Epigenetically distinct synaptic architecture in clonal compartments in the teleostean dorsal pallium

  1. Yasuko Isoe
  2. Ryohei Nakamura
  3. Shigenori Nonaka
  4. Yasuhiro Kamei
  5. Teruhiro Okuyama
  6. Naoyuki Yamamoto
  7. Hideaki Takeuchi
  8. Hiroyuki Takeda

  • Curated by eLife

    eLife logo

    eLife assessment:

    This important paper highlights the clonal organization of the dorsal telencephalon, a major region of the vertebrate brain, and analyzes the distinctive gene expression and chromatin accessibility present in each clonal using the adult teleost fish medaka. High-quality data were collected using convincing and solid methods and these were used to identify synaptic genes with a distinct chromatin landscape and expression in one of the regions of the dorsal pallium, with the goal of ascribing an evolutionary origin to these neurons.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

The dorsal telencephalon (i.e. the pallium) exhibits high anatomical diversity across vertebrate classes. The non-mammalian dorsal pallium accommodates various compartmentalized structures among species. The developmental, functional, and evolutional diversity of the dorsal pallium remain unillustrated. Here, we analyzed the structure and epigenetic landscapes of cell lineages in the telencephalon of medaka fish ( Oryzias latipes ) that possesses a clearly delineated dorsal pallium (Dd2). We found that pallial anatomical regions, including Dd2, are formed by mutually exclusive clonal units, and that each pallium compartment exhibits a distinct epigenetic landscape. In particular, Dd2 possesses a unique open chromatin pattern that preferentially targets synaptic genes. Indeed, Dd2 shows a high density of synapses. Finally, we identified several transcription factors as candidate regulators. Taken together, we suggest that cell lineages are the basic components for the functional regionalization in the pallial anatomical compartments and that their changes have been the driving force for evolutionary diversity.

Article activity feed

  1. Version published to 10.7554/elife.85093 on eLife
  2. eLife

    Author Response

    Reviewer #1 (Public Review):

    In the present study, Yasuko Isoe, Ryohei Nakamura & colleagues follow a lineage analysis study aiming at identifying the clonal organization of the dorsal telencephalon. The authors use the teleost fish medaka to conduct their experiments since it displays a clearly delineated dorsal pallium. After identifying the clonal units that constitute the dorsal telencephalon, they analyze the epigenetic landscape in each unit. The authors identify then differential open chromatin patterns that they relate to functional aspects of each unit, and additionally, use the epigenetic landscape to infer the identity of transcription factors operating as putative regulators. Overall, the study consists of an impressive amount of data that shed light on the organization of a central brain region in vertebrates.

    The findings in the manuscript are organized into two main sections: lineage analysis and epigenetic organization. The authors combine genetic tools with laser dissections of specific clones and ATAC-seq and RNA-seq analysis in multiple samples, an approach that is very elegant and follows high technical standards. For lineage analysis, the authors used a basic, but appropriate, tool to induce and follow clones generated in early embryos, with the side note that lineages are followed using a non-ubiquitous promoter so that the authors restrict their analysis to neural progenitors. My overall impression is that the authors have collected a massive amount of high-quality data, which unfortunately is not properly integrated or discussed in the manuscript. There is only a superficial effort in incorporating the two main findings, which contrasts with the depth of acquired data.

    The observation of clonal sectors in the pallium is a great finding that deserves a more detailed analysis in terms of their developmental and evolutionary origin: How many progenitors are used to set up the entire pallium? What is the smallest clone that contributes to it? Is there any laterality bias in the clonal architecture?

    Thank you for the question. We interpret the first question as, “how many neural progenitors (or neural stem cells) at the early developmental stage contribute to the adult pallium?”. Based on the number of clonal units visualized in the pallium, we assume that there are around 50 neural stem cells at the neurula stage that provide cells in the pallium.

    In terms of the smallest clone, we found a dozen of cells in the anterior lateral pallium region (Dla) as the smallest clone. But since the HuC promoter activity is not strong in Dla (shown in Figure 1 – figure supplement 2B), we didn’t observe the clones in a reproducible way, so we removed the clones in Dla from the comprehensive structural analysis. The second smallest clone is the cells in the Dcpm, in which only a few dozens of cells were labeled at once.

