Genome of venomous caterpillar Doratifera vulnerans reveals recruitment of immune peptides and their adaptation as pain-inducing toxins.

Genome of venomous caterpillar Doratifera vulnerans reveals recruitment of immune peptides and their adaptation as pain-inducing toxins - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Proc Natl Acad Sci U S A . 2025 Dec 1;122(49):e2513640122. doi: 10.1073/pnas.2513640122 Genome of venomous caterpillar Doratifera vulnerans reveals recruitment of immune peptides and their adaptation as pain-inducing toxins Mohaddeseh H Goudarzi a, b , Samuel D Robinson a , Fernanda C Cardoso a , Kushal Suryamohan c , Nicole Lawrence a, b , David A Eagles a, b , Huy N Hoang a, b , Irina Vetter a, d , David P Fairlie a, b , Somasekar Seshagiri c, e , Glenn F King a, b, 1 , Andrew A Walker a, b, 1 Author information Article notes Copyright and License information a Institute for Molecular Bioscience, The University of Queensland, St. Lucia, QLD 4072, Australia b Australian Research Council Centre of Excellence for Innovations in Protein and Peptide Science, The University of Queensland, St. Lucia, QLD 4072, Australia c Research and Development, MedGenome Inc., Foster City, CA 94404 d School of Pharmacy, The University of Queensland, Woolloongabba, QLD 4102, Australia e Research and Development, Science, Society, Education and Research Foundation, San Carlos, CA 94070 1 To whom correspondence may be addressed. Email: glenn.king@imb.uq.edu.au or a.walker@imb.uq.edu.au . Edited by Michael Strand, University of Georgia, Athens, GA; received May 29, 2025; accepted September 10, 2025 Received 2025 May 29; Accepted 2025 Sep 10; Issue date 2025 Dec 9. Copyright © 2025 the Author(s). Published by PNAS. This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND) . PMC Copyright notice PMCID: PMC12704794  PMID: 41325521 Significance The modes by which venom toxins evolve is debated, but sparse detailed genomic data are available for venomous animal groups other than snakes to inform this debate. The high-quality genome data reported in this study allows a precise reconstruction of the evolutionary history of toxin-encoding genes in an emerging model species of venomous caterpillar. We find that the previously described pain-causing toxins in the venom of this caterpillar are derived from an immune peptide family, the cecropins, that are usually known for their antimicrobial activity. Following recruitment as venom peptides, strong evolutionary pressures have shaped these peptides into potent pain-causing defensive toxins, revealing a unique signature of venom toxin evolution. Keywords: molecular evolution, venom peptide, insect venom, Lepidoptera, Limacodidae Abstract Gene duplication followed by adaptation to new selection pressures has been proposed to be of central importance in the evolution of venom toxins. Coupling high-quality genome data with quantitative bioactivity readouts can be used to understand how venom toxins evolved, but such studies are rare. Here, we report a near chromosomal-level genome assembly for Doratifera vulnerans (Lepidoptera: Limacodidae), which is venomous in the larval stage. We identify 115 gene loci that produce the polypeptide toxins in venom, including numerous multigene families as well as single-copy genes. Previously described membrane-permeabilizing peptides that cause pain are shown to be encoded by a gene cluster on chromosome 7 that also encodes multiple members of the cecropin family, which are insect innate immunity peptides. These data reveal the origin of defensive toxins from immune peptides followed by strong sequence divergence driven by positive selection pressures. Dv13, which is present as a trace component in the venom, and which has sequence features conserved with canonical cecropin A, potently inhibited the growth of Gram-negative bacteria and fungi, but only weakly permeabilized the membranes of mammalian neurons with EC 50 values >100 µM. In contrast, peptides Dv11 and Dv12 are abundant in the venom, have sequence features divergent from Dv13, and potently disrupt mammalian neuronal membranes with EC 50 values as low as 190 nM, but have reduced antimicrobial activity. These data provide evidence for the adaptation of innate immune peptides as specialized defensive venom toxins and provide a natural experiment informing the structure–activity relationships of cecropin family peptides. The evolution of a venomous animal from a nonvenomous ancestor requires genetic substrates for producing toxic molecules. It is accepted that nonvenomous animals possess a range of peptides and proteins with innate toxin-like properties, although the modes by which venom toxins evolve continue to be debated ( 1 – 6 ). The concept of toxin gene “recruitment” ( 7 ) is closely related to Ohno’s classic model of gene neofunctionalization ( 8 ) in which random gene duplication leads to two resulting copies with redundant functions, allowing one copy to undergo mutations that produce a new function without incurring a fitness disadvantage to the organism. For example, scorpion ion-channel modulating peptide toxins are thought to have evolved by neofunctionalization of antimicrobial cystine-stabilized α/β-defensins ( 9 ), and it has been demonstrated that ion-channel-modulating activity can be conferred by making relatively minor changes to antimicrobial CS-α/β-defensins ( 10 , 11 ). The concept of toxin “restriction” instead highlights subfunctionalization, the process of specialization of daughter genes to optimize different subactivities present in the ancestral gene ( 1 , 4 ). There are few clear examples of subfunctionalization during venom toxin evolution, but the pattern of domain loss that occurred in snake venom metalloproteinases to produce toxins with diverse functions is one example ( 12 , 13 ). Variations of neo- and subfunctionalization models are distinguished by whether the final function(s) of daughter genes are new or were present as subfunctions of the ancestral gene, and if gene duplication and subsequent mutations to the daughter genes were neutral or provided fitness advantages ( 14 , 15 ). However, actual toxin evolution is likely to be more complex than these theoretical models suggest. There is strong evidence to suggest that gene duplication contributes strongly to the evolution of venom toxins ( 16 , 17 ), but emerging studies have highlighted that venom evolution does not always require gene duplication, since simple mutation in cis -regulatory elements can be sufficient for recruitment of single-copy genes as venom toxins ( 18 , 19 ). There are a dearth of studies empirically evaluating toxin evolution via neo- or subfunctionalization that combine analysis of high-quality genome data with quantification of the activity of toxic gene products. Currently, our understanding of the genomic basis of venom production depends on a handful of species ( 20 , 21 ), most of which are snakes ( 17 , 22 , 23 ). Further genomic studies are required to understand how venom toxins evolved in other lineages. Doratifera vulnerans is a moth belonging to the family Limacodidae. About half of the 600 species in this lepidopteran family produce venom in the larval (caterpillar) stages, including D. vulnerans ( 24 ). The limited data available suggest that venom production by limacodid caterpillars is adaptive due to causing avoidance and/or pain when injected, and thereby reducing predation by both vertebrate and invertebrate predators ( 25 – 27 ). Recently, we used a proteotranscriptomic approach to show that the venom of D. vulnerans contains 151 different peptides and proteins, including neurohormone-like peptides, cecropin-like peptides, and inhibitor cystine knot (ICK)-like peptides, which we grouped into 62 different toxin families based on primary structure ( 28 ). Experiments to determine which toxin(s) in D. vulnerans venom are responsible for inducing pain in mammals, which comprised both assays of primary murine peripheral sensory neuron activation and behavioral assays after injection of synthetic peptides into mice, led to the identification of two peptide toxins, Dv12 and Dv11 ( 28 ). These and nine other members of this family (Limacoditoxin Family 2) showed sequence similarity to the cecropin family of innate immunity peptides. Cecropin A from the nonvenomous moth Hyalophora cecropia was the first insect antimicrobial peptide (AMP) to be purified ( 29 ). Related peptides have since been identified in other lepidopterans, dipterans, and coleopterans ( 30 ). Cecropin A is a linear 37-residue AMP that forms an α-helical structure with an amphipathic