Introduction

Interoception is a major path through which the nervous system assays internal states (Wyart et al., 2023; Verdonk et al., 2025). The cerebrospinal fluid (CSF) is a complex solution in which the central nervous system baths. The CSF is produced by the choroid plexus and changes composition as a function of age, time of the day, and physiological conditions (Parnetti et al., 2019; Fagan et al., 2021; Hermann et al., 2021) so that spinal taps are often used to find out about the nature of an infection or a neurological disease. Cells in contact with the cerebrospinal fluid include ependymal cells, radial glia, tanycytes, secretory cells from the choroid plexi, the subcommissural organ or the pineal gland as well as ciliated neurons. In the last decade, a sensory pathway involving ciliated neurons detecting changes of the cerebrospinal fluid was discovered as important to adjust posture during challenging behaviors, locomotor speed, as well as morphogenesis and even innate immunity (Wyart et al., 2023). A century ago, Kolmer and Agduhr described ciliated neurons at the level of the central canal in the spinal cord of over a hundred vertebrate species that contact the cerebrospinal fluid (CSF) (Kolmer, 1921; Agduhr, 1922; Dale et al., 1987). These CSF-contacting neurons (CSF-cNs) are dense GABAergic neurons in the spinal cord of macaques, mice, turtles or zebrafish (Djenoune and Wyart, 2017).

During development, CSF-cNs are organised in two rows originating from different progenitor domains, pMN and p3 in zebrafish (Shin et al., 2007; Yang et al., 2010; Huang et al., 2012; Yang et al., 2020) versus p3 and p2 in mice (Petracca et al., 2016; Di Bella et al., 2019). Early during embryonic development, a bidirectional flow of CSF in the central canal can enable the effective transport of large particles (Thouvenin et al., 2020). Later in the central canal of mature animals, the CSF typically flows from anterior to posterior and its flow can be enhanced by muscle contractions (Thouvenin et al., 2020). By interacting with the Reissner fiber (Reissner, 1860), a polymer under tension in the CSF, CSF-cNs detect spinal compression (Bohm et al., 2016; Jalalvand et al., 2016b; Orts-Del’Immagine et al., 2020). This process requires the channel TRPP2 or PKD2L1 that is a highly specific marker of CSF-cNs across species (Sternberg et al., 2018; Djenoune et al., 2014; Petracca et al., 2016; Gerstmann et al., 2022; Nakamura et al., 2023). In return, CSF-cNs acutely modulate locomotion, its speed and fine postural control (Wyart et al., 2009; Fidelin et al., 2015; Hubbard et al., 2016; Wu et al., 2021). On longer time scales, these cells, together with the Reissner fiber, control morphogenesis (Sternberg et al., 2018; Cantaut-Belarif et al., 2018; Troutwine et al., 2020; Rose et al., 2020; Marie-Hardy et al., 2023) via release of peptides from the Urotensin 2 family (Zhang et al., 2018; Bearce et al., 2022; Gaillard et al., 2023). In addition to urotensin related peptides, CSF-cNs are GABAergic neurons strikingly synthesizing numerous monoamines and peptides that differ between species: somatostatin in fish and lamprey, dopamine in lamprey, serotonin transiently in fish (Djenoune et al., 2017), trace amines in rodents (Hökfelt et al., 1973; Jaeger et al., 1983; Ren et al., 2017), urotensin related peptides in fish (Quan et al., 2015) among many others (Djenoune and Wyart, 2017).

The chemosensory functions of CSF-cNs are however the first ones that have been investigated. Over 20 years ago, Stoeckel et al. showed that these cells express specific P2X receptors for ATP (Stoeckel et al., 2003). Zuker’s team found that CSF-cNs firing is modulated by pH (Huang et al., 2006), which is later confirmed that the activity of CSF-cN is strongly modulated by changes in pH and osmolarity in mice and lamprey (Orts-Del’Immagine et al., 2012; Jalalvand et al., 2016a). CSF-cNs express numerous chemoreceptors and respond to bacterial metabolites upon invasion of the CSF by pathogenic bacteria triggering meningitis (Prendergast et al., 2023). Recent evidence for expression of opioid receptors in CSF-cNs indicate a role in spinal cord injury (Yue et al., 2024). Yet, the extent of their chemosensory abilities, i.e. which receptors CSF-cNs express and molecules they can detect in the CSF, is not fully understood.