    And for the last question, we didn’t find any lateral bias in the clonal architecture in the telencephalon (shown in Figure 1- figure supplement 3A, 3C)

    We added the explanation above in the revised manuscript. (page 29, line 591 - 595)

    Is the clonal architecture exclusive for progenitors or does it extend to neurons as well?

    Though we used HuC promoter to visualize the clones which should label the neural progenitors, we observed long axonal projections from Dp to the olfactory bulb, which suggest that this transgenic line labels both neural progenitors and young mature neurons, at least in some brain regions. So yes, we assume this clonal architecture extends to neurons as well, and we added descriptions to the revised manuscript. (page 10, line 205-207)

    How has the clonal architecture impacted the morphological diversity of the pallium among teleosts? What are possible evolutionary paths to explain this phenomenon? The authors' discussion on this point circles around one concept, illustrated in the following sentence: " (The clonal architecture) ... possibly explains how the difference in diversity between the pallium and subpallium has emerged: the subpallium is conserved because cells belong to various clonal units intertwined with each other, which has constrained their modification during evolution; whereas the pallium is diverse because of the modular nature of the clonal units which allows for the emergence of diversity". This is the concept that I have the most problems with. The authors' reason that a more defined clonal structure (pallium) makes a system more prone to evolutionary novelties, while a region where clones intermingle (subpallium) is more rigid and therefore more conserved between species. Is there experimental data that backs up this statement in any other systems? If there is, I urge the authors to share these here. If this is just a speculation, then the argument would benefit from an explanation of how this clonal organization allows for evolutionary novelty.

    We appreciate the reviewer’s question. In order to make our point, we added the following paragraph to the revised manuscript,

    “Our structural analysis in the adult medaka telencephalon revealed that the clonal architecture between the pallium and subpallium differs in the distribution of cells in clonal units: clonal units in the subpallium intertwine with each other, whereas the pallium is formed by the compartmentalized clonal units, giving rise to a modular structure. Modular structure is frequently seen in the animal body, including brain; central complex in insect 40, cerebellum in vertebrates 41. And the modularity of cell populations or organs is generally thought to contribute to evolutionary flexibility; one module can acquire a new phenotype without impacting the others.42, 43, 44 . We assume that the modular nature of the clonal units in the pallium plays a key role in the diversity across teleost.” (page 23, line 448-452)

    Would it happen by the appearance of more clones at the early stages of development? The authors leave this central point untouched even when discussing the evolutionary origin of the pallium in teleosts.

    Thank you for the comment. As shown in the previous report, when the Cre-loxp recombination was induced at the early developmental stage, a wider expression of GFP is observed across the whole brain (Okuyama et al. 2013). This suggests that the neural stem cells at the earlier developmental stage generate daughter neural stem cells which produce neural progenitors later. We added a few sentences mentioning this in the revised manuscript. (page 7, line 146-149)

    Having shown the clonal architecture of the pallium and conducted a detailed epigenetic analysis in clones, the authors could also speculate on what is special about this type of organization. Particularly, how they envision that cells belonging to the same clone inherit a common epigenetic landscape that will define their function later on.

    Thank you for the comment. To explain the epigenetic feature of this pallial organization, we added the following paragraph in the revised manuscript.

    “As shown in mammals, the epigenetic landscape can be inherited from apical progenitors, which have a multipotency, to the late neural progenitors during development 37. Since the teleost exhibit post-hatch neurogenesis in the entire life, we think that the common epigenetic landscape is inherited in each clonal unit in the adult medaka telencephalon. And as a result, we make the assumption that function and characteristic of each clonal unit is defined already in progenitors by specific regulators (e.g. TFs), and those progenitors continuously produce neurons that possess the same property to function in a coordinated manner.” (page 22, line 433-439)

    There is little analysis of the cellular organization of each clone, mainly because the authors labeled only a subset of the real, genetic clone. The authors present images of entire brains and optical horizontal and transverse sections, which largely sustain their claims for a clonal organization. The difference in the clonal arrangements between the Dld and the Vd is clear, but the authors could provide a higher-resolution image of some clones in the telencephalon to get an idea of the cellular composition of the regions they use for their analysis.

