HGT is widespread in insects and contributes to male courtship in lepidopterans

Abstract

Horizontal gene transfer (HGT) is an important evolutionary force shaping prokaryotic and eukaryotic genomes. HGT-acquired genes have been sporadically reported in insects, a lineage containing >50% of animals. We systematically examined HGT in 218 high-quality genomes of diverse insects and found that they acquired 1,410 genes exhibiting diverse functions, including many not previously reported, via 741 distinct transfers from non-metazoan donors. Lepidopterans had the highest average number of HGT-acquired genes. HGT-acquired genes containing introns exhibited substantially higher expression levels than genes lacking introns, suggesting that intron gains were likely involved in HGT adaptation. Lastly, we used the CRISPR-Cas9 system to edit the prevalent unreported gene LOC105383139, which was transferred into the last common ancestor of moths and butterflies. In diamondback moths, males lacking LOC105383139 courted females significantly less. We conclude that HGT has been a major contributor to insect adaptation.

Keywords: HGT; adaptive evolution; biodiversity; comparative genomics; intron gain; male courtship behavior; symbionts.

Figure 1.. Distribution of the 1,410 putative HGT-acquired genes on the maximum likelihood phylogeny of 218 insects.

We sampled 218 insects representing 11 / 19 species-rich orders (i.e., orders with >1,000 described species) (Stork, 2018). The phylogeny was a concatenated ML tree inferred from analysis of 1,367 single-copy BUSCO genes. These 1,410 putative HGT-acquired genes were likely acquired through 741 distinct HGT events, of which 588 were species-specific and the other 153 were present in two or more species. The stacked bars indicate the number of HGT-acquired genes from the different HGT donor resources (red: Bacteria, orange: Fungi, green: Plants, blue: Viruses, gray: Others). Images representing taxa were taken from PhyloPic (http://phylopic.org). See also Figures S1 and S2 and Tables S1 and S2.

Figure 2.. Robustness of HGT inference.

A) Validation of HGT-acquired genes using PCR and Sanger sequencing experiments. Since it is challenging to validate all 1,410 HGT-acquired genes in 218 insects due to limitation of insect genomic DNA, we examined 54 HGT-acquired genes in 16 insect species representing 8 of 11 orders. Note that the uneven sampling of insects for this analysis might not fully reflect the accuracy of HGT inference across orders. For each of 54 HGT-acquired genes, two separate PCR reactions followed by Sanger sequencing of the amplicons were used to validate the presence of the HGT-acquired gene in the insect genome (see details in Methods). B) Distributions of sequence lengths of genomic contigs with and without HGT-acquired genes. The darker distribution is that of the sequence lengths of the 928 contigs that contain the 1,410 HGT-acquired genes, and the lighter distribution is that of the sequence lengths of the 89,481 contigs that do not contain HGT-acquired genes. C) Distribution of proportions of HGT-acquired genes in each of the 928 contigs that harbor the 1,410 inferred HGT-acquired genes. D) Distribution of the protein sequence similarity between the sequence in the insect recipient genome and its closest hit in a non-metazoan donor genome for all 1,410 HGT-acquired genes.

Figure 3.. Symbionts of host insects were likely to be involved in the transitions of foreign genes into insect genomes.

These 1,410 foreign genes were likely horizontally acquired through 741 distinct HGT events from 670 putative HGT donor species. A) The distribution of 670 putative donor species. B) The association in abundance between HGT donor species and known insect symbionts. We calculated the relative abundance of each HGT donor genus as well as the relative abundance of each known symbiont genus in 20 insects (from 7 / 11 orders) in SymGenDB. Pearson’s correlation coefficient was used to test whether these two variables are significantly correlated. C) Gene ontology (GO) term analysis of 1,410 HGT-acquired genes in our study and of 193 HGT-acquired genes reported in previous studies in terms of biological processes.

Figure 4.. Repeat-rich intron gains from native insect genomes were likely involved in the adaptation of HGTs in insects.

