SCODiA Report by MLBI Lab

Single-Cell Transcriptomic Profiling Reveals Distinct and Shared Cellular and Pathway Dysregulation in Familial and Sporadic Alzheimer's Disease Brain

This comprehensive single-cell analysis of human brain tissue reveals distinct and shared cellular and molecular pathologies in familial (E280A) and sporadic Alzheimer's disease (AD) compared to controls. Key findings include a shift in microglial populations from homeostatic to activated M2-like states in both AD forms, alongside a notable increase in fibroblasts in sporadic AD samples, indicating extensive neurovascular and extracellular matrix remodeling. Furthermore, while familial AD is characterized by a general dampening of synaptic communication, sporadic AD exhibits robust activation of endothelial-mediated interactions. These insights highlight the heterogeneous nature of AD and pinpoint specific cellular dysfunctions and pathways that could serve as targets for subtype-specific therapeutic interventions.

Contents

  1. Dataset overview
  2. UMAP Visualization of Cell Populations by Condition, Sample, and Cell Type Hierarchies
  3. Major Cell Type Score and Annotation Mapping on UMAP
  4. Overall Celltype_subset Marker Expression Dot Plot Analysis
  5. Cell Type Population Analysis in Brain Samples
  6. Microglial Subpopulation Shifts in Alzheimer's Disease Conditions
  7. Microglial Subset Population Shifts in Neurodegenerative Conditions
  8. Cell-Cell Interaction Analysis in Control Brain Tissue
  9. Condition-Specific Cell-Cell Interaction Patterns in Brain Tissue
  10. Microglia Condition-Specific Surfaceome Markers in Alzheimer's Disease
  11. Endothelial Cell Gene Ontology (GSA) Analysis in Alzheimer's Disease Conditions
  12. 뇌 세포 유형별 유전자 세트 농축 분석 결과 (GSEA)
  13. Discussion
  14. Query List

0. Dataset overview

데이터셋 요약

데이터 유형: 단일 세포 RNA 시퀀싱 데이터 (AnnData 형식)

세포 및 유전자 수: 43,743개 세포, 26,318개 유전자

종: 인간

조직: 뇌

변수 (var columns): variable_genes

조건: E280A, Control, Sporadic

1. UMAP Visualization of Cell Populations by Condition, Sample, and Cell Type Hierarchies

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis presents five UMAP (Uniform Manifold Approximation and Projection) plots derived from single-cell RNA sequencing data of human brain tissue. Each plot visualizes the same cellular embedding but is colored by different metadata attributes: overall condition (Control, E280A, Sporadic), individual sample, major cell type, minor cell type, and granular cell type subset. These visualizations provide a foundational overview of the dataset's structure, cell type composition, and the distribution of samples and conditions across the identified cell populations.

Visual Summary

Condition UMAP

The UMAP colored by condition reveals that cells from all three conditions (Control, E280A, Sporadic) largely intermingle across many regions of the UMAP space, indicating a shared underlying cellular architecture. However, certain regions show enrichment for specific conditions. For example, some areas within the large neuronal cluster appear to have a higher density of E280A (yellow) or Sporadic (purple) cells, suggesting potential condition-specific cellular states or compositional differences. Control cells (maroon) are broadly distributed, often overlapping with the disease conditions but also occupying distinct zones.

Sample UMAP

The sample colored UMAP shows a generally good mixing of cells from different individual samples (each patient representing a sample) across the UMAP projection. This indicates that major batch effects related to individual samples have been largely mitigated during data integration. While there is broad mixing, some subtle patterns emerge where specific samples or groups of samples from the same condition show localized enrichment, reflecting inter-individual biological variability or subtle remaining sample-specific effects.

Celltype_major UMAP

This UMAP clearly delineates major cell types into distinct clusters.

Celltype_minor UMAP

The celltype_minor UMAP provides a finer resolution of cell types, refining the major clusters.

Celltype_subset UMAP

The celltype_subset UMAP presents the most granular level of cell type annotation, revealing substantial cellular heterogeneity.

Biological Interpretation

The UMAP plots provide a robust framework for understanding the cellular landscape of the human brain in the context of Alzheimer's disease (E280A and Sporadic forms) and healthy controls.

The clear segregation of major and minor cell types confirms that the single-cell RNA-seq data effectively captures the distinct transcriptomic profiles of various brain cell populations, including neurons, astrocytes, oligodendrocytes, and microglia, as well as endothelial and stromal components. This detailed cell type annotation is crucial for downstream analyses, as the brain is a highly heterogeneous organ.

The presence of specific microglial activation states (M0, M1, M2a, M2b, M2c) is particularly relevant for studying neuroinflammation, a hallmark of Alzheimer's disease. M1 microglia are generally considered pro-inflammatory, while M2 subtypes (M2a, M2b, M2c) are often associated with anti-inflammatory, repair, or tissue remodeling functions. Their distinct clustering suggests that their proportions or specific states might differ across conditions, warranting further investigation into their roles in AD pathology PubMed search: microglia M1 M2 Alzheimer's disease.

The extensive diversity of neuronal subtypes captured, from Glutamatergic to Dopaminergic and Cholinergic neurons, allows for targeted studies on neuronal vulnerability or resilience in specific AD pathologies. Different neuronal populations are known to be differentially affected in neurodegenerative diseases. For instance, cholinergic neurons are particularly impacted in Alzheimer's disease GeneCards: CHAT.

The observed overlap but also distinct distribution of cells from E280A, Sporadic, and Control conditions across the UMAP space suggests two key aspects:

  1. Shared cellular identity: Many core cell types and states are conserved across healthy and diseased brains.
  2. Disease-specific perturbations: Regions where disease conditions are enriched likely represent cell states or compositions that are altered in Alzheimer's disease, either due to pathological processes (e.g., reactive astrocytes, activated microglia) or changes in cell proportions. This indicates a strong biological signal related to disease status within the dataset, paving the way for differential expression and cell-cell interaction analyses.

