SCODiA Report by MLBI Lab

Single-Cell Atlas of Human Pancreas Reveals Dynamic Cellular and Molecular Changes in Type 1 Diabetes Progression

This single-cell analysis of the human pancreas investigates cellular and molecular alterations associated with Type 1 Diabetes (T1D) and autoantibody positivity. We reveal robust cell type identification and observe an expanded cell-cell interaction network in autoantibody-positive individuals, suggesting early compensatory or dysregulatory responses. In established T1D, there is evidence of beta cell loss, extensive pancreatic inflammation affecting both endocrine and exocrine compartments, and significant stromal remodeling, providing a comprehensive view of disease progression.

Contents

0. Dataset overview
1. UMAP Visualization of Pancreatic Cell Atlas by Condition, Sample, and Cell Type Hierarchies
2. Major Cell Type Score Visualization and Annotation Overview
3. Marker Gene Expression Validation for Pancreatic Cell Subsets
4. Pancreatic Minor Cell Type Population Analysis Across Diabetic Conditions
5. Pancreatic Cell-Cell Interaction Patterns Across Type 1 Diabetes Progression Stages
6. Condition-Specific Cell-Cell Interaction Patterns in Pancreatic Cells Across Type 1 Diabetes Progression
7. Beta Cell Condition-Specific Surfaceome Markers in Pancreas
8. Gene Ontology (GSA) Analysis for Alpha Cells in Type 1 Diabetes Progression
9. Gene Set Enrichment Analysis (GSEA) of Pancreatic Cell Types in Type 1 Diabetes and Autoimmunity
10. Discussion
11. Query List

0. Dataset overview

Dataset Summary

• This AnnData object contains single-cell RNA-seq data with 87,200 cells and 30,501 genes.
• The data is from human Pancreas tissue.
• Key observational columns (obs) include sample, donor, condition, sex, celltype_major, celltype_minor, and celltype_subset, along with quality control metrics.
• Gene information (var) includes gene_symbol and gene_id.
• Experimental conditions are type1_diabetes, non_diabetic, and autoantibody_positive.
• Cell type annotations are available at major, minor, and subset levels, covering cell types such as Acinar cell, Ductal cell, Stromal cell, Alpha cell, Beta cell, and Endothelial cell.
• The reference condition for differential analyses is 'non_diabetic'.

Precomputed Results

• Cell-Cell Interaction (CCI): Results are available per condition and per sample.
• Differential Expression Genes (DEG): Results are precomputed for each celltype_minor, comparing conditions against the rest of the dataset or against the 'non_diabetic' reference condition.
• Gene Set Enrichment Analysis (GSEA): Results are available for each celltype_minor, comparing conditions against the rest of the dataset or against the 'non_diabetic' reference condition.
• Gene Ontology (GO/GSA): Results are precomputed for each celltype_minor, comparing conditions against the rest of the dataset or against the 'non_diabetic' reference condition.
• CNV estimates are not available in this dataset.

Inferred Biological Context

• The dataset is relevant to Type 1 Diabetes and related autoimmunity, indicated by conditions like 'type1_diabetes' and 'autoantibody_positive' within the pancreas tissue, which contains affected cell types like 'Alpha cell' and 'Beta cell'.

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

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Uniform Manifold Approximation and Projection (UMAP) plots of single-cell RNA-sequencing data from the human pancreas. These visualizations project high-dimensional gene expression data into a 2D space, revealing underlying cellular relationships. The UMAPs are colored by various metadata attributes including condition (disease status), sample (individual donor), celltype_major, celltype_minor, and celltype_subset (hierarchical cell type annotations), allowing for an assessment of data integration quality, cell type resolution, and condition-specific patterns.

Visual Summary

The UMAPs provide a comprehensive overview of the cellular landscape of the pancreas, highlighting the distribution of cells based on different categorical annotations:

• Condition UMAP: The plot colored by condition (autoantibody_positive, non_diabetic, type1_diabetes) shows considerable overlap among the three conditions across the entire embedding space. While cells from all conditions are broadly distributed, there is no strong, global separation of distinct clusters solely driven by condition. This suggests that the fundamental cellular architecture captured by the UMAP is largely preserved across these conditions, though subtle shifts or localized enrichments might exist.
• Sample UMAP: This plot displays cells colored by individual sample identifiers. The mixing of various sample colors throughout the different cell clusters indicates good integration of data from different donors. This suggests that potential batch effects originating from individual samples or experiments have been largely mitigated during data processing, and the UMAP primarily reflects biological rather than technical variation.
• Celltype_major UMAP: This plot reveals distinct and well-separated clusters corresponding to major pancreatic cell types. Acinar cells form a very large, dense cluster, dominating the central and lower-right regions. Alpha and Beta cells cluster distinctly in the upper-left, reflecting their endocrine islet cell identity. Ductal, Endothelial, and Stromal cells also form clearly defined, separate clusters. The 'unassigned' cells are sparsely distributed, indicating most cells have received a robust major cell type label.
• Celltype_minor UMAP: This UMAP provides a finer resolution of cell types, refining the categories observed in celltype_major. Within the broader 'Stromal cell' cluster, distinct populations such as Fibroblasts, Stellate cells, and Smooth muscle cells (SMC) emerge as well-defined sub-clusters. The main islet (Alpha, Beta), acinar, ductal, and endothelial cell clusters remain coherent and distinct, demonstrating consistent and robust annotation at this level of granularity.
• Celltype_subset UMAP: This highest level of annotation detail further resolves specific cell populations, particularly within the endothelial compartment. Lymphatic Endothelial cells (Endo Lymp) and Endothelial tip cells (Endo tip) are now discernible as distinct, smaller clusters, differentiating them from the general 'Endo' population. Other core cell types like Acinar, Alpha, Beta, Ductal, Fibroblast, SMC, and Stellate cells retain their clear boundaries as observed in the celltype_minor plot.

Biological Interpretation

The UMAP visualizations provide crucial insights into the cellular composition and organization of the human pancreas in the context of Type 1 Diabetes and related autoimmunity.

• Robust Cell Type Identification: The clear separation of cell populations across celltype_major, celltype_minor, and celltype_subset UMAPs confirms high-quality cell type annotation. The hierarchical nature of these annotations (e.g., distinguishing general Endothelial cells into Lymphatic and Tip cells) is biologically meaningful and allows for detailed analysis of specific cellular functions. For instance, the distinct clustering of Alpha and Beta cells, the primary cell types affected in Type 1 Diabetes, validates their accurate identification, which is critical for downstream differential expression or cell-cell interaction analyses.
• Islet Cell Composition: The well-defined Alpha and Beta cell clusters are consistent with the known distinct endocrine functions of these cell types within the pancreatic islets. The proximity of Alpha and Beta cells on the UMAP reflects their co-localization within islets and potential functional interactions.
• Pancreatic Microenvironment: The presence and distinct clustering of Acinar, Ductal, Stromal (Fibroblast, Stellate, SMC), and Endothelial cells highlight the diverse cellular ecosystem of the pancreas. These non-islet cells play critical roles in supporting islet function, extracellular matrix remodeling, and inflammatory responses, which are highly relevant in the context of autoimmune diseases like Type 1 Diabetes.
• Condition-Agnostic Global Architecture: The substantial overlap of type1_diabetes, non_diabetic, and autoantibody_positive cells across the UMAP indicates that the major cell type composition and their global transcriptional states are not drastically altered to the point of forming entirely separate UMAP clusters based solely on disease condition. This implies that while Type 1 Diabetes involves significant pathological changes, the fundamental cellular identities and their general relationships within the pancreas remain largely recognizable. Any condition-specific effects are likely more subtle, involving changes in cell proportions, cell state shifts within existing clusters, or differential gene expression rather than wholesale alterations of cellular populations, which would manifest as distinct condition-driven clusters. Further analysis, such as differential gene expression or cell-cell interaction analysis within specific cell types, would be necessary to uncover these more nuanced, condition-specific changes.
• Data Integration Quality: The effective mixing of cells from different sample origins across the UMAP demonstrates successful data integration. This is important for ensuring that observed biological differences are truly related to disease conditions or cell types rather than technical variations introduced by different donors or experimental batches.

Annotation Notes

The UMAP plots demonstrate high-quality data integration and robust cell type annotation. The clear separation of cell populations at major, minor, and subset levels, along with good mixing of samples, provides a solid foundation for further in-depth biological investigations. The 'unassigned' cells represent a minor fraction, suggesting a high confidence in most cell type assignments. Future analyses should focus on identifying subtle condition-specific changes within these well-defined cell populations.

2. Major Cell Type Score Visualization and Annotation Overview

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the "HiCAT_major_score" for various major cell types across the UMAP embedding, alongside the final celltype_major assignments. The purpose is to assess the spatial distribution and distinctness of cell types based on their computed scores and to confirm the quality and consistency of the assigned major cell type annotations within the single-cell RNA-seq dataset from the Pancreas, with particular attention to conditions related to Type 1 Diabetes and autoimmunity.

Visual Summary

The UMAP plots illustrate the distribution of cells highly scoring for specific major cell types.

Pancreatic Parenchymal Cells:

• Acinar cell scores are highest in a large, central-lower region of the UMAP, indicating a dominant population.
• Ductal cell scores are concentrated in a distinct cluster positioned above the Acinar cell cluster.
• Alpha cell, Beta cell, Delta cell, Epsilon cell, and Gamma (PP) cell scores are strongly localized to several tightly clustered regions on the left side of the UMAP, which are characteristic of pancreatic islet cell populations. These endocrine cells form distinct, albeit sometimes adjacent, clusters.

Stromal and Endothelial Cells:

• Stromal cell scores are prominent in a region adjacent to the main Acinar cell cluster, towards the upper-left, suggesting their supportive role within the tissue.
• Endothelial cell scores are also localized to a distinct, smaller cluster to the upper-left, often co-localizing with stromal components as expected for vascular structures.

Immune Cells:

• T cell, B cell, Myeloid cell, and Mast cell scores are generally lower and appear in relatively smaller, more diffuse clusters, often at the periphery of the main cell mass or within specific minor branches. This suggests their less abundant or more scattered presence compared to the major pancreatic cell types.

