Advertisement
Current Biology
This journal offers authors two options (open access or subscription) to publish research

Blind cavefish retain functional connectivity in the tectum despite loss of retinal input

Open AccessPublished:August 03, 2022DOI:https://doi.org/10.1016/j.cub.2022.07.015

      Highlights

      • Cavefish retain functional connectivity in their visual center despite the loss of eyes
      • Inhibitory connectivity is largely lost, whereas excitatory connectivity is retained
      • Shared genetic architecture underlies multiple types of neurons in the tectum

      Summary

      Sensory systems display remarkable plasticity and are under strong evolutionary selection. The Mexican cavefish, Astyanax mexicanus, consists of eyed river-dwelling surface populations and multiple independent cave populations that have converged on eye loss, providing the opportunity to examine the evolution of sensory circuits in response to environmental perturbation. Functional analysis across multiple transgenic populations expressing GCaMP6s showed that functional connectivity of the optic tectum largely did not differ between populations, except for the selective loss of negatively correlated activity within the cavefish tectum, suggesting positively correlated neural activity is resistant to an evolved loss of input from the retina. Furthermore, analysis of surface-cave hybrid fish reveals that changes in the tectum are genetically distinct from those encoding eye loss. Together, these findings uncover the independent evolution of multiple components of the visual system and establish the use of functional imaging in A. mexicanus to study neural circuit evolution.

      Graphical abstract

      Keywords

      Introduction

      Sensory systems are highly tuned through evolution to an animal’s environment and under stringent evolutionary pressure.
      • Schlosser G.
      A short history of nearly every sense—the evolutionary history of vertebrate sensory cell types.
      ,
      • Wylie D.R.
      • Gutiérrez-Ibáñez C.
      • Iwaniuk A.N.
      Integrating brain, behavior, and phylogeny to understand the evolution of sensory systems in birds.
      Adaptation to novel environments is frequently associated with modification of sensory-processing governing vital behaviors, including foraging and avoidance of predation.
      • Lloyd E.
      • Olive C.
      • Stahl B.A.
      • Jaggard J.B.
      • Amaral P.
      • Duboué E.R.
      • Keene A.C.
      Evolutionary shift towards lateral line dependent prey capture behavior in the blind Mexican cavefish.
      ,
      • Paz A.
      • McDole B.
      • Kowalko J.E.
      • Duboue E.R.
      • Keene A.C.
      Evolution of the acoustic startle response of Mexican cavefish.
      The reliance on different sensory cues has been widely documented to be accompanied by changes in the anatomy of the primary sense organ, such as a tradeoff between expansion of the eye or olfactory epithelium or changes in cortical area devoted to processing sensory information.
      • Hinaux H.
      • Devos L.
      • Blin M.
      • Elipot Y.
      • Bibliowicz J.
      • Alié A.
      • Rétaux S.
      Sensory evolution in blind cavefish is driven by early embryonic events during gastrulation and neurulation.
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      • Gross J.B.
      The complex origin of Astyanax cavefish.
      • Bronchti G.
      • Schönenberger N.
      • Welker E.
      • Van der Loos H.
      Barrelfield expansion after neonatal eye removal in mice.
      • Touj S.
      • Gallino D.
      • Chakravarty M.M.
      • Bronchti G.
      • Piché M.
      Structural brain plasticity induced by early blindness.
      However, little is known about the evolution of neural circuitry in downstream sensory-processing centers.
      The Mexican tetra, Astyanax mexicanus, provides a powerful biological system to study the genetic and neural basis of sensory evolution, consisting of eyed surface populations that inhabit rivers in Northeast Mexico and southern Texas, and at least 30 cavefish populations.
      • Gross J.B.
      The complex origin of Astyanax cavefish.
      ,
      • Jeffery W.R.
      Regressive evolution in Astyanax cavefish.
      Cavefish populations have evolved a series of morphological, physiological, and behavioral changes including eye loss, albinism, reduced sleep, loss of social behaviors, and expansion of non-visual sensory organs, including olfaction, taste, and mechanosensation.
      • Kowalko J.
      Utilizing the blind cavefish Astyanax mexicanus to understand the genetic basis of behavioral evolution.
      ,
      • Yoshizawa M.
      Behaviors of cavefish offer insight into developmental evolution.
      Many of these phenotypes and behaviors have evolved repeatedly in geographically isolated cavefish populations, positioning A. mexicanus as a model to investigate the genetic and ecological factors that shape the evolution of traits.
      • Chin J.S.R.
      • Gassant C.E.
      • Amaral P.M.
      • Lloyd E.
      • Stahl B.A.
      • Jaggard J.B.
      • Keene A.C.
      • Duboue E.R.
      Convergence on reduced stress behavior in the Mexican blind cavefish.
      • Kowalko J.E.
      • Rohner N.
      • Linden T.A.
      • Rompani S.B.
      • Warren W.C.
      • Borowsky R.
      • Tabin C.J.
      • Jeffery W.R.
      • Yoshizawa M.
      Convergence in feeding posture occurs through different genetic loci in independently evolved cave populations of Astyanax mexicanus.
      • Duboué E.R.
      • Keene A.C.
      • Borowsky R.L.
      Evolutionary convergence on sleep loss in cavefish populations.
      One of the most striking traits is convergence on eye loss, presumably to conserve energy in a perpetually dark environment.
      • Jeffery W.R.
      Adaptive evolution of eye degeneration in the Mexican blind cavefish.
      • Krishnan J.
      • Rohner N.
      Cavefish and the basis for eye loss.
      • McGaugh S.E.
      • Gross J.B.
      • Aken B.
      • Blin M.
      • Borowsky R.
      • Chalopin D.
      • Hinaux H.
      • Jeffery W.R.
      • Keene A.
      • Ma L.
      • et al.
      The cavefish genome reveals candidate genes for eye loss.
      This phenotype has been studied for over 80 years; however, the concomitant physiological changes in visual-processing centers have not been investigated.
      In teleosts, the majority of retinal inputs project to the optic tectum, a brain region that is homologous to the mammalian superior colliculus.
      • Krauzlis R.J.
      • Lovejoy L.P.
      • Zénon A.
      Superior colliculus and visual spatial attention.
      Although the tectum primarily receives sensory inputs from the retina, minor inputs from auditory and lateral line modalities have been reported.
      • Thompson A.W.
      • Vanwalleghem G.C.
      • Heap L.A.
      • Scott E.K.
      Functional profiles of visual-auditory-and water flow-responsive neurons in the zebrafish tectum.
      In sighted species, visual processing within the tectum is essential for generating appropriate visuomotor behaviors, including prey capture and predator avoidance.
      • Bianco I.H.
      • Engert F.
      Visuomotor transformations underlying hunting behavior in zebrafish.
      • Förster D.
      • Helmbrecht T.O.
      • Mearns D.S.
      • Jordan L.
      • Mokayes N.
      • Baier H.
      Retinotectal circuitry of larval zebrafish is adapted to detection and pursuit of prey.
      • Barker A.J.
      • Baier H.
      Sensorimotor decision making in the zebrafish tectum.
      In comparison to surface fish, the size of the tectum is reduced in cavefish from multiple populations.
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      This may be indicative of co-evolution of reduced inputs of the retina and its target centers within the brain. Alternatively, this may be an epiphenomena in which degradation of the retina and the optic tectum occurred independently of one another. Recently, we generated a quantitative neuroanatomical atlas of surface fish and multiple cavefish populations,revealing that the optic tectum is reduced by ∼20% across populations.
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      ,
      • Loomis C.
      • Peuß R.
      • Jaggard J.B.
      • Wang Y.
      • McKinney S.A.
      • Raftopoulos S.C.
      • Raftopoulos A.
      • Whu D.
      • Green M.
      • McGaugh S.E.
      • et al.
      An adult brain atlas reveals broad neuroanatomical changes in independently evolved populations of Mexican cavefish.
      Although this represents a significant reduction in neural tissue, the presence of the remaining 80% raises the possibility that this system still contributes to brain computations, given the energetic cost of neural tissue.
      • Herculano-Houzel S.
      Scaling of brain metabolism with a fixed energy budget per neuron: implications for neuronal activity, plasticity and evolution.
      ,
      • Hyder F.
      • Rothman D.L.
      • Bennett M.R.
      Cortical energy demands of signaling and nonsignaling components in brain are conserved across mammalian species and activity levels.
      Therefore, the maintenance of the tectum despite the apparent loss of visual input provides a system to examine how brain structures are impacted by evolved differences in sensory input.
      The optic tectum of the zebrafish, Danio rerio, is an established model to study development and function of visual sensory systems.
      • Bollmann J.H.
      The zebrafish visual system: from circuits to behavior.
      ,
      • Niell C.M.
      • Smith S.J.
      Functional imaging reveals rapid development of visual response properties in the zebrafish tectum.
      Two-photon Ca2+ imaging has been used to identify cell types that respond to different properties of visual stimuli and infer the circuit's functional connectivity.
      • Gabriel J.P.
      • Trivedi C.A.
      • Maurer C.M.
      • Ryu S.
      • Bollmann J.H.
      Layer-specific targeting of direction-selective neurons in the zebrafish optic tectum.
      • Preuss S.J.
      • Trivedi C.A.
      • Vom Berg-Maurer C.M.
      • Ryu S.
      • Bollmann J.H.
      Classification of object size in retinotectal microcircuits.
      • Romano S.A.
      • Pietri T.
      • Pérez-Schuster V.
      • Jouary A.
      • Haudrechy M.
      • Sumbre G.
      Spontaneous neuronal network dynamics reveal circuit’s functional adaptations for behavior.
      • Pietri T.
      • Romano S.A.
      • Pérez-Schuster V.
      • Boulanger-Weill J.
      • Candat V.
      • Sumbre G.
      The emergence of the spatial structure of tectal spontaneous activity is independent of visual inputs.
      Recently, we have implemented transgenesis in A. mexicanus, permitting the application of approaches previously limited to zebrafish.
      • Stahl B.A.
      • Peuß R.
      • McDole B.
      • Kenzior A.
      • Jaggard J.B.
      • Gaudenz K.
      • Krishnan J.
      • McGaugh S.E.
      • Duboue E.R.
      • Keene A.C.
      • Rohner N.
      Stable transgenesis in Astyanax mexicanus using the Tol2 transposase system.
      The development of these tools has provided the first opportunity to investigate how neural responsiveness and functional connectivity of sensory circuits are shaped by evolutionarily derived changes in sensory function. Here, we used two-photon scanning microscopy to image surface and cavefish pan-neuronally expressing GCaMP to directly define differences in neuronal dynamics within the tectum across independently evolved populations of A. mexicanus.

