Genotype and Gene Expression Associations with Immune Function in Drosophila

Open Access
Publisher Website
Google Scholar
Abstract
It is now well established that natural populations of Drosophila melanogaster harbor substantial genetic variation associated with physiological measures of immune function. In no case, however, have intermediate measures of immune function, such as transcriptional activity of immune-related genes, been tested as mediators of phenotypic variation in immunity. In this study, we measured bacterial load sustained after infection of D. melanogaster with Serratia marcescens, Providencia rettgeri, Enterococcus faecalis, and Lactococcus lactis in a panel of 94 third-chromosome substitution lines. We also measured transcriptional levels of 329 immune-related genes eight hours after infection with E. faecalis and S. marcescens in lines from the phenotypic tails of the test panel. We genotyped the substitution lines at 137 polymorphic markers distributed across 25 genes in order to test for statistical associations among genotype, bacterial load, and transcriptional dynamics. We find that genetic polymorphisms in the pathogen recognition genes (and particularly in PGRP-LC, GNBP1, and GNBP2) are most significantly associated with variation in bacterial load. We also find that overall transcriptional induction of effector proteins is a significant predictor of bacterial load after infection with E. faecalis, and that a marker upstream of the recognition gene PGRP-SD is statistically associated with variation in both bacterial load and transcriptional induction of effector proteins. These results show that polymorphism in genes near the top of the immune system signaling cascade can have a disproportionate effect on organismal phenotype due to the amplification of minor effects through the cascade. Genetic variation for resistance to infection is widespread among insects and other organisms. However, the extent to which this variation in resistance is mediated by changes in infection-induced gene expression is not known. In this study, we assayed expression of immune system genes and bacterial load after infection in a genotyped panel of lines of the model insect Drosophila melanogaster. We find that statistical associations between genetic variants and bacterial load tend to cluster in genes encoding proteins involved in microbial recognition. Variation in suppression of bacterial growth is also determined in part by genetic variation in the expression of downstream components of the immune system that function to directly kill bacteria, despite finding no genetic variation in any single of these effector gene significantly associated with phenotype. Instead, it appears that activity differences in upstream components of the pathway have a cascading effect that results in larger variation in the expression of coordinately regulated downstream effector genes. These results imply that the interactions among genes need to be taken into account when assessing the phenotypic consequences of genetic variation, as signaling cascades such as those in the immune response have the potential to amplify the phenotypic effects of minor genetic variation in individual genes.