bioRxivNeurosciencePreprint
Non-canonical odor coding in the mosquito
Meg A. YoungeriD, Margaret HerreiD, Olivia V. GoldmaniD, Tzu‐Chiao LuiD, Gabriela Caballero-VidaliD, Yanyan QiiD, Zachary N. GilbertiD, Zhongyan GongiD, Takeshi MoritaiD, Saher RahieliD, Majid GhaniniaiD, Rickard IgnelliD, Benjamin John MatthewsiD, Hongjie LiiD, Leslie B. VosshalliD
Abstract
Female Aedes aegypti mosquitoes are a persistent human foe, transmitting arboviruses including dengue and yellow fever when they bite us to obtain a blood meal. Mosquitoes are intensely attracted to human-emitted body odor, heat, and carbon dioxide, which they detect using three different large multi-gene families encoding odor-gated ion channels. Genetic mutations that cause profound disruptions to the olfactory system have modest effects on human attraction, suggesting significant redundancy in odor coding. The canonical view is that olfactory sensory neurons each express a single chemosensory receptor that defines its ligand selectivity. Using immunostaining, RNA in situ hybridization, and single nucleus RNA sequencing, we discovered that Aedes aegypti uses an entirely different organizational principle, with many neurons co-expressing multiple chemosensory receptor genes. In vivo electrophysiology demonstrates that the broad ligand-sensitivity of mosquito olfactory neurons is due to this non-canonical co-expression. The redundancy afforded by an olfactory system in which many neurons co-express multiple receptors with different chemical sensitivity may greatly increase the robustness of the mosquito olfactory system and explain our longstanding inability to engineer new compounds that disrupt the detection of human body odor by mosquitoes.
- organism: Arabidopsis
- assay: snRNA-seq
- assay: imaging
- assay: bulk RNA-seq
- research focus: clustering
- research focus: cell type annotation
- approach: experimental assay
In-depth summary
Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (5 Jan 2022) · CC BY 4.0
Aedes aegypti olfactory neurons often co-express OR, IR and sometimes GR receptors; in the maxillary palp, one neuron type responds to odorants via both OR and IR pathways.
Aedes aegypti expresses at least twice as many chemosensory receptors as it has antennal lobe glomeruli, which suggests an organization different from the one-receptor-to-one-neuron-to-one-glomerulus rule described for other species. The paper tests receptor co-expression directly with knock-in reporter genetics, in situ hybridization, immunostaining, snRNA-seq and in vivo recordings.
Key results
SuggestedAedes aegypti expresses many more chemosensory receptors than it has antennal lobe glomeruli (at least twice as many), suggesting its olfactory organization must differ from the canonical one-receptor-per-glomerulus scheme.
Glomeruli: 64.9 ± 0.9 (mean±SEM, range 60-72 per lobe, 12 female brains). Genome: 117 OR, 135 IR and 72 GR genes. Reanalysed RNA-seq: 102, 134, 156 and 178 receptors expressed at 5, 2, 1 and 0.5 TPM. · Mismatch in chemosensory receptor and olfactory…
ShownOrco and Ir25a are widely co-expressed in single olfactory neurons of the antenna, maxillary palp and proboscis, in both sexes. Far fewer neurons co-express Orco with Ir8a or Ir76b.
Orco-DBD plus Ir25a-AD Split-QF2 labelled nearly half of glomeruli; neither half alone labelled any (n=3 per genotype). Ir25a-QF2 neurons innervated 89.9 ± 1.4% of glomeruli (n=3). In situ: almost half of Orco cells co-express Ir25a and vice versa; few co-express Ir76b. · Co-expression of Orco and Ir25a in the mosquito…; Co-expression of Orco and Ir25a in the mosquito…; Coordinated co-expression of chemosensory…; Additional technical notes on expression and…
ShownsnRNA-seq of the female antenna shows many receptor combinations that clustering hides. A single ligand-selective receptor marks several distinct cell types, often mixing OR and IR genes.
14,161 nuclei from two batches, 6,645 classed as neurons, in 54 neuron clusters. Using a normalized expression threshold of 0.5, Or82 marked at least 6 cell types and Ir64a at least 6, each with different extra receptors. · Single nucleus RNA sequencing reveals that many…
ShownThe maxillary palp is more complex than the A/B/C neuron model. Ir25a is in all Orco and all Gr3 cells, and several IRs and ORs are co-expressed with Or8 or Or49.
In situ (n=5 palps): no Orco-Gr3 overlap; Or8 and Or49 in separate populations; Ir93a and Ir100a in subsets of Or49 cells. snRNA-seq: 2,298 nuclei, 630 neurons, 4 neuron classes. Only snRNA-seq showed low Orco in the Gr3 cluster. · Coordinated co-expression of chemosensory…; Single nucleus RNA sequencing of maxillary palp…
SuggestedMaxillary palp B neurons respond to 1-octen-3-ol through the OR pathway and to two amines partly through the IR pathway, consistent with OR and IR function in the same neuron.
