Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Celsr3 drives development and connectivity of the acoustic startle hindbrain circuit

  • Joy H. Meserve ,

    Contributed equally to this work with: Joy H. Meserve, Maria F. Navarro

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Cell and Developmental Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States of America

    ⨯
  • Maria F. Navarro ,

    Contributed equally to this work with: Joy H. Meserve, Maria F. Navarro

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Cell and Developmental Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States of America

    ⨯
  • Elelbin A. Ortiz,

    Roles Funding acquisition, Investigation, Methodology, Writing – review & editing

    Affiliation Department of Cell and Developmental Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States of America

    ⨯
  • Michael Granato

    Roles Conceptualization, Funding acquisition, Writing – review & editing

    granatom@pennmedicine.upenn.edu

    Affiliation Department of Cell and Developmental Biology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, United States of America

    ⨯

Abstract

In the developing brain, groups of neurons organize into functional circuits that direct diverse behaviors. One such behavior is the evolutionarily conserved acoustic startle response, which in zebrafish is mediated by a well-defined hindbrain circuit. While numerous molecular pathways that guide neurons to their synaptic partners have been identified, it is unclear if and to what extent distinct neuron populations in the startle circuit utilize shared molecular pathways to ensure coordinated development. Here, we show that the planar cell polarity (PCP)-associated atypical cadherins Celsr3 and Celsr2, as well as the Celsr binding partner Frizzled 3a/Fzd3a, are critical for axon guidance of two neuron types that form synapses with each other: the command-like neuron Mauthner cells that drive the acoustic startle escape response, and spiral fiber neurons which provide excitatory input to Mauthner cells. We find that Mauthner axon growth towards synaptic targets is vital for Mauthner survival. We also demonstrate that symmetric spiral fiber input to Mauthner cells is critical for escape direction, which is necessary to respond to directional threats. Moreover, we identify distinct roles for Celsr3 and Celsr2, as Celsr3 is required for startle circuit development while Celsr2 is dispensable, though Celsr2 can partially compensate for loss of Celsr3 in Mauthner cells. This contrasts with facial branchiomotor neuron migration in the hindbrain, which requires Celsr2 while we find that Celsr3 is dispensable. Combined, our data uncover critical and distinct roles for individual PCP components during assembly of the acoustic startle hindbrain circuit.

Author summary

The assembly of neuronal circuits that drive behavior requires coordination of molecular pathways that govern neuron development and connectivity. Disruption of circuit development can lead to behavioral dysfunction, but understanding how circuits develop in vivo, especially in the immensely complex human brain, poses a significant challenge. Here, we leverage a well-characterized circuit driving a conserved protective behavior, the acoustic startle response, in the developing zebrafish to investigate molecular pathways driving circuit assembly. We discovered that proteins in the planar cell polarity (PCP) pathway are critical for development of two neuron types in the startle hindbrain circuit: the large paired “Mauthner neurons,” which mediate the acoustic startle fast escape, and a group of specialized Mauthner-activating “spiral fiber neurons.” We found that the PCP atypical cadherin Celsr3 guides Mauthner axon growth towards and into the spinal cord, where Mauthner axons form connections with motor neurons essential for the startle response. Additionally, we uncovered unique and overlapping roles for Celsr3 and the related Celsr2 protein in hindbrain circuit development, shedding light on their specific functions in neurodevelopment. Overall, our findings in zebrafish illustrate how the conserved PCP pathway orchestrates circuit assembly and influences behavior.

Citation: Meserve JH, Navarro MF, Ortiz EA, Granato M (2024) Celsr3 drives development and connectivity of the acoustic startle hindbrain circuit. PLoS Genet 20(10): e1011415. https://doi.org/10.1371/journal.pgen.1011415

Editor: Cecilia Moens, Fred Hutchinson Cancer Research Center, UNITED STATES OF AMERICA

Received: May 8, 2024; Accepted: September 2, 2024; Published: October 21, 2024

Copyright: © 2024 Meserve et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.

Funding: This work was funded by National Institutes of Health (nih.gov) National Institute of Mental Health (F32MH115434 to JHM), Eunice Kennedy Shriver National Institute of Child Health and Human Development (T32HD083185 to MFN), National Institutes of Health/National Institute on Deafness and Other Communication Disorders (T32DC016903 to EAO), and National Institute of Neurological Disorders and Stroke (R01NS097914 and R01NS118921 to MG). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The startle response is an evolutionary conserved behavior that enables animals to adopt a protective stance or escape a dangerous situation. In humans, the startle response is elicited by sudden and intense stimuli, which may be auditory, somatosensory, and/or visual, and manifests with an eyeblink and bilateral contraction of facial and neck muscles [1]. In small prey animals including insects [2–4] and zebrafish [5,6], as well as other teleost fish [7], an intense stimulus can elicit a fast escape. This escape can be directional, away from the perceived stimulus, or in a stereotyped direction, such as forward or backward. The speed and movements of these startle responses depend upon the underlying neural circuit.

In zebrafish, the acoustic startle circuit assembles during embryogenesis [5,8,9]. Central to the circuit are the Mauthner cells, a pair of large reticulospinal neurons located in the hindbrains of some amphibians [10] and fish [7], including zebrafish. These neurons are functionally similar to startle-associated giant reticulospinal neurons in the mammalian caudal pontine reticular nucleus [11,12]. In five day old larval zebrafish (5 days post-fertilization, dpf), intense acoustic stimuli perceived by hair cells in the ear and lateral line robustly elicit Mauthner-mediated fast escape responses. Acoustic stimulation of hair cells in the inner ear is conveyed to ipsilateral Mauthners via the eighth cranial nerve (Fig 1A). Mauthner axons cross the midline and extend posteriorly throughout the entire length of the spinal cord, where they synapse onto motor neurons that induce muscle contraction and a sharp turn away from the acoustic stimulus. This initial turn, called the “C1-bend,” is followed by a second turn and subsequent fast swimming to escape (Fig 1B). The circuit also includes feedback inhibitory neurons, which provide both recurrent inhibition (to prevent multiple action potentials from the same Mauthner) and reciprocal inhibition (to prevent both Mauthners from firing) [13]. The Mauthner also receives excitatory input at its axon initial segment from spiral fiber neurons, which are activated in response to acoustic stimuli [14]. These spiral fiber synapses, along with feedback inhibitory synapses and surrounding glia, form the Mauthner axon cap [15]. Unilateral ablations of spiral fiber or Mauthner neurons result in larvae that only escape in a single direction, driven by an initial left C1-bend or right C1-bend [14,16], which could be towards the source of the acoustic stimulus and danger rather than away. The dire consequences of a misdirected escape demand that proper development of the startle circuit occurs.

thumbnail
Download:
Fig 1. Celsr3 is required for directionally unbiased acoustic startle escapes.

A) Schematic of a 5 dpf (days post-fertilization) larva (dorsal view) with a simplified diagram of the hindbrain circuit that drives the acoustic startle fast escape. +/- symbols indicate excitatory/inhibitory synapses. Dotted line indicates unknown connection. In response to a startling acoustic stimulus, Mauthner firing results in direct activation of contralateral motor neurons, leading to unilateral muscle contraction and a sharp turn (C1-bend). An iteration of this circuit model was previously used by the authors [87]. B) Representative right and left acoustic startle fast escapes in wild type 5 dpf larvae. Pre-Stimulus and C1-Bend: still images of when the acoustic stimulus was delivered (0 ms) and the highest angle of the Mauthner-dependent initial turn (C1-bend). Middle panel: 90 ms temporal projection of the response, including the C1-bend (begins arrow 1), counter bend (begins arrow 2), and subsequent swimming (begins arrow 3). Active Mauthner: predicted Mauthner that fired to drive right or left escape. C) Percentage of twenty acoustic stimuli that elicited any response (fast escape or response with >17 ms latency, with latency measured as first head movement following stimulus delivery). Each dot represents one fish. n = 126 siblings (mix of celsr3+/+ and celsr3-/+), 154 celsr3-/- mutants. D) For larvae from (C), percentage of twenty acoustic stimuli that elicited a fast escape (latency <17 ms). 47% of mutant larvae responded within the sibling range (88–100%). E) For larvae from (D) that performed at least six fast escapes, initial C1-bend direction ratio (right versus left) for individual larva. Arrows indicate larvae that only turn one direction (“100% biased”; 24% of mutants). n = 126 siblings, 133 mutants.

https://doi.org/10.1371/journal.pgen.1011415.g001

Neural circuit formation requires coordination of numerous developmental steps. Neurons must properly differentiate, migrate, if necessary, to their correct location, extend axons, and form synapses with appropriate targets. Directing all these processes are molecular pathways and proteins. Here, we focus on the planar cell polarity (PCP) pathway, which plays numerous roles in nervous system development. These roles include tissue polarization during neural tube closure [17–19], polarization of interacting cells during neuron migration [20–23], axon guidance [24,25], dendrite development [26–28], and synapse formation [29]. These processes do not necessarily require all PCP core proteins or follow the classic epithelial patterning model of protein interactions [30]. Celsrs (Cadherin EGF LAG seven-pass G-type receptors) are core PCP pathway components homologous to Drosophila Flamingo/Starry night [31–33]. Mammals have three CELSR genes (CELSR1, CELSR2, and CELSR3) [34], while zebrafish have four (celsr1a, celsr1b, celsr2, and celsr3) [35]. Results from numerous studies suggest CELSR1 primarily regulates epithelial PCP, while CELSR2 and CELSR3 have both overlapping and unique roles in neural development [33,36,37]. Celsr2 and Celsr3 complex with Frizzled receptors, particularly Frizzled/Fzd3, which regulate neurodevelopment through interaction with signaling molecules including Wnts and Ephrin receptors [24,33,38,39]. The precise roles PCP-pathway proteins play in neural development, and whether these mechanisms mirror those of PCP-dependent epithelial patterning, are complex questions not yet fully understood.

