SCODiA Report by MLBI Lab

Single-Cell Insights into Pancreatic Cellular Dynamics and Intercellular Communication in Type 2 Diabetes Progression

This report leverages single-cell RNA sequencing data to characterize the human pancreas across non-diabetic, prediabetic, and type 2 diabetic conditions. Key findings reveal a progressive decline in Beta cell proportion and function, accompanied by widespread cellular stress and active cell cycle arrest in established diabetes. Pancreatic macrophages undergo significant metabolic reprogramming and polarization towards a pro-inflammatory state, while pancreatic stellate cells exhibit enhanced fibrotic signaling and altered interactions that contribute to tissue remodeling. These cellular shifts and disrupted communication networks provide critical insights into the pathogenesis of type 2 diabetes.

Contents

0. Dataset overview
1. UMAP Visualization of Pancreatic Single-Cell RNA-Seq Data by Condition, Sample, and Cell Type
2. Pancreatic Cell Type Annotation and Marker Gene Expression Analysis on UMAP
3. Celltype_subset Marker Gene Expression Dot Plot in Human Pancreas
4. Pancreatic Minor Cell Type Population Analysis in Diabetes Progression
5. Pancreatic Macrophage Subset Population Shifts in Type 2 Diabetes Progression
6. Macrophage Subset Population Shifts Across Diabetes Progression in the Pancreas
7. Pancreatic Cell-Cell Interaction Landscape Across Diabetes Progression
8. Cell-Cell Interaction Analysis of Immune Checkpoint and Cell Cycle Related Genes in Pancreatic Cell Types Across Diabetes Progression
9. Pancreatic Immune and Stromal Cell-Cell Interaction Patterns Across Diabetes Conditions
10. Macrophage Condition-Specific Surfaceome Markers in Pancreatic Tissue
11. Pancreatic Beta Cell Cycle Gene Expression Shifts in Prediabetes and Type 2 Diabetes
12. Pancreatic Cell-Type Specific Gene Ontology Upregulation in Prediabetes and Type 2 Diabetes
13. Gene Set Enrichment Analysis Reveals Condition-Specific Pathway Dysregulation in Pancreatic Cell Types
14. Discussion
15. Query List

0. Dataset overview

Dataset Summary

• Cells and Genes: The dataset contains 90,006 cells and 28,920 genes.
• Species and Tissue: The data is from human Pancreas tissue.
• Conditions: Three experimental conditions are available: type2_diabetes, non_diabetic, and prediabetes. The reference condition for differential analyses is 'non_diabetic'.

Cell Types: Cells are annotated at three hierarchical levels

• Major: Ductal cell, Alpha cell, Acinar cell, Beta cell, Delta cell, Stromal cell, Gamma (PP) cell, unassigned, Myeloid cell, Endothelial cell, Mast cell.
• Minor: Ductal cell, Alpha cell, Acinar cell, Beta cell, Delta cell, Stellate cell, Gamma (PP) cell, unassigned, Macrophage, Endothelial cell, Dendritic cell, Smooth muscle cell, Fibroblast, Mast cell.
• Subset: Ductal cell, Alpha cell, Acinar cell, Beta cell, Delta cell, Stellate cell, Gamma (PP) cell, unassigned, Macrophage (M1), Endothelial cell, Endothelial tip cell, Macrophage (M2C), Macrophage (M2D), Macrophage (M2B), Macrophage (M2A), Smooth muscle cell, Lymphatic Endothelial cell, Fibroblast, Mast cell.

Available Precomputed Results:

• Cell-Cell Interaction (CCI): Results are stored per condition and per sample.
• Differential Expression Genes (DEG): Results are available for each 'celltype_minor', comparing one condition versus the rest, or one condition versus the reference ('non_diabetic').
• Gene Set Enrichment Analysis (GSEA): Results are available for each 'celltype_minor', comparing one condition versus the rest, or one condition versus the reference.
• Gene Set Analysis (GSA/GO): Results are available for each 'celltype_minor', comparing one condition versus the rest, or one condition versus the reference, specifically for upregulated genes.

1. UMAP Visualization of Pancreatic Single-Cell RNA-Seq Data by Condition, Sample, and Cell Type

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Uniform Manifold Approximation and Projection (UMAP) plots of single-cell RNA sequencing data from human pancreas, displaying the distribution of cells across various metadata categories: disease condition, individual sample origin, major cell types, minor cell types, and detailed cell type subsets. These visualizations are crucial for understanding the overall structure of the dataset, assessing the quality of cell type annotations, and evaluating the presence of technical variations (e.g., batch effects).

Visual Summary

• Condition UMAP: Cells from different disease conditions (non_diabetic, prediabetes, type2_diabetes) are largely intermingled across the UMAP embedding. While there appears to be a general overlap, subtle enrichments or depletions of specific conditions might be observed in certain regions, indicating potential condition-specific changes in cell type composition or transcriptional states within particular cell populations.
• Sample UMAP: Individual samples (e.g., MS17001, MS19003) show broad mixing across the UMAP. This indicates that technical variations or "batch effects" related to individual samples have been effectively mitigated, and the primary drivers of cell clustering are biological differences rather than sample-specific technical artifacts.
• Celltype_major UMAP: The UMAP clearly resolves distinct clusters corresponding to major pancreatic cell types (e.g., Acinar cell, Alpha cell, Beta cell, Ductal cell, Endothelial cell, Stromal cell, Myeloid cell). These clusters are well-separated, suggesting robust and accurate identification of these broad cell populations based on their unique gene expression profiles.
• Celltype_minor UMAP: This visualization provides a finer-grained resolution, showing distinct clusters for minor cell types such as Stellate cells (within Stromal), Macrophages (within Myeloid), and different endocrine cell types (Alpha, Beta, Delta, Gamma PP cells). The separation remains clear, demonstrating the ability to differentiate more specific cell identities.
• Celltype_subset UMAP: The most detailed level of annotation reveals further heterogeneity within minor cell types, particularly evident in the distinct sub-clusters for macrophage polarization states (Mac_M1, M2a, M2b, M2c, M2d) and endothelial cell subtypes (Endo Lymp, Endo tip). This deep annotation highlights the complex cellular landscape of the pancreas.

Biological Interpretation

The UMAP plots provide a high-level overview of the cellular heterogeneity in the human pancreas and how it relates to disease conditions.

• Robust Cell Type Annotation: The progressive and clear segregation of cells from celltype_major to celltype_minor and celltype_subset across the UMAP embedding strongly supports the quality and confidence of the cell type annotations. This hierarchical classification allows for comprehensive analysis, from broad cellular compartments to highly specific subpopulations like macrophage subsets, which are known to play distinct roles in inflammation and tissue remodeling in diabetes [1].
• Effective Batch Correction: The successful intermixing of cells from different sample origins across the UMAP indicates that computational methods have effectively corrected for potential batch effects. This is critical for ensuring that observed biological differences are genuine and not artifacts of experimental processing.
• Disease-Related Shifts: While disease condition does not appear to be the dominant factor driving overall cell clustering, the overlay of conditions on the UMAP allows for a visual assessment of potential shifts in cell proportions or states. The apparent overlap suggests that disease pathology might induce transcriptional changes *within* specific cell types rather than causing wholesale re-clustering of cell populations. This warrants further investigation through cell-type-specific differential gene expression (DEG) and pathway enrichment (GSEA/GSA) analyses. For example, changes in the proportion of specific endocrine cells (e.g., Beta cells) or immune cell subsets (e.g., macrophage polarization) are well-documented in type 2 diabetes pathophysiology [2].
• Pancreatic Cellular Complexity: The detailed celltype_minor and celltype_subset annotations highlight the intricate cellular composition of the human pancreas beyond the classical endocrine (islet) and exocrine (acinar/ductal) compartments, including various stromal and immune populations. These accessory cells are increasingly recognized for their roles in pancreatic homeostasis and disease progression [3].

Annotation Notes

• Annotation Consistency: The distinct clustering of cell types at all hierarchical levels (major, minor, subset) suggests a high degree of consistency and accuracy in the cell type identification, indicating that the chosen markers and computational methods are effective.
• Unassigned Cells: The presence of a relatively small and dispersed "unassigned" population is a positive indicator, suggesting that most cells have been confidently classified. Further examination of these "unassigned" cells might reveal novel or rare cell types, or cells of lower quality.

---

References:

1. Macrophage Polarization in Diabetes: A search on PubMed for "macrophage polarization diabetes pancreas" can provide relevant literature. PubMed search: macrophage polarization diabetes pancreas
2. Cellular Changes in Type 2 Diabetes: A general review on cellular changes in type 2 diabetes. PubMed search: pancreatic cell type changes type 2 diabetes
3. Pancreatic Accessory Cells: A general search on PubMed for "pancreas stromal immune cells diabetes" can provide context. PubMed search: pancreas stromal immune cells diabetes

2. Pancreatic Cell Type Annotation and Marker Gene Expression Analysis on UMAP

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the UMAP embedding of single-cell RNA-seq data from the human pancreas, showcasing the expression patterns of a panel of known marker genes and the corresponding minor cell type annotations. The primary goal is to assess the quality of cell type annotation by correlating gene expression patterns with pre-assigned cell identities.

Visual Summary

The UMAP plots clearly delineate distinct cell populations based on their transcriptional profiles. The celltype_minor annotation plot (bottom right) provides a spatial reference for various pancreatic cell types. Overlaying the expression of selected marker genes on the UMAP allows for a direct visual assessment of their cell type specificity:

Immune Cell Markers:

• CD3D, CD4, CD8A (T-cell markers): Show highly localized expression in distinct, relatively small clusters that primarily correspond to cells labeled as "unassigned" in the celltype_minor annotation. Specifically, a cluster towards the lower-middle left and another smaller cluster slightly above show strong expression.
• CD79A, MS4A1 (B-cell markers): Exhibit concentrated expression in another "unassigned" cluster, spatially distinct from the T-cell clusters, located more centrally in the lower-middle region of the UMAP.
• MZB1 (Plasma cell marker): Shows very specific and strong expression in a tiny subset of cells within the "unassigned" region, overlapping with the B-cell rich area, indicating a highly differentiated immune cell subset.
• CD14, LYZ (Myeloid cell/Macrophage markers): Display robust expression in the cluster identified as "Macrophage" in the celltype_minor annotation, located near the center-left.

Stromal Cell Markers:

• FBLN1 (Fibroblast marker): Shows strong expression in the cluster annotated as "Stellate cell", which represents a major stromal component of the pancreas.
• NOTCH3 (Pericyte/Vascular smooth muscle cell marker, also expressed in some stromal cells): Primarily expressed in the "Stellate cell" cluster, with some broader expression that might encompass closely related stromal or endothelial populations.

Epithelial Cell Markers:

• EPCAM, MUC1 (General epithelial markers): Demonstrate widespread and strong expression across the majority of the pancreatic epithelial cell clusters, including "Acinar cell", "Alpha cell", "Beta cell", "Delta cell", "Ductal cell", and "Gamma (PP) cell". This pattern is consistent with their role as pan-epithelial markers.

