Detail View

Comparative transcriptomic analysis of non-model fishes identifies spinal neuronal correlates of divergent locomotor strategies

Citations

WEB OF SCIENCE

Citations

SCOPUS

Metadata Downloads

DC Field Value Language
dc.contributor.advisor 백명인 -
dc.contributor.author Tery Yun -
dc.date.accessioned 2026-09-01T19:29:30Z -
dc.date.available 2026-09-01T19:29:30Z -
dc.date.issued 2026 -
dc.identifier.uri https://scholar.dgist.ac.kr/handle/20.500.11750/60723 -
dc.identifier.uri http://dgist.dcollection.net/common/orgView/200001006684 -
dc.description comparative transcriptomics,single-cell RNA sequencing,spatial transcriptomics,spinal cord,locomotion -
dc.description.abstract Locomotor behaviors have diversified extensively during vertebrate evolution, yet how conserved spinal cord developmental programs are adjusted to generate species-specific locomotor strategies remains unclear. Here, comparative single-cell RNA sequencing and Stereo-seq spatial transcriptomics were performed on the spinal cord of two non-model actinopterygian fishes with contrasting locomotor strategies—the body- undulating Siberian sturgeon (Acipenser baerii) and the fin-propelled big-belly seahorse (Hippocampus abdominalis)—and compared with a published mouse spinal cord reference and a re-analyzed zebrafish dataset to examine how conserved cardinal neuronal subtypes are remodeled in association with divergent locomotor modes. Cross-species cell-type matching showed that cardinal spinal neuronal subtypes are broadly conserved between actinopterygian fishes and mammals, whereas species-specific divergence was concentrated in subtype abundance and spatial distribution. The sturgeon spinal cord showed a marked depletion of ventral motor- associated populations, consistent with reduced initial neuronal production and subsequent postmitotic depletion, together with little to no detectable Raldh2 expression. In contrast, the seahorse dorsal fin level exhibited substantial expansion of motor neurons and ventral interneurons. Motor neuron subclustering identified three molecularly distinct subtypes—axial MN-like, fin MN-like, and PGC-like populations—that correspond to the mammalian MMC, LMC, and PGC programs. The Foxp1- centered LMC-like transcriptional core, including Hox6/Hox10-associated regional signatures, was deployed at both the pectoral and dorsal fin levels in seahorse, whereas Raldh2 expression—required for retinoic acid– mediated specification of LMC-lateral identity in tetrapods—remained restricted to the pectoral fin level. This dissociation suggests that the LMC-like program is a partially decomposable molecular module: its Foxp1- centered appendicular core can be redeployed across paired and unpaired anatomical contexts, while Raldh2- dependent LMC-lateral specification remains more strictly associated with paired appendage systems. Cross-species positional analysis further revealed that Hox-based regional