Our interest in the roles of PCP proteins in startle circuit development began with a fortuitous observation: while studying the role of celsr3 in axon regeneration [40], we discovered uninjured celsr3 mutant larvae display marked escape direction bias in response to repeated acoustic stimuli. We demonstrate here that this behavioral deficit in celsr3 mutants is due to Celsr3’s role in axon growth and guidance of both Mauthner cells and spiral fiber neurons. Additionally, we find that Celsr2 and Fzd3a function with Celsr3 to regulate startle circuit development and subsequent behavior. Our data also reveals diverging roles for Celsr3 and Celsr2 in hindbrain development, as Celsr2 plays a well-documented critical role in facial branchiomotor neuron migration [20,22,41] while we find that Celsr3 is dispensable for this process. Combined, our results reveal partially overlapping and unique roles for Celsrs in hindbrain development and a clear role for the PCP pathway in directing assembly of pre- and post-synaptic partners within the acoustic startle circuit.

Results

Celsr3 is required for unbiased acoustic startle escapes

Behavioral output is a powerful indicator of circuit function and integrity. This is especially true for the acoustic startle escape response in zebrafish (Fig 1A and 1B), as neurons comprising the underlying circuit have been well-characterized, and the behavioral phenotypes following manipulation (ablation, silencing, and/or activation) of many of these neurons are known [14,16,42–46]. In response to intense and sudden acoustic stimuli, 5 dpf larvae have a high response frequency (Fig 1C) and primarily respond with fast escapes (Fig 1D). Directional acoustic stimuli elicit escapes away from the startling stimulus, driven by the neural circuitry underlying the response (Fig 1A and 1B). Following a non-directional acoustic stimulus, as in our assay, larvae perform an escape maneuver to the left or to the right, following the initial Mauthner-dependent C1-bend direction, without bias (“unbiased escapes”). Over multiple stimuli, wild type and celsr3 sibling larvae on average perform equal numbers of escapes to the left and to the right (measured as ratio of right C1-bends versus left C1-bends for individual larvae; Fig 1E). In contrast, the C1-bend direction ratio is unbalanced in celsr3 mutant larvae, with 29% only initiating escapes in one direction over multiple stimuli (“biased escapes,” Fig 1E). Additionally, celsr3 mutants on average exhibit a reduction in total acoustic response frequency (includes fast escapes and slower responses; Fig 1C) and fast escape frequency (Fig 1D), though 47% of celsr3 mutant larvae react at sibling levels (within one standard deviation from sibling average, i.e. 88–100% fast escape frequency) (Fig 1D). The reduction in response frequency may be caused by defects in acoustic acuity upstream of the Mauthner circuit or by additional developmental defects, as celsr3 mutants lack a swim bladder, exhibit reduced spontaneous movements (sibling bouts/min average = 72, n = 27; mutant bouts/min average = 31, n = 21; p-value<0.001), and are not viable past ~11 dpf (celsr3fh339 is a presumed null allele [40]). Nonetheless, the biased escapes we observe in celsr3 mutants suggest celsr3 has a specific role in startle circuit development and/or function.

Celsr3 is required for Mauthner cell development

We first assessed Celsr3’s role in Mauthner cell development, since the initial startle escape turn direction is dependent on whether the left or right Mauthner cell fires (Fig 1A and 1B). Given that celsr3 mutants were more likely to perform escapes in one direction, we hypothesized that celsr3 mutants may have deficits in Mauthner cell development and/or connectivity. To visualize Mauthner cells, we utilized the Tol-056 transgenic line [47], which expresses cytosolic GFP in Mauthners, and the neurofilament antibody 3A10, which labels Mauthner axons. At 5 dpf, 100% of wild type and celsr3 sibling larvae have two Mauthner cells (Fig 2A and 2C), while ~10% of celsr3 mutant larvae lack one Mauthner cell (Fig 2B and 2C). We do not observe any severe morphological defects in Mauthner somas or hindbrain axons in celsr3 mutants with one or two Mauthner cells (Fig 2B), and the major lateral and ventral dendrites, which receive sensory input [15,48–50], are intact. We next assessed Mauthner axons in the spinal cord in celsr3 mutants. During normal development, each Mauthner projects a single axon that crosses the midline in the hindbrain and extends posteriorly into the spinal cord (Fig 2D). By 5 dpf, the Mauthner axon measures ~2700μM in total length and extends along the entire spinal cord, which is surrounded by repeating muscle segments. We used these segments as landmarks to quantify Mauthner axon length. At 5 dpf, Mauthner axons in sibling larvae extend at least to segment 30 (Fig 2D, 2F and 2H). In contrast, Mauthner axons in celsr3 mutants are significantly shorter, as 50% of axons do not reach segment 30 (Fig 2E, 2G and 2H), and some were as short as segment 15 (Fig 2H). These results demonstrate celsr3 is critical for proper Mauthner development and axon growth.

thumbnail
Download:
Fig 2. Celsr3 is required for Mauthner development.

A,B) Tol-056 transgene expression of cytosolic GFP in Mauthner neurons in celsr3 sibling (A) and mutant (B) larvae. Mauthner axons are labeled with α-3A10 (neurofilament antibody) staining. All images and quantifications in figure are of 5 dpf larvae. Scale bar = 20μM, Z-projection. C) Mauthner cell number in celsr3 siblings (100% 2 M-cells; n = 51) and celsr3 mutants (12% 1 M-cell, 88% 2 M-cells; n = 67). D,E) Mauthner somas and axons in Tol-056 celsr3 sibling (D) and mutant (E). The Mauthner/spinal cord is pseudo-colored green for ease of visualization, while non-Mauthner/spinal cord GFP expression is magenta. Only z-frames with the Mauthner axon in focus are included. Arrows indicate where the Mauthner axons end and corresponding muscle segments, numbered anterior to posterior. Scale bar = 500μM, Z-projection. F,G) Magnified Mauthner axon ends in Tol-056 celsr3 sibling (F) and mutant (G). The top axon in (G) continues to segment 30. Scale bar = 50 μM, Z-projection. H) Mauthner axon lengths in celsr3 siblings and mutants measured by muscle segment containing the Mauthner axon end. Average sibling = segment 32 (n = 22 axons), mutant = segment 29 (n = 66 axons). I) Escape C1-bend ratio for celsr3 siblings and mutants that performed at least six fast escapes in response to 20 acoustic stimuli, normalized so all fish turning only one direction, left or right, appear at the same y position. n = 76 siblings, 6 mutants with one Mauthner (1M), 109 mutants with two Mauthners (2M). J) Escape frequency (Escape%) for individual Mauthners was calculated using total Escape% and turn bias ratio for individual fish; left C1-bend escape was assigned to right Mauthner, right C1-bend escape assigned to left Mauthner. n = 20 sibling Mauthners, 253 celsr3 mutant Mauthners. Points outlined in black indicate celsr3 mutant Mauthners from larvae with a single Mauthner (1M). Individual Mauthner escape% and axon length are not significantly correlated (simple linear regression, p = 0.63) in celsr3 mutants.

https://doi.org/10.1371/journal.pgen.1011415.g002

We next asked whether Mauthner defects in celsr3 mutants are correlated with biased escapes. Similar to previous work in larvae with only a single Mauthner following ablation of the second Mauthner [14,16], celsr3 mutant larvae with a single Mauthner cell initiate escapes to only one side, contralateral to the remaining Mauthner (Fig 2I). Unexpectedly, celsr3 mutant larvae with two Mauthner cells also exhibit biased escapes (Fig 2I). To explore the possibility that Mauthner axon length might relate to Mauthner activity, we calculated escape frequency for individual Mauthners. We used the escape frequency for an individual larva and that larva’s turn bias to calculate response frequency for individual Mauthners, with the assumption that left escapes were driven by right Mauthner firing, and vice versa. Mauthner escape frequency and Mauthner axon length do not correlate, and there is no clear correlation between mutants with single Mauthners and Mauthner axon length or escape% (Fig 2J). We conclude that shortened Mauthner axons do not affect escape frequency or direction bias in celsr3 mutants. In contrast, loss of Mauthner cells does contribute to biased escapes, though 81% of celsr3 mutants with completely biased escapes have two Mauthners. Therefore, we predicted that celsr3 is required in additional startle circuit neurons critical for unbiased escapes.

Celsr3 is required for spiral fiber axon guidance

Mauthner cells receive excitatory input from the eighth cranial nerve and from spiral fiber neurons that are activated by acoustic stimuli [14]. Spiral fibers form mixed electrical and glutamatergic synapses on the Mauthner axon initial segment (AIS) [15,51,52]. Excitatory spiral fiber synapses and competing feedback inhibitory synapses at the Mauthner AIS are surrounded by astrocyte-like glial cells, which all together comprise the Mauthner axon cap [53] (Fig 3A–3J). To visualize spiral fiber somas and axons, we utilized the transgenic reporters hcrt:Gal4, UAS:Kaede (hcrt::Kaede) [14], which stochastically labels spiral fiber neurons (Fig 3D and 3I), and j1229a [54], which expresses cytosolic GFP in various neuron populations, including spiral fiber and Mauthner neurons. In wild type larvae, approximately ten spiral fiber soma per brain hemisphere are organized in two bilateral groups rostroventral to the Mauthner cells [52] (Fig 3A–3D). These spiral fiber neurons project axons across the midline, then turn posteriorly to establish synapses at the Mauthner AIS (Fig 3A, 3B and 3D). At the axon cap (Fig 3F–3J), a nuclei-free region (Fig 3G) is surrounded by S100B+ [55] glia (Fig 3H) and contains spiral fiber axons (Fig 3I) wrapping around the Mauthner AIS (Fig 3J). In celsr3 mutants, while some spiral fiber axons project correctly to the Mauthner AIS, we also observe misprojected spiral fiber axons that form ectopic bundles reminiscent of those at the endogenous axon cap (Fig 3K and 3L). To further characterize spiral fiber defects in celsr3 mutants, we utilized an antibody against the potassium channel subunit Kv1.1, which is highly expressed in spiral fiber axons [56], particularly at the axon cap (Fig 3M). We find that Kv1.1 localizes to the axon cap in all celsr3 mutants (n = 48), suggesting at least some spiral fibers axons have reached the Mauthner (Fig 3N). In addition to this wild-type-like innervation pattern, we also observe ectopic accumulations of Kv1.1 rostroventral to the Mauthner AIS along the path of spiral fiber projections (Fig 3N), indicative of misprojecting spiral fiber axons. These ectopic accumulations, which we term “ectopic caps” for their resemblance to spiral fiber axons and Kv1.1 at endogenous Mauthner axon caps, are present in celsr3 mutant brain hemispheres with or without a corresponding Mauthner cell (Fig 3O and 3P), most commonly at the position of spiral fiber somas along the anterior-posterior axis (Fig 3Q and 3R). These ectopic caps also co-localize with the gap junction protein Connexin 35/Cx35 (Fig 4B–4D) which marks spiral fiber synapses with the Mauthner [57], indicating misguided axons form synapses with inappropriate targets. These data indicate celsr3 is required for spiral fiber axon guidance to the Mauthner AIS.

thumbnail
Download:
Fig 3. Celsr3 is required for spiral fiber axon guidance and glial localization to the axon cap.

