A coffee berry borer (Hypothenemus hampei) genome assembly reveals a reduced chemosensory receptor gene repertoire and male-specific genome sequences

Abstract

Coffee berry borer-CBB (Hypothenemus hampei) is a globally important economic pest of coffee (Coffea spp.). Despite current insect control methods for managing CBB, development of future control strategies requires a better understanding of its biology and interaction with its host plant. Towards this objective, we performed de novo CBB genome and transcriptome sequencing, improved CBB genome assembly and predicted 18,765 protein-encoding genes. Using genome and transcriptome data, we annotated the genes associated with chemosensation and found a reduced gene repertoire composed by 67 odorant receptors (ORs), 62 gustatory receptors (GRs), 33 ionotropic receptors (IRs) and 29 odorant-binding proteins (OBPs). In silico transcript abundance analysis of these chemosensory genes revealed expression enrichment in CBB adults compared with larva. Detection of differentially expressed chemosensory genes between males and females is likely associated with differences in host-finding behavior between sexes. Additionally, we discovered male-specific genome content and identified candidate male-specific expressed genes on these scaffolds, suggesting that a Y-like chromosome may be involved in the CBB's functional haplodiploid mechanism of sex determination.

Figure 1

Number of chemosensory genes in Coleoptera insects. The digits in brackets close to histograms represent the number of odorant receptors (ORs), gustatory receptors (GRs), ionotropic receptors (IRs) and odorant-binding proteins (OBPs), respectively, obtained from genome annotations for Hypothenemus hampei (this study) (Curculionidae), Dendroctonus ponderosae (Curculionidae), Leptinotarsa decemlineata (Chrysomelidae), Anoplophora glabripennis (Cerambycidae), Tribolium castaneum (Tenebrionidae) and Agrilus planipennis (Buprestidae).

Figure 2

Phylogeny of odorant receptor (OR) family. OR protein sequences from Hypothenemus hampei (HhamOR), Dendroctonus ponderosae (DponOR) and Agrilus planipennis (AplaOR) were clustered by Maximum-Likelihood tree-building. Branch supports (aLRT; approximate likelihood ratio test) are shown as colored circles (yellow to red transition). Colored arcs indicate the clusters for OR families 1–3, 5 and 7; and the conserved OR coreceptor (Orco) clade.

Figure 3

Phylogeny of gustatory receptor (GR) family. GR protein sequences from Hypothenemus hampei (HhamGR), Dendroctonus ponderosae (DponGR) and Agrilus planipennis (AplaGR) were clustered by Maximum-Likelihood tree-building. Branch supports (aLRT; approximate likelihood ratio test) are shown as colored circles (yellow to red transition). Thick colored arcs indicate the clusters for the conserved GRs for fructose, sugar and CO₂. The remaining GRs are likely bitter receptors. The recently discovered “GR215 clade” is also indicated by a colored arc. The black arc indicates likely GR gene expansions in H. hampei.

Figure 4

Phylogeny of ionotropic receptors (IR) family. IR protein sequences from Hypothenemus hampei (HhamIR), Dendroctonus ponderosae (DponIR), Agrilus planipennis (AplaIR), Tribolium castaneum (TcIr) and Leptinotarsa decemlineata (LdecIr) were clustered by Maximum-Likelihood tree-building. Branch supports (aLRT; approximate likelihood ratio test) are shown as colored circles (yellow to red transition). Colored arcs indicate the widely conserved lineages of antennal IRs (IR8a, IR25a, IR21a, IR40a, IR41a, IR68a, IR75 and IR76b) and the Divergent IR clade.

Figure 5

Phylogeny of odorant-binding protein (OBP) family. OBP protein sequences from Hypothenemus hampei (HhamOBP), Dendroctonus ponderosae (DponOBP), Leptinotarsa decemlineata (LdecOBP) and Tribolium castaneum (TcasOBP) were clustered by Maximum-Likelihood tree-building. Branch supports (aLRT; approximate likelihood ratio test) are shown as colored circles (yellow to red transition). Colored arcs indicate the conserved OBP classes Minus-C, Plus-C and antennal binding protein II (ABPII). The remaining OBPs are indicated as classic OBPs by an orange arc.

Figure 6

In silico mRNA abundance of chemosensation-related genes. Heat maps represent the abundance of mRNA reads for odorant receptors (ORs), gustatory receptors (GRs), ionotropic receptors (IRs) and odorant-binding proteins (OBPs) in RNA-seq libraries from females (F), males (M) and larvae (L) as estimated by Kallisto. Read abundance is expressed as Log2(TPM + 1). Black asterisks represent chemosensory genes with significant differences at mRNA abundance between females and males as calculated by Kallisto-Sleuth pipeline (False Discovery Rate [FDR] adjusted p-value: * < 0.01; ** < 0.001).

Figure 7

Identification of male-specific genome scaffolds. (A) The Chromosome Quotient (CQ) was plotted as Log10(CQ) across the genome scaffolds [Log10(Scaffold Length)]. Each black dot represents the Log10(CQ) of single scaffolds. Dots below − 0.7 (threshold indicated by a dotted line) were considered as male-specific. (B) PCR-DNA marker analysis for selected candidate male-specific scaffolds using genomic DNA from males (M) and females (F). Cropped images from several electrophoresis gels were combined for B. The full-length electrophoresis gel images are shown in Supplementary Fig. S5. (C) RT-PCR for candidate Y-linked genes using total RNA from males (M) and females (F). Control RT-PCR assays for gDNA contamination were performed using primers for Hh00g129860, with no male-RNA template (-RNA) or lacking retro-transcriptase enzyme (-RT). Cropped images from the same electrophoresis gel were combined for C. The full-length electrophoresis gel image is shown in Supplementary Fig. S6.