A whole transcriptomal linkage analysis of gene co-regulation in insecticide resistant house flies, Musca domestica

Abstract

Background: Studies suggest that not only is insecticide resistance conferred via multiple gene up-regulation, but it is mediated through the interaction of regulatory factors. However, no regulatory factors in insecticide resistance have yet been identified, and there has been no examination of the regulatory interaction of resistance genes. Our current study generated the first reference transcriptome from the adult house fly and conducted a whole transcriptome analysis for the multiple insecticide resistant strain ALHF (wild-type) and two insecticide susceptible strains: aabys (with morphological recessive markers) and CS (wild type) to gain valuable insights into the gene interaction and complex regulation in insecticide resistance of house flies, Musca domestica.

Results: Over 56 million reads were used to assemble the adult female M. domestica transcriptome reference and 14488 contigs were generated from the de novo transcriptome assembly. A total of 6159 (43%) of the contigs contained coding regions, among which 1316 genes were identified as being co-up-regulated in ALHF in comparison to both aabys and CS. The majority of these up-regulated genes fell within the SCOP categories of metabolism, general, intra-cellular processes, and regulation, and covered three key detailed function categories: redox detailed function category in metabolism, signal transduction and kinases/phosphatases in regulation, and proteases in intra-cellular processes. The redox group contained detoxification gene superfamilies, including cytochrome P450s, glutathione S-transferases, and esterases. The signal transduction and kinases/phosphatases groups contained gene families of rhodopsin-like GPCRs, adenylate and guanylate cyclases, protein kinases and phosphatases. The proteases group contained genes with digestive, catalytic, and proteinase activities. Genetic linkage analysis with house fly lines comparing different autosomal combinations from ALHF revealed that the up-regulation of gene expression in the three key SCOP detailed function categories occurred mainly through the co-regulation of factors among multiple autosomes, especially between autosomes 2 and 5, suggesting that signaling transduction cascades controlled by GPCRs, protein kinase/phosphates and proteases may be involved in the regulation of resistance P450 gene regulation.

Conclusion: Taken together, our findings suggested that not only is insecticide resistance conferred via multi-resistance mechanisms or up-regulated genes, but it is mediated through the trans and/or cis co-regulations of resistance genes.

Figure 1

Schematic for the generation of the M. domestica combination strains used in our study. Strain ALHF is a highly insecticide-resistant strain while the aabys strain is an insecticide-susceptible strain that possesses five recessive morphological markers, with each morphological marker being uniquely present on one autosome. The images along the bottom row show the recessive morphological markers unique to each of the combination strains.

Figure 2

Data analysis pipeline for the generation of the Musca domestica predicted gene set and differential gene expression testing. Hexagons represent the raw data collected for this study, while terms within boxes represent either the programs, or the filtering steps used in the data analysis. The directions of the arrows indicate the data processing flow.

Figure 3

Nucleotide length distributions for the Musca domestica ALHF strain raw assembled contigs and predicted coding regions (CDS). Coding region lengths were predicted using Augustus (version 2.5.5) under the “fly” model and include both complete (5469 sequences) open reading frames (ORFs) and partial ORFs (690 sequences). A partial ORF means any sequence that is predicted to be missing either the start or the stop codon, but not both.

Figure 4

Venn diagram for the annotation obtained for the Musca domestica ALHF strain predicted gene set. Overlapping ellipses represent predicted genes from the Pfam-A (v26.0), the Drosophila melanogaster proteome (v. r5.46), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) automatic annotation server (KAAS) that could be annotated in two or more of the databases used to predict gene function. An e-value threshold for homology detection was fixed at 10^-20 for the Pfam and blastx analyses and at 10^-5 for KEGG. The circles excluded from the overlapping ellipses represent sequences which contained the coding region, but had no homologs in any of the three databases used for gene prediction.

Figure 5

Box and whisker plots representing the interquartile ranges (IQR) for the nucleotide coverage of each of the ALHF strain predicted genes in each of the Musca domestica strains tested. The dependent axis has been broken to make the IQR and median values discernible. The solid line within each of the boxes represents the median value for the gene coverage for each house fly strain.

Figure 6

Correlation of the gene expression levels (FPKM) for all of the Musca domestica strains tested versus the ALHF pyrethroid-resistant strain (upper panels). Scatterplots represent the differential gene expression compared to the ALHF strain (lower panels). In the upper panels, the points closest to the 1:1 line represent genes with the same gene expression value as the ALHF strain and the tested M. domestica strain. In the lower panels, each point represents a gene, with red points below the central axis indicating the genes that were down-regulated in the tested M. domestica strain compared to the ALHF strain, thus the red points below the horizontal axis on the lower panels represent the genes that were up-regulated in the pyrethroid-resistant ALHF strain and putatively linked to insecticide resistance.

Figure 7

Heat map of the gene expression values (within gene) relative to aabys for each of the genes tested by qPCR to validate the gene expression levels within the different Musca domestica lines and the parental ALHF and aabys strains. Colors scaled from yellow to red indicate low to higher gene expression, respectively, relative to aabys.

Figure 8

Linkage of genes up-regulated in the ALHF strain of Musca domestica. The overlapping areas between the ellipses indicate the autosomal interaction for those genes that were up-regulated in the ALHF strain for two or more of autosomes.

Figure 9

Allele-specific RT-PCR autosomal mapping of the Musca domestica genes. PCR fragments were generated using the allele-specific primer set according to the sequence of each gene from ALHF. The absence of a PCR product in a house fly line indicated that the gene was located on the corresponding autosome from aabys (i.e. the absence of a band in the A1234 line indicates that the gene was present on autosome 5).