A) The axon cap (dotted red box in circuit diagram) is comprised of excitatory spiral fiber axons (blue) and feedback inhibitory axons (pink) synapsing with the Mauthner (green) axon initial segment (AIS), surrounded by astrocyte-like glia (dotted line). B-E) A wild type larva with j1229a transgene expression of cytosolic GFP in Mauthner and spiral fiber neurons (B,C), hcrt:Gal4, UAS:Kaede/hcrt::Kaede transgene expression in a subset of spiral fibers (B,D; protein mostly converted from green to red), and α-3A10 staining (B,E). White brackets in (B) indicate spiral fiber somas. Yellow brackets in (D) indicate spiral fiber axons at the axon cap. White arrows in (E) indicate spiral fiber axons at midline crossing. All images and quantifications in figure are of 5 dpf larvae. Scale bar = 20μM, Z-projection. F-J) Magnification of box in (B) showing both Mauthner axon caps with TO-PRO-3 labeling nuclei (F,G), α-S100B labeling glia (F,H), Kaede labeling spiral fiber axons (F,I), and α-3A10 labeling Mauthner axons (F,J). Yellow arrowheads in (J) indicate Mauthner AISs. Scale bar = 20μM, single Z-frame. K,L) celsr3 mutant larva with Tol-056:GFP (K) and hcrt::Kaede (K,L) transgene expression. White asterisks indicate spiral fiber soma (note: only a subset of spiral fiber neurons are labeled due to variegated transgene expression). Yellow brackets indicate normal spiral fiber axon targeting to Mauthner axon cap. Red arrow indicates inappropriate spiral fiber axon targeting. Scale bar = 20μM, Z-projection. M,N) Mauthner axon cap region of sibling (M) and celsr3 mutant (N) Tol-056:GFP larvae. Kv1.1 localizes to Mauthner axon caps (yellow brackets) that spiral fiber axons target and accumulates at ectopic positions (red arrows) where spiral fiber axons frequently misproject. Scale bar = 20μM, Z-projection (dorsal range included for posterior Mauthner axon cap region and ventral range included for anterior spiral fiber axon misprojections, for clarity). O) Schematic for quantifying spiral fiber misprojections, measured as accumulation of Kv1.1 in ectopic “axon caps,” as in (N) (red arrows), within 60μM anterior to Mauthner AIS and 60μM lateral to midline (boundaries of spiral fiber soma). Left brain hemisphere is normal, with 0 ectopic caps, while right brain hemisphere has 1 ectopic spiral fiber cluster. For quantification in (P), this example would be counted as 0 (for left) and 1 (for right). P) Quantification of “ectopic caps,” presumed to represent spiral fiber misprojections, assessed by counting numbers of large Kv1.1 accumulation (>5μM diameter) rostroventral to position of normal axon cap, as in (O). For each larva, the left and right hindbrain were counted independently (two “hemispheres” per larva). In addition to ectopic Kv1 accumulation, all sibling and celsr3 mutant hemispheres with a Mauthner present had Kv1.1 accumulation at the endogenous axon cap region surrounding the Mauthner AIS. celsr3 mutant brain hemispheres where Mauthner neurons were absent only had ectopic caps. n = 10 sibling hemispheres, 58 celsr3 mutant hemispheres with Mauthners, 9 celsr3 mutant hemispheres without Mauthners. Q) Schematic for quantifying anterior-posterior position of ectopic caps quantified in (P). 1 = positioned at Mauthner AIS, 2 = <20μM from correct position, 3 = posterior to spiral fiber soma, 4 = at posterior spiral fiber midline crossing, 5 = at anterior spiral fiber midline crossing. For example, ectopic cap in (O) would be counted as 4. When the Mauthner neuron was absent, additional axon tracts labeled by Kv1.1 were used to identify where correct axon cap positioning was expected. R) Quantification of anterior-posterior position of ectopic caps from (P). For brain hemispheres with two ectopic caps, each position was counted separately. n = 63 celsr3 mutant ectopic caps in hemispheres with Mauthners, 17 celsr3 mutant ectopic caps in hemispheres without Mauthners. S-X) Sibling (S-U) and celsr3 mutant (V-X) larvae with Glast:myrGFP-p2A-H2AmCherry (membrane GFP, nuclear mCherry) transgene labeling astrocytes, as well as Tol-056:GFP and hcrt::Kaede (partially converted from green to red) transgenes. Green channel (myrGFP, GFP, and Kaede) in S,V,T,W; red channel (H2AmCherry, Kaede) in S,V,U,X. White asterisks indicate spiral fiber soma (note: only a subset of spiral fiber neurons are labeled due to variegated transgene expression). Yellow brackets indicate endogenous Mauthner axon cap, red brackets indicate ectopic localization of glia, red arrow indicates misprojected spiral fiber axons, and yellow arrowheads indicate two example glia localized correctly to the endogenous axon cap. Scale bar = 20μM, Z-projection. Y) Quantification of percent of sibling or celsr3 mutant larvae with zero, one (left or right), or two (left and right) anterior ectopic glial clusters. All larvae for quantification had two Mauthners. n = 16 siblings, 15 celsr3 mutants. Z) For larvae in (Y), glia on one side of the hindbrain (left or right hemisphere) were categorized as correctly localized to only the endogenous axon cap at the Mauthner AIS, localized to both the endogenous axon cap and an anterior ectopic location, localized only to an ectopic location, or not observed/missing. n = 32 sibling hemispheres, 30 celsr3 mutant hemispheres.

https://doi.org/10.1371/journal.pgen.1011415.g003

Previous work has shown that astrocyte-like glia surrounding spiral fiber and inhibitory synapses arrive at the axon cap after spiral fiber axons appear [15,58], yet the signals guiding them to this location are unknown. We utilized a transgenic line labeling astrocytes, slc1a3b/Glast:myrGFP-p2A-H2AmCherry [59], to visualize axon cap glia (Fig 3S–3X). In celsr3 mutants, glia co-localize with spiral fiber axons at the bonafide Mauthner axon cap but also localize to rostroventral positions and with misprojected spiral fiber axons (Fig 3V–3Z). The glia mislocalization phenotype is variable in celsr3 mutants, similar to misguided spiral fiber axons, and we observed at least one ectopic glial cluster in the region of the Mauthner and spiral fiber somas in all celsr3 mutants imaged (Fig 3Y). While we always observe Kv1.1 localization to the Mauthner axon cap in celsr3 mutants, we failed to detect Glast+ glia at the axon cap in ~45% of celsr3 mutant brain hemispheres with Mauthner cells (Fig 3Z). This suggests spiral fiber axons may play a role in attracting glia to the axon cap, such that misguided spiral fiber axons can redirect glia away from their proper position. Thus, celsr3 is required for glial guidance to the axon cap, either indirectly by promoting spiral fiber axon growth and guidance or directly by functioning in axon cap glia guidance.

Spiral fiber input at the axon cap is correlated with startle turn bias

Spiral fiber input is required for a robust, fast startle escape [14]. Larvae with unilaterally ablated spiral fiber soma only execute fast escapes in one direction [14], similar to the defects observed in celsr3 mutants. We wondered if stochastic defective excitatory input to the left and right Mauthners correlates with and might be causative of biased escapes. In celsr3 siblings, Kv1.1 and Cx35 localize to both the left and right Mauthner axon caps (Fig 4A). Similarly, in mutants with unbiased C1-bend direction ratio (40–60% right escapes), Kv1.1 and Cx35 are symmetrically localized to the left and right axon caps (n = 9/12 larvae with symmetrical Kv1.1 at caps; Fig 4B). In contrast, in celsr3 mutants that are 100% left-biased, indicating only the right Mauthner fires, the left axon cap has reduced Kv1.1 and Cx35 localization (n = 8/8 larvae with less Kv1.1 at left versus right cap; Fig 4C). Similarly, mutant larvae that are 100% right-biased have diminished localization to the right axon cap (n = 10/11 larvae with less Kv1.1 at right versus left caps; Fig 4D). We interpret from this data that spiral fiber axon misguidance in celsr3 mutants stochastically leads to asymmetric input to the left and right Mauthner cells such that the Mauthner with the most spiral fiber input is preferentially activated, resulting in escapes only initiated in one direction. Thus, celsr3 is required for proper spiral fiber targeting and maintaining balance within the startle circuit that enables animals to appropriately adjust their response to the spatial origin of startling stimuli.

thumbnail
Download:
Fig 4. Asymmetric spiral fiber input to Mauthner neurons in celsr3 mutants correlates with biased escapes.