arrangement of hydrophobic and cationic amino acids. Consistent with other AMPs, these features contribute to killing bacteria by permeabilizing membranes that are rich in negatively charged phospholipids. Due to this function, cecropins preferentially disrupt prokaryotic membranes, including those of multidrug-resistant Gram-negative bacteria such as Acinetobacter baumanii and Pseudomonas aeruginosa ( 31 ), over more neutrally charged eukaryotic membranes. The structure–activity relationships of cecropins have been studied with the aim of developing analogs that selectively target microbes ( 32 – 36 ) or parasites ( 37 , 38 ), or for use as food additives ( 39 , 40 ). Structurally, cecropin A is divided by a central Pro residue into an amphiphilic, cationic N terminus, and a more hydrophilic C-terminus, with an amidated C-terminal residue ( 41 ). The N- and C-terminal halves of the peptide form α-helices in the presence of phospholipids ( 42 – 44 ), which are bisected by the central helix-breaking Pro residue. Though the exact mechanism of pore formation is debated, interactions between cationic side-chains and phospholipid head groups, and between hydrophobic side-chains and phospholipid tail groups, are likely important. The mechanism of pore formation by cecropins is likely similar to those of other AMPs as well as those of venom peptide toxins produced by ants, bees, and wasps ( 45 ). In this study, we combined Oxford Nanopore Technology long-read sequencing, Omni-C, and Illumina short-read datasets to assemble a chromosomal-level genome of D. vulnerans , which we used to analyze the genetic basis of venom production. We found numerous instances of toxin-encoding genes occurring in clusters or tandem arrays, providing evidence for the importance of gene duplication in venom evolution, as well as many toxin-encoding genes occurring as single copies. We demonstrate that cecropin family peptides are either functionally specialized as pain-causing venom toxins or retain their original canonical functional specialization as AMPs, though both kinds of peptides possess weak activity for their nonspecialized role. These data provide strong support for evolution of defensive venom toxins from innate immunity peptides, which is similar to but more complex than the classical notions of neofunctionalization and subfunctionalization. 1. Results 1.1. Genome Sequencing. We assembled the nuclear genome of D. vulnerans by combining Oxford Nanopore Technology (ONT), Illumina, and Omni-C sequencing (raw reads are available from the Sequence Read Archive with Bioproject Accession PRJNA1218917) ( 46 ). The final genome of 1.09 Gb (DDBJ/ENA/GenBank accession code JBLKOF000000000 , JBLKOF010000000 ) ( 47 ) consists of 12 main scaffolds comprising 985 Mb and 1,294 smaller scaffolds, representing a near-chromosomal assembly ( Fig. 1 A and B and SI Appendix , Figs. S1 and S2 and Table S1 ). This genome was highly complete, with 95.1% of Insecta BUSCOs recovered. Comparison with other high-quality lepidopteran genomes indicated high levels of synteny with a moth from the same superfamily (Zygaenoidea, the six-spot burnet moth Zygaena filipendulae ) but extensive chromosomal rearrangements compared to moths from even closely related superfamilies (i.e., the diamondback moth Plutella xylostella , Yponomeutoidea; and the cabbage white butterfly Pieris rapae , Papilionoidea) ( Fig. 1 C and D ). Fig. 1. Open in a new tab Genome of D. vulnerans and chromosomal rearrangement among Lepidoptera. ( A ) Snail plot showing high genome completeness (95.1% Insecta BUSCOs) and contiguity (N 50 54.7 Mb). ( B ) Link density histogram generated from Omni-C data. Each square shows linkage between the 12 longest scaffolds, with red diagonals indicating strong linkage due to adjacent locations on the same scaffold. ( C ) De novo genome assembly of D. vulnerans . Outermost to innermost tracks show scaffolds ordered by length, repeat content, gene density, and GC content (%). The average repeat content is indicated by the red line. All data were plotted in 100-kb windows. ( D ) Chromosome painting depicting synteny between D. vulnerans and the six-spot burnet moth Z. filipendulae (GCA_907165275.2), diamondback moth P. xylostella (GCF_932276165.1), and cabbage white butterfly Pieris rapae (GCF_905147795.1) illustrating extensive chromosomal rearrangement within lepidopteran genomes. 1.2. Genes Encoding Venom Toxins Occur in Clusters or as Single Genes. To examine the genetic basis of venom toxin production, we mapped the coding sequences of venom polypeptides that were detected by proteomics ( 28 ) against the D. vulnerans genome. Of the 151 toxin-encoding sequences previously obtained by RNA-Seq ( 28 ), 149 were mapped to 115 loci in the genome ( Fig. 2 and Dataset S1 ). This result suggests that some previously identified toxin variants represent allelic variants rather than the products of paralogous genes. Of the 115 loci, 35 (30%) represent toxin families consisting of a single member. The remaining 80 toxins (70%) are products of genes that occur in a cluster or tandem array of at least two genes. Several peptide toxins that were not previously grouped into families by sequence similarity ( 28 ) were found to be encoded by genes occurring in a cluster ( Fig. 2 , pink circles), suggesting that they also most likely originated by duplication and sequence divergence. Fig. 2. Open in a new tab Genetic basis of venom toxin production in D. vulnerans . ( A ) Loci encoding venom toxins. Blue circles indicate the position of single-copy genes. Orange and pink circles indicate the position of gene clusters with area proportional to the number of gene products detected in venom. Orange gene clusters represent families previously classified on the basis of sequence similarity, while pink indicates genes that are newly grouped based on genomic clustering. CHH, crustacean hyperglycaemic hormone; OBP, odorant binding protein; ICK, inhibitor cystine knot; ACP, adipokinetic hormone/corazonin-like peptide; MBF2, similar to multiprotein bridging factor 2. ( B ) Heatmap showing expression of venom toxins [log 2 (TPM+1)] in larval tissues (venom spines, ventral surface, internal organs) and ontogenic stages (whole larva, whole adult). ( C ) Arrangement of the cluster of seven genes encoding cecropin family peptides on chromosome 7. ( D ) Shared exon/intron architecture of genes encoding cecropin family peptides. Bm, Bombyx mori gene beginning at 10505632 on NC_051383.1 ( Dataset S2 ). All 12 major scaffolds contributed to venom production, with scaffolds 4 and 10 in particular containing numerous toxin-encoding genes. The largest family of toxins in D. vulnerans venom, the ICK-like peptides (Family 3), consisted of 41 previously identified toxins ( 28 ). These toxins were the product of 29 gene loci, 26 of which spanned the region 71,973,022 to 78,831,324 bp on chromosome 4. Genes encoding adipokinetic hormone/corazonin-like peptide (ACP)-like toxins (Family 1) and peptide toxins in Families 5, 7, 13, 14, 24, 47, 50 to 52, and 58 were located close to this ICK peptide cluster, spanning the region 76,389,289 to 80,173,872 bp, but they do not share sequence similarity with the ICK-like peptides. Although D. vulnerans venom contains ACP-like peptides, nontoxin ACP, and ACP receptor genes located on different chromosomes to the ACP-like toxin-encoding genes are also present in the genome. Toxin-encoding genes were more highly expressed in venom spines compared to whole larva, whole adult, larval ventral surface, or internal organs and were more highly expressed in whole larva (including spines) compared to whole adult ( Fig. 2 B ). However, some venom components were more highly expressed in other tissues or ontogenic stages compared to the larval venom spines, suggesting that they have multiple roles, some of which are distinct from venom physiology. These results are consistent with other studies showing that venom-encoding genes evolve by multiple mechanisms, including gene duplication and neofunctionalization as well as co-option of single-copy genes ( 1 , 2 , 18 , 19 , 48 ). 