Annotation Notes

The comprehensive and hierarchical cell type annotations (major, minor, and subset) appear to be well-defined and align logically with the UMAP structure, indicating high confidence in cell identity assignments. The good mixing of samples across the UMAP suggests that the data integration effectively minimized major batch effects, enhancing the biological interpretability of condition-specific differences. The presence of an 'unassigned' cluster, while small, indicates a small proportion of cells that might require further investigation or exclusion, depending on the focus of subsequent analyses.

2. Major Cell Type Score and Annotation Mapping on UMAP

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a Uniform Manifold Approximation and Projection (UMAP) visualization of single-cell RNA-seq data from human brain tissue. The UMAP plots serve two main purposes: first, to display a continuous "HiCAT_major_score" for each major cell type, indicating the strength of identity for that cell type across all cells; and second, to show the discrete celltype_major annotations for each cell. This comparison allows for a visual assessment of the coherence between the computed cell type scores and the assigned cell type labels.

Visual Summary

The UMAP embedding reveals a complex yet structured landscape of cell populations, with several distinct clusters.

HiCAT_major_score Plots (Top 6 Panels):

celltype_major Plot (Bottom-Left Panel):

Biological Interpretation

The strong visual alignment between the continuous "HiCAT_major_score" and the discrete celltype_major annotations indicates a robust and reliable cell type classification for this single-cell RNA-seq dataset from the human brain.

Annotation Notes

The consistency between the quantitative cell type scores and the qualitative cell type labels is a key indicator of high-quality annotation. The clear separation of most major cell types on the UMAP, both by continuous scoring and discrete labeling, suggests that the cell type assignments are well-supported by the underlying gene expression profiles. This provides a solid foundation for downstream analyses, such as differential gene expression or cell-cell interaction studies, ensuring that comparisons are made between genuinely distinct cell populations.

3. Overall Celltype_subset Marker Expression Dot Plot Analysis

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a dot plot illustrating the expression of marker genes across various celltype_subset populations identified in the single-cell RNA-seq data from human brain tissue. The plot serves to validate the assigned cell type annotations by demonstrating distinct and expected gene expression patterns for each subset. Dot size indicates the fraction of cells within a given subset expressing a particular gene, while color intensity reflects the mean expression level of that gene within the subset.

Visual Summary

The dot plot clearly shows well-defined clusters of marker genes, with high expression (dark red, large dots) predominantly localized to specific celltype_subset populations. This pattern strongly supports the distinct identities of the annotated cell types.

Biological Interpretation

The observed marker gene expression patterns align remarkably well with the known biology of the celltype_subset populations in the human brain, providing strong evidence for the accuracy of the cell type annotations.

Annotation Notes

The comprehensive display of marker gene expression across celltype_subset populations provides strong validation for the current cell type annotations. The distinct and biologically relevant marker sets for each subset confirm their identity and differentiation. The clear separation of expression patterns and minimal ambiguous signals suggest high confidence in the quality of the cell type assignments for this dataset. This plot serves as a fundamental step in ensuring the reliability of downstream analyses, such as differential gene expression or cell-cell interaction studies, which rely on accurate cell type identification.

4. Cell Type Population Analysis in Brain Samples

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis presents a bar plot illustrating the relative proportions of minor cell types within individual samples across three conditions: Control, E280A, and Sporadic. The data is derived from single-cell RNA sequencing of human brain tissue. This visualization provides an initial overview of potential compositional changes in the brain cellular landscape associated with different disease contexts.

Visual Summary

The stacked bar plot shows the percentage composition of various minor cell types for each sample, grouped by condition.

Condition-Specific Differences:

Biological Interpretation

The observed shifts in cell type proportions offer initial biological insights into the distinct conditions, particularly in the context of brain pathology:

Clinical or Translational Implications

5. Microglial Subpopulation Shifts in Alzheimer's Disease Conditions

Report figure

[Analysis Visualization Results]...

Analysis Overview

이 분석은 단일 세포 RNA 시퀀싱(scRNA-seq) 데이터를 사용하여 Microglia 세포의 하위 유형(M0, M1, M2a, M2b, M2c) 분포를 Control, E280A (가족성 알츠하이머병 변이), 그리고 Sporadic (산발성 알츠하이머병) 조건별로 비교한 결과를 막대그래프 형태로 시각화합니다. 각 막대는 개별 샘플을 나타내며, 전체 Microglia 개체군 내에서 각 하위 유형이 차지하는 비율을 보여줍니다.

Visual Summary

Biological Interpretation

뇌의 주요 면역 세포인 Microglia는 질병 상태에서 다양한 표현형으로 활성화될 수 있습니다. 본 분석 결과는 알츠하이머병(AD)의 두 가지 형태인 가족성(E280A)과 산발성(Sporadic) AD에서 Microglia의 하위 유형 분포에 특징적인 변화가 있음을 보여줍니다.

종합적으로, E280A 및 Sporadic AD 모두에서 Microglia는 항상성 M0 상태에서 벗어나 염증 조절, 손상 복구 및 잔해 제거와 관련된 M2 유사 표현형으로 전환되는 경향을 보입니다. 이는 AD 뇌에서 Microglia가 병리 진행에 반응하여 적극적으로 변화하고 있음을 명확히 보여줍니다.

Clinical or Translational Implications

6. Microglial Subset Population Shifts in Neurodegenerative Conditions

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates potential statistically significant differences in the proportion of specific microglial subsets, namely Microglia (M0) and Microglia (M2b), across different conditions: Sporadic (likely sporadic neurodegenerative disease), E280A (a familial Alzheimer's disease mutation), and Control. The goal is to identify cell-state shifts in the microglial population that may be associated with disease conditions.