      Results

      Development of transgenic cavefish for brain imaging

      Multiple populations of cavefish have converged on reduced eye size and volume of the optic tectum;
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      ,
      • Jeffery W.R.
      Adaptive evolution of eye degeneration in the Mexican blind cavefish.
      ,
      • Loomis C.
      • Peuß R.
      • Jaggard J.B.
      • Wang Y.
      • McKinney S.A.
      • Raftopoulos S.C.
      • Raftopoulos A.
      • Whu D.
      • Green M.
      • McGaugh S.E.
      • et al.
      An adult brain atlas reveals broad neuroanatomical changes in independently evolved populations of Mexican cavefish.
      ,
      • Wilkens H.
      Regressive evolution: ontogeny and genetics of cavefish eye rudimentation.
      however, little is known about the underlying impacts on neuronal function. To investigate the changes in function of the optic tectum associated with eye loss, we sought to compare the functional connectivity and the responsiveness to light in the tectum of river-dwelling surface fish and two independently evolved populations of cavefish (Figure 1A). The Pachón population originates from old surface stock and is located in the Sierra del Abra region, whereas the Molino population evolved from new surface stock and is located in the Sierra de Guatemala range.
      • Bradic M.
      • Beerli P.
      • García-de León F.J.
      • Esquivel-Bobadilla S.
      • Borowsky R.L.
      Gene flow and population structure in the Mexican blind cavefish complex (Astyanax mexicanus).
      • Herman A.
      • Brandvain Y.
      • Weagley J.
      • Jeffery W.R.
      • Keene A.C.
      • Kono T.J.Y.
      • Bilandžija H.
      • Borowsky R.
      • Espinasa L.
      • O’Quin K.
      • et al.
      The role of gene flow in rapid and repeated evolution of cave-related traits in Mexican tetra, Astyanax mexicanus.
      • Mitchell R.W.
      • Russell W.H.
      • Elliott W.R.
      Geological and genomic evidence suggest that these two populations of cavefish have independently evolved loss of eyes.
      • Herman A.
      • Brandvain Y.
      • Weagley J.
      • Jeffery W.R.
      • Keene A.C.
      • Kono T.J.Y.
      • Bilandžija H.
      • Borowsky R.
      • Espinasa L.
      • O’Quin K.
      • et al.
      The role of gene flow in rapid and repeated evolution of cave-related traits in Mexican tetra, Astyanax mexicanus.
      • Mitchell R.W.
      • Russell W.H.
      • Elliott W.R.
      • Borowsky R.
      Restoring sight in blind cavefish.
      To study changes in the function of the optic tectum in A. mexicanus, we generated surface and cave populations harboring the genetically encoded Ca2+ indicator (GCaMP6s) of the pan-neuronal promoter elavl3/HuC.
      • Freeman J.
      • Vladimirov N.
      • Kawashima T.
      • Mu Y.
      • Sofroniew N.J.
      • Bennett D.V.
      • Rosen J.
      • Yang C.T.
      • Looger L.L.
      • Ahrens M.B.
      Mapping brain activity at scale with cluster computing.
      The surface and Molino populations have previously been described;
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      here, we also utilized a HuC:GCaMP6s Pachón line, allowing for direct comparison of brain function in independently evolved cavefish populations. Fish from the surface, Pachón, and Molino populations expressed the reporter as early as 1.5 days post-fertilization (dpf) through 8.5 dpf (Figures 1B and S1A–S1C), confirming stable, brain-wide expression throughout early development across all three A. mexicanus populations. Therefore, these lines recapitulate expression in zebrafish and can be applied to measure neuronal activity throughout the brain.
      Figure thumbnail gr1
      Figure 1Generation of surface and cave fish transgenic lines expressing pan-neuronal HuC:GCaMP6s
      (A) Left: a map of the Sierra del Abra region, identifying the location of the Pachón and Molino caves. Right: representative images of adult fish from the surface environment (top) and from the Molino (middle) and Pachón (bottom) caves. Scale bars, 0.5 cm.
      (B) Representative images of 6 dpf surface, Molino, and Pachón larvae, demonstrating pan-neuronal expression (top, bright-field image; bottom, fluorescent image). Scale bars, 0.5 mm.
      (C) Quantification of number of neurons in the tectum at the level of imaging, showing a reduction in tectal neuronal number in Molino and Pachón cavefish with respect to surface fish. Surface, 607.5 ± 126.0 neurons; Molino, 466.6 ± 85.5 neurons; Pachón, 491.7 ± 74.51 neurons, p = 0.0006; ordinary one-way ANOVA followed by post hoc Tukey test. Blue, surface, n = 12; orange, Molino, n = 20; green, Pachón, n = 17 (∗∗p < 0.01, ∗∗∗p < 0.001).
      See also .