1.3. D. vulnerans Produces Canonical and Venom-Adapted Cecropins. Cecropin family peptides (Family 2) are encoded by a cluster of seven genes spanning between base pairs 16,968,623 and 17,127,600 on chromosome 7 ( Fig. 2 C ). These genes comprise two exons separated by an intron within the mature peptide sequence, an architecture that is typical of other lepidopteran cecropin genes ( Fig. 2 D and Dataset S2 ). Comparison of the cecropin-encoding genes of D. vulnerans with those of nonvenomous species revealed that all species possess multiple cecropin-encoding genes, ranging from two for Z. filipendulae , which is a member of the same superfamily as D. vulnerans (Zygaenoidea), to nine for the silkworm Bombyx mori ( Dataset S2 ). The single exception is a nonvenomous limacodid, Apoda limacodes , whose genome ( 49 ) contains no cecropin-like genes. These data strongly suggest that the seven D. vulnerans genes encoding cecropin family peptides evolved by shared descent from a common ancestor and in a process involving gene duplication. Despite this evidence of common ancestry, the cecropin-family peptides of D. vulnerans show either a canonical primary structure typical of cecropin A, or divergent sequences that we propose are venom-adapted ( Fig. 3 A ). Canonical cecropins are represented by two genes encoding peptides that differ by a single amino acid and closely correspond to the sequence of the previously identified peptide Dv13 ( 28 ). Venom-adapted cecropins are represented by the five remaining genes, from which the remaining 10 cecropin family peptides detected in venom originate, including Dv11 and Dv12 which were previously identified as key pain-inducing peptides in D. vulnerans venom ( Fig. 2 C ) ( 28 ). The primary structure of Dv13 is highly similar to canonical cecropins with 76% identity and 89% similarity to Zf_cec1, a cecropin of the nonvenomous zygenoid caterpillar Z. filipendulae , whereas Dv11 and Dv12 have <20% and <60% similarity to this peptide ( Dataset S2 ). Moreover, the venom-adapted peptides are abundant in venom, whereas Dv13 is present at much lower levels ( Fig. 3 B ). Fig. 3. Open in a new tab Canonical and venom-adapted cecropins produced by D. vulnerans. ( A ) Sequence alignment of selected cecropin family peptides from nonvenomous lepidopterans, D. vulnerans , and synthetic analogs. Cationic, anionic, and hydrophobic residues are shown in blue, red, and pink, respectively. Pro residues are highlighted in yellow. # indicates C-terminal amidation. ( B ) Comparison of the levels of expression of each cecropin family peptide-encoding gene in venom spines with the abundance of the encoded peptides in venom. ( C ) While all cecropins feature a positively charged (blue) N terminus, only venom-adapted cecropins show characteristic negatively charged (red) C-termini. The Inset shows the charge over the normalized length for all venom-adapted peptides. The primary structure of canonical lepidopteran cecropins such as Dv13 comprises a cationic N-terminal half and a more hydrophobic α-amidated C-terminal half separated by a central Pro residue ( Fig. 3 A ). In contrast, the venom-adapted cecropins have a strongly anionic C-terminal region without C-terminal amidation ( Fig. 3 A and C ). Dv11 and Dv12 are slightly shorter than Dv13 and lack the central Pro residue, while the other venom-adapted peptides Dv14–Dv19 are substantially longer (mature length 48 to 85 residues compared to 31 to 44 residues for the other cecropin family peptides shown in Fig. 3 A ), but they retain the central Pro residue. The higher expression levels in venom spines and atypical primary structures of Dv11,12, and Dv14–19 support the notion that they are adapted as venom toxins. Phylogenetic analysis of cecropin-encoding nucleotide sequences of D. vulnerans and other related lepidopterans using maximum likelihood ( Fig. 4 ) indicated that transcripts encoding the venom-adapted cecropins belong to a single clade (bootstrap 100%) which is sister to the clade of transcripts encoding the canonical D. vulnerans cecropins (bootstrap 99%). While transcripts encoding the canonical D. vulnerans cecropins have similar branch lengths compared to cecropins from nonvenomous species, the venom-adapted cecropins have longer branch lengths that reflect their high degree of sequence divergence. To investigate whether the venom-adapted cecropins evolved under positive selection, we estimated the ratio of nonsynonymous to synonymous mutations, ω, using the Phylogenetic Analysis by Maximum Likelihood (PAML) software package. First, we estimated the average ω for the mature functional peptide separately from the secretion signal peptide and prepropeptide, which are regions that are cleaved off during maturation and do not contribute to the function of the mature peptide. We found that, on average, the mature sequences of venom-adapted cecropins evolved under mildly positive selection (ω = 1.05), consistent with strong selection at some sites and toxin neofunctionalization. In contrast, the mature sequences of the D. vulnerans canonical cecropins, and cecropins from nonvenomous species, evolved under strongly negative selection (ω = 0.00 and 0.04, respectively), consistent with selection for conservation of their canonical antimicrobial function, as well as previous results indicating strong negative selection dominating the evolution of the cecropin family in insects ( 50 ). While clear signatures of positive and negative selection are present in sequences encoding the mature regions of venom-adapted and canonical cecropins, respectively, they are missing from regions that do not appear in the mature peptide: The ω ratio for the signal sequence plus prepropeptide for D. vulnerans venom-adapted cecropins, D. vulnerans canonical cecropins, and cecropins from nonvenomous species was 0.39, 0.15, and 0.14, respectively. Fig. 4. Open in a new tab Evolution of D. vulnerans cecropins. The tree shows the maximum likelihood phylogeny of lepidopteran cecropins. Node labels indicate % bootstrap support. Panels on right indicate the outcome of tests for positive selection using PAML. ss, signal sequence. Second, we estimated ω for each codon to detect sites that evolved under positive selection ( P > 0.95). We found 12 such codons within the coding region of the mature sequence of D. vulnerans venom-adapted cecropins, compared to just two sites in the secretion signal peptide plus prepropeptide. No positively selected sites were detected in any regions of the canonical D. vulnerans cecropins ( Fig. 4 , Right ). These data indicate that the venom-adapted cecropins evolved under positive selection pressures, in contrast to canonical cecropins whose evolution is dominated by negative selection pressures ( 50 ). 1.4. Dv11 and Dv12 Are Adapted to Target Eukaryotic Cell Membranes. We hypothesized that some D. vulnerans cecropin-family peptides are adapted as defensive toxins and would have increased potency in disrupting eukaryotic cell membranes. To test this hypothesis, we produced synthetic versions of Dv11, Dv12, and Dv13a, as well as a series of analogs with sequence features intermediate between the canonical and venom-adapted forms ( Fig. 3 A , Lower ). Dv11[E24A, E32A, E33A] and Dv12[E31A, E36A] were designed to test the functional role of anionic side-chains near the C-terminus, Dv13[P24S] to test the contribution of the central Pro residue, and Dv12[S21P]-NH 2 to examine the contribution of the central Pro and C-terminal amidation. As a measure of the algogenic capacity of the synthetic cecropin-family peptides, we tested their ability to activate mammalian dorsal root ganglion (DRG) peripheral sensory neurons in vitro. In this assay, the intracellular calcium concentration [Ca 2+ ] i of DRG neurons was measured as an indicator of neuronal activation that would accompany pain signaling ( 51 ). We previously showed that the addition of 10 µM Dv12 robustly activated DRG neurons, and caused leakage of calcium-sensitive dye from cells and lack of subsequent response to KCl (positive control) ( 28 ) ( Fig. 5 A , Upper ). In contrast, the canonical Dv13 peptide and its proline-deficient analog Dv13[P24S] did not activate DRG neurons at the high concentration of 10 µM, and subsequent addition of KCl caused strong calcium influx ( Fig. 5 A , Lower ). Addition of the remaining modified cecropin family peptides at the same concentration resulted in potent activation of these sensory neurons similar to Dv12 ( Fig. 5 C ). Fig. 5. Open in a new tab Activation of mammalian sensory neurons by cecropin-family peptides. ( A ) Fluorescence photomicrographs of mouse DRG cells before application of peptide ( Left ), 20 s after application of 10 µM peptide ( Middle ), and after application of positive control (20 mM KCl, Right ). ( B ) Concentration–response curves for peptide-induced fluorescence increases from SH-SY5Y cells in FLIPR experiments. Data points show the mean of three technical replicates ± SD. Curves are four-parameter nonlinear regressions. Left , nucleic acid exposure to propidium iodide; AUC, area under curve. Right , changes in