Visual Summary

The box plots illustrate the celltype proportion (percentage) of Microglia (M0) and Microglia (M2b) for each condition. Each black dot represents an individual sample. Statistical significance (p-value) for pairwise comparisons between conditions is indicated above the plots.

Microglia (M0)

Microglia (M2b)

Biological Interpretation

Microglia are the primary immune cells of the central nervous system, playing critical roles in maintaining brain homeostasis, immune surveillance, and responding to injury or disease [1]. They exhibit remarkable plasticity, adopting different functional phenotypes, often broadly categorized as M0 (resting/homeostatic), M1 (pro-inflammatory), and M2 (anti-inflammatory/pro-resolving/reparative). The M2 category itself is heterogeneous, including subtypes like M2a, M2b, M2c.

Clinical or Translational Implications

These findings highlight distinct microglial phenotypic shifts in neurodegenerative conditions.

---

References:

  1. Microglial functions in health and disease: PubMed search for "microglia function neurodegeneration" https://pubmed.ncbi.nlm.nih.gov/?term=microglia+function+neurodegeneration
  2. Microglial activation in Alzheimer's disease: PubMed search for "microglial activation Alzheimer's disease" https://pubmed.ncbi.nlm.nih.gov/?term=microglial+activation+Alzheimer%27s+disease
  3. M2b microglia phenotype: PubMed search for "M2b microglia" https://pubmed.ncbi.nlm.nih.gov/?term=M2b+microglia
  4. Targeting microglia for neurodegenerative diseases: PubMed search for "microglia therapeutic target neurodegeneration" https://pubmed.ncbi.nlm.nih.gov/?term=microglia+therapeutic+target+neurodegeneration

7. Cell-Cell Interaction Analysis in Control Brain Tissue

Report figure

[Analysis Visualization Results]...

This analysis characterizes the intricate cell-cell communication landscape within the Control human brain tissue, providing a baseline for understanding physiological interactions. The dot plot visualizes the top 80 most significant and strongest ligand-receptor interactions between different brain cell types.

Analysis Overview

Cell-cell interaction (CCI) analysis using CellPhoneDB identifies potential ligand-receptor mediated communication between different cell populations. For the Control condition, this analysis highlights key communicative pathways essential for maintaining brain homeostasis. The visualization prioritizes interactions based on their statistical significance (p-value) and interaction strength (mean expression of the ligand-receptor pair), filtering for interactions with pval < 0.05 and mean > 0.01.

Visual Summary

The dot plot displays a matrix where the y-axis represents interacting cell type pairs (e.g., "Neuron|Astrocyte"), and the x-axis lists specific ligand-receptor pairs (e.g., "Glutamate_byGLS2_and_SLC17A7_GRM3"). Each dot signifies a detected interaction:

Key observations from the plot for Control samples include:

Biological Interpretation

The observed cell-cell interactions in the Control human brain tissue reflect fundamental processes required for normal brain function:

  1. Excitatory Neurotransmission and Neuromodulation: The abundance of glutamatergic signaling pathways (e.g., Glutamate_byGLS2_and_SLC17A7_GRM3) involving neurons and astrocytes highlights their coordinated activity in excitatory neurotransmission. Glutamate, released by neurons, can bind to astrocytic glutamate receptors, initiating gliotransmission or modulating synaptic activity. SLC17A7 (also known as VGLUT1) is a vesicular glutamate transporter, and GRM3 is a metabotropic glutamate receptor. GLS2 (glutaminase 2) converts glutamine to glutamate. This interplay is critical for learning, memory, and cognitive functions.

References:

  1. Synaptic Plasticity and Connectivity: The prominent interactions involving Neurexins (NRXN), Neuroligins (NLGN), and LRRTMs underscore their role in establishing and maintaining synaptic integrity. These protein families form trans-synaptic bridges that regulate synapse formation, maturation, and plasticity. Their presence in Neuron|Neuron interactions is expected, given their direct involvement in synaptic structure. Interactions with glial cells suggest broader roles in shaping synaptic circuits and supporting neuronal health.

References:

  1. Myelination and Oligodendrocyte Function: Interactions involving oligodendrocytes and OPCs are crucial for myelination, a process where myelin sheaths are formed around axons to facilitate rapid signal transmission. The involvement of Neurexins and LRRTMs in Oligodendrocyte|Neuron interactions hints at the complex signaling required for proper myelination and axonal support. NRG3-ERBB4 signaling is known to be involved in oligodendrocyte development and myelination.

References:

  1. Neuroimmune Modulation: Microglial interactions, such as those involving APP-TNFRSF21, suggest their role in immune surveillance and responses, potentially influencing neuronal survival or pruning processes under physiological conditions. APP (Amyloid Precursor Protein) is highly relevant in neurodegenerative contexts, and its interaction with TNFRSF21 (Death Receptor 6) could be involved in regulating neuronal apoptosis or stress responses.

References:

  1. Neurovascular Coupling and Axon Guidance: Endothelial cell interactions with neurons via NTN1-UNC5C and NTN4-NTRK2 suggest roles in maintaining the neurovascular unit and potentially influencing neuronal guidance or survival. Netrins (NTN1, NTN4) are well-known guidance cues, and UNC5C and NTRK2 (TrkB) are their respective receptors.

References:

Clinical or Translational Implications

Understanding these baseline cell-cell interactions in Control brain tissue is paramount for contextualizing changes observed in disease states like E280A or Sporadic conditions. These interactions represent potential physiological hubs that, if dysregulated, could contribute to pathology and thus serve as therapeutic targets.