programs in motor neurons and V2a interneurons align with mouse forelimb-versus-hindlimb signatures, whereas non-Hox positional genes showed weak cross-species concordance. This indicates a hierarchical organization of evolutionary constraint, with strong conservation of the Hox framework and lineage-specific divergence of downstream effector programs. Together, these findings reveal that the vertebrate spinal cord retains a deeply conserved cell-type identity architecture while permitting species-specific adjustment along three orthogonal axes: quantitative remodeling of neuronal composition, modular redeployment of subtype identity programs beyond their canonical anatomical contexts, and divergence of downstream effector programs within a stable Hox-based framework. This supports a developmental tuning model in which vertebrate locomotor evolution proceeds not primarily through the invention of novel spinal neuron classes, but through coordinated modulation of conserved cardinal subtypes. Keywords: comparative transcriptomics, single-cell RNA sequencing, spatial transcriptomics, spinal cord, locomotion.|척추동물의 운동 행동은 진화 과정에서 매우 다양하게 분화되어 왔지만, 보존된 척수 발달 프로그램이 어떻게 조정되어 종 특이적인 운동 전략을 형성하는지는 아직 명확히 밝혀지지 않았다. 본 연구에서는 서로 다른 운동 전략을 보이는 두 비모델 조기어류인, 몸통의 파동 운동을 통해 유영하는 시베리아 철갑상어(Acipenser baerii)와 몸통을 비교적 고정한 채 지느러미 운동에 의존하는 빅벨리 해마(Hippocampus abdominalis)의 척수를 대상으로 단일세포 RNA 시퀀싱과 Stereo-seq 기반 공간 전사체 분석을 수행하였다. 또한 공개된 마우스 척수 참조 데이터와 재분석한 제브라피쉬 데이터를 함께 비교함으로써, 보존된 척수 신경세포 아형들이 서로 다른 운동 방식에 따라 어떻게 재구성되는지를 분석하였다.
종간 세포 유형 매칭 분석 결과, 마우스에서 정의된 주요 척수 신경세포 아형들은 조기어류와 포유류 사이에서 전반적으로 보존되어 있었다. 반면 종 특이적인 차이는 새로운 신경세포 유형의 출현보다는 각 아형의 상대적 비율과 공간적 분포의 변화에 집중되어 있었다. 철갑상어 척수에서는 운동과 관련된 복측 신경세포 집단이 현저히 감소되어 있었으며, 이는 초기 신경세포 생성 감소와 이후 postmitotic 단계에서의 세포 집단 감소 가능성과 연관되어 있었다. 또한 철갑상어에서는 Raldh2 발현이 거의 검출되지 않았다. 반대로 해마의 등지느러미 수준 척수에서는 운동신경세포와 복측 인터뉴런 집단이 뚜렷하게 확장되어 있었다.
운동신경세포를 세부적으로 재분류한 결과, 축성 운동신경세포 유사 아형(axial MN-like), 지느러미 운동신경세포 유사 아형(fin MN-like), 그리고 PGC 유사 아형(PGC-like)의 세 가지 분자적으로 구분되는 집단이 확인되었으며, 이들은 각각 포유류의 MMC, LMC, PGC 프로그램에 대응되었다. 특히 Foxp1 중심의 LMC 유사 전사체 프로그램은 해마의 가슴지느러미뿐만 아니라 등지느러미 수준의 운동신경세포에서도 나타났으나, 사지형 척추동물의 LMC-lateral identity 형성에 관여하는 Raldh2 발현은 가슴지느러미 수준에만 제한적으로 관찰되었다. 이러한 결과는 Foxp1 중심의 부속지형 운동신경세포 프로그램이 서로 다른 지느러미 운동계에서 재배치될 수 있는 반면, Raldh2 의존적 프로그램은 보다 제한적으로 paired appendage-associated system과 연관되어 있음을 시사한다.
추가적인 종간 위치성 분석에서는 운동신경세포와 V2a 인터뉴런의 Hox 기반 전후축 지역성 프로그램이 마우스의 앞다리와 뒷다리 수준에서 관찰되는 전사체 특징과 대응되는 반면, non-Hox 위치성 유전자들은 종간 일치성이 상대적으로 약한 것으로 나타났다. 이는 척수 지역성 프로그램이 Hox 기반의 안정적인 상위 조절 체계와, 계통 특이적으로 변화하는 하위 실행 유전자 프로그램으로 구성된 계층적 구조를 가진다는 점을 보여준다.
종합하면, 본 연구는 척추동물 척수가 깊이 보존된 세포 유형 정체성 체계를 유지하면서도, 종 특이적인 운동 전략에 맞추어 세 가지 축에서 조정될 수 있음을 보여준다. 첫째, 신경세포 조성의 양적 재구성, 둘째, 기존 아형 정체성 프로그램의 새로운 해부학적 맥락으로의 재배치, 셋째, 안정적인 Hox 기반 틀 안에서의 하위 실행 프로그램의 분화가 그것이다. 이러한 결과는 척추동물의 운동 진화가 완전히 새로운 척수 신경세포 유형의 발명보다는, 보존된 cardinal 신경세포 아형들의 조정과 재배치를 통해 이루어진다는 발달 조율 모델을 제시한다.
-
dc.description.tableofcontents List of Contents

