The role of peptidoglycan in pathogenesis
Introduction
Peptidoglycan (PG) is a highly complex and essential macromolecule of bacterial cell walls (except in mycoplasma) that enables bacteria to resist osmotic pressure. Although it was often regarded as an inert structure surrounding bacteria, the PG layer is a highly dynamic and tightly regulated macromolecule that is constantly remodeled to allow cell growth and division. Despite a rather simple building block (the PG precursor) consisting of a disaccharide N-acetyl-d-glucosamine-(β-1,4)-N-acetylmuramic acid (GlcNAc-MurNAc) pentapeptide, its complexity is the collective result of a variety of different enzymes, including glycosyltransferases, transpeptidases (also known as penicillin-binding proteins), d,d-carboxypeptidases and hydrolases, that shape its three-dimensional structure [1]. Although the peptides can vary extensively in composition from species to species, in general, the glycan backbone is regarded as mostly invariant [2]. The archetypal stem peptide is l-alanine-γ-d-glutamate-diamino acid-d-alanine-d-alanine with the diamino acid being most frequently meso-diaminopimelic acid (mesoDAP) or l-lysine. However, other diamino acids can be found such as l-ornithine and l,l- or d,d-diaminopimelic acid. Furthermore, in Gram-positive bacteria it is very frequent to find cross-bridges linking two distinct stem peptides, increasing even more the complexity of the PG structure.
Interestingly, PG has been shown to be a powerful biological effector with different stimulatory activities such as adjuvanticity, activation of macrophages, cytotoxicity, induction of arthritis, or modulation of slow-wave sleep (Table 1). Most of the elicitor molecules are small PG fragments called muropeptides that are derived from the macromolecule as a result of cell wall turnover. These muropeptides result from the action of d,d-carboxypeptidases and hydrolases. However, until recently little was known about how PG could elicit such a variety of host responses. In 1999, several reports indicated that Toll-like receptor (TLR) 2 could be the sought-after mammalian receptor for PG [3, 4, 5]. However, it became readily apparent that TLR2 was not the receptor for muropeptides [6] and therefore could not account for most of the biological activities of PG degradation products. Furthermore, recent results show that TLR2 does not recognize PG but rather recognizes minor contaminants, such as lipoproteins or lipoteichoic acids, in PG preparations [7].
In 2003, major advances were made with the identification of the cytosolic receptors known as nucleotide-binding oligomerization domain (Nod) 1 and Nod2 as sensors of unique muropeptides [8••, 9••, 10•, 11••, 12••]. Hence, Nod1 recognizes GlcNAc-MurNAc-l-alanine-γ-d-glutamate-mesoDAP (GMTriPDAP) and Nod2 recognizes GlcNAc-MurNAc-l-alanine-d-glutamate (GMDiP). These results have several implications for infectious diseases. Since TLR2 was supposed to recognize the intact PG macromolecule, theoretically TLR2 could only detect Gram-positive PG because the outer membrane shields the Gram-negative PG layer. However, since only the Nods are clearly PG sensors and these recognize degradation products, detection of PG by the host requires prior processing of the PG layer either by bacterial endogenous hydrolases or by the host, in order for the Nod proteins to detect their agonists. Consequently, any mechanism that prevents release of those muropeptides can constitute an escape mechanism from the innate immune system. In light of these recent advances, this review tries to put into a new perspective results from the literature addressing PG and bacterial virulence.