To identify new pathways enabling long range signaling in the CSF via CSF-contacting neurons, we investigated here in the CSF-cN transcriptome in larval zebrafish which novel putative chemoreceptors may be expressed and enriched in these cells. To confirm expression in CSF-cNs and characterize the spatial expression pattern, we used hybridization chain reaction (HCR) on 2-3 days post fertilization (dpf) larvae and found that both dorsolateral and ventral CSF-cNs expressed the low-density lipoprotein (LDL) receptor 2 ldlrad2, glutamate receptor 2 grm2a and ptprna. In contrast, somatostatin receptor 2 sstr2a was specifically expressed in the ventral population while somatostatin itself is only present in the dorsal population. Interestingly, multiple of these receptors were also found in other cells contacting the CSF such as ependymal radial glia (ERGs) or floor plate neuroepithelial cells (FP). Altogether, our results indicate that numerous chemosensory signaling pathways can enable long distance communication at the CSF interface.

Results

Selection of receptors combining enrichment in CSF-cNs and high expression

To identify novel chemoreceptors in CSF-cNs, we investigated the transcriptome of these cells performed out of 5 replicates run after sorting for GFP+ cells in the 3-day old guillotined double transgenic Tg(pkd2l1:GAL4;UAS:GFP) larvae (Prendergast et al., 2023). We selected receptors who were enriched (Log Fold Change > 1.15) and had an absolute number of reads in the GFP+ CSF-cNs population above 10 Fragments Per Kilobase per Million mapped fragments (FPKM) (Table 1). Four receptors caught our attention: the low-density lipoprotein (LDL) receptor 2 (ldlrad2), the somatostatin receptor 2a (sstr2a), the glutamate metabotropic receptor 2 (grm2a) and the ptprn receptor (ptprna).

Table 1.

Somatostatin receptor sstr2a is expressed in ventral CSF-cNs and unknown dorsal spinal cells in zebrafish larvae

To confirm the expression of sstr2a in CSF-cNs, we performed hybridization chain reaction (HCR) for sstr2a and pkd2l1 transcripts on 3-day post fertilization (dpf) AB mifta ^-/- larvae (n = 9 fish, Figure 1). We used DAPI for delaminating the ventral and dorsal boundaries of the spinal cord and highlighting the central canal (Figure 1A1). The pkd2l1 HCR probe nicely labels as shown before (Djenoune et al., 2014) both dorsolateral (triangle) and ventral (arrowhead) CSF-cNs (Figure 1A2, A2i). We observed all along the spinal cord co-expression of the receptor sstr2a with pkd2l1+ cells located ventrally from the central canal, indicating that ventral CSF-cNs (arrowhead, Figure 1A3, A3i, A4, A4i, C, D) but not dorsolateral CSF-cNs (triangle, Figure 1B3, B3i, B4, B4i, C, D) express the sstr2a receptor. In addition, we noticed high expression of sstr2a is also found in unknown dorsal most cells on the lateral edge of the spinal cord (asterisk, Figure 1B3, B4, C). The total number of pkd2l1⁺ CSF-cNs was similar between ventral (18 ± 2.0 per 100 µm) and dorsolateral (18 ± 1.2 per 100 µm) CSF-cN populations. Quantification confirmed that sstr2a expression was largely confined to ventral CSF-cNs: on average, 12 ± 1.2 ventral pkd2l1⁺ CSF-cNs per 100 µm expressed sstr2a, compared to only 1 ± 0.1 dorsolateral cell. These results demonstrate that sstr2a expression is largely restricted to ventral CSF-cNs among the CSF-cN population.

Figure 1.