intracellular Ca 2+ levels. ( C ) Summary of experiments on DRGs. Data points show median of 80 to 100 cells ± interquartile range. The dotted line indicates no change. ( D ) EC 50 values from curves shown in panel B ± 95% CI. Arrows indicate EC 50 values > 100 µM. Orange and green boxes indicate results from venom-adapted and canonical cecropins, respectively. To more precisely quantify the membrane-permeabilizing potency of each peptide on mammalian neurons, we applied a dilution series of each peptide to the neuroblastoma cell line SH-SY5Y ( Fig. 5 B ). We simultaneously measured intracellular calcium levels ([Ca 2+ ] i ) and membrane permeability to propidium iodide (PI) using a FLIPR PENTA duplex assay ( 52 – 54 ). The venom-adapted Dv12 and Dv11 had potent membrane-permeabilizing activities, with EC 50 values (± bounds of 95% CI) of 0.19 ± 0.06 µM and 1.90 ± 0.85 µM based on [Ca 2+ ] i measurements and 1.87 ± 0.26 µM and 34.06 µM ± 2.32 based on PI measurements, respectively ( Fig. 5 B and D ). In contrast, the canonical Dv13 had minimal (<10%) membrane-permeabilizing activity at 100 µM. Analogs of venom peptides with inserted canonical features (neutralizing acidic residues, central Pro, and/or amidation of the C-terminus) all had decreased membrane permeabilization activity compared to the natural venom-adapted Dv11 and Dv12 ( Fig. 5 B and D ). This suggests that each of the modified features of the venom-adapted peptides (loss of C-terminal amidation, presence of C-terminal anionic residues, and loss of the central proline) is an adaptation that favors the disruption of eukaryotic neuronal cell membranes. The markedly increased potency with which Dv11 and Dv12 disrupt eukaryotic membranes compared to Dv13a provides support for the hypothesis that Dv11 and Dv12 are adapted as pain-causing defensive toxins. 1.5. Dv11 and Dv12 Have Reduced Activity against Gram-Negative Bacteria and Fungi. To investigate whether the venom-adapted peptides Dv11 and Dv12 and their synthetically modified analogs have retained or lost the antimicrobial effects for which cecropins are known ( 32 , 37 , 55 – 58 ), we submitted them and Dv13a to the Community for Antimicrobial Drug Discovery (CO-ADD) ( 59 ), where they were tested against a range of Gram-negative and Gram-positive bacteria and fungi. Consistent with previous reports, the ancestral-like cecropin Dv13a showed potent antimicrobial activity, inhibiting the growth of all Gram-negative bacteria and fungi tested with minimum inhibitory concentrations (MICs) equal or lower than the lowest concentration tested, 62 nM (0.25 µg/mL) ( Table 1 ). In contrast, the venom-adapted peptides Dv11 and Dv12 had reduced activity on Gram-negative bacteria and the yeast Candida albicans ( Table 1 ). However, Dv12 had increased activity compared to Dv13 on the Gram-positive bacterium Staphylococcus aureus , with MIC ≤62 nM. Analogs with neutralized C-terminal anionic side-chains (Dv12[E31A, E36A] and Dv11[E24A, E32A, E33A]) showed minimal differences compared to the wild-type venom peptides. Amidation of the C-terminus of Dv12[E31A, E36A] produced a peptide (Dv12[E31A, E36A]-NH 2 ) with strong potency against all microbes tested. Substitution of the central proline for serine in Dv13[P24S] reduced its activity against C. albicans , but not the other microbes tested ( Table 1 ). These data are consistent with the evolution of specialized defensive toxins from canonical antimicrobial cecropins. Table 1. Effects of D. vulnerans cecropin-family peptides and synthetic peptides on bacterial and fungal pathogens Peptide Gram-positive Gram-negative Fungi S. aureus Escherichia coli Klebsiella pneumoniae P. aeruginosa Acinetobacter baumannii C. albicans Cryptococcus neoformans Dv13 >7,978 ≤62 ≤62 ≤62 ≤62 ≤62 ≤62 Dv13[P24S] >7,998 ≤62 ≤62 ≤62 ≤62 >7,998 ≤62 Dv12[S21P-NH 2 ] 999 3,996 >7,993 ≤62 ≤62 >7,993 ≤62 Dv12[E31A, E36A-NH 2 ] ≤64 ≤64 ≤64 ≤64 ≤64 ≤64 ≤64 Dv11[E24A, E32A, E33A] >9,813 2,453 4,906 4,907 ≤76 >9,813 >9,813 Dv12[E31A, E36A] 1,031 ≤64 516 1,031 ≤64 >8,250 ≤64 Dv11 4,658 >9,316 >9,316 >9,316 ≤72 >9,316 >9,316 Dv12 ≤62 501 4,005 2,002 ≤63 >8,010 ≤62 Open in a new tab MIC are shown in nM. Bold and italic text denote results from venom-adapted and canonical cecropins, respectively. 1.6. Biophysical Properties of Peptides Underlying the Ability to Permeabilize Neurons. We used circular dichroism (CD) spectropolarimetry and surface plasmon resonance (SPR) to investigate how the biophysical properties of the cecropin peptides correlate with their activity. All peptides were unstructured in aqueous solution as indicated by CD minima below 200 nm ( SI Appendix , Fig. S1 C ). In the presence of 20 µM SDS micelles, the CD spectra displayed minima at 208 and 222 nm and a maximum near 190 nm, characteristics indicative of α-helical secondary structure ( SI Appendix , Fig. S1 D ). This structural transition from random coil to α-helix upon interaction with biological membranes or membrane mimetics is typical of many hymenopteran venom peptides, including melittin, the most abundant toxin in honeybee ( Apis mellifera ) venom ( 45 , 60 ). Dv11 and Dv12 were more helical in the presence of SDS micelles (81 to 91%) than the ancestral-type Dv13a (55%), consistent with the presence of the helix-breaking central Pro residue in the latter but not the former ( SI Appendix , Fig. S1 F ). Adding a central Pro to amidated Dv12 (Dv12[S21P]-NH 2 ) resulted in reduced helicity in SDS of 34%. However, removing the central Pro residue from Dv13 (Dv13[P24S]) did not increase helicity in SDS but further reduced it, from 55 to 24%. Analogs with neutralized acidic residues showed intermediate to high helicity of 49 to 89%. We hypothesized that the venom peptides Dv11 and Dv12 might be adapted to bind eukaryotic membranes, whereas ancestral cecropins might be better adapted for binding to bacterial membranes. To test this hypothesis, we used SPR to examine peptide binding to lipid bilayers. POPC was used to represent the neutral membranes of healthy mammalian cells, and a mixture of POPC with POPG (4:1) was used to model the more negatively charged surface of bacterial cell membranes. However, while these experiments unequivocally demonstrated that Dv11 and Dv12 bind phospholipids, they did not reveal that venom peptides and canonical cecropins have different phospholipid preferences. Instead, all peptides bound to the negatively charged bacterial-like POPC/POPG (4:1) bilayers with higher affinity than the eukaryotic-like POPC membranes ( SI Appendix , Fig. S1 A , B , and E ). Dv12 displayed stronger lipid binding than Dv13 or Dv11, which was supported by molecular dynamics (MD) simulations ( SI Appendix , Fig. S2 ). To investigate how peptide structural features, biophysical properties, and bioactivities (EC 50 values in FLIPR experiments) are related, we compared peptide helicity (CD) and peptide–lipid binding (SPR) data alongside peptide structural features and bioactivity data ( SI Appendix , Table S2 ). Since the different structural features of peptides assayed in this study are not independently distributed among peptides, these correlations are suggestive rather than indicative. Nevertheless, they suggest that increased charge asymmetry, increased isoelectric point, replacement of Lys with Arg, loss of C-terminal amidation, higher helical propensity, and higher phospholipid binding affinity are among the specific structural alterations that underlie the adaptation of Dv11 and Dv12 as pain-causing defensive toxins. 