A) After the integration of the 1,410 HGT-acquired genes into insect genomes, 849 gained 1,534 introns ≥ 100 bp in length (orange), 42 lost 53 introns (blue), and 519 had no intron gain or loss (gray). B) The origins of 1,534 gained introns. 1,013 introns are highly similar to sequences present in native insect genomes (green), while the sequences of the remaining 521 introns do not show similarity to the native insect genomes and were likely acquired from other organisms (gray). C) The transposable element (TE) compositions of 1,013 introns gained from native insect genomes. D) Comparisons of characteristics between transferred genes in HGT donor species (orange), foreign genes in recipient species (red), and native genes in recipient species (green). The left three boxes correspond to gene length, CDS length, and intron length, respectively. The right box corresponds to number of introns per gene. E) Characterizing changes in gene structures (gene length, CDS length, intron length, and number of introns) for the HGT-acquired genes in the context of relative divergence times. For these analyses, we examined the 822 / 1,410 HGT-acquired genes that were inferred to have been acquired in the common ancestor of two or more of the 218 species included in our study. The relative divergence times were inferred by the RelTime in MEGA7 (Kumar et al., 2016) using the ML tree in Figure 1. F) Comparison of average expression level between HGT-acquired genes containing introns (I: first row) and HGT-acquired genes lacking introns (II: second row) for each of 32 insects representing 6 of 11 orders in our study. Note that we compared HGT-acquired genes with and without introns only within each transcriptomic dataset (e.g., only using transcriptome data from the same stage and the same tissue for a given species). We used 90 publicly available transcriptome datasets to calculate the expression levels of HGT-acquired genes containing introns and HGT-acquired genes lacking introns within each transcriptomic dataset. The information of developmental stage and tissue for the transcriptome data for each species is given in Table S3. The phylogeny of 32 insects was taken from the full phylogeny of 218 insects in Figure 1. For a given species, a red star indicates that the average expression level of HGT-acquired genes containing introns is higher than that of HGT-acquired genes lacking introns, while a white star indicates that the average expression levels of HGT-acquired genes containing introns is lower than that of HGT-acquired genes lacking introns. See also Table S3.

Figure 5.. The prevalent HGT-acquired gene LOC105383139 enhances male courtship behavior in lepidopterans.

The prevalent HGT-acquired gene LOC105383139, which belongs to the large protein family of zinc-binding alcohol dehydrogenases, is present in nearly all moths and butterflies in our study, except for the moth Chilo suppressalis and the butterfly Leptidea sinapis. A) A simplified gene family phylogeny of the gene LOC105383139. Red branches indicate moths and butterflies, while green branches indicate Bacteria. B) A simplified schematic diagram of the generation of the homozygous mutant line (MT-139) using the CRISPR-case9 systems with two sgRNAs to edit single-copy gene LOC105383139 in Plutella xylostella. Three representative mutant lines are given in the below box. C) Comparison of numbers of eggs produced by wild-type males + wild-type females (MT, n=15 pairs), knockout males + knockout females (WT-139, n=15 pairs), wild-type virgin females (WT-virgin, n=15 females), and knockout virgin (WT-139-virgin, n=15 females) in 48 hours. D) Characterizing four developmental stages of P. xylostella from egg to hatching within 70 hours. Red arrows in the upper box indicate changes through four developmental stages. Stacked bars in the below box indicate the proportions of each of four stages for four different treatments. Note that all eggs produced by 4 / 26 (15.4%) pairs of wild-type moths (WT) and by 20 / 30 (66.7%) pairs of knockout moths (MT-139) were completely stuck in the stage I (no gastrulation). E) Percentage of successfully courted pairs of adult females and adult males during 48 hours. Courtship index is the percentage of successfully courted pairs, in which the male moves toward the female with flapping wings and tipping the abdomen, in a given time period. F) Percentage of successfully mated pairs of adult females and adult males during 48 hours. Mating index is the percentage of successfully mated pairs in which the male copulates with the female for approximately one hour. We used four treatments to conduct behavioral assays: wild-type males (WT♂) + wild-type females (WT♀), wild-type males (WT♂) + knockout females (MT-139♀), knockout males (MT-139 ♂) + knockout females (MT-139♀), and knockout males (MT-139♂) + wild-type females (WT♀). Each treatment had three replicates using 24 pairs of 1-day-old male and female adult moths. G) Comparison of gene expression of the gene LOC105383139 in 15 pairs (males ♂ and females♀) of adult lepidopterans. See also Figures S3–S5, Table S4, and Video S1.