Section snippets
The PG recycling pathway
With the arrival of the genomic era, large-scale studies of virulence have been performed for different pathogens. Surprisingly, a large number of genes implicated in cell metabolism have been identified as necessary for full virulence, including some involved in PG metabolism (see Table 2). In general, these have been considered to affect virulence by their effects on bacterial fitness. However, some of these genes are predicted to be involved in PG maturation, turnover or recycling, and
The Listeria monocytogenes example
In Gram-positive bacteria, recycling of PG might occur only for the stem peptide via the olipeptide permease (Opp) system. Therefore, PG turnover results in shedding of the cell wall into the environment, where it is readily available for detection by the host innate immune system. Despite the diverse biological activities of PG fragments from Gram-positive bacteria, most studies have addressed the structure-activity relationship without considering its role in the bacteria's life-cycle.
d,d-carboxypeptidases and the tracheal cytotoxin
Another interesting example of N. meningitidis attenuated mutants concerns the putative d,d-carboxypeptidase PBP3, which is predicted to transform the PG subunit GlcNAc-MurNAc-l-alanine-γ-d-glutamate-mesoDAP-d-alanine-d-alanine (GMPentaPDAP) once incorporated in the PG layer into GlcNAc-MurNAc-l-alanine-γ-d-glutamate-mesoDAP-d-alanine (GMTetraPDAP). During PG turnover, the major PG fragment released is an anhydrous variant of GMTetraPDAP, G(anh)MTetraPDAP, also known as the tracheal cytotoxin
Modifications of PG and resistance to lysozyme
Besides endogenous hydrolases, the host has its own set of hydrolases capable of degrading insoluble or large polymeric PG fragments into muropeptides and stem peptides; lysozyme is the best-characterized of these hydrolases. Since 1959, it has been known that bacteria have developed ingenious strategies such as O-acetylation of the C-6 atom of MurNAc residues to counteract the hydrolytic activity of lysozyme, human sera and polymorphonuclear enzymes (for a review see [28]). Again, although it
Conclusions
It is becoming apparent that pathogens have developed during evolution a myriad of strategies to evade host surveillance. Structural modifications of lipolysaccharides or secondary polysaccharides and antigenic variation of virulence factors have been acknowledged for some time as mechanisms to escape the innate and adaptive immune systems. However, the same was not applied to PG. The identification of the Nod proteins as intracellular receptors for PG fragments has brought the PG field into
Update
Recent work addressing the role of the recycling pathway in Salmonella enterica serovar Typhimurium, shows that an ampD mutant induces a stronger host response [32]. However, the authors also provide evidence that the attenuated virulence could be a combination of reduced bacterial fitness and stronger host response as ampG or ampG/ampD double mutants are not affected for intercellular survival in macrophages.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Acknowledgements
I would like to thank Catherine Chaput and Anne Derbise for critical reading of the manuscript. Ivo G Boneca is an INSERM (Institut National de la Santé et de la Recherche Médicale) Research Associate. Our research is supported by an Institut Pasteur grant PTR153 and an ACI (Action Concertée Incitative Microbiology) Grant from the Ministère chargé de la Recherche (INSERM N° MIC 0321).
References (57)
- et al.
Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by toll-like receptor 2
J Biol Chem
(1999) - et al.
Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components
Immunity
(1999) - et al.
Host recognition of bacterial muramyl dipeptide mediated through NOD2. Implications for Crohn's disease
J Biol Chem
(2003) Why does Escherichia coli recycle its cell wall peptides?
Mol Microbiol
(1995)- et al.
Arthropathic properties related to the molecular-weight of peptidoglycan-polysaccharide polymers of steptococcal cell walls
Infect Immun
(1982) - et al.
The pgdA gene encodes for a peptidoglycan N-acetylglucosamine deacetylase in Streptococcus pneumoniae
J Biol Chem
(2000) - et al.
Lipophilic derivative of muramyl dipeptide is more active than muramyl dipeptide in priming macrophages to release superoxide anion
Infect Immun
(1980) - et al.
G(Anh)MTetra, a natural bacterial cell wall breakdown product, induces interleukin-1 beta and interleukin-6 expression in human monocytes. A study of the molecular mechanisms involved in inflammatory cytokine expression
J Biol Chem
(1994) - et al.
Activation of the alternate complement pathway by peptidoglycan from streptococcal cell wall
Infect Immun
(1978) - et al.
Arthropathic properties of gonococcal peptidoglycan fragments: implications for the pathogenesis of disseminated gonococcal disease
Infect Immun
(1986)