    Here, we added a new panel in Figure 2 which is a combination of previous supplemental figures S3-1,2,3 to show our analysis on the cellular organization of each clone. We showed how the pallial regions, other than Dld, are formed by multiple genetic clones in different colors, and also the projection from each clone. (page 9, Figure 2B)

    What is the extent of non-GFP cells in the regions they use for RNAseq and ATACseq? From the images shown it is very difficult to realize whether all cells in the clonal sector do indeed belong to the clone.

    Thank you for your question. In our revised manuscript, we analyzed the ratio of cells labeled in this transgenic line (HuC:loxp-DsRed-loxp-GFP). We found that a large portion of cells (around 60-70% cells) are DsRed positive in our transgenic line (Figure 1 - figure supplement 2B). (page 7, line 142-143)

    Reviewer #2 (Public Review):

    In this study, Isoe and team produced an atlas of the telencephalon of the adult medaka fish with which they better defined pallial and subpallial regions, characterized the expression of neurotransmitters, and performed clonal analysis to address their organization and maintenance during the continuous neurogenesis. They show that pallial anatomical regions are formed by independent clonal units. Furthermore, the authors demonstrate that pallial compartments exhibit region-specific chromatin landscapes, suggesting that gene expression is differentially regulated. Specifically, synaptic genes have a distinct chromatin landscape and expression in one of the regions of the dorsal pallium, the Dd2. Using the region-specific RNA expression and chromatin accessibility data they have generated; the authors propose several transcription factors as candidate regulators of Dd2 specification. Lastly, the authors use the enrichment of transcription factor binding motifs to establish homology between medaka and human telencephalon, aiming to describe an evolutionary origin for the Dd2 region.

    Overall, the study carefully describes diverse aspects of neurogenesis in the telencephalon of the adult medaka fish. As such, the manuscript has the potential to contribute insights to the understanding of circuits and neurogenesis in teleosts and the medaka fish, as well as the evolution of cellular heterogeneity and organization of the telencephalon. Furthermore, the atlas, if easily accessible to the broader community, could be a substantial resource to researchers interested in medaka and teleosts neuroscience. However, there are some conceptual and technical concerns that should be addressed to strengthen this work.

    Improving the atlas: The different interpretations of the imaging data generated remain isolated or fragmented and could be better integrated to describe anatomical, connectivity, and ontogeny differences through pallial and subpallial regions.

    In the revision process, we described the details of anatomical, and connectivity differences in the adult pallial and subpallial regions in Table 2. This document includes the description of comparing the brain regions with previous atlases.

    In terms of the ontogeny differences, we described the neural stem cells localization in the telencephalon in Figure 1 figure supplement 4. “The cell-body distribution in the pallium and subpallium is consistent with the pattern of the neural stem cell (radial glial) (Figure 1 – figure supplement 4). In the teleost telencephalon, the cell bodies of radial glia are located in the surface of the hemispheres and project inside the telencephalon 15. Since neural progenitors migrate along those axons, it is consistent that the cell bodies of the pallial clonally-related units are clustered along those axons in a cylindrical way.”(page 8, line 175-179; page 22, line 427-431))

    Molecular differences across regions and species: Differential gene expression and chromatin accessibility throughout regions should be better and more deeply characterized and presented, exhibiting more region-specific features, and leading to a better description of candidate transcription factors that could differentially regulate regional gene expression.

    The comparison between medaka fish and human telencephalon regions would benefit from a more extensive molecular analysis. Comparison of gene expression and accessible regions could expand the analysis together with TF-binding motif enrichment.

