Multivalent alteration of quorum sensing in Staphylococcus aureus

Article abstract of DOI:10.1039/c3cc41645c

Virulence in Staphylococcus aureus is strongly and positively correlated with local cell density. Here we present an effective approach to modulate this group behaviour using multivalent peptide-polymer conjugates. Our results show that by attaching multiple AIP-4′ units to macromolecular scaffolds, the agr QS response in S. aureus was affected strongly, while displaying a clear multivalency effect.

Full text of DOI:10.1039/c3cc41645c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Multivalent alteration of quorum sensing in
Staphylococcus aureus†‡
Sarit Melamed Yerushalmi,a Maren E. Buck,b David M. Lynn,bc N. Gabriel Lemcoff*a
and Michael M. Meijler*ad
Cite this: Chem. Commun., 2013,
49, 5177
Received 4th March 2013,
Accepted 16th April 2013
DOI: 10.1039/c3cc41645c
www.rsc.org/chemcomm
Virulence in Staphylococcus aureus is strongly and positively correlated not known how far apart different AgrC proteins are located on the
with local cell density. Here we present an effective approach to cell membrane, even in the case that only one AgrC receptor is bound
modulate this group behaviour using multivalent peptide–polymer by the AIP–polymer, this first event will increase the likelihood that a
conjugates. Our results show that by attaching multiple AIP-40 units to nearby AIP moiety will be bound upon release of the first unit,
macromolecular scaffolds, the agr QS response in S. aureus was thereby effectively raising the affinity of AgrC for the whole polymer
affected strongly, while displaying a clear multivalency effect.
construct. Nature frequently uses multivalency to achieve strong
binding in situations where univalent receptor–ligand interactions
are weak.8 One good example of such a phenomenon, for instance, is
the recognition of carbohydrate ligands by bacterial and mammalian
lectins. Recently, Janda and coworkers described multivalent den-
drimers based on the quorum sensing molecule AI-2.9 Here, we
present the first example of quorum sensing attenuation based on
and strengthened by multivalent polymeric constructs in S. aureus.
Multivalent ligand-induced binding of signaling molecules can
result in antagonist- or agonist-like effects, thereby forming multi-
valent inhibitors or effectors, respectively. Multivalent inhibitors were
shown before to effectively block receptor function. On the other
hand, multivalent effectors were also demonstrated to activate
various cellular processes.10 Moreover, the use of polymeric scaffolds
for synthetic multivalent ligands has been explored previously. For
instance, polymers that were generated by ring opening metathesis
polymerization (ROMP) were used to investigate cell–surface inter-
actions, multivalent binding of histidine-tagged proteins and DNA
binding analysis.11–13 This synthetic method not only allows control
over the degree of polymerization and over the display and density of
ligands,14,15 it is also a straightforward procedure that allows the
assembly of polymer constructs in a single step.16
In order to assess potential multivalent interactions with the AgrC
receptor, we prepared constructs based on non-native AIP-4 (Fig. SI-3,
ESI;‡ we replaced methionine with norleucine, yielding a more stable
isosteric analog (AIP-40) that is identical in activity to the native AIP-4
EC50 33). Next, a modular approach to synthetic copolymers
functionalized with multiple AIP-40 units was designed (Fig. 1). As
the N-terminal end would be the most convenient site of attachment
to the polymer, an N-terminal modified AIP-40 was synthesized to
assess the effect on its activity. For a monomeric norbornene label
only a minor change in the EC50 value was observed (Fig. SI-4, ESI‡).
Consequently, several AIP-40 containing polymers, differing in poly-
mer backbone structure and in the average number of AIP-40 units per
Staphylococcus aureus is a Gram-positive human pathogen that can
cause problems ranging from minor skin infections to life-threatening
health conditions.1 Pathogenesis is caused mainly by the secretion of a
wide array of toxins and tissue-degrading enzymes.2 One of the
primary mechanisms for controlling the secretion of these virulence
factors is a quorum sensing (QS) system called the accessory
gene regulator (agr).3,4 The agr system monitors the extracellular
concentration of unique autoinducing peptides (AIPs) that are
produced and secreted by S. aureus. Each bacterium secretes AIPs
that then accumulate extracellularly. Once these ligands reach a
certain threshold concentration they bind to the transmembrane
domain receptor AgrC, the agr QS system. Taking into account that
coordinated virulence factor synthesis in S. aureus is largely under the
control of the agr QS system, significant efforts have been made to
identify effective inhibitors of AgrC.5,6
As mentioned above, QS responses are only initiated upon reach-
ing a certain threshold concentration of the appropriate receptor
ligand. Therefore, exploiting multivalency effects may be a promising
strategy to achieve a marked increase in ‘‘effective ligand/inhibitor
concentration’’. There are two main types of multivalency in bio-
molecular recognition: the first is dependent on multiple binding
events, and the second relies on high local concentration.7 While it is
a Department of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
E-mail: lemcoff@bgu.ac.il, meijler@bgu.ac.il
b Department of Chemistry, University of Wisconsin-Madison, Wisconsin, USA
c Department of Chemical and Biological Engineering, University of Wisconsin-
Madison, Wisconsin, USA
d The National Institute for Biotechnology in the Negev, Ben-Gurion University of
the Negev, Beer-Sheva, Israel
† We thank Richard P. Novick and the Network on Antimicrobial Resistance in
Staphylococcus aureus (NARSA) for the bacterial strains.
