Evolution of chemoreception in relation to land-to-water transition(s) in Adephaga beetles – Evo-AquaSense

Aim 1: Developing the diving beetle Rhantus suturalis (Fig. 2A) as a model aquatic insect species for functional study of chemoreception Establishment of a panel of odorant molecules that may represent RsutOR ligands based on biological relevance, chemical structure and commercial availability; search for novel candidate ligands in natural extracts of mosquito larvae by gas chromatography and HPLC linked to mass spectrometry; functional study of R. suturalis ORs (and possibly some IRs) in Xenopus laevis oocytes using the two-electrode voltage clamp method with candidate odorant molecules; electro-antennogram (EAG) recordings on isolated antennae submitted to odorants, to establish dose-response curves; behavioural assays ; transcriptome sequencing and analysis of larval chemosensory organs to compare the gene repertoires with those expressed in the adult; detailed neuro-anatomical exploration using SEM, TEM and immunofluorescence; gene expression studies by in situ hybridisation for co-receptors of ORs and IRs, and immunofluorescence. Aim 2: Uncovering the diversity of chemosensory genes and organs across terrestrial and aquatic Adephaga Field samplings and micro-dissection of adult chemosensory organs for two phylogenetically distant species per extant Adephaga family (for Dytiscidae, five species from different subfamilies), including key taxa to be collected in Guadeloupe, French Guyana, Madagascar and South Africa; generation of RNAseq data for antennae and palps from these species; transcriptome assembly and annotation of all members of known chemosensory gene families, including candidate receptors (ORs, IRs, GRs), Odorant-Binding Proteins (OBPs) and odorant-degrading enzymes (notably CYP450 and carboxylesterases); read mapping and gene expression level estimates; for comparative morphology: SEM exploration and character formalisation; expression studies in selected species of the main families of the co-receptors of ORs and IRs by in situ hybridisation. Aim 3: Enlightening the evolution of chemosensory genes and organs in Adephaga in relation to ecological transitions Establishment of orthologue clusters conserved among Coleoptera, filtering and trimming, construction of a multigene supermatrix for phylogenetic analyses of Adephaga using Bayesian inference with site-heterogeneous models; study of key fossils using environmental SEM and nano-tomography at the SOLEIL synchrotron followed by 3D reconstruction imaging; character scoring, Ancestral States Reconstructions of the characters, under a model-based probabilistic framework (Mk model); correlations tests through phylogenetic ANOVA; gene tree / species tree reconciliation to infer duplications and losses of chemosensory genes; estimation of dN/dS ratios to identify amino acids evolving under positive selection.

For the part of the project related to the study of the chemosensory system in a selected diving beetle model, Rhantus suturalis, we identified many molecules emitted into the water by the diving beetles and potential prey. These molecules, along with various volatile organic compounds, were used to stimulate the beetle’s antennae and record electro-antennograms, probably a first for an aquatic beetle. Most compounds were detected and about twenty hydrophobic molecules elicited strong antennal responses. To functionally characterise the ORs, we selected 15 candidates ORs based on their expression pattern and phylogenetic position, and active molecules were identified for one of them. Finally, protocols were developed to analyse the behavioural response of diving beetles to olfactory signals. Overall, one major finding is that hydrophobic molecules seem to play a prominent role in the chemical ecology of diving beetles.

In another diving beetle, Cybister lateralimarginalis, we found that larvae express fewer chemosensory genes than adults but larva-specific genes displayed a unique expression pattern, not previously observed in any other holometabolous insect, with 5 larva-specific ORs all having a close paralogue which is adult-specific, 14 larva-specific IRs belonging to a single gene expansion and no larva-specific GR. The expression of a substantial repertoire of ORs specifically in the antennae of the larva further suggests that hydrophobic chemical cues are important for long-range chemodetection in freshwater, contrary to prevailing views.

For the evolutionary analysis of chemosensory genes, a large-scale transcriptomic dataset was generated from antennae and palps of 24 species of terrestrial and aquatic Adephaga, from 6 families. A detailed analysis of these data showed that ORs comprise highly conserved gene lineages as well as a few rapidly-evolving groups that expanded in aquatic families. Such pattern is known to be often adaptive in the context of response to environmental change. Furthermore, unlike land beetles where ORs are expressed mainly in antennae, we found a group of ORs expressed in maxillary palps of the aquatic families Gyrinidae, Hygrobiidae, and Dytiscidae. In R. suturalis, these OR genes are expressed in flat multiporous sensilla on the maxillary palps, which are common in aquatic Adephaga beetles but not in terrestrial ones. This suggests a strong link to the adaptation of chemical senses to aquatic life.

A detailed SEM study of sensilla performed on 23 species from 7 Adephaga families revealed that sensilla structure and distribution provide informative characters at all levels of the phylogeny. We have published the first description of chemosensory sensilla in Hygrobiidae, uncovering some features unique to this relict family as well as synapomorphies shared between Hygrobiidae and Dytiscidae. In R. suturalis, we have performed detailed analyses of the internal structure of palpal sensilla using array tomography.

Evo-AquaSense was an exploratory project, basically starting from a terra incognita, and involving multiple approaches and techniques, on a wide diversity of water beetle taxa. It generated a huge amount of experimental and comparative data, most of which remain at this stage essentially descriptive. Therefore, the opened perspectives for short and longer term research are huge.