Other Cell Types:

• Pancreatic progenitor cell and Schwann cell scores are also localized to specific, smaller regions within the UMAP, indicating the presence of these less abundant cell populations.

• Overall Annotation Consistency: The final celltype_major UMAP plot, colored by assigned cell type, shows clear and largely non-overlapping clusters that correspond well with the regions of high scores for each individual HiCAT_major_score plot. For instance, the large burgundy cluster representing Acinar cells in celltype_major matches the high-score region for HiCAT_major_score: Acinar cell. Similarly, the distinct endocrine cell clusters (Alpha, Beta, Delta, Epsilon, Gamma) are well-represented by distinct colors in the celltype_major plot and show high scores in their respective UMAP regions.

Biological Interpretation

The UMAP plots demonstrate a robust and well-resolved clustering of major cell types from the Pancreas tissue, consistent with known pancreatic cellular heterogeneity.

• Pancreatic Architecture: The clear separation of endocrine islet cells (Alpha, Beta, Delta, Epsilon, Gamma) from exocrine cells (Acinar, Ductal) on the UMAP reflects the distinct functional and anatomical compartments of the pancreas. The close proximity and intermingling of different islet cell types, yet their distinct clustering, are biologically accurate, reflecting the multi-hormonal nature of the islets.
• Annotation Quality: The high degree of concordance between the individual cell type scores and the final celltype_major annotations indicates a high quality of cell type assignment. Cells belonging to a specific major type show strong enrichment for their corresponding score, with minimal spill-over into other cell type score regions. This suggests that the gene expression profiles used to define these scores are highly specific to each cell population.
• Immune Infiltration: The presence of various immune cell types (T, B, Myeloid, Mast cells) is noteworthy, particularly in the context of the inferred Type 1 Diabetes and autoimmunity. Their clear, albeit smaller, clusters confirm their identification and suggest their potential involvement in disease pathology, which warrants further investigation into their specific states and interactions. For instance, T cells are central to the autoimmune destruction of beta cells in Type 1 Diabetes [PubMed Search: T cells type 1 diabetes pancreas].

Annotation Notes

• The visualizations clearly demonstrate that the celltype_major annotations are well-supported by the underlying HiCAT cell type scores. Each major cell type forms a distinct and coherent cluster or set of clusters in the UMAP embedding.
• The embedding structure effectively separates distinct pancreatic cell lineages, highlighting the success of the dimensionality reduction and clustering steps.
• Minor overlap or diffusion of scores is minimal, suggesting a strong discriminatory power of the cell type markers used for scoring. The 'unassigned' category in celltype_major is relatively small and appears to comprise cells with lower or ambiguous scores for specific major types, which is an expected outcome in complex single-cell datasets.

3. Marker Gene Expression Validation for Pancreatic Cell Subsets

[Analysis Visualization Results]...

Analysis Overview

This dot plot provides a visualization of marker gene expression across different celltype_subset populations derived from single-cell RNA-seq data of the human pancreas. Each row represents a distinct cell type, while each column corresponds to a specific gene. The intensity of the dot color signifies the mean expression level of a gene within a given cell group, and the size of the dot indicates the fraction of cells within that group expressing the gene. Marker genes that are significantly enriched in each cell type are grouped together and highlighted by red bounding boxes. This analysis is crucial for evaluating the quality and specificity of the assigned cell type annotations by examining how well the observed gene expression patterns align with known biological markers.

Visual Summary

The dot plot effectively showcases the distinct and, in most cases, highly specific expression profiles of marker genes across the various celltype_subset populations.

• Acinar cells (70181 cells): These cells exhibit robust and highly specific expression of a broad panel of acinar markers, including PNLIPRP1, CPA2, PRSS1, CTRB1, and numerous other digestive enzyme-related genes (e.g., CELA3A, AMY2B). The high mean expression (dark red) and large fraction of expressing cells (large dots) for these markers, coupled with their minimal presence in other cell types, strongly supports the acinar cell annotation.
• Alpha cells (2310 cells): Alpha cells are clearly identified by the strong and specific expression of *GCG* (glucagon), a canonical marker for this endocrine population. Other associated markers like RFX6 and SMARCA1, along with the neuroendocrine marker *CHGA*, also show distinct expression patterns, reinforcing the alpha cell identity.
• Beta cells (2111 cells): This population displays exceptionally high and specific expression of *INS* (insulin) and *SLC30A8* (zinc transporter 8), which are quintessential markers for pancreatic beta cells. Additional markers such as CASR, STXBP5, and PCSK2 further corroborate the beta cell annotation with unique expression profiles.
• Ductal cells (8838 cells): Ductal cells are well-characterized by the specific expression of *CFTR* (Cystic Fibrosis Transmembrane Conductance Regulator) and ANXA4, both established markers for pancreatic ductal epithelium. The transcription factor HNF1B also shows strong and specific expression, consistent with ductal cell identity.
• Fibroblasts (958 cells): This mesenchymal population demonstrates a characteristic profile of extracellular matrix (ECM) components, including DCN (Decorin), COL1A1, COL1A2, COL3A1, COL5A1, and FBLN1. These genes are strongly expressed in fibroblasts with high specificity, validating their annotation.
• Stellate cells (1168 cells): Pancreatic stellate cells, another mesenchymal type, share some expression patterns with fibroblasts, particularly for ECM genes like COL6A2, COL1A2, DCN, COL5A1, COL3A1, and FBLN1. However, *PDGFRB* also shows specific enrichment in stellate cells, helping to distinguish them from general fibroblasts.
• Endothelial tip cells (46 cells) and Lymphatic Endothelial cells (79 cells): These two smaller endothelial populations exhibit specific, though sometimes overlapping, marker patterns.
  • Lymphatic Endothelial cells are robustly identified by the highly specific and strong expression of *PROX1* and *PDPN*, which are canonical markers for the lymphatic endothelium.
  • Endothelial tip cells show expression of various collagen genes (COL4A1, COL6A3, COL6A2) and growth factor receptors like PDGFRA and PDGFRB. However, some of these markers (e.g., collagens) are also expressed in fibroblasts and stellate cells, indicating a potential overlap or less unique specificity. PROX1 and PDPN are notably absent or minimally expressed in "Endothelial tip cells," confirming their distinction from lymphatic endothelial cells. Given the small number of cells in this group, the robustness of some of these markers might be limited.

Biological Interpretation

The analysis of marker gene expression provides strong biological evidence affirming the assigned identities of most major pancreatic cell types.

1. High Confidence in Exocrine and Endocrine Identities: The clear and highly specific expression of well-established markers for Acinar, Alpha, Beta, and Ductal cells provides high confidence in these annotations. The distinction between glucagon-producing alpha cells and insulin-producing beta cells by their respective *GCG* and *INS* expression is biologically fundamental for islet biology [PubMed search: Pancreatic islet cell markers]. The distinct exocrine functions of acinar cells (digestive enzymes) and ductal cells (CFTR for fluid and bicarbonate secretion) are also clearly reflected in their gene expression profiles.
2. Mesenchymal Cell Heterogeneity and Overlap: The expression patterns for Fibroblasts and Stellate cells are consistent with their roles in producing and remodeling the extracellular matrix. While they share several ECM-related genes, which is expected given their mesenchymal origin and function, the subtle differences, such as the relatively higher expression of PDGFRB in stellate cells, contribute to their distinct annotation. Pancreatic stellate cells are key players in pancreatic fibrosis due to their ECM production [GeneCards: PDGFRB].
3. Endothelial Subtype Annotation Considerations: The annotation of Lymphatic Endothelial cells is well-supported by the strong and specific expression of *PROX1* and *PDPN*, both recognized lymphatic markers [UniProt: PROX1]. For the "Endothelial tip cell" population, while some general endothelial and ECM-related markers are observed, the panel does not present highly unique or canonical "tip cell" specific markers that are entirely distinct from other endothelial or mesenchymal populations. The low cell count (46 cells) for this group also warrants cautious interpretation regarding the exclusivity of its marker profile. Further investigation with more specific, known "tip cell" markers, or re-evaluating if it represents a specific state rather than a distinct lineage, could enhance the confidence in this annotation.

Annotation Notes

The celltype_subset annotations for Acinar cell, Alpha cell, Beta cell, Ductal cell, Fibroblast, and Stellate cell are strongly supported by their highly specific and canonical marker gene expression patterns, indicating high confidence in these cell type assignments.

For the "Lymphatic Endothelial cell" annotation, the specific expression of *PROX1* and *PDPN* provides robust evidence for its identity. However, the "Endothelial tip cell" annotation, given its small cell count and the observed overlap of its highlighted markers with general endothelial or even other mesenchymal populations, appears less definitively supported by uniquely specific markers in this visualization. While it might represent a functional state, its precise and exclusive identity based solely on the presented markers could benefit from further validation or refinement with more targeted marker analysis.

4. Pancreatic Minor Cell Type Population Analysis Across Diabetic Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis provides a detailed view of the relative proportions of different minor cell types across individual pancreatic samples, grouped by their clinical condition: 'autoantibody_positive', 'non_diabetic' (reference), and 'type1_diabetes'. This population bar plot is critical for understanding the cellular landscape of the pancreas and identifying potential shifts in cellular composition associated with Type 1 Diabetes and related autoimmune processes.

Visual Summary

The visualization consists of three stacked bar charts, each representing a specific condition. Within each chart, individual bars correspond to distinct samples (e.g., MM_556, MM_426), displaying the percentage contribution of each minor cell type to the total cellularity of that sample.

• Dominant Cell Type: Acinar cells (dark red) are consistently the most abundant cell type across all samples and conditions, frequently comprising over 70% of the total cells. This aligns with the known composition of the exocrine pancreas.
• Islet Cell Proportions: Alpha cells (red) and Beta cells (orange), key endocrine cells of the pancreas, are present in smaller proportions, as expected given their distribution within the islets of Langerhans.
• Other Pancreatic Cells: Ductal cells (light orange) are also consistently observed in moderate proportions. Endothelial cells (yellow), Fibroblasts (light yellow), Smooth muscle cells (light green), and Stellate cells (teal) constitute smaller fractions, reflecting the vascular and stromal components of the tissue.
• Unassigned Cells: A small percentage of cells are categorized as 'unassigned' (blue), indicating cells that did not clearly cluster with the defined minor cell types.
• Sample-to-Sample Variability: Significant variability in the proportions of less abundant cell types is evident across individual samples within each condition, highlighting the biological heterogeneity among donors.

