Ciliomotor circuitry underlying whole-body coordination of ciliary activity in the Platynereis larva

Abstract

Ciliated surfaces harbouring synchronously beating cilia can generate fluid flow or drive locomotion. In ciliary swimmers, ciliary beating, arrests, and changes in beat frequency are often coordinated across extended or discontinuous surfaces. To understand how such coordination is achieved, we studied the ciliated larvae of Platynereis dumerilii, a marine annelid. Platynereis larvae have segmental multiciliated cells that regularly display spontaneous coordinated ciliary arrests. We used whole-body connectomics, activity imaging, transgenesis, and neuron ablation to characterize the ciliomotor circuitry. We identified cholinergic, serotonergic, and catecholaminergic ciliomotor neurons. The synchronous rhythmic activation of cholinergic cells drives the coordinated arrests of all cilia. The serotonergic cells are active when cilia are beating. Serotonin inhibits the cholinergic rhythm, and increases ciliary beat frequency. Based on their connectivity and alternating activity, the catecholaminergic cells may generate the rhythm. The ciliomotor circuitry thus constitutes a stop-and-go pacemaker system for the whole-body coordination of ciliary locomotion.

Keywords: P. dumerilii; acetylcholine; catecholamines; ciliary nerve; connectomics; neuroscience; serotonin; zooplankton.

Figure 1.. Coordinated ciliary beating and arrests in Platynereis larvae.

(A–B,D) Scanning electron micrographs of 72 hr-post-fertilization larvae, (A) ventral view, (B) lateral view, (D) anterior view, close up. The structures dorsal to the crescent cell are non-motile sensory cilia. (C) Serial TEM reconstruction of ciliary band cells, lateral view. Nuclei of ciliary band cells are shown as spheres. The nervous system (brain and ventral nerve cord) is shown in light grey. (E) Kymographs of spontaneous ciliary activity from an immobilized 72 hr-post-fertilization larva. Arrowheads indicate the beginning of ciliary arrests. (F) Ventral view of the left half of a 72 hr-post-fertilization larva in a calcium-imaging experiment (left panel) with the corresponding differential interference contrast (DIC) image (right panel). (G) GCaMP6s signal from the prototroch and paratroch ciliary bands. A kymograph of ciliary activity in paratroch III is shown below. (H) GCaMP6s signal recorded from the prototroch of a 48 hr-post-fertilization larva at 45 frames per second. White areas in the kymograph correspond to periods of arrest. The boxed area is shown enlarged in the inset. Scale bars, 50 μm (A), 10 μm (D). DOI: http://dx.doi.org/10.7554/eLife.26000.003

Figure 2.. Anatomy and connectivity of biaxonal cholinergic ciliomotor neurons.

(A) ssTEM reconstruction of the MC neuron (blue), ventral view. Ciliated cells are shown in grey. Circles represent position of cell body, lines represent axonal track. Presynaptic sites along the axon are marked in red. (B) The MC neuron in anterior view. (C) ssTEM reconstruction of the Loop neurons, ventral view. (D) The right Loop neuron in lateral view. (E) Synaptic connectivity of the MC neuron and the prototroch cells. (F) Synaptic connectivity of the Loop neurons and ciliary band cells. (G) Transgenic labelling of the right Loop neuron with an sGCβ1:palmi-3xHA-tdTomato construct. Expression was detected with anti-HA antibody staining. Right panel: ventral view, left panel: lateral view. Asterisks indicate background signal in the spinning glands. In the network graphs, nodes represent individual or grouped cells, as indicated by the numbers in square brackets. The edges show the direction of signalling from presynaptic to postsynaptic cells. Edges are weighted by the number of synapses. The number of synapses is indicated on the edges. Scale bar, 40 μm (G). Abbreviation: prn, prototroch ring nerve. DOI: http://dx.doi.org/10.7554/eLife.26000.008

Figure 2—figure supplement 1.. Anatomy and connectivity of eyespotPRC^R3, ventral MN, and MN^ant ciliomotor neurons.