With regard to the functional study of the ORs, concerns about poor expression in the heterologous system chosen meant that we were unable to go as far as we had hoped, in particular to link the expression of the ORs in specific chemosensory organs, the evolution of these ORs and their function. An important first step will be to continue this functional study. Functional studies of the chemosensory receptors specifically expressed at the larval stage would be particularly interesting. First, the larvae of diving beetles are strictly and exclusively aquatic, whereas the adults are amphibious. Therefore, larva-specific genes are certainly functional underwater. Second, all larva-specific ORs identified in this project are specialised variants of adult-specific genes, calling for deciphering the functional significance of this specialisation.

The experimental techniques used have allowed us to show that, contrary to what was thought, the hydrophobic molecules commonly used in airborne communication also play an important role in the chemical ecology of diving beetles and other aquatic Adephaga. Following on from this project, it will be necessary to investigate the presence in aquatic environments of signals belonging to very different chemical classes and possibly specific to these environments. Similarly, an important aspect of this future research will be to gain a better understanding of the behaviour of aquatic beetles in response to chemical signals, particularly in the search for prey and sexual partners.

In situ hybridisation study in Rhantus suturalis antenna and palps could be performed for OR and IR co-receptors, two tuning ORs and three OBPs. We expect to produce a more comprehensive dataset of expression patterns, down to the cellular level, for ORs and other chemosensory gene families.

The large-scale comparative investigation of molecular actors of chemodetection in Adephaga beetles could be performed only for Odorant Receptors, in the context of this project. An obvious follow-up will be to study the evolution of other key chemosensory gene families, such as other receptor families (GRs, IRs), OBPs, odor-degrading enzymes…

Evolution of sensilla morphology and distribution in Adephaga beetles is another fertile terrain for follow-ups to Evo-AquaSense. Perspectives include investigations in families that could not be studied in the context of this project (Trachypachidae, Meruidae, Amphizoidae), and lower taxonomic scale comparative studies, especially within Dytiscidae.

Publications international scientific journals:

Montagné N., Jager M., Chertemps T., Persyn E., Jaszczyszyn Y., Meslin C., Jacquin-Joly E. & Manuel M. (2021) The chemosensory transcriptome of a diving beetle. Frontiers in Ecology and Evolution (en révision).

Jager M., Errais W., Trichet M. and Manuel M. (2021) Morphology of antennal and palpal sensilla of Hygrobia hermanni adults and larvae. Insects (soumis).

Oral communications

Michaël Manuel. Evolution of chemoreception and water-to-land transition in insects: focus on diving beetles (Dytiscidae). 10th edition of the EFOR Meeting, 11.05.2021.

Chemical senses play a key role in species adaptation to new ecological niches, notably in insects. Understanding how the fundamentally aerial chemosensory equipment of insects was remodelled in lineages that underwent land-to-freshwater transition is of major interest to evolutionary ecology but remains unexplored. Our project aims to understand how and when chemosensory genes and organs became adapted to life underwater in Adephaga coleopterans, an ideal model to tackle this question as they display various levels of affinity to the aquatic medium, including fully terrestrial, semi-aquatic and highly specialised aquatic insects (the latter being prominent predators of mosquito larvae).
We will combine functional approaches in the diving beetle Rhantus suturalis (family Dytiscidae) with comparative studies of chemosensory genes and structures in a phylogenetic framework at the Adephaga level.
In Task 1, we will identify odour ligands of the ~30 odorant receptors (ORs) that our preliminary RNAseq analyses of R. suturalis adult antennae and palps have identified, by establishing their response spectra to a large panel of odorants including candidate pheromones and molecules separated by chromatography from mosquito larval extracts. This will be complemented by in vivo functional assays (eletro-antennography and behavioural experiments on adult R. suturalis), comparison of the chemosensory gene repertoires of the strictly aquatic larva and of the amphibious adult, and detailed description of the R. suturalis chemosensory neuro-anatomy.
In Task 2, the repertoires of chemosensory genes expressed in antenna and palps will be compared by RNAseq analyses between representatives of the 11 extant families of Adephaga. Furthermore, we will perform comprehensive comparative studies of chemosensory organ structure (sensillum types and distribution) in those families, looking for trends in relation to the more or less pronounced aquatic life style. Finally, OR-expressing sensilla (associated in aerial insects with detection of hydrophobic odorants) and ionotropic receptors(IR)-expressing sensilla (detecting hydrophilic molecules in aerial insect) will be identified in antennae and palps of selected species by in situ hybridisation study of their respective co-receptors.
In Task 3, we will firstly resolve the as yet uncertain phylogenetic relationships between Adephaga families through a phylogenomic approach. In parallel, the morphology of fossils of key extinct families of Mesozoic Adephaga will be revisited in order to re-evaluate their phylogenetic placement, and the paleontological data will also be used to calibrate the molecular tree. We will then perform ancestral character state reconstructions onto the phylogeny, in order to infer the history of land-to-water transition(s), the evolution of chemosensory sensilla distribution and morphology, events of duplications and losses in chemosensory gene families, and signatures of selection in their sequences.
Altogether, the expected results will give an unprecedented picture of the evolutionary changes in chemosensory structures and genes that accompanied land-to-freshwater transition in a lineage of insects, and will allow disentangling in this context the relative contributions of adaptation and exaptation.