drugs 2015, 24, 689. (c) Dickey, S. W.; Cheung, G. Y. C.; Otto,
M. Nat. Rev. Drug Discovery 2017, 16, 457.
(3) (a) Novick, R. P.; Geisinger, E. Annu. Rev. Genet. 2008, 42, 541.
(b) Thoendel, M.; Kavanaugh, J. S.; Flack, C. E.; Horswill, A. R. Chem.
Rev. 2011, 111, 117. (c) Le, K. Y.; Otto, M. Front. Microbiol. 2015, 6,
1174. (d) Wang, B.; Muir, T. W. Cell Chem. Biol. 2016, 23, 214.
(4) Mayville, P.; Ji, G.; Beavis, R.; Yang, H.; Goger, M.; Novick, R. P.;
Muir, T. W. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 1218.
(5) (a) Nielsen, A.; Mansson, M.; Bojer, M. S.; Gram, L.; Larsen, T.
̊
O.; Novick, R. P.; Frees, D.; Frøkiær, H.; Ingmer, H. PLoS One 2014,
9, e84992. (b) Sully, E. K.; Malachowa, N.; Elmore, B. O.; Alexander,
S. M.; Femling, J. K.; Gray, B. M.; DeLeo, F. R.; Otto, M.; Cheung, A.
L.; Edwards, B. S.; Sklar, L. A.; Horswill, A. R.; Hall, P. R.; Gresham,
H. D. PLoS Pathog. 2014, 10, e1004174. (c) Baldry, M.; Kitir, B.;
Frøkiær, H.; Christensen, S. B.; Taverne, N.; Meijerink, M.; Franzyk,
H.; Olsen, C. A.; Wells, J. M.; Ingmer, H. PLoS One 2016, 11,
e0145618.
(6) Ji, G.; Beavis, R. C.; Novick, R. P. Proc. Natl. Acad. Sci. U. S. A.
1995, 92, 12055.
(7) Novick, R. P.; Ross, H. F.; Projan, S. J.; Kornblum, J.; Kreiswirth,
B.; Moghazeh, S. EMBO J. 1993, 12, 3967.
Figure 3. Chemical shift changes of α-protons observed in DMSO-d6.
In summary, we have developed an efficient method for on-
resin cyclization of thiodepsipeptides with concomitant
cleavage from the solid support and applied this strategy to
enable SAR investigation of an autoinducing peptide from S.
schleiferi. This furnished the highly potent agr-I inhibitor Ss-AIP
G6A (16), highlighting the potential of exploring nonhuman
pathogens for identification of novel QS modulators. Finally,
synthesis of microbe-specific signaling molecules (such as AIPs)
will enable a deeper understanding of the intriguing cross-talk
between pathogens of different species and their effects on
colonization.
(8) Ji, G.; Beavis, R.; Novick, R. P. Science 1997, 276, 2027.
(9) (a) Otto, M.; Echner, H.; Voelter, W.; Gotz, F. Infect. Immun.
̈
2001, 69, 1957. (b) Ji, G.; Pei, W.; Zhang, L.; Qiu, R.; Lin, J.; Benito,
Y.; Lina, G.; Novick, R. P. J. Bacteriol. 2005, 187, 3139. (c) Olson, M.
E.; Todd, D. A.; Schaeffer, C. R.; Paharik, A. E.; Van Dyke, M. J.;
Buttner, H.; Dunman, P. M.; Rohde, H.; Cech, N. B.; Fey, P. D.;
̈
Horswill, A. R. J. Bacteriol. 2014, 196, 3482. (d) Canovas, J.; Baldry,
M.; Bojer, M. S.; Andersen, P. S.; Grzeskowiak, P. K.; Stegger, M.;
Damborg, P.; Olsen, C. A.; Ingmer, H. Front. Microbiol. 2016, 7, 1733;
Correction. Front. Microbiol. 2017, 8, 1075; Correction. Front.
Microbiol. 2017, 8, 1949.
(10) (a) McDowell, P.; Affas, Z.; Reynolds, C.; Holden, M. T. G.;
Wood, S. J.; Saint, S.; Cockayne, A.; Hill, P. J.; Dodd, C. E. R.; Bycroft,
B. W.; Chan, W. C.; Williams, P. Mol. Microbiol. 2001, 41, 503.
(b) Lyon, G. J.; Wright, J. S.; Muir, T. W.; Novick, R. P. Biochemistry
2002, 41, 10095. (c) Scott, R. J.; Lian, L.-Y.; Muharram, S. H.;
Cockayne, A.; Wood, S. J.; Bycroft, B. W.; Williams, P.; Chan, W. C.
Bioorg. Med. Chem. Lett. 2003, 13, 2449. (d) George, E. A.; Novick, R.
P.; Muir, T. W. J. Am. Chem. Soc. 2008, 130, 4914. (e) Tal-Gan, Y.;
Stacy, D. M.; Foegen, M. K.; Koenig, D. W.; Blackwell, H. E. J. Am.
Chem. Soc. 2013, 135, 7869.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b02550.
■
S
Supporting figures, experimental details, compound
characterization data, and copies of HPLC traces and
1H and 13C NMR spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: cao@sund.ku.dk.
ORCID
(11) (a) Gordon, C. P.; Olson, S. D.; Lister, J. L.; Kavanaugh, J. S.;
Horswill, A. R. J. Med. Chem. 2016, 59, 8879. (b) Yang, T.; Tal-Gan,
Y.; Paharik, A. E.; Horswill, A. R.; Blackwell, H. E. ACS Chem. Biol.
2016, 11, 1982.
Christian A. Olsen: 0000-0002-2953-8942
Notes
(12) Otto, M.; Sußmuth, R.; Jung, G.; Gotz, F. FEBS Lett. 1998, 424,
̈
̈
89.
The authors declare no competing financial interest.
(13) Blanco-Canosa, J. B.; Dawson, P. E. Angew. Chem., Int. Ed. 2008,
47, 6851.
(14) (a) Tal-Gan, Y.; Ivancic, M.; Cornilescu, G.; Blackwell, H. E.
Org. Biomol. Chem. 2016, 14, 113. (b) Vasquez, J. K.; Tal-Gan, Y.;
Cornilescu, G.; Tyler, K. A.; Blackwell, H. E. ChemBioChem 2017, 18,
413.
(15) Blanco-Canosa, J. B.; Nardone, B.; Albericio, F.; Dawson, P. E. J.
Am. Chem. Soc. 2015, 137, 7197.
(16) Acosta, G. A.; Royo, M.; de la Torre, B. G.; Albericio, F.
Tetrahedron Lett. 2017, 58, 2788.
ACKNOWLEDGMENTS
■
This work was supported by the Carlsberg Foundation (2013-
01-0333 and CF15-011; to C.A.O.), the University of
Copenhagen (to C.A.O.), a CSC scholarship (to P.P.), and
funding from DNRF (120; to H.I.). The NMR Center−DTU
and the Villum Foundation are acknowledged for access to 600
and 800 MHz spectrometers. We thank Ms. Anne Hector
(DTU) for technical assistance and Martin S. Bojer (University
of Copenhagen) for assistance with fluorescence assays.
REFERENCES
■
(1) (a) WHO Antimicrobial Resistance Global Report on Surveillance
2014; WHO; Switzerland, 2014. (b) Boucher, H. W.; Talbot, G. H.;
Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice, L. B.; Scheld, M.;
D
DOI: 10.1021/acs.orglett.7b02550
Org. Lett. XXXX, XXX, XXX−XXX