Biological Interpretation

The observed cell type proportions provide important insights into pancreatic biology in the context of Type 1 Diabetes:

• Expected Pancreatic Composition: The overwhelming dominance of Acinar cells confirms the identity of the tissue as pancreas and reflects the primary function of the exocrine pancreas in digestive enzyme production [GeneCards: PNLIP]. This serves as a robust internal control for tissue sampling and annotation.
• Islet Cell Changes in Type 1 Diabetes: Type 1 Diabetes is characterized by the autoimmune destruction of insulin-producing Beta cells. While not statistically quantified here, visual inspection suggests that the proportion of Beta cells (orange segment) might be lower in some samples from the 'type1_diabetes' and 'autoantibody_positive' groups compared to 'non_diabetic' controls. For example, several T1D samples (e.g., MM_355, MM_358) appear to have particularly reduced Beta cell percentages. However, there is also variability within the 'non_diabetic' group. Alpha cells, while not the primary target in T1D, can also show altered proportions or function in the disease context.
• Stromal and Vascular Components: The presence and proportions of stromal cells (Fibroblasts, Stellate cells, Smooth muscle cells) and Endothelial cells can reflect the overall tissue microenvironment. Changes in these populations could indicate processes like inflammation, fibrosis, or altered vascularization associated with disease progression or autoimmune attack [PubMed search: pancreatic fibrosis type 1 diabetes].
• Heterogeneity: The considerable sample-to-sample variability in cell type proportions underscores the importance of considering individual biological differences and potential technical variations when interpreting single-cell data.

Clinical or Translational Implications

• Disease Biomarker Potential: Quantitative assessment of Beta cell proportions, particularly relative to other islet cells (Alpha cells) or total pancreatic cells, could serve as a valuable biomarker for Type 1 Diabetes progression or the effectiveness of immunomodulatory therapies. This plot provides a qualitative indication that warrants rigorous statistical follow-up.
• Pathogenic Mechanisms: Deviations from the normal cellular composition, such as a reduction in Beta cell mass, are central to the pathogenesis of Type 1 Diabetes [PubMed search: beta cell mass type 1 diabetes pathogenesis]. Further analysis focusing on the statistical significance of these changes across conditions would be crucial.
• Understanding Autoimmunity: The 'autoantibody_positive' condition represents a pre-diabetic or early stage of the disease. Observing cellular shifts in these individuals could provide insights into the early immunological and cellular events preceding overt diabetes.
• Limitations: This visualization provides relative proportions, not absolute cell counts. Changes in relative percentages might be influenced by changes in the absolute number of other cell types. Detailed statistical analysis is required to confirm significant differences between conditions.

5. Pancreatic Cell-Cell Interaction Patterns Across Type 1 Diabetes Progression Stages

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interaction (CCI) patterns within the human pancreas across three conditions: 'non_diabetic', 'autoantibody_positive', and 'type1_diabetes'. Utilizing CellPhoneDB results, dot plots visualize the strength (mean expression, color) and significance (p-value, dot size) of ligand-receptor interactions between various pancreatic cell types. The focus is on identifying key interaction changes that may characterize the progression of Type 1 Diabetes (T1D), with an emphasis on ligand-receptor biology relevant for therapeutic targeting and experimental validation.

Visual Summary

The provided dot plots illustrate cell-cell interaction landscapes for 'autoantibody_positive', 'non_diabetic', and 'type1_diabetes' conditions, displaying at most 80 significant interactions per group.

• 'autoantibody_positive' Condition: This plot exhibits the most extensive and diverse network of cell-cell interactions. Notably, it includes numerous interactions involving Beta cells (e.g., Beta|Beta, Beta|Alpha, Beta|Acinar, Acinar|Beta) and Alpha cells (e.g., Alpha|Alpha, Alpha|Acinar, Acinar|Alpha), alongside interactions between Acinar and Ductal cells. Many of these interactions demonstrate high mean expression (indicated by yellow/bright green colors) and strong statistical significance (larger dot sizes). Key ligand-receptor pairs appearing prominent in this state include several BMPs (BMP5, BMP1R, BMP2, BMP4), NRG2/3/4-ERBB3/4, FGF10-FGFR2, JAG1-NOTCH2, LAMC1_integrin_a6b1_complex, Cholesterol_byCEL_RORA, and various Glutamate-related interaction pathways.
• 'non_diabetic' Condition: This plot presents a much sparser interaction profile. The top 80 interactions are predominantly confined to Ductal|Ductal, Ductal|Acinar, Acinar|Ductal, and Acinar|Acinar cell pairs. Significantly, there are no interactions involving Beta or Alpha cells among the most prominent interactions shown in this healthy control group. Notable interactions include BMPR1B_BMPR2, Cholesterol_byCEL_RORA (particularly strong in Acinar|Acinar), and FGF10_FGFR2.
• 'type1_diabetes' Condition: Similar to the 'non_diabetic' condition, the 'type1_diabetes' plot also shows a relatively sparse interaction landscape, primarily involving Ductal and Acinar cell interactions. Crucially, Beta cell and Alpha cell interactions are absent from the top 80 most significant interactions in this condition, mirroring the 'non_diabetic' state in this regard. Specific interactions like APP_SORL1 (Ductal|Ductal) and various TGFB2-related ligand-receptor pairs emerge, alongside persistent BMPR1B_BMPR2, Cholesterol_byCEL_RORA, and FGF10_FGFR2 interactions also seen in the non-diabetic state.

Biological Interpretation

The contrasting CCI landscapes provide crucial insights into the pancreatic microenvironment in health, pre-diabetes, and overt T1D.

1. Expanded Interaction Network in Autoantibody Positivity: The 'autoantibody_positive' condition stands out with a significantly richer and more diverse set of cell-cell interactions, particularly involving Beta and Alpha cells. This suggests a highly active and potentially dysregulated microenvironment during the pre-diabetic or early autoimmune phase. The presence of numerous Beta cell-mediated interactions (e.g., Beta|Beta, Beta|Alpha, Beta|Acinar) at this stage is particularly significant, as Beta cells are the primary targets of autoimmune destruction in T1D. This expanded network could represent:
   • Early stress response: Beta cells might be actively communicating with their neighbors (Alpha, Acinar, and other Beta cells) in response to autoimmune attack or metabolic stress, attempting to adapt or repair.
   • Immune cell recruitment/activation: While not directly visible here (as immune cells are not explicitly cell types in the plot's axes), these altered local interactions could modulate the recruitment or activity of immune cells, contributing to disease progression.
   • Dysfunctional communication: Some of these "activated" interactions might be maladaptive, contributing to Beta cell dysfunction or susceptibility.
2. Loss of Beta Cell-Centric Interactions in Overt T1D and Non-Diabetic States: The striking absence of Beta cell-involved interactions among the top 80 in both 'non_diabetic' and 'type1_diabetes' conditions suggests distinct scenarios.
   • In the 'non_diabetic' state, the top interactions are primarily focused on maintaining the structural and functional integrity of the exocrine pancreas (Acinar and Ductal cells). Beta cell-specific communication might be subtle or less dominant than the high-magnitude interactions of the exocrine compartment, hence not appearing in the top 80.
   • In 'type1_diabetes', the absence of Beta cell interactions among the top hits could reflect significant Beta cell loss due to autoimmune destruction. If Beta cells are severely depleted or functionally impaired, their contribution to the overall CCI landscape would diminish. This shift implies a profound reorganization of the pancreatic islet microenvironment following disease onset.
3. Role of Specific Ligand-Receptor Systems:
   • NRG-ERBB signaling (e.g., NRG2/3/4-ERBB3/4): Prominent in 'autoantibody_positive' Beta cell interactions. ERBB receptors are involved in cell proliferation, survival, and differentiation. Dysregulated NRG-ERBB signaling could indicate Beta cell stress, attempts at regeneration, or altered survival pathways in early autoimmunity. PubMed Search: NRG ERBB pancreas diabetes
   • BMP signaling (e.g., BMPR1B_BMPR2): Consistently present across conditions, particularly in Ductal and Acinar cells. BMPs are crucial for pancreatic development and maintenance. Alterations in BMP signaling could impact tissue homeostasis. GeneCards: BMPR1B
   • FGF10-FGFR2: Observed across all conditions, primarily between Acinar and Ductal cells. FGF10 is involved in pancreatic development and regeneration. Its persistence might reflect ongoing maintenance or compensatory regenerative processes. UniProt: FGF10
   • TGFB2-TGFBR1/2: Emerges in the 'type1_diabetes' condition. TGF-beta signaling is a critical regulator of immune responses, fibrosis, and cell survival. Its activation in T1D could contribute to islet inflammation, fibrosis, or attempts at immune regulation. PubMed Search: TGFB2 type 1 diabetes pancreas

Clinical or Translational Implications

1. Early Biomarkers of Disease Progression: The distinct and expanded CCI network observed in 'autoantibody_positive' individuals, particularly those involving Beta cells, could serve as a rich source of novel biomarkers for early detection or prediction of T1D onset. Monitoring the activity of specific ligand-receptor pairs (e.g., NRG-ERBB, specific BMPs) in accessible biological fluids or even via advanced imaging could indicate early pancreatic stress or autoimmune activity.
2. Therapeutic Target Identification:
   • Modulating Beta cell interactions in autoantibody positivity: Interventions aimed at restoring healthy Beta cell communication or mitigating maladaptive interactions (e.g., through ERBB receptor antagonists/agonists, or targeting specific glutamate pathways) could potentially slow down or prevent Beta cell destruction in autoantibody-positive individuals.
   • Targeting inflammation and fibrosis in T1D: The emergence of TGFB2-TGFBR1/2 interactions in overt T1D suggests that modulating this pathway could be a therapeutic strategy to reduce inflammation or fibrosis within the remnant pancreatic tissue, potentially preserving residual Beta cell function or aiding transplantation efforts. PubMed Search: TGFB signaling type 1 diabetes therapy
3. Experimental Validation: The identified critical ligand-receptor pairs and cell-cell interactions warrant further experimental validation.
   • _In vitro_ co-culture models: Co-culturing pancreatic cell types (Beta, Alpha, Acinar, Ductal) under various stress conditions (e.g., inflammatory cytokines) could validate the functional significance of selected interactions observed in the 'autoantibody_positive' state.
   • _Ex vivo_ human islet studies: Analyzing human islets from donors with different T1D stages could confirm the CCI changes and their impact on islet function and survival.
   • Genetically engineered mouse models: Utilizing mouse models of T1D (e.g., NOD mice) to manipulate specific ligand-receptor pathways identified (e.g., NRG-ERBB, TGFB2) could provide _in vivo_ evidence for their roles in disease pathogenesis and as therapeutic targets.