Glutamate metabotropic receptor grm2a is expressed in ventral and dorsolateral CSF-cNs and additional dorsal spinal cells

To confirm the expression of grm2a in CSF-contacting neurons (CSF-cNs), we performed HCR for grm2a and pkd2l1 transcripts on 2 dpf AB mifta ^-/- larvae (n = 6 fish; Figure 2). DAPI staining was used to delineate the ventral and dorsal boundaries of the spinal cord and to highlight the central canal (Figure 2A1). The pkd2l1 HCR probes labeled both dorsolateral (triangle) and ventral (arrowhead) CSF-cNs (Figure 2A2, B2). Along the entire spinal cord, we observed co-expression of grm2a with all pkd2l1⁺ cells, indicating that both ventral (arrowhead; Figure 2A3, A4, A4i, B3, B4, B4i) and dorsolateral (triangle) CSF-cNs express the grm2a receptor. In addition, a few grm2a⁺ cells not expressing pkd2l1 near the central canal were labeled (asterisk; Figure 2B4, B4i), as well as unidentified dorsalmost located spinal cells (asterisk; Figure 2A3, A4, B1). The total number of pkd2l1⁺ CSF-cNs was (19 ± 1.3 per 100 µm) for ventral and (14 ± 2.5 per 100 µm) for dorsolateral populations. Quantification confirmed that grm2a was expressed in both ventral and dorsolateral CSF-cNs (on average, 15 ± 1.7 per 100 µm for ventral CSF-cNs and 10 ± 2.8 per 100 µm for dorsolateral pkd2l1⁺ CSF-cNs per 100 µm were grm2a⁺).

Figure 2.

The phosphatase receptor ptprna is expressed predominantly in both ventral and dorsolateral CSF-cNs in 3 day old larval zebrafish

To confirm the expression of ptprna in CSF-cNs, we performed HCR for ptprna and pkd2l1 transcripts on 3 dpf AB mifta ^-/- larvae (n = 5 fish, Figure 3). DAPI was used to delaminate the ventral and dorsal boundaries of the spinal cord and highlight the central canal (Figure 3A1). The pkd2l1 HCR probe labels both dorsolateral (triangle) and ventral (arrowhead) CSF-cNs (Figure 3A2, B2). We observed all along the spinal cord co-expression of the receptor ptprna and pkd2l1 in both ventral CSF-cNs (arrowhead, Figure 3A3, A4, A4i, B3, B4, B4i)) and dorsolateral CSF-cNs (triangle, Figure 3A3, A4, A4i, B3, B4, B4i). In addition, we noticed ptprna expression in unknown spinal cells ventral to the central canal (asterisk, Figure 3A3, A4, B1).

Figure 3.

Ldlrad2 is expressed in ventral and dorsolateral CSF-cNs and neighboring cells surrounding the central canal

To confirm the expression of ldlrad2 in CSF-cNs, we performed HCR for ldlrad2 and pkd2l1 transcripts on 3 dpf AB mifta^-/- larvae (n = 6 fish, Figure 4). DAPI was added to delaminate the ventral and dorsal boundaries of the spinal cord and highlight the central canal (Figure 4A1). The pkd2l1 receptor is expressed in both dorsolateral (triangle) and ventral (arrowhead) CSF-cNs (Figure 4A2). We observed all along the spinal cord co-expression of the receptor ldlrad2 with pkd2l1 (Figure 4A3, A4), indicating that dorsolateral (triangle, Figure 4A2ii-A4ii) and ventral (arrowhead) CSF-cNs (Figure 4A2i-A4i) express the ldlrad2 receptor. In addition, we noticed that numerous other cells in contact with the cerebrospinal fluid (asterisk, Figure 4A2i-A4i; 4A2ii-A4ii) presumably ependymal radial glia cells (ERGs) (Jalalvand et al., 2014; Becker and Becker, 2022) also expressed the ldlrad2 receptor (Figure 4A3, A4, 4A2ii-A4ii, asterisk).

Figure 4.

Discussion

In this study, we first confirmed expression of four chemoreceptors in CSF-cNs previously identified in the transcriptome from bulk population of CSF-cNs sorted by fluorescence (Prendergast et al., 2023): the metabotropic somatostatin receptor sstr2a, the metabotropic glutamate receptor grm2a, the receptor ptprna and the LDL receptor 2 ldlrad2.