  1. Therapeutic Target Prioritization: Ligand-receptor pairs that are highly active and significant in the Control brain, such as those involved in glutamatergic signaling or synaptic adhesion (Neurexins/Neuroligins/LRRTMs), could be critical for maintaining neuronal health. If these pathways are found to be perturbed in disease, they represent attractive targets for interventions aimed at restoring synaptic function, reducing excitotoxicity, or promoting neuroprotection. For example, strategies to modulate specific glutamate receptor subtypes or enhance synaptic stability could be explored.
  2. Biomarker Discovery and Experimental Validation: Identifying specific cell-cell communication axes that are robust in control but altered in disease could lead to the discovery of novel biomarkers for disease progression or therapeutic response. Further experimental validation, perhaps using in vitro co-culture models or in vivo studies with specific genetic manipulations, could then focus on deciphering the precise functional consequences of these interactions. For instance, investigating how specific NRXN-NLGN interactions are affected in E280A Alzheimer's models could shed light on early synaptic dysfunction.
  3. Understanding Disease Mechanisms: The Control CCI profile provides a crucial reference. Deviations in glutamatergic signaling, synaptic adhesion, or neuro-glial support in neurodegenerative conditions (like E280A Alzheimer's) could pinpoint key molecular mechanisms underlying neuronal dysfunction, cognitive decline, or demyelination. For example, reduced Neurexin-Neuroligin interactions might suggest impaired synapse formation or maintenance, which is a hallmark of many neurological disorders.
  4. Pharmacological Modulations: The specific ligand-receptor pairs identified offer direct targets for pharmacological modulation. For instance, drugs targeting specific metabotropic glutamate receptors (like GRM3) or pathways involving NRG3-ERBB4 could be developed to enhance neuronal function, reduce inflammation, or promote myelination in conditions where these pathways are compromised.

8. Condition-Specific Cell-Cell Interaction Patterns in Brain Tissue

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates statistically significant differences in cell-cell interactions (CCI) across three conditions: Control, E280A (a familial form of Alzheimer's Disease), and Sporadic (sporadic Alzheimer's Disease). The interactions were assessed among a defined set of cell types, including Microglia, Astrocyte, Endothelial cell, Fibroblast, and Smooth muscle cell, interacting with themselves or other cell types present in the brain tissue (e.g., Neurons, Oligodendrocytes). The visualization highlights the top 25 most differentially interacting ligand-receptor pairs for each condition, ordered by significance, to reveal condition-specific communication landscapes.

Visual Summary

The dot plot effectively visualizes condition-specific CCI patterns, where dot color intensity represents the standardized mean interaction strength (darker red indicates stronger interaction) and dot size represents the statistical significance (-log10(p-value), larger dots indicate higher significance).

Note on Fibroblast and Smooth muscle cell interactions: While these cell types were included in the target list for the analysis, specific interactions explicitly labeled with "Fibroblast" or "Smooth muscle cell" were not among the top 25 most significant items displayed in this plot. This suggests their differential interactions, if any, were either less significant than the presented ones or involved in different pathways not captured in these top entries.

Biological Interpretation

The observed condition-specific CCI patterns provide critical insights into the underlying cellular mechanisms in different forms of Alzheimer's Disease (AD).

Clinical or Translational Implications

Targeted Therapeutic Strategies

9. Microglia Condition-Specific Surfaceome Markers in Alzheimer's Disease

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies condition-specific surfaceome markers in Microglia cells, comparing Control, E280A (familial Alzheimer's Disease), and Sporadic (sporadic Alzheimer's Disease) conditions. The plot_markers_and_expression_dot tool was used to visualize the expression of the top 50 surfaceome markers for each condition. Surfaceome markers are particularly relevant as they represent proteins accessible on the cell surface, making them potential targets for diagnostic imaging, cell sorting, or therapeutic interventions.

Visual Summary

The dot plot visualizes the expression of selected surfaceome genes across individual samples, grouped by condition (Control, E280A, Sporadic). Each row represents a sample, and each column represents a gene.

Three distinct clusters of highly expressed genes are evident, enclosed by red boxes, corresponding to each condition:

  1. Control-specific markers: A cluster of genes on the left shows high expression and prevalence primarily in Control samples (e.g., *CX3CR1, MRC1, PMEPA1, MRC2, MILR1, EPHB2, SLC29A3, SLC26A3, SUSD3, SELPLG, LYVE1*).
  2. E280A-specific markers: A central cluster of genes is prominently expressed in E280A samples (e.g., *DSCAM, OLR1, OPRM1, PLP1, CD163*). Some overlap with Control is seen for *LINGO1*, but it appears stronger in Controls. *CD163* also appears to be shared with Sporadic AD.
  3. Sporadic-specific markers: A rightmost cluster of genes shows high expression and prevalence mainly in Sporadic samples (e.g., *SLC2A9, ADGRE2, PLXNC1, TNFRSF13B, PLB1, PTPRG, ESR1*). *CD163* expression is also prominent in Sporadic AD.

Overall, the plot clearly demonstrates distinct surfaceome profiles for Microglia across the three conditions, suggesting condition-specific microglial states.

Biological Interpretation

The identified surfaceome markers highlight significant shifts in microglial phenotypes associated with different AD forms and healthy aging. Microglia are key immune cells in the brain, and their activation states are highly dynamic, influencing disease progression.

Control Microglia Phenotype

E280A Microglia Phenotype

Sporadic Microglia Phenotype

Clinical or Translational Implications

The identification of condition-specific surfaceome markers in Microglia offers several promising clinical and translational avenues:

  1. Diagnostic and Prognostic Biomarkers: These surface markers could serve as a panel for distinguishing different forms of AD (E280A vs. Sporadic) from healthy controls. For example, high *CX3CR1* levels might indicate a healthier state, while increased *OLR1* or *ESR1* could be indicative of disease progression. These could potentially be detected via advanced imaging techniques (e.g., PET ligands targeting these receptors) or in biofluids if they are shed.
  2. Therapeutic Targets: Given their surface localization, these proteins are excellent candidates for targeted drug delivery or immunomodulatory therapies.
  1. Experimental Validation and Disease Modeling: These markers can be used to isolate specific microglial populations (e.g., using flow cytometry or magnetic bead sorting) from brain tissue or iPSC-derived models to study their functional differences in vitro. This would enable deeper mechanistic studies into how these distinct microglial states contribute to AD pathophysiology.
  2. Understanding Disease Heterogeneity: The distinct markers for E280A and Sporadic AD highlight potential differences in underlying microglial pathology between familial and sporadic forms of the disease. This could lead to more personalized treatment strategies tailored to the specific microglial dysfunction observed.

