Comments
Research Report: A genetic basis for potential nociception in cochineal bugs (Dactylopius coccus; Hemiptera: Dactylopiidae)
8 min read
Prepared by: Claudinéia Pereira Costa, PhD; Meghan Barrett, PhD
Note: This post is very similar to our own prior EA Forum report on the genetic basis for nociception in another hemipteran, the lac bug (Kerria lacca, Hemiptera: Kerridae). In several places, we directly use the same text as we wrote in the first report to save time and ensure all limitations/caveats of the project are clearly, consistently reported. Consider this an attribution statement to our prior work for any directly copied text found within this post. All results for D. coccus are new and the methodology was slightly altered due to the genomic/proteomic resources available for this species.
Note: This report has not been peer-reviewed and has only been generated for the Forum. Results should be interpreted with caution, though we do follow peer-reviewed methodologies.
Cochineal bugs (Dactylopius coccus; Hemiptera: Dactylopiidae) are wild-harvested or farmed and killed as part of the production of carmine dye, which is used in food, pharmaceuticals, fashion, and cosmetics. A report by Rethink Priorities estimates 4.6 to 22 trillion cochineal bug deaths annually (Rowe 2020). This wide range largely relates to poor reporting of the number of animals used and significant uncertainties about how many animals die as nymphs (juveniles).
Cochineal bugs belong to the Hemiptera - the order of ‘true bugs’. Despite the importance of this group, they have been so poorly studied for nociception, neurobiology, and cognition that they were not included among the major orders reviewed in the Gibbons et al. (2022) paper on pain in insects. (However, we are in the midst of a review of the evidence in Hemiptera now and will make a summary of it available on the Forum once it is complete and the paper has been preprinted.)
Adult female cochineal bugs, like lac bugs (also hemipterans), are sessile. As we describe in greater detail in the introduction of the lac bug post, this reduced agency coupled with differences in hemipteran neuroanatomy compared to other orders of insects (Strausfeld et al. 2009) could make nociception less adaptive. Therefore, it makes sense to investigate if the capacity for nociception is present in these groups, despite a strong background assumption that nociception is likely present across all the insects (Gibbons et al. 2022). While data to support the capacity for nociception must go beyond genetic evidence to include morphological and functional evidence, here we aim to assess the publicly-available genome for late-stage adult female D. coccus for genetic evidence that the receptor proteins associated with nociception in other species are present in cochineal bugs.
Generally speaking, we repeated the process performed in Goldberg et al. (2024) for other farmed insects. These genes include (text quoted from the methods of Goldberg):
However, we also added a search for three other receptors/gene families that our further research highlighted may be of interest. This included:
This is the first, simplest check for evidence of nociception. Below, we present our findings; methods are briefly provided for those interested at the end. Importantly, finding evidence of any of these gene families does not prove nociception is present (much less sentience!) in D. coccus, as genes may have evolved to serve different functions, these genes may not be expressed in nociceptors, or nociceptive neurons may be absent. Further, finding evidence that all of these gene families are absent does not prove nociception is absent (much less sentience!) in D. coccus, as different insect taxa have evolved novel genes that have nociceptive functionality (e.g., HsTRPA in Hymenoptera, to replace the lost ancient TRPA1 channel; Matsuura et al. 2009, Kohno et al. 2010, Wang et al. 2018).
This analysis was an initial in silico screening step, and results presented here should thus be considered provisional. Targeted sequencing and functional assays are strongly recommended to confirm the gene presence and genomic structure as well as the biological roles.
That said, apparently intact candidate genes were identified across eight of the ten target gene families (using the criteria described in our methodology section at the end). There was strong support for a single copy of pain, NompC , trpm, Stj, and TRPA1, as well as two copies of Piezo (strong canonical channel and a divergent, second Piezo-family member), two copies of DEG/ENaC channels in the ppk/rpk/ppk26 gene family, and three copies of Pyrexia. This suggests the genetic architecture that supports heat, chemical, and mechanical nociception in other insect species has strong support in D. coccus.
While an orthologous locus was identified for TRPA5 in the D. coccus genome, a stop codon and several frameshifts suggest it may be nonfunctional; we have thus labeled it as a ‘disrupted’ locus.