Endothelial Cell Marker:

• CD34 (Endothelial/Hematopoietic stem cell marker): Shows distinct and high expression localized to the cluster annotated as "Endothelial cell", located in the lower-left region of the UMAP.

Biological Interpretation

The observed gene expression patterns largely align with established cell type-specific markers in the human pancreas, confirming the biological identity and robust clustering of the various cell populations.

• Pancreatic Epithelium: Genes like EPCAM and MUC1 serve as broad markers for the major parenchymal cells of the pancreas (acinar, ductal, islet cells), highlighting their epithelial origin.
• Immune Cell Heterogeneity: The expression of T-cell (CD3D, CD4, CD8A), B-cell (CD79A, MS4A1), and plasma cell (MZB1) markers in "unassigned" clusters indicates the presence of diverse lymphoid populations within the pancreatic microenvironment. The distinct localization of these markers on the UMAP suggests successful separation of these immune subsets, even if not explicitly named in the celltype_minor annotation. Macrophages are clearly identified by CD14 and LYZ expression, consistent with their innate immune functions GeneCards: LYZ.
• Stromal Components: The specific expression of FBLN1 and NOTCH3 in the "Stellate cell" cluster confirms their identity as a key stromal cell type, likely pancreatic stellate cells, which play crucial roles in extracellular matrix remodeling and disease progression in conditions like pancreatitis and pancreatic cancer PubMed Search: pancreatic stellate cells FBLN1 NOTCH3.
• Vascular Endothelium: CD34 expression effectively identifies the endothelial cell population, crucial for maintaining vascular integrity and facilitating nutrient/hormone exchange within the pancreas UniProt: CD34.

Annotation Notes

The UMAP visualization of marker gene expression strongly supports the quality and accuracy of the celltype_minor annotations for most of the major pancreatic cell types. The "unassigned" category effectively captures immune cell subsets that, while clearly identifiable by specific markers (e.g., T cells, B cells, plasma cells), may not have been granularly labeled in the celltype_minor metadata. This indicates that while the current celltype_minor annotation is good for broad categories, further sub-clustering and specific naming of immune cell types within the "unassigned" population could enhance the resolution of the dataset.

3. Celltype_subset Marker Gene Expression Dot Plot in Human Pancreas

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the expression of key marker genes across different celltype_subset populations identified in human pancreas single-cell RNA sequencing data. The plot serves as an essential quality control and validation step, allowing us to assess the specificity and robustness of the cell type annotations. Marker genes were identified with a focus on surfaceome genes, which is important for understanding cell-cell interactions and potential therapeutic targeting.

Visual Summary

The dot plot effectively illustrates the expression profile of 140 selected marker genes across 16 distinct celltype_subset populations. Each row represents a cell type, and each column represents a marker gene.

• Dot size corresponds to the fraction of cells within that group expressing the gene (percentage of non-zero counts).
• Dot color intensity (red scale) indicates the mean expression level of the gene within that cell type group.
• Diagonal blocks of expression: The plot clearly shows distinct clusters of highly expressed and broadly detected markers (dark, large red dots) for most cell types along the diagonal. This indicates good specificity of the identified markers for their respective cell populations.
• Cell Counts: The right-hand side bar indicates the number of cells belonging to each celltype_subset, ranging from 59 (Macrophage M2A) to 27558 (Beta cell), providing context for the robustness of marker identification for each group.
• Surfaceome Focus: The markers plotted are biased towards surface proteins, as indicated by the surfaceome_only: True parameter in the find_cfg.

Biological Interpretation

The marker gene expression patterns largely align with known biological functions and identities of pancreatic cell types, reinforcing the accuracy of the celltype_subset annotations:

• Acinar Cells: Show strong expression of pancreatic digestive enzymes and related proteins such as *REG1A*, *PRSS2*, *SPINK1*, *CTRB2*, and *CELA3A* GeneCards: PRSS2.
• Alpha Cells: Characterized by high expression of *GCG* (Glucagon), *ALDH1A1*, and *SLC30A8* GeneCards: GCG. SLC30A8 is notably involved in zinc transport and insulin crystallization, important for islet biology.
• Beta Cells: Distinctly marked by the expression of *INS* (Insulin), *IAPP* (Amylin), and *PCSK1* (Proprotein Convertase Subtilisin/Kexin Type 1) GeneCards: INS. These are quintessential markers for insulin production and secretion.
• Delta Cells: Identified by the expression of *SST* (Somatostatin) GeneCards: SST, reflecting their role in regulating other islet hormones.
• Ductal Cells: Display expression of epithelial markers like *KRT7*, *ANXA4*, and functional markers such as *CFTR* (Cystic Fibrosis Transmembrane Conductance Regulator) GeneCards: CFTR, consistent with their role in pancreatic fluid and bicarbonate secretion.
• Gamma (PP) Cells: Specifically marked by *PPY* (Pancreatic Polypeptide) GeneCards: PPY, their namesake hormone.
• Endothelial Tip Cells: Show expression of *ESM1* and *ANGPT2*, markers associated with endothelial cell activation and angiogenesis, consistent with a tip cell phenotype.
• Stellate Cells: Exhibit typical stromal markers, including collagen genes (*COL6A1*, *COL3A1*, *COL4A1*), *PDGFRA*, and extracellular matrix-related genes like *FN1* (Fibronectin 1) GeneCards: FN1, reflecting their role in pancreatic fibrosis.
• Mast Cells: Clearly identified by specific immune markers like *TPSAB1* and *CPA3*, consistent with their immune surveillance and inflammatory roles.
• Macrophage Subtypes (M1, M2A, M2B, M2C, M2D): While all macrophage subtypes show some common macrophage markers, M1 macrophages display a more distinct set of interferon-responsive genes (*IFNGR1*, *IFNGR2*, *STAT1*). The M2 subtypes (M2A, M2B, M2C, M2D) appear to share more markers among themselves, and their distinction based solely on surfaceome markers, as identified and plotted here, is less pronounced compared to the other major cell types. This suggests that further functional or intracellular markers might be needed for a finer resolution of M2 subtypes, or that their surface protein profiles are highly similar in this context.

Annotation Notes

The dot plot provides strong evidence for the fidelity of the majority of celltype_subset annotations within this single-cell dataset. The distinct and expected expression of well-established marker genes for Acinar, Alpha, Beta, Delta, Ductal, Gamma (PP), Stellate, Endothelial tip, and Mast cells significantly validates their assigned identities.

The observation regarding the macrophage subtypes, where distinct surfaceome markers are less apparent across M2 sub-classifications, indicates a potential area for further investigation. While general macrophage identity is clear, the ability to sharply delineate M2 subtypes based on the selected surfaceome markers alone might be limited, suggesting that these distinctions might rely on more subtle transcriptional differences, intracellular markers, or functional assays not captured in this surfaceome-focused analysis. This finding informs the confidence level of these specific sub-annotations and highlights where additional validation might be beneficial.

4. Pancreatic Minor Cell Type Population Analysis in Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis presents the relative proportions of minor cell types within individual pancreatic samples, grouped by diabetes conditions: non_diabetic, prediabetes, and type2_diabetes. The stacked bar plots normalize cell type counts to 100% for each sample, allowing for a clear visualization of changes in cellular composition across disease states. The primary objective is to identify shifts in pancreatic cellular architecture associated with the progression of Type 2 Diabetes (T2D).

Visual Summary

The stacked bar plots effectively illustrate the cellular landscape of the pancreas across the non_diabetic, prediabetes, and type2_diabetes conditions. Each bar represents a single sample, with different colors indicating various minor cell types.

• Dominant Cell Types: Acinar cells (dark red) and Alpha cells (bright red) consistently represent a large fraction of cells across all samples and conditions. Beta cells (orange-red) also constitute a substantial proportion, particularly in non_diabetic samples.
• Beta Cell Decline in Diabetes Progression: A notable trend observed is the apparent decrease in the relative proportion of Beta cells (orange-red) as the disease progresses. They appear most abundant in non_diabetic samples, show a reduction in prediabetes, and are further diminished in type2_diabetes samples. This reduction is visually consistent across most samples within the prediabetes and type2_diabetes groups.
• Other Endocrine Cell Trends: Alpha cells maintain a relatively stable high proportion across all conditions. Delta cells (orange) and Gamma (PP) cells (light green) remain minor components with no clear proportional changes.
• Exocrine and Stromal Cells: Acinar cells show some variability but generally remain a major component. Ductal cells (light orange) maintain a moderate proportion. Stromal cell types like Stellate cells (teal) and Fibroblasts (very light yellow) remain minor fractions, with no dramatic shifts immediately apparent, though subtle increases might be present in type2_diabetes samples.
• Immune Cells: Macrophages (green) and other immune cells like Dendritic cells (light orange) and Mast cells (mint green) are present in small proportions across all conditions, reflecting the baseline immune surveillance and inflammatory potential within the pancreas. No large-scale shifts are immediately discernible visually for these populations.
• Unassigned Cells: The "unassigned" fraction (blue) represents a small portion of cells, indicating that most cells were confidently classified into known minor cell types.

Biological Interpretation

The most striking biological insight from this population analysis is the progressive reduction in the relative proportion of Beta cells with increasing severity of diabetes.

• Beta Cell Loss in T2D: The observed decline in Beta cell proportion from non_diabetic to prediabetes and type2_diabetes is a hallmark of T2D pathophysiology. T2D is characterized by both insulin resistance and a progressive failure of pancreatic Beta cells to produce sufficient insulin to compensate for this resistance. This Beta cell dysfunction and mass reduction are critical drivers of hyperglycemia [1]. The visual evidence here strongly supports Beta cell vulnerability and loss as a key feature of diabetes progression in this cohort.
• Islet Cell Balance: While Beta cells decrease, Alpha cells maintain a relatively stable proportion. This can contribute to the imbalanced glucagon-to-insulin ratio often seen in T2D, where relative hyperglucagonemia (due to relatively preserved or even increased alpha cell activity/mass compared to beta cells) exacerbates hyperglycemia.
• Pancreatic Microenvironment: The stable proportions of Acinar and Ductal cells suggest that the exocrine compartment might be relatively less affected at the population level compared to the endocrine islets, though functional changes within these cells are still possible. Small increases in stromal cells (e.g., Fibroblasts, Stellate cells) or immune cells (e.g., Macrophages) in the diabetic conditions could indicate increased fibrosis and chronic inflammation, respectively, which are known contributors to pancreatic damage and Beta cell dysfunction in T2D [2, 3]. However, these changes are not overtly dramatic at the population level in this plot and would warrant further investigation with specific marker analysis.