      Ongoing spontaneous activity in the optic tectum

      Correlation in the spontaneous activity between neurons has been used as a proxy of the circuit’s functional connectivity. In zebrafish, the ongoing spontaneous activity of the optic tectum is organized in assemblies composed of highly correlated neurons with similar spatial tuning curves. These neuronal assemblies show attractor-like dynamics and may improve the detection of prey-like visual stimuli.
      • Romano S.A.
      • Pietri T.
      • Pérez-Schuster V.
      • Jouary A.
      • Haudrechy M.
      • Sumbre G.
      Spontaneous neuronal network dynamics reveal circuit’s functional adaptations for behavior.
      ,
      • Pietri T.
      • Romano S.A.
      • Pérez-Schuster V.
      • Boulanger-Weill J.
      • Candat V.
      • Sumbre G.
      The emergence of the spatial structure of tectal spontaneous activity is independent of visual inputs.
      To quantify changes in the tectal functional circuitry that co-evolve with eye degeneration, we compared spontaneous activity in the tectum between surface and cave populations. For this purpose, we performed two-photon Ca2+ imaging of surface and cave transgenic GCaMP6s larvae (6 dpf) for a period of 30 min while limiting as much as possible the presence of sensory stimuli (Figures S1D–S1G). Quantification of the number of neurons in the imaged region of the tectum by automated segmentation showed a reduction of ∼20% in the cavefish population, consistent with previous reports of reduced cavefish tectum size (Figure 1C).
      • Jaggard J.B.
      • Lloyd E.
      • Yuiska A.
      • Patch A.
      • Fily Y.
      • Kowalko J.E.
      • Appelbaum L.
      • Duboue E.R.
      • Keene A.C.
      Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
      We then analyzed the resulting fluorescence traces to detect significant Ca2+ transients as previously described
      • Romano S.A.
      • Pérez-Schuster V.
      • Jouary A.
      • Boulanger-Weill J.
      • Candeo A.
      • Pietri T.
      • Sumbre G.
      An integrated calcium imaging processing toolbox for the analysis of neuronal population dynamics.
      (Figures 2A–3C ). We first quantified the average frequency and duration of individual Ca2+ events across the tectal circuit. No differences were detected between surface fish and Pachón or Molino cavefish (Figure S2). Therefore, the overall levels of spontaneous activity within the tectum do not show evolutionary changes in two independently evolved cave populations. These findings reveal that the overall levels of spontaneous neuronal activity in the tectum are unaffected by evolutionary eye loss.
      Figure thumbnail gr2
      Figure 2The structure of spontaneous activity is conserved between surface and cavefish
      (A–C) Top: examples of raster plots of spontaneous neuronal activity recorded during 30 min in single surface (A), Molino (B), and Pachón (C) larvae, at a frame rate of 3.96 Hz. Color bar: relative change of fluorescence ΔF/F. The raster plots were sorted using the k-means algorithms with 10 clusters and Euclidean distance as a measure of similarity between neurons. Bottom: proportion of significant Ca2+ events.
      (D–F) Example correlation matrices in single larvae corresponding to the raster plots in (A)–(C). Blue: negative correlations, red: positive correlations. The correlation matrices were not sorted according to the k-means clustering solution used in (A)–(C).
      (G–I) Probability density distribution of pairwise spontaneous neuronal activity correlations across larvae in surface (G, blue curve, n = 12), Molino (H, orange curve, n = 20), and Pachón (I, green curve, n = 17). In black is the average of the same distributions after circularly shifting each neuron’s Ca2+ activity time series by a random lag (between 0 and 30 min). This ensures that the temporal structure of each time series is preserved, while pairwise correlations are destroyed (null model). Shaded area: mean ± SD.
      (J and K) For each larva, we measured the probability density function of pairwise spontaneous correlations and quantified the difference with the null model by computing the Jensen-Shannon divergence (JSD) between the two distributions for positive (J) (surface, 0.035 ± 0.018 JSD; Molino, 0.036 ± 0.021 JSD; Pachón, 0.037 ± 0.030 JSD, p = 0.84; Kruskal-Wallis non-parametric ANOVA) and negative (K) (surface, 0.0062 ± 0.0073 JSD; Molino, 0.001 ± 0.0020 JSD; Pachón, 0.002 ± 0.0047 JSD, p = 0.052; Kruskal-Wallis non-parametric ANOVA) correlations. Statistical significance was assessed at the group level using the non-parametric Kruskal-Wallis one-way ANOVA (p = 0.841 for positive correlations, p = 0.0525 for negative correlations), followed by post hoc testing using the Tukey-Kramer procedure between each pair (p < 0.05; ns, not significant). Blue, surface, n = 12; orange, Molino, n = 20; green, Pachón, n = 17.
      (L) Average distribution of pairwise correlations against distance across larvae. For each pair of neurons, we measured the pairwise correlation coefficient as well as the physical distance in the optical section between those two neurons in micrometers. We averaged the pairwise correlations within distance bins across larvae in surface (blue, n = 12), Molino (orange, n = 20), and Pachón (green, n = 17). Shaded gray area: mean ± SD.
      See also .
      Figure thumbnail gr3
      Figure 3Partial loss of neuronal response to visual stimuli in cavefish populations
      (A) Representative traces of the five types of light response.
      (B–D) Left: raster plots of normalized neuronal activity over the whole population of surface (B, n = 6), Molino (C, n = 5), and Pachón (D, n = 3), averaged around the onset of whole-field visual stimulation. Visual stimuli were delivered by an array of white LEDs for a duration of 30 s (represented by horizontal light blue bar above the raster), with an inter-stimulus interval of 120 s, and neuronal activity was recorded with a frame rate of 1.96 Hz. Each experiment contains between 8 and 20 visual stimuli. Neurons responsive to the visual stimuli were detected using a linear regression model and classified into 5 categories: “On” cells (blue), “Off” cells (turquoise), “On/Off” cells (green), “sustained” cells (yellow), and “inhibited” cells (orange). Right: ΔF/F averaged across neurons and stimuli repetitions for each response type (surface, 593 neurons; Molino, 170 neurons; Pachón, 70 neurons). Average slope of the response onset for surface, 0.106 ± 0.008 ΔF/F·s−1; Molino, 0.009 ± 0.005 ΔF/F·s−1; Pachón, 0.007 ± 0.002 ΔF/F·s−1; R-squared surface, 0.826 ± 0.012; Molino, 0.853 ± 0.077; Pachón, 0.356 ± 0.197, p = 0.008 Kruskal-Wallis. Shaded gray area: mean ± SD.
      (E) Proportion of neurons responding to the whole-field visual stimulus in one hemi-tectum, averaged per neuron. Surface, 19.55% ± 6.26%; Molino, 6.10% ± 3.78%; Pachón, 4.20% ± 3.10%, mean ± SD, p = 0.0107, Kruskal-Wallis non-parametric ANOVA. Statistical significance was assessed at the group level using the non-parametric Kruskal-Wallis one-way ANOVA (p = 0.0107), followed by post hoc testing using the Tukey-Kramer procedure between each pair (p < 0.05; ns, not significant).
      (F) Distribution of the peak values for the amplitude of the neuronal response (ΔF/F) to whole-field visual stimuli, averaged per neuron (surface, 0.491 ± 0.432 ΔF/F; Molino, 0.212 ± 0.192 ΔF/F; Pachón, 0.241 ± 0.212 ΔF/F, mean ± SD, p < 10−4 Kruskal-Wallis non-parametric ANOVA). Kruskal-Wallis one-way ANOVA (p < 10−4 for positive correlations, p = 0.0525 for negative correlations), Tukey-Kramer post hoc test (∗∗∗p < 0.001; ns, not significant).
      See also and .
      To determine whether correlated activity represents statistically significant interactions that are indicative of neural circuit connectivity in the optic tectum, we applied Pearson’s pairwise correlations between the spontaneous activity of the tectal neurons (Figures 2D–2F). The probability distribution of the correlations in spontaneous activity was compared with a null model that randomly shifted the time series of every neuron (the null model represents the correlations that can be explained by chance). To quantify the difference between the two distributions of the pairwise correlations (the correlations in the data that can not be explained by chance), we used the Jensen-Shannon divergence (JSD). JSD measures the distance between two probability distributions (STAR Methods; Figures 2G–2I). Briefly, JSD allows assessing the similarity between two distributions: a low JSD value (close to 0) indicates that the distributions are very similar, whereas large JSD values indicate that the distributions are different. By comparing the JSD value of the dataset and the null model, we can assess the level of significant correlations (those that can not be explained by chance). Thus, we computed for each larva of the Surface, Molino, and Pachón populations, the distribution of pairwise correlations as well as the distribution obtained from the null model, and computed the JSD value between these two. Using this approach, we first evaluated the functional excitatory connectivity by assessing only the “positive” pairwise correlations. We found no significant differences between the tectal neurons of surface fish, or Pachón and Molino cavefish (Figure 2J). These findings suggest that functional tectal connectivity was retained in cavefish despite eye loss. However, analysis of the “negative” correlations, which can reveal the functional inhibitory connectivity, showed a significant reduction in the Molino cavefish, with Pachón approaching significance (Figure 2K). Given the importance of the inhibitory/excitatory balance to visual processing, the reduction in negative correlations in cavefish may reflect a physiological correlate of relaxed selection on the visual circuit in blind cavefish that is driven by a lack of retinal input. Finally, analysis of the physical distance between correlated neurons revealed that both surface and cave populations exhibit spatially structured correlations, with an exponential decrease in correlations with the Euclidean distance between neurons (Figure 2L). Therefore, positive correlations are resistant to the evolved loss of visual function, suggesting that excitatory and inhibitory connectivity are under partially independent evolutionary selection.

      Visual responses in the optic tectum

      In zebrafish and other teleosts, the optic tectum is the main visual-processing center.
      • Bollmann J.H.
      The zebrafish visual system: from circuits to behavior.
      ,
      • Basso M.A.
      • May P.J.
      Circuits for action and cognition: a view from the superior colliculus.
      ,
      • Helmbrecht T.O.
      • dal Maschio M.
      • Donovan J.C.
      • Koutsouli S.
      • Baier H.
      Topography of a visuomotor transformation.
      Cavefish retain a diminished eye, and they behaviorally respond to some light cues; however, the eye is suggested to be non-functional as cavefish exhibit severely reduced retinotectal connections by adulthood.
      • Soares D.
      • Yamamoto Y.
      • Strickler A.G.
      • Jeffery W.R.
      The lens has a specific influence on optic nerve and tectum development in the blind cavefish Astyanax.
      To examine whether cavefish retain a light-evoked response in the optic tectum, we used two-photon microscopy to image light-evoked activity in transgenic GCaMP6s surface and cavefish (6 dpf). Briefly, fish were immobilized in low-melting point agarose and imaged during visual stimulus presentation, which consisted of a white-light LED located in front of the larvae. The LED was turned on for 30 s and presented approximately ten times, separated by 60 s (STAR Methods; Figure S3). Using a multivariate linear regression model, we categorized the Ca2+ responses of individual tectal neurons during stimulus presentation and identified neurons with multiple distinct response profiles (Figure 3A). In surface fish, the most common were cells that responded to either the onset (“On” cells) or offset (“Off” cells) of light, with another subset that responded to both the onset and offset of light (“On/Off” cells). In addition, we identified a small number of neurons that exhibited a sustained activation throughout the period of the light stimulus (“sustained” cells) and another that showed a sustained decrease in activity during light stimulus (“inhibited” cells). Therefore, the tectal light response in A. mexicanus surface fish contains cell types previously observed in zebrafish.
      • Robles E.
      • Filosa A.
      • Baier H.
      Precise lamination of retinal axons generates multiple parallel input pathways in the tectum.
      ,
      • Robles E.
      • Fields N.P.
      • Baier H.
      The zebrafish visual system transmits dimming information via multiple segregated pathways.
      Although cavefish have previously been reported to be blind with the exception of response to looming stimulus and lack physiological response to light in the tectum,
      • Voneida T.J.
      • Fish S.E.
      Central nervous system changes related to the reduction of visual input in a naturally blind fish (Astyanax hubbsi ).
      ,
      • Yoshizawa M.
      • Jeffery W.R.
      Shadow response in the blind cavefish Astyanax reveals conservation of a functional pineal eye.
      we found that both populations of cavefish showed significant, although severely reduced, Ca2+ responses to the presented visual stimuli (Figures 3B–3D). In comparison to surface fish, Pachón and Molino larvae showed a significantly smaller number of responsive neurons to the visual stimulus, representing a reduction of responding neurons of ∼75% in Molino and ∼90% in Pachón (Figure 3E). The amplitude of the Ca2+ responses of the cavefish to the visual stimulus was also reduced by ∼60% (Figure 3F). The probability of the neurons to respond to visual stimuli was also significantly smaller in cavefish with respect to surface fish (Figure S3). Moreover, the “On/Off” neurons were absent in both Pachón and Molino larvae, and the inhibited neurons showed very weak responses that were almost absent in Pachón (Figures 3B–3D).The onset precision of the visual response with respect to the onset of the stimulus was also significantly more variable in the cavefish with respect to that of the surface fish (Figures S4C and S4D). The extent of the reduction in light response was severe enough that responses to light in cavefish tecta were undetectable by the naked eye, requiring rigorous statistical analyses to definitively determine whether neurons were truly responsive (STAR Methods). Together, these findings reveal that although cavefish exhibit severe reductions in visual tectal function, they maintain the genetic and cellular functions necessary to process visual stimulus.
      The presence of light responsiveness in the cavefish tectum raises the possibility that the retina is still functional in larval cavefish. Alternatively, responses in the tectum may be downstream of extra-retinal light-responsive organs such as the pineal gland.
      • Vígh B.
      • Manzano M.J.
      • Zádori A.
      • Frank C.L.
      • Lukáts A.
      • Röhlich P.
      • Szél A.
      • Dávid C.
      Nonvisual photoreceptors of the deep brain, pineal organs and retina.
      To distinguish between these possibilities, we imaged the retinal response in GCaMP-expressing Molino and Pachón larvae while presenting a series of 30 s light stimuli. Due to pigmentation, recording of the intact retina in surface fish was not possible. In the Pachón population, we identified strong spontaneous retinal waves known to play a role in the development of retinotectal pathways (Video S1).
      • Firth S.I.
      • Wang C.T.
      • Feller M.B.
      Retinal waves: mechanisms and function in visual system development.
      • Stafford B.K.
      • Sher A.
      • Litke A.M.
      • Feldheim D.A.
      Spatial-temporal patterns of retinal waves underlying activity-dependent refinement of retinofugal projections.
      • Ge X.
      • Zhang K.
      • Gribizis A.
      • Hamodi A.S.
      • Sabino A.M.
      • Crair M.C.
      Retinal waves prime visual motion detection by simulating future optic flow.
      However, no light-evoked activity was detected (Video S1; Figures 4A–4C ). The Molino retina showed neither retinal waves nor visually induced responses. These findings suggest that although the retina is still alive and active, at least in the case of the Pachón population, it does not respond to light. Therefore, light responsiveness observed in the optic tectum of cavefish must stem from a non-visual source (e.g., the pineal gland or deep-brain internal photoreceptors).
      • Vígh B.
      • Manzano M.J.
      • Zádori A.
      • Frank C.L.
      • Lukáts A.
      • Röhlich P.
      • Szél A.
      • Dávid C.
      Nonvisual photoreceptors of the deep brain, pineal organs and retina.
      ,
      • Fernandes A.M.
      • Fero K.
      • Arrenberg A.B.
      • Bergeron S.A.
      • Driever W.
      • Burgess H.A.
      Deep brain photoreceptors control light-seeking behavior in zebrafish larvae.
      Figure thumbnail gr4
      Figure 4Cavefish show spontaneous retinal waves, but not retinal visual responses
      (A) Top: raster plot of the retinal activity of a Pachón larva during periods of visual stimulation (flashes of light with a duration of 30 s and inter-stimulus interval of 120 s, gray bars). Middle: averaged ΔF/F of all neurons. Gray patches: visual stimuli. Bottom: zoom-in on the red dotted region in the top panel to better appreciate the temporal dynamics of the retinal waves. Gray scale bar: ΔF/F.
      (B) Top: raster plot of the averaged neuronal response across visual stimulation trials (same larva as in A). Bottom: ΔF/F averaged over cells and trials. Gray patch: visual stimulation.
      (C) Average ΔF/F across visual stimulation trials (during Stim) and a matching 30 s period of spontaneous activity before the stimulation onset (before Stim) for n = 3 Pachón larvae. Statistical significance was assessed using a unilateral, non-parametric paired Wilcoxon signed-rank test (p = 0.13).
      See also .

