The role of antimicrobial compounds in the life cycle of the symbiotic bacterium, Xenorhabdus nematophila.

摘要

The bacterium Xenorhabdus nematophila maintains a mutualistic relationship with the entomopathogenic nematode Steinernema carpocapsae and is also pathogenic towards insect larvae. X. nematophila possesses a large number of gene clusters potentially involved in antimicrobial production. Several antibiotics, including xenocoumacin (Xcn) produced at high levels in broth cultures, have been characterized. In this study I established that during nematode invasion of the insect body cavity (hemocoel) gut microbiota enter the hemocoel representing potential competitors for X. nematophila. As infection progressed some transient species, such as Staphylococcus saprophyticus disappeared early in infection, while other persistent species such as Enterococcus faecalis proliferated. S. saprophyticus was found to be highly sensitive towards X. nematophila antibiotics and E. faecalis was more resistant. S. saprophyticus was eliminated when co-injected with X. nematophila into the insect host, Manduca sexta. In contrast, E. faecalis proliferated when co-injected with X. nematophila. The induction of transcripts for cecropin, an insect antimicrobial peptide, by E. faecalis was suppressed by the presence of X. nematophila suggesting that E. faecalis proliferation was due in part to a combination of immune suppression and relatively high antibiotic resistance. Injection of E. faecalis into M. sexta caused mortality suggesting that E. faecalis may contribute to, but is not required for, virulence in an insect infected with X. nematophila. The role of antibiotics in interspecies competition was assessed using various antibiotic-deficient strains of X. nematophila co-inoculated in LB broth with either S. saprophyticus or E. faecalis. Antibiotics are produced at high levels in LB broth. During the course of this study I discovered a new non-ribosomal peptide synthetase (NRPS) cluster (cluster F) that produced antibiotic activity. The elimination of S. saprophyticus required Xcn but not compound F. In contrast, elimination of E. faecalis was not dependent on either Xcn or compound F. When competitions were carried out in a more biologically relevant medium (Grace's medium) based on lepidopteran insect hemolymph, both the competitors grew better than X. nematophila due to lower production of antibiotics in Grace's medium and faster growth rate of the competitors. S. saprophyticus was eliminated when inoculated into growing cultures of either the xcn or F strains but grew in the presence of a strain (ngrA) completely devoid of antibiotic activity suggesting that antibiotics other than Xcn and compound F were required to eliminate the competitor. In contrast, E. faecalis was not eliminated in competition with any of the X. nematophila strains consistent with its relatively high antibiotic resistance. S. saprophyticus was eliminated when co-injected into M. sexta with either the xcn or ngrA strain while growth of E. faecalis was facilitated by co-inoculation with both of the mutant strains. Finally, when nematodes carrying the ngrA strain were used for natural infection of M. sexta, nematode reproduction was significantly reduced suggesting that NRPS-derived compounds may function as developmental signals. Together, these findings establish the competitors for X. nematophila and the role of antimicrobials in differential competition and nematode reproduction.