Clinical or Translational Implications

The clear evidence of Beta cell loss across diabetes progression has significant clinical and translational implications:

• Disease Progression Marker: The relative proportion of Beta cells could serve as a valuable biomarker for monitoring the progression of prediabetes to T2D and assessing disease severity.
• Therapeutic Targets: Strategies aimed at preserving Beta cell mass and function are central to T2D management and drug development. Understanding the molecular mechanisms driving Beta cell loss in prediabetes could open avenues for early intervention to prevent or delay overt T2D.
• Heterogeneity in Islet Dysfunction: While Beta cell decline is a general trend, individual sample variability highlights the heterogeneous nature of diabetes progression, which may influence patient responses to therapies.
• Inflammation and Fibrosis: If further analyses confirm an increase in immune or stromal cells, these populations could represent targets for therapies aimed at mitigating the inflammatory and fibrotic microenvironment that contributes to Beta cell damage in T2D.

References

1. Beta cell dysfunction in T2D: A search on PubMed for "beta cell dysfunction type 2 diabetes" provides extensive literature. For example: https://pubmed.ncbi.nlm.nih.gov/?term=beta+cell+dysfunction+type+2+diabetes
2. Pancreatic inflammation in T2D: A search on PubMed for "pancreatic inflammation type 2 diabetes" provides relevant reviews. For example: https://pubmed.ncbi.nlm.nih.gov/?term=pancreatic+inflammation+type+2+diabetes
3. Pancreatic fibrosis in T2D: A search on PubMed for "pancreatic fibrosis type 2 diabetes" provides relevant literature. For example: https://pubmed.ncbi.nlm.nih.gov/?term=pancreatic+fibrosis+type+2+diabetes

5. Pancreatic Macrophage Subset Population Shifts in Type 2 Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis visualizes the relative proportions of different macrophage subsets (M1, M2A, M2B, M2C, M2D) within the total macrophage population across individual pancreatic samples. These samples are grouped by disease condition: non-diabetic, prediabetes, and type2_diabetes. The plot provides insights into how the macrophage landscape in the pancreas changes during the progression of Type 2 Diabetes (T2D).

Visual Summary

The stacked bar plots display the fractional abundance of five distinct macrophage subsets for each sample. Each bar represents a single patient sample, and the segments within the bar correspond to the proportion of each macrophage subset.

• Macrophage (M1), represented by dark red, consistently forms the largest proportion of macrophages across all conditions. Notably, its relative abundance shows a clear increasing trend from non-diabetic samples (ranging approximately 25-60%) to prediabetes (25-75%) and further into type2_diabetes (frequently reaching 70-80% in many samples). This suggests a progressive enrichment of the M1 phenotype with disease advancement.
• Macrophage (M2B), indicated by light yellow, appears to be the second most abundant subset in non-diabetic individuals (up to 30%). However, its proportion diminishes significantly in prediabetes and even more so in type2_diabetes samples, where it often contributes less than 20% to the total macrophage population.
• Macrophage (M2A), shown in orange, is present in relatively small proportions across all conditions and does not show a pronounced trend.
• Macrophage (M2C), in light green, and Macrophage (M2D), in teal/mint green, show moderate representation in non-diabetic samples but appear to decrease in proportion as the disease progresses through prediabetes to type2_diabetes, becoming minor components in many T2D samples.

In summary, the most striking observation is the increased prevalence of pro-inflammatory M1 macrophages and a concomitant reduction in generally anti-inflammatory or regulatory M2 macrophage subsets (M2B, M2C, M2D) in the pancreas as individuals transition from a non-diabetic state to prediabetes and eventually to type2_diabetes.

Biological Interpretation

The observed shift in macrophage populations within the pancreas offers significant biological insight into the immune pathology of Type 2 Diabetes.

• Dominance of M1 Macrophages in T2D: Macrophage (M1) cells are classically activated macrophages known for their pro-inflammatory functions, including the production of cytokines like TNF-α, IL-1β, and IL-6. Their increasing dominance in the pancreas of prediabetic and type2_diabetic individuals suggests a heightened and persistent inflammatory state. Chronic low-grade inflammation in the pancreatic islets, often referred to as "insulitis," is a critical driver of beta-cell dysfunction and death, contributing to insulin resistance and progressive loss of insulin secretion characteristic of T2D PubMed search: Pancreatic inflammation type 2 diabetes macrophages.
• Reduction of M2 Subsets: The decrease in M2B, M2C, and M2D macrophage subsets is equally important. M2 macrophages are generally associated with anti-inflammatory responses, tissue repair, immune regulation, and metabolic homeostasis.
• M2B macrophages are known for their immunoregulatory roles, often producing both pro- and anti-inflammatory mediators.
• M2C macrophages typically have immunosuppressive and tissue remodeling functions, secreting IL-10 and TGF-β.
• M2D macrophages (also termed "regulatory" or "tumor-associated" in some contexts) are involved in immune suppression and angiogenesis.

The reduction of these M2 populations implies a diminished capacity for the pancreas to resolve inflammation, repair tissue damage, or maintain immune homeostasis, thereby perpetuating the inflammatory cycle detrimental to beta-cell survival and function.

This overall re-polarization of pancreatic macrophages from a more balanced or potentially anti-inflammatory/regulatory state towards a pro-inflammatory M1 phenotype strongly implicates immune dysregulation in the pathogenesis and progression of T2D.

Clinical or Translational Implications

The findings highlight critical aspects of T2D pathogenesis with potential clinical and translational relevance:

• Biomarker Potential: The ratio of M1 to M2 macrophage subsets in pancreatic tissue could serve as a valuable biomarker for assessing disease stage, progression risk (e.g., in prediabetes), or responsiveness to therapeutic interventions aimed at modulating inflammation. Non-invasive methods to assess this polarization, perhaps through circulating markers reflecting pancreatic immune status, could be explored.
• Therapeutic Targets: Modulating macrophage polarization represents a promising therapeutic strategy for T2D. Interventions aimed at:
• Inhibiting M1 macrophage activation or pro-inflammatory signaling pathways (e.g., targeting specific cytokines or their receptors).
• Promoting the differentiation or survival of beneficial M2 macrophage subsets.

Such strategies could help alleviate pancreatic inflammation, preserve beta-cell mass and function, and improve glucose homeostasis in individuals with prediabetes and T2D PubMed search: Macrophage polarization diabetes therapy.

• Disease Monitoring: Understanding these shifts provides a cellular-level mechanism for the chronic inflammatory component of T2D, which could inform patient stratification and personalized treatment approaches.

6. Macrophage Subset Population Shifts Across Diabetes Progression in the Pancreas

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the proportional changes of specific macrophage subsets within the human pancreas across different conditions: prediabetes, type 2 diabetes, and non-diabetic controls. Using single-cell RNA-seq data, box plots with superimposed stripplots were generated to visualize the celltype proportion of Mac (M2B) and Mac (M1) macrophages, with statistical significance determined by comparing conditions against each other (p-value cutoff < 0.1).

Visual Summary

The box plots illustrate the distribution of celltype proportions for Mac (M2B) and Mac (M1) macrophage subsets across the three conditions. Each black dot represents the proportion from an individual sample.

Mac (M2B) Proportions:

• The median proportion of Mac (M2B) cells appears higher in both the prediabetes and type2_diabetes groups compared to the non_diabetic group.

Statistically significant differences (p < 0.1) were observed

Prediabetes vs. non_diabetic: p = 0.06

Type2_diabetes vs. non_diabetic: p = 0.06

• No statistically significant difference was found between prediabetes and type2_diabetes (p = 0.88), suggesting that the elevation of M2B macrophages might occur early and persist into established type 2 diabetes.

Mac (M1) Proportions:

• The median proportion of Mac (M1) cells appears lower in the type2_diabetes group compared to the non_diabetic group.

A statistically significant difference (p < 0.1) was observed

Type2_diabetes vs. non_diabetic: p = 0.07

• No significant differences were found between prediabetes and non_diabetic (p = 0.40), or between prediabetes and type2_diabetes (p = 0.52).

Biological Interpretation

Macrophages are critical immune cells in the pancreas, playing complex roles in both the initiation and progression of type 2 diabetes (T2D). Macrophages can polarize into different functional states, broadly categorized as M1 (pro-inflammatory) and M2 (anti-inflammatory, reparative, or pro-fibrotic), with further nuances within M2 subtypes.

• Increased Mac (M2B) in Prediabetes and T2D: The significant increase in Mac (M2B) proportions in both prediabetes and type 2 diabetes, compared to non-diabetic individuals, suggests a notable shift in the macrophage landscape early in disease progression. M2B macrophages are a less-studied M2-like subset that can be involved in antigen presentation, immune regulation, and inflammation resolution, but also potentially contribute to chronic inflammation and fibrosis in metabolic diseases depending on the context [PubMed search for M2B macrophage diabetes: PubMed Search]. Their sustained elevation from prediabetes to T2D may indicate an adaptive response that ultimately becomes maladaptive, failing to resolve inflammation or actively contributing to tissue remodeling and insulin resistance within the pancreatic environment.
• Decreased Mac (M1) in T2D: The observed trend of decreased M1 macrophage proportion in established type 2 diabetes compared to non-diabetic controls is intriguing. M1 macrophages are typically associated with acute, pro-inflammatory responses and are often implicated in the initial damage to pancreatic beta cells (insulitis) [GeneCards entry for M1 macrophage markers: GeneCards]. This decrease might suggest a shift from an acute M1-driven pro-inflammatory phase to a more chronic, potentially M2-dominated, inflammatory environment in the later stages of T2D progression, or it could reflect the exhaustion or repolarization of M1 cells over time.

Collectively, these findings indicate a dynamic re-orchestration of pancreatic macrophage populations during diabetes development. The simultaneous increase in M2B and decrease in M1 macrophages in T2D suggests a complex interplay where different macrophage subsets contribute to the chronic low-grade inflammation and tissue dysfunction characteristic of the disease.

Clinical or Translational Implications

Understanding these specific shifts in macrophage subsets could have significant clinical implications for T2D:

• Biomarker Potential: Changes in the proportions of Mac (M2B) and Mac (M1) could potentially serve as biomarkers for assessing disease stage or progression, particularly in distinguishing prediabetes and T2D from non-diabetic states.
• Therapeutic Targets: Modulating macrophage polarization and specific subset abundance represents a promising therapeutic avenue. For example, if M2B macrophages are found to contribute to pancreatic fibrosis or insulin resistance, strategies aimed at inhibiting their detrimental functions or repolarizing them to a more beneficial state could be explored [PubMed search for macrophage polarization in diabetes therapy: PubMed Search].
• Disease Mechanisms: Further investigation into the specific roles and molecular mechanisms of Mac (M2B) macrophages in the diabetic pancreas is warranted to elucidate their contribution to disease pathology and identify potential intervention points.

7. Pancreatic Cell-Cell Interaction Landscape Across Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interactions (CCIs) within the human pancreas across different metabolic conditions: non_diabetic, prediabetes, and type2_diabetes. Utilizing single-cell RNA sequencing data and CellPhoneDB, the plot_cci_dots tool identified and visualized the most significant ligand-receptor interactions between various pancreatic cell types for each condition. The visualization highlights the strength (dot size, representing mean expression of the interacting pair) and statistical significance (dot color, representing -log10(p-value)) of up to 80 prominent cell-cell communication pairs. This allows for a comparative understanding of how intercellular communication networks adapt and change during the progression to type 2 diabetes.

