Antibiotic-resistant pathogens are profoundly important to human health, but the environmental reservoirs of resistance determinants are poorly understood. The origins of antibiotic resistance in the environment is relevant to human health because of the increasing importance of zoonotic diseases as well as the need for predicting emerging resistant pathogens. This Review explores the presence and spread of antibiotic resistance in non-agricultural, non-clinical environments and demonstrates the need for more intensive investigation on this subject.

Once thought to be a process that occurred only in a few human pathogens, release of biologically active peptidoglycan fragments during growth by Gram-negative bacteria controls many types of bacterial interaction, including symbioses and interactions between microorganisms. This Perspective explores the role of peptidoglycan fragments in mediating a range of microbial-host interactions, and discusses the many systems in which peptidoglycan fragments released during bacterial growth might be active.

Peptidoglycan fragments released by Neisseria gonorrhoeae contribute to the inflammation and ciliated cell death associated with gonorrhea and pelvic inflammatory disease. However, little is known about the production and release of these fragments during bacterial growth. Previous studies demonstrated that one lytic transglycosylase, LtgA, was responsible for the production of approximately half of the released peptidoglycan monomers. Systematic mutational analysis of other putative lytic transglycosylase genes identified lytic transglycosylase D (LtgD) as responsible for release of peptidoglycan monomers from gonococci. An ltgA ltgD double mutant was found not to release peptidoglycan monomers and instead released large, soluble peptidoglycan fragments. In pulse-chase experiments, recycled peptidoglycan was not found in cytoplasmic extracts from the ltgA ltgD mutant as it was for the wild-type strain, indicating that generation of anhydro peptidoglycan monomers by lytic transglycosylases facilitates peptidoglycan recycling. The ltgA ltgD double mutant showed no growth abnormalities or cell separation defects, suggesting that these enzymes are involved in pathogenesis but not necessary for normal growth.Peptidoglycan (PG) fragments released during growth contribute to the pathogenesis of multiple bacterial infections, including those of Bordetella pertussis, Helicobacter pylori, and Neisseria gonorrhoeae (6,23,34). PG fragments induce the production of inflammatory cytokines, cause ciliated cell damage and fluid efflux, and trigger the Nod signaling cascade (reviewed in reference 5). Although PG fragments have been studied biochemically and for immunologic effects in multiple systems, the repertoire of genes and enzymes involved in PG fragment production and release from growing bacteria is unknown.PG fragments released from gram-negative bacterial pathogens are predicted to be produced by the action of lytic transglycosylases. Lytic transglycosylases cleave the N-acetylmuramic acid-␤-1,4-N-acetylglucosamine linkage in PG and catalyze the formation of a 1,6-anhydro bond on the N-acetylmuramic acid (16). PG monomers released from N. gonorrhoeae and B. pertussis were shown to have the 1,6-anhydro bond, indicating that they were generated by lytic transglycosylases (26, 30). To identify genes for PG monomer production, we systematically mutated the genes for lytic transglycosylase homologues in N. gonorrhoeae. Mutation of lytic transglycosylase A (ltgA) resulted in a substantial decrease in PG monomers released (3). Mutations in lytic transglycosylase B (ltgB) or lytic transglycosylase C (ltgC) genes had no effect on PG monomer release (4, 19), although the ltgC mutant showed a severe defect in cell separation. These findings suggested the presence of other lytic transglycosylases in N. gonorrhoeae involved in release of PG monomers.Here we show that lytic transglycosylase LtgD is involved in the release of PG monomers. Additionally, we generated and characterized an N. gonorrhoeae strain deleted for both ltgA and...

Type IV secretion systems require peptidoglycan lytic transglycosylases for efficient secretion, but the function of these enzymes is not clear. The type IV secretion system gene cluster of Neisseria gonorrhoeae encodes two peptidoglycan transglycosylase homologues. One, LtgX, is similar to peptidoglycan transglycosylases from other type IV secretion systems. The other, AtlA, is similar to endolysins from bacteriophages and is not similar to any described type IV secretion component. We characterized the enzymatic function of AtlA in order to examine its role in the type IV secretion system. Purified AtlA was found to degrade macromolecular peptidoglycan and to produce 1,6-anhydro peptidoglycan monomers, characteristic of lytic transglycosylase activity. We found that AtlA can functionally replace the lambda endolysin to lyse Escherichia coli. In contrast, a sensitive measure of lysis demonstrated that AtlA does not lyse gonococci expressing it or gonococci cocultured with an AtlA-expressing strain. The gonococcal type IV secretion system secretes DNA during growth. A deletion of ltgX or a substitution in the putative active site of AtlA severely decreased DNA secretion. These results indicate that AtlA and LtgX are actively involved in type IV secretion and that AtlA is not involved in lysis of gonococci to release DNA. This is the first demonstration that a type IV secretion peptidoglycanase has lytic transglycosylase activity. These data show that AtlA plays a role in type IV secretion of DNA that requires peptidoglycan breakdown without cell lysis.