      Analysis of visual response in hybrid populations

      It is possible that the decrease in the evoked tectal light response in cavefish is due to the partial loss of eyes that, at 6 dpf, are still in the evolutionary process of degenerating.
      • Emam A.
      • Yoffe M.
      • Cardona H.
      • Soares D.
      Retinal morphology in Astyanax mexicanus during eye degeneration.
      The ability to generate surface-cave hybrid fish has been widely used to examine the functional and genetic relationship between traits.
      • McGaugh S.E.
      • Kowalko J.E.
      • Duboué E.
      • Lewis P.
      • Franz-Odendaal T.A.
      • Rohner N.
      • Gross J.B.
      • Keene A.C.
      Dark world rises: the emergence of cavefish as a model for the study of evolution, development, behavior, and disease.
      To determine whether changes in the anatomy of the tectum are related to the loss of eyes, we examined the relationship between traits in genetically and morphologically variable hybrid fish. By crossing transgenic HuC:GCaMP6s surface fish to Pachón or Molino cavefish and re-crossing the siblings for one generation, we generated genetically variable F2 surface × cave hybrids harboring the HuC:GCaMP6s transgene (Figure 5A). To determine whether there is a developmental or genetic relationship between eye size and the number of neurons in the tectum, we examined the relationship between these two traits. We found an association between eye diameter and the number of neurons in the tectum in hybrids from the Molino, but not in Pachón, caves (Figure 5B). These findings suggest that the reductions in the size of the tectum and eyes of Molino cavefish is controlled by shared genetic architecture, whereas these factors are independent in Pachón cavefish, suggesting that the evolutionary mechanisms that confer eye loss and/or tectal size in these two populations differ. The finding that tectal development is linked to eye size in Molino hybrids led us to question whether the tectum’s functional response is also related to eye size. We measured eye size in F2 fish, performed Ca2+ imaging in response to changes in illumination, and examined correlations between the two. As the tectal response in cavefish is so slight as to be nearly undetectable relative to the surface responses, we presumed the majority of the tectal response to be visual in nature and derived from genetic contributions from the surface parent. We found no significant relationship between eye size and tectal response, in any of the cell types previously identified (Figure S5). Therefore, there is a functional relationship between eye and tectum size in Molino, but eye size does not impact light-induced tectum activity in either cavefish population.
      Figure thumbnail gr5
      Figure 5Analysis of light response in a hybrid F2 population
      (A) Examples of 6 dpf F2 crosses, demonstrating variability in eye size and pigmentation.
      (B) Molino F2s (orange) exhibit a correlation between eye size and tectal neuron number (F1,15 = 10.06, p = 0.0063), whereas Pachón F2s (green) do not (F1,19 = 3.399, p = 0.08).
      (C) There is no relationship between the proportion of “On” and “Off” cells in the tectum of either Molino (F1,15 = 1.013, p = 0.33) or Pachón (F1,19 = 0.711, p = 0.41) F2s.
      (D) There is a significant correlation between the proportion of “On” and “On/Off” cells in the tectum of both Molino (F1,15 = 12.56, p = 0.003) and Pachón (F1,19 = 5.261, p = 0.03) F2s.
      (E) There is a significant correlation between the proportion of “Off” and “On/Off” cells in the tectum of both Molino (F1,15 = 16.03, p = 0.001) and Pachón (F1,19 = 14.97, p = 0.001) F2s.
      See also .
      To determine whether there is a shared genetic basis underlying the development of distinct neuronal functional populations in the tectum, we quantified different neuronal response types in surface-cave hybrids. In both Pachón and Molino cavefish, there was no relationship between the number of “On” cells and “Off” cells, suggesting that each population is independently regulated (Figure 5C). This finding suggests that distinct genetic pathways guide the development of these cell types within the tectum. However, a relationship was identified between each singly responsive cell type and dual-responsive “On/Off” cells, suggesting a shared genetic basis (Figures 5D and 5E). Additional correlation analyses between incidence of cell types, including the “sustained” and “inhibited” cells, yielded no significant correlations (data not shown). Taken together, these findings suggest that shared genetic architecture underlies the loss of “On/Off” cells and cells that exclusively respond to light onset or offset across both Molino and Pachón populations of cavefish. These findings are indicative of convergence on shared developmental processes contributing to changes in tectum function across independently evolved cavefish populations through different genetic mechanisms. Future studies building on these findings may provide a more in-depth understanding of the genetic factors underlying the independent evolution of the visual system across caves.