Visual Summary

The three dot plots visualize cell-cell interactions for non_diabetic, prediabetes, and type2_diabetes conditions. The Y-axis displays specific cell-pair interactions (e.g., Beta|Beta, Alpha|Stellate), while the X-axis lists ligand-receptor gene pairs (e.g., TGFB1-TGFBR2, SPP1-integrin_avb3_complex).

1. Non_diabetic Condition: The plot for non_diabetic individuals serves as a baseline, showing a diverse array of interactions. Prominent features include widespread collagen and fibronectin-integrin interactions, particularly involving pancreatic stellate cells (Stellate|Stellate, Acinar|Stellate, Ductal|Stellate), indicative of baseline extracellular matrix (ECM) maintenance. Key islet cell interactions (e.g., Beta|Beta, Alpha|Beta) involving signaling pathways like TGFB1-TGFBR2/TGFBR3 and SPP1-integrin_avb3_complex are also observed.
2. Prediabetes Condition: The prediabetes plot appears largely similar to the non_diabetic condition in terms of the overall patterns of the top 80 interactions. While many core ECM and islet interactions persist, subtle shifts in the significance or strength of specific interactions might be present, suggesting an early stage of pancreatic adaptation or stress.
3. Type2_diabetes Condition: In the type2_diabetes plot, several critical changes are evident compared to the non_diabetic and prediabetes states:

• TGFB1 Signaling Dominance: Interactions involving TGFB1-TGFBR2 and TGFB1-TGFBR3 show a marked increase in both statistical significance (brighter yellow/green colors) and mean expression (larger dot sizes), particularly in interactions involving Beta cells (Beta|Beta, Beta|Alpha, Beta|Delta, Beta|Ductal) and Delta|Stellate cells. This indicates a heightened activation of the TGF-beta pathway.
• Persistent ECM Interactions: Collagen-integrin and fibronectin-integrin interactions, especially with Stellate cells, remain highly prevalent, likely reflecting ongoing ECM remodeling processes.
• WNT Signaling Dynamics: Some WNT family interactions (e.g., WNT7B-FZD2) involving Beta and Stellate cells appear more pronounced, suggesting altered WNT pathway activity in T2D.
• SPP1-Integrin: The SPP1-integrin_avb3_complex interaction remains active, consistent with a sustained inflammatory environment.

Biological Interpretation

The observed changes in cell-cell interactions provide crucial insights into the biological mechanisms underlying type 2 diabetes progression in the pancreas.

• Extracellular Matrix Remodeling and Fibrosis: The consistent and strong presence of collagen-integrin and fibronectin-integrin interactions across all conditions, especially those involving Stellate cells, highlights the fundamental role of the ECM in pancreatic architecture. The dramatic increase in TGFB1-TGFBR2/TGFBR3 signaling in type2_diabetes is a critical finding. TGF-beta is a master regulator of fibrosis and inflammation. Its upregulation in interactions involving Beta cells, and between Delta and Stellate cells, strongly suggests increased pancreatic fibrosis, impaired beta-cell function, and chronic inflammation associated with advanced diabetes. Activated pancreatic stellate cells, which produce ECM components, are known to be driven by TGF-beta signaling.
• [PubMed search for "TGFB1 pancreatic fibrosis diabetes": PubMed Search]
• Islet Cell Dysfunction and Communication: The dynamic changes in interactions involving Alpha, Beta, and Delta cells are highly relevant to islet homeostasis and dysfunction in diabetes.
• The heightened TGFB1 signaling in Beta-Beta, Beta-Alpha, Beta-Delta, and Beta-Ductal interactions in type2_diabetes indicates significant stress on beta cells and perturbed intercellular communication within the islet, potentially contributing to insulin secretion defects and beta-cell dedifferentiation or loss.
• SPP1-integrin_avb3_complex interactions, consistently observed in Alpha-Beta pairs, indicate ongoing inflammation and immune cell recruitment within the islet, as SPP1 (Osteopontin) is a known pro-inflammatory cytokine.
• [GeneCards for SPP1: GeneCards]

Altered Growth Factor and Signaling Pathways:

• While IGFBP3-IGF1R interactions are noted in Beta-Beta cells in earlier stages, their relative prominence may shift in T2D, potentially reflecting altered insulin-like growth factor axis regulation.
• [GeneCards for IGFBP3: GeneCards]
• Changes in specific WNT ligand-receptor pairs, such as WNT7B-FZD2 in Beta-Stellate and Alpha-Stellate interactions in type2_diabetes, suggest that WNT signaling, critical for pancreatic development and regeneration, is also perturbed. This could influence cell fate, proliferation, or the fibrotic response of stellate cells.

Clinical or Translational Implications

The findings from this cell-cell interaction analysis have direct implications for understanding type 2 diabetes pathology and identifying potential therapeutic targets:

• Targeting TGF-beta Signaling: The strong activation of TGFB1-TGFBR2/TGFBR3 signaling in type2_diabetes positions this pathway as a prime candidate for therapeutic intervention. Inhibiting TGF-beta signaling could potentially reduce pancreatic fibrosis, mitigate beta-cell stress, and improve islet function in diabetic patients. Clinical trials for TGF-beta inhibitors in fibrotic diseases are ongoing, and their potential in T2D warrants further exploration.
• Modulating ECM Remodeling: The sustained and altered ECM interactions, driven in part by TGF-beta, suggest that strategies aimed at inhibiting pancreatic stellate cell activation or modulating specific ECM components could improve pancreatic health and reduce fibrosis in T2D.
• Addressing Inflammation via SPP1: The persistent SPP1-integrin_avb3_complex interactions highlight ongoing inflammation in the diabetic pancreas. Targeting SPP1 or its receptors could represent a strategy to reduce chronic inflammation and its detrimental effects on islet cells.
• Investigating WNT Pathway Modulation: The dynamic shifts in specific WNT ligand-receptor interactions warrant further investigation. Precise modulation of specific WNT axes might offer novel approaches to preserve beta-cell mass, enhance regeneration, or control fibrosis in the diabetic pancreas.

These findings provide a basis for prioritizing specific ligand-receptor pairs and associated pathways for experimental validation in in vitro and in vivo models of diabetes, with the ultimate goal of developing more targeted and effective therapies.

8. Cell-Cell Interaction Analysis of Immune Checkpoint and Cell Cycle Related Genes in Pancreatic Cell Types Across Diabetes Progression

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates cell-cell interactions (CCI) within the human pancreas across different conditions: non_diabetic, prediabetes, and type2_diabetes. The user specifically queried for genes related to immune checkpoint and cell cycle pathways. The plot_cci_dots tool was used to visualize significant ligand-receptor interactions, with dot size representing interaction significance (-log10(p-value)) and dot color indicating the mean expression level of the interacting genes (log2(mean)). While the input target_genes covered a broad spectrum of immune checkpoint and cell cycle molecules, the visualized ligand-receptor pairs are those known to mediate cell-surface communication and influence these processes, specifically focusing on EGFR, IFNGR, TGF-beta, and Integrin pathways.

Visual Summary

The dot plots illustrate active cell-cell communication networks across various pancreatic cell types, including Stellate, Ductal, Endothelial, Alpha, Beta, and Acinar cells.

• Dominant Pathways: The most prominent interactions involve the TGF-beta signaling pathway (e.g., TGFB1, TGFB2, TGFB3 interacting with their respective receptors) and integrin-mediated interactions (e.g., TGFB1_integrin_aVb6_complex). These interactions are particularly strong and widespread for cell pairs involving Stellate cells (e.g., Stellate|Stellate, Stellate|Ductal, Stellate|Alpha, Stellate|Acinar). EGFR signaling (AREG_EGFR, HBEGF_EGFR, TGFA_EGFR) also shows activity, primarily involving Endothelial and Ductal cells, but less widespread than TGF-beta. CD93_IFNGR1 interactions are less prominent.
• Stellate Cell Hyperactivity: Stellate cells consistently display a high number and strength of interactions, especially with other Stellate cells and Ductal cells, centered around TGF-beta ligands and receptors, and integrin alphaVbeta6. These interactions show high mean expression (yellow/green dots) and high significance (large dots) across all conditions.

Changes Across Diabetes Progression:

• Loss of Acinar-Ductal Communication: A notable observation is the disappearance of the Acinar|Ductal interaction involving TGFB1_TGFbeta_receptor2 in both prediabetes and type2_diabetes conditions, which was present in non_diabetic individuals. This suggests a potential disruption in inter-cellular communication pathways critical for pancreatic homeostasis early in disease development.
• Reduced Beta Cell Interactions: Interactions involving Beta cells, particularly Beta|Beta and Ductal|Beta via TGFB1 and TGFB2 signaling, appear to be reduced in significance and/or strength (smaller, less intense dots) as the disease progresses from non_diabetic to prediabetes and further in type2_diabetes.
• Persistent Stellate Cell Interactions: The robust TGF-beta and integrin signaling involving Stellate cells remains largely consistent across all conditions, including type2_diabetes, indicating their sustained activity regardless of disease stage.

Biological Interpretation

The observed cell-cell interactions provide critical insights into the pancreatic microenvironment during diabetes progression.

• Role of Pancreatic Stellate Cells (PSCs) in Fibrosis: The consistent and strong activation of TGF-beta and integrin signaling, particularly involving Stellate cells (Stellate|Stellate, Stellate|Ductal), is highly indicative of their central role in pancreatic fibrosis. Pancreatic stellate cells are key mediators of extracellular matrix deposition, and activation of TGF-beta is a primary driver of their fibrogenic phenotype. Integrin alphaVbeta6 is known to activate latent TGF-beta, thus forming a feed-forward loop that amplifies fibrotic signals [1]. This sustained fibrogenic activity may contribute to the progressive structural and functional changes in the diabetic pancreas.
• Immune Modulation and Cell Cycle Impact: While the visualized pathways are not exclusively "immune checkpoint" or "cell cycle" proteins in a strict sense, they are potent modulators. TGF-beta is a well-established immunosuppressive cytokine that can regulate immune cell differentiation and function [2]. It also plays a role in regulating cell proliferation and differentiation, thereby indirectly impacting cell cycle progression. EGFR signaling can drive cell proliferation and survival, while also influencing inflammatory responses. The observed interactions thus point to pathways that broadly influence both the immune landscape and cellular dynamics within the pancreas.
• Disruption of Beta Cell Niche and Pancreatic Homeostasis: The reduction in TGF-beta-mediated interactions involving Beta cells (Beta|Beta, Ductal|Beta) in diabetic conditions is noteworthy. TGF-beta signaling can influence beta cell survival, proliferation, and differentiation [3]. A reduction could indicate altered signaling crucial for beta cell maintenance or a changing cellular state. More critically, the loss of Acinar|Ductal TGFB1_TGFbeta_receptor2 interaction in prediabetes and T2D suggests a specific breakdown in communication between exocrine (Acinar) and ductal compartments, which might impair overall pancreatic homeostasis and recovery in disease.
• Inflammatory Context: Although IFNGR1 interactions are less prominent, inflammation is a known component of type 2 diabetes. The broader roles of TGF-beta and EGFR signaling in immune cell recruitment and activation also contribute to the inflammatory environment.