Receptor expression to specific cells contacting the CSF

While the ldlrad2, grm2a and ptprna receptors are present in all CSF-cNs, we find interesting specialization with specific receptors found in subtypes of CSF-cNs: the somatostatin receptor 2a is mainly expressed in ventral spinal CSF-cNs. This is particularly interesting as somatostatin 1 sst1.1 is solely expressed by dorsal CSF-cNs: somatostatin1 could therefore coordinate the activity between dorsolateral and ventral CSF-cNs via the metabotropic somatostatin receptor 2a that acts as an inhibitor of secretion onto ventral CSF-cNs via the G protein Gαi (Rodrigues et al., 2018). The somatostatin receptor sstr2a was also expressed in dorsal spinal cord where it could modulate sensory processing such as itching in rodents (Flauaus et al., 2022) and chronic pain perception as shown previously via interaction with corticostatin (Morell et al., 2014) in addition to its role in psychiatric and neurodegenerative diseases (Beneyto et al., 2012; Ádori et al., 2015).

Furthermore, for most receptors investigated, sparse expression is also present in unknown spinal cells. For the LDL receptor 2 ldlrad2, the expression is also high in other cells surrounding the central canal that possibly correspond to ependymal radial glia (Becker and Becker 2014), which bear a motile cilium in the central canal (Bellegarda et al., 2023). Interestingly, the Reissner fiber in the central canal also exhibits multiple ligand-binding region of the LDL receptor family (Gobron et al., 1996; Sepúlveda et al., 2021).

Note that our list of receptors is of course not exhaustive. We selected some of the receptors that were only highly expressed and enriched in CSF-cNs versus all other cells of the trunk (Log Fold Change > 1.1). We could therefore have missed numerous receptors not enriched in CSF-cNs, highly expressed in other cells of the trunk (such as muscles, notochord, gut), or just expressed at low levels in CSF-cNs.

Physiological context: when does the CSF change composition and could modulate these receptors?

In fish, lamprey and mouse, CSF-cNs have been involved in diverse physiological functions in the context of motor control, posture and locomotion (Wyart et al., 2009; Hubbard et al., 2016; Bohm et al., 2016; Quan et al., 2020; Wu et al., 2021; Gerstmann et al., 2022; Nakamura et al., 2023; Jalalvand et al., 2016a; Jalalvand et al., 2016b). In addition, this interoceptive system has been also involved in morphogenesis to strengthen the body axis and spine (Sternberg et al., 2018; Cantaut-Belarif et al., 2018; Zhang et al., 2018; Bearce et al., 2022; Gaillard et al., 2023), detection of pathogen invasion to enhance host defence (Prendergast et al., 2023) as well as modulation of the stem cells niche after spinal cord injury (Wyart et al., 2023; Kathe et al., 2022; Yue et al. 2024).

Our data here suggest that somatostatin may act as a signaling molecule from dorsolateral to ventral CSF-cNs. Dorsolateral CSF-cNs specifically express somatostatin (Djenoune et al., 2017). During locomotion, the tail bends from side to side, which activates ipsilateral dorsolateral CSF-cNs that respond to left / right curvature of the trunk (Böhm et al., 2016), which can dampen the oscillations and reduce bout duration (Quan et al., 2020). We show here that ventral CSF-cNs express the somatostatin receptors sstr2a. Therefore, upon activation of dorsolateral CSF-cNs during locomotion, somatostatin could act to reduce intracellular calcium in ventral CSF-cNs, and thereby decreasing the release of numerous other peptides and secreted proteins such as urp1, urp2, msmp2, or nppc (Quan et al., 2015; Prendergast et al., 2023).