6. Condition-Specific Cell-Cell Interaction Patterns in Pancreatic Cells Across Type 1 Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies cell-cell interactions (CCIs) that significantly differ between 'autoantibody_positive', 'non_diabetic', and 'type1_diabetes' conditions within the human pancreas. The plot_dot_for_cci_with_signif_difference tool was used to visualize these interactions, focusing on major pancreatic cells (Acinar, Ductal, Alpha, Beta) and Stromal cells (Fibroblast, Stellate). The plot displays the scaled log strength of interaction (color intensity) and its statistical significance (-log10(p-value), dot size) for each sample and condition. The analysis aimed to uncover condition-specific communication patterns relevant to Type 1 Diabetes and autoimmunity.

Visual Summary

The dot plot effectively illustrates distinct patterns of cell-cell communication across the three conditions.

• Overall Variability: A substantial number of unique cell-cell interactions exhibit significant differences across the conditions, indicating dynamic changes in pancreatic cellular communication during disease progression.

Condition-Specific Clusters:

• autoantibody_positive samples (left panel, particularly the upper block, highlighted by the blue box) show a high frequency of strong and significant interactions, including various COL-integrin pairs, APP-TNFRSF21, PTPRF-LRFN5, and particularly prominent INS-INSR and Glutamate signaling interactions involving Beta cells. This suggests an active remodeling and communication network potentially preceding overt diabetes.
• non_diabetic samples (middle panel, notably the middle blue box) exhibit a diverse range of strong interactions. Many of these overlap with those seen in the autoantibody_positive group (e.g., INS-INSR), but also include specific WNT4-FZD6, CDH1-integrin, TENM3-ADGRL2, and TGFB2-TGFbeta_receptor interactions. This represents the baseline communication network in healthy pancreatic tissue.
• type1_diabetes samples (right panel, particularly the lower block, highlighted by the blue box) demonstrate a more heterogeneous pattern. While some interactions are dampened compared to other conditions (e.g., many INS-INSR interactions are notably weaker or absent), a distinct set of interactions appears strong and significant. These include specific COL-integrin (Fibroblast-Fibroblast), JAG1-NOTCH2 (Acinar-Alpha), NRG2-ERBB4 (Alpha-Alpha), CXCL12-DPP4 (Fibroblast-Alpha), GAS6-AXL (Ductal-Beta), FGF7-FGFR2 (Stellate-Ductal), and ADCYAP1-VIPR2 (Beta-Beta) interactions. This indicates a shift in cellular communication reflecting beta cell stress, loss, and potential compensatory or pathological remodeling within the diabetic pancreas.

• Sample Heterogeneity: Within each condition, there is still notable sample-to-sample variability, suggesting different disease stages or individual responses.

Biological Interpretation

The observed condition-specific CCI patterns provide crucial insights into the pathophysiology of Type 1 Diabetes (T1D) and its autoimmune precursor state.

1. Impaired Islet Homeostasis: Reduced INS-INSR Signaling in T1D
   • Interaction: INS-INSR (Beta cell | Beta cell, Beta cell | Alpha cell)
   • Condition Pattern: Strong and highly significant in autoantibody_positive and non_diabetic samples, but strikingly weaker or absent in many type1_diabetes samples.
   • Signaling/Mechanism: This represents autocrine and paracrine insulin signaling within and between islet endocrine cells. Insulin (INS) binding to its receptor (INSR) on beta cells (and to a lesser extent alpha cells) is critical for glucose metabolism, cell survival, proliferation, and maintaining islet architecture.
   • T1D Relevance: The diminished INS-INSR signaling in T1D samples directly reflects the progressive loss of functional beta cells, a hallmark of the disease. This impacts not only systemic glucose regulation but also local islet cell communication and survival. The relatively preserved signal in autoantibody-positive individuals might indicate remaining beta cell mass and function, even amidst early autoimmune attack.
2. Stromal Remodeling and Fibrosis: Altered Integrin-Collagen Interactions
   • Interaction: COL1A1_integrin_a1b1_complex (Fibroblast | Fibroblast) and other COL-integrin pairs involving Stromal cells (Fibroblast, Stellate).
   • Condition Pattern: Prominent and strong in the type1_diabetes condition, particularly in specific samples, compared to non_diabetic and autoantibody_positive groups.
   • Signaling/Mechanism: Integrins are cell surface receptors mediating cell-extracellular matrix (ECM) interactions. Collagens are major components of the ECM. These interactions are fundamental for cell adhesion, migration, proliferation, and differentiation, contributing to tissue architecture and mechanotransduction.
   • T1D Relevance: Increased stromal-stromal (fibroblast/stellate) interactions via collagens and integrins in T1D suggest enhanced ECM remodeling and pancreatic fibrosis. This fibrotic process can encapsulate and further damage islets, impairing beta cell function and access to nutrients and oxygen. PubMed search: pancreatic fibrosis type 1 diabetes integrin
3. Islet Cell Plasticity and Regeneration: JAG1-NOTCH2 Signaling

Interaction: JAG1_NOTCH2 (Acinar cell | Alpha cell)

• Condition Pattern: Strong and distinct in a subset of type1_diabetes samples.
• Signaling/Mechanism: Notch signaling, activated by ligands like Jagged-1 (JAG1), is a critical pathway for cell fate determination, proliferation, and differentiation during development and regeneration.
• T1D Relevance: Altered Notch signaling between acinar and alpha cells in T1D may indicate attempts at cellular plasticity or compensatory regeneration (e.g., acinar-to-islet cell transdifferentiation) in response to beta cell loss. However, dysregulation could also contribute to maladaptive processes or impaired islet architecture. PubMed search: Notch signaling pancreas regeneration type 1 diabetes

4. Immune Cell Homing and Regulation: CXCL12-DPP4 Axis

Interaction: CXCL12_DPP4 (Fibroblast | Alpha cell)

• Condition Pattern: Notably strong in type1_diabetes samples.
• Signaling/Mechanism: CXCL12 (SDF-1) is a potent chemokine that attracts immune cells (T cells, B cells) to sites of inflammation by binding to its receptor CXCR4. DPP4 (CD26) is a cell surface peptidase that cleaves and inactivates CXCL12, thus regulating its bioavailability and the chemotactic gradient. This interaction likely represents stromal regulation of local chemokine activity or alpha cell-mediated immune responses.
• T1D Relevance: The CXCL12-CXCR4 axis is critical for immune cell trafficking and infiltration into the pancreatic islets, a central event in T1D pathogenesis. Altered activity of the CXCL12-DPP4 axis could modulate the recruitment and retention of autoreactive immune cells, or influence alpha cell function and survival in the inflammatory milieu. GeneCards: CXCL12, GeneCards: DPP4

5. Beta Cell Stress Response and Survival: ADCYAP1-VIPR2 Signaling

Interaction: ADCYAP1_VIPR2 (Beta cell | Beta cell)

• Condition Pattern: Strong and present in type1_diabetes samples.
• Signaling/Mechanism: ADCYAP1 (Pituitary adenylate cyclase-activating polypeptide, PACAP) is a neuropeptide that binds to VIPR2 (VPAC2 receptor). This signaling typically activates adenylate cyclase, increasing intracellular cAMP, and is known for its neuroprotective, anti-inflammatory, and insulinotropic effects.
• T1D Relevance: The strong ADCYAP1-VIPR2 interaction in T1D samples might represent an endogenous compensatory mechanism to protect remaining beta cells from apoptosis and inflammatory damage, or an adaptive response to stress. PACAP is known to enhance beta cell survival and function. GeneCards: ADCYAP1, GeneCards: VIPR2

Clinical or Translational Implications

The identification of condition-specific cell-cell interactions offers potential avenues for therapeutic intervention in Type 1 Diabetes:

Targeting Beta Cell Survival and Function:

• The marked reduction in INS-INSR signaling in T1D underscores the urgent need for therapies that preserve or restore beta cell mass and function. While exogenous insulin is standard, strategies aimed at enhancing endogenous beta cell survival (e.g., anti-apoptotic agents, beta cell growth factors) or promoting regeneration could be transformative. GLP-1 receptor agonists, currently used in type 2 diabetes, are being explored for beta cell preservation in T1D due to their ability to improve insulin sensitivity and beta cell function. PubMed search: GLP-1 receptor agonists type 1 diabetes beta cell
• The sustained ADCYAP1-VIPR2 signaling in T1D beta cells suggests an intrinsic protective mechanism. PACAP and its analogues are being investigated for their therapeutic potential in diabetes due to their cytoprotective and insulinotropic effects on beta cells, which could be exploited to bolster remaining beta cells.

Modulating Pancreatic Fibrosis:

• The prominence of COL-integrin interactions in T1D indicates significant stromal remodeling. Therapies targeting specific integrins or collagen synthesis pathways could mitigate pancreatic fibrosis, thereby preserving islet integrity and function. Anti-fibrotic agents, currently in development or use for other fibrotic diseases (e.g., pirfenidone, nintedanib), might warrant investigation in T1D.

Controlling Immune Infiltration:

• The increased CXCL12-DPP4 interaction in T1D samples highlights a potential mechanism influencing immune cell homing. DPP4 inhibitors (e.g., Sitagliptin, Saxagliptin), approved for type 2 diabetes, could be explored in T1D to modulate immune cell infiltration by altering CXCL12 bioavailability and potentially impacting islet inflammation and destruction. PubMed search: DPP4 inhibitors type 1 diabetes immunomodulation

Harnessing Regenerative Potential:

• The strong JAG1-NOTCH2 signaling in T1D suggests active modulation of cell plasticity. While Notch signaling is complex and can have dual roles, targeted manipulation could potentially promote beta cell regeneration from ductal or acinar precursors. However, therapeutic modulation of Notch pathways requires careful consideration due to their pleiotropic effects.