      Discussion

      Cave animals have evolved numerous phenotypic changes that are likely adaptive to a cave environment, including eye loss and expansion of olfactory and mechanoreceptive organs.
      • Yoshizawa M.
      Behaviors of cavefish offer insight into developmental evolution.
      ,
      • Blin M.
      • Tine E.
      • Meister L.
      • Elipot Y.
      • Bibliowicz J.
      • Espinasa L.
      • Rétaux S.
      Developmental evolution and developmental plasticity of the olfactory epithelium and olfactory skills in Mexican cavefish.
      ,
      • Yoshizawa M.
      • Gorički S.
      • Soares D.
      • Jeffery W.R.
      Evolution of a behavioral shift mediated by superficial neuromasts helps cavefish find food in darkness.
      These changes occur in diverse cavefish species throughout the world, suggesting that the shared ecological features in cave environments, such as perpetual darkness and reduced food availability, underlie the evolved changes in sensory processing.
      • Culver D.C.
      • Pipan T.
      The Biology of Caves and Other Subterranean Habitats.
      The repeated evolution of eye loss in the different A. mexicanus cavefish populations has provided a model to investigate the genetic basis of evolution. Several studies have identified numerous factors underlying eye loss, including a critical role for the degeneration of the lens and retina.
      • McGaugh S.E.
      • Gross J.B.
      • Aken B.
      • Blin M.
      • Borowsky R.
      • Chalopin D.
      • Hinaux H.
      • Jeffery W.R.
      • Keene A.
      • Ma L.
      • et al.
      The cavefish genome reveals candidate genes for eye loss.
      • Pottin K.
      • Hinaux H.
      • Rétaux S.
      Restoring eye size in Astyanax mexicanus blind cavefish embryos through modulation of the Shh and Fgf8 forebrain organising centres.
      • Sifuentes-Romero I.
      • Ferrufino E.
      • Thakur S.
      • Laboissonniere L.A.
      • Solomon M.
      • Smith C.L.
      • Keene A.C.
      • Trimarchi J.M.
      • Kowalko J.E.
      Repeated evolution of eye loss in Mexican cavefish: evidence of similar developmental mechanisms in independently evolved populations.
      • Yamamoto Y.
      • Byerly M.S.
      • Jackman W.R.
      • Jeffery W.R.
      Pleiotropic functions of embryonic sonic hedgehog expression link jaw and taste bud amplification with eye loss during cavefish evolution.
      • Yamamoto Y.
      • Stock D.W.
      • Jeffery W.R.
      Hedgehog signalling controls eye degeneration in blind cavefish.
      Furthermore, genetic complementation studies have revealed that different genetic mechanisms underlie eye loss across cave populations.
      • Borowsky R.
      Restoring sight in blind cavefish.
      ,
      • Wilkens H.
      Genetic interpretation of regressive evolutionary processes: studies on hybrid eyes of two Astyanax populations (Characidae, Pisces).
      Despite our detailed understanding of genetic and morphological changes that occur in sensory organs, strikingly little is known about evolutionary adaptations in sensory-processing centers of the brain. We utilized GCaMP-expressing transgenic A. mexicanus to examine the changes in the main visual-processing center in surface and cavefish, the optic tectum.
      In teleost fish, the optic tectum is the main visual-processing region; it is involved in the detection and processing of visual stimuli to generate goal-directed motor behaviors such as prey capture.
      • Marachlian E.
      • Avitan L.
      • Goodhill G.J.
      • Sumbre G.
      Principles of functional circuit connectivity: insights from spontaneous activity in the zebrafish optic tectum.
      Similar to our findings in A. mexicanus, numerous cellular response types have been previously identified in the zebrafish tectum, including those which respond selectively to changes in light level.
      • Barker A.J.
      • Helmbrecht T.O.
      • Grob A.A.
      • Baier H.
      Functional, molecular and morphological heterogeneity of superficial interneurons in the larval zebrafish tectum.
      We identified shared cellular response types across surface and both cave populations of A. mexicanus, although the number and sensitivity of responsive cells was severely reduced across cave populations. Imaging from the cavefish retina did not identify responses to flashes of light, raising the possibility that the light-induced activity that we observed in the tectum derives from non-visual sources such as internal photoreceptors and/or the pineal body.
      • Fernandes A.M.
      • Fero K.
      • Driever W.
      • Burgess H.A.
      Enlightening the brain: linking deep brain photoreception with behavior and physiology.
      These differences support the hypothesis that the light response in the tectum is vestigial in cavefish, suggesting a transformation in the function of the cavefish tectum. Also, in support of this hypothesis is the finding that fish from the evolutionarily older Pachón population exhibit more severe deficits in tectal light response relative to fish from the Molino caves.
      In zebrafish, the tectum is primarily devoted to visual processing but also receives mechanosensory and auditory information. Previous studies have identified a laminar organization of the zebrafish tectum, with different layers corresponding to distinct functional specializations.
      • Gabriel J.P.
      • Trivedi C.A.
      • Maurer C.M.
      • Ryu S.
      • Bollmann J.H.
      Layer-specific targeting of direction-selective neurons in the zebrafish optic tectum.
      ,
      • Gebhardt C.
      • Auer T.O.
      • Henriques P.M.
      • Rajan G.
      • Duroure K.
      • Bianco I.H.
      • Del Bene F.
      An interhemispheric neural circuit allowing binocular integration in the optic tectum.
      ,
      • Thompson A.W.
      • Scott E.K.
      Characterisation of sensitivity and orientation tuning for visually responsive ensembles in the zebrafish tectum.
      Numerous cellular response types have been previously identified in zebrafish, including those which respond selectively to changes in light level.
      • Barker A.J.
      • Helmbrecht T.O.
      • Grob A.A.
      • Baier H.
      Functional, molecular and morphological heterogeneity of superficial interneurons in the larval zebrafish tectum.
      Their identification in A. mexicanus provides an opportunity to investigate the genetic basis of tectal wiring through hybridization experiments. Our findings show that the prevalence of “On” cells and “Off” cells is uncorrelated in the tectum of hybrid individuals, suggesting that separate genetic mechanisms regulate these traits, and that likely distinct axon guidance molecules regulate the wiring of the tectum.
      • Culverwell J.
      • Karlstrom R.O.
      Making the connection: retinal axon guidance in the zebrafish.
      ,
      • Robles E.
      • Baier H.
      Assembly of synaptic laminae by axon guidance molecules.
      Conversely, the correlation of both of these cell types with dual-responsive “On/Off” cells suggests a shared developmental basis, e.g., overlapping expression of axon guidance molecules. Further experiments in surface-cave F2 hybrids, in combination with quantitative trait locus (QTL) mapping, could elucidate the genetic basis of these, and other, functional cell types in the tectum.
      Our analysis of the spatiotemporal correlations in the spontaneous activity within the tectum suggests cavefish have maintained functional connectivity despite the complete loss of a light-responsive retina and the near-complete loss of tectal light response. In the zebrafish optic tectum, spontaneous activity reflects the functional connectivity of the circuit, which is adapted to improve visual detection of prey-like visual stimuli.
      • Romano S.A.
      • Pietri T.
      • Pérez-Schuster V.
      • Jouary A.
      • Haudrechy M.
      • Sumbre G.
      Spontaneous neuronal network dynamics reveal circuit’s functional adaptations for behavior.
      These findings suggest that eye loss has not influenced the tectum’s functional connectivity. However, this trend did not hold true for the negative correlations within the tectum, which tended to be reduced in both Molino and Pachón populations. This suggests that tectal inhibition is tightly linked to visual processing or that only positive correlations and not inhibition play a role in the potential repurposing of the optic tectum in cavefish.
      It is possible that the visual circuitry of the sighted teleost has been repurposed for processing a different sensory modality in cavefish. This hypothesis is supported by the fact that in contrast to the significant reduction in visual response, which may represent a vestige of visual degeneration evolution, the functional connectivity of surface and cavefish did not show significant differences. Systematic characterization of the response of the tectum to a battery of sensory stimuli can determine whether this brain region has been repurposed to respond predominantly to non-visual sensory stimuli.
      The functional imaging described here represents the first use of genetically encoded Ca2+ indicators to examine evolved differences in brain function in a vertebrate model. In addition to loss of vision, cavefish are widely used to study many different behaviors including enhanced lateral line function, sleep loss, prey capture, olfactory processing, and stress.
      • Kowalko J.
      Utilizing the blind cavefish Astyanax mexicanus to understand the genetic basis of behavioral evolution.
      ,
      • McGaugh S.E.
      • Kowalko J.E.
      • Duboué E.
      • Lewis P.
      • Franz-Odendaal T.A.
      • Rohner N.
      • Gross J.B.
      • Keene A.C.
      Dark world rises: the emergence of cavefish as a model for the study of evolution, development, behavior, and disease.
      ,
      • Yoshizawa M.
      • Robinson B.G.
      • Duboué E.R.
      • Masek P.
      • Jaggard J.B.
      • O'Quin K.E.
      • Borowsky R.L.
      • Jeffery W.R.
      • Keene A.C.
      Distinct genetic architecture underlies the emergence of sleep loss and prey-seeking behavior in the Mexican cavefish.
      The application of whole-brain imaging approaches that have been developed in zebrafish to evolutionary models including A. mexicanus has the potential to identify neural mechanisms associated with evolved differences in these behaviors.
      • Ahrens M.B.
      • Engert F.
      Large-scale imaging in small brains.
      Therefore, this manuscript provides a framework from applying comparative functional imaging to identify variability in brain function.

      STAR★Methods

      Key resources table

      Tabled 1
      REAGENT or RESOURCESOURCEIDENTIFIER
      Chemicals, peptides, and recombinant proteins
      Mivacurium chlorideSigma-AldrichCAS: 106861-44-3
      Deposited data
      Raw dataThis paperhttps://doi.org/10.17632/y9j999yk25.1
      Experimental models: Organisms/strains
      Surface A. mexicanus Tg(Huc:H2B-GCaMP6s)This paperN/A
      Molino A. mexicanus Tg(Huc:H2B-GCaMP6s)This paperN/A
      Pachón A. mexicanus Tg(Huc:H2B-GCaMP6s)This paperN/A
      Software and algorithms
      MATLAB 2020aMathworkshttps://mathworks.com/
      Octave 6.2.0GNUhttps://www.gnu.org/software/octave/
      Other
      Low-melting point agaroseThermo-FisherCAT#17856
      Nunc glass bottom dishThermo-FisherCAT#150680
      Arduino UnoArduinohttps://www.arduino.cc/

      Resource availability

      Lead contact

      Further information and requests should be directed to and will be fulfilled by the lead contact, Alex Keene ( akeene@bio.tamu.edu ).

      Materials availability

      A. mexicanus lines used in this study can be made available by request to the lead contact.

      Experimental model and subject details

      Fish

      Animal husbandry was carried out as previously described
      • Borowsky R.
      Handling Astyanax mexicanus eggs and fry.
      and all protocols were approved by the IACUC Florida Atlantic University. Fish were housed in the Florida Atlantic University core facilities at 23 °C ± 1 °C constant water temperature throughout rearing for behavior experiments. Lights were kept on a 14:10 h light-dark cycle that remained constant throughout the animal's lifetime. Adult fish were fed a diet of blood worms to satiation 2-3x daily (Aquatic Foods, Fresno, CA), and standard flake fish food during periods when fish were not being used for breeding (Tetramin Pro).
      Larval fish used were collected as embryos from adult tanks following breeding, and raised in a temperature-controlled incubator at 23 °C ± 1 °C until 6 dpf, when they were used in imaging experiments.

