Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Nov 3;9(44):eadh7693.
doi: 10.1126/sciadv.adh7693. Epub 2023 Nov 1.

A telencephalon cell type atlas for goldfish reveals diversity in the evolution of spatial structure and cell types

Affiliations

A telencephalon cell type atlas for goldfish reveals diversity in the evolution of spatial structure and cell types

Muhammad Tibi et al. Sci Adv. .

Abstract

Teleost fish form the largest group of vertebrates and show a tremendous variety of adaptive behaviors, making them critically important for the study of brain evolution and cognition. The neural basis mediating these behaviors remains elusive. We performed a systematic comparative survey of the goldfish telencephalon. We mapped cell types using single-cell RNA sequencing and spatial transcriptomics, resulting in de novo molecular neuroanatomy parcellation. Glial cells were highly conserved across 450 million years of evolution separating mouse and goldfish, while neurons showed diversity and modularity in gene expression. Specifically, somatostatin interneurons, famously interspersed in the mammalian isocortex for local inhibitory input, were curiously aggregated in a single goldfish telencephalon nucleus but molecularly conserved. Cerebral nuclei including the striatum, a hub for motivated behavior in amniotes, had molecularly conserved goldfish homologs. We suggest elements of a hippocampal formation across the goldfish pallium. Last, aiding study of the teleostan everted telencephalon, we describe substantial molecular similarities between goldfish and zebrafish neuronal taxonomies.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.. Composition of goldfish telencephalon cells.
(A) Outline of this study. Adult goldfish telencephala were analyzed in parallel using spatial- and single-cell transcriptomics, integrated to a spatially mapped cell type atlas, and followed by species comparison. (B) Images of fresh dissected goldfish brain and coronal telencephalon sections, describing the microdissection scheme. OT, optic tecum; SC, spinal cord. (C) t-distributed stochastic neighbor embedding (t-SNE) visualization of all cells in the goldfish forebrain. Each dot represents a cell, colored by class. (D) Density scatter visualization on t-SNE of all cells [as in (C)] per microdissection origin. Yellow, dense; blue, sparse. (E) Left; Contribution of four microdissections to each class; visualized as pie charts. Right: Contribution of cell class to each dissection; visualized as bar plot. (F) Dendrogram overview of all cell types. Bottom; Top marker gene expression visualized as dotplot. Dot size represents the percentage of cells in a cluster expressing each gene; dot color represents cell type.
Fig. 2.
Fig. 2.. Axial and regional parcellation of goldfish telencephalon.
(A) Molecular cell type–based neuroanatomy of the goldfish telencephalon. Top: Color scheme shown above for GABAergic and glutamatergic cell types (dendrogram). Center: Eight coronal goldfish sections (goldfish 1), sampled for Visium ST, overlaid with a weighted color map that integrates all goldfish telencephalon cell types. Bottom: Regional parcellation based on color map differences above, each color indicates a different region, and suggested nomenclature annotated by similarity region names according to Northcutt (23). D, area dorsalis; V, area ventralis; Dc, large-celled subdivision of Dm; Vsst, ventral Sst; Ppa, nucleus preopticus parvocellularis anterioris; a, anterior; p, posterior; d, dorsal; v, ventral; m, medial; l, lateral. (B) Heatmaps of SD (normalized per row) along lateral-medial (left) and dorsal-ventral (right) axes for top axial pattern genes, for goldfish 1 and 2; dots indicate spatial enrichment according to axial color scheme shown above; gray without dot, no enrichment. Right: Summary of axial score per gene (mean of enriched sections). (C) Expression of 14 axial-patterned genes across the goldfish telencephalon. Gray, low; red, high.
Fig. 3.
Fig. 3.. GABAergic neurons in the goldfish telencephalon.
(A) t-SNE visualization of GABAergic neurons in the goldfish forebrain. Each dot represents a cell, colored by cell type assignment. Right: Expression of three branch-organizing genes. (B) All GABA types arranged in dendrogram order (GABA1 to GABA40), with top marker gene expression visualized as heatmap (white, high; black, low). Middle: Violin plots, where each dot represents a single cell; maximum expression (UMI) indicated on the right. Bottom: Contribution of four microdissections to each cell type, visualized as pie charts. (C) Expression of three branch-organizing genes [as (A)] and, in ST, eight anterior-posterior telencephalon coronal hemisphere sections. (D) Examples across the GABAergic dendrogram for spatial correlation of Visium spots: five scRNA-seq cell types (columns) across eight a.-p. coronal sections (rows).
Fig. 4.
Fig. 4.. Glutamatergic neurons in the goldfish telencephalon.