A-D) Region of hindbrain with Mauthner and spiral fiber neurons from 5 dpf Tol-056:GFP larvae in (A) siblings with unbiased escapes (40–60% right/left C1-bend direction ratio; all categories >6 escapes over 20 stimuli), (B) celsr3 mutants with unbiased escapes, (C) celsr3 mutants that only made left C1-bends (right Mauthner predicted active), and (D) celsr3 mutants that only made right C1-bends (left Mauthner predicted active). Composite images are Z-projections with ventral range for anterior portion and dorsal range for posterior portion. Kv1.1 (left and middle panels) localizes to endogenous axon caps (brackets) and ectopic caps (red arrows in B-D) presumed to represent spiral fiber misprojections. α-Cx35 (left and right panels) labels spiral fiber synapses at endogenous (A-D) and ectopic (B-D) caps. White dashed brackets in (C,D) indicate the axon cap with reduced Kv1.1/Cx35 signal, which corresponds to the inactive Mauthner, based on escape bias. Scale bar = 20 μM.

https://doi.org/10.1371/journal.pgen.1011415.g004

celsr2 partially compensates for loss of celsr3 in Mauthner development

In zebrafish, the Celsr gene family includes celsr1a, celsr1b, celsr2, and celsr3, which have overlapping and unique expression patterns during nervous system development [35]. Previous studies have shown that Celsr3 and Celsr2 have partially overlapping roles in some neurodevelopmental contexts [23,37]. We hypothesized that the variable Mauthner phenotypes in celsr3 mutants could be due to compensation by celsr2. celsr2 mutants do not have reduced Mauthner cell numbers at 5 dpf (Fig 5A–5C and 5G), suggesting this gene is not required for Mauthner development. Similarly, celsr2 mutants do not exhibit reduced escape frequencies or biased escapes (Fig 5H–5J). However, loss of a single copy of celsr2 in homozygous celsr3 mutants strongly enhances the Mauthner cell loss phenotype (Fig 5G), and in >90% of celsr3;celsr2 double mutant larvae, both Mauthner cells are absent at 5 dpf (Fig 5D–5G). Spiral fiber neurons, however, are still present in celsr3;celsr2 double mutants (Fig 5D–5F). Additionally, we find that celsr3;celsr2 mutants display a marked reduction of acoustic stimulus-induced responses (Fig 5H) and escapes (Fig 5I). This reduction is likely not due to the absence of Mauthners, as wild type larvae with bilateral Mauthner ablations perform Mauthner-independent, Mauthner-homolog mediated escapes with longer latencies than Mauthner-dependent escapes [60]. The occasional fast (<17 ms latency) escapes we observe celsr3;celsr2 mutants lacking both Mauthners have a longer latency than those from celsr3;celsr2 mutants with single Mauthners (Fig 5K). This indicates that Mauthner-homolog mediated escapes are at least partially intact in celsr3;celsr2 mutants in lacking both Mauthners, and single Mauthner cells in celsr3;celsr2 mutants are functional, as the escape latency in these larvae is within the range of Mauthner-dependent escapes. Combined, these data indicate that while celsr2 is dispensable for normal Mauthner development, it can partially compensate for loss of celsr3.

thumbnail
Download:
Fig 5. celsr2 compensates for loss of celsr3 during Mauthner development.

A-F) j1229a:GFP transgene expression (A,B,D,E) labels Mauthner (A,B) and spiral fiber (asterisks; A,B,D,E) neurons in celsr2-/- (A-C) and celsr3-/-;celsr2-/- (D-F) larvae. α-3A10 (A,C,D,F) labels spiral fiber axons at midline crossing (white arrowheads) and Mauthner axons (A,C). White dashed brackets (D-F) indicate positions where Mauthner neurons are absent. All images and quantifications in figure are of 5 dpf larvae. Scale bar = 20μM, Z-projection. G) Mauthner cell number in celsr3 sib;celsr2-/- (100% 2 M-Cells; n = 44), celsr3-/-;celsr2+/+ (5% 1 M-cell, 95% 2 M-cells; n = 20), celsr3-/-;celsr2+/- (20% 0 M-cell, 39% 1 M-cell, 41% 2 M-cells; n = 41), and celsr3-/-;celsr2-/- (91% 0 M-cell, 9% 1 M-cell; n = 22). p-values calculated with Chi-square tests. H) Percentage of twenty acoustic stimuli that elicited any response for celsr3 sib;celsr2-/- (n = 28) and celsr3-/-;celsr2-/- (n = 17) larvae. I) For larvae from (H), percentage of twenty acoustic stimuli that elicited a fast escape (latency <17 ms). Points outlined in black indicate celsr3-/-;celsr2-/- larvae with one Mauthner (5/17), while the remaining celsr3-/-;celsr2-/- larvae had zero Mauthners. J) For celsr3 sib;celsr2-/- larvae from (I) that performed at least six escapes (100% of larvae), escape C1-bend ratio, normalized so all fish turning only one direction, left or right, appear at the same y position. K) Latency (time to first head movement) for individual fast escapes in celsr3-/-;celsr2-/- larvae with one (1M) or zero (0M) Mauthners. n = 19 responses 1M, 12 responses 0M.

https://doi.org/10.1371/journal.pgen.1011415.g005

Celsr3 is dispensable for facial branchiomotor neuron migration

Our findings that Celsr3 is required for development of two distinct neuronal populations, Mauthners and spiral fibers, prompted us to investigate Celsr3’s role in hindbrain neurons outside the startle circuit. Other PCP proteins, including Celsr2, are required in the hindbrains of both zebrafish [20] and mice [23] for facial branchiomotor neuron (FBMN) migration. The cell bodies of islet-1/isl1+ FBMNs originate in hindbrain rhombomere 4 (r4), where Mauthner soma reside. By 48 hours post-fertilization (hpf), FBMNs have migrated to rhombomere 6 (r6) (Fig 6A) [61,62]. In zebrafish celsr2 mutants, FBMNs fail to migrate and remain in r5 at 48 hpf [20] (Fig 6B and 6E). In contrast, we find that by 48 hpf, FBMNs in celsr3 mutants have appropriately migrated to r6 (Fig 6C and 6E). In addition, we fail to detect differences in the severity of migration deficits between celsr2 single mutants and celsr3;celsr2 double mutants (Fig 6D and 6E). Importantly, the lack of detectable differences between celsr2 mutants and celsr3;celsr2 mutants is unlikely to result from a ceiling effect in severity of migration as celsr2 mutant larvae injected with a morpholino against celsr1a/b have a more severe migration phenotype than celsr2 mutants alone [20]. Thus, while we cannot exclude a role for celsr3 in FBMNs at later stages in development, as reported in mouse [23], our data suggests that in zebrafish, celsr3 is dispensable for FBMN migration.

thumbnail
Download:
Fig 6. FBMN migration is independent of celsr3.

A-D) isl1:GFP transgene expression of cytosolic GFP in 48 hours post-fertilization (hpf) facial branchiomotor neurons (FBMNs) in sibling (A), celsr2-/- (B), celsr3-/- (C), and celsr3-/-;celsr2-/- (D) embryos. α-3A10 labels Mauthner somas and axons (left panels). Approximate rhombomere boundaries (r4, r5, r6) are indicated with brackets (right panels). Asterisks in (C,D) denote presumptive dying Mauthners based on soma morphology. Scale bars = 20 μM, Z-projection. E) Percent of larvae with full (e.g. A,C), intermediate (e.g. D), and minimal (e.g. B) FBMN migration. n = 75 sib, 32 celsr3-/-, 11 celsr2-/-, 10 celsr3-/-;celsr2-/-. p-value calculated with Fisher’s exact test.

https://doi.org/10.1371/journal.pgen.1011415.g006

Regional disruption of Celsr/Fzd signaling reveals a direct role for Celsr3 in spiral fiber axon guidance

Our results reveal that Celsr3 is required for both Mauthner development and spiral fiber axon guidance. One possible mechanism is that spiral fiber axon misguidance in mutants is a downstream effect of Mauthner cell developmental defects, rather than a direct role for Celsr3 in spiral fiber axon guidance. To distinguish between an indirect and direct role for Celsr3 in spiral fiber axon guidance, we sought to disrupt Celsr signaling specifically in rhombomere four (r4), where the Mauthner cell develops, and not r3, where spiral fiber soma reside. To achieve this, we utilized the Celsr3 binding partner, Frizzled 3a (Fzd3a) [24,39], and regionally expressed a C-terminally truncated Fzd3a (Fzd3aΔC) that acts as a dominant negative [20,22]. Zebrafish have two fzd3 genes, fzd3a and fzd3b, and fzd3a is the predominantly expressed homolog during embryogenesis [63]. fzd3a functions in a number of neurodevelopmental processes [20,22,64–66], so we first asked if fzd3a is involved in startle circuit development. Similar to celsr2 mutants, fzd3a single mutant larvae display normal startle behavior and Mauthner cell number (Fig 7A–7C). In contrast, celsr3;fzd3a mutant larvae display similar, although milder, phenotypes compared to those we observe in celsr3;celsr2 mutant larvae, with ~69% of celsr3;fzd3a mutants lacking one or both Mauthner cells (Fig 7A–7C). These data indicate Celsrs and Fzds work together to promote Mauthner development, and we predict disruption of Fzd3a function through expression of Fzd3aΔC will affect this Celsr/Fzd pathway.

thumbnail
Download:
Fig 7.

Spiral fiber axon guidance is partially independent from Mauthner development A) Quantification of percent of twenty acoustic stimuli that resulted in a fast escape (latency <17 ms). All images and quantifications in figure are of 5 dpf larvae. n = 55 celsr3 siblings, 51 celsr3 sib;fzd3a-/-, 56 celsr3-/-, 92 celsr3-/-;fzd3a-/+, 45 celsr3-/-;fzd3a-/-. B) Mauthner cell number in larvae from (A). n = 53 celsr3 sib;fzd3a-/- (100% 2 M-cells), 55 celsr3-/- (15% 1 M-cell, 85% 2 M-cells), 86 celsr3-/-;fzd3a+/- (7% 1 M-cell, 93% 2 M-cells), 45 celsr3-/-;fzd3a-/- (36% 0 M-cells, 33% 1 M-cell, 31% 2 M-cells). C) Escape C1-bend ratio for larvae from (A) with two Mauthners (B) that performed at least six fast escapes in response to 20 acoustic stimuli, normalized so all fish turning only one direction, left or right, appear at the same y position. n = 53 celsr3 siblings, 48 celsr3 sib;fzd3a-/-, 41 celsr3-/-, 71 celsr3-/-;fzd3a-/+, 10 celsr3-/-;fzd3a-/-. D) Mauthner cell number in control (no hoxb1a:Gal4, UAS:fzd3aΔC/hoxb1a::fzd3aΔC transgenes) and hoxb1a::fzd3aΔC larvae. n = 28 control (100% 2 M-cells), 52 hoxb1a::fzd3aΔC (12% 0 M-cells, 27% 1 M-cell, 62% 2 M-cells). E) Quantification of percent of twenty acoustic stimuli that resulted in a fast escape (latency <17 ms). n = 30 control, 65 hoxb1a::fzd3aΔC. F) Escape C1-bend ratio for larvae from (E) with two Mauthners that performed at least six fast escapes in response to 20 acoustic stimuli, normalized so all fish turning only one direction, left or right, appear at the same y position. n = 23 control, 27 hoxb1a::fzd3aΔC. G,H) j1229a:GFP labels Mauthner cells and spiral fiber neurons in hoxb1a::fzd3aΔC larvae while α-S100B labels glia. Yellow brackets indicate correctly positioned GFP+ spiral fiber axons at the Mauthner AIS, while white dashed brackets indicate displaced spiral fiber axon caps, due to either an absent Mauthner (G) or mispositioned Mauthner AIS (H). Yellow arrowheads in (H) denote aberrant path of left Mauthner axon, which continues posteriorly out of the Z-range displayed. Scale bars = 20 μM, Z-projection (ventral anterior range and dorsal posterior range for GFP, for clarity; division marked by dotted line). I) Quantification of “ectopic caps,” as in Fig 3O and 3P. In brain hemispheres with no Mauthners, any large Kv1.1 accumulations were considered “ectopic” since no endogenous Mauthner AIS was present. n = 16 control hemispheres, 60 hoxb1a::fzd3aΔC hemispheres with Mauthners, 32 hemispheres without Mauthners. J) Quantification of anterior-posterior position of ectopic caps from (K), as in Fig 3Q and 3R. n = 32 hoxb1a::fzd3aΔC ectopic caps in hemispheres without Mauthners.