Clinical or Translational Implications

The findings from this CCI analysis have significant implications for understanding diabetes pathophysiology and identifying potential therapeutic targets.

• Therapeutic Targeting of Fibrosis: The sustained and robust TGF-beta and integrin signaling involving pancreatic stellate cells represents a critical therapeutic axis to mitigate pancreatic fibrosis, a common complication and contributor to dysfunction in diabetes. Strategies to inhibit TGF-beta signaling or specifically block integrin alphaVbeta6, which activates TGF-beta, could be explored to slow disease progression and preserve pancreatic function [4].
• Biomarker Discovery: The observed disappearance of the Acinar|Ductal TGFB1_TGFbeta_receptor2 interaction in prediabetes could serve as a potential early diagnostic or prognostic biomarker for diabetes progression, warranting further investigation in larger cohorts. Changes in other Beta cell-related interactions might also indicate disease status.
• Restoring Beta Cell Microenvironment: Understanding the specific changes in ligand-receptor interactions involving Beta cells can guide strategies aimed at restoring a supportive microenvironment to protect beta cell mass and function. Modulating the balance of TGF-beta signaling around Beta cells could be crucial for regenerative approaches or improving islet transplantation outcomes.
• Precision Medicine: Identifying specific cell-cell communication hubs that are altered in diabetes provides a foundation for developing targeted therapies that precisely interfere with detrimental interactions or bolster protective ones, thereby advancing precision medicine approaches for diabetes management.

---

References:

1. Integrin alphaVbeta6 and TGF-beta activation: A review on integrin alphaVbeta6's role in TGF-beta activation. [PubMed Search: "integrin alphaVbeta6 TGF-beta activation fibrosis"]
2. TGF-beta in immune suppression: A broad overview of TGF-beta's role in immune regulation. [PubMed Search: "TGF-beta immune suppression review"]
3. TGF-beta and beta cell biology: A review or research article on the effects of TGF-beta on pancreatic beta cells. [PubMed Search: "TGF-beta pancreatic beta cell proliferation"]
4. Targeting fibrosis in diabetes: Research on therapeutic strategies for pancreatic fibrosis in diabetes. [PubMed Search: "pancreatic fibrosis diabetes treatment"]

9. Pancreatic Immune and Stromal Cell-Cell Interaction Patterns Across Diabetes Conditions

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies statistically significant differences in cell-cell interactions (CCIs) among major immune (Macrophage, Dendritic cell, Mast cell) and stromal (Stellate cell, Smooth muscle cell, Fibroblast) cell types within the human pancreas, across three conditions: non-diabetic, prediabetes, and type2_diabetes. Utilizing single-cell RNA-seq data, CellPhoneDB was employed to infer ligand-receptor interactions, and the results are visualized as a dot plot, highlighting the top 25 most significantly elevated CCIs for each condition. The plot shows the standardized mean interaction strength (color intensity) and the statistical significance (-log10(p-value), dot size) for each sample, grouped by condition.

Visual Summary

The dot plot, titled "Condition-specific CCI pattern," effectively visualizes the differential activation of cell-cell interactions across individual samples and disease conditions.

• Overall Structure: The y-axis represents individual samples, grouped by condition (non_diabetic, prediabetes, type2_diabetes). The x-axis lists specific cell-cell interaction indices, which are ligand-receptor pairs between two cell types. These interactions are clustered into three main blocks, corresponding to interactions predominantly elevated in the non-diabetic, prediabetes, and type2_diabetes conditions, respectively.
• Color and Size Encoding: The color intensity (red scale) indicates the standardized mean interaction strength (darker red = higher strength). The size of the dot represents the negative logarithm of the p-value (-log10(p)), where larger dots signify higher statistical significance (smaller p-value).

Condition-Specific Patterns

• Non-diabetic: This condition exhibits a robust set of CCIs, characterized by relatively strong interaction strengths (dark red dots) and high significance (large dot sizes). Interactions predominantly involve Stellate cells and Endothelial cells, with prominent roles for COL (Collagen) integrin complexes (e.g., COL18A1_integrin_a1b1_complex--Stellate/Endothelial cell) and DLL4-NOTCH signaling (e.g., DLL4_NOTCH1--Endothelial cell/Stellate cell). These likely represent healthy tissue maintenance and basic signaling.
• Prediabetes: This condition shows a notably less dense pattern of highly significant interactions compared to the other two. While some interactions are present, they generally appear less widespread across samples and often with lower mean strength or significance. Interactions such as LAMC1_integrin_a6b1_complex--Stellate/Ductal cell and CDH6_CDH6--Ductal cell/Stellate cell are present, indicating early changes in cell adhesion and potentially matrix interactions. Macrophage interactions (e.g., VEGFA_NRP1--Macrophage/Stellate cell) also emerge here.
• Type2_diabetes: This condition displays the most distinct and extensive pattern of highly active and significant CCIs. A large cluster of interactions, predominantly involving Pancreatic Stellate cells interacting with various pancreatic cells (Beta cell, Alpha cell, Delta cell, Acinar cell, Ductal cell), stands out with uniformly dark red, large dots across most samples. Key interactions include:
• TNFSF12-TNFRSF12A (TWEAK-Fn14 axis): Highly prominent between Stellate cells and most endocrine (Alpha, Beta, Delta), exocrine (Acinar), and Ductal cells.
• BMP signaling (BMP5-ACVR1/BMPR2 complex): Especially involving Stellate-Beta cell interactions.
• HBEGF-EGFR: Noted between Stellate cells and Beta cells.
• Cholesterol metabolism-related interactions: Such as Desmosterol_byDHCR7_NR1H2 (LXRβ) with Stellate cells and Acinar/Ductal cells.

Biological Interpretation

The analysis reveals a dynamic rewiring of cell-cell communication networks in the pancreas during the progression from a non-diabetic state to type 2 diabetes.

• Pancreatic Stellate Cells (PSCs) as Central Players: PSCs are consistently involved in significant interactions across all conditions, but their interaction profile dramatically shifts in type2_diabetes. PSCs are known to regulate the pancreatic microenvironment, and their activation is strongly implicated in pancreatic fibrosis and inflammation, which are hallmarks of T2D pathology.
• ECM Remodeling and Basal Signaling in Non-Diabetic Pancreas: The prevalence of COL-integrin interactions and Notch signaling in non-diabetic individuals suggests active extracellular matrix (ECM) maintenance and homeostatic signaling between stromal and endothelial cells, essential for normal tissue structure and function.
• Early Changes in Prediabetes: The relatively sparse yet distinct CCI profile in prediabetes points to initial, perhaps subtle, alterations in pancreatic cell communication. Cadherin-mediated adhesion (CDH6) and early involvement of macrophages (VEGFA-NRP1) could signify the onset of cellular stress, inflammation, or remodeling processes preceding overt diabetes.
• Inflammatory and Fibrotic Signature in Type 2 Diabetes: The overwhelming dominance of specific CCIs in type2_diabetes highlights pathways associated with chronic inflammation, fibrosis, and metabolic dysregulation:
• TNFSF12-TNFRSF12A (TWEAK-Fn14) Axis: This pathway is a potent inducer of inflammation, apoptosis, and fibrosis. Its widespread activation between PSCs and pancreatic endocrine/exocrine/ductal cells in T2D is a strong indicator of an exacerbated inflammatory and pro-fibrotic environment contributing to islet dysfunction and damage [1].
• Altered BMP Signaling: BMPs play diverse roles in pancreatic development and function. Dysregulation of BMP5 signaling, particularly involving PSCs and Beta cells, could contribute to impaired beta-cell mass and function in T2D [2].
• HBEGF-EGFR Signaling: The activation of HBEGF-EGFR between PSCs and Beta cells might reflect aberrant growth factor signaling that could impact beta-cell proliferation or survival, potentially contributing to pathological remodeling [3].
• Lipid Metabolism Dysregulation (Desmosterol-LXR/RXR): The prominence of interactions involving Desmosterol_byDHCR7_NR1H2 (LXRβ) signifies significant alterations in cholesterol and lipid metabolism within the pancreas. LXRβ plays a crucial role in regulating lipid homeostasis and inflammation, and its dysregulation can contribute to lipotoxicity and insulin resistance in T2D [4].

Clinical or Translational Implications

The identified condition-specific cell-cell interaction patterns provide valuable insights into the pathobiology of type 2 diabetes and suggest potential therapeutic targets.

• Therapeutic Targeting of PSCs: The central role of pancreatic stellate cells and their highly active, distinct interactions in type2_diabetes positions them as a critical therapeutic target. Interventions aimed at modulating PSC activation or their specific interactions could mitigate pancreatic inflammation and fibrosis, potentially preserving islet function.
• TNFSF12-TNFRSF12A Axis as a Drug Target: Given its strong activation and pro-inflammatory/pro-fibrotic role in T2D, blocking the TNFSF12-TNFRSF12A axis could be a promising strategy to reduce pancreatic damage and improve disease outcomes.
• Addressing Metabolic Dysregulation: The prominent cholesterol-related interactions (e.g., Desmosterol-LXR/RXR) underscore the importance of lipid metabolism in T2D pathogenesis. Therapies that restore lipid homeostasis within the pancreas, perhaps through LXR modulators, could offer a novel approach to prevent or treat the disease.
• Biomarker Discovery: The identified condition-specific ligand-receptor pairs could serve as novel biomarkers for early detection of prediabetes progression or for monitoring disease severity and therapeutic response in type 2 diabetes.

---

References:

1. TNFSF12 (TWEAK) and TNFRSF12A (Fn14) in diabetes and inflammation: [PubMed Search: TWEAK Fn14 diabetes pancreas]
2. BMP signaling in pancreatic function and diabetes: [GeneCards: BMP5]
3. HBEGF-EGFR in pancreatic disease: [GeneCards: HBEGF]
4. LXRβ (NR1H2) and cholesterol metabolism in diabetes: [UniProt: NR1H2] ; [PubMed Search: LXR beta diabetes pancreas]

10. Macrophage Condition-Specific Surfaceome Markers in Pancreatic Tissue

[Analysis Visualization Results]...

Analysis Overview

This analysis identifies condition-specific surfaceome markers in Macrophages derived from human pancreatic single-cell RNA-seq data. The dot plot visualizes the expression of these markers across individual samples, grouped by 'non_diabetic' and 'prediabetes' conditions. The goal is to highlight surface proteins that are distinctly expressed on Macrophages in these different metabolic states, which could serve as potential biomarkers or therapeutic targets.

Visual Summary

The dot plot displays the mean expression level (color intensity, redder indicates higher) and the fraction of cells expressing a gene (dot size, larger indicates higher) for a set of identified surfaceome markers across different pancreatic samples. The samples are grouped into 'non_diabetic' (e.g., MS20017, MS19038, MS17002) and 'prediabetes' (e.g., MS20005, MS19023, MS17005, MS17006, MS17004) conditions.