CSF lipoproteins transport lipids and associated proteins that contribute to lipid homeostasis in the central nervous system and support neural development (Tsujita et al., 2024; Merrill et al. 2023). This pathway could be relevant to morphogenesis regulated by CSF-cNs and the Reissner fiber (Wyart et al., 2023). The Reissner fiber is indeed formed by the glycoprotein SCO-spondin, which contains low-density lipoprotein receptor type A (LDLrA) domains (Meiniel and Meiniel, 2007). In vitro experiments have shown that LDL can bind SCO-spondin and modulate its neurogenic activity (Vera et al., 2015). In addition, LDL-family particles can carry signaling molecules, including morphogens from the Hedgehog (Hh) and Wnt families (Panáková et al., 2005; Willnow et al., 2007). SCO-spondin has been proposed to facilitate morphogen distribution in the CSF by interacting with lipoproteins (Vera et al., 2015). LDL receptor 2a (ldlr2a) appears to be expressed by cells in contact with the cerebrospinal fluid, including CSF-cNs, and most likely ependymal radial glia and floor plate cells. These cells could therefore use LDLR2a to capture LDL/associated cargo (lipids and/or morphogens) from the CSF, potentially facilitated by interactions with the Reissner fiber during morphogenesis.

During spinal cord injury or when pathogens invade the CSF during an infection of the central nervous system, glutamate can be released from the cytoplasm following cell death. In human bacterial meningitis, elevated CSF glutamate levels have been associated with clinical outcomes (Spranger et al., 1996). One potential pathway is that excess glutamate is detected by CSF-cNs via activation of this Gi/o-coupled grm2a receptor. Receptor activation inhibits adenylyl cyclase (AC), reduces intracellular cAMP, and can suppress voltage-gated calcium channel (VGCC) activity, thereby decreasing Ca²⁺ influx (Reiner and Levitz, 2018). Such a reduction in Ca²⁺ entry could be neuroprotective by limiting Ca²⁺-dependent cellular damage under inflammatory conditions. Consistent with reduced excitability, intraventricular injection of Streptococcus pneumoniae in zebrafish silenced spontaneous global Ca²⁺ activity in CSF-cNs (Prendergast et al., 2023). In the context of spinal cord injury, glutamate may be sensed by CSF-cNs via the metabotropic glutamate receptor Grm2a, which can promote GABA release (Corti et al., 2007) and modulate the stem cell niche (New et al., 2023). Because infection and injury converge on excitotoxic and inflammatory cascades, Grm2a-dependent CSF-cN signaling may couple acute changes in excitability to longer-term repair programs, consistent with regenerative responses reported in mice (Kathe et al., 2022; Yue et al., 2024).

Ptprn is localized to dense-core secretory vesicles in neuroendocrine cells and is involved in secretory granule biogenesis and regulation of secretion (Cai et al., 2011). CSF-cNs are secretory cells containing dense-core vesicles (Djenoune et al., 2017) and express numerous genes encoding neuropeptides and other secreted factors implicated in locomotion, morphogenesis and innate immunity (Prendergast et al., 2023; Wyart et al., 2023). The expression of ptprna in CSF-cNs therefore suggests that this receptor contributes to the regulation of secretory machinery and may participate in controlling the release of bioactive compounds into the cerebrospinal fluid.

Altogether, our study opens new paths for investigation of chemosensory signaling in the spinal cord and at the interface with the cerebrospinal fluid that can carry long range signaling between the nervous system and other systems and organs in the body.

Materials and Methods

Zebrafish housing, mutant and strain

Animal handling and procedures were validated by the Paris Brain Institute (ICM) and the French National Ethics Committee (Comité National de Réflexion Éthique sur l’Expérimentation Animale; APAFIS no. 2018071217081175 and APAFIS no. #38209-2022080517223947 and APAFIS no. #51743-2024102515029150) in agreement with European Union legislation. To avoid pigmentation, all experiments were performed on Danio rerio larvae of AB strain carrying the mitfa^−/− mutation. Adult zebrafish were reared at a maximum density of six animals per liter in a 14/10 h light–dark cycle environment at 28.5 °C. Larval zebrafish were typically raised in Petri dishes filled with system water under the same conditions in terms of temperature and lighting as for adults.

