Genomic landscape and genetic manipulation of the black soldier fly Hermetia illucens, a natural waste recycler

Abstract

The black soldier fly (BSF), Hermetia illucens (Diptera: Stratiomyidae), is renowned for its bioconversion of organic waste into a sustainable source of animal feed. We report a high-quality genome of 1.1 Gb and a consensus set of 16,770 gene models for this beneficial species. Compared to those of other dipteran species, the BSF genome has undergone a substantial expansion in functional modules related to septic adaptation, including immune system factors, olfactory receptors, and cytochrome P450s. We further profiled midgut transcriptomes and associated microbiomes of BSF larvae fed with representative types of organic waste. We find that the pathways related to digestive system and fighting infection are commonly enriched and that Firmicutes bacteria dominate the microbial community in BSF across all diets. To extend its potential practical applications, we further developed an efficient CRISPR/Cas9-based gene editing approach and implemented this to yield flightless and enhanced feeding capacity phenotypes, both of which could expand BSF production capabilities. Our study provides valuable genomic and technical resources for optimizing BSF lines for industrialization.

Fig. 1

Biology and genomic features of the black soldier fly. N50, the minimum length of scaffolds that cover 50% of the genome assembly; AMP antimicrobial peptide, OR olfactory receptor, P450 cytochrome P450s

Fig. 2

Relationships of Diptera species and pathways that differ between BSF and parallel lineages. a Phylogenetic relationship and assigned orthology across ten dipteran species. The maximum likelihood phylogenomic tree was calculated based on 814 single-copy universal genes. Nodes labeled with a brown circle indicate those with high bootstrap support (at least 82 of 100 replicates). The colored histogram indicates category of orthology as follows: “1:1:1”, single-copy universal genes; “N:N:N”, multi-copy universal genes; “Brachyera only”, genes specific to Brachyera species; “Nematocera only”, genes specific to Nematocera species; “Species-specific”, genes without an ortholog in any other species; “Specific duplication”, genes with species-specific duplication; “Patchy”, orthologs in some species; “Homology”, homologs detected in other species (E < 1e−5) but not grouped in the orthology analysis. The species used in the analysis were: Aedes aegypti, the yellow fever mosquito; Anopheles gambiae, the African malaria mosquito; Belgica antarctica, the Antarctic midge; Mayetiola destructor, the hessian fly; H. illucens, BSF; Zeugodacus cucurbitae, the melon fly; Drosophila melanogaster, the fruit fly; Glossina morsitans, the tsetse fly; Musca domestica, the house fly; and Lucilia cuprina, the blow fly. b Identification of pathways that have rapidly evolved in BSF. dN/dS ratios were calculated independently in two parallel evolutionary lineages, M. domestica and D. melanogaster, using BSF as the common ancestor. Each dot indicates the median dN/dS ratios of all related genes in the corresponding pathway. Significantly enriched (FDR-adjusted P < 0.05), rapidly evolving genes in KEGG pathways are highlighted in red

Fig. 3

Expansions in gene families related to BSF environmental adaptation. a Number of gene copies in the indicated families related to environmental adaptation in dipteran species. The area size of each pie indicates the relative gene number in each family. b–e Phylogenetic relationships across three dipteran species for gene families with prominent expansions in BSF: gram-negative binding proteins (b), cecropin antimicrobial peptides (c), Olfactoery receptors (d), cytochrome P450s (e). Phylogenetic trees were estimated using the maximum likelihood method

Fig. 4

Intestinal transcriptome in BSF larvae fed with organic waste. Midguts of BSF larvae fed with food waste (FW), poultry manure (PM), dairy manure (DM), or swine manure (SM) were sampled on days 4, 6, 8, and 12 of feeding with the indicated diet. The samples were subjected to RNA-seq. a Distributions of expressed genes (n = 9417) across 16 samples: Genes expressed at each time point under each type of diet are labeled “All”; those expressed in 15 out of 16 samples are labeled “Almost all”; genes commonly expressed under each diet but not at every time point are labeled “Broad”; genes only expressed in one sample are labeled “Orphan”; genes only expressed by larvae fed with manure are labeled “Manure”; and genes only expressed in larvae fed with food waste are labeled “Waste”. b Principal component analysis of intestinal samples based on their overall expression profiles. The first two eigenvectors that explained 34.2% and 20.4% of the variance are plotted. c Venn diagram of the 500 most highly expressed genes (~5% of all expressed genes), selected for each type of diet based on the average expression values across all time points. A total of 326 genes were expressed by larvae fed all four diets. d The 326 genes expressed by larvae fed all four diets were subjected to KEGG enrichment analysis. Pathways in blue belong to digestive systems, and pathways in red indicate those related to infectious diseases. Gene counts are presented as histograms. Hypergeometric test (FDR-adjusted): *P < 0.05, ***P < 0.005, ****P < 0.001. e A representative gene cluster specific to BSF and highly expressed in larvae fed with organic waste. Genomic organization in BSF and the homologous region in D. melanogaster are shown. Homolog pairs between these species are linked by lines. Genes in green and blue indicate BSF-specific genes that belong to two ortholog groups. These 14 genes do not have homology to genes of any other sequenced invertebrate species. Note that this cluster is located in the end of an assembled BSF scaffold. The heatmap shows the expression pattern of corresponding genes in BSF larvae fed with the other diets at each of the four time points

Fig. 5

Microbiome of BSF larvae fed with different types of organic waste. a Within-sample diversity estimates of the bacterial communities in larvae fed with the indicated diets. b Constrained principal coordinate analysis of between-sample diversity. Bray-Curtis distances between samples constrained by diets plotted for the first two CPCoAs. c The dynamic landscape of OTUs across all communities at a phylum level. OTU richness is indicated by the area of corresponding symbols. Symbols indicate counts of contained sequences. Colors indicate the fraction of target OTUs relative to all OTUs of the corresponding sample

Fig. 6

Mutagenesis of Ptth leads to increased feeding capacity in BSF larvae. The CRISPR/Cas9 system was used to induce mutations at the HiPtth locus in H. illucens. a Schematic representation of the exon/intron boundaries of the HiPtth gene. Exons are shown as boxes; thin lines represent introns; numbers are fragment lengths in base pairs (bp). Target site (TS) locations are noted and PAM sequences are shown in red. b Sequences of the targeted region in the HiPtth locus in the mutants. The PAM sequence is in red. The numbers of nucleotides deleted in each line are indicated on the right. c Morphology of HiPtth mutants showing their greater size relative to wild type (WT) controls. d Average body weights of mutants and control (n = 30; mean values ± SEM)

Fig. 7

Mutagenesis of Vg in BSF eliminates wings in adults. a Schematic representation of the exon/intron boundaries of HiVg. Exons are shown as boxes and thin lines represent the introns. Target site (TS) locations are noted and PAM sequences are shown in red. b Sequences of the targeted region in the corresponding loci of Vg mutants. The PAM sequence is in red. The numbers of nucleotides deleted in each line are indicated on the right. c Phenotypic images show that Vg mutants lack wings in the adult stage