• Non-diabetic Macrophage Markers: Several genes, including CD69, GPR35, TMX4, CD300C, CD1D, and PLXNB2, show high mean expression and are expressed in a substantial fraction of cells primarily within the 'non_diabetic' samples. These genes appear to be downregulated or expressed at much lower levels in the 'prediabetes' condition.
• Prediabetes Macrophage Markers: A distinct and larger set of genes, such as CD84, PTPRN, ADCY3, a series of Solute Carrier (SLC) family members (e.g., SLC30A1, SLC2A1, SLC15A4, SLC39A6, SLC7A2, SLC33A1, SLC39A10, SLC2A13), CPM, SCARB1, CSF3R, SCAP, FURIN, SUCNR1, DLK1, PKD2L1, PTGER4, and PLPP3, are prominently expressed in 'prediabetes' samples. These markers show high mean expression and are detected in a notable fraction of cells specifically under prediabetic conditions, with minimal or no expression in non-diabetic samples.
• Sample Variability: While most markers show consistent patterns within their respective conditions, some degree of sample-to-sample variability in expression levels and cell fraction is observable, particularly for certain markers in the 'prediabetes' group.

Biological Interpretation

The distinct sets of surfaceome markers highlight significant phenotypic differences in pancreatic Macrophages between non-diabetic and prediabetic states. Macrophages are key immune cells known for their plasticity and their critical role in both immune surveillance and metabolic regulation, particularly in the context of insulin resistance and type 2 diabetes development.

• Non-diabetic Macrophage Signature: The markers enriched in non-diabetic macrophages, such as CD69 (an early activation marker and regulator of tissue residency) and CD1D (involved in presenting lipid antigens to NKT cells for immune regulation), may reflect a homeostatic, tissue-resident, or immunoregulatory macrophage phenotype in a healthy pancreatic microenvironment. GPR35 and TMX4 could point to baseline metabolic or redox signaling.
• Prediabetes Macrophage Signature - Metabolic Reprogramming and Inflammation: The substantial upregulation of a diverse array of surfaceome markers in prediabetic macrophages strongly suggests metabolic reprogramming and an altered inflammatory state.
• Metabolic Transporters (SLCs): The numerous SLC family members (e.g., SLC2A1/GLUT1, SLC15A4, various zinc transporters like SLC30A1 and SLC39A6) point to significantly altered nutrient, ion, and metabolite transport dynamics. Upregulation of SLC2A1 (GLUT1) is particularly noteworthy, as GLUT1 is often induced in activated immune cells to support increased glucose uptake, fueling metabolic and inflammatory processes characteristic of metabolic stress in prediabetes [1]. The changes in zinc transporters may indicate altered zinc homeostasis, which is crucial for pancreatic islet function and immune responses [2].
• Lipid Metabolism and Scavenger Receptors: The expression of SCARB1 (Scavenger receptor class B type 1), involved in high-density lipoprotein (HDL) metabolism, and SCAP (SREBP cleavage-activating protein), crucial for cholesterol biosynthesis, indicates significant shifts in lipid handling by macrophages. This is highly relevant, given the dyslipidemia associated with prediabetes and the known role of macrophages in lipid accumulation and foam cell formation in metabolic diseases [3].
• Inflammatory and Activation Markers: CD84 (SLAM family receptor, involved in immune activation), CPM (carboxypeptidase M, involved in processing inflammatory peptides), CSF3R (receptor for G-CSF, involved in myeloid cell proliferation and differentiation), and PTGER4 (PGE2 receptor, mediating inflammatory responses) suggest an activated or inflammatory macrophage phenotype in prediabetes. SUCNR1 (succinate receptor 1) links macrophage activation directly to metabolic changes, as succinate is a pro-inflammatory metabolite that accumulates during metabolic stress and activates immune cells [4].
• Other Noteworthy Markers: PTPRN (protein tyrosine phosphatase) could be involved in altered signaling pathways relevant to cell adhesion and migration, while FURIN (proprotein convertase) plays roles in processing many precursor proteins, including those involved in inflammation and metabolism. DLK1 has been implicated in adipogenesis and metabolic regulation.

These findings suggest that during prediabetes, pancreatic macrophages undergo a substantial shift towards a phenotype characterized by altered nutrient sensing, lipid metabolism, and inflammatory activation, consistent with their proposed role in driving pancreatic inflammation and insulin resistance.

Clinical or Translational Implications

The identified condition-specific surfaceome markers in pancreatic macrophages offer several translational avenues:

• Early Diagnostic/Prognostic Biomarkers: The distinct expression profiles could potentially be exploited as early diagnostic or prognostic biomarkers for prediabetes. If these markers are detectable in circulating monocytes or accessible tissue biopsies, they could indicate the progression of metabolic dysfunction in the pancreas.
• Therapeutic Targets for Prediabetes: As these are surfaceome markers, they represent excellent candidates for targeted therapeutic interventions.
• For example, inhibition of SLC2A1 (GLUT1) activity on prediabetic macrophages could reduce their glucose uptake and subsequent pro-inflammatory activation, thereby ameliorating metabolic stress.
• Modulating SCARB1 or SCAP activity could impact lipid metabolism within these cells, potentially reducing lipid accumulation and inflammatory signaling.
• Targeting SUCNR1 or PTGER4 could dampen the inflammatory responses initiated by metabolic intermediates or prostaglandins.
• These markers could serve as targets for cell-specific drug delivery, utilizing antibodies or other ligands to deliver anti-inflammatory or metabolism-modulating agents directly to prediabetic macrophages, minimizing off-target effects.
• Experimental Validation: Further experimental validation is warranted to confirm the functional roles of these markers in pancreatic macrophage biology in prediabetes. Techniques such as flow cytometry, immunohistochemistry, and functional assays (e.g., macrophage polarization assays, metabolic flux analysis) could elucidate their precise contribution to disease progression and their potential as therapeutic targets.

References

1. SLC2A1 (GLUT1) in activated immune cells: PubMed Search: "GLUT1 macrophage activation inflammation" https://pubmed.ncbi.nlm.nih.gov/?term=GLUT1+macrophage+activation+inflammation
2. Zinc transporters in immunity and metabolism: UniProt: SLC30A1 https://www.uniprot.org/uniprotkb/O43831/entry, UniProt: SLC39A6 https://www.uniprot.org/uniprotkb/O43272/entry
3. SCARB1 and macrophage lipid metabolism in metabolic disease: GeneCards: SCARB1 https://www.genecards.org/cgi-bin/carddisp.pl?gene=SCARB1
4. SUCNR1 and succinate in inflammation: PubMed Search: "SUCNR1 succinate macrophage inflammation" https://pubmed.ncbi.nlm.nih.gov/?term=SUCNR1+succinate+macrophage+inflammation

11. Pancreatic Beta Cell Cycle Gene Expression Shifts in Prediabetes and Type 2 Diabetes

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates the differential expression of 12 selected cell cycle-related genes in pancreatic Beta cells across three conditions: non_diabetic, prediabetes, and type2_diabetes. The expression levels are compared using box plots, with statistical significance indicated by p-values. The reference condition for comparisons is non_diabetic. This allows us to identify how cell cycle regulation in beta cells changes with the progression of diabetes.

Visual Summary

The box plots display the gene expression (sample mean) for each of the 12 selected cell cycle genes across the three conditions. Key observations include:

• Prediabetes vs. Non-diabetic: A consistent and statistically significant downregulation of many cell cycle-promoting genes is observed in the prediabetes condition compared to non_diabetic. Genes such as ANAPC13, ANAPC5, BUB3, CCND3, CDC27, E2F4, HDAC2, MAD2L2, and TFDP1 all show lower expression in prediabetes. GADD45G also exhibits lower expression in prediabetes.
• Type 2 Diabetes vs. Non-diabetic: In contrast to prediabetes, beta cells in type2_diabetes show a significant *upregulation* of several cell cycle inhibitor genes. Specifically, CDKN1A (p21) and CDKN2A (p16INK4a) are markedly increased in type2_diabetes compared to non_diabetic. GADD45G is also significantly upregulated in type2_diabetes. For other proliferative genes, expression in type2_diabetes is generally similar to non_diabetic, or significantly higher than prediabetes (e.g., ANAPC13, ANAPC5, CCND3, E2F4, GADD45G, HDAC2, MAD2L2, TFDP1).
• General Trend: The most striking pattern is a widespread suppression of cell cycle activity in the prediabetes stage, followed by a state of active cell cycle arrest, possibly senescence, in type2_diabetes.

Biological Interpretation

Pancreatic beta cells are critical for maintaining glucose homeostasis, and their ability to proliferate and function is severely compromised in diabetes. The observed differential expression of cell cycle genes provides insights into the distinct beta cell pathologies in prediabetes and type 2 diabetes:

1. Impaired Proliferative Capacity in Prediabetes:

• Downregulation of Proliferative Drivers: The significant decrease in genes like CCND3 (Cyclin D3) [GeneCards], a key G1 cyclin, indicates a reduced drive for cell cycle progression in beta cells during prediabetes.
• Reduced Anaphase-Promoting Complex (APC/C) and Spindle Assembly Checkpoint (SAC) Activity: Lower expression of ANAPC13, ANAPC5, and CDC27 (all APC/C components) [PubMed Search: Anaphase-Promoting Complex cell cycle] along with BUB3 and MAD2L2 (SAC proteins) [GeneCards] suggests that not only is the entry into the cell cycle suppressed, but also the proper progression through mitosis may be compromised or the cells are not entering these phases.
• Altered Transcription Factor Regulation: Decreased E2F4 and TFDP1 expression further supports a shift towards a less proliferative or quiescent state, as these factors are crucial for regulating genes involved in DNA synthesis and cell cycle progression.
• Overall, this pattern suggests that beta cells in prediabetes may be failing to adequately proliferate to compensate for increasing insulin demand, entering a state of functional decline and reduced plasticity.

2. Active Cell Cycle Arrest and Stress Response in Type 2 Diabetes:

• Upregulation of Cell Cycle Inhibitors: The significant increase in CDKN1A (p21) [GeneCards] and CDKN2A (p16INK4a) [GeneCards] is a strong indicator of activated cell cycle arrest pathways. p21 is often induced by p53 in response to DNA damage or stress, while p16INK4a is a well-known marker and mediator of cellular senescence. This suggests that beta cells in established T2D are not merely failing to proliferate but are actively arresting their cell cycle, likely due to chronic metabolic stress, inflammation, and DNA damage.
• DNA Damage Response: The upregulation of GADD45G (Growth Arrest and DNA Damage-inducible protein gamma) [GeneCards] in type 2 diabetes further supports the presence of cellular stress and DNA damage, which can trigger cell cycle arrest and potentially senescence or apoptosis.
• This pattern points to a more advanced stage of beta cell dysfunction in T2D, characterized by irreversible cell cycle exit (senescence) and/or heightened stress responses, contributing to the progressive loss of beta cell mass and function.