7. Beta Cell Condition-Specific Surfaceome Markers in Pancreas

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies surfaceome markers specifically expressed or upregulated in Beta cells under different conditions related to Type 1 Diabetes (T1D): 'autoantibody_positive' (pre-diabetic/risk state) and 'type1_diabetes' (established disease), with 'non_diabetic' serving as the reference. The visualization, a dot plot, presents the mean expression level (color intensity) and the fraction of cells expressing each marker (dot size) across individual samples (donors). The markers displayed were specifically selected based on their differential expression (nz_pct_score and nz_pct_FC criteria) when compared to the 'non_diabetic' condition or other conditions, aiming to pinpoint surface proteins relevant to the autoimmune processes and disease progression in T1D.

Visual Summary

The dot plot reveals a heterogeneous yet insightful landscape of surfaceome marker expression in pancreatic Beta cells across the studied conditions and individual donors.

• Heterogeneity: Significant variability in marker expression is observed both between the 'autoantibody_positive' and 'type1_diabetes' conditions, and among individual samples within each group, indicating diverse cellular responses or disease stages.
• 'autoantibody_positive' Signature (Upper Left Block): A subset of samples within the 'autoantibody_positive' group shows notable upregulation of certain markers. Genes such as IL6R, LRRTM3, SLC8A1, LDLR, and SGCE display increased expression and prevalence across several samples. IL6R is particularly prominent, showing high mean expression and presence in a large fraction of Beta cells in samples like MM_357, MM_407, and MM_554. This suggests an early inflammatory or stress response.
• 'type1_diabetes' Signature (Lower Right Block, Notably MM_555, MM_406, MM_401, MM_405): A pronounced cluster of samples within the 'type1_diabetes' condition exhibits robust and widespread upregulation of a broad panel of surfaceome markers. Markers like LSAMP, TMEM132D, SLC8A1, EFNA5, CNTN5, NALCN, PTPRK, NCAM1, and many others in this dense region show very high mean expression (dark red) and are expressed in a large fraction of Beta cells (large dot size) in these specific donors. This indicates a more extensive phenotypic shift or a stronger immune-driven response in Beta cells during established T1D.
• Shared Markers: Some markers, including TMEM132D and SLC8A1, appear to be elevated in both 'autoantibody_positive' and 'type1_diabetes' conditions, potentially reflecting shared or progressive pathophysiological mechanisms common to both stages.
• Cell Counts: The bar chart on the right indicates the total number of cells contributing to the data for each sample, providing context for the observed expression patterns.

Biological Interpretation

The differential expression of these surfaceome markers in Beta cells offers critical insights into the molecular and cellular mechanisms underpinning Type 1 Diabetes and its autoimmune prodrome. The 'autoantibody_positive' state appears to capture early immunological and cellular alterations, while established T1D is characterized by a more diverse and pronounced set of changes, indicative of advanced pathology.

Here are biological interpretations for a selection of prominent markers:

IL6R (Interleukin-6 Receptor)

• Signaling Pathways/Mechanisms: IL6R is the receptor for Interleukin-6 (IL-6), a pleiotropic cytokine pivotal in inflammation and immune responses. Upon IL-6 binding, IL6R activates the JAK/STAT (Janus kinase/Signal Transducer and Activator of Transcription) pathway, leading to gene expression changes that mediate inflammatory and immune responses.
• Pancreatic/T1D Context: IL-6 is a well-documented pro-inflammatory cytokine in T1D pathogenesis. Its elevated expression in the pancreatic microenvironment during insulitis contributes to beta cell dysfunction and apoptosis. The observed upregulation of IL6R on Beta cells, especially in autoantibody-positive individuals, suggests that Beta cells are directly responsive to the inflammatory signals, potentially exacerbating their stress and contributing to their demise.
• Reference: GeneCards - IL6R

LRRTM3 (Leucine Rich Repeat Transmembrane Neuronal 3)

• Signaling Pathways/Mechanisms: LRRTM proteins are neuronal transmembrane proteins known to play roles in excitatory synapse formation and cell adhesion through interactions with presynaptic neurexins. They modulate cell-cell recognition.
• Pancreatic/T1D Context: While primarily studied in the nervous system, Beta cells share some neuronal characteristics and participate in complex intercellular communication within the islet. Upregulation of LRRTM3 could indicate altered cell adhesion properties, changes in Beta cell plasticity, or modified interactions with neighboring cells (e.g., other islet cells, nerves, immune cells) in response to autoimmune stress, potentially affecting islet integrity or function.
• Reference: GeneCards - LRRTM3

LSAMP (Limbal Sensory Axon Guidance Related Protein)

• Signaling Pathways/Mechanisms: LSAMP is a neuronal cell adhesion molecule belonging to the immunoglobulin superfamily. It is involved in axon guidance and neuronal differentiation, often through homophilic or heterophilic cell surface interactions, affecting cell adhesion and motility.
• Pancreatic/T1D Context: The pronounced upregulation of LSAMP in Beta cells from some T1D individuals points to significant alterations in Beta cell identity, adhesion characteristics, or interactions with their microenvironment. This could impact islet architecture, cellular communication, and susceptibility to immune-mediated destruction. Such changes may also reflect a stress-induced shift in beta cell phenotype.
• Reference: GeneCards - LSAMP

EFNA5 (Ephrin A5)

• Signaling Pathways/Mechanisms: EFNA5 is a membrane-bound ligand that interacts with EphA receptor tyrosine kinases. Eph/ephrin signaling is a crucial bidirectional cell-cell communication system regulating processes like cell adhesion, repulsion, migration, and tissue patterning (e.g., during development and angiogenesis). It influences cytoskeletal dynamics and downstream signaling cascades (e.g., MAPK, Akt pathways).
• Pancreatic/T1D Context: Eph/ephrin interactions are vital for normal pancreatic development and maintaining islet architecture. Upregulation of EFNA5 in T1D Beta cells suggests potential disruption in intra-islet communication, altered Beta cell migration, or modified interactions with endothelial cells (impacting vascularization) or infiltrating immune cells. This could contribute to the pathological remodeling of the islet microenvironment during autoimmune insulitis.
• Reference: GeneCards - EFNA5

SLC8A1 (Solute Carrier Family 8 Member A1 / NCX1)

• Signaling Pathways/Mechanisms: SLC8A1, also known as Sodium/Calcium Exchanger 1 (NCX1), is a pivotal plasma membrane transporter that maintains intracellular calcium (Ca2+) homeostasis by exchanging one Ca2+ ion for three sodium (Na+) ions. This process is critical for various cellular functions, including hormone secretion, muscle contraction, and neurotransmission.
• Pancreatic/T1D Context: Calcium signaling is fundamental for glucose-stimulated insulin secretion (GSIS) in Beta cells. Dysregulation of intracellular calcium dynamics is a hallmark of Beta cell dysfunction in diabetes. The consistent upregulation of SLC8A1 in both pre-diabetic and diabetic states could represent a compensatory mechanism by stressed Beta cells to manage altered calcium levels (e.g., by increasing Ca2+ efflux) or, conversely, it could contribute to an imbalance in Ca2+ homeostasis, leading to impaired GSIS or increased susceptibility to apoptosis.
• Reference: GeneCards - SLC8A1

NCAM1 (Neural Cell Adhesion Molecule 1)

• Signaling Pathways/Mechanisms: NCAM1 is a cell surface glycoprotein of the immunoglobulin superfamily involved in homophilic and heterophilic cell-cell interactions. It modulates signaling pathways crucial for cell growth, differentiation, and survival, often by interacting with fibroblast growth factor receptor (FGFR) and activating Fyn kinase.
• Pancreatic/T1D Context: NCAM1 is expressed on pancreatic Beta cells and plays a role in islet development, morphology, and maintaining Beta cell integrity. Its upregulation in T1D Beta cells might indicate stress-induced phenotypic changes, altered cell-cell adhesion within the islet, or changes in how Beta cells interact with infiltrating immune cells. NCAM1 expression can also be associated with Beta cell dedifferentiation, a phenomenon observed in diabetes where Beta cells lose their mature functional identity.
• Reference: GeneCards - NCAM1

Clinical or Translational Implications

The identified surfaceome markers offer promising avenues for therapeutic intervention, disease monitoring, and deepening our understanding of Type 1 Diabetes pathogenesis and its autoimmune precursors.

IL6R as a Therapeutic Target for Inflammation:

• The robust upregulation of IL6R in Beta cells, particularly in autoantibody-positive individuals, underscores the critical role of IL-6-mediated inflammation in driving early Beta cell pathology in T1D.
• Therapeutic Potential: Blocking IL-6 signaling, for example, using antibodies against IL6R (such as tocilizumab, an agent approved for other autoimmune diseases like rheumatoid arthritis), represents a strong therapeutic strategy. Such an approach could potentially mitigate inflammation, reduce Beta cell stress, and help preserve Beta cell function and mass in individuals at risk or in early stages of T1D. Clinical trials targeting IL-6 in T1D are ongoing or have been explored, highlighting its clinical relevance.
• Therapeutic Agents: Tocilizumab (Actemra) is a leading example of an anti-IL6R antibody. Other IL-6 pathway inhibitors may also be considered. PubMed search for Tocilizumab Type 1 Diabetes

SLC8A1 (NCX1) for Beta Cell Calcium Homeostasis Modulation:

• The consistent upregulation of SLC8A1 suggests a persistent perturbation in calcium homeostasis within Beta cells across both pre-diabetic and diabetic states.
• Therapeutic Potential: Modulating NCX1 activity to restore optimal calcium dynamics in Beta cells could theoretically improve glucose-stimulated insulin secretion and Beta cell survival. However, NCX1 plays a complex, context-dependent role (calcium efflux vs. influx), necessitating careful investigation to ensure any targeted modulation achieves beneficial rather than detrimental effects.
• Therapeutic Agents: While specific NCX1 modulators are not currently in clinical use for T1D, strategies aimed at restoring calcium homeostasis in Beta cells are an active area of diabetes research.