References:

[1] CX3CR1 GeneCards. https://www.genecards.org/cgi-bin/carddisp.pl?gene=CX3CR1

[2] MRC1 (CD206) UniProt. https://www.uniprot.org/uniprot/P22897

[3] OLR1 GeneCards. https://www.genecards.org/cgi-bin/carddisp.pl?gene=OLR1

[4] CD163 UniProt. https://www.uniprot.org/uniprot/Q86VB7

[5] SLC2A9 GeneCards. https://www.genecards.org/cgi-bin/carddisp.pl?gene=SLC2A9

[6] Estrogen Receptor Alpha and Microglial Activation PubMed search. https://pubmed.ncbi.nlm.nih.gov/?term=estrogen+receptor+alpha+microglia+activation

10. Endothelial Cell Gene Ontology (GSA) Analysis in Alzheimer's Disease Conditions

Report figure

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Ontology (GO) enrichment results for endothelial cells, comparing three conditions (Control, E280A, and Sporadic) against the other conditions within the dataset. Endothelial cells in the brain form the critical blood-brain barrier (BBB) and play vital roles in nutrient transport, waste removal, and regulating neuroinflammation. The goal is to identify biological pathways that are significantly upregulated in endothelial cells specific to each condition, providing insights into their cellular state and functional alterations in the context of Alzheimer's Disease (AD). The AnnData context indicates that E280A is likely referring to the *PSEN1* E280A mutation, a known cause of early-onset familial AD, while Sporadic refers to the more common form of late-onset AD.

Visual Summary

The bar plots illustrate the top 60 enriched GO terms (pathways) based on their statistical significance (-log(p-val) and -log(q-val)) for Endothelial cells under different conditions. A higher bar indicates greater statistical significance for the enrichment of that pathway.

Control vs. others

Endothelial cells from Control samples show significant enrichment for fundamental cellular processes such as Ribosome (protein synthesis) and Oxidative phosphorylation (energy metabolism). Interestingly, several neurodegenerative disease-related pathways, including Parkinson disease, Pathways of neurodegeneration, Prion disease, Amyotrophic lateral sclerosis, Alzheimer disease, and Huntington disease, are also highly enriched. This suggests that in healthy individuals, endothelial cells may maintain robust baseline functions or protective mechanisms associated with these processes, which might be altered in disease states.

E280A vs. others

Endothelial cells from E280A samples display enrichment for pathways related to both fundamental cellular functions and a significant inflammatory/stress response. Similar to controls, Ribosome and several neurodegenerative disease pathways (e.g., Parkinson disease, Alzheimer disease) are observed. However, a prominent signature of infectious disease responses emerges, including Coronavirus disease, Shigellosis, Salmonella infection, and Pathogenic Escherichia coli infection. Other enriched terms like Non-alcoholic fatty liver disease and Pathways in cancer point towards metabolic dysregulation and altered growth signaling, while Focal adhesion suggests changes in cell-matrix interactions.

Sporadic vs. others

Endothelial cells from Sporadic samples exhibit a distinct and striking enrichment profile. The most significant term is Pathways in cancer, indicating a strong signal for aberrant cell proliferation or survival mechanisms. Focal adhesion is also highly ranked, alongside critical inflammatory and cell survival signaling pathways such as MAP kinase signaling pathway, TNF signaling pathway, Rap1 signaling pathway, and PI3K-Akt signaling pathway. Furthermore, a broad range of viral and bacterial infection-related pathways are enriched, including Human papillomavirus infection, Human T-cell leukemia virus 1 infection, Hepatitis B, Human immunodeficiency virus 1 infection, Yersinia infection, and Epstein-Barr virus infection. Ubiquitin mediated proteolysis also shows enrichment, suggesting alterations in protein quality control.

Biological Interpretation

Endothelial Cell Homeostasis in Control

The enrichment of Ribosome and Oxidative phosphorylation pathways in Control endothelial cells suggests they are metabolically active and capable of robust protein synthesis, maintaining fundamental cellular functions. The seemingly counter-intuitive enrichment of various neurodegenerative disease pathways (e.g., Alzheimer disease, Parkinson disease) in control samples, when compared to AD conditions, could indicate that healthy endothelial cells actively engage in or possess compensatory mechanisms related to the biological processes implicated in these diseases. Alternatively, it might imply that these fundamental pathways are *downregulated or dysfunctional* in the E280A and Sporadic AD samples, leading to their relative upregulation in controls. This highlights the importance of baseline endothelial function in brain health and potentially in resisting neurodegeneration.

Inflammatory and Stress Responses in E280A AD

Endothelial cells from E280A AD patients show an activated immune and stress response profile. The enrichment of multiple infectious disease pathways suggests these cells are under significant inflammatory stress, possibly responding to pathogen-associated molecular patterns (PAMPs) or damage-associated molecular patterns (DAMPs) characteristic of the AD microenvironment. Such responses can compromise the integrity of the blood-brain barrier (BBB), leading to increased permeability and further neuroinflammation. The inclusion of Non-alcoholic fatty liver disease hints at broader metabolic dysregulation, which is increasingly recognized as a contributor to AD pathogenesis. Changes in Focal adhesion are critical as they directly impact endothelial cell junctions and thus BBB function. The *PSEN1* E280A mutation is known to alter amyloid precursor protein processing, and this inflammatory signature in endothelial cells could be a downstream consequence of increased amyloid-beta pathology [1].