(A) ssTEM reconstruction of the eyespotPRC^R3 neurons, anterior view. Ciliated cells are shown in grey or in teal if they receive synapses from eyespotPRC^R3. (B) Synaptic connectivity of the eyespotPRC^R3 neurons and prototroch. (C) ssTEM reconstruction of the MN^r1 neurons, ventral view. (D) Synaptic connectivity of the MN^r1 neurons, ciliary band cells, and muscles. (E) ssTEM reconstruction of the MN^r2 neurons, ventral view. (F) Synaptic connectivity of the MN^r2 neurons, ciliary band cells, and muscles. (G) ssTEM reconstruction of the MN^r3 neurons, ventral view. (H) Synaptic connectivity of the MN^r3 neurons and ciliary band cells. (I–L) ssTEM reconstruction of the MN^ant neurons, (I) anterior view, (J,K) ventral view. (L) Synaptic connectivity of the MN^ant neurons and ciliary band cells. In the network graphs, nodes represent individual or grouped cells, as indicated by the numbers in square brackets. The edges show the direction of signalling from presynaptic to postsynaptic cells. Edges are weighted by the number of synapses. The number of synapses is indicated on the edges. DOI: http://dx.doi.org/10.7554/eLife.26000.009

Figure 2—figure supplement 2.. sGCbeta1 is a cholinergic marker.

Average registered gene expression pattern of (A) ChAT and (B) sGCbeta1. (C) Overlap of the ChAT and sGCbeta1 gene expression domains. (D–F) Transgenic labeling of MN^l3 neuron with the sGCbeta1:palmi-3xHA-tdTomato construct. Larvae in (D,E) were counterstained for acetylated tubulin (white). Expression was detected with anti-HA antibody staining. Scale bar in (D,F) 30 µm. DOI: http://dx.doi.org/10.7554/eLife.26000.010

Figure 3.. Anatomy and connectivity of serotonergic ciliomotor neurons.

(A–B) ssTEM reconstruction of the right (A) and left (B) head serotonergic ciliomotor (Ser-h1) neuron, anterior view. (C) Transgenic labelling of the right Ser-h1 neuron with the TrpH:palmi-3xHA-tdTomato construct, anterior view. Expression was detected with anti-HA antibody staining. (D–E) ssTEM reconstruction of the right (D) and left (E) trunk serotonergic ciliomotor (Ser-tr1) cell, anterior view. (F) Transgenic labelling of the left Ser-tr1 neuron with the TrpH:palmi-3xHA-tdTomato construct, ventral view. (G) ssTEM reconstruction of the pygPB^bicil serotonergic sensory-ciliomotor neuron. Ciliated cells are shown in grey. Circles represent position of cell body, lines represent axonal track. Presynaptic sites along the axon are marked in red. (H) Synaptic connectivity of the serotonergic ciliomotor neurons and the ciliary band cells. Nodes represent individual or grouped cells, as indicated by the numbers in square brackets. Edges show the direction of signalling from presynaptic to postsynaptic cells. Edges are weighted by the number of synapses. Synapse number is indicated. Edges with one synapse are not shown. Scale bars, 30 μm (C, F). Abbreviation: prn, prototroch ring nerve. DOI: http://dx.doi.org/10.7554/eLife.26000.012

Figure 3—figure supplement 1.. Serotonergic neurons in Platynereis larvae.

(A–C) Immunostaining for serotonin and acetylated tubulin in (A) 48 hr-post-fertilization, (B) 72 hr-post-fertilization, and (C) 50 hr-post-fertilization larvae. Anterior (A,B) and ventral views (C). Scale bars, 30 μm (A,C), 20 μm (B). Abbreviations: cPRC, ciliary photoreceptor cell. DOI: http://dx.doi.org/10.7554/eLife.26000.013

Figure 3—figure supplement 2.. Parallel axons of the Loop and Ser-tr1 neurons.

(A) Electron micrographs of Loop and Ser-tr1 neuron axon cross-sections. Axon outlines are shown. The Loop neurons contain several dense core vesicles. The Ser-tr1 neurons have a clearer cytoplasm and no dense core vesicles. (B) ssTEM reconstruction of the Loop and Ser-tr1 neurons, ventral view. Loop and Ser-tr1 axons run in parallel throughout the body. Scale bar in (A) 1 μm. DOI: http://dx.doi.org/10.7554/eLife.26000.014

Figure 4.. Anatomy and connectivity of catecholaminergic/peptidergic ciliomotor neurons.