      Method details

      Molecular cloning and transgenesis

      The A. mexicanus Tg(Huc:H2B-GCaMP6s) was generated using the previously published zebrafish Tg(elav3:H2B-GCaMP6s).
      • Freeman J.
      • Vladimirov N.
      • Kawashima T.
      • Mu Y.
      • Sofroniew N.J.
      • Bennett D.V.
      • Rosen J.
      • Yang C.T.
      • Looger L.L.
      • Ahrens M.B.
      Mapping brain activity at scale with cluster computing.
      Briefly, the zebrafish elavl3 promoter was cloned to a Tol2 vector with the subcloned H2B, which restricts expression to the nucleus. The transgene plasmid and transposase RNA were injected into one to four cell-stage embryos at 25 ng/μl to a 1-nl volume, as described previously.
      • Stahl B.A.
      • Jaggard J.B.
      • Chin J.S.R.
      • Kowalko J.E.
      • Keene A.C.
      • Duboué E.R.
      Manipulation of gene function in Mexican cavefish.
      Injected (F0) founders were grown to sexual maturity and bred, and transgenic lines were isolated by identification of high expression of fluorescence in the F1 and F2 generations. All transgenic animals were raised under standard conditions.

      Functional Imaging

      Prior to imaging experiments, larvae were paralyzed by 60 s of immersion in 0.5mg/ml mivacurium chloride, rinsed thoroughly with system water, and embedded in 2% low-melting point agarose (Thermo-Fisher, 17856), in clear plastic 35mm dishes (Nunc, Thermo-Fisher, 150680), with a strip of white LED lights (Lepro) affixed to the mounting dish rostral to the larvae. The microscope was shielded to prevent unintended light stimulus. Following a minimum of 1 minute of recording to capture basal activity levels, a 30 second visual stimulus was applied. In follow-up experiments, multiple 30 second stimuli were applied, separated by additional 1 minute intervals. Visual stimulus was controlled by custom code running on Octave 6.2.0, utilizing the Arduino package. All visual stimulus experiments were acquired on a Nikon A1 confocal microscope, at 20X magnification, 2X digital zoom, at a frame rate of 1.96 Hz.

      Two-photon Ca2+ imaging

      To capture spontaneous activity in the tectum, larvae were prepared as described above, and Ca2+ activity in the tectum was monitored for a period of 30 minutes, at a frame rate of 3.96 Hz on a Nikon A1R Multiphoton microscope, with a Chameleon Vision II Ti:Sapphire tunable laser at an excitation wavelength of 940 nm. Images were acquired with a 25x objective, and 1.5X digital zoom.

      Quantification and statistical analysis

      Ca2+ activity analysis

      Ca2+ activity of individual cells was extracted using a MATLAB toolbox developed by Romano et al.
      • Romano S.A.
      • Pérez-Schuster V.
      • Jouary A.
      • Boulanger-Weill J.
      • Candeo A.
      • Pietri T.
      • Sumbre G.
      An integrated calcium imaging processing toolbox for the analysis of neuronal population dynamics.
      Briefly, an average image was generated from acquired time-lapse images, and a watershed algorithm was applied to generate ROIs outlining individual cells. These were used to extract raw fluorescence data from the time-lapse images, and calculate the ΔF/F values for individual cells (Figure S1). Cell counts (for Figure 1C) were derived from the automated segmentation.

      Correlation analysis

      In order to compute the average probability density function (pdf) for correlations in surface, Molino and Pachón, we estimated the correlation pdf for each larva by counting the number of correlations in regularly spaced (0.05) correlation bins between -1 and 1. We then averaged those pdf together within each population. In order to quantify the amount of correlation that emerged from the spontaneous rate of activity, we built a null model by shifting circularly the neuron’s activity time series by a random lag between 0 and 30min. The latter corresponds to the total duration of the experiments. This keeps the time series untouched while destroying the temporal correlations between time series. We performed 1000 repetitions of this procedure to obtain the null model.
      To compare the distributions of correlations between the surface, Molino and Pachón populations, we quantified the distance between the null-model distribution and the distribution of correlations in our dataset using the Jensen-Shannon divergence for each larva. The discrete Kullback-Leibler divergence DKL, or relative entropy, quantifies the similarity between probability distributions P and Q defined on some probability space X.
      DKL(PQ)=xXP(x)log(P(x)Q(x))


      The Jensen-Shannon divergence DJS is based on the Kullback-Leibler divergence, but is symmetric, such that DJS(PQ)=DJS(QP). It can be defined as:
      DJS(PQ)=12DKL(PM)+12DKL(QM)


      where
      M=12(P+Q)


      Finally, to investigate the spatial structure of correlations in the optic tectum of surface, Molino and Pachón, we averaged the correlation between pairs of neurons falling into regularly spaced (20 μm) distance bins between 0 and 200 μm to obtain the distribution of correlations against distance.

      Identification and categorization of cells responsive to whole-field visual stimuli using linear regression

      To identify and categorize neurons that are responsive to visual stimuli, we used a multivariate linear regression model. To search for neurons with specific responding profiles, we used a regressor-based analysis. We used 3 binary regressors based on the timings of the onset and offset the visual stimuli (R1) or the offset (R2) of the visual stimulation, or a sustained activation or inhibition during the whole duration of the stimulus (R3). To model the slow dynamics of the H2B-GCaMP6s Ca2+ sensor, we convolved those regressors with a decaying exponential kernel with characteristic time constant τ = 3.5s.
      • Migault G.
      • van der Plas T.L.
      • Trentesaux H.
      • Panier T.
      • Candelier R.
      • Proville R.
      • Englitz B.
      • Debrégeas G.
      • Bormuth V.
      Whole-brain calcium imaging during physiological vestibular stimulation in larval zebrafish.
      More specifically, if we denote yn the time series of the relative change of fluorescence over time for neuron n, we have:
      yn=β0n+β1n×R1+β2n×R2+β3n×R3+ϵn


      Where β0n is the intercept, βin are the regression coefficients associated to each regressor Ri, and ϵn represent the residuals for neuron n.
      We compared the fraction of variance explained by the linear model R2, which is a measure of the goodness of fit and compared this to a null model. We also compared the regression coefficient βi obtained for each regressor, which captured the intensity of the response of a neuron, against the distribution of regression coefficients under the null model.
      A neuron was classified as responsive if its R2 value exceeded the 95th percentile of the distribution of R2 over the whole population of recorded neurons in the null model (R952), and if exactly one of its regression coefficient was also higher than the 95th percentile of the null β distribution for this neuron (βin,95), or lower than the 5th percentile in the case of inhibition (βin,5). This procedure ensures that silent neurons, which may present a very good fit (high R2 values) but low β value are excluded from the population of responsive neurons, and also allows to categorize the neurons into 5 response profiles:
      • ON cells: Rn2R952β1β1n,95β2<β2n,95β3<β3n,95β3>β3n,5
      • OFF cells: Rn2R952β1<β1n,95β2β2n,95β3<β3n,95β3>β3n,5
      • ON/OFF cells: Rn2R952β1β1n,95β2β2n,95β3<β3n,95β3>β3n,5
      • Sustained cells: Rn2R952β1<β1n,95β2<β2n,95β3β3n,95
      • Inhibited cells: Rn2R952β1<β1n,95β2<β2n,95β3β3n,5

      Detection of significant Ca2+ events

      In order to detect significant Ca2+ events in the fluorescence time series, we used the method described in Romano et al.
      • Romano S.A.
      • Pietri T.
      • Pérez-Schuster V.
      • Jouary A.
      • Haudrechy M.
      • Sumbre G.
      Spontaneous neuronal network dynamics reveal circuit’s functional adaptations for behavior.
      This enables us to compute the fraction of neurons active at any time point during the recording.

      Quantification of the temporal precision of the neural response to the onset of the visual stimulus

      Due to the level of noise in the data, especially in the Molino and Pachón populations, we could not reliably detect the onset of the visually induced Ca2+ response for each neuron. Therefore, we chose to estimate the variability of the onset of the response by averaging together the across-trial probability of response (Figures S3G–S3I) for all cells (ON, ON/OFF and Sustained cells). Then, we performed a linear regression of the obtained average in a small time window (-2s to +2s) around the onset of the stimulus. As a proof of principle of this method, we computed the slope of the linear regression from synthetic data with different jittering levels (S6A,B).

      Statistical inference and data analysis

      Non parametric statistical tests were performed when the normality assumption could not be upheld. Data analysis was performed using custom scripts written in MATLAB (R2020a).

      Data and code availability

      All data used for analysis has been uploaded to Mendeley Data and is publicly available as of the date of publication. The DOI is listed in the key resources table. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

      Acknowledgments

      This work was supported by funding from an HFSP award ( HFSP-RGP0062 ) to A.C.K. and G.S., NSF award ( 1923372 ) to E.R.D., and NIH awards ( 1R01GM127872 , OD030214 , and NS122166 ) to A.C.K. We would like to thank members of the Keene, Sumbre, and Duboué labs for technical assistance and helpful discussion.

      Author contributions

      A.C.K., G.S., and E.L. designed the experiments. E.L. and B.M. performed the experiments. E.L., M.P., and B.M. analyzed the data. E.L., B.M., and J.B.J. generated the transgenic lines used in all experiments. A.C.K., G.S., E.R.D., E.L., and M.P. wrote the manuscript. E.R.D. and J.B.J. provided additional help with the experimental design and data analysis.

