Cocoonase is indispensable for Lepidoptera insects breaking the sealed cocoon

Abstract

Many insects spin cocoons to protect the pupae from unfavorable environments and predators. After emerging from the pupa, the moths must escape from the sealed cocoons. Previous works identified cocoonase as the active enzyme loosening the cocoon to form an escape-hatch. Here, using bioinformatics tools, we show that cocoonase is specific to Lepidoptera and that it probably existed before the occurrence of lepidopteran insects spinning cocoons. Despite differences in cocooning behavior, we further show that cocoonase evolved by purification selection in Lepidoptera and that the selection is more intense in lepidopteran insects spinning sealed cocoons. Experimentally, we applied gene editing techniques to the silkworm Bombyx mori, which spins a dense and sealed cocoon, as a model of lepidopteran insects spinning sealed cocoons. We knocked out cocoonase using the CRISPR/Cas9 system. The adults of homozygous knock-out mutants were completely formed and viable but stayed trapped and died naturally in the cocoon. This is the first experimental and phenotypic evidence that cocoonase is the determining factor for breaking the cocoon. This work led to a novel silkworm strain yielding permanently intact cocoons and provides a new strategy for controlling the pests that form cocoons.

Fig 1. Phylogenetic analysis of cocoonase homologs.

The phylogenetic tree was constructed using the Maximum Likelihood method, based on the Poisson correction model with 1,000 bootstrap replicates, and bootstrap support values more than 50% were shown in the tree. The 295 cocoonase homologs from 12 classes, identified with different colored fonts, are divided into 14 clades, showed in the outer ring. And 70 protein sequences from the Lepidoptera were aggregated with the clade containing silkworm cocoonase, shown in the orange curved area.

Fig 2. Syntenic analysis of cocoonase genes in insects.

Syntenic analysis of cocoonase genes in lepidopteran species marked with purple clades and three outgroups with available linkage information. Orthologous genes are identified by a border with same color, and non-homologous genes and genes without functional annotation are indicated by ORFs and marked by a gray border. The green rectangles represent predicted transposons. The Arabic numerals in the figure showed the copy number of the cocoonase gene in each species. The two cocoonase genes of Plutella xylostella were located at different scaffold, and the IDs of the two enzymes were shown next to the gene on the scaffold diagram.

Fig 3. The diverse cocooning behaviors of lepidopteran insects.

Based on the habits of cocooning or not and the characteristics of the cocoons, lepidopteran insects not spinning a cocoon, spinning a sealed cocoon, or having an unsealed cocoon were classified roughly and displayed on the right panels.

Fig 4. Temporal expression profile of BmCoc throughout the life cycle.

Temporal expression profile of BmCoc was investigated throughout the life cycle from the stages of early embryo to the adult moth, and the silkworm ribosomal protein L3 (RpL3) was used as the internal reference. I144, I156, I168h: the time after incubation; 1^st 1d: day 1 of the first instar; 1^st m: the molting stage of the first instar; 2^nd 1d: day 1 of the second instar; 2^nd 2d: day 2 of the second instar; 2^nd m: the molting stage of the second instar; 3^rd 1d: day 1 of the third instar; 3^rd m: the molting stage of the third instar; 4^th 1d: day 1 of the fourth instar; 4^th m: the molting stage of the fourth instar; 5^th 1d: day 1 of the fifth instar; W0: the time the wandering stage was initiated; P1–P8: days 1–8 of pupae; M: moth.

Fig 5. Knockout of BmCoc in silkworm by CRISPR/Cas9 system.

(A) Gene structure with gRNA targeting exon 3 of BmCoc. (B) qRT-PCR analysis of the relative expression level of BmCoc in the wildtype (BmCoc^+/+) and homozygous individuals (BmCoc^-/-) on day 8 of the pupal stage, using cDNAs of heads as templates. The expression level was compared between BmCoc^-/- and BmCoc^+/+ by two-tailed Student’s t-tests. ** P < 0.01. (C) The SDS-PAGE analysis for the proteins contained in the silkmoth-vomiting fluid collected from wild type and BmCoc^-/- individuals. The gel was stained with Coomassie brilliant blue R-250. The protein band indicated by the red arrow is the mature cocoonase with a predicted molecular weight of 23.7 kDa. M indicates a protein marker and H1-H8 lanes represent different fluid samples from eight homozygous individuals.

Fig 6. Effects of knocking out BmCoc on moth eclosion and emergence.

(A) Unpierced cocoon rate of BmCoc^+/+ and BmCoc^-/- individuals. One hundred individuals of wild type and BmCoc^-/- homozygotes were randomly selected for analysis. (B) Morphological observations of BmCoc^+/+ and BmCoc^-/- cocoons. BmCoc^+/+ cocoons with pierced holes caused by the moth escaping from the cocoon are shown on the top image. Intact cocoons of BmCoc^-/- and split cocoons with a successfully metamorphosed moth inside are shown below. Bar = 1 cm. (C) Morphological observations of homozygotes naturally stored for more than six months. Inside the cocoons are dried moths that have died naturally. Bar = 2 cm.