Fluorescent in situ hybridization using high chain reaction (HCR)

HCR probes were custom-designed and synthesized by Molecular Instruments (Los Angeles, CA, USA) based on NCBI mRNA reference sequences: pkd2l1 (XM_690312), sstr2a (XM_005170121.5), ldlrad2 (XM_003199510.6), ptprna (XM_017357652.3), and grm2a boosted version (XM_005166117.4). HCR reagents, including hairpins and buffers, were obtained from the same supplier. Double-staining HCR was performed for all experiments using a probe for pkd2l1 and a probe for the receptor of interest. The HCR hairpins were labeled with Alexa Fluor 488, Alexa Fluor 546, or Alexa Fluor 647. The protocol of « whole-mount zebrafish embryos and larvae » used for HCR was modified based on Shainer et al., Science Advances 2022. Briefly, embryos/larvae (1–10 per mL) were fixed in 1 mL freshly prepared 4% paraformaldehyde (PFA) in Dulbecco’s phosphate-buffered saline (DPBS) overnight at 4 °C with gentle shaking to preserve tissue integrity. Fixed samples were washed 3 × 5 min in DPBST (DPBS + 0.1% Tween 20) and permeabilized for 10 min in prechilled 100% methanol at −20 °C. For rehydration, embryos/larvae were washed for 5 min in 50% methanol/50% DPBST, 5 min in 25% methanol/75% DPBST, and 5 × 5 min in DPBST. To optimize the probe permeabilization, 2-dpf embryos were treated with 10 µg/mL Proteinase K in 1× DPBST for 20 min and 3-dpf larvae for 30 min at room temperature, followed by post-fixation in 4% PFA for 20 min. To increase the signaling for weakly expressed metabotropic receptor genes, the concentration of the probe was prepared for 32 or 64nM in final solutions (8 or 16 μl of 1 μM probe stock). The standard concentration 16nM was used for pkd2l1 (4 μl of 1 μM probe stock). After prehybridization in 250 μL prewarmed hybridization buffer for 30 min at 37 °C, embryos/larvae in probe solution were incubated 12–16 h at 37 °C. Excess probes were removed by four 30 min washes in 250 μL prewarmed probe wash buffer at 37 °C, followed by two 20 min washes in 5× SSCT (sodium chloride sodium citrate + 0.1% Tween 20) at room temperature. Embryos/larvae were preamplified in 250 μL amplification buffer for 30 min at room temperature. Hairpins h1 and h2 (60 pmol each) were prepared by snap-cooling 10 μL of 3 μM stock at 95 °C for 90 s, then cooling in the dark for 30 min. For double labeling, 5 μL of each hairpin and corresponding amplifiers were included. Hairpins were combined in a 250 μL amplification buffer with DAPI, and embryos/larvae were incubated in this solution overnight (12– 16 h) in the dark at room temperature. Following removal of the amplification buffer, embryos/larvae were incubated in hairpin solution containing DAPI (1:1000) for 12–16 h in the dark at room temperature. The next day, excess hairpins were removed by 3 × 20 min washes in 500 μL 5× SSCT at room temperature. Samples were mounted in ibidi Mounting Medium (Ref: 50001, ibidi GmbH, Germany) for imaging.

Confocal Imaging

Imaging was performed on a Leica SP8 DLS (inverted) or SP8 X White Light Laser confocal microscope (Leica Microsystems, Wetzlar, Germany) equipped with a 40 X oil-immersion objective (NA = 1.3). Laser lines used were 405, 488, 552, and 638 nm (SP8 DLS) or 405 nm and a white laser (470–670 nm; SP8 X), applied in sequential scanning mode to avoid spectral bleed-through. Emission windows were 415–465 nm (DAPI), 498–520 nm (488 nm excitation), 568– 600 nm (546 nm excitation), and ≥657 nm (647 nm excitation). Image acquisition was performed with LAS X software (Leica Microsystems) and processed in Fiji (https://imagej.net/software/fiji/downloads, Schindelin et al., 2012).