Pro-inflammatory and Dysplastic Signatures in Sporadic AD

The most striking finding in Sporadic AD endothelial cells is the highly significant enrichment of Pathways in cancer, suggesting a dysregulated cellular state that might involve uncontrolled growth, survival, or altered cell cycle. This could reflect a chronic attempt at vascular remodeling or an aberrant proliferative response in a sustained inflammatory environment. The co-enrichment of core inflammatory signaling pathways (e.g., MAP kinase, TNF, PI3K-Akt) further supports a state of chronic neuroinflammation, known to be a key driver in sporadic AD progression [2]. TNF signaling is particularly relevant as it plays a crucial role in BBB dysfunction and neurotoxicity in AD [3]. The wide array of enriched viral/bacterial infection pathways points to a robust and possibly dysregulated immune response by endothelial cells, potentially linked to pathogen exposure, immune senescence, or chronic low-grade infections contributing to AD [4]. Alterations in Ubiquitin mediated proteolysis suggest impairments in protein quality control, a fundamental aspect of AD pathology involving the accumulation of misfolded proteins like amyloid-beta and tau.

Implications for Blood-Brain Barrier (BBB) Function

Across both AD conditions, changes in pathways like Focal adhesion, and the strong inflammatory/infectious disease signatures, strongly implicate a dysfunctional BBB. Endothelial cells are the primary components of the BBB, and their altered state, whether through active inflammation (E280A) or a more proliferative/dysplastic and chronically inflamed state (Sporadic), would inevitably lead to increased BBB permeability, impaired brain homeostasis, and exacerbated neuroinflammation, contributing to neurodegeneration.

Clinical or Translational Implications

The distinct GO profiles in endothelial cells across AD conditions offer crucial insights into disease mechanisms and potential therapeutic targets.

  1. Biomarker Discovery: The specific pathway signatures observed in E280A (infectious/metabolic stress) and Sporadic AD (cancer/proliferation/chronic inflammation) could serve as novel endothelial cell-specific biomarkers for diagnosing or monitoring disease progression in different AD subtypes.
  2. Targeting Neuroinflammation: The prominent inflammatory and infectious disease signatures in both AD conditions highlight endothelial cells as key mediators of neuroinflammation. Therapeutic strategies aimed at modulating endothelial immune responses, reducing inflammation, or restoring BBB integrity could be beneficial. Inhibitors of TNF, MAPK, or PI3K-Akt pathways in endothelial cells might be particularly relevant for sporadic AD.
  3. BBB Restoration: Pathways related to Focal adhesion and general cell-matrix interactions are critical for BBB integrity. Understanding how these pathways are dysregulated in AD could lead to strategies for pharmacologically restoring BBB function, thereby reducing the influx of harmful substances and immune cells into the brain.
  4. Differentiating AD Subtypes: The unique enrichment of "Pathways in cancer" in Sporadic AD endothelial cells, compared to E280A, suggests fundamental differences in cellular responses, which might inform subtype-specific therapeutic approaches.

---

References

  1. PSEN1 E280A mutation and AD pathology: Review on familial Alzheimer's disease mutations and mechanisms.

PubMed Search: PSEN1 E280A Alzheimer's Disease mechanism

  1. Neuroinflammation in sporadic AD: General review on the role of inflammation in Alzheimer's disease.

PubMed Search: Neuroinflammation sporadic Alzheimer's Disease

  1. TNF signaling and BBB in AD: Review on TNF-alpha in Alzheimer's disease and its impact on the blood-brain barrier.

PubMed Search: TNF alpha blood-brain barrier Alzheimer's Disease

  1. Infections and AD: Review articles exploring the link between infections and Alzheimer's disease.

PubMed Search: Infection Alzheimer's Disease pathology

11. 뇌 세포 유형별 유전자 세트 농축 분석 결과 (GSEA)

Report figure

[Analysis Visualization Results]...

Analysis Overview

제공된 분석 결과는 단일 세포 RNA 시퀀싱(scRNA-seq) 데이터를 기반으로 한 유전자 세트 농축 분석(GSEA)의 닷 플롯입니다. 이 분석은 뇌 조직 내 주요 세포 유형인 신경세포(Neuron), 성상교세포(Astrocyte), 미세아교세포(Microglia), 희소돌기아교세포(Oligodendrocyte) 각각에서 세 가지 질병 조건(E280A, Sporadic)과 대조군(Control) 간의 유전자 발현 변화가 특정 생물학적 경로에 미치는 영향을 비교합니다. 각 조건은 다른 두 조건을 포함한 "나머지" 그룹과 비교되었습니다. 닷 플롯은 정규화된 농축 점수(NES, 색상: RdBu_r, 양수 NES는 적색, 음수 NES는 청색)와 통계적 유의성(-log10(p-value), 점의 크기)을 통해 경로의 활성화 및 억제 정도를 시각화합니다.

Visual Summary

GSEA 닷 플롯은 E280A 및 Sporadic 질병 상태와 관련된 다양한 세포 유형에서 광범위한 경로 변화를 보여줍니다.

Biological Interpretation

이러한 GSEA 결과는 E280A 및 Sporadic 조건에서 뇌의 세포 유형별로 특이적이고 광범위한 생물학적 변화가 있음을 명확히 보여줍니다.

Clinical or Translational Implications

이러한 발견은 질병 메커니즘을 이해하고 잠재적인 치료 표적을 식별하는 데 중요한 임상적, 번역적 함의를 가집니다.