    In order to check the gene expression across brain regions in the different vertebrate species, we examined the mammal gene expression data (in situ hybridization) from the Allen Institute database. We analyzed the expression of all the Dd-specific expressing genes (809 genes) across the mammalian brain regions (12 regions), but we could not observe strong correlations with any specific brain regions in mice. Therefore, we have revised our conclusions regarding the correspondence between medaka's Dd2 and mammalian brain regions to be more cautious. (page 20, line 396 - page 21, line 401)

    Lineage tracing: The authors claim that the functional compartmentalization of the pallium relies on different cell lineages, which also mostly share connectivity patterns and, at least to some extent, expression patterns. It would be interesting to see how homogenous these lineages are at the molecular level and whether their compartmentalization is retained when neurons reach maturity.

    Thank you for the comment. We think single-cell RNA-seq in cell lineages in the future will allow us to see how homogenous cells that derived from the same lineages are at the molecular level and to assess the cell-type of the cells.

  3. eLife

    eLife assessment:

    This important paper highlights the clonal organization of the dorsal telencephalon, a major region of the vertebrate brain, and analyzes the distinctive gene expression and chromatin accessibility present in each clonal using the adult teleost fish medaka. High-quality data were collected using convincing and solid methods and these were used to identify synaptic genes with a distinct chromatin landscape and expression in one of the regions of the dorsal pallium, with the goal of ascribing an evolutionary origin to these neurons.

  4. eLife

    Reviewer #1 (Public Review):

    In the present study, Yasuko Isoe, Ryohei Nakamura & colleagues follow a lineage analysis study aiming at identifying the clonal organization of the dorsal telencephalon. The authors use the teleost fish medaka to conduct their experiments since it displays a clearly delineated dorsal pallium. After identifying the clonal units that constitute the dorsal telencephalon, they analyze the epigenetic landscape in each unit. The authors identify then differential open chromatin patterns that they relate to functional aspects of each unit, and additionally, use the epigenetic landscape to infer the identity of transcription factors operating as putative regulators. Overall, the study consists of an impressive amount of data that shed light on the organization of a central brain region in vertebrates.

    The findings in the manuscript are organized into two main sections: lineage analysis and epigenetic organization. The authors combine genetic tools with laser dissections of specific clones and ATAC-seq and RNA-seq analysis in multiple samples, an approach that is very elegant and follows high technical standards. For lineage analysis, the authors used a basic, but appropriate, tool to induce and follow clones generated in early embryos, with the side note that lineages are followed using a non-ubiquitous promoter so that the authors restrict their analysis to neural progenitors. My overall impression is that the authors have collected a massive amount of high-quality data, which unfortunately is not properly integrated or discussed in the manuscript. There is only a superficial effort in incorporating the two main findings, which contrasts with the depth of acquired data.

    The observation of clonal sectors in the pallium is a great finding that deserves a more detailed analysis in terms of their developmental and evolutionary origin: How may progenitors are used to set up the entire pallium? What is the smallest clone that contributes to it? Is there any laterality bias in the clonal architecture? Is the clonal architecture exclusive for progenitors or does it extend to neurons as well? How has the clonal architecture impacted the morphological diversity of the pallium among teleosts? What are possible evolutionary paths to explain this phenomenon? The authors' discussion on this point circles around one concept, illustrated in the following sentence: " (The clonal architecture) ... possibly explains how the difference in diversity between the pallium and subpallium has emerged: the subpallium is conserved because cells belong to various clonal units intertwined with each other, which has constrained their modification during evolution; whereas the pallium is diverse because of the modular nature of the clonal units which allows for the emergence of diversity". This is the concept that I have the most problems with. The authors' reason that a more defined clonal structure (pallium) makes a system more prone to evolutionary novelties, while a region where clones intermingle (subpallium) is more rigid and therefore more conserved between species. Is there experimental data that backs up this statement in any other systems? If there is, I urge the authors to share these here. If this is just a speculation, then the argument would benefit from an explanation of how this clonal organization allows for evolutionary novelty. Would it happen by the appearance of more clones at the early stages of development? The authors leave this central point untouched even when discussing the evolutionary origin of the pallium in teleosts.