Abstract i
List of contents ii
List of tables vi
List of figures vii

Ⅰ. Introduction 1
1.1 Vertebrate Locomotor Diversity and Spinal Circuit Organization 1
1.2 Developmental Origins of Spinal Cord Cell Diversity 5
1.2.1 Dorsoventral patterning and the cardinal neuron classes 5
1.2.2 Rostrocaudal patterning by Hox genes 5
1.2.3 Motor neuron columnar organization and subtype diversification 6
1.2.4 Evolutionary variation in neuronal composition 7
1.3 Cross-Species Comparative Transcriptomics 13
1.3.1 Single-cell RNA sequencing and cell type classification 13
1.3.2 Cross-species comparison: challenges and computational strategies 13
1.3.3 Spatial transcriptomics: linking molecular identity to tissue architecture 14
1.4 Main Question and Aims of This Study 15

II. Materials and Methods 20
2.1 Animals 20
2.2 Phylogenomic analysis and ortholog inference 20
2.2.1 Species selection and proteome acquisition 20
2.2.2 Phylogenetic tree construction 21
2.2.3 Ortholog identification 21
2.3 Motor neuron retrograde tracing 22
2.4 Single-cell RNA sequencing: sample preparation 22
2.4.1 Cell-sorting and library preparation 22
2.4.2 Reference dataset acquisition and preprocessing 23
2.5 scRNA-seq data processing 23
2.5.1 Read alignment and reference genome construction 23
2.5.2 Quality control and filtering 24
2.5.3 Clustering and cell type annotation 24
2.6 scRNA-seq downstream analysis 24
2.6.1 Neuronal and MN subclustering 24
2.6.2 Validation of neuronal cluster robustness 25
2.6.3 Pseudo-bulk differential expression analysis 26
2.6.4 Cross-species positional DEG concordance analysis 27
2.7 CAME-based cross-species cell type matching 28
2.8 Stereo-seq spatial transcriptomics 29
2.8.1 Library preparation and sequencing 29
2.8.2 Data processing 30
2.8.3 Downstream analysis in Seurat 30
2.8.4 Spatial identification of dorsal horn and ventral domains 31
2.9 Histology and imaging 33
2.9.1 RNA in situ hybridization 33
2.9.2 Immunohistochemistry 34
2.9.3 Image acquisition 34
2.9.4 Neuron soma size measurement 35
2.9.5 Spatial mapping and quantification of marker-expressing cells 35
2.10 Behavioral recording and kinematic analysis 36
2.11 Statistical analysis and visualization 36