Clinical or Translational Implications

The distinct patterns of cell cycle gene expression in prediabetes and type 2 diabetes have significant clinical implications:

• Early Intervention Targets: In prediabetes, the widespread suppression of proliferative genes suggests that strategies aimed at stimulating beta cell regeneration or improving their functional capacity at this early stage might be beneficial. Understanding the triggers for this early decline in proliferative drive could identify novel therapeutic targets.
• Beta Cell Senescence in T2D: The upregulation of senescence markers like CDKN1A and CDKN2A in T2D beta cells reinforces the concept of cellular senescence contributing to beta cell failure. Senolytic drugs, which selectively remove senescent cells, or senomorphic drugs, which modulate the senescent phenotype, could represent future therapeutic avenues for preserving beta cell function in established type 2 diabetes [PubMed Search: beta cell senescence type 2 diabetes therapeutics].
• Disease Progression Markers: These differentially expressed cell cycle genes could serve as biomarkers to monitor the progression of beta cell dysfunction from prediabetes to type 2 diabetes. Their distinct patterns might allow for more precise staging of the disease and personalized treatment strategies.

12. Pancreatic Cell-Type Specific Gene Ontology Upregulation in Prediabetes and Type 2 Diabetes

[Analysis Visualization Results]...

Analysis Overview

This analysis presents Gene Ontology (GO) enrichment results (GSA_up) for Ductal cells, Acinar cells, and Beta cells in the pancreas. The results compare each disease state (non_diabetic, prediabetes, type2_diabetes) against all other conditions combined within each cell type. The visualization is a dot plot, where the size and color intensity of each dot represent the significance of pathway upregulation, quantified as -log(P-value). Larger and darker red dots indicate more highly significant enrichment of a pathway in a given cell type and condition.

Visual Summary

The dot plot effectively illustrates the differential enrichment of various biological pathways across three key pancreatic cell types (Ductal, Acinar, Beta) and disease conditions (non_diabetic, prediabetes, type2_diabetes), with each condition compared against all others.

1. Baseline (non_diabetic): For all three cell types, the "non_diabetic_vs_others" comparisons show relatively few and less significant (smaller, lighter dots) enriched pathways, suggesting a more homeostatic state or fewer stark differences compared to the collective "others" group.
2. Prediabetes and Type 2 Diabetes: In contrast, the "prediabetes_vs_others" and "type2_diabetes_vs_others" columns for all three cell types show a striking and widespread upregulation of numerous pathways, often with very high significance (large, dark red dots). This indicates significant biological shifts occurring during disease progression.
3. Beta Cell Specificity: Beta cells exhibit the most pronounced and extensive enrichment of highly significant pathways, particularly in both prediabetes and type 2 diabetes conditions. This includes a broad spectrum of pathways related to cellular stress, metabolism, protein handling, and signaling.
4. Commonly Enriched Pathways: Pathways frequently and highly enriched across prediabetes and type 2 diabetes in all three cell types include: "Apoptosis", "Oxidative phosphorylation", "Protein processing in endoplasmic reticulum", "Ribosome", "Tight junction", "Ubiquitin mediated proteolysis", and "mTOR signaling pathway". These pathways are indicative of increased cellular stress, altered metabolic activity, and perturbed protein homeostasis.
5. Disease Relevance: Several disease-specific pathways like "Diabetic cardiomyopathy", "Pancreatic cancer", and "Pathways of neurodegeneration" are also prominently enriched, especially in Beta cells during disease states.

Biological Interpretation

The observed shifts in upregulated GO pathways provide critical insights into the cellular responses of different pancreatic cell types during the progression of type 2 diabetes.

1. Pancreatic Cellular Stress Response: The consistent and highly significant enrichment of pathways such as "Apoptosis", "Protein processing in endoplasmic reticulum", "Ubiquitin mediated proteolysis", "Autophagy", and "Cellular senescence" across Ductal, Acinar, and particularly Beta cells in prediabetes and type 2 diabetes, points towards a widespread cellular stress response within the pancreas.

• "Protein processing in endoplasmic reticulum" enrichment suggests increased ER stress, a known contributor to beta-cell dysfunction and death in diabetes [PMID: 29555627]. The cell tries to cope with misfolded proteins, and if this fails, "Apoptosis" (programmed cell death) is triggered.
• "Ubiquitin mediated proteolysis" also relates to protein quality control and degradation, indicating a burden on cellular machinery.
• "Autophagy" represents a cellular recycling process, often upregulated in response to stress to maintain homeostasis or remove damaged components [PubMed Search: "Autophagy diabetes beta cell" PubMed Search].

2. Metabolic Dysregulation: The strong enrichment of "Oxidative phosphorylation" suggests alterations in mitochondrial function and energy metabolism. While essential for ATP production, dysregulated oxidative phosphorylation can lead to increased reactive oxygen species and oxidative stress, contributing to cellular damage in diabetes [PMID: 32367803]. Pathways like "Valine, leucine and isoleucine degradation" also highlight metabolic changes, as branched-chain amino acid metabolism is often perturbed in insulin resistance and type 2 diabetes.
3. Signaling Pathway Dysregulation: Upregulation of critical signaling cascades like "mTOR signaling pathway", "PI3K-Akt signaling pathway", and "MAPK signaling pathway" signifies profound changes in cell growth, proliferation, survival, and nutrient sensing. Dysregulation of these pathways is directly implicated in beta-cell failure and insulin resistance [GeneCards: MTOR GeneCards, GeneCards: PIK3CA GeneCards]. The "HIF-1 signaling pathway" involvement points to potential hypoxia or pseudohypoxia responses, often observed in inflammatory or metabolic stress conditions.
4. Cellular Integrity and Architecture: Enrichment of "Adherens junction", "Focal adhesion", "Tight junction", and "Regulation of actin cytoskeleton" pathways indicates potential remodeling or compromise of cellular structure and cell-cell interactions. This could impact tissue integrity and function, particularly in the pancreatic islets.
5. Beta Cell Vulnerability: The significantly higher number and greater significance of enriched pathways in Beta cells underscore their central role and extreme vulnerability in diabetes pathogenesis. The pervasive cellular stress, metabolic dysfunction, and altered signaling in Beta cells directly contribute to their loss of function and mass, which is a hallmark of type 2 diabetes. While Ductal and Acinar cells also show significant stress responses, the magnitude in Beta cells is particularly striking.

Clinical or Translational Implications

These findings have several important clinical and translational implications:

1. Biomarker Identification: The identified significantly upregulated pathways and their constituent genes could serve as valuable biomarkers for the early detection or monitoring of prediabetes and type 2 diabetes progression. Cell-type-specific biomarkers could offer more precise diagnostic and prognostic tools.
2. Therapeutic Target Identification: Pathways involved in cellular stress ("Apoptosis", "ER stress", "Autophagy"), metabolic dysregulation ("Oxidative phosphorylation", "mTOR signaling pathway", "PI3K-Akt signaling pathway"), and inflammation represent promising therapeutic targets. Developing interventions that modulate these pathways specifically in affected pancreatic cell types, especially Beta cells, could help preserve beta-cell function and mass, thereby slowing or halting disease progression. For instance, modulating ER stress has been actively explored as a strategy to protect beta cells [PubMed Search: "ER stress diabetes therapy" PubMed Search].
3. Understanding Disease Heterogeneity: Recognizing that multiple pancreatic cell types are affected, albeit to varying degrees, emphasizes the complex nature of type 2 diabetes. Therapeutic strategies might need to consider a multi-pronged approach addressing the dysfunction in different cell populations.
4. Drug Repurposing: The overlap of enriched pathways with other disease contexts (e.g., "Pathways in cancer", "Pathways of neurodegeneration") might open avenues for drug repurposing, where existing drugs targeting these shared mechanisms could be investigated for their potential in diabetes.

13. Gene Set Enrichment Analysis Reveals Condition-Specific Pathway Dysregulation in Pancreatic Cell Types

[Analysis Visualization Results]...

Analysis Overview

This analysis investigates gene set enrichment (GSEA) across key pancreatic cell types—Beta cells, Alpha cells, Ductal cells, and Macrophages—under different metabolic conditions: non_diabetic, prediabetes, and type2_diabetes. For each cell type, the analysis compares gene expression in a given condition against the combined expression from the other two conditions (e.g., non_diabetic vs. [prediabetes + type2_diabetes]). The results are visualized as a dot plot, where the color intensity of each dot reflects the Normalized Enrichment Score (NES) (red for positive enrichment, blue for negative enrichment) and the dot size indicates the statistical significance (-log(p-value)). This approach highlights pathways that are uniquely activated or suppressed in specific cell types and disease states within the pancreatic microenvironment.

Visual Summary

The dot plot effectively displays the enrichment patterns of 80 distinct gene sets across the selected cell types and conditions.

• Insulin secretion pathway stands out, showing strong positive enrichment (red, large dots) in non_diabetic Beta cells, but significant negative enrichment (blue, large dots) in Beta cells from both prediabetes and type2_diabetes conditions. This is a crucial observation indicating a progressive decline in insulin secretory capacity.
• Apoptosis pathway exhibits positive enrichment in type2_diabetes Beta cells, suggesting increased cell death in this critical cell population. Alpha cells in type2_diabetes also show enrichment for apoptosis.
• Lipid and atherosclerosis pathways show widespread positive enrichment (red) across almost all cell types (Beta, Alpha, Ductal, Macrophage) in prediabetes and type2_diabetes conditions, indicating systemic metabolic stress.
• Glycolysis / Gluconeogenesis pathway shows an interesting inverse pattern in Beta cells, being enriched in non_diabetic but depleted in prediabetes and type2_diabetes. In contrast, macrophages show positive enrichment for glycolysis-related pathways in prediabetes and type2_diabetes.
• Immune-related pathways, such as "Antigen processing and presentation," "PD-L1 expression and PD-1 checkpoint pathway in cancer," and "Viral protein interaction with cytokine and cytokine receptor," are consistently enriched in Macrophages across all conditions, highlighting their persistent immune surveillance and activation roles. Some "infection" pathways also show enrichment, particularly in macrophages and ductal cells in diabetic states.
• Neuroactive ligand-receptor interaction pathway is consistently enriched (red, large dots) across all cell types and conditions, suggesting active and complex neuro-endocrine communication within the pancreas.

Biological Interpretation

Beta Cell Dysfunction and Apoptosis

The most prominent finding is the progressive impairment of "Insulin secretion" in Beta cells as conditions shift from non_diabetic to prediabetes and type2_diabetes. This is a hallmark of type 2 diabetes pathophysiology, where Beta cells fail to produce and secrete sufficient insulin. Concurrently, the positive enrichment of "Apoptosis" in type2_diabetes Beta cells indicates increased Beta cell death, a key contributor to Beta cell mass reduction and further exacerbation of insulin deficiency PubMed search: beta cell apoptosis type 2 diabetes. The shift from positive to negative enrichment in "Glycolysis / Gluconeogenesis" and "Fatty acid degradation" in diabetic Beta cells suggests a metabolic inflexibility or a shift away from efficient glucose utilization and lipid breakdown, potentially contributing to lipotoxicity and glucotoxicity. Impaired "Autophagy" in diabetic Beta cells could also contribute to cellular stress and accumulation of damaged organelles.