https://doi.org/10.1371/journal.pgen.1011415.g007

We next used Tg(hoxb1a:Gal4) [67] to drive UAS:fzd3aΔC expression throughout r4 from ~10–48 hpf [22] and assessed whether Mauthner and spiral fiber neurons were disrupted. By 5 dpf, 39% of hoxb1a::fzd3aΔC larvae are missing one or both Mauthner cells (Fig 7D and 7G). As this phenotype was not observed in fzd3a single mutants, we hypothesize from these results that the dominant negative Fzd3a protein disrupts PCP signaling by preventing Celsr from acting with other Fzd proteins, such as Fzd3b. In hoxb1a::fzd3aΔC larvae with two Mauthners, we did not detect defects in escape frequency (Fig 7E) or turn bias (Fig 7F), suggesting that spiral fiber neurons are functional. In the majority of hoxb1a::fzd3aΔC brain hemisphere with a Mauthner, Kv1.1+ ectopic caps were absent (n = 59/60, quantified in Fig 7I), suggesting spiral fiber axons were not dramatically misguided as in celsr3 mutants. In the rare hoxb1a::fzd3aΔC larvae with an ectopic or misplaced axon cap (n = 4/60), the Mauthner soma was morphologically abnormal and lateral to its correct position (Fig 7H). When the Mauthner soma is incorrectly positioned and the spiral fiber axon cap is displaced, glia associate with the spiral fiber axon cap and not the Mauthner AIS (Fig 7G). We do not observe ectopically localized glia in hoxb1a::fzd3aΔC larvae (n = 10). Additionally, in hoxb1a::fzd3aΔC brain hemispheres where the Mauthner was absent, a single “ectopic” cap was observed (Fig 7I and 7J), in contrast to celsr3 mutants lacking a Mauthner cell (Fig 3P). This ectopic cap also forms closer to the correct position in hoxb1a::fzd3aΔC larvae (Fig 7J) than in celsr3 mutant larvae (Fig 3R) and is associated with glia (Fig 7H). Together, these data indicate that spiral fiber defects present in celsr3 mutants are unlikely a mere consequence of Mauthner cell defects. Instead, we conclude that celsr3 is directly required for spiral fiber axon guidance.

The PCP pathway promotes Mauthner anterior-posterior axon guidance

To identify whether the Celsr/Fzd pathway is required for the initial development or subsequent maintenance of Mauthner cells, we examined hoxb1a::fzd3aΔC embryos at 30 hpf. In age matched wild type animals, Mauthner cells have differentiated and their axons have extended across the midline and grown posteriorly into the spinal cord [68]. At 5 dpf, 39% of hoxb1a::fzd3aΔC larvae are missing one or both Mauthner cells (Fig 7D), while at 30 hpf, all hoxb1a::fzd3aΔC embryos have two Mauthner cells (Fig 8A–8C). However, many Mauthner axons in these embryos have grown incorrectly in the anterior direction (Fig 8A–8C). Since all fish have two Mauthners at 30 hpf, but only 39% still have two Mauthners at 5 dpf, we predict some Mauthner cells die between 30 hpf and 5 dpf due to axon guidance defects. We hypothesize that Mauthner cells that project axons anteriorly and fail to correct likely die, as we do not observe Mauthner axons growing solely anteriorly at 5 dpf. Consistent with this hypothesis, the ratio of 30 hpf embryos with two correct posteriorly growing axons versus one correct posterior and one incorrect anterior versus two incorrectly growing anterior is similar to the ratios of two versus one versus zero Mauthners in 5 dpf larvae (Figs 7D and 8C). We conclude that expression of Fzd3aΔC in r4 does not disrupt Mauthner specification but instead affects the process of axon growth and guidance, suggesting Celsr/Fzd signaling is required for Mauthner axon guidance.

thumbnail
Download:
Fig 8. The PCP pathway directs A-P axon guidance of Mauthner cells.

A-B) Wild type hindbrain with α-3A10 labeling Mauthner somas and axons. Mauthner axons grow posteriorly, towards the bottom of the image. All images and quantification in figure are from 30 hpf embryos. B) Two representative examples of hoxb1a::fzd3aΔC-GFP transgenic hindbrains with Mauthner somas and axons labeled with α-3A10. Yellow arrowheads indicate axons from left Mauthners and red arrows indicate axons from right Mauthners. In top panels, the left axon grows correctly posteriorly while the right axon grows incorrectly towards the anterior, and in bottom panels, both axons grow incorrectly towards the anterior. Scale bars = 20 μM, Z-projection. C) Quantification of Mauthner axon phenotypes in hoxb1a::fzd3aΔC positive (n = 18) and celsr3-/-;celsr2-/- (n = 19) embryos. D,E) Sibling (D) and celsr3-/-;celsr2-/- (E) embryos with hoxb1a in situ (magenta) labeling hindbrain rhombomere 4 and nefma in situ (green) labeling reticulospinal neurons, including Mauthner cells (arrowheads). Scale bars = 20 μM, Z-projection. F,G) Mauthner soma and axons are labeled with α-3A10 in siblings (F) and celsr3-/-;celsr2-/- (G) embryos. Yellow arrowheads indicate axon from left Mauthner growing correctly towards the spinal cord but displaced laterally. Red arrowheads indicate axon from right Mauthner growing incorrectly towards the anterior. Dotted lines indicate midline and encircle developing ears, for reference. Scale bars = 20 μM, Z-projection.

https://doi.org/10.1371/journal.pgen.1011415.g008

Finally, we asked whether the Mauthner defects in celsr3;celsr2 mutants were similarly correlated with axon guidance defects in early development. At 30 hpf, all celsr3;celsr2 mutants have two Mauthner cells by gene expression (in situ for neurofilament medium chain a/nefma and hoxb1a, Fig 8D and 8E), indicating proper specification, and α-3A10 labeling (Fig 8F and 8G). While Mauthners are present and all Mauthner axons crossed the midline, many axons turned anteriorly rather than posteriorly (Fig 8C and 8G). Unlike hoxb1a::fzd3aΔC, in celsr3;celsr2 mutants, more Mauthner cells are absent at 5 dpf (Fig 5G) than predicted if only Mauthner cells with anteriorly growing axons at 30 hpf die (Fig 8C). This suggests Celsrs promote additional developmental processes, such as progressive axon growth (Fig 2H), that ensure Mauthner cell survival. We conclude that the PCP pathway, particularly Celsr3 and Celsr2, promotes proper anterior-posterior Mauthner axon guidance.

Discussion

Neural circuit assembly during development depends on numerous interacting molecular pathways driving axon targeting and connection of synaptic partners. Whether synapsing pairs of neurons within the same circuit utilize shared pathways to ensure connectivity, and what the consequences on circuit output are for altering but not abolishing these connections, is not fully understood. Here, we uncover a previously unappreciated role for the PCP pathway in the development and function of the acoustic startle hindbrain circuit (Table 1). We find that Celsr3, along with Celsr2 and Fzd3a, are critical for axon growth and guidance of the Mauthner cells and spiral fiber neurons. Additionally, we demonstrate that disruption of spiral fiber guidance that leads to asymmetric input to the paired Mauthner cells has behavioral consequences leading to larvae unable to acutely determine escape direction, which is critical for escaping directional threats. Displaced spiral fibers axons also displace glia associated with spiral fiber-Mauthner synapses, suggesting a role for spiral fiber axons in glial recruitment to their target area, the axon cap. Finally, we discover divergent roles for Celsr3 and Celsr2 for hindbrain development with Celsr3 being the primary player in startle circuit development while Celsr2 plays a compensatory role, and Celsr2 being critical for FBMN migration while Celsr3 seems dispensable. Combined, these results provide compelling evidence for a critical role of the PCP pathway in assembly and connectivity of the acoustic startle circuit.

thumbnail
Download:
Table 1. Summary of hindbrain circuit and behavior phenotypes in PCP mutants.

https://doi.org/10.1371/journal.pgen.1011415.t001

How do shared molecular pathways promote neural circuit connectivity? Cadherins, including those in the Celsr/Flamingo family, play various roles in axon growth and targeting. The clustered protocadherins have been especially well-studied in mammals for roles in self-avoidance and neurite tiling through cis and trans heterophilic and homophilic interactions, as well as in synaptogenesis and neural connectivity through preventing inappropriate connections [69]. Drosophila Flamingo similarly plays roles in neurite tiling [27,28] and preventing ectopic synapse formation [70], as well as axonal fasciculation [71], as does the C. elegans Flamingo homolog FMI-1 [72]. In vertebrates, distinct roles in axon guidance and growth have been described for different Celsrs and different neural populations. In mammals, Celsr3 is required for anterior-posterior axis axon guidance in monoaminergic axons in the brainstem [73,74] and commissural axons in the dorsal spinal cord [75], likely through intercellular interactions between PCP proteins promoting growth cone turning [24,76,77]. Celsr3 has also been implicated in synapse stabilization [29] and inter-growth cone interactions that promote axon organization and progressive axon growth [74]. In future studies, it will be interesting to determine whether Celsr3 homophilic interactions between Mauthner and spiral fibers neurons similarly promote their synaptic connectivity, especially as this connection offers the opportunity to investigate Celsrs in axo-axonic synapse formation.