III. Results 37
3.1 Establishing a Comparative Framework and Spinal Cord Atlas 37
3.1.1 Phylogenomic positioning of sturgeon and seahorse within ray-finned fishes 37
3.1.2 Contrasting locomotor strategies 37
3.1.3 Fin-associated motor neuron abundance differs between species 38
3.1.4 Conserved Hox-based regional identity defines the comparative framework 38
3.2 Single-Cell and Spatial Transcriptomic Atlas of the Spinal Cord 45
3.2.1 scRNA-seq profiling of sturgeon and seahorse spinal cord 45
3.2.2 Stereo-seq spatial mapping of major cell populations 45
3.3 Shared Spinal Neuron Classes and Their Quantitative Divergence 61
3.3.1 Identification of twelve cardinal neuron classes across four species 61
3.3.2 CAME-based cross-species cell type matching 61
3.3.3 Spatial and in situ validation of neuronal identities 62
3.3.4 Species- and region-specific variation in neuronal class proportions 62
3.4 Divergent Dorsoventral Organization Reflects Locomotor Strategy 82
3.4.1 Divergent dorsoventral composition between sturgeon and seahorse 82
3.4.2 Histological validation of the dorsoventral shift 83
3.4.3 Graded ventral enrichment across fin levels within seahorse 83
3.5 Developmental Remodeling Underlies Ventral Depletion in Sturgeon 91
3.5.1 Progressive motor neuron decline after initial generation in sturgeon 91
3.5.2 Olig2 progenitor domain switching and retinoic acid signaling in sturgeon 91
3.5.3 Selective loss of ventral inhibitory interneurons 91
3.5.4 Motor neuron ablation in mouse does not cause ventral interneuron loss 92
3.6 Motor Neuron Subtype Diversity and Deployment of an LMC-like Program 97
3.6.1 Three MN subtypes in seahorse: Axial, Fin, and PGC MN-like 97
3.6.2 Validation of PGC MN-like identity by retrograde labeling 97
3.6.3 Divergent LMC-like deployment across fin levels 97
3.6.4 LMC-like matching across mouse rostral and caudal spinal levels 98
3.6.5 Conservation of the tripartite MN subtype structure in zebrafish 99
3.7 Regional Transcriptional Programs and Hox-Based Positional Conservation 105
3.7.1 Pectoral versus dorsal DEGs in seahorse spinal cord neurons 105
3.7.2 Hox genes as the dominant regional markers in seahorse spinal neurons 105
3.7.3 Cross-species comparison reveals Hox-restricted positional conservation 106
3.7.4 Non-Hox regional DEGs include ion channel and signaling genes 108

IV. Discussion 119
4.1 Overview 119
4.2 Neuronal Remodeling for Locomotor Diversity 119
4.2.1 Quantitative remodeling of ventral populations and motor neurons 119
4.2.2 Conserved spinal neuron repertoire despite locomotor divergence 121
4.2.3 Motor neuron subtype diversity and deployment of an LMC-like program 122
4.2.4 Hox-based positional conservation and downstream divergence 124
4.2.5 Molecular candidates for tuning fin motor patterns 125
4.2.6 An integrative model for spinal locomotor evolution 127
4.3 Hierarchical Evolutionary Architecture Across Spinal Cord Cell Types 128
4.4 Limitations and Future Directions 128

V. Conclusion 131
VI. References 153
VII. Abstract in Korean 164
-
dc.format.extent 164 -
dc.language eng -
dc.publisher DGIST -
dc.title Comparative transcriptomic analysis of non-model fishes identifies spinal neuronal correlates of divergent locomotor strategies -
dc.title.alternative 비모델 어류의 비교 전사체 분석을 통한 보존된 척수 신경세포 아형 규명 및 운동 방식 차이에 따른 조성 변화 분석 -
dc.type Thesis -
dc.identifier.doi 10.22677/THESIS.200001006684 -
dc.description.degree Doctor -
dc.contributor.department Department of Brain Sciences -
dc.contributor.coadvisor Byung-Chang Suh -
dc.date.awarded 2026-08-01 -
dc.publisher.location Daegu -
dc.description.database dCollection -
dc.citation XT.BD 윤88 202608 -
dc.date.accepted 2026-07-21 -
dc.contributor.alternativeDepartment 뇌과학과 -
dc.subject.keyword comparative transcriptomics,single-cell RNA sequencing,spatial transcriptomics,spinal cord,locomotion -
dc.contributor.affiliatedAuthor Tery Yun -
dc.contributor.affiliatedAuthor Myungin Baek -
dc.contributor.affiliatedAuthor Byung-Chang Suh -
dc.contributor.alternativeName 윤태리 -
dc.contributor.alternativeName Myungin Baek -
dc.contributor.alternativeName 서병창 -
dc.rights.embargoReleaseDate 2031-08-31 -
Show Simple Item Record

File Downloads

  • There are no files associated with this item.

공유

qrcode
공유하기

Total Views & Downloads

???jsp.display-item.statistics.view???: , ???jsp.display-item.statistics.download???: