Prior infection of Manduca sexta with non-pathogenic Escherichia coli elicits immunity to pathogenic Photorhabdus luminescens: Roles of immune-related proteins shown by RNA interference

Abstract

Prior infection of Manduca sexta caterpillars with the non-pathogenic bacterium Escherichia coli elicits effective immunity against subsequent infection by the usually lethal and highly virulent insect pathogen Photorhabdus luminescens TT01. Induction of this protective effect is associated with up-regulation of both microbial pattern recognition protein genes ( hemolin, immulectin-2 and peptidoglycan recognition protein) and anti-bacterial effector genes ( attacin, cecropin, lebocin, lysozyme and moricin). We used RNA interference to knock down over-transcription of members of both these sets of genes one at a time. Interfering with expression of individual recognition proteins had a drastic adverse effect on the E. coli elicited immunity. RNAi knock-down of immulectin-2 caused the greatest reduction in immunity, followed by hemolin and peptidoglycan recognition protein (PGRP) in that order, to the extent that knock-down of any one of these three proteins left the insects more susceptible to P. luminescens infection than insects that had not experienced prior infection with E. coli. Interfering with the expression of individual antibacterial effector proteins and peptides had a much less marked effect on immunity. Knock-down of attacin, cecropin or moricin caused treated insects to be more susceptible to P. luminescens infection than controls that had been pre-infected with E. coli but which had not received the specific RNAi reagents, but they were still less susceptible than insects that had not been pre-infected with E. coli. RNAi knock-down with expression of lebocin or lysozyme had no effect on E. coli-induced immunity to P. luminescens, indicating that these effectors are not involved in the response. By bleeding pre-infected caterpillars and growing the pathogen directly within cell-free insect haemolymph, we showed that at least part of the protection elicited by previous exposure to E. coli is due to the presence of factors within the blood plasma that inhibit the growth of P. luminescens. The production of these factors is inhibited by RNAi treatment with ds-RNA reagents that knock down hemolin, immulectin-2, and PGRP. These results demonstrate that the insect immune system can be effectively primed by prior infection with non-pathogenic bacteria against subsequent infection by a highly virulent pathogen. Given the continuous normal exposure of insects to environmental and symbiotic bacteria, we suggest that prior infection is likely to play a significant and underestimated role in determining the level of insect immunity found in nature.