‡ Electronic supplementary information (ESI) available: Experimental procedures
for compounds 1–6, and the biological protocols. See DOI: 10.1039/c3cc41645c
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5177--5179 5177

View Article Online
Communication
ChemComm
Table 1 Activation and inhibition by 3, 5 and free AIP-40
Compound
AIP-40 units per compound
EC50 [nM]
IC50 [nM]
AIP-40
3a
3b
3c
5a
1
1
5
10
4
20
40
33 ꢀ 13
149 ꢀ 26
52 ꢀ 20
35 ꢀ 4
Fig. 1 Polymeric quorum sensing constructs based on AIP-4.
22 ꢀ 13
5b
5c
145 ꢀ 59
89 ꢀ 41
polymer chain were prepared. To control the latter, a ‘‘graft-to’’
approach was adopted, linking the appropriate amount of AIP-40
peptide moieties to N-hydroxysuccinimide (NHS) active sites in the
polymer backbone. Any unreacted active groups on the polymer were
saturated with 3,6,9-trioxadecylamine (MEA) to increase solubility and
minimize undesired subsequent side reactions. Thus, polymer 2 was
synthesized by ROMP of a NHS- activated norbornene carboxylate
monomer (1)17 using a Grubbs 3rd generation catalyst.18 The degree of
polymerization (DP) of polymer 2 was determined by triple-detector
GPC to be around 100. Consequently, three different polymers,
containing 1, 5 or 10% AIP-40 peptide (i.e. 1, 5 and 10 average units
of AIP-40 per polymer) were prepared using the NHS strategy. The
disappearance of free peptide was monitored by LC-MS. Once all
monomeric peptides had reacted, a reaction with excess of MEA was
carried out to afford polymers 3a–c as presented in Scheme 1.
In order to also analyze higher molecular weight polymers,
a second type of polymer (5) was prepared by attaching 1, 5
or 10% AIP-40 peptide to poly(2-vinyl-4,4-dimethylazlactone)
(DP B 400) (4)19,20 by azlactone ring opening, followed by
reaction with excess MEA (Scheme 2).
Due to the higher degree of polymerization and concomitant
monomer content, it was possible to achieve a much higher number
of AIP-40 moieties in polymers 5a–c than in polymers 3a–c. Notably, for
concentrations up to 10 mM none of the studied polymers showed
growth inhibitory effects on S. aureus. In order to determine whether
the polymers could attenuate activation of the agr QS system, a
modified strain of S. aureus class-4 (RN9371), containing a plasmid-
carried P3 promoter fused to a b-lactamase reporter gene,4 was grown
inthe presence of AIP-40 and varying concentrations of polymers 3 or 5
(both polymers without AIP-40 did not show either activation or
EC50/IC50 values for AIP-40, 3 and 5. All results were determined through
b-lactamase assays by incubating S. aureus strain RN9371 with AIP-40
containing compounds. Control experiments with buffer alone or
polymers without AIP-40 showed no activation above background.
inhibition of b-lactamase activity). In order to assess whether the
increase in AIP-40 loading per polymer results in a true multivalency
effect, we plotted b-lactamase activity versus effective AIP-40 molarity –
and not polymer concentration (Fig. SI-5 and SI-6, ESI‡ and Table 1).