• Adhesion Molecules (LRRTM3, LSAMP, EFNA5, NCAM1) as Biomarkers and Modulators of Islet Architecture:
  • The observed changes in adhesion molecules like LRRTM3, LSAMP, EFNA5, and NCAM1 highlight significant alterations in Beta cell interaction, plasticity, and potentially dedifferentiation during T1D progression.
  • Therapeutic Potential: While these markers may not be immediate direct therapeutic targets, they offer valuable insights for novel strategies:
    • Restoring Islet Architecture: Modulating the activity of pathways involving these adhesion molecules, such as Eph/ephrin (EFNA5) or NCAM1 signaling, could potentially help maintain or restore the structural integrity of islets and their crucial interactions with the microenvironment. This might protect Beta cells from immune attack or enhance their regenerative capacity.
    • Biomarkers: Changes in the expression of these surface markers could serve as early biomarkers for Beta cell stress, dedifferentiation, or the progression of autoimmune destruction. These could potentially be detected through non-invasive means like circulating exosomes or advanced imaging techniques.
  • Therapeutic Agents: Currently, no specific agents directly targeting these adhesion molecules are approved for T1D, but research into cell adhesion modulation is a broad field with implications for various diseases.

In summary, the detailed surfaceome profiling of Beta cells provides a window into the dynamic changes occurring during T1D pathogenesis, identifying key inflammatory drivers and fundamental alterations in cellular physiology and interactions. These findings are crucial for developing targeted therapies that address specific facets of Beta cell pathology in Type 1 Diabetes.

8. Gene Ontology (GSA) Analysis for Alpha Cells in Type 1 Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Ontology (GO) enrichment results for Alpha cells, identifying biological processes and pathways that are significantly upregulated in specific conditions. The analysis compares Alpha cells from individuals with autoantibody positivity (autoantibody_positive) and individuals with confirmed Type 1 Diabetes (type1_diabetes) against a non-diabetic control group (non_diabetic). Additionally, a comparison of non_diabetic Alpha cells against a pooled 'other' group (comprising type1_diabetes and autoantibody_positive individuals) is included to highlight pathways characteristic of the healthy state. The results are based on gene set enrichment analysis (GSA) using up-regulated differentially expressed genes (DEGs) and are visualized as bar plots, with bars representing the negative logarithm of the p-value and q-value.

Visual Summary

The three bar plots display the top enriched Gene Ontology terms based on upregulated genes in Alpha cells for different comparisons:

1. GSA for Alpha cell: autoantibody_positive_vs_non_diabetic: This plot highlights pathways upregulated in Alpha cells from autoantibody-positive individuals compared to non-diabetic controls. Prominent terms with high significance (high -log(p-val) and -log(q-val)) include "Neurotrophin signaling pathway," "ErbB signaling pathway," "Dopaminergic synapse," "Chemokine signaling pathway," "Insulin signaling pathway," and "Ubiquitin mediated proteolysis." Several key cellular signaling cascades like "mTOR signaling pathway," "PI3K-Akt signaling pathway," and "MAPK signaling pathway" are also significantly enriched.
2. GSA for Alpha cell: non_diabetic_vs_others: This plot shows pathways upregulated in non-diabetic Alpha cells compared to a combined group of autoantibody-positive and type 1 diabetic individuals. The most significant terms include fundamental cellular processes such as "Spliceosome," "Protein processing in endoplasmic reticulum," "RNA transport," and "Ribosome biogenesis in eukaryotes." "Pancreatic secretion" and "Notch signaling pathway" are also highly ranked, indicating robust baseline cellular functions in healthy Alpha cells.
3. GSA for Alpha cell: type1_diabetes_vs_non_diabetic: This plot focuses on pathways upregulated in Alpha cells from Type 1 Diabetes patients compared to non-diabetic controls. Similar to the autoantibody-positive comparison, "Neurotrophin signaling pathway," "Dopaminergic synapse," "Ubiquitin mediated proteolysis," and "ErbB signaling pathway" are highly significant. Interestingly, terms related to viral infections such as "Coronavirus disease" and "Human cytomegalovirus infection" also appear among the top enriched pathways, alongside various hormonal and metabolic signaling pathways like "Insulin secretion," "Thyroid hormone signaling pathway," and "Growth hormone synthesis, secretion and action."

Across all plots, the x-axis represents the significance of enrichment as -log(p-val) and -log(q-val), where larger values indicate higher statistical significance.

Biological Interpretation

The Gene Ontology analysis provides critical insights into the biological changes occurring in pancreatic Alpha cells during the progression of Type 1 Diabetes.

Alpha Cell Dysregulation in Autoimmunity (autoantibody_positive vs non_diabetic)

Alpha cells from autoantibody-positive individuals, representing a pre-diabetic or early autoimmune stage, exhibit significant alterations:

• Immune and Stress Responses: The strong enrichment of the "Chemokine signaling pathway" suggests active immune cell recruitment and inflammatory processes occurring even before overt diabetes. This is consistent with the autoimmune nature of Type 1 Diabetes where immune cells infiltrate the pancreatic islets PubMed search: "chemokine signaling islet inflammation type 1 diabetes". Pathways like "Neurotrophin signaling pathway" and "ErbB signaling pathway" often play roles in cell survival, stress response, and tissue remodeling, indicating Alpha cells are under stress.
• Metabolic and Signaling Perturbations: "Insulin signaling pathway" and "Insulin resistance" are enriched, which is crucial as Alpha cells respond to insulin to modulate glucagon secretion. Dysregulation here can lead to inappropriate glucagon release, contributing to hyperglycemia. Core cellular signaling pathways like "mTOR signaling," "PI3K-Akt signaling," and "MAPK signaling" are also activated, indicating altered cell growth, metabolism, and survival pathways in early autoimmunity.
• Protein Homeostasis: "Ubiquitin mediated proteolysis" being upregulated suggests altered protein turnover and quality control, potentially in response to cellular stress.

Healthy Alpha Cell Function (non_diabetic vs others)

The comparison highlighting pathways upregulated in non-diabetic Alpha cells provides a benchmark for healthy function, indicating processes that may be compromised in diseased states:

• Robust Cellular Machinery: High enrichment for "Spliceosome," "Protein processing in endoplasmic reticulum," "RNA transport," and "Ribosome biogenesis" indicates a strong and efficient protein synthesis and processing machinery in healthy Alpha cells. These fundamental processes are essential for maintaining cell function and responsiveness.
• Pancreatic Specific Functions: "Pancreatic secretion" is highly enriched, reflecting the healthy secretory capacity of these cells (e.g., glucagon). "Notch signaling pathway" is involved in pancreatic development and cell fate, suggesting its normal activity is sustained in healthy Alpha cells.

Alpha Cell Changes in Overt Type 1 Diabetes (type1_diabetes vs non_diabetic)

In established Type 1 Diabetes, Alpha cells show a more pronounced and distinct set of activated pathways:

• Sustained Stress and Neuro-hormonal Dysregulation: Similar to the autoantibody-positive stage, "Neurotrophin signaling pathway," "ErbB signaling pathway," and pathways related to neuronal components ("Dopaminergic synapse," "Axon guidance") are upregulated, suggesting persistent cellular stress responses and potentially altered neural regulation of islet function.
• Altered Protein Turnover and Autophagy: "Ubiquitin mediated proteolysis" and "Autophagy" are enriched, pointing to significant cellular remodeling and quality control mechanisms activated under chronic stress conditions in T1D.
• Metabolic and Hormonal Cross-talk: "Insulin secretion" is observed, which in Alpha cells can indirectly indicate their perturbed metabolic state or interaction with the diseased islet microenvironment, influencing glucagon secretion. Other hormonal signaling pathways like "Thyroid hormone signaling pathway" and "Growth hormone synthesis, secretion and action" hint at broader systemic metabolic influences on Alpha cells.
• Immune/Viral Response: The notable appearance of "Coronavirus disease" and "Human cytomegalovirus infection" pathways suggests a heightened antiviral immune response or the activation of general inflammatory pathways that mimic viral responses. While not necessarily indicating direct viral causation for all T1D cases, viral infections are known triggers for autoimmunity, and these pathways reflect an ongoing immune activation state PubMed search: "viral infection autoimmunity type 1 diabetes".

Clinical or Translational Implications

The GSA results for Alpha cells provide several potential clinical and translational implications:

• Biomarker Discovery: The enrichment of pathways like "Chemokine signaling pathway" and specific stress response pathways (e.g., "Neurotrophin signaling") in autoantibody-positive individuals could identify early markers of Alpha cell dysfunction, potentially aiding in predicting progression to overt Type 1 Diabetes.
• Therapeutic Targets: Understanding the specific dysregulated signaling pathways (e.g., mTOR, PI3K-Akt, MAPK) and stress responses in Alpha cells may open avenues for targeted therapies to preserve Alpha cell function, reduce inappropriate glucagon secretion, or mitigate their contribution to hyperglycemia in Type 1 Diabetes.
• Disease Pathogenesis: These findings underscore that Alpha cells are active participants, not passive bystanders, in Type 1 Diabetes pathogenesis. Their altered function, stress responses, and engagement with immune/inflammatory pathways contribute to the overall islet pathology and metabolic dysregulation, providing a more comprehensive view of the disease beyond solely Beta cell destruction.

9. Gene Set Enrichment Analysis (GSEA) of Pancreatic Cell Types in Type 1 Diabetes and Autoimmunity

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Set Enrichment Analysis (GSEA) results, visualized as a dot plot, comparing various pancreatic cell types under different conditions (autoantibody_positive, type1_diabetes) against a non-diabetic reference. The purpose is to identify pathways and biological processes that are significantly enriched (upregulated) or repressed (downregulated) in these conditions, providing insight into the molecular mechanisms underlying Type 1 Diabetes (T1D) pathogenesis and related autoimmunity. Each dot represents a gene set (pathway), with its color indicating the Normalized Enrichment Score (NES) – red for positive enrichment (upregulation) and blue for negative enrichment (downregulation). The size of the dot reflects the statistical significance (-log(P-value)).