No support was found for copies of Pkd2, similar to lac bugs but not some other hemipterans (see Costa and Barrett, 2026 and Goldberg et al. 2024). Pkd2 is used in fruit fly larval class III multidendritic neurons (which respond to both gentle touch and noxious cold) to directly sense noxious cold temperatures (Turner et al. 2017). NompC and trpm are also used for modulating responses (perhaps, calcium homeostasis) to noxious cold in CIII neurons in fruit flies (Turner et al. 2017). As we describe in greater detail in our post on lac bugs, Pkd2 is the ‘most absent’ gene family in prior searches for nociceptive ion channels in insects (Goldberg et al. 2024); the other relevant genes for cold sensing were present in the cochineal genome (NompC and trpm); and the lack of Pkd2 may not mean a lack of noxious cold-sensing in cochineal bugs.
Thus, at this time, we can infer with reasonably high confidence that D. coccus does not use the Pkd2 mechanism for noxious cold sensing - but we cannot infer that cold nociception is thereby necessarily absent in the species.
Ultimately, further research confirming the nociceptive functions of the proteins encoded by these genes, as well as their expression in nociceptors, is needed to confirm nociception in D. coccus. Therefore, follow-up morphological, physiological, and functional studies would be required to conclusively demonstrate nociception in the species. For more on this, see the final results section of the post on lac bugs.
Obtaining the Genome
We obtained the genome of D. coccus here (and genome link here). The authors state the genome was obtained from the whole body of an adult female.
Gene Family Analysis
The genome assembly of D. coccus (GCA_977009245.1) was used to identify nociception-related ion channel and auxiliary-subunit gene families. Candidate genes were identified directly from the genomic sequence because public gene annotation and a predicted proteome were not available. Genome quality was first evaluated using BUSCO v6.1.0 with the Hemiptera-specific hemiptera_odb12 lineage dataset (Manni et al., 2021), which indicated high assembly completeness (94.2%). We followed the comparative approach used for K. lacca described by Goldberg et al. (2024) alongside a few new targets. Ten nociception-associated gene family targets were examined: Painless, NompC, TRPM, Piezo, Pkd2, ppk/rpk/ppk26, TRPA1, TRPA5, Pyrexia, and Straightjacket.
Validated insect protein sequences were used as queries in the TBLASTN (Altschul et al., 1990). We searched the complete D. coccus genome sequence with protein sequences from Drosophila melanogaster (Diptera: Drosophilidae) and multiple hemipteran species, including K. lacca, Planococcus citri (Hemiptera: Pseudococcidae), Acyrthosiphon pisum (Hemiptera: Aphididae), Bemisia tabaci (Hemiptera: Aleyrodidae), and Rhodnius prolixus (Hemiptera: Reduviidae), and additional taxa were incorporated for gene families requiring phylogenetic resolution. Candidate genomic regions were evaluated using E-value, alignment length and coverage, consistency among reference proteins, and the organization of neighboring TBLASTN hits. Protein-to-genome alignments were then reconstructed with Miniprot (Li, 2023) to evaluate exon structure and coding-sequence continuity. Predicted proteins were inspected for expected length and for obvious disruptions such as frameshifts or internal stop codons. We did not use a fixed percentage identity cutoff; instead, we interpreted sequence identity together with alignment length, coverage, E-value, and conserved domain presence/absence. Genes for which no clear evidence was found were reported as not detected.
Additional phylogenetic analyses were performed for gene families in which pairwise similarity alone was insufficient for confident classification, particularly the ppk-related DEG/ENaC genes, TRPA5, Pyrexia/WTRW-related sequences, and Straightjacket-related α2δ paralogs. Candidate and reference protein sequences were aligned using MAFFT (Katoh and Standley, 2013), and their evolutionary relationships were analyzed using maximum-likelihood phylogenetics with IQ-TREE (Minh et al., 2020). Candidates were classified according to their combined genomic and phylogenetic support rather than annotation name alone.
Code Availability
Code has been made available here on the Barrett lab’s GitHub. To reduce errors in software implementation and improve code quality, we used a LLM only for code review and debugging. All code logic was written by humans, with AI suggestions manually checked and tested to ensure accuracy and reproducibility. All text in this document, and the interpretation of findings from the in silico screen, were written entirely by humans.
Thank you to Macroscopic Ventures and The Navigation Fund. Their funding made Dr. Costa’s position in the Barrett lab possible and thus supported the research that generated this post. The funders had no control over experimental design, methodology, interpretation of results, or decision to publish this work. This post should not be taken to reflect the opinions, values, or beliefs of the funders.