In addition to divergent roles for different classes of cadherins, there are also distinct functions within the Celsr family. Here, we report that Celsr3 and Celsr2 are differentially required in two developmental processes in the hindbrain: startle circuit assembly and FBMN migration. This may reflect differences in developmental timing for these two processes, as Mauthner axon turning along the anterior-posterior axis occurs several hours before FBMN migration, or differential expression in different neuron populations, though celsr3 and celsr2 expression in the hindbrain largely overlap at this point in development [35]. Alternatively, these distinct requirements may reflect differential functions of the proteins. While Celsrs have similar protein structures [33,35], differences do exist, particularly in the C-terminal intracellular domain. Though the function of these intracellular domains is largely unknown, C-terminally truncated Celsr2 in zebrafish is sequestered in the golgi, and expression of just the C-terminal domain with a membrane tag disrupts convergence/extension in embryos [78]. Additionally, recent work has demonstrated that mammalian CELSRs differentially engage G proteins, which may result in differential G protein-dependent signaling [79]. Future experiments in zebrafish could address whether the differential roles for Celsr3 and Celsr2 in hindbrain development can be attributed to differences in protein sequence.

Intracellular signaling in response to PCP pathway activity is thought to support growth cone turning by promoting cytoskeletal rearrangement [24]. The ligand that induces this PCP activity is not entirely clear, and may differ for different neuron populations, but Wnt [80,81] and ephrinA [39] proteins act in specific axon guidance processes. Both Wnt proteins, which interact with Fzd receptors, and Eph/Ephrin pairs, which can act as a co-receptors for Fzds, are expressed in distinct rhombomere and rhombomere boundaries in the zebrafish hindbrain [82,83], making them potential upstream regulators for Celsr3-mediated Mauthner and spiral fiber axon guidance. Downstream of PCP receptor/ligand interactions, the intracellular signaling that leads to cytoskeletal rearrangements in growth cones likely overlaps with proteins involved in other PCP-dependent processes, including Disheveled proteins [76,77]. Formin proteins, which help regulate actin polymerization, are also implicated as downstream PCP regulators [84,85]. Interestingly, formin2b/fmn2b zebrafish morphants and crispants have reduced spiral fiber commissures and perform reduced short latency escapes [86], similar to the phenotype of wild type fish with ablated spiral fibers [14]. This data suggests studies on spiral fiber axon development may provide insight on molecular and genetic interactions between celsr3 and fmn2b, as well as additional proteins proposed to act during axon guidance. Additionally, the ability to tag proteins and visualize axonal dynamics in vivo, particularly in the uniquely large Mauthner cell, may offer an invaluable in vivo insight to localization of various PCP proteins during the process of growth cone turning and axon growth.

Neural circuit assembly is orchestrated by a myriad of genetic and molecular pathways. Simple and well-characterized sensorimotor circuits like the acoustic startle circuit offer an opportunity to untangle these complex molecular interactions and identify relationships between synapsing pairs of neurons and specific cellular pathways. In addition to identifying the first, to our knowledge, pathway regulating anterior-posterior guidance in Mauthner cells, our work also provides compelling evidence that the zebrafish hindbrain is a prime model for developmental studies on molecular mechanisms for PCP-directed circuit assembly.

Materials and methods

Ethics statement

All animal protocols were approved by the University of Pennsylvania Institutional Animal Care and Use Committee (IACUC), Protocol #805167. Embryos were raised at 29°C on a 14-h:10-h light:dark cycle in E3 media. Mutant and transgenic lines used in this study are as follows:

celsr3fh339 (ZDB-ALT-111010-10)

celsr2sa19274 (ZDB-ALT-230530-16)

fzd3arw689 (ZDB-ALT-061025-2)

Tg(10xUAS:fzd3a-GFP)fh447Tg (ZDB-ALT-170224-9)

Tg(hoxb1a-Mmu.Hbb:GAL4-VP16)um60Tg (ZDB-ALT-120611-5)

Tg(-6.7Tru.Hcrtr2:GAL4-VP16)a150Tg (ZDB-ALT-151028-3)

Tg(UAS-E1b:Kaede)s1999t (ZDB-ALT-070314-1)

Tol056/zf206Et (ZDB-ALT-110217-6)

j1229aGt (ZDB-ALT-100212-17)

Tg(isl1a:GFP)rw0Tg (ZDB-ALT-030919-2)

Tg(slc1a3b:MYRGFP-2A-H2AmCherry)vo80Tg (ZDB-ALT-200915-2)

Behavior testing

Behavioral experiments were performed using 5 dpf larvae and analyzed using Flote software [16]. Larvae were tested in groups in a 6x6 laser cut grid. ~25.9 dB acoustic stimuli were used for acoustic startle experiments, based on larval response rates and previous measurements of the speaker [43]. 20 stimuli were given with a 20 second interstimulus interval. Videos of acoustic startle response were acquired at 1000 frames per second (fps). Before Flote tracking and analysis, NIH ImageJ/Fiji was used to remove background from videos to prevent erroneous measurements from the outline of the wells. Larvae with >2 untracked responses due to tracking errors were excluded from further analysis. Escape% per Mauthner was calculated with the following formula: [Escape% for a left Mauthner] = [Escape%, larva] x [%Right turns, larva] and [Escape% for a right Mauthner] = [Escape%, larva] x [%Left turns, larva]. For example, for a larva with 80% escape response, and 60% right turns, we calculate a left Mauthner escape% of 48% and a right Mauthner escape% of 32%.

Immunohistochemistry

5 dpf larvae were fixed with either 2% trichloroacetic acid (TCA) for 3 hours at room temperature (Kv1.1 staining) or 4% paraformaldehyde (PFA) overnight at 4°C (all other staining) and stained as previously described [87]. Before mounting in Vectashield, TCA-fixed brains and spines were peeled [88], and PFA-fixed brains were dissected. Antibodies and dilutions are as follows:

Anti-3A10 (1/50); 3A10 (supernatant), Developmental Studies Hybridoma Bank

Anti-GFP (1/500); GFP-1010, AvesLabs

Anti-Kv1.1 (1/200); AB5174, MilliporeSigma

Anti-Connexin 35/36 (1/200); MAB3045, MilliporeSigma

Anti-S100B (1/100); HPA015768, MilliporeSigma

Donkey Anti-Chicken 488 (1/500); 703-545-155, Jackson ImmunoResearch

Goat Anti-Mouse Cy3 (1/500); 115-165-003, Jackson ImmunoResearch

Donkey Anti-Mouse 633 (1/500); A21082, Invitrogen

Goat Anti-Mouse 750 (1/500); A21037, Invitrogen

Goat Anti-Rabbit 546 (1/500); A11035, Invitrogen

Goat Anti-Rabbit 633 (1/500); A21071, Invitrogen

TO-PRO-3 stain 642/661 (1/500); T3605, Thermo Fisher

Imaging and quantification

Mauthner soma numbers were quantified in live Tol-056 larvae using an Olympus MVX10 fluorescent stereo microscope. Mauthner axon lengths and glia localization were quantified in live Tol-056 larvae using an Olympus IX81 spinning disk confocal with a 40x lens. Fixed samples for figure images and quantification were acquired on a Zeiss LSM 880 or 980 laser scanning confocal with a 20x lens.

Genotyping

fzd3arw689 mutants were identified by PCR amplification and restriction enzyme digest (Primers:

TTTAGCTGTGAGCCCATCACATTGAGGCTG / CCTCCATGGCGAGAGCAGCAGTCTG; DdeI cuts mutant allele). celsr3fh339 and celsr2sa19274 genotypes were identified using KASP assays (LGC Biosearch Technologies), with the following sequence targets ([wt allelic sequence/mutant allelic sequence]):

celsr3fh339: CGGGAGTTTGCCTTTGCCAAGACACCAACTCAGCCCTTATTAAAAGCCTA[T/A]TCTTATTCTAATGTGGCTTCAGACGCCAATAAATTGCACAAAGCGACAGC

celsr2sa19274:

GTGTGTGTTCTGTGGTCTGCAGATGGTGTGCACTCGGTGTCGGCTCAGTG[T/A]CTGCTGCAGGTCACCGTCATCACGGATGAGATGCTCTCCAACAGCATCAC

Statistics

Statistical analyses were performed using GraphPad Prism (Dotmatics). P-values were calculated using non-parametric Mann-Whitney tests, unless otherwise noted in the figure legend. Bar graphs display mean and standard deviation.

Supporting information

S1 Data. Raw data numbers for all graphs.

https://doi.org/10.1371/journal.pgen.1011415.s001

(XLSX)

Acknowledgments

The authors would like to thank the labs of Drs. Cecilia Moens, Greg Walsh, Kelly Monk, Mary Mullins, and David Schoppik for zebrafish lines; Dr. Philip Campbell for assistance with spontaneous movement measurements; the UPenn Cell & Developmental Biology Microscopy Core; and members of the Granato lab for feedback regarding the manuscript.