Polymer 3a, the short polymer that contains on average 1 unit of AIP-40
peptide, acts as an agonist with an EC50 of 149 ꢀ 26 nM. In comparison
with the activity of free AIP-40 it appears that the polymeric structure
somewhat interferes with the affinity of AIP-40 for its receptor, even
though this effect is not very prominent. Interestingly, as loading of
AIP-40 on the polymer increases, the EC50 values of polymers 3b and 3c
decrease. This suggests that the loss of affinity caused by the polymer
construct was compensated for by a positive effect providing the first
evidence that ligand multivalency can affect QS responses. In the
longer polymers (5a–c), the average number of AIP-40 units per chain is
4, 20 and 40 respectively. Polymer 5a, loaded with 1% AIP-40, activates
the QS system to a similar degree as free AIP-40 (EC50 of 33 ꢀ 13 nM –
reported EC50 values for native AIP-4 are 13–20 nM5,21,22). At higher
loadings, however, we observed a remarkable switch from agonist to
antagonist behavior. This change, also observed at very high concen-
trations (>5 mM) of free AIP-40, could be due to either significantly
higher local AIP concentrations and/or higher affinity of the peptide in
the vicinity of the AgrC receptor, as would be predicted for multivalent
systems. Addition of the AIP-40 rich polymers 5b and 5c to S. aureus
resulted in strong inhibition of its QS system with IC50 values of 145 ꢀ
59 nM and 89 ꢀ 41 nM respectively. While not unprecedented,23
inhibition of QS by high concentrations of autoinducers is poorly
understood. One possible explanation could be that a very high
autoinducer concentration could signal to the bacteria that their
population density is too high and coordinated group behaviour such
as increased virulence factor expression would not add to their benefit
at this point or might even harm the long-term survival of the
population. We currently use the multivalent high-load polymer (5c)
as a tool to further unravel this phenomenon.
To continue the analysis of multivalent binding of the polymers to
the AgrC receptors, fluorescent probes were prepared. Thus, fluores-
cent polymers 6a and 6b, based on polymer 5b conjugated to
rhodamine-6G, were designed and synthesized (Scheme 3). These
fluorescently labeled polymers could be utilized for direct receptor
binding studies. Polymer 6a contains on average 20 AIP-40 units, while
polymer 6b is void of any AIP-40 (and most likely void of any QS activity,
although its fluorescence prevented proper determination of its
activity). In order to quantify binding of polymer to the receptor,
bacteria (OD600 = 0.7) were incubated in the presence of polymers 6a
and 6b, and the fluorescence of both the supernatant and pellet were
Scheme 1 Synthesis of polymers 3a–c.
Scheme 2 Synthesis of polymers 5a–c.
c
5178 Chem. Commun., 2013, 49, 5177--5179
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
wild-type group IV strain, RN8242, was incubated in the presence of
different concentrations of free AIP-40 and polymers 5b and 5c. The
hemolytic activity was measured following the release of heme from
sheep red blood cells (RBCs). We verified that free AIP-40, as expected,
acts as activator at low concentrations (13 nM and beyond, Fig. SI-6,
ESI‡) and as inhibitor at high concentrations (>5 mM). However,
polymers 5b and 5c showed some inhibition at both low and high
concentrations, while polymers devoid of AIP-40 did not. The inhibi-
tory pattern of 5b and 5c was quite unexpected and is currently further
investigated. It should be noted that while the agr system attenuates
hemolytic activity of S. aureus, it is not the sole or even primary
regulator of hemolysis, and as such full inhibition is not expected
upon agr inhibition.
Scheme 3 Synthesis of polymers 6a and 6b.
This is the first description of a multivalent polymeric inhibitor
of QS in S. aureus, using its autoinducing peptide as the active
entitiy on two different types of polymers. This system may be a
launching pad for creating new types of inhibitors by using
agonists or antagonists as the recognition elements. In addition,
it could deepen our understanding of multimeric receptor–ligand
interactions in QS and aid in studying the importance of bacterial
QS receptor topology on bacterial cell surfaces.
This work was supported by the Israel Science Foundation
(Grant 749/09, MMM) and by The Edmond J. Safra Center for
the Design and Engineering of Functional Biopolymers in the
Negev (MMM and NGL).
Fig. 2 Fluorimetric binding assay. S. aureus cells were incubated for 30 min at
37 1C in the presence of 0.5 mM polymer 6b, 0.5 mM polymer 6a, or 0.5 mM polymer
6a + 0.5 mM AIP-40. Cells were then washed with TRIS buffer and fluorescence of
both pellet and supernatant was measured (lexc 490 nm, lem 520 nm), and their
ratios are shown. ** and * denote statistically significant differences between 6b
and 6a (p o 0.02) and between 6a and 6a + AIP-40 (p o 0.05), respectively.
measured. As shown in Fig. 2, polymer 6a shows significantly higher
affinity to the cells compared to polymer 6b. Moreover, in a competi-
tion assay between polymer 6a and free AIP-40 fluorescence levels are
comparable to those with polymer 6b. This suggests that the multi-
valency effect described in Table 1 is very likely due to an increase in
the affinity of the multivalent ligands to the receptor. The competition
assay also indicates that the general affinity of polymer 6a to AgrC is
lower than that of AIP-40 (Fig. 2).