Visual Summary

The dot plot effectively illustrates complex patterns of pathway enrichment and repression across multiple pancreatic cell types and disease states.

• Overall Pattern: There is a clear and widespread pattern of both positive (red) and negative (blue) enrichment across many pathways and cell types.
• Disease-Specific Signatures: The comparisons involving type1_diabetes_vs_non_diabetic and autoantibody_positive_vs_non_diabetic show distinct, often strong, enrichment signals.
• Immune and Inflammatory Pathways: Several immune-related pathways, such as "Antigen processing and presentation", "Chemokine signaling pathway", "TNF signaling pathway", "Th17 differentiation", and "Complement and coagulation cascades," are frequently and strongly enriched (large red dots) across almost all analyzed cell types in type1_diabetes compared to non_diabetic states.
• Metabolic and Functional Pathways: Pathways like "Insulin secretion" and "Pancreatic secretion" show strong negative enrichment (large blue dots) specifically in Beta cells and Acinar/Ductal cells, respectively, in the type1_diabetes condition. "Ribosome" and "Oxidative phosphorylation" pathways are generally downregulated (blue dots) across many cell types in T1D.
• Apoptosis: The "Apoptosis" pathway is notably enriched (red dots) in most cell types during type1_diabetes.
• Differences between Conditions: The autoantibody_positive_vs_non_diabetic comparisons often show similar, but generally weaker, trends in enrichment and repression compared to the type1_diabetes_vs_non_diabetic comparisons, particularly for immune and functional pathways in endocrine cells.
• Cell-Type Specificity: While common inflammatory signatures are observed, some pathways show cell-type specific enrichment, such as the specific downregulation of "Insulin secretion" in Beta cells and "Glucagon signaling pathway" in Alpha cells.

Biological Interpretation

1. Widespread Pancreatic Inflammation and Autoimmunity in T1D: The most striking observation is the consistent and highly significant upregulation of key immune and inflammatory pathways (e.g., "Antigen processing and presentation" [NCBI], "Chemokine signaling pathway" [NCBI], "TNF signaling pathway" [GeneCards]) across almost all pancreatic cell types (Acinar, Alpha, Beta, Ductal, Endothelial, Fibroblast, Stellate cells) in individuals with Type 1 Diabetes compared to non-diabetic controls. This indicates that insulitis, the inflammatory process central to T1D, is not confined to immune cells or even solely to β-cells, but impacts the entire pancreatic microenvironment. Non-endocrine cells appear to be actively responding to or participating in the inflammatory milieu.
2. Beta-Cell Dysfunction and Destruction: As expected in T1D, Beta cells show a profound downregulation of the "Insulin secretion" pathway (large blue dot in Beta cell: type1_diabetes_vs_non_diabetic). Concomitantly, "Apoptosis" is highly enriched in Beta cells, directly reflecting the immune-mediated destruction of insulin-producing cells. The general downregulation of "Ribosome" and "Oxidative phosphorylation" further suggests a decline in overall cellular metabolism and protein synthesis, indicating severe stress and functional impairment in Beta cells.
3. Early Pathological Changes in Autoantibody-Positive Individuals: In autoantibody-positive individuals (a pre-diabetic state), Beta and Alpha cells already exhibit significant changes, albeit generally less pronounced than in overt T1D. For instance, "Antigen processing and presentation" and "Apoptosis" are enriched, while "Insulin secretion" is downregulated in Beta cells, signifying that immune activation and early cell damage begin before clinical diagnosis. This highlights the importance of the autoantibody-positive stage as a window for potential intervention.
4. Involvement of Non-Endocrine Pancreatic Cells: Acinar, Ductal, Endothelial, Fibroblast, and Stellate cells all show strong upregulation of immune and inflammatory pathways in T1D.
   • Acinar and Ductal cells: Upregulation of immune pathways, alongside downregulation of "Pancreatic secretion" and "Protein digestion and absorption," suggests that exocrine pancreatic function is also compromised by the inflammatory environment [PubMed Search].
   • Endothelial cells: Strong immune pathway enrichment indicates their active role in mediating inflammation, potentially facilitating immune cell infiltration and contributing to vascular changes in the pancreas.
   • Fibroblasts and Stellate cells (Stromal cells): Their robust activation of inflammatory pathways suggests a critical role for the stromal compartment in shaping the immune response and potentially contributing to fibrosis or tissue remodeling in the T1D pancreas [PubMed Search].
5. Metabolic Reprogramming and Stress: Beyond specific immune responses, the widespread downregulation of "Ribosome" and "Oxidative phosphorylation" across many cell types suggests a general suppression of protein synthesis and energy metabolism, possibly a broader stress response or energy reallocation in the inflammatory context. The upregulation of "Glycolysis / Gluconeogenesis" in Alpha and Beta cells might reflect a shift in metabolic strategy under stress or altered nutrient availability.
6. Potential Viral Triggers: The occasional enrichment of pathways related to "Viral infections" (e.g., "Herpes simplex virus 1 infection", "Human cytomegalovirus infection") in some cell types within T1D conditions hints at a possible role for viral exposure or persistent viral responses in exacerbating pancreatic inflammation and autoimmune attack, a hypothesis well-documented in T1D research [NCBI].

Clinical or Translational Implications

1. Broadening Therapeutic Targets: The finding that immune and inflammatory pathways are activated across nearly all pancreatic cell types in T1D suggests that therapeutic strategies should extend beyond solely protecting β-cells. Targeting broader inflammatory networks or cellular interactions involving non-endocrine cells (e.g., modulating stromal cell activity or endothelial cell responses to inflammation) could be crucial for more effective disease management and prevention.
2. Early Intervention in Autoantibody-Positive Individuals: The observation of significant pathological changes (e.g., early immune activation and β-cell dysfunction) in autoantibody-positive individuals underscores the value of early identification and intervention in this at-risk population. Therapies aimed at modulating the immune response or protecting β-cell function could potentially halt or slow disease progression before overt diabetes develops.
3. Biomarker Discovery: Pathways consistently and strongly altered (e.g., "Antigen processing and presentation", "Apoptosis", "Insulin secretion") represent potential sources for novel biomarkers to monitor disease activity, progression, or therapeutic response in a cell-type specific manner.
4. Understanding Disease Heterogeneity: The cell-type specific responses provide a foundation for understanding the intricate interplay between different pancreatic cell populations during T1D. This nuanced view can inform the development of more targeted and personalized therapeutic approaches.

10. Discussion

The integrated single-cell analysis of the human pancreas across non-diabetic, autoantibody-positive, and Type 1 Diabetes (T1D) conditions reveals a complex and dynamic landscape of cellular and molecular changes, extending beyond the traditionally emphasized beta cell destruction.

Initial UMAP visualizations and marker expression analyses confirm robust cell type identification and successful data integration, showing that the global cellular architecture of the pancreas remains largely preserved across conditions, though subtle shifts or localized enrichments are evident. The presence of various immune cell types is observed, underscoring their role in the autoimmune context. While Lymphatic Endothelial cells are well-defined by canonical markers, the annotation of 'Endothelial tip cells' warrants cautious interpretation due to their low cell count and potential marker overlap. Analysis of cell type proportions visually suggests a reduction in Beta cells in Type 1 Diabetes and autoantibody-positive samples, highlighting the heterogeneity among donors and the need for rigorous statistical quantification.

A notable finding from cell-cell interaction (CCI) analysis is the expanded and diverse interaction network observed specifically in autoantibody-positive individuals, particularly involving Beta and Alpha cells. This 'activated' communication landscape, featuring interactions like NRG-ERBB, BMP, and Glutamate pathways, suggests a pre-diabetic state characterized by active cellular responses to stress or early autoimmune attack, potentially reflecting attempts at adaptation or repair, or early maladaptive signaling. This is a significant deviation from the sparser interaction profiles seen in both non-diabetic and overt T1D states, highlighting the autoantibody-positive stage as a critical window of dynamic molecular activity.

In established T1D, the CCI analysis reveals a striking reduction or absence of Beta cell-centric interactions (e.g., INS-INSR signaling) among the most significant interactions, directly reflecting the progressive loss and dysfunction of beta cells. Concomitantly, pathways related to stromal remodeling and fibrosis, such as various COL-integrin interactions, become more prominent in T1D, indicating an altered tissue microenvironment. Other condition-specific interactions in T1D, like JAG1-NOTCH2 (Acinar-Alpha), suggest attempts at cellular plasticity or regeneration, while CXCL12-DPP4 (Fibroblast-Alpha) points to modulation of immune cell homing. The sustained ADCYAP1-VIPR2 signaling (Beta-Beta) in T1D might represent an endogenous protective mechanism for remaining beta cells.

Surfaceome analysis of Beta cells further distinguishes disease stages. Upregulation of IL6R in autoantibody-positive Beta cells points to an early inflammatory response. In overt T1D, a broader panel of markers like LSAMP, EFNA5, and NCAM1 are upregulated, indicating significant alterations in Beta cell identity, adhesion, and microenvironment interactions. Consistently, SLC8A1 (NCX1) is elevated in both pre-diabetic and diabetic states, suggesting persistent perturbations in Beta cell calcium homeostasis.

Gene Ontology (GSA) and Gene Set Enrichment Analysis (GSEA) reveal widespread, multi-cellular involvement in T1D pathology. Alpha cells in autoantibody-positive individuals show enrichment in "Chemokine signaling," "Neurotrophin signaling," "ErbB signaling," and core metabolic pathways (mTOR, PI3K-Akt, MAPK), indicative of stress and early inflammation. In overt T1D, Alpha cells exhibit similar stress responses along with altered protein turnover and signs of antiviral responses. The GSEA, encompassing all pancreatic cell types, demonstrates consistent and significant upregulation of core immune and inflammatory pathways (e.g., "Antigen processing and presentation," "Chemokine signaling," "TNF signaling") across Acinar, Alpha, Beta, Ductal, Endothelial, Fibroblast, and Stellate cells in T1D. This emphasizes that insulitis impacts the entire pancreatic microenvironment, not just the islets. Beta cells in T1D consistently show profound downregulation of "Insulin secretion" and upregulation of "Apoptosis," confirming their destruction and functional impairment. Exocrine and stromal cells also exhibit immune activation and metabolic reprogramming, highlighting their active participation in the disease progression. Early pathological changes, including immune activation and beta-cell dysfunction, are already evident in autoantibody-positive individuals, underscoring the critical window for intervention.