References

  1. 1. Koch M. The neurobiology of startle. Prog Neurobiol. 1999;59(2):107–28. pmid:10463792
  2. 2. Allen MJ, Godenschwege TA, Tanouye MA, Phelan P. Making an escape: development and function of the Drosophila giant fibre system. Semin Cell Dev Biol. 2006;17(1):31–41. pmid:16378740
  3. 3. Comer CM, Dowd JP. Escape turning behavior of the cockroach. Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology. 1987;160(5):571–583.
  4. 4. Kanou M, Ohshima M, Inoue J. The Air-puff Evoked Escape Behavior of the Cricket Gryllus bimaculatus and its Compensational Recovery after Cercal Ablations. Zoological Science. 1999;16(1):71–79.
  5. 5. Eaton RC, Farley RD. Development of the mauthner neurons in embryos and larvae of the zebrafish brachydanio rerio. Copeia. 1973;4:673–682.
  6. 6. Kimmel CB, Patterson J, Kimmel RO. The development and behavioral characteristics of the startle response in the zebra fish. Dev Psychobiol. 1974;7(1):47–60. pmid:4812270
  7. 7. Eaton RC, Bombardieri RA, Meyer DL. The Mauthner-initiated startle response in teleost fish. J Exp Biol. 1977;66(1):65–81. pmid:870603
  8. 8. Eaton RC, Farley RD, Kimmel CB, Schabtach E. Functional development in the Mauthner cell system of embryos and larvae of the zebra fish. J Neurobiol. 1977;8(2):151–72. pmid:856948
  9. 9. Kimmel CB, Hatta K, Metcalfe WK. Early axonal contacts during development of an identified dendrite in the brain of the zebrafish. Neuron. 1990;4(4):535–45. pmid:2322459
  10. 10. Will U. Amphibian Mauthner cells. Brain Behav Evol. 1991;37(5):317–32. pmid:1657273
  11. 11. Zheng A, Schmid S. A review of the neural basis underlying the acoustic startle response with a focus on recent developments in mammals. Neurosci Biobehav Rev. 2023;148:105129. pmid:36914078
  12. 12. Davis M, Gendelman DS, Tischler MD, Gendelman PM. A primary acoustic startle circuit: lesion and stimulation studies. J Neurosci. 1982;2(6):791–805. pmid:7086484
  13. 13. Takahashi M, Narushima M, Oda Y. In vivo imaging of functional inhibitory networks on the mauthner cell of larval zebrafish. J Neurosci. 2002;22(10):3929–38. pmid:12019312
  14. 14. Lacoste AM, Schoppik D, Robson DN, Haesemeyer M, Portugues R, Li JM, et al. A convergent and essential interneuron pathway for Mauthner-cell-mediated escapes. Curr Biol. 2015;25(11):1526–34. pmid:25959971
  15. 15. Kimmel CB, Sessions SK, Kimmel RJ. Morphogenesis and synaptogenesis of the zebrafish Mauthner neuron. J Comp Neurol. 1981;198(1):101–20. pmid:7229136
  16. 16. Burgess HA, Granato M. Sensorimotor gating in larval zebrafish. J Neurosci. 2007;27(18):4984–94. pmid:17475807
  17. 17. Goodrich LV, Strutt D. Principles of planar polarity in animal development. Development. 2011;138(10):1877–92. pmid:21521735
  18. 18. Wang Y, Guo N, Nathans J. The role of Frizzled3 and Frizzled6 in neural tube closure and in the planar polarity of inner-ear sensory hair cells. J Neurosci. 2006;26(8):2147–56. pmid:16495441
  19. 19. Ybot-Gonzalez P, Savery D, Gerrelli D, Signore M, Mitchell CE, Faux CH, et al. Convergent extension, planar-cell-polarity signalling and initiation of mouse neural tube closure. Development. 2007;134(4):789–99. pmid:17229766
  20. 20. Wada H, Tanaka H, Nakayama S, Iwasaki M, Okamoto H. Frizzled3a and Celsr2 function in the neuroepithelium to regulate migration of facial motor neurons in the developing zebrafish hindbrain. Development. 2006;133(23):4749–59. pmid:17079269
  21. 21. Davey CF, Moens CB. Planar cell polarity in moving cells: think globally, act locally. Development. 2017;144(2):187–200. pmid:28096212
  22. 22. Davey CF, Mathewson AW, Moens CB. PCP Signaling between Migrating Neurons and their Planar-Polarized Neuroepithelial Environment Controls Filopodial Dynamics and Directional Migration. PLoS Genet. 2016;12(3):e1005934. pmid:26990447
  23. 23. Qu Y, Glasco DM, Zhou L, Sawant A, Ravni A, Fritzsch B, et al. Atypical cadherins Celsr1-3 differentially regulate migration of facial branchiomotor neurons in mice. J Neurosci. 2010;30(28):9392–401. pmid:20631168
  24. 24. Chai G, Goffinet AM, Tissir F. Celsr3 and Fzd3 in axon guidance. Int J Biochem Cell Biol. 2015;64:11–4. pmid:25813877
  25. 25. Feng J, Xian Q, Guan T, Hu J, Wang M, Huang Y, et al. Celsr3 and Fzd3 Organize a Pioneer Neuron Scaffold to Steer Growing Thalamocortical Axons. Cereb Cortex. 2016;26(7):3323–34. pmid:27170656
  26. 26. Shima Y, Kengaku M, Hirano T, Takeichi M, Uemura T. Regulation of dendritic maintenance and growth by a mammalian 7-pass transmembrane cadherin. Dev Cell. 2004;7(2):205–16. pmid:15296717
  27. 27. Kimura H, Usui T, Tsubouchi A, Uemura T. Potential dual molecular interaction of the Drosophila 7-pass transmembrane cadherin Flamingo in dendritic morphogenesis. J Cell Sci. 2006;119(6):1118–29. pmid:16507587
  28. 28. Gao FB, Kohwi M, Brenman JE, Jan LY, Jan YN. Control of dendritic field formation in Drosophila: the roles of flamingo and competition between homologous neurons. Neuron. 2000;28(1):91–101. pmid:11086986
  29. 29. Thakar S, Wang L, Yu T, Ye M, Onishi K, Scott J, et al. Evidence for opposing roles of Celsr3 and Vangl2 in glutamatergic synapse formation. Proc Natl Acad Sci U S A. 2017;114(4):E610–E618. pmid:28057866
  30. 30. Yang Y, Mlodzik M. Wnt-Frizzled/planar cell polarity signaling: cellular orientation by facing the wind (Wnt). Annu Rev Cell Dev Biol. 2015;31:623–46. pmid:26566118
  31. 31. Lu B, Usui T, Uemura T, Jan L, Jan YN. Flamingo controls the planar polarity of sensory bristles and asymmetric division of sensory organ precursors in Drosophila. Curr Biol. 1999;9(21):1247–50. pmid:10556092
  32. 32. Usui T, Shima Y, Shimada Y, Hirano S, Burgess RW, Schwarz TL, et al. Flamingo, a seven-pass transmembrane cadherin, regulates planar cell polarity under the control of Frizzled. Cell. 1999;98(5):585–95. pmid:10490098
  33. 33. Goffinet AM, Tissir F. Seven pass Cadherins CELSR1-3. Semin Cell Dev Biol. 2017;69:102–110. pmid:28716607
  34. 34. Tissir F, De-Backer O, Goffinet AM, Lambert de Rouvroit C. Developmental expression profiles of Celsr (Flamingo) genes in the mouse. Mech Dev. 2002;112(1–2):157–60. pmid:11850187
  35. 35. Joshi B, Gaur H, Hui SP, Patra C. Celsr family genes are dynamically expressed in embryonic and juvenile zebrafish. Dev Neurobiol. 2022;82(2):192–213. pmid:35213071
  36. 36. Tissir F, Goffinet AM. Shaping the nervous system: role of the core planar cell polarity genes. Nat Rev Neurosci. 2013;14(8):525–35. pmid:23839596
  37. 37. Qu Y, Huang Y, Feng J, Alvarez-Bolado G, Grove EA, Yang Y, et al. Genetic evidence that Celsr3 and Celsr2, together with Fzd3, regulate forebrain wiring in a Vangl-independent manner. Proc Natl Acad Sci U S A. 2014;111(29): E2996–3004. pmid:25002511
  38. 38. Vivancos V, Chen P, Spassky N, Qian D, Dabdoub A, Kelley M, et al. Wnt activity guides facial branchiomotor neuron migration, and involves the PCP pathway and JNK and ROCK kinases. Neural Dev. 2009;4:7. pmid:19210786
  39. 39. Chai G, Zhou L, Manto M, Helmbacher F, Clotman F, Goffinet AM, et al. Celsr3 is required in motor neurons to steer their axons in the hindlimb. Nat Neurosci. 2014;17(9):1171–9. pmid:25108913
  40. 40. Bremer J, Marsden KC, Miller A, Granato M. The ubiquitin ligase PHR promotes directional regrowth of spinal zebrafish axons. Commun Biol. 2019;2:195. pmid:31149640
  41. 41. Wada H, Iwasaki M, Sato T, Masai I, Nishiwaki Y, Tanaka H, et al. Dual roles of zygotic and maternal Scribble1 in neural migration and convergent extension movements in zebrafish embryos. Development. 2005;132(10):2273–85.
  42. 42. Ali DW, Drapeau P, Legendre P. Development of spontaneous glycinergic currents in the Mauthner neuron of the zebrafish embryo. J Neurophysiol. 2000;84(4):1726–36. pmid:11024065
  43. 43. Marsden KC, Granato M. In Vivo Ca(2+) Imaging Reveals that Decreased Dendritic Excitability Drives Startle Habituation. Cell Rep. 2015;13(9):1733–40.
  44. 44. Shimazaki T, Tanimoto M, Oda Y, Higashijima SI. Behavioral Role of the Reciprocal Inhibition between a Pair of Mauthner Cells during Fast Escapes in Zebrafish. J Neurosci. 2019;39(7):1182–1194. pmid:30578342
  45. 45. Liu KS, Fetcho JR. Laser ablations reveal functional relationships of segmental hindbrain neurons in zebrafish. Neuron. 1999;23(2):325–35. pmid:10399938
  46. 46. Issa FA, O’Brien G, Kettunen P, Sagasti A, Glanzman DL, Papazian DM. Neural circuit activity in freely behaving zebrafish (Danio rerio). J Exp Biol. 2011;214(6):1028–38. pmid:21346131