12. Discussion

The comprehensive single-cell transcriptomic analysis of human brain tissue provides a high-resolution view of the cellular and molecular landscapes altered in familial (E280A) and sporadic Alzheimer's disease (AD). A central finding is the dynamic remodeling of the brain's immune and vascular compartments, demonstrating both shared and distinct pathological signatures across disease etiologies.

Microglia, the brain's resident immune cells, undergo significant phenotypic shifts in both E280A and sporadic AD. Consistent patterns include a marked decrease in homeostatic M0 microglia accompanied by a significant increase in M2b and M2c-like activated states (Sections 5, 6). While M2 microglia are often associated with anti-inflammatory and reparative functions, their sustained presence alongside evidence of increased inflammatory pathways (Section 11, Astrocytes & Microglia GSEA) suggests a dysfunctional M2-like state that may fail to effectively clear pathology or even contribute to chronic neuroinflammation. The specific surfaceome markers identified for microglia, such as increased OLR1 and PLP1 in E280A, and SLC2A9, ADGRE2, TNFRSF13B, and ESR1 in sporadic AD (Section 9), underscore the nuanced and condition-specific activation profiles, suggesting different molecular drivers and consequences of microglial engagement with disease.

A striking and potentially overlooked aspect is the increased presence of Fibroblasts in sporadic AD samples, a population largely absent in controls and E280A (Section 4). While typically sparse in healthy brain parenchyma, their emergence points to significant extracellular matrix (ECM) remodeling, perivascular reactivity, or compromised blood-brain barrier (BBB) integrity, all of which are critical contributors to sporadic AD pathogenesis. This is further corroborated by the prominent endothelial cell-mediated interactions involving Integrins and Collagens in sporadic AD (Section 8), indicating an active and likely pathological reorganization of the neurovascular unit.

Neuronal compartments in both E280A and sporadic AD exhibit a consistent decline in pathways related to synaptic function (e.g., cholinergic, GABAergic synapses, neuroactive ligand-receptor interaction), calcium signaling, and energy metabolism (Section 11). This widespread dampening of critical neuronal processes aligns with the known synaptic loss and functional decline in AD. However, the cell-cell interaction analysis revealed a more generalized reduction in synaptic adhesion (Neurexin-Neuroligin) and GABAergic signaling interactions in E280A (Section 8), implying a broad impairment of neural circuit communication.

The endothelial cell analysis also highlights distinct pathological roles. While control endothelial cells maintain robust metabolic and potentially protective functions, E280A endothelial cells show activation of infectious disease pathways, metabolic dysregulation, and altered focal adhesion (Section 10), indicative of inflammatory stress and BBB compromise. Sporadic AD endothelial cells, in particular, exhibit a highly significant enrichment in "Pathways in cancer," alongside inflammatory (TNF, MAPK, PI3K-Akt) and viral/bacterial infection pathways, suggesting a dysregulated proliferative, pro-inflammatory, and chronically stressed state at the neurovascular interface (Section 10). This "cancer-like" signature in sporadic AD endothelial cells is a notable finding, potentially reflecting aberrant vascular remodeling or a sustained attempt at repair that veers into dysfunctional proliferation.

Finally, widespread disruption of protein homeostasis pathways (Proteasome, Protein processing in ER, Ubiquitin-mediated proteolysis) is observed across multiple cell types in both AD conditions (Section 11, Oligodendrocytes). This shared molecular pathology underscores a fundamental cellular stress response to misfolded protein accumulation, a hallmark of AD, and suggests that therapeutic interventions targeting protein quality control mechanisms could have broad utility. The active ferroptosis pathways in astrocytes, microglia, and oligodendrocytes (Section 11) also point to a specific form of iron-dependent cell death contributing to neurodegeneration in both AD types.

In summary, this study delineates critical cellular adaptations and dysfunctions in AD. Familial AD (E280A) primarily manifests as a broad synaptic and neuronal communication failure, coupled with microglial and endothelial inflammatory responses. Sporadic AD, while sharing microglial activation and protein homeostasis disruption, is uniquely characterized by significant neurovascular remodeling, fibroblast expansion, and a dysplastic endothelial cell phenotype, suggesting a more complex interplay between immune, vascular, and ECM components.

Hypotheses:

  1. Microglial M2-like activation in Alzheimer's disease represents a dysfunctional rather than purely beneficial response, contributing to chronic neuroinflammation and inadequate clearance of pathological aggregates.
  2. The increased presence of fibroblasts and active neurovascular remodeling, specifically the upregulation of endothelial cell-mediated Integrin-Collagen interactions, is a key driver of blood-brain barrier dysfunction and disease progression in sporadic Alzheimer's disease.
  3. The widespread reduction in synaptic adhesion (Neurexin-Neuroligin) and GABAergic signaling interactions is a primary pathological mechanism underlying early synaptic loss and cognitive decline in familial Alzheimer's disease (E280A).
  4. Dysregulation of protein quality control pathways and activation of ferroptosis in neurons, astrocytes, microglia, and oligodendrocytes represent core, shared pathological mechanisms contributing to cell death and neurodegeneration in both familial and sporadic Alzheimer's disease.
  5. Endothelial cells in sporadic Alzheimer's disease adopt a 'cancer-like' proliferative and pro-inflammatory phenotype, actively contributing to aberrant vascular remodeling and neuroinflammation rather than solely reacting to neuronal pathology.