Hypotheses:

1. The expanded cell-cell interaction network in autoantibody-positive individuals reflects early, possibly compensatory or dysregulatory, responses of pancreatic cells to autoimmune attack, rather than a quiescent pre-diabetic state.
2. The progressive loss of Beta cells in Type 1 Diabetes is directly correlated with a quantifiable reduction in Beta cell-centric cell-cell interactions and a shift towards increased stromal remodeling and fibrosis within the pancreas.
3. Upregulation of specific surfaceome markers (e.g., IL6R) on Beta cells in autoantibody-positive individuals serves as an early indicator of inflammatory stress, preceding overt Beta cell destruction.
4. Alpha cells are active participants in Type 1 Diabetes pathogenesis, exhibiting distinct stress responses, altered metabolic pathways, and engagement with inflammatory signals, contributing to overall islet dysfunction and hyperglycemia.
5. The widespread activation of immune and inflammatory pathways across both endocrine and exocrine pancreatic cell types in T1D indicates that the disease involves a global inflammatory response impacting the entire pancreatic microenvironment, not just the islets.

Potential therapeutic targets:

1. IL6R: Upregulated on Beta cells in autoantibody-positive individuals, indicating direct responsiveness to pro-inflammatory IL-6. Blocking IL-6 signaling can mitigate inflammation, reduce Beta cell stress, and preserve function. Evidence: IL-6 is a pro-inflammatory cytokine in T1D pathogenesis; its elevated expression contributes to beta cell dysfunction and apoptosis. Clinical trials targeting IL-6 in T1D are ongoing or have been explored. Validation: Tocilizumab (anti-IL6R antibody) is approved for other autoimmune diseases. In vitro studies blocking IL6R on beta cells under inflammatory conditions could validate its protective effect.
2. NRG-ERBB signaling (e.g., NRG2/3/4-ERBB3/4): Prominent in autoantibody-positive Beta cell interactions. Dysregulated NRG-ERBB signaling could indicate Beta cell stress or altered survival pathways in early autoimmunity. Modulating this could slow or prevent Beta cell destruction. Evidence: ERBB receptors are involved in cell proliferation, survival, and differentiation. Validation: In vitro co-culture models of pancreatic cells under stress with ERBB receptor antagonists/agonists. Ex vivo human islet studies.
3. TGFB2-TGFBR1/2 signaling: Emerges in overt T1D. TGF-beta signaling is a critical regulator of immune responses, fibrosis, and cell survival. Modulating this pathway could reduce inflammation or fibrosis, preserving residual Beta cell function. Evidence: TGF-beta signaling contributes to islet inflammation and fibrosis. Validation: In vitro studies on pancreatic cells, ex vivo human islets, or genetically engineered mouse models.
4. Integrin-Collagen interactions (e.g., COL1A1_integrin_a1b1_complex): Prominent in T1D fibroblasts, indicating enhanced ECM remodeling and pancreatic fibrosis, which damages islets. Targeting could mitigate fibrosis, preserving islet integrity and function. Evidence: Pancreatic fibrosis contributes to islet damage in T1D. Anti-fibrotic agents (e.g., pirfenidone, nintedanib) are in development or use for other fibrotic diseases. Validation: Investigation of anti-fibrotic agents in T1D models.
5. CXCL12-DPP4 axis: Strong in T1D Fibroblast-Alpha cell interactions. CXCL12 is a chemokine attracting immune cells, while DPP4 cleaves and inactivates it. Modulating this axis could control immune cell infiltration. Evidence: The CXCL12-CXCR4 axis is critical for immune cell trafficking and infiltration into pancreatic islets in T1D pathogenesis. DPP4 cleaves CXCL12. Validation: DPP4 inhibitors (e.g., Sitagliptin, Saxagliptin), approved for type 2 diabetes, could be explored in T1D to modulate immune cell infiltration.
6. ADCYAP1-VIPR2 signaling: Strong in T1D Beta cells. This signaling, typically activating adenylate cyclase and increasing cAMP, is known for neuroprotective, anti-inflammatory, and insulinotropic effects. Could bolster remaining beta cells from apoptosis and inflammatory damage. Evidence: PACAP (ADCYAP1) is a neuropeptide known to enhance beta cell survival and function. Validation: PACAP and its analogues are being investigated for their therapeutic potential in diabetes due to cytoprotective and insulinotropic effects.
7. SLC8A1 (NCX1): Upregulated in Beta cells in both autoantibody-positive and T1D conditions, suggesting persistent perturbation in calcium homeostasis. Modulating NCX1 activity could improve glucose-stimulated insulin secretion and Beta cell survival. Evidence: Calcium signaling is fundamental for glucose-stimulated insulin secretion (GSIS) in Beta cells; dysregulation of intracellular calcium dynamics is a hallmark of Beta cell dysfunction in diabetes. Validation: While specific NCX1 modulators are not currently in clinical use for T1D, strategies aimed at restoring calcium homeostasis in Beta cells are an active area of diabetes research.
8. Broader Inflammatory/Immune Networks (e.g., Antigen processing and presentation, Chemokine signaling pathway, TNF signaling pathway): GSEA shows widespread activation across multiple pancreatic cell types in T1D. Targeting these broader inflammatory networks, involving non-endocrine cells, could be crucial for more effective disease management and prevention, rather than solely protecting β-cells. Evidence: Insulitis, the inflammatory process central to T1D, impacts the entire pancreatic microenvironment. Non-endocrine cells appear to be actively responding to or participating in the inflammatory milieu. Validation: Clinical trials for TNF inhibitors or chemokine receptor antagonists in T1D are ongoing or have been explored in other autoimmune diseases, supporting the therapeutic strategy.

Follow-up validation ideas:

1. Conduct in vitro co-culture models to validate the functional significance of specific ligand-receptor pairs (e.g., NRG-ERBB, BMP, Glutamate, TGFB2-TGFBR1/2) in pancreatic cell types under inflammatory stress conditions.
2. Perform ex vivo human islet studies from donors with different T1D stages to confirm observed CCI changes and their impact on islet function and survival.
3. Utilize genetically engineered mouse models (e.g., NOD mice) to manipulate specific ligand-receptor pathways (e.g., NRG-ERBB, TGFB2, JAG1-NOTCH2, CXCL12-DPP4, ADCYAP1-VIPR2) in vivo to assess their roles in disease pathogenesis.
4. Investigate the functional impact of modulating IL6R, SLC8A1, LSAMP, EFNA5, or NCAM1 expression on beta cell survival, function, and interactions with immune cells using in vitro functional assays.
5. Validate protein expression of identified surface markers (e.g., IL6R, LSAMP, EFNA5, SLC8A1, NCAM1) in pancreatic tissue or isolated beta cells from human donors at different T1D stages using immunohistochemistry or flow cytometry.
6. Confirm alpha cell functional dysregulation, as indicated by GSA, by measuring glucagon secretion in isolated alpha cells or islets under various metabolic or inflammatory stimuli.
7. Further investigate the 'Endothelial tip cell' annotation with more specific, known markers or re-evaluate if it represents a specific state rather than a distinct lineage, to enhance confidence in its identity.
8. Conduct rigorous statistical analysis to confirm significant differences in Beta cell proportions across conditions, as suggested by the population bar plots.

Limitations:

1. The annotation of 'Endothelial tip cell' is based on a small cell count (46 cells) and observed marker overlap with general endothelial or mesenchymal populations, requiring further validation for precise and exclusive identity.
2. Population bar plots provide relative proportions, not absolute cell counts. Changes in relative percentages might be influenced by changes in the absolute number of other cell types, and detailed statistical analysis is required to confirm significant differences between conditions.
3. In non-diabetic conditions, the absence of Beta cell-involved interactions among the top 80 might reflect subtle Beta cell-specific communication being less dominant than high-magnitude interactions of the exocrine compartment, rather than a complete absence of interaction.
4. The appearance of viral infection pathways (e.g., 'Coronavirus disease', 'Human cytomegalovirus infection') in GSA for Alpha cells and GSEA across cell types suggests a heightened antiviral response or the activation of general inflammatory pathways that mimic viral responses. This does not necessarily imply direct viral causation for all T1D cases but reflects an ongoing immune activation state.
5. The findings are from a specific cohort and may not be fully generalizable to all T1D patients or populations, especially given the sample-to-sample variability observed.
6. The analyses identify associations and enrichments, but do not definitively establish causality. Experimental validation is needed to confirm functional roles.
7. While immune cells are mentioned, the analysis primarily focuses on pancreatic stromal and epithelial cell types. A deeper dive into immune cell subsets and their specific interactions would provide further insights.

11. Query List

1. Show and save UMAPs for condition, sample, major cell type, minor cell type, and celltype_subset in two columns.
2. Show major cell type scores on UMAP and save the result.
3. Show and save a marker expression dot plot for celltype_subset. SET target_cell = None and var_group_rotation = 45. Leave all other arguments at their default values.
4. Show and save a population bar plot of minor cell types.
5. Show and save cell-cell interaction patterns. Select at most 80 cell-cell interactions per group.
6. Find cell-cell interactions involving major Pancreas cells, stromal cells and immune cells that differ significantly between conditions, show them as a dot plot, and save the result. Set max_n_items_per_group to 60. For each interaction identified under each condition, using the provided dataset context, describe the associated signaling pathways and mechanisms and, where relevant, its potential as a therapeutic target and related therapeutic agents.
7. Show marker expression for Beta cells in a dot plot, and save the result. Use only surfaceome markers, with at most 50 markers per condition. For each marker identified under each condition, using the provided dataset context, describe the associated signaling pathways and biological mechanisms within the Pancreas and, where relevant, its potential as a therapeutic target for Type 1 Diabetes and related therapeutic agents.
8. Show Gene Ontology (GSA) analysis results for Alpha cells as a bar plot, and save the result.
9. Show Gene Set Enrichment Analysis results as a dot plot, and save it. Set the color map to RdBu_r, n_pws_to_show to 120, and target_cells to None.