2. Discussion The high analytical power and relatively low cost of venom proteomics and venom-gland transcriptomics ( 61 ) have led to their extensive use to study venom composition and resolve toxin primary structures ( 62 – 64 ). Lacking from most studies, however, is high-quality genome data that allow deeper analysis of toxin evolution, including the relative loci of toxin-encoding genes and those adjacent to them, their intron/exon architectures, and the ability to distinguish between toxin variants due to multiple gene loci, alternative splicing, and allelic variation ( 17 , 20 – 23 ). In this study, we assembled a nuclear genome of the venomous lepidopteran D. vulnerans that is near complete and highly contiguous. Analysis of this genome together with venom proteomic and venom spine transcriptomic data allowed us to identify 115 gene loci encoding venom toxins and unambiguously assign 149 of the 151 previously reported toxin variants ( 28 ) as either allelic variants or the products of distinct gene loci. We uncovered strong evidence for toxin production from tandem arrays and duplicated genes, as well as single-copy genes. These data are consistent with a “many mechanisms” model based on emerging data from multiple groups of venomous animals that highlights venom toxin recruitment and evolution proceeding by diverse genetic processes in parallel ( 2 , 18 , 19 ). In this study, genomic data were crucial to convincingly demonstrate the shared ancestry of canonical and venom-adapted cecropins, which might otherwise share similar primary structure due solely to convergent evolution. Both the genomic loci and the exon/intron architecture of D. vulnerans cecropins strongly support their shared ancestry within the cecropin family of peptides that are ubiquitous in nonvenomous Lepidoptera. Cecropins are one of several groups of AMPs characterized by their cationic and amphipathic nature, propensity to form α-helices, and ability to disrupt lipid bilayers. AMPs from the skin of frogs, such as maculatin ( 65 ), the cathelicidin-derived human peptide LL37 ( 66 ), and venom peptides from hymenopteran insects ( 45 ), are well-known examples of AMPs with similar properties that evolved independently. We found that the primary structures of cecropins from nonvenomous lepidopteran species have been strongly conserved since the ancestors of P. xylostella and the other species examined in this study diverged approximately 130 Mya ( 67 ), revealing that primary structural adaptations more sophisticated than simply having a cationic/amphipathic nature are important for cecropin function. Previous structure–function analyses have indicated that C-terminal amidation is a key feature for the potent antimicrobial activity of cecropins ( 68 , 70 , 71 ) and other cationic amphipathic peptides ( 72 , 73 ). In contrast, we found that cecropins with carboxyl termini have greater potency against the membranes of mammalian neuronal cells. Several studies have focused on the importance of the cationic residues of cecropins that are essential for antimicrobial activity ( 69 , 74 ) whereas few have tested the effect of anionic residues despite these being prominent features of some membrane-disrupting venom peptides such as MGTX 1 -Mg1a from the bull ant Myrmecia gulosa ( 75 ). It has previously been reported that anionic residues reduce antimicrobial activity ( 69 , 76 ), whereas we found that the presence of anionic residues near the C-terminus of venom-adapted cecropins is a key feature conferring greater potency to disrupt the membranes of mammalian neurons. Other features of canonical cecropins reported to be important for antimicrobial activity, such as a conserved Trp residue in the second position ( 43 , 74 ) are not present in venom-adapted cecropins. The functional significance of a Pro-containing hinge region is more difficult to interpret: flexible hinge regions may be either present ( 77 – 79 ) or absent ( 66 ) in AMPs and their functional effect depends on the remaining structure of the peptide ( 41 , 78 , 80 – 85 ). Among D. vulnerans venom-adapted cecropins, the hinge-associated Pro residue is absent from the shorter peptides Dv11 and Dv12 but retained in the longer forms. The activity of the D. vulnerans venom-adapted cecropins is consistent with their evolution under selection to potently disrupt eukaryotic and/or neuronal membranes, combined with relaxation of selection pressures for selective and potent activity against microbial membranes. We observed that Dv13 had antimicrobial activity typical of canonical cecropins, with potent activity against Gram-negative bacteria and fungi and weaker activity against Gram-positive bacteria ( 30 , 86 , 88 , 89 ). Given the low activity of canonical cecropins against mammalian neuronal membranes, it is perhaps surprising they were ever recruited as venom toxins, but this evolutionary barrier might have been lowered by previously higher concentrations in venom or synergy with other toxins. In contrast to canonical cecropins, venom-adapted cecropins have much weaker antimicrobial activity, although Dv12 unexpectedly was found to have increased potency against Gram-positive S. aureus . These data are consistent with reports that cecropin chimeras with melittin, magainin, or LL37 can have potent activity against Gram-positive bacteria ( 83 , 87 ). The reason that cecropins do not already possess optimized antimicrobial activity ( 69 , 83 , 87 ) may be because they are evolving under pleiotropic selection not only for antimicrobial activity but also to avoid disrupting the cell membranes of the insect producing them. Animals that produce toxins, either for innate immunity or as venom toxins, must employ strategies to protect themselves from toxicity. Such protective mechanisms include production of toxin inhibitors ( 90 ), storage in inactive form ( 91 ), mutations in their own ion channels that confer resistance ( 92 ), alteration of lipid bilayer compositions ( 93 ), and compartmentalization of toxins separate from body tissues. In the case of limacodid venom peptides, the venom is compartmentalized into the cuticle-lined venom reservoir inside venom spines, preventing the toxin from coming into contact with cells other than the secretory cells that produce them ( 28 ). Thus, compared to canonical cecropins, venom-adapted cecropins may also be released from pressure to avoid activity against animal cells. Consistent with this hypothesis, and with previously reported data ( 83 ), we found that primary structure alterations that confer increased potency against animal cells also confer activity against S. aureus . An alternative hypothesis is that neofunctionalization of cecropins to target Gram-positive bacteria might have been an intermediate step in the evolution of D. vulnerans venom-adapted cecropins. In any case, the antimicrobial activities of D. vulnerans cecropins provides insights into the design of antimicrobial peptides for use in biotechnology. Neofunctionalization and subfunctionalization are two key evolutionary models used to explain how venom toxins have evolved from endogenous, nonvenom peptides and proteins ( 1 , 2 , 5 ). Neofunctionalization of cecropins has also been noted in the fruit fly Drosophila melanogaster , where andropin, a cecropin-related antimicrobial peptide, plays a crucial role in safeguarding seminal fluid and the male reproductive tract against microbial contamination ( 94 ). In this study, we obtained both high-quality genome data, and quantitative readouts of gene-product functions associated with distinct physiological roles: antimicrobial activity and disruption of sensory neuron membranes. The combination of these datasets allows a rare empirical investigation of the ability of models such as neo- and subfunctionalization to explain venom toxin recruitment and evolution. We found that venom-adapted cecropins have evolved through a process that bears resemblance to both models but that neither model is sufficiently complex. For example, these models typically begin with duplication of a single ancestral gene ( 5 , 8 , 14 , 15 ), whereas antimicrobial cecropins of nonvenomous Lepidoptera are typically encoded by a tandem array that is evolving in a “birth-and-death” mode prior to recruitment of the gene products as venom toxins ( 95 – 99 ). Furthermore, we found that canonical cecropins have potent activity against fungi and Gram-negative bacteria, but only weak ability to disrupt mammalian sensory neurons. Venom-adapted cecropins have evolved potent activity against neuronal membranes, but their antimicrobial activity has been ameliorated or changed in nature rather than entirely lost. Subfunctionalization is usually associated with segregation of multiple functions present in the ancestral gene product into distinct daughter genes and does not well capture the massive increases in potency against neuronal membranes that venom-adapted cecropins have achieved. Neofunctionalization infers the loss of ancestral gene function and gain of a new function, neither of which is entirely satisfied in this case. This study provides a real-world example of the recruitment of innate immunity peptides into venom and their adaptation as pain-causing defensive toxins. 