Image Quantification

We quantified all cells expressing the receptors and identified whether they were ventral or dorsolateral CSF-cNs, other CSF-contacting cells (floor plate, roof plate, ependymal radial glia) or other unknown cells in the dorsal spinal cord. Three segments, each comprising four somites in the rostral, middle, and caudal regions of the trunk, were imaged within a single larva. For each segment, Z-stacks were acquired from 30–40 μm-thick sections using a step size of 1 μm, resulting in 30–40 slices with a field of view measuring 291 × 145 μm. The pkd2l1+ cells were distinguished in two groups: (i) Ventral CSF-cNs: a ventral row of cells adjacent to the central canal (ii) dorsolateral CSF-cNs: a dorsal row of cells adjacent to the central canal. The receptor-expressing cells were counted in ventral and dorsolateral CSF-cNs and in other pkd2l1-cells. The proportion of receptor-expressing cells among pkd2l1+ cells was assessed separately for ventral and dorsolateral CSF-cNs. The proportion of receptor-expressing cells among other cell types was calculated separately for the dorsal-to-central canal region and the P3 domain. Cells were counted per 100 µm for each location along the rostrocaudal axis (rostral, middle, and caudal) within a single larva. For each receptor, three double-labeled larvae were analyzed.

Statistics

All values provided in the text are given as mean +/- standard deviation.

Data availability

All cell counts are included in the Supplementary table 2.

Acknowledgements

We would like to thank for their inputs and camaraderie spirit the Wyart lab “SIBBIL” (https://wyartlab.org; https://parisbraininstitute.org/paris-brain-institute-research-teams/sibbil-navigation-sensorimotor-integration-brain-body-integration-lab) for providing great insight along the way, particularly Dr. Giulia Messa, Dr. Emma Partiot, Dr. Clothilde Colart for constructive discussions that helped shape the manuscript. We thank A. Arneau, N. Jezequel, C. Lejeune, S. Nunes Figueiredo, B. Daboval of the core facility Pheno-Zfish for fish care. We thank the 2020 Fondation Bettencourt-Schueller (FBS-don-0031) award (“Identity and organization or neuronal networks controlling exploration”), a New York Stem Cell Foundation (NYSCF) Robertson Award 2016 research grant (NYSCF-R-NI39), the 2020 Prize Equipe ‘Fondation pour la Recherche Médicale’ (FRM-EQU202003010612) ‘Neuronal circuits underlying navigation: from genes to behavioral models’, a 2020 European Research Council Consolidator grant no. 101002870 (2021– 2026) ‘Exploratome: Circuit mechanisms underlying sensory-evoked navigation’ and a National Institutes of Health grant no. 1U19NS104653-01 awarded to C.W., the European Union’s Horizon 2020 research and innovation programme under a Marie Skłodowska-Curie grant no. 813457 awarded to C.W. The project benefited as well from the support of the Agence Nationale pour la Recherche (ANR) ANR-22-CE37-0023 named LOCOCONNECT, ANR-23-CE16-0017-02 named RocSMAP, ANR-24-CE16-7992 named CIRCOLOCO, ANR-21-CE14-0042 named MOTOMYO and ANR-21-CE13-0008 named ASCENTS.

Additional information

Author Contributions

Conceptualization: CW & FQ; HCR execution and optimization: EV, LM, LT; Quantification: EV; Analysis: EV; Data visualization: EV; Writing of initial draft: CW, FQ, EV; Supervision: CW & FQ; Funding: CW.

Funding

EC | European Research Council (ERC)

https://doi.org/10.3030/101002870

• Claire Wyart

• Louise Moizan

• Emily Verran

• Loeva Tocquer

Agence Nationale de la Recherche (ANR) (ANR-24-CE16-7992 CIRCOLOCO)

• Claire Wyart

Agence Nationale de la Recherche (ANR) (ANR-23-CE16-0017-02 RocSMAP)

• Claire Wyart

Agence Nationale de la Recherche (ANR) (ANR-22-CE37-0023 LOCOCONNECT)

• Claire Wyart

Agence Nationale de la Recherche (ANR) (ANR-21-CE14-0042 MOTOMYO)

• Claire Wyart

Richard Mille Fund (DBS: targeting resilient motor circuits in Parkinson's disease)

• Claire Wyart

• Fenq Q Quan

Agence Nationale de la Recherche (ANR) (ANR-21-CE13-0008 ASCENTS)

• Claire Wyart

Additional files

Supplementary file 1.

Supplementary file 2.