Potential therapeutic targets:

  1. OLR1 (Oxidized Low-Density Lipoprotein Receptor 1): Upregulated in E280A microglia, OLR1 is involved in lipid uptake and inflammation, suggesting its role in lipid dyshomeostasis and amyloid-beta pathology. Modulating OLR1 could impact microglial inflammatory responses and lipid processing. Evidence: Section 9: High expression in E280A-specific microglial surfaceome markers, involved in lipid uptake and inflammation (GeneCards). Validation: Inhibition of OLR1 in AD microglial cell models or E280A animal models to assess its effect on amyloid-beta clearance, inflammatory cytokine production, and neuronal survival. Develop small molecule inhibitors or antibody-based therapies targeting OLR1.
  2. CD163 (Scavenger Receptor Cystein-Rich Type 1 Protein M130): Highly expressed in both E280A and Sporadic M2-like microglia, CD163 marks a reparative or immunosuppressive phenotype. Modulating CD163 activity could re-balance microglial states to enhance beneficial functions (e.g., phagocytosis) or dampen detrimental chronic inflammation. Evidence: Section 9: Prominent expression in E280A and Sporadic-specific microglial surfaceome markers. Section 5, 6: Increased M2-like microglia in AD. Validation: Targeted delivery of agonists or antagonists to CD163 in AD models to evaluate effects on microglial phagocytosis, inflammation resolution, and overall neuroprotection. Flow cytometry-based sorting of CD163+ microglia for in-depth functional studies.
  3. Integrin-Collagen interactions in Endothelial cells: Robust upregulation of Integrin-Collagen interactions (e.g., COL4A1_integrin_a10b1_complex) in sporadic AD endothelial cells indicates significant neurovascular unit remodeling and compromised BBB integrity. Targeting these interactions could stabilize the BBB and reduce harmful influx into the brain. Evidence: Section 8: Prominent Integrin-Collagen interactions in Sporadic condition. Section 10: Focal adhesion (cell-matrix interactions) enriched in Sporadic endothelial cells. Validation: Develop small molecule inhibitors or neutralizing antibodies against specific integrin subunits or collagen-integrin binding sites to test their efficacy in restoring BBB integrity and reducing neuroinflammation in sporadic AD models. Utilize in vitro BBB models to screen candidate compounds.
  4. Ferroptosis pathway components (e.g., GPX4, FSP1): Ferroptosis-related pathways are significantly increased in astrocytes, microglia, and oligodendrocytes in both E280A and sporadic AD. Inhibiting ferroptosis could mitigate iron-dependent cell death in multiple key brain cell types, thereby slowing neurodegeneration. Evidence: Section 11: 'Ferroptosis' pathway is significantly increased in Astrocytes, Microglia, and Oligodendrocytes in E280A and Sporadic conditions. Validation: Test known ferroptosis inhibitors (e.g., ferrostatin-1, liproxstatin-1) or novel compounds targeting key ferroptosis regulators (e.g., GPX4, FSP1) in AD cell culture models or animal models to assess effects on cell survival, neuroinflammation, and cognitive function.

Follow-up validation ideas:

  1. Utilize high-resolution spatial transcriptomics or multiplexed immunostaining to validate the spatial localization of increased fibroblasts in sporadic AD brains and their proximity to vascular structures or amyloid plaques.
  2. Perform functional perturbation experiments in iPSC-derived microglial models from AD patients to assess the impact of modulating OLR1, CD163, or ESR1 expression on phagocytic activity, cytokine secretion, and neurotoxicity.
  3. Employ in vitro co-culture systems of neurons and endothelial cells to study the functional consequences of altered Integrin-Collagen interactions on BBB integrity and neuronal health under AD-like conditions, using targeted gene knockdown or overexpression.
  4. Conduct targeted proteomic analysis (e.g., mass spectrometry) to quantify levels of key protein quality control components (e.g., proteasome subunits, ER chaperones) and ferroptosis markers (e.g., lipid peroxidation products, iron accumulation) in specific cell types isolated from AD brain tissue.
  5. Validate the reduced synaptic interactions in E280A AD using electron microscopy to quantify synapse density or in vivo electrophysiology to assess synaptic plasticity in AD animal models carrying the E280A mutation.

Limitations:

This single-cell RNA sequencing analysis provides valuable insights into cellular and molecular changes in Alzheimer's disease. However, it presents several limitations. The cross-sectional nature of the data does not allow for inference of causality or the dynamic progression of cellular changes over time. While detailed cell type annotations are provided, some 'unassigned' cells remain, and the exact functional states of identified cell subsets (e.g., M2 microglia subtypes) require further functional validation beyond transcriptional profiling. The observed compositional shifts and pathway enrichments might be influenced by regional variations within the brain or specific stages of disease progression represented in the sampled cohorts. Furthermore, single-cell dissociation can introduce stress-induced gene expression changes or lead to the loss of fragile cell types. The identified cell-cell interactions and pathway alterations are inferred from gene expression levels and require experimental validation to confirm functional significance and direct physical interactions.

13. Query List

  1. Show UMAPs colored by condition, sample, celltype_major, celltype_minor, and celltype_subset in 2 columns, and save them.
  2. Show major celltype scores on UMAP and save it.
  3. Show the overall Celltype_subset marker expression dot plot and save it. Keep target_cell as None and keep the other arguments at their default values.
  4. Show population bar plot for minor cell types and save it.
  5. Show a subset population barplot for Microglia and save it.
  6. If there are statistically significant differences in Microglia subset population between conditions, show a boxplot and save it. Set ncols appropriately based on the total number of panels.
  7. Show cell-cell interactions by condition and save the result. Select at most 80 cell-cell interactions for each condition.
  8. Find statistically significant differences in cell-cell interactions between conditions for Microglia, Astrocyte, Endothelial cell, Fibroblast, Smooth muscle cell, show a dot plot, and save it. Set max_n_items_per_group = 25.
  9. Extract condition-specific markers for Microglia, show a dot plot, and save it. Include only surfaceome markers, up to 50 per condition.
  10. Show gene ontology (GSA) analysis results bar plot for Endothelial cell and save it.
  11. Show Gene Set Enrichment Analysis results dot plot for Neuron, Astrocyte, Microglia, and Oligodendrocyte, and save it. Set color map to RdBu_r and n_pws_to_show = 80.
Top