To detect and localize the polymers in live cells, bacteria (O.D.600-
0.5) were incubated for 15 min at 37 1C in the presence of polymers 6a
and 6b. Images were acquired immediately after 2 washes with TRIS
buffer. Polymer 6a (containing AIP-40) was found attached to the
bacteria (Fig. 3b). In contrast, none of the cells was labeled with
polymer 6b (Fig. 3d). Following these results, the inhibitory effects of
AIP-40 containing polymers 5b and 5c on QS activation and their effect
on hemolytic activity of S. aureus was probed.5 Accordingly, S. aureus
Notes and references
1 A. C. Fluit, C. L. Wielders, J. Verhoef and F. J. Schmitz, J. Clin.
Microbiol., 2001, 39, 3727.
2 W. C. Chan, B. J. Coyle and P. Williams, J. Med. Chem., 2004, 47, 4633.
3 G. Ji, R. Beavis and R. P. Novick, Science, 1997, 276, 2027–2030.
4 G. J. Lyon, P. Mayville, T. W. Muir and R. P. Novick, Proc. Natl. Acad.
Sci. U. S. A., 2000, 97, 13330.
5 E. Geisinger, E. A. George, T. W. Muir and R. P. Novick, J. Biol.
Chem., 2008, 283, 8930.
6 B. C. Gorske and H. E. Blackwell, Org. Biomol. Chem., 2006, 4, 1441.
7 D. J. Welsh and D. K. Smith, Org. Biomol. Chem., 2011, 9, 4795.
8 L. L. Kiessling, J. E. Gestwicki and L. E. Strong, Angew. Chem., Int.
Ed., 2006, 45, 2348.
9 A. L. Garner, J. Park, J. S. Zakhari, C. A. Lowery, A. K. Struss, D. Sawada,
G. F. Kaufmann and K. D. Janda, J. Am. Chem. Soc., 2011, 133, 15934.
10 M. J. Ludden, D. N. Reinhoudt and J. Huskens, Chem. Soc. Rev.,
2006, 35, 1122.
11 H. D. Maynard, S. Y. Okada and R. H. Grubbs, J. Am. Chem. Soc.,
2001, 123, 1275.
12 R. Samudrala, X. Zhang, R. M. Wadkins and D. L. Mattern, Bioorg.
Med. Chem., 2007, 15, 186.
13 R. V. Weatherman, K. H. Mortell, M. Chervenak, L. L. Kiessling and
E. J. Toone, Biochemistry, 1996, 35, 3619.
14 C. W. Cairo, J. E. Gestwicki, M. Kanai and L. L. Kiessling, J. Am.
Chem. Soc., 2002, 124, 1615.
15 J. E. Gestwicki, C. W. Cairo, L. E. Strong, K. A. Oetjen and
L. L. Kiessling, J. Am. Chem. Soc., 2002, 124, 14922.
16 S. Iyer, S. Rele, G. Grasa, S. Nolan and E. L. Chaikof, Chem. Commun.,
2003, 151.
17 A. Carrillo, M. J. Yanjarappa, K. V. Gujraty and R. S. Kane, J. Polym.
Sci., Part A: Polym. Chem., 2006, 44, 928.
18 A. Leitgeb, J. Wappel and C. Slugovc, Polymer, 2010, 51, 2927.
19 M. E. Buck and D. M. Lynn, ACS Appl. Mater. Interfaces, 2010, 2, 1421.
20 B. Sun, X. Liu, M. E. Buck and D. M. Lynn, Chem. Commun., 2010, 46, 2016.
21 J. S. Kavanaugh, M. Thoendel and A. R. Horswill, Mol. Microbiol.,
2007, 65, 780.
22 G. J. Lyon, J. S. Wright, T. W. Muir and R. P. Novick, Biochemistry, 2002,
41, 10095.
23 M. M. Meijler, L. G. Hom, G. F. Kaufmann, K. M. McKenzie, C. Z. Sun, J. A.
Moss, M. Matsushita and K. D. Janda, Angew. Chem., Int. Ed., 2004, 43, 2106.
Fig. 3 Imaging of polymers 6a, b in living cells. S. aureus cultures were incubated in the
presence of 1 mM polymer 6a (a, b), or in the presence of 1 mM polymer 6b (c, d). Left
panels, bright field; right panels, rhodamine channel. The scale bar represents 10 mm.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5177--5179 5179