      Declaration of interests

      The authors declare no competing interests.

      Supplemental information

      References

        • Schlosser G.
        A short history of nearly every sense—the evolutionary history of vertebrate sensory cell types.
        Integr. Comp. Biol. 2018; 58: 301-316
        • Wylie D.R.
        • Gutiérrez-Ibáñez C.
        • Iwaniuk A.N.
        Integrating brain, behavior, and phylogeny to understand the evolution of sensory systems in birds.
        Front. Neurosci. 2015; 9: 281
        • Lloyd E.
        • Olive C.
        • Stahl B.A.
        • Jaggard J.B.
        • Amaral P.
        • Duboué E.R.
        • Keene A.C.
        Evolutionary shift towards lateral line dependent prey capture behavior in the blind Mexican cavefish.
        Dev. Biol. 2018; 441: 328-337
        • Paz A.
        • McDole B.
        • Kowalko J.E.
        • Duboue E.R.
        • Keene A.C.
        Evolution of the acoustic startle response of Mexican cavefish.
        J. Exp. Zool. B Mol. Dev. Evol. 2020; 334: 474-485
        • Hinaux H.
        • Devos L.
        • Blin M.
        • Elipot Y.
        • Bibliowicz J.
        • Alié A.
        • Rétaux S.
        Sensory evolution in blind cavefish is driven by early embryonic events during gastrulation and neurulation.
        Development. 2016; 143: 4521-4532
        • Jaggard J.B.
        • Lloyd E.
        • Yuiska A.
        • Patch A.
        • Fily Y.
        • Kowalko J.E.
        • Appelbaum L.
        • Duboue E.R.
        • Keene A.C.
        Cavefish brain atlases reveal functional and anatomical convergence across independently evolved populations.
        Sci. Adv. 2020; 6: 3126-3142
        • Gross J.B.
        The complex origin of Astyanax cavefish.
        BMC Evol. Biol. 2012; 12: 105
        • Bronchti G.
        • Schönenberger N.
        • Welker E.
        • Van der Loos H.
        Barrelfield expansion after neonatal eye removal in mice.
        Neuroreport. 1992; 3: 489-492
        • Touj S.
        • Gallino D.
        • Chakravarty M.M.
        • Bronchti G.
        • Piché M.
        Structural brain plasticity induced by early blindness.
        Eur. J. Neurosci. 2021; 53: 778-795
        • Jeffery W.R.
        Regressive evolution in Astyanax cavefish.
        Annu. Rev. Genet. 2009; 43: 25-47
        • Kowalko J.
        Utilizing the blind cavefish Astyanax mexicanus to understand the genetic basis of behavioral evolution.
        J. Exp. Biol. 2020; 223: jeb208835
        • Yoshizawa M.
        Behaviors of cavefish offer insight into developmental evolution.
        Mol. Reprod. Dev. 2015; 82: 268-280
        • Chin J.S.R.
        • Gassant C.E.
        • Amaral P.M.
        • Lloyd E.
        • Stahl B.A.
        • Jaggard J.B.
        • Keene A.C.
        • Duboue E.R.
        Convergence on reduced stress behavior in the Mexican blind cavefish.
        Dev. Biol. 2018; 441: 319-327
        • Kowalko J.E.
        • Rohner N.
        • Linden T.A.
        • Rompani S.B.
        • Warren W.C.
        • Borowsky R.
        • Tabin C.J.
        • Jeffery W.R.
        • Yoshizawa M.
        Convergence in feeding posture occurs through different genetic loci in independently evolved cave populations of Astyanax mexicanus.
        Proc. Natl. Acad. Sci. USA. 2013; 110: 16933-16938
        • Duboué E.R.
        • Keene A.C.
        • Borowsky R.L.
        Evolutionary convergence on sleep loss in cavefish populations.
        Curr. Biol. 2011; 21: 671-676
        • Jeffery W.R.
        Adaptive evolution of eye degeneration in the Mexican blind cavefish.
        J. Hered. 2005; 96: 185-196
        • Krishnan J.
        • Rohner N.
        Cavefish and the basis for eye loss.
        Philos. Trans. R. Soc. Lond. B Biol. Sci. 2017; 372: 20150487
        • McGaugh S.E.
        • Gross J.B.
        • Aken B.
        • Blin M.
        • Borowsky R.
        • Chalopin D.
        • Hinaux H.
        • Jeffery W.R.
        • Keene A.
        • Ma L.
        • et al.
        The cavefish genome reveals candidate genes for eye loss.
        Nat. Commun. 2014; 5: 5307
        • Krauzlis R.J.
        • Lovejoy L.P.
        • Zénon A.
        Superior colliculus and visual spatial attention.
        Annu. Rev. Neurosci. 2013; 36: 165-182
        • Thompson A.W.
        • Vanwalleghem G.C.
        • Heap L.A.
        • Scott E.K.
        Functional profiles of visual-auditory-and water flow-responsive neurons in the zebrafish tectum.
        Curr. Biol. 2016; 26: 743-754
        • Bianco I.H.
        • Engert F.
        Visuomotor transformations underlying hunting behavior in zebrafish.
        Curr. Biol. 2015; 25: 831-846
        • Förster D.
        • Helmbrecht T.O.
        • Mearns D.S.
        • Jordan L.
        • Mokayes N.
        • Baier H.
        Retinotectal circuitry of larval zebrafish is adapted to detection and pursuit of prey.
        eLife. 2020; 9: 1-26
        • Barker A.J.
        • Baier H.
        Sensorimotor decision making in the zebrafish tectum.
        Curr. Biol. 2015; 25: 2804-2814
        • Loomis C.
        • Peuß R.
        • Jaggard J.B.
        • Wang Y.
        • McKinney S.A.
        • Raftopoulos S.C.
        • Raftopoulos A.
        • Whu D.
        • Green M.
        • McGaugh S.E.
        • et al.
        An adult brain atlas reveals broad neuroanatomical changes in independently evolved populations of Mexican cavefish.
        Front. Neuroanat. 2019; 13: 88
        • Herculano-Houzel S.
        Scaling of brain metabolism with a fixed energy budget per neuron: implications for neuronal activity, plasticity and evolution.
        PLoS One. 2011; 6e17514
        • Hyder F.
        • Rothman D.L.
        • Bennett M.R.
        Cortical energy demands of signaling and nonsignaling components in brain are conserved across mammalian species and activity levels.
        Proc. Natl. Acad. Sci. USA. 2013; 110: 3549-3554
        • Bollmann J.H.
        The zebrafish visual system: from circuits to behavior.
        Annu. Rev. Vis. Sci. 2019; 5: 269-293
        • Niell C.M.
        • Smith S.J.
        Functional imaging reveals rapid development of visual response properties in the zebrafish tectum.
        Neuron. 2005; 45: 941-951
        • Gabriel J.P.
        • Trivedi C.A.
        • Maurer C.M.
        • Ryu S.
        • Bollmann J.H.
        Layer-specific targeting of direction-selective neurons in the zebrafish optic tectum.
        Neuron. 2012; 76: 1147-1160
        • Preuss S.J.
        • Trivedi C.A.
        • Vom Berg-Maurer C.M.
        • Ryu S.
        • Bollmann J.H.
        Classification of object size in retinotectal microcircuits.
        Curr. Biol. 2014; 24: 2376-2385
        • Romano S.A.
        • Pietri T.
        • Pérez-Schuster V.
        • Jouary A.
        • Haudrechy M.
        • Sumbre G.
        Spontaneous neuronal network dynamics reveal circuit’s functional adaptations for behavior.
        Neuron. 2015; 85: 1070-1085
        • Pietri T.
        • Romano S.A.
        • Pérez-Schuster V.
        • Boulanger-Weill J.
        • Candat V.
        • Sumbre G.
        The emergence of the spatial structure of tectal spontaneous activity is independent of visual inputs.
        Cell Rep. 2017; 19: 939-948
        • Stahl B.A.
        • Peuß R.
        • McDole B.
        • Kenzior A.
        • Jaggard J.B.
        • Gaudenz K.
        • Krishnan J.
        • McGaugh S.E.
        • Duboue E.R.
        • Keene A.C.
        • Rohner N.
        Stable transgenesis in Astyanax mexicanus using the Tol2 transposase system.
        Dev. Dyn. 2019; 248: 679-687
        • Wilkens H.
        Regressive evolution: ontogeny and genetics of cavefish eye rudimentation.
        Biol. J. Linn. Soc. Lond. 2007; 92: 287-296
        • Bradic M.
        • Beerli P.
        • García-de León F.J.
        • Esquivel-Bobadilla S.
        • Borowsky R.L.
        Gene flow and population structure in the Mexican blind cavefish complex (Astyanax mexicanus).
        BMC Evol. Biol. 2012; 12: 9https://doi.org/10.1186/1471-2148-12-9
        • Herman A.
        • Brandvain Y.
        • Weagley J.
        • Jeffery W.R.
        • Keene A.C.
        • Kono T.J.Y.
        • Bilandžija H.
        • Borowsky R.
        • Espinasa L.
        • O’Quin K.
        • et al.
        The role of gene flow in rapid and repeated evolution of cave-related traits in Mexican tetra, Astyanax mexicanus.
        Mol. Ecol. 2018; 27: 4397-4416
        • Mitchell R.W.
        • Russell W.H.
        • Elliott W.R.
        Mackey C. Barnett G.E. Mexican Eyeless Characin Fishes, Genus Astyanax: Environment, Distribution, and Evolution. Texas Tech Press, 1977
        • Borowsky R.
        Restoring sight in blind cavefish.
        Curr. Biol. 2008; 18: R23-R24
        • Freeman J.
        • Vladimirov N.