(A–C) ssTEM reconstruction of the cMN^d (green) (A), cMN^ATO (orange) (B), and cMN^vl (blue) (C) ciliomotor neurons, anterior view. Ciliated cells are shown in grey. Circles represent position of cell bodies, lines represent axonal tracks. Presynaptic sites along the axon are marked in red. (D) Synaptic connectivity of the cMN neurons with the prototroch and the MC cell. Nodes represent individual or grouped cells, as indicated by the numbers in square brackets. Edges show the direction of signalling from presynaptic to postsynaptic cells. Edges are weighted by the number of synapses and indicate the direction of signalling. Synapse number is indicated. Abbreviation: prn, prototroch ring nerve. DOI: http://dx.doi.org/10.7554/eLife.26000.017

Figure 4—figure supplement 1.. Catecholaminergic and neuropeptide marker expression in cMN neurons.

(A–C) In situ hybridisation for TyrH (A), dbh (B), and allatotropin proneuropeptide (C). (D) Immunostaining with an anti-allatotropin antibody. The samples were counterstained for acetylated tubulin (white). Scale bar in (A) 30 µm. DOI: http://dx.doi.org/10.7554/eLife.26000.018

Figure 5.. Overall connectivity of ciliomotor neurons and ciliary bands.

(A) Synaptic connectivity graph of all ciliomotor neurons and ciliary band cells. The network is partitioned into three modules. The neurons belonging to each module are shown in the anatomical reconstructions. The ciliary band cells are grouped into anatomical units. (B) Matrix representation of the connectivity of ciliomotor neurons to all ciliated cells. Colour intensity is proportional to the number of synapses. (C) Matrix representation of the connectivity of ciliomotor neurons amongst themselves. Colour intensity is proportional to the number of synapses. (D–E) Graph representations of the ciliomotor circuit. Nodes with <4 synapses are not shown. In (D) nodes are grouped by cell type (teal, cholinergic; red, serotonergic; orange, cMN; grey, ciliated cells). In (E) all cMN, serotonergic, cholinergic neurons, and ciliary band cells are represented as one node each. In (D, E) edges with <4 synapses are not shown. Abbreviations: proto, prototroch; meta, metatroch; para, paratroch; akro, akrotroch. DOI: http://dx.doi.org/10.7554/eLife.26000.019

Figure 6.. Activity of cMN neurons.

(A) ssTEM reconstruction (left) and correlation (Pearson's r) map of neuronal activity (right) of cMN^vl, anterior view. Correlation values were calculated relative to the prototroch (outlined). (B) Calcium-imaging trace of cMN^vl and the prototroch (C) ssTEM reconstruction (left) and correlation (Pearson’s r) map of neuronal activity (right) of cMN^ATO. Correlation values were calculated relative to the prototroch (outlined). (D) Calcium imaging trace of cMN^ATO and the prototroch. (E) ssTEM reconstruction (left) and correlation (Pearson’s r) map of neuronal activity (right) of cMN^d. Correlation values were calculated relative to the prototroch (outlined). (F) Calcium-imaging trace of cMN^d and the prototroch. (G) Correlation of GCaMP signals of cMN neurons with GCaMP signals measured from the prototroch ciliary band (CB). Data points represent measurements from different larvae, n > 12 larvae. Mean and standard deviation are shown. All sample medians are different from 0 with p-values≤0.0002 as determined by Wilcoxon Signed Rank Test. DOI: http://dx.doi.org/10.7554/eLife.26000.021

Figure 7.. Activity of cholinergic neurons.