    Having shown the clonal architecture of the pallium and conducted a detailed epigenetic analysis in clones, the authors could also speculate on what is special about this type of organisation. Particularly, how they envision that cells belonging to the same clone inherit a common epigenetic landscape that will define their function later on. There is little analysis of the cellular organization of each clone, mainly because the authors labeled only a subset of the real, genetic clone. The authors present images of entire brains and optical horizontal and transverse sections, which largely sustain their claims for a clonal organization. The difference in the clonal arrangements between the Dld and the Vd is clear, but the authors could provide a higher-resolution image of some clones in the telencephalon to get an idea of the cellular composition of the regions they use for their analysis. What is the extent of non-GFP cells in the regions they use for RNAseq and ATACseq? From the images shown it is very difficult to realize whether all cells in the clonal sector do indeed belong to the clone.

  5. eLife

    Reviewer #2 (Public Review):

    In this study, Isoe and team produced an atlas of the telencephalon of the adult medaka fish with which they better defined pallial and subpallial regions, characterized the expression of neurotransmitters, and performed clonal analysis to address their organization and maintenance during the continuous neurogenesis. They show that pallial anatomical regions are formed by independent clonal units. Furthermore, the authors demonstrate that pallial compartments exhibit region-specific chromatin landscapes, suggesting that gene expression is differentially regulated. Specifically, synaptic genes have a distinct chromatin landscape and expression in one of the regions of the dorsal pallium, the Dd2. Using the region-specific RNA expression and chromatin accessibility data they have generated; the authors propose several transcription factors as candidate regulators of Dd2 specification. Lastly, the authors use the enrichment of transcription factor binding motifs to establish homology between medaka and human telencephalon, aiming to describe an evolutionary origin for the Dd2 region.

    Overall, the study carefully describes diverse aspects of neurogenesis in the telencephalon of the adult medaka fish. As such, the manuscript has the potential to contribute insights to the understanding of circuits and neurogenesis in teleosts and the medaka fish, as well as the evolution of cellular heterogeneity and organization of the telencephalon. Furthermore, the atlas, if easily accessible to the broader community, could be a substantial resource to researchers interested in medaka and teleosts neuroscience. However, there are some conceptual and technical concerns that should be addressed to strengthen this work.

    Improving the atlas: The different interpretations of the imaging data generated remain isolated or fragmented and could be better integrated to describe anatomical, connectivity, and ontogeny differences through pallial and subpallial regions.
    Molecular differences across regions and species: Differential gene expression and chromatin accessibility throughout regions should be better and more deeply characterized and presented, exhibiting more region-specific features, and leading to a better description of candidate transcription factors that could differentially regulate regional gene expression. The comparison between medaka fish and human telencephalon regions would benefit from a more extensive molecular analysis. Comparison of gene expression and accessible regions could expand the analysis together with TF-binding motif enrichment.
    Lineage tracing: The authors claim that the functional compartmentalization of the pallium relies on different cell lineages, which also mostly share connectivity patterns and, at least to some extent, expression patterns. It would be interesting to see how homogenous these lineages are at the molecular level and whether their compartmentalization is retained when neurons reach maturity.

  6. eLife

    Reviewer #3 (Public Review):

    In this manuscript, the authors characterized the clonal composition in the medaka pallium and found the dorsal pallium region to be a compartment constituting repeatedly identifiable clonal units. By performing ATAC-seq on the clonal units as well as RNA-seq on different subregions of the pallium, the dorsal pallium was further identified as a unique region with open chromatin regions with regulatory elements enriched for synapse-related genes. Further experimental and bioinformatic evidence supports the region's putative function of synapse generation, with similar TF-binding motifs to homologous brain regions in human. Although the "uniqueness" of the dorsal pallium might be a coincidence of the timing of clonal tracing, the conclusions in the manuscript are largely supported by experimental evidence. The study showcases an elegant model of how anatomical, molecular, and functional diversity arises in a previously under-characterized brain region. The enriched genes in Dd2 are an interesting candidate for future investigation, and the function of the dorsal pallium of teleost fish is of general interest for studies of brain evolution.

  7. Version published to 10.1101/2022.10.01.510385v2 on bioRxiv
  8. Version published to 10.1101/2022.10.01.510385v1 on bioRxiv