        • Kawashima T.
        • Mu Y.
        • Sofroniew N.J.
        • Bennett D.V.
        • Rosen J.
        • Yang C.T.
        • Looger L.L.
        • Ahrens M.B.
        Mapping brain activity at scale with cluster computing.
        Nat. Methods. 2014; 11: 941-950
        • Romano S.A.
        • Pérez-Schuster V.
        • Jouary A.
        • Boulanger-Weill J.
        • Candeo A.
        • Pietri T.
        • Sumbre G.
        An integrated calcium imaging processing toolbox for the analysis of neuronal population dynamics.
        PLOS Comput. Biol. 2017; 13e1005526
        • Basso M.A.
        • May P.J.
        Circuits for action and cognition: a view from the superior colliculus.
        Annu. Rev. Vis. Sci. 2017; 3: 197-226https://doi.org/10.1146/annurev-vision-102016-061234
        • Helmbrecht T.O.
        • dal Maschio M.
        • Donovan J.C.
        • Koutsouli S.
        • Baier H.
        Topography of a visuomotor transformation.
        Neuron. 2018; 100: 1429-1445.e4
        • Soares D.
        • Yamamoto Y.
        • Strickler A.G.
        • Jeffery W.R.
        The lens has a specific influence on optic nerve and tectum development in the blind cavefish Astyanax.
        Dev. Neurosci. 2004; 26: 308-317
        • Robles E.
        • Filosa A.
        • Baier H.
        Precise lamination of retinal axons generates multiple parallel input pathways in the tectum.
        J. Neurosci. 2013; 33: 5027-5039
        • Robles E.
        • Fields N.P.
        • Baier H.
        The zebrafish visual system transmits dimming information via multiple segregated pathways.
        J. Comp. Neurol. 2021; 529: 539-552
        • Voneida T.J.
        • Fish S.E.
        Central nervous system changes related to the reduction of visual input in a naturally blind fish (Astyanax hubbsi ).
        Am. Zool. 1984; 24: 775-782
        • Yoshizawa M.
        • Jeffery W.R.
        Shadow response in the blind cavefish Astyanax reveals conservation of a functional pineal eye.
        J. Exp. Biol. 2008; 211: 292-299
        • Vígh B.
        • Manzano M.J.
        • Zádori A.
        • Frank C.L.
        • Lukáts A.
        • Röhlich P.
        • Szél A.
        • Dávid C.
        Nonvisual photoreceptors of the deep brain, pineal organs and retina.
        Histol. Histopathol. 2002; 17: 555-590
        • Firth S.I.
        • Wang C.T.
        • Feller M.B.
        Retinal waves: mechanisms and function in visual system development.
        Cell Calcium. 2005; 37: 425-432
        • Stafford B.K.
        • Sher A.
        • Litke A.M.
        • Feldheim D.A.
        Spatial-temporal patterns of retinal waves underlying activity-dependent refinement of retinofugal projections.
        Neuron. 2009; 64: 200-212
        • Ge X.
        • Zhang K.
        • Gribizis A.
        • Hamodi A.S.
        • Sabino A.M.
        • Crair M.C.
        Retinal waves prime visual motion detection by simulating future optic flow.
        Science. 2021; 373eabd0830
        • Fernandes A.M.
        • Fero K.
        • Arrenberg A.B.
        • Bergeron S.A.
        • Driever W.
        • Burgess H.A.
        Deep brain photoreceptors control light-seeking behavior in zebrafish larvae.
        Curr. Biol. 2012; 22: 2042-2047
        • Emam A.
        • Yoffe M.
        • Cardona H.
        • Soares D.
        Retinal morphology in Astyanax mexicanus during eye degeneration.
        J. Comp. Neurol. 2020; 528: 1523-1534
        • McGaugh S.E.
        • Kowalko J.E.
        • Duboué E.
        • Lewis P.
        • Franz-Odendaal T.A.
        • Rohner N.
        • Gross J.B.
        • Keene A.C.
        Dark world rises: the emergence of cavefish as a model for the study of evolution, development, behavior, and disease.
        J. Exp. Zool. B Mol. Dev. Evol. 2020; 334: 397-404
        • Blin M.
        • Tine E.
        • Meister L.
        • Elipot Y.
        • Bibliowicz J.
        • Espinasa L.
        • Rétaux S.
        Developmental evolution and developmental plasticity of the olfactory epithelium and olfactory skills in Mexican cavefish.
        Dev. Biol. 2018; 441: 242-251
        • Yoshizawa M.
        • Gorički S.
        • Soares D.
        • Jeffery W.R.
        Evolution of a behavioral shift mediated by superficial neuromasts helps cavefish find food in darkness.
        Curr. Biol. 2010; 20: 1631-1636
        • Culver D.C.
        • Pipan T.
        The Biology of Caves and Other Subterranean Habitats.
        Oxford University Press, 2009
        • McGaugh S.E.
        • Gross J.B.
        • Aken B.
        • Blin M.
        • Borowsky R.
        • Chalopin D.
        • Hinaux H.
        • Jeffery W.R.
        • Keene A.
        • Ma L.
        • et al.
        The cavefish genome reveals candidate genes for eye loss.
        Nat. Commun. 2014; 5: 5307
        • Pottin K.
        • Hinaux H.
        • Rétaux S.
        Restoring eye size in Astyanax mexicanus blind cavefish embryos through modulation of the Shh and Fgf8 forebrain organising centres.
        Development. 2011; 138: 2467-2476
        • Sifuentes-Romero I.
        • Ferrufino E.
        • Thakur S.
        • Laboissonniere L.A.
        • Solomon M.
        • Smith C.L.
        • Keene A.C.
        • Trimarchi J.M.
        • Kowalko J.E.
        Repeated evolution of eye loss in Mexican cavefish: evidence of similar developmental mechanisms in independently evolved populations.
        J. Exp. Zool. B Mol. Dev. Evol. 2020; 334: 423-437
        • Yamamoto Y.
        • Byerly M.S.
        • Jackman W.R.
        • Jeffery W.R.
        Pleiotropic functions of embryonic sonic hedgehog expression link jaw and taste bud amplification with eye loss during cavefish evolution.
        Dev. Biol. 2009; 330: 200-211
        • Yamamoto Y.
        • Stock D.W.
        • Jeffery W.R.
        Hedgehog signalling controls eye degeneration in blind cavefish.
        Nature. 2004; 431: 844-847
        • Wilkens H.
        Genetic interpretation of regressive evolutionary processes: studies on hybrid eyes of two Astyanax populations (Characidae, Pisces).
        Evolution. 1971; 25: 530-544
        • Marachlian E.
        • Avitan L.
        • Goodhill G.J.
        • Sumbre G.
        Principles of functional circuit connectivity: insights from spontaneous activity in the zebrafish optic tectum.
        Front. Neural Circuits. 2018; 12: 46
        • Barker A.J.
        • Helmbrecht T.O.
        • Grob A.A.
        • Baier H.
        Functional, molecular and morphological heterogeneity of superficial interneurons in the larval zebrafish tectum.
        J. Comp. Neurol. 2021; 529: 2159-2175
        • Fernandes A.M.
        • Fero K.
        • Driever W.
        • Burgess H.A.
        Enlightening the brain: linking deep brain photoreception with behavior and physiology.
        Bioessays. 2013; 35: 775-779
        • Gebhardt C.
        • Auer T.O.
        • Henriques P.M.
        • Rajan G.
        • Duroure K.
        • Bianco I.H.
        • Del Bene F.
        An interhemispheric neural circuit allowing binocular integration in the optic tectum.
        Nat. Commun. 2019; 10: 5471
        • Thompson A.W.
        • Scott E.K.
        Characterisation of sensitivity and orientation tuning for visually responsive ensembles in the zebrafish tectum.
        Sci. Rep. 2016; 6: 34887
        • Culverwell J.
        • Karlstrom R.O.
        Making the connection: retinal axon guidance in the zebrafish.
        Semin. Cell Dev. Biol. 2002; 13: 497-506
        • Robles E.
        • Baier H.
        Assembly of synaptic laminae by axon guidance molecules.
        Curr. Opin. Neurobiol. 2012; 22: 799-804
        • Yoshizawa M.
        • Robinson B.G.
        • Duboué E.R.
        • Masek P.
        • Jaggard J.B.
        • O'Quin K.E.
        • Borowsky R.L.
        • Jeffery W.R.
        • Keene A.C.
        Distinct genetic architecture underlies the emergence of sleep loss and prey-seeking behavior in the Mexican cavefish.
        BMC Biol. 2015; 13: 15
        • Ahrens M.B.
        • Engert F.
        Large-scale imaging in small brains.
        Curr. Opin. Neurobiol. 2015; 32: 78-86
        • Borowsky R.
        Handling Astyanax mexicanus eggs and fry.
        Cold Spring Harb. Protoc. 2008; 2008 (pdb.prot5093)
        • Stahl B.A.
        • Jaggard J.B.
        • Chin J.S.R.
        • Kowalko J.E.
        • Keene A.C.
        • Duboué E.R.
        Manipulation of gene function in Mexican cavefish.
        J. Vis. Exp. 2019; 146e59093
        • Migault G.
        • van der Plas T.L.
        • Trentesaux H.
        • Panier T.
        • Candelier R.
        • Proville R.
        • Englitz B.
        • Debrégeas G.
        • Bormuth V.
        Whole-brain calcium imaging during physiological vestibular stimulation in larval zebrafish.
        Curr. Biol. 2018; 28: 3723-3735.e6