(A) t-SNE visualization of glutamatergic neurons in the goldfish forebrain. Each dot represents a cell, colored by cell type. (B) All glutamatergic types, in dendrogram order (GLUT1 to GLUT48), with top marker gene expression visualized as heatmap (white, high; black, low). Middle: Violin plots, where each dot represents a single cell; maximum expression (UMI) indicated on the right. Bottom: Contribution of four microdissections to each cluster, visualized as pie charts. (C) Expression of two branch-organizing genes, NR2F2 and CNR1, visualized on t-SNE [as (A)] and, in ST, eight anterior-posterior telencephalon coronal hemisphere sections. (D) Examples across the glutamatergic dendrogram for spatial correlation of Visium spots: four scRNA-seq cell types (columns) across eight anterior-posterior coronal sections (rows).
Fig. 5.
Fig. 5.. Region-specific marker genes and cell types.
(A) Top enriched region-specific marker genes as identified by ST, presented as dotplot: Circle size represents normalized mean expression of each gene in each region. (B) Representative examples of genes from (A), with spatial distribution of expession across eight anterior-posterior telencephalon sections. Gray, low; red, high. (C) Enrichment of cell types to regions shown as dotplot, where dot size represents the correlation between cell type expression profile and region expression profile. (D) Validation of a switch in territories for NEUROD6 expression along the telencephalon anterior-posterior axis in matched sections of ST and HCR-FISH. Section overviews for HCR are displayed segmented, with normalized expression scale below. Scale bars, 0.5 mm (ST and HCR-overview). (E) Weighted spatial cell type enrichment for two groups of glutamatergic clusters only (green, GLUT20, GLUT21, GLUT30, and GLUT31; red, GLUT5, GLUT12, GLUT16, and GLUT18), indicating a Dm-Dl territory switch for both groups that would result in 3D bends/horns.
Fig. 6.
Fig. 6.. Goldfish and mouse cross species comparison of the telencephalon cell types.
(A) Scheme of comparative species analysis between goldfish and mouse telencephalon (33). Per cell class, both species’ datasets are integrated (Harmony), followed by density-based spatial clustering of applications with noise (DBSCAN) clustering or a KNN classifier. (B) Nonneuronal cells species comparison visualized as heatmap with conserved cell types outlined (KNN classifier). (C) Nonneuronal cells species comparison visualized as t-SNE. Dots represent cells, colored by goldfish cell type. gray, mouse cell; gray outlines, integrated cluster (see fig. S7); pink outlines and labels, mouse cell class; with mouse Schwann cells highlighted. (D) GABA class species comparison (KNN classifier), with examples for consistency with integrated clusters highlighted. (E) Distribution of SST interneuron types by spatial correlation with goldfish ST and mouse ISH (33), with zoom-in to dispersed Sst expression in isocortex (58) (Allen Mouse Brain ISH Atlas). (F) Distribution of (putative) striatal neuron types by spatial correlation with goldfish ST (left) and mouse ISH (right) (33). (G) Expression of two mouse striatum markers in goldfish ST (left) and mouse ISH (right) (33). (H) Distribution of (putative) pallidal neuron types by spatial correlation with goldfish ST (left) and mouse ISH (right) (33). (I) GLUT class species comparison (KNN classifier), with examples for consistency with integrated clusters highlighted. (J) Distribution of (putative) retrohippocampal neuron types by spatial correlation with goldfish ST (left) and mouse ISH (right) (33). (K) Expression of five genes in goldfish ST and mouse ISH, in (putative) subiculum/retrohippocampal formation.
Fig. 7.
Fig. 7.. Cross-species comparison of teleostean telencephalon cell types.
(A and B) t-SNE visualizing comparative species analysis between goldfish and zebrafish (46) telencephalon, with cell types highlighted per species; (A) goldfish and (B) zebrafish. Per cell class, both species’ datasets are integrated (Harmony), followed by DBSCAN clustering. (C) Per cell type comparison, scored using KNN classifier. (D) Expression of CBLN1, PENK, and SST in the integrated teleostean dataset; dots (cells) colored by species origin. (E) Validation of gene expression detected in ST using HCR-FISH. Top row: Corresponding section overviews of genes detected in Visium (left) and HCR (right), where each spot represents a segmented fluorescent cell, colored by normalized expression. Bottom row: Raw fluorescent signal in zoom-ins, as indicated in overview sections.

References

    1. C. Salas, C. Broglio, F. Rodríguez, Evolution of forebrain and spatial cognition in vertebrates: Conservation across diversity. Brain Behav. Evol. 62, 72–82 (2003). - PubMed
    1. C. Broglio, F. Rodríguez, A. Gómez, J. L. Arias, C. Salas, Selective involvement of the goldfish lateral pallium in spatial memory. Behav. Brain Res. 210, 191–201 (2010). - PubMed
    1. T. Fremouw, P. Jackson-Smith, R. P. Kesner, Impaired place learning and unimpaired cue learning in hippocampal-lesioned pigeons. Behav. Neurosci. 111, 963–975 (1997). - PubMed
    1. J. C. López, F. Rodríguez, Y. Gómez, J. P. Vargas, C. Broglio, C. Salas, Place and cue learning in turtles. Anim. Learn. Behav. 28, 360–372 (2000).
    1. F. Rodríguez, J. C. López, J. P. Vargas, Y. Gómez, C. Broglio, C. Salas, Conservation of spatial memory function in the pallial forebrain of reptiles and ray-finned fishes. J. Neurosci. 22, 2894–2903 (2002). - PMC - PubMed