Alpha Cell Changes and Broader Islet Stress

Alpha cells, responsible for glucagon secretion, also show signs of stress. Similar to Beta cells, "Apoptosis" is enriched in Alpha cells in type2_diabetes, suggesting that islet cell death is not restricted to Beta cells but may affect other endocrine populations in advanced disease. The enrichment of "Lipid and atherosclerosis" pathways in Alpha cells further points to broader metabolic dysregulation affecting islet integrity.

Ductal Cell Responses

Ductal cells exhibit enrichment for "Lipid and atherosclerosis" and "VEGF signaling pathway" in type2_diabetes. VEGF signaling is crucial for angiogenesis and tissue repair, suggesting that ductal cells might be involved in responding to microvascular complications or local tissue remodeling within the diabetic pancreas. Enrichment of certain "infection" or inflammatory pathways in ductal cells, particularly in prediabetes and type2_diabetes, could indicate their role in initiating or propagating inflammatory responses within the exocrine pancreas.

Macrophage Activation and Pancreatic Inflammation

Macrophages show a consistently high level of activation, with strong enrichment for "Antigen processing and presentation", "PD-L1 expression and PD-1 checkpoint pathway in cancer", and "Viral protein interaction with cytokine and cytokine receptor" across all conditions. This underscores their role as key immune surveillance cells in the pancreas. The enrichment of "Lipid and atherosclerosis" and "Glycolysis / Gluconeogenesis" pathways in macrophages in prediabetes and type2_diabetes is noteworthy. Increased glycolysis in macrophages is often associated with a pro-inflammatory (M1-like) phenotype, suggesting that pancreatic macrophages adopt a metabolically reprogrammed, inflammatory state in diabetic conditions, contributing to insulitis and tissue damage PubMed search: macrophage glycolysis inflammation diabetes pancreas.

Systemic Metabolic Dysregulation

The widespread enrichment of pathways like "Lipid and atherosclerosis" across multiple pancreatic cell types (Beta, Alpha, Ductal, Macrophage) in prediabetes and type2_diabetes underscores the systemic nature of metabolic dysfunction in diabetes. This indicates that dyslipidemia and related inflammatory processes affect not only the endocrine cells but also structural and immune cells within the pancreatic microenvironment. The continuous strong enrichment of "Neuroactive ligand-receptor interaction" suggests persistent, complex neuro-endocrine communication that attempts to maintain or adapt pancreatic function, albeit unsuccessfully in the context of progressive disease.

Clinical or Translational Implications

These GSEA findings offer significant insights into the cellular and molecular mechanisms underlying the progression of type 2 diabetes in the human pancreas.

1. Biomarker Discovery and Early Intervention: The clear decrease in "Insulin secretion" and increase in "Apoptosis" pathways in Beta cells during prediabetes and type2_diabetes could highlight potential early biomarkers for disease progression or targets for interventions aimed at preserving Beta cell function and mass.
2. Therapeutic Targets: Pathways involved in lipid metabolism ("Lipid and atherosclerosis"), glucose metabolism ("Glycolysis / Gluconeogenesis"), and autophagy, particularly in Beta cells, could represent novel therapeutic targets to improve metabolic health and Beta cell survival in diabetes.
3. Inflammation in Diabetes: The consistent activation of inflammatory and immune-related pathways in macrophages reinforces the role of pancreatic inflammation in diabetes pathogenesis. Targeting specific macrophage activation pathways (e.g., those involving PD-L1 or viral interaction signaling) could be explored as a strategy to mitigate insulitis and preserve islet function.
4. Multi-cellular Crosstalk: The pervasive "Neuroactive ligand-receptor interaction" enrichment suggests that targeting neuro-endocrine signaling could modulate overall pancreatic function and inter-cellular communication in diabetes.
5. Disease Heterogeneity: The cell-type-specific pathway enrichments emphasize the importance of considering cellular heterogeneity when developing diabetes therapies, moving beyond a Beta-cell-centric view to include other crucial pancreatic cell types like Alpha cells, Ductal cells, and immune cells.

14. Discussion

The analysis of single-cell RNA sequencing data from the human pancreas provides a comprehensive view of cellular and molecular changes occurring during the progression of type 2 diabetes (T2D). A central and consistent finding across multiple analyses is the progressive decline in the relative proportion of Beta cells from non-diabetic to prediabetic and overtly diabetic states. This quantitative reduction is underpinned by profound changes in Beta cell intrinsic programs: in prediabetes, Beta cells exhibit a widespread suppression of cell cycle-promoting genes, suggesting an early impairment in their proliferative capacity to compensate for increased insulin demand. This contrasts with a commonly held view that beta cell proliferation might be initially upregulated to compensate before exhaustion. Furthermore, in established T2D, Beta cells show a significant upregulation of cell cycle inhibitors and senescence markers like CDKN1A (p21) and CDKN2A (p16INK4a), along with GADD45G, indicating an active and irreversible cell cycle arrest, contributing to their functional decline and mass loss. This transition from impaired proliferation to active senescence is further supported by the clear negative enrichment of 'Insulin secretion' and positive enrichment of 'Apoptosis' in diabetic Beta cells from GSEA, alongside widespread cellular stress pathways from GO analysis.

Beyond the endocrine compartment, immune and stromal cells play critical roles. Pancreatic macrophages undergo a significant re-orchestration in their subset composition and metabolic state. While visual inspection of population plots might suggest an M1 dominance in T2D, statistical analysis reveals an increase in Mac (M2B) proportions in prediabetes and T2D compared to non-diabetic controls, alongside a modest decrease in M1 macrophages in T2D versus non-diabetic. This suggests a more nuanced shift than a simple M1 prevalence, pointing to an adaptive or maladaptive role for M2B macrophages. Furthermore, prediabetic macrophages display a distinct surfaceome signature, with marked upregulation of metabolic transporters like SLC2A1 (GLUT1) and genes involved in lipid metabolism (SCARB1, SCAP), alongside inflammatory activation markers such as SUCNR1. This strongly indicates a metabolic reprogramming towards a pro-inflammatory phenotype, consistent with GSEA findings of enriched glycolysis in macrophages in diabetic conditions, fueling local inflammation and insulitis, which could exacerbate Beta cell damage.

Pancreatic Stellate Cells (PSCs), key mediators of fibrosis, are consistently implicated in altered cell-cell interactions. In T2D, there is a marked increase in TGFB1-TGFBR2/TGFBR3 signaling involving PSCs and various pancreatic cells (Beta, Alpha, Delta, Ductal), driving extracellular matrix remodeling and fibrosis. Notably, the TNFSF12-TNFRSF12A (TWEAK-Fn14) axis emerges as highly prominent in T2D, connecting PSCs with endocrine, exocrine, and ductal cells, suggesting a robust inflammatory and pro-fibrotic signaling cascade previously less emphasized in the context of T2D pancreatic pathology. A particularly striking finding from CCI analysis is the disappearance of the Acinar|Ductal TGFB1_TGFbeta_receptor2 interaction in both prediabetes and T2D, implying a critical breakdown in inter-compartmental communication essential for overall pancreatic homeostasis early in disease development.

Finally, widespread enrichment of 'Lipid and atherosclerosis' pathways across Beta, Alpha, Ductal, and Macrophage cells in diabetic conditions highlights the pervasive systemic metabolic dysregulation affecting the entire pancreatic microenvironment. The consistent presence of 'Neuroactive ligand-receptor interaction' pathways further suggests complex, yet perhaps perturbed, neuro-endocrine modulation of pancreatic function throughout disease progression.

Hypotheses:

1. Beta cell senescence, characterized by the upregulation of CDKN1A and CDKN2A, is a primary mechanism driving Beta cell mass loss and functional decline in type 2 diabetes.
2. The increased proportion of Mac (M2B) macrophages in prediabetes and type 2 diabetes, combined with metabolic reprogramming (e.g., elevated SLC2A1/GLUT1), contributes to a maladaptive inflammatory or pro-fibrotic environment within the pancreas.
3. Hyperactivation of TGFB1-TGFBR2/3 and the TNFSF12-TNFRSF12A (TWEAK-Fn14) axis in pancreatic stellate cells is a central driver of pancreatic fibrosis and contributes directly to islet dysfunction in type 2 diabetes.
4. Disruption of Acinar-Ductal cell communication, specifically the loss of TGFB1-TGFbeta_receptor2 interaction, impairs pancreatic resilience and exacerbates disease progression in prediabetes and type 2 diabetes.
5. Altered lipid metabolism and cholesterol-related interactions (e.g., Desmosterol-LXR/RXR) in pancreatic stellate cells and macrophages drive chronic inflammation and tissue damage in type 2 diabetes.

Potential therapeutic targets:

1. TGF-beta signaling pathway (TGFB1-TGFBR2/TGFBR3 axis and integrin alphaVbeta6): Markedly activated in type 2 diabetes, particularly involving pancreatic stellate cells, driving fibrosis and contributing to islet dysfunction. Integrin alphaVbeta6 specifically activates latent TGF-beta. Evidence: CCI analyses show strong upregulation of TGFB1-TGFBR2/3 interactions (Section 7, 8, 9). Stellate cells are consistently central to these fibrotic interactions. Validation: Test TGF-beta receptor kinase inhibitors or integrin alphaVbeta6 blocking antibodies in preclinical models of diabetes (e.g., diet-induced obesity mice) to assess reduction in pancreatic fibrosis, improved beta-cell function, and glucose homeostasis.
2. TNFSF12-TNFRSF12A (TWEAK-Fn14 axis): Highly prominent in type 2 diabetes, mediating inflammation and fibrosis between pancreatic stellate cells and various endocrine/exocrine/ductal cells. Evidence: CCI analyses show strong activation of TNFSF12-TNFRSF12A interactions in type 2 diabetes, predominantly involving stellate cells (Section 9). Validation: Evaluate the efficacy of Fn14-blocking antibodies or small molecule inhibitors in reducing pancreatic inflammation, fibrosis, and preserving islet function in diabetic animal models.
3. Macrophage metabolic reprogramming (e.g., GLUT1/SLC2A1, SUCNR1): Prediabetic macrophages show upregulation of SLC2A1 (GLUT1) and SUCNR1, indicative of metabolic shifts fueling a pro-inflammatory phenotype contributing to insulitis. Evidence: Condition-specific surfaceome marker analysis shows significant upregulation of SLC2A1 and SUCNR1 in prediabetic macrophages (Section 10). GSEA shows glycolysis enrichment in diabetic macrophages (Section 13). Validation: Investigate specific inhibitors of SLC2A1 or SUCNR1 in macrophage cultures to attenuate pro-inflammatory responses and test their effect on glucose homeostasis and pancreatic inflammation in diabetic animal models.
4. Beta cell senescence (