  47. 47. Satou C, Kimura Y, Kohashi T, Horikawa K, Takeda H, Oda Y, et al. Functional role of a specialized class of spinal commissural inhibitory neurons during fast escapes in zebrafish. J Neurosci. 2009;29(21):6780–93. pmid:19474306
  48. 48. Zottoli SJ, Faber DS. Properties and distribution of anterior VIIIth nerve excitatory inputs to the goldfish Mauthner cell. Brain Res. 1979;174(2):319–23. pmid:39661
  49. 49. Zottoli SJ, Faber DS. Review: The Mauthner Cell: What Has it Taught us? The Neuroscientist. 2000;6(1):26–38.
  50. 50. Zottoli SJ, Hordes AR, Faber DS. Localization of optic tectal input to the ventral dendrite of the goldfish Mauthner cell. Brain Res. 1987;401(1):113–21. pmid:3815088
  51. 51. Kimmel CB, Sessions SK, Kimmel RJ. Target recognition in neurogenesis: formation of the Mauthner axon cap. Proc Natl Acad Sci U S A. 1979;76(9):4691–4. pmid:291997
  52. 52. Koyama M, Kinkhabwala A, Satou C, Higashijima S, Fetcho J. Mapping a sensory-motor network onto a structural and functional ground plan in the hindbrain. Proc Natl Acad Sci U S A. 2011;108(3):1170–5. pmid:21199937
  53. 53. Triller A, Korn H. Glio-axonic junctional like complexes at the Mauthner cell’s axon cap of teleosts: a possible morphological basis for field effect inhibitions. Neurosci Lett. 1980;18(3):275–81. pmid:7052498
  54. 54. Jain RA, Bell H, Lim A, Chien CB, Granato M. Mirror movement-like defects in startle behavior of zebrafish dcc mutants are caused by aberrant midline guidance of identified descending hindbrain neurons. J Neurosci. 2014;34(8):2898–909. pmid:24553931
  55. 55. März M, Chapouton P, Diotel N, Vaillant C, Hesl B, Takamiya M, et al. Heterogeneity in progenitor cell subtypes in the ventricular zone of the zebrafish adult telencephalon. Glia. 2010 May;58(7):870–88. pmid:20155821
  56. 56. Nelson JC, Witze E, Ma Z, Ciocco F, Frerotte A, Randlett O, et al. Acute Regulation of Habituation Learning via Posttranslational Palmitoylation. Curr Biol. 2020;30(14):2729–2738.e4. pmid:32502414
  57. 57. Jabeen S, Thirumalai V. Distribution of the gap junction protein connexin 35 in the central nervous system of developing zebrafish larvae. Front Neural Circuits. 2013;7:91. pmid:23717264
  58. 58. Ito R, Kohno K. Development of the axon cap neuropil of the Mauthner cell in the goldfish. Cell Tissue Res. 1984;237(1):49–55. pmid:6478484
  59. 59. Chen J, Poskanzer KE, Freeman MR, Monk KR. Live-imaging of astrocyte morphogenesis and function in zebrafish neural circuits. Nat Neurosci. 2020;23(10):1297–1306. pmid:32895565
  60. 60. Kohashi T, Oda Y. Initiation of Mauthner- or non-Mauthner-mediated fast escape evoked by different modes of sensory input. J Neurosci. 2008;28(42):10641–53. pmid:18923040
  61. 61. Higashijima S, Hotta Y, Okamoto H. Visualization of cranial motor neurons in live transgenic zebrafish expressing green fluorescent protein under the control of the islet-1 promoter/enhancer. J Neurosci. 2000;20(1):206–18. pmid:10627598
  62. 62. Chandrasekhar A, Moens CB, Warren JT Jr, Kimmel CB, Kuwada JY. Development of branchiomotor neurons in zebrafish. Development. 1997;124(13):2633–44. pmid:9217005
  63. 63. Nikaido M, Law EW, Kelsh RN. A systematic survey of expression and function of zebrafish frizzled genes. PLoS One. 2013;8(1):e54833. pmid:23349976
  64. 64. Sun SD, Purdy AM, Walsh GS. Planar cell polarity genes Frizzled3a, Vangl2, and Scribble are required for spinal commissural axon guidance. BMC Neurosci. 2016;17(1):83. pmid:27955617
  65. 65. Hofmeister W, Devine CA, Rothnagel JA, Key B. Frizzled-3a and slit2 genetically interact to modulate midline axon crossing in the telencephalon. Mech Dev. 2012;129(5–8):109–24. pmid:22609481
  66. 66. Hofmeister W, Key B. Frizzled-3a and Wnt-8b genetically interact during forebrain commissural formation in embryonic zebrafish. Brain Res. 2013;1506:25–34. pmid:23438515
  67. 67. Choe SK, Nakamura M, Ladam F, Etheridge L, Sagerström CG. A Gal4/UAS system for conditional transgene expression in rhombomere 4 of the zebrafish hindbrain. Dev Dyn. 2012;241(6):1125–32. pmid:22499412
  68. 68. Mendelson B. Development of reticulospinal neurons of the zebrafish. II. Early axonal outgrowth and cell body position. J Comp Neurol. 1986;251(2):172–84. pmid:3782497
  69. 69. Pancho A, Aerts T, Mitsogiannis MD, Seuntjens E. Protocadherins at the Crossroad of Signaling Pathways. Front Mol Neurosci. 2020;13:117. pmid:32694982
  70. 70. Bao H, Berlanga ML, Xue M, Hapip SM, Daniels RW, Mendenhall JM, et al. The atypical cadherin flamingo regulates synaptogenesis and helps prevent axonal and synaptic degeneration in Drosophila. Mol Cell Neurosci. 2007;34(4):662–78. pmid:17321750
  71. 71. Berger-Müller S, Suzuki T. Seven-pass transmembrane cadherins: roles and emerging mechanisms in axonal and dendritic patterning. Mol Neurobiol. 2011;44(3):313–20. pmid:21909747
  72. 72. Steimel A, Wong L, Najarro EH, Ackley BD, Garriga G, Hutter H. The Flamingo ortholog FMI-1 controls pioneer-dependent navigation of follower axons in C. elegans. Development. 2010;137(21):3663–73. pmid:20876647
  73. 73. Fenstermaker AG, Prasad AA, Bechara A, Adolfs Y, Tissir F, Goffinet A, et al. Wnt/planar cell polarity signaling controls the anterior-posterior organization of monoaminergic axons in the brainstem. J Neurosci. 2010;30(47):16053–64. pmid:21106844
  74. 74. Onishi K, Tian R, Feng B, Liu Y, Wang J, Li Y, et al. LRRK2 mediates axon development by regulating Frizzled3 phosphorylation and growth cone-growth cone communication. Proc Natl Acad Sci U S A. 2020;117(30):18037–18048. pmid:32641508
  75. 75. Lyuksyutova AI, Lu CC, Milanesio N, King LA, Guo N, Wang Y, et al. Anterior-posterior guidance of commissural axons by Wnt-frizzled signaling. Science. 2003;302(5652):1984–8. pmid:14671310
  76. 76. Shafer B, Onishi K, Lo C, Colakoglu G, Zou Y. Vangl2 promotes Wnt/planar cell polarity-like signaling by antagonizing Dvl1-mediated feedback inhibition in growth cone guidance. Dev Cell. 2011;20(2):177–91. pmid:21316586
  77. 77. Onishi K, Shafer B, Lo C, Tissir F, Goffinet AM, Zou Y. Antagonistic functions of Dishevelleds regulate Frizzled3 endocytosis via filopodia tips in Wnt-mediated growth cone guidance. J Neurosci. 2013;33(49):19071–85. pmid:24305805
  78. 78. Carreira-Barbosa F, Kajita M, Morel V, Wada H, Okamoto H, Martinez Arias A, et al. Flamingo regulates epiboly and convergence/extension movements through cell cohesive and signalling functions during zebrafish gastrulation. Development. 2009;136(3):383–92. pmid:19091770
  79. 79. Bui DLH, Roach A, Li J, Bandekar SJ, Orput E, Raghavan R, et al. The adhesion GPCRs CELSR1-3 and LPHN3 engage G proteins via distinct activation mechanisms. Cell Rep. 2023;42(6):112552. pmid:37224017
  80. 80. Avilés EC, Stoeckli ET. Canonical wnt signaling is required for commissural axon guidance. Dev Neurobiol. 2016;76(2):190–208. pmid:26014644
  81. 81. Marfull-Oromí P, Onishi K, Han X, Yates JR3rd, Zou Y. The Fragile X Messenger Ribonucleoprotein 1 Participates in Axon Guidance Mediated by the Wnt/Planar Cell Polarity Pathway. Neuroscience. 2023;508:76–86. pmid:36191829
  82. 82. Cayuso J, Xu Q, Addison M, Wilkinson DG. Actomyosin regulation by Eph receptor signaling couples boundary cell formation to border sharpness. Elife. 2019;8:e49696. pmid:31502954
  83. 83. Tambalo M, Mitter R, Wilkinson DG. A single cell transcriptome atlas of the developing zebrafish hindbrain. Development. 2020;147(6):dev184143. pmid:32094115
  84. 84. Habas R, Kato Y, He X. Wnt/Frizzled activation of Rho regulates vertebrate gastrulation and requires a novel Formin homology protein Daam1. Cell. 2001;107(7):843–54. pmid:11779461
  85. 85. Gombos R, Migh E, Antal O, Mukherjee A, Jenny A, Mihály J. The Formin DAAM Functions as Molecular Effector of the Planar Cell Polarity Pathway during Axonal Development in Drosophila. J Neurosci. 2015;35(28):10154–67. pmid:26180192
  86. 86. Nagar D, James TK, Mishra R, Guha S, Burgess SM, Ghose A. The Formin Fmn2b Is Required for the Development of an Excitatory Interneuron Module in the Zebrafish Acoustic Startle Circuit. eNeuro. 2021;8(4):ENEURO.0329-20.2021. pmid:34193512
  87. 87. Meserve JH, Nelson JC, Marsden KC, Hsu J, Echeverry FA, Jain RA, et al. A forward genetic screen identifies Dolk as a regulator of startle magnitude through the potassium channel subunit Kv1.1. PLoS Genet. 2021;17(6):e1008943. pmid:34061829
  88. 88. Martin EA, Ijaz S, Pereda AE, Miller AC. Trichloroacetic Acid Fixation and Antibody Staining of Zebrafish Larvae. Bio Protoc. 2022;12(2):e4289. pmid:35127979