A high-quality de novo genome assembly based on nanopore sequencing of a wild-caught coconut rhinoceros beetle (Oryctes rhinoceros)

Background

An optimal starting point for relating genome function to organismal biology is a high-quality nuclear genome assembly, and long-read sequencing is revolutionizing the production of this genomic resource in insects. Despite this, nuclear genome assemblies have been under-represented for agricultural insect pests, particularly from the order Coleoptera. Here we present a de novo genome assembly and structural annotation for the coconut rhinoceros beetle, Oryctes rhinoceros (Coleoptera: Scarabaeidae), based on Oxford Nanopore Technologies (ONT) long-read data generated from a wild-caught female, as well as the assembly process that also led to the recovery of the complete circular genome assemblies of the beetle’s mitochondrial genome and that of the biocontrol agent, Oryctes rhinoceros nudivirus (OrNV). As an invasive pest of palm trees, O. rhinoceros is undergoing an expansion in its range across the Pacific Islands, requiring new approaches to management that may include strategies facilitated by genome assembly and annotation.

Results

High-quality DNA isolated from an adult female was used to create four ONT libraries that were sequenced using four MinION flow cells, producing a total of 27.2 Gb of high-quality long-read sequences. We employed an iterative assembly process and polishing with one lane of high-accuracy Illumina reads, obtaining a final size of the assembly of 377.36 Mb that had high contiguity (fragment N50 length = 12 Mb) and accuracy, as evidenced by the exceptionally high completeness of the benchmarked set of conserved single-copy orthologous genes (BUSCO completeness = 99.1%). These quality metrics place our assembly ahead of the published Coleopteran genomes, including that of an insect model, the red flour beetle (Tribolium castaneum). The structural annotation of the nuclear genome assembly contained a highly-accurate set of 16,371 protein-coding genes, with only 2.8% missing BUSCOs, and the expected number of non-coding RNAs. The number and structure of paralogous genes in a gene family like Sigma GST is lower than in another scarab beetle (Onthophagus taurus), but higher than in the red flour beetle (Tribolium castaneum), which suggests expansion of this GST class in Scarabaeidae. The quality of our gene models was also confirmed with the correct placement of O. rhinoceros among other members of the rhinoceros beetles (subfamily Dynastinae) in a phylogeny based on the sequences of 95 protein-coding genes in 373 beetle species from all major lineages of Coleoptera. Finally, we provide a list of 30 candidate dsRNA targets whose orthologs have been experimentally validated as highly effective targets for RNAi-based control of several beetles.

Conclusions

The genomic resources produced in this study form a foundation for further functional genetic research and management programs that may inform the control and surveillance of O. rhinoceros populations, and we demonstrate the efficacy of de novo genome assembly using long-read ONT data from a single field-caught insect.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-022-08628-z.

Diverse Host Immune Responses of Different Geographical Populations of the Coconut Rhinoceros Beetle to Oryctes Rhinoceros Nudivirus (OrNV) Infection

Incursions of the coconut rhinoceros beetle (CRB), Oryctes rhinoceros, into different islands in the South Pacific have been detected in recent years. It has been suggested that this range expansion is related to an O. rhinoceros haplotype reported to show reduced susceptibility to the well-established classical biocontrol agent, Oryctes rhinoceros nudivirus (OrNV). Our understanding of the genetic characteristics which distinguish the population of O. rhinoceros that has recently established in Solomon Islands from other well-established populations across the region is very limited. Here, we hypothesized that the recently established O. rhinoceros population should have greater innate immune responses when challenged by OrNV than those of well-established and native O. rhinoceros populations. We used the RNA sequencing (RNA-Seq) approach to generate gene expression profiles of midgut tissue from OrNV-infected and noninfected individuals collected in the Solomon Islands (recent incursion), Papua New Guinea and Fiji (previously established), and the Philippines (within the native range). The collections included individuals from each of the three major mitochondrial lineages (CRB-G, CRB-PNG, and CRB-S) known to the region, allowing us to explore the specific responses of each haplotype to infection. Although insects from the Philippines and Solomon Islands that were tested belong to the same mitochondrial lineage (CRB-G), their overall responses to infection were different. The number of differentially expressed genes between OrNV-infected and noninfected wild-caught individuals from the four different locations varied from 148 to 252. Persistent OrNV infection caused a high level of induced antimicrobial activity and immune responses in O. rhinoceros, but the direction and magnitude of the responses were population specific. The insects tested from the Solomon Islands displayed extremely high expression of genes which are known to be involved in immune responses (e.g. coleoptericin, cecropin, and serpin). These variations in the host immune system among insects from different geographical regions might be driven by variations in the virulence of OrNV isolates, and this requires further investigation. Overall, our current findings support the importance of immunity in insect pest incursion and an expansion of the pest's geographic range. IMPORTANCE Oryctes rhinoceros nudivirus (OrNV) is a double-stranded DNA (dsDNA) virus which has been used as a biocontrol agent to suppress coconut rhinoceros beetle (CRB) in the Pacific Islands. Recently a new wave of CRB incursions in Oceania is thought to be related to the presence of low-virulence isolates of OrNV or virus-tolerant haplotypes of beetles (CRB-G). Our comparative analysis of OrNV-infected and noninfected CRBs revealed that specific sets of genes were induced by viral infection in the beetles. This induction was much stronger in beetles collected from the Solomon Islands, a newly invaded country, than in individuals collected from within the beetle's native range (the Philippines) or from longer-established populations in its exotic range (Fiji and Papua New Guinea [PNG]). Beetles from the Philippines and the Solomon Islands that were tested in this study all belonged to the CRB-G haplotype, but the country-specific responses of the beetles to OrNV infection were different.