3. Methods We combined Oxford Nanopore Technology long reads, Illumina gDNA short reads, and Dovetail Omni-C technology to sequence and assemble the nuclear genome of D. vulnerans and RNA-Seq of diverse tissues and life stages to annotate it. D. vulnerans cecropin nucleotide and amino acid sequences were analyzed phylogenetically using the IQ-TREE 2 and PAML software packages ( 100 , 101 ). Cecropin peptides and analogs were produced using solid phase peptide synthesis using Fmoc chemistry. Peptides were tested for bioactivity using fluorescence imaging assays of primary murine peripheral neurons (dorsal root ganglia cells) and the neuroblastoma cell line SH-SY5Y as previously described ( 52 , 102 ). Antimicrobial activity was measured in growth inhibition assays by the Centre for Open Antimicrobial Drug Discovery (Co-ADD) ( 59 ). Lipid binding by synthetic peptides was assessed using SPR assays ( 103 ) and MD simulations. For full details refer to SI Appendix , Supplementary Methods . Supplementary Material Appendix 01 (PDF) pnas.2513640122.sapp.pdf (3.8MB, pdf) Dataset S01 (XLSX) pnas.2513640122.sd01.xlsx (70.2KB, xlsx) Dataset S02 (XLSX) pnas.2513640122.sd02.xlsx (24.5KB, xlsx) Acknowledgments This work was funded by the Australian Research Council through Discovery Project DP200102867 to A.A.W. and Centre of Excellence grant CE200100012 to D.P.F. and G.F.K.; by the Australian National Health and Medical Research Council through Investigator Grants 2017461 to S.D.R. and 2035090 to G.F.K.; and by The University of Queensland (UQ) through a PhD scholarship to M.H.G. N.L. was supported by the US Department of Defense Congressionally Directed Medical Research Program grant W81XWH-22-1-0219. The antimicrobial screening performed by CO-ADD (The Community for Antimicrobial Drug Discovery) was funded by the Wellcome Trust (UK) and UQ. Author contributions M.H.G., S.D.R., F.C.C., K.S., N.L., H.N.H., S.S., G.F.K., and A.A.W. designed research; M.H.G., S.D.R., F.C.C., K.S., N.L., D.A.E., H.N.H., and A.A.W. performed research; I.V., D.P.F., S.S., G.F.K., and A.A.W. contributed new reagents/analytic tools; M.H.G., S.D.R., F.C.C., K.S., N.L., D.A.E., H.N.H., I.V., S.S., G.F.K., and A.A.W. analyzed data; and M.H.G., S.D.R., F.C.C., K.S., N.L., D.A.E., H.N.H., I.V., D.P.F., S.S., G.F.K., and A.A.W. wrote the paper. Competing interests The authors declare no competing interest. Footnotes This article is a PNAS Direct Submission. Contributor Information Glenn F. King, Email: glenn.king@imb.uq.edu.au. Andrew A. Walker, Email: a.walker@imb.uq.edu.au. Data, Materials, and Software Availability This Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession code JBLKOF000000000 ( 47 ). The version described in this paper is version JBLKOF010000000 . Raw gDNA and RNA-Seq reads generated in this project have been deposited in the Sequence Read Archive with BioProject Accession PRJNA1218917 ( 46 ). Supporting Information References 1. Jackson T. N. W., Koludarov I., How the toxin got its toxicity. Front. Pharmacol. 11, 574925–574925 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Walker A. A., The evolutionary dynamics of venom toxins made by insects and other animals. Biochem. Soc. Trans. 38, 1353–1365 (2020). [ DOI ] [ PubMed ] [ Google Scholar ] 3. Hughes A. L., The evolution of functionally novel proteins after gene duplication. Proc. R. Soc. Lond. Ser. B: Biol. Sci. 256, 119–124 (1997). [ DOI ] [ PubMed ] [ Google Scholar ] 4. Hargreaves A. D., Swain M. T., Hegarty M. J., Logan D. W., Mulley J. F., Restriction and recruitment—Gene duplication and the origin and evolution of snake venom toxins. Genome Biol. Evol. 6, 2088–2095 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Fry B. G., et al. , The toxicogenomic multiverse: Convergent recruitment of proteins into animal venoms. Annu. Rev. Genom. Hum. Genet. 10, 483–511 (2009). [ DOI ] [ PubMed ] [ Google Scholar ] 6. Barua A., Mikheyev A. S., An ancient, conserved gene regulatory network led to the rise of oral venom systems. Proc. Natl. Acad. Sci. U.S.A. 118, e2021311118 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Fry B. G., Wuster W., Assembling an arsenal: Origin and evolution of snake venom proteome inferred from phylogenetic analysis of toxin sequences. Mol. Biol. Evol. 21, 870–883 (2004). [ DOI ] [ PubMed ] [ Google Scholar ] 8. Ohno S., Evolution by Gene Duplication (Georg Allen and Unwin Ltd., London, 1970). [ Google Scholar ] 9. Bontems F., Roumestand C., Gilquin B., Menez A., Toma F., Refined structure of charybdotoxin: Common motifs in scorpion toxins and insect defensins. Science 254, 1521–1523 (1991). [ DOI ] [ PubMed ] [ Google Scholar ] 10. Jin J., et al. , Weaponisation ‘on the fly’: Convergent recruitment of knottin and defensin peptide scaffolds into the venom of predatory assassin flies. Insect Biochem. Mol. Biol. 118, 103310 (2020). [ DOI ] [ PubMed ] [ Google Scholar ] 11. Zhu S., et al. , Experimental conversion of a defensin into a neurotoxin: Implications for origin of toxic function. Mol. Biol. Evol. 31, 546–559 (2014). [ DOI ] [ PubMed ] [ Google Scholar ] 12. Casewell N. R., Wagstaff S. C., Harrison R. A., Renjifo C., Wüster W., Domain loss facilitates accelerated evolution and neofunctionalization of duplicate snake venom metalloproteinase toxin genes. Mol. Biol. Evol. 28, 2637–2649 (2011). [ DOI ] [ PubMed ] [ Google Scholar ] 13. Brust A., et al. , Differential evolution and neofunctionalization of snake venom metalloprotease domains. Mol. Cell. Proteomics 12, 651–663 (2013). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Sikosek T., Chan H. S., Bornberg-Bauer E., Escape from adaptive conflict follows from weak functional trade-offs and mutational robustness. Proc. Natl. Acad. Sci. U.S.A. 109, 14888–14893 (2012). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Copley S. D., Evolution of new enzymes by gene duplication and divergence. FEBS J. 287, 1262–1283 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Pineda S. S., et al. , Structural venomics reveals evolution of a complex venom by duplication and diversification of an ancient peptide-encoding gene. Proc. Natl. Acad. Sci. U.S.A. 117, 11399–11408 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Suryamohan K., et al. , The Indian cobra reference genome and transcriptome enables comprehensive identification of venom toxins. Nat. Genet. 52, 106–117 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Martinson E. O., Mrinalini, Kelkar Y. D., Chang C.-H., Werren J. H., The evolution of venom by co-option of single-copy genes. Curr. Biol. 27, 2007–2013 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Drukewitz S. H., Bokelmann L., Undheim E. A. B., von Reumont B. M., Toxins from scratch? Diverse, multimodal gene origins in the predatory robber fly Dasypogon diadema indicate a dynamic venom evolution in dipteran insects. Gigascience 8, giz081 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Sanggaard K. W., et al. , Spider genomes provide insight into composition and evolution of venom and silk. Nat. Commun. 5, 3765 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Peng C., et al. , The first Conus genome assembly reveals a primary genetic central dogma of conopeptides in C. betulinus . Cell Discov. 7, 11 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Vonk F. J., et al. , The king cobra genome reveals dynamic gene evolution and adaptation in the snake venom system. Proc. Natl. Acad. Sci. U.S.A. 110, 20651 (2013). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Shibata H., et al. , The habu genome reveals accelerated evolution of venom protein genes. Sci. Rep. 8, 11300 (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Epstein M. E., Revision and phylogeny of the limacodid-group families, with evolutionary studies on slug caterpillars (Lepidoptera: Zygaenoidea). Smithson. Contrib. Zool. 582, 1–102 (1996). [ Google Scholar ] 25. Murphy S. M., Leahy S. M., Williams L. S., Lill J. T., Stinging spines protect slug caterpillars (Limacodidae) from multiple generalist predators. Behav. Ecol. 21, 153–160 (2010). [ Google Scholar ] 26. Trébouet F., Reichard U. H., Pinkaew N., Malaivijitnond S., Extractive foraging of toxic caterpillars in wild northern pig-tailed macaques ( Macaca leonina ). Primates 59, 185–196 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 27. Southcott R. V., “Lepidopterism in the Australian region” in Records of the Adelaide Children’s Hospital, Sims E. B., Simpson D., Verco P. W., Eds. (1978). vol. 2, pp. 87–173. [ Google Scholar ] 28. Walker A. A., et al. , Production, composition, and mode of action of the painful defensive venom produced by a limacodid caterpillar, Doratifera vulnerans . Proc. Natl. Acad. Sci. U.S.A. 118, e2023815118 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Hultmark D., Steiner H., Rasmuson T., Boman H. G., Insect immunity. Purification and properties of three inducible bactericidal proteins from hemolymph of immunized pupae of Hyalophora cecropia . Eur. J. Biochem. 