(A–B) GCaMP6s signal revealing cell morphologies of the active MC (anterior view) (A) and Loop (ventral/posterior view) (B) neurons in 2 days-post-fertilisation larvae. (C) Correlation (Pearson’s r) map of neuronal activity of the MC, vMN^sync and vMN^async neurons. Correlation values were calculated relative to the prototroch. (D) Correlation of GCaMP signals of cholinergic neurons with GCaMP signals measured from the prototroch ciliary band (CB). Data points represent measurements from different larvae, n > 12 larvae. Mean and standard deviation are shown. All sample medians are different from 0 with p-values≤0.0034 as determined by Wilcoxon Signed Rank Test. (E–F) Effect of MC neuron ablation on prototroch activity and ciliary closures. Red bar represents the time of the ablation. GCaMP signals in the MC neuron, in a vMN^sync neuron, and in the prototroch before (E) and after (F) MC neuron ablation. (G) Number of ciliary arrests in the prototroch before and after MC neuron ablation. (H) Calcium signals measured from the MC neuron and the prototroch following the addition of 50 µM alpha-bungarotoxin. Abbreviation: prn, prototroch ring nerve. DOI: http://dx.doi.org/10.7554/eLife.26000.023

Figure 7—figure supplement 1.. Characterisation of MC neuron activity.

(A) Normalised calcium signals in the MC neuron (n = 96 larvae). (B) Cycle length of MC neurons as determined by Fourier analysis of the traces shown in (A). (C) Frequency distribution of cycle length of MC neurons, same data as in panel (B). DOI: http://dx.doi.org/10.7554/eLife.26000.026

Figure 7—figure supplement 2.. Effect of MC neuron ablation on prototroch activity and ciliary closures.

(A–F) Calcium signals in the MC neuron, a vMN^rsync neuron, and in the prototroch, before and after MC neuron ablation. Kymographs showing arrests of prototroch cilia are also shown. Arrowheads indicate the beginning of arrests. (A–F) show examples of MC neuron ablations in different larvae. DOI: http://dx.doi.org/10.7554/eLife.26000.028

Figure 8.. Activity of serotonergic neurons.

(A) ssTEM reconstruction (left) and correlation (Pearson’s r) map of neuronal activity (right) of the Ser-h1 neurons, anterior view. Correlation values were calculated relative to the prototroch. (B) Calcium signals measured from a Ser-h1 neuron and the MC cell. (C) ssTEM reconstruction (left) and neuronal activity correlation (right) of the Ser-tr1 neurons, ventral view. Correlation values were calculated relative to the prototroch. (D) Calcium signals measured from a Ser-tr1 neuron and a Loop neuron. (E) Correlation of GCaMP signals of serotonergic neurons with GCaMP signals measured from the prototroch ciliary band (CB). Data points represent measurements from different larvae, n > 9. Mean and standard deviation are shown. All sample medians are different from 0 with p-values≤0.002 as determined by Wilcoxon Signed Rank Test. (F) Ciliary beat frequency of prototroch cilia in the absence and presence of serotonin. (G) Ciliary closures of prototroch cilia in the absence and presence of serotonin. P-values of an unpaired (F) and paired (G) t-test are shown. In (F,G) n > 9 larvae. Samples in (F,G) passed the D'Agostino and Pearson omnibus normality test (alpha = 0.05). DOI: http://dx.doi.org/10.7554/eLife.26000.030

Figure 8—figure supplement 1.. In vivo labelling of a Ser-tr1 cell by photoactivation during calcium imaging.

(A) Correlation map of neuronal activity of a Ser-tr1 and a Loop neuron, ventral view. (B) Calcium signals measured from the Ser-tr1 and Loop neurons shown in (A). (C) Serotonin antibody staining (cyan) and mCherry1 fluorescence (orange) in the Ser-tr1 cell following local photoactivation of PAmCherry1 during calcium imaging (in the larva shown in A and B). DOI: http://dx.doi.org/10.7554/eLife.26000.034

Figure 9.. The stop-and-go ciliomotor system of the Platynereis larva.

(A) ssTEM reconstruction of ciliomotor neurons that are activated synchronously (orange) and asynchronously (teal) with the ciliary bands. (B) Schematic of the circuitry, consisting of a pacemaker system, as well as cholinergic and serotonergic ciliomotor neurons that span the entire body and innervate all multiciliated cells. Abbreviation: prn, prototroch ring nerve. DOI: http://dx.doi.org/10.7554/eLife.26000.035