Gypsy moth genome provides insights into flight capability and virus-host interactions

Abstract

Since its accidental introduction to Massachusetts in the late 1800s, the European gypsy moth (EGM; Lymantria dispar dispar) has become a major defoliator in North American forests. However, in part because females are flightless, the spread of the EGM across the United States and Canada has been relatively slow over the past 150 years. In contrast, females of the Asian gypsy moth (AGM; Lymantria dispar asiatica) subspecies have fully developed wings and can fly, thereby posing a serious economic threat if populations are established in North America. To explore the genetic determinants of these phenotypic differences, we sequenced and annotated a draft genome of L. dispar and used it to identify genetic variation between EGM and AGM populations. The 865-Mb gypsy moth genome is the largest Lepidoptera genome sequenced to date and encodes ∼13,300 proteins. Gene ontology analyses of EGM and AGM samples revealed divergence between these populations in genes enriched for several gene ontology categories related to muscle adaptation, chemosensory communication, detoxification of food plant foliage, and immunity. These genetic differences likely contribute to variations in flight ability, chemical sensing, and pathogen interactions among EGM and AGM populations. Finally, we use our new genomic and transcriptomic tools to provide insights into genome-wide gene-expression changes of the gypsy moth after viral infection. Characterizing the immunological response of gypsy moths to virus infection may aid in the improvement of virus-based bioinsecticides currently used to control larval populations.

Keywords: Lepidoptera; Lymantria dispar; gypsy moth; virus–host interactions.

Fig. 1.

The Gypsy moth as a model organism. (A) Morphology of L. dispar populations. For each specimen, dorsal (Left) and ventral (Right) sides are shown and their voucher codes are (left to right, top to bottom): NVG-17104G03, -17104G10, -17104H08, -17105A01, -17104H01, -17105A06, -18028H06, and -18028H07. See Dataset S1 for specimen data. (B) Orthology assignment of nine insect and two mammalian genomes. Bars are subdivided to represent different types of ortholog relationships: “1:1:1” indicates universal single-copy genes present in all species; “Diptera” indicates Dipteran-specific genes, present in both D. melanogaster and Aedes aegypti; “Hymenoptera” indicates Hymenopteran-specific genes, present in N. vitripennis, Apis mellifera, and Pogonomyrmex barbatus genomes; “Insect” indicates all other insect-specific orthologs; “Mammal” indicates mammalian-specific orthologs; “N:N:N” indicates other universal genes, but absence in a single genome is tolerated; “Patchy” indicates orthologs that are present in at least one insect and one mammalian genome; “Species” indicates species-specific genes. The phylogeny on the Left is a maximum-likelihood tree of a concatenated alignment of 1,756 single-copy proteins from the 1:1:1 subgroup. The tree was rooted using mammals as the outgroup. CEGMA: these are essential genes and the presence of them in a genome is used to evaluate the quality of an assembly.

Fig. 2.

Comparative analysis of L. dispar and other Lepidoptera species. (A) The phylogeny on the Left is a maximum-likelihood tree of a concatenated alignment of 1,756 single-copy proteins from the 1:1:1 subgroup and was rooted using Plutella xylostella as outgroup. (B) Duplication of myc genes in L. dispar. (C) Duplication of TLR genes in L. dispar. Abbreviation of the species are used and proteins from moths are colored by species. Proteins from gypsy moth are colored red. Abbreviations: atr, Amyelois transitella; aly, Achalarus lyclades; bmo, Bombyx mori; cce, Calycopis cecrops; cne, Calephelis nemesis; dpl, Danaus plexippus; hml, Heliconius melpomene; lac, Lerema accuius; ldi, Lymantria dispar; mci, Melitaea cinxia; mse, Manduca sexta; obr, Operophtera brumata; pgl, Pterourus glaucus; pra, Pieris rapae; pse, Phoebis sennae; pxu, Papilio xuthus; pxy, Plutella xylostella; tni, Trichoplusia ni.

Fig. 3.

Comparison of mitochondrial and nuclear genomes among L. dispar specimens. (A) Phylogeny of mitogenomes of 26 wild-caught specimens from continental Asia, Japan, Iran, Europe, and North America are calculated by maximum likelihood. The LD652 cell line mitogenome is also included in this analysis. Populations are presented by different colors. Iran, blue; Asian, red; European, orange; United States (North America), yellow. (B) PCA of nuclear genomes of the same 26 specimens onto first two PC axes.

Fig. 4.

Divergence between EGM and AGM populations. (A) Distribution of ratio of divergent positions between EGM and AGM populations in 1-kb windows. (B) The divergent positions ratio in different categories. “Adjacent” indicates a 100-bp segment upstream of genes; “Exon” indicates protein-coding regions; “Intergenic” indicates regions between genes while repetitive regions are excluded; “Intron” indicates introns excluding repetitive regions; “Repetitive” indicates repetitive regions. (C) GO-term analysis of proteins with elevated divergence between EGM and AGM populations. Related GO terms are connected by lines. The size of the GO-term circle is proportional to the number of Drosophila proteins associated with this term; the color indicates the level of significance with darker colors indicating a higher degree of significance.

Fig. 5.

An aldehyde oxidase shows elevated divergence between EGM and AGM populations. (A) L. dispar aldehyde oxidase (ldi971.1) modeled on a human aldehyde oxidase template (4UHW) is depicted. Colors indicate secondary structure: helix (cyan), strand (yellow), and loop (green) with FeS centers (orange/yellow spheres), FAD redox cofactor (black stick), molybdenum cofactor (black stick), and substrate (gray stick). Population-specific sites map to the model surface (G/S17, E/K304, C/Y366, L/Q511, C/F516, T/S730, and D/G976; Cα positions are shown with magenta/pink spheres), with two of the surface residues contributing to the dimer interface (D/G976 and T/S730; Cα positions shown with pink spheres). Two sites form the hydrophobic core of their respective subdomains (T/A725 and V/I1056; Cα positions shown with red spheres). (B and C) Zoom-in of two polymorphic residues that map near active sites (blue spheres): F/L244, which lines the FAD-binding site (black stick) (B), and L/M852, which lines the substrate-binding site (gray stick) in the molybdenum cofactor (black stick)-binding domain (C).