106, 7–16 (1980). [ DOI ] [ PubMed ] [ Google Scholar ] 30. Brady D., Grapputo A., Romoli O., Sandrelli F., Insect cecropins, antimicrobial peptides with potential therapeutic applications. Int. J. Mol. Sci. 20, 5862 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Lee E., Shin A., Kim Y., Anti-inflammatory activities of Cecropin A and its mechanism of action. Arch. Insect Biochem. Physiol. 88, 31–44 (2015). [ DOI ] [ PubMed ] [ Google Scholar ] 32. Shrestha A., Duwadi D., Jukosky J., Fiering S. N., Cecropin-like antimicrobial peptide protects mice from lethal E. coli infection. PLoS One 14, e0220344 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Hancock R. E. W., Lehrer R., Cationic peptides: A new source of antibiotics. Trends Biotechnol. 16, 82–88 (1998). [ DOI ] [ PubMed ] [ Google Scholar ] 34. Hancock R. E. W., Sahl H. G., Antimicrobial and host-defense peptides as new anti-infective therapeutic strategies. Nat. Biotechnol. 24, 1551–1557 (2006). [ DOI ] [ PubMed ] [ Google Scholar ] 35. Wei X., et al. , A novel cecropin-LL37 hybrid peptide protects mice against EHEC infection-mediated changes in gut microbiota, intestinal inflammation, and impairment of mucosal barrier functions. Front. Immunol. 11, 1361 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Li B., et al. , Cecropin AD ameliorates pneumonia and intestinal injury in mice with Mycoplasma pneumoniae by mediating gut microbiota. BMC Vet. Res. 21, 39 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Gwadz R. W., et al. , Effects of magainins and cecropins on the sporogonic development of malaria parasites in mosquitoes. Infect. Immun. 57, 2628–2633 (1989). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Lou J., et al. , Antimalarial activity of cecropin antimicrobial peptides derived from Anopheles mosquitoes. Antimicrob. Agents Chemother. 68, e00311-24 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Alhadrami H. A., Sayed A. M., Hassan H. M., Rateb M. E., Abdelkader K., Cecropin A improves the antibacterial activity of hen egg white lysozyme against challenging Salmonella enterica serovars. Pharmaceutics 14, 2201 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Wen L.-F., He J.-G., Dose–response effects of an antimicrobial peptide, a cecropin hybrid, on growth performance, nutrient utilisation, bacterial counts in the digesta and intestinal morphology in broilers. Br. J. Nutr. 108, 1756–1763 (2012). [ DOI ] [ PubMed ] [ Google Scholar ] 41. Fink J., Boman A., Boman H. G., Merrifield R. B., Design, synthesis and antibacterial activity of cecropin-like model peptides. Int. J. Pept. Protein Res. 33, 412–421 (1989). [ DOI ] [ PubMed ] [ Google Scholar ] 42. Romoli O., et al. , Enhanced silkworm cecropin B antimicrobial activity against Pseudomonas aeruginosa from single amino acid variation. ACS Infect. Dis. 5, 1200–1213 (2019). [ DOI ] [ PubMed ] [ Google Scholar ] 43. Lee E., et al. , Functional roles of aromatic residues and helices of papiliocin in its antimicrobial and anti-inflammatory activities. Sci. Rep. 5, 12048 (2015). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Holak T. A., et al. , The solution conformation of the antibacterial peptide cecropin A: A nuclear magnetic resonance and dynamical simulated annealing study. Biochemistry 27, 7620–7629 (1988). [ DOI ] [ PubMed ] [ Google Scholar ] 45. Chen J., Guan S.-M., Sun W., Fu H., Melittin, the major pain-producing substance of bee venom. Neurosci. Bull. 32, 265–272 (2016). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Goudarzi M. H., et al. , Doratifera vulnerans isolate:GDNA001 Genome sequencing. Sequence Read Archive. https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1218917 . Deposited 7 February 2025. 47. Goudarzi M. H., et al. , Doratifera vulnerans GDNA001. GenBank Whole Genome Shotgun. https://www.ncbi.nlm.nih.gov/nuccore/JBLKOF000000000 . Deposited 6 February 2025. 48. Lynch V. J., Inventing an arsenal: Adaptive evolution and neofunctionalization of snake venom phospholipase A2 genes. BMC Evol. Biol. 7, 2 (2007). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Broad G., et al. , The genome sequence of the festoon, Apoda limacodes (Hufnagel, 1766). Wellcome Open Res. 8, 19 (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Tennessen J. A., Molecular evolution of animal antimicrobial peptides: Widespread moderate positive selection. J. Evol. Biol. 18, 1387–1394 (2005). [ DOI ] [ PubMed ] [ Google Scholar ] 51. Osteen J. D., et al. , Subtype-selective spider toxins implicate Na V 1.1 voltage-gated sodium channels in mechanical pain. Nature 534, 494–499 (2016). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Kramer S., et al. , High-content fluorescence bioassay investigates pore formation, ion channel modulation and cell membrane lysis induced by venoms. Toxicon: X 21, 100184 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Walker A. A., et al. , Horizontal gene transfer underlies the painful stings of asp caterpillars (Lepidoptera: Megalopygidae). Proc. Natl. Acad. Sci. U.S.A. 120, e2305871120 (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Cardoso F. C., “High-throughput bioassays applied to phylogenetic and envenomation studies of spider and snake venoms” in Venom Week VIII, Norton R., Tambourgi D., Eds. (Toxicon, Scottsdale, AZ, 2022). [ Google Scholar ] 55. Bulet P., Hetru C., Dimarcq J.-L., Hoffmann D., Antimicrobial peptides in insects; structure and function. Dev. Comp. Immunol. 23, 329–344 (1999). [ DOI ] [ PubMed ] [ Google Scholar ] 56. Yi H.-Y., Chowdhury M., Huang Y.-D., Yu X.-Q., Insect antimicrobial peptides and their applications. Appl. Microbiol. Biotechnol. 98, 5807–5822 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Wu J., et al. , A cecropin-like antimicrobial peptide with anti-inflammatory activity from the black fly salivary glands. Parasit. Vectors 8, 561 (2015). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Akuffo H., Hultmark D., Engstöm A., Frohlich D., Kimbrell D., Drosophila antibacterial protein, cecropin A, differentially affects non-bacterial organisms such as Leishmania in a manner different from other amphipathic peptides. Int. J. Mol. Med. 1, 77–159 (1998). [ DOI ] [ PubMed ] [ Google Scholar ] 59. Blaskovich M. A. T., Zuegg J., Elliott A. G., Cooper M. A., Helping chemists discover new antibiotics. ACS Infect. Dis. 1, 285–287 (2015). [ DOI ] [ PubMed ] [ Google Scholar ] 60. Habermann E., Bee and wasp venoms. Science 177, 314–322 (1972). [ DOI ] [ PubMed ] [ Google Scholar ] 61. Walker A. A., Robinson S. D., Hamilton B. F., Undheim E. A. B., King G. F., Deadly proteomes: A practical guide to proteotranscriptomics of animal venoms. Proteomics 20, 1900324 (2020). [ DOI ] [ PubMed ] [ Google Scholar ] 62. Walker A. A., et al. , Melt with this kiss: Paralyzing and liquefying venom of the assassin bug Pristhesancus plagipennis (Hemiptera: Reduviidae). Mol. Cell. Proteomics 16, 552–566 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 63. Walker A. A., et al. , The assassin bug Pristhesancus plagipennis produces two distinct venoms in separate gland lumens. Nat. Commun. 9, 755 (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 64. Drukewitz S., et al. , A dipteran’s novel sucker punch: Evolution of arthropod atypical venom with a neurotoxic component in robber flies (Asilidae, Diptera). Toxins 10, 29 (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Patocka J., Nepovimova E., Klimova B., Wu Q., Kuca K., Antimicrobial peptides: Amphibian host defense peptides. Curr. Med. Chem. 26, 5924–5946 (2019). [ DOI ] [ PubMed ] [ Google Scholar ] 66. Bals R., Wilson J. M., Cathelicidins—A family of multifunctional antimicrobial peptides. Cell. Mol. Life Sci. 60, 711–720 (2003). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Kawahara A. Y., et al. , Phylogenomics reveals the evolutionary timing and pattern of butterflies and moths. Proc. Natl. Acad. Sci. U.S.A. 116, 22657–22663 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Callaway J. E., et al. , Modification of the C terminus of cecropin is essential for broad-spectrum antimicrobial activity. Antimicrob. Agents Chemother. 37, 1614–1619 (1993). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Tian Y., et al. , Antibacterial mechanism and structure–activity relationships of Bombyx mori cecropin A. Insect Mol. Biol. 33, 708–721 (2024). [ DOI ] [ PubMed ] [ Google Scholar ] 70. Hara S., Taniai K., Kato Y., Yamakawa M., Isolation and α-amidation of the non-amidated form of cecropin D from larvae of Bombyx mori . Comp. Biochem. Physiol. B: Biochem. Mol. Biol. 108, 303–308 (1994). [ Google Scholar ] 71. Vunnam S., Juvvadi P., Merrifield R. B., Synthesis and antibacterial action of cecropin and proline-arginine-rich peptides from pig intestine. J. Pept. Res. 49, 59–66 (1997). [ DOI ] [ PubMed ] [ Google Scholar ] 72. da Silva A. V. R., et al. , The effects of the C-terminal amidation of mastoparans on their biological actions and interactions with membrane-mimetic systems. Biochim. Biophys. Acta Biomembr. 