The 1-octen-3-ol response was lost in Orco mutants only. Amine responses persisted in Gr3 and Orco mutants. In wild type (LVP), all 17 neurons responded to triethyl amine and 16 of 17 to hexyl amine. In Ir25a mutants, 78.3% of 23 did not respond, but 5 of 23 responded more strongly than wild type. · Receptor co-expression expands the functional…
Methods, data and reproducibility
Built CRISPR knock-in QF2 and Split-QF2 driver lines for chemosensory co-receptor genes. Combined antennal lobe imaging, immunostaining, in situ hybridization, snRNA-seq of antenna and maxillary palp, and single sensillum recordings in wild-type and receptor-mutant females.
- Antennal lobe glomerulus counts from brains stained for synaptic neuropil · Aedes aegypti · n = 12 female brains (left hemisphere)
- Female antenna snRNA-seq (10x Genomics), two batches · Aedes aegypti · n = 14,161 nuclei after filtering; 6,645 neurons · GEO: GSE192978
- Female maxillary palp snRNA-seq (10x Genomics) · Aedes aegypti · n = 2,298 nuclei; 630 neurons · GEO: GSE192978
- Reanalysis of published antenna and maxillary palp RNA-seq (TPM, AaegL5 annotation) · Aedes aegypti
- Antennal whole-mount immunostaining and RNA in situ cell counts · Aedes aegypti · n = n=7 antennal segments (immunostaining); n=4 (in situ)
- Maxillary palp in situ hybridization cell counts · Aedes aegypti · n = n=5 maxillary palps
- Single sensillum recordings, maxillary palp, wild type and Gr3, Orco and Ir25a mutants · Aedes aegypti · n = n=4-16 per condition; 17 wild-type and 23 Ir25a mutant neurons
- Methods & tools
- CRISPR-Cas9 knock-in, Q-system and Split-QF2, HCR RNA in situ hybridization, single sensillum recording, Cell Ranger 6.0.2, DecontX 1.8.1, DoubletFinder 2.0.3, Seurat 4.0.5, DropletUtils 1.12.3, ImageJ/FIJI, Imaris, ComplexHeatmap
- Data & accessions
- https://zenodo.org/record/5818543#.YdX7VWjMIuUhttps://zenodo.org/record/5818952#.YdX7KWjMIuUhttps://github.com/VosshallLab/AGWG-AaegL5
- Reagents & environment
- Cell Ranger 6.0.2, DecontX 1.8.1, DropletUtils 1.12.3, Seurat 4.0.5, DoubletFinder 2.0.3; genome AaegL5.0. Donor plasmids at Addgene #162520-162526.
- Reproducibility gaps
- Raw data are in Data File 1 (supplement). Availability of the mosquito strains is not stated.
Limitations (4)
- Co-convergence of distinct neuron types onto shared glomeruli is likely but untested. Split-QF2 results do not rule it out, and variable glomerular anatomy prevents identifying most glomeruli across animals. · stated by the authors
- snRNA-seq detected low Orco expression in the Gr3 neuron cluster. This conflicts with the other three methods and is not understood by the authors. · stated by the authors
- Single-nucleus co-expression calls depend on normalized-expression thresholds. Transcripts detected in one nucleus do not show that OR and IR complexes assemble and function in that neuron. · noted in this summary
- Functional evidence covers only the maxillary palp A and B cells. The suggestion that two B-neuron types explain the residual amine responses in Ir25a mutants is a hypothesis. No behavioural test of redundancy or robustness is presented. · noted in this summary
Read this if you work on insect olfaction or chemosensory receptor expression and want snRNA-seq plus genetic and electrophysiological evidence against the one-receptor-per-neuron model in a mosquito.
- Citations
- 43
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Olfactory receptor coexpression and co-option in the dengue mosquito
Elisha David Adavi, Vitor L. dos Anjos, … Carolyn S. McBride·Nature CommunicationsPeer-reviewed·Aug 15, 2026
- The co-receptors Orco and Ir8a are required for coordinated expression of chemosensory genes in the antennae of the yellow fever mosquito, Aedes aegypti.
Matthew M. Cooke, Michael Solomon Chembars, Ronald Jason Pitts·bioRxiv·Apr 23, 2025
- Synergistic Coding of Carbon Dioxide and a Human Sweat Odorant in the Mosquito Brain
Shruti Shankar, Genevieve M. Tauxe, … Conor J. McMeniman·bioRxiv·Nov 2, 2020
- Chemical signatures of human odour generate a unique neural code in the brain of Aedes aegypti mosquitoes
Zhilei Zhao, Jessica L. Zung, … Carolyn S. McBride·bioRxiv·Nov 1, 2020
- When non-canonical olfaction is optimal
Caitlin Lienkaemper, Meg A. Younger, Gabriel Koch Ocker·Proceedings of the National Academy of SciencesPeer-reviewed·Oct 7, 2025
- Cross-modal sensory compensation increases mosquito attraction to humans
Takeshi Morita, Nia G. Lyn, … Leslie B. Vosshall·Science AdvancesPeer-reviewed·Jan 1, 2025