1838, 2357–2368 (2014). [ DOI ] [ PubMed ] [ Google Scholar ] 73. Strandberg E., et al. , Influence of C-terminal amidation on the antimicrobial and hemolytic activities of cationic α-helical peptides. Pure Appl. Chem. 79, 717–728 (2007). [ Google Scholar ] 74. Andreu D., Merrifield R. B., Steiner H., Boman H. G., N-terminal analogs of cecropin A: Synthesis, antibacterial activity, and conformational properties. Biochemistry 24, 1683–1688 (1985). [ DOI ] [ PubMed ] [ Google Scholar ] 75. Robinson S. D., et al. , A comprehensive portrait of the venom of the giant red bull ant, Myrmecia gulosa , reveals a hyperdiverse hymenopteran toxin gene family. Sci. Adv. 4, eaau4640 (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Oñate-Garzón J., et al. , Antimicrobial activity and interactions of cationic peptides derived from Galleria mellonella cecropin D-like peptide with model membranes. J. Antibiot. 70, 238–245 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 77. Park C. B., Yi K.-S., Matsuzaki K., Kim M. S., Kim S. C., Structure–activity analysis of buforin II, a histone H2A-derived antimicrobial peptide: The proline hinge is responsible for the cell-penetrating ability of buforin II. Proc. Natl. Acad. Sci. U.S.A. 97, 8245–8250 (2000). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 78. Sani M.-A., Lee T.-H., Aguilar M.-I., Separovic F., Proline-15 creates an amphipathic wedge in maculatin 1.1 peptides that drives lipid membrane disruption. Biochim. Biophys. Acta Biomembr., 2277–2289 (1848). [ DOI ] [ PubMed ] [ Google Scholar ] 79. Guha S., et al. , Applications and evolution of melittin, the quintessential membrane active peptide. Biochem. Pharmacol. 193, 114769 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Rex S., A Pro→Ala substitution in melittin affects self-association, membrane binding and pore-formation kinetics due to changes in structural and electrostatic properties. Biophys. Chem. 85, 209–228 (2000). [ DOI ] [ PubMed ] [ Google Scholar ] 81. Dempsey C. E., et al. , Contribution of proline-14 to the structure and actions of melittin. FEBS Lett. 281, 240–244 (1991). [ DOI ] [ PubMed ] [ Google Scholar ] 82. Forfar M. J., et al. , Single amino acid changes impact the ability of Drosophila melanogaster cecropins to inhibit growth of Providencia pathogens. ACS Omega 10, 5403–5414 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 83. Shin S. Y., Kang J. H., Hahm K. S., Structure-antibacterial, antitumor and hemolytic activity relationships of cecropin A-magainin 2 and cecropin A-melittin hybrid peptides. J. Pept. Res. 53, 82–90 (1999). [ DOI ] [ PubMed ] [ Google Scholar ] 84. Erviana R., et al. , Novel antimicrobial peptides from a cecropin-like region of Heteroscorpine-1 from Heterometrus laoticus venom with membrane disruption activity. Molecules 26, 5872 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Shin S. Y., et al. , Effects of the hinge region of cecropin A(1–8)–magainin 2(1–12), a synthetic antimicrobial peptide, on liposomes, bacterial and tumor cells. Biochim. Biophys. Acta Biomembr. 1463, 209–218 (2000). [ DOI ] [ PubMed ] [ Google Scholar ] 86. Ryu S., et al. , Antimicrobial and anti-inflammatory effects of cecropin A(1–8)–magainin 2(1–12) hybrid peptide analog P5 against Malassezia furfur infection in human keratinocytes. J. Invest. Dermatol. 131, 1677–1683 (2011). [ DOI ] [ PubMed ] [ Google Scholar ] 87. Wei X.-B., et al. , Novel hybrid peptide cecropin A (1–8)-LL37 (17–30) with potential antibacterial activity. Int. J. Mol. Sci. 17, 983 (2016). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 88. Mukherjee K., et al. , Anti- Listeria activities of Galleria mellonella hemolymph proteins. Appl. Environ. Microbiol. 77, 4237–4240 (2011). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 89. Liu D., et al. , A potential food biopreservative, CecXJ-37N, non-covalently intercalates into the nucleotides of bacterial genomic DNA beyond membrane attack. Food Chem. 217, 576–584 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 90. Gomes C. M., et al. , Differential transcript profile of inhibitors with potential anti-venom role in the liver of juvenile and adult Bothrops jararaca snake. PeerJ 5, e3203 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 91. Moura-da-Silva A. M., et al. , Processing of snake venom metalloproteinases: Generation of toxin diversity and enzyme inactivation. Toxins 8, 183 (2016). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 92. Tarvin R. D., et al. , Interacting amino acid replacements allow poison frogs to evolve epibatidine resistance. Science 357, 1261–1266 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 93. Meinardi E., Florinchristensen M., Paratcha G., Azcurra J. M., Florinchristensen J., The molecular basis of the self/non-self selectivity of a coelenterate toxin. Biochem. Biophys. Res. Commun. 216, 348–354 (1995). [ DOI ] [ PubMed ] [ Google Scholar ] 94. Samakovlis C., Kylsten P., Kimbrell D. A., Engström A., Hultmark D., The andropin gene and its product, a male-specific antibacterial peptide in Drosophila melanogaster . EMBO J. 10, 163–169 (1991). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 95. Quesada H., Ramos-Onsins S. E., Aguadé M., Birth-and-death evolution of the cecropin multigene family in Drosophila . J. Mol. Evol. 60, 1–11 (2005). [ DOI ] [ PubMed ] [ Google Scholar ] 96. Fry B. G., From genome to venome: Molecular origin and evolution of the snake venom proteome inferred from phylogenetic analysis of toxin sequences and related body proteins. Genome Res. 15, 403–420 (2005). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 97. Casewell N. R., et al. , Solenodon genome reveals convergent evolution of venom in eulipotyphlan mammals. Proc. Natl. Acad. Sci. U.S.A. 116, 25745–25755 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 98. Reyes-Velasco J., et al. , Expression of venom gene homologs in diverse python tissues suggests a new model for the evolution of snake venom. Mol. Biol. Evol. 32, 173–183 (2015). [ DOI ] [ PubMed ] [ Google Scholar ] 99. Barua A., Koludarov I., Mikheyev A. S., Co-option of the same ancestral gene family gave rise to mammalian and reptilian toxins. BMC Biol. 19, 268 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 100. Minh B. Q., et al. , IQ-tree 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol. Biol. Evol. 37, 1530–1534 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 101. Yang Z., PAML 4: Phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24, 1586–1591 (2007). [ DOI ] [ PubMed ] [ Google Scholar ] 102. Goudarzi M. H., et al. , Phylogeny, envenomation syndrome, and membrane permeabilising venom produced by Australia’s electric caterpillar Comana monomorpha . Sci. Rep. 14, 14172 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 103. Henriques S. T., et al. , Decoding the membrane activity of the cyclotide Kalata B1: The importance of phosphatidylethanolamine phospholipids and lipid organization on hemolytic and anti-HIV activities. J. Biol. Chem. 286, 24231–24241 (2011). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Appendix 01 (PDF) pnas.2513640122.sapp.pdf (3.8MB, pdf) Dataset S01 (XLSX) pnas.2513640122.sd01.xlsx (70.2KB, xlsx) Dataset S02 (XLSX) pnas.2513640122.sd02.xlsx (24.5KB, xlsx) Data Availability Statement This Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession code JBLKOF000000000 ( 47 ). The version described in this paper is version JBLKOF010000000 . Raw gDNA and RNA-Seq reads generated in this project have been deposited in the Sequence Read Archive with BioProject Accession PRJNA1218917 ( 46 ). Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences ACTIONS View on publisher site PDF (6.6 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• De Novo Assembly and Characterization of Venom Gland Transcriptome for Rhabdophis lateralis.#140735
• Synthetic Snake Venom: The Novel Creation of <em>Crotalid</em> Phospholipase A2 Using Genetic Engineering#982911
• Untargeted metabolomics and proteomics reveals cocoa-mediated mitigation of valproic acid-induced dysregulation in a zebrafish model of autism: pilot study.#297118
• Engineered bacteria in disease diagnosis and therapy: A synthetic biology perspective#304157
• Engineered bacteria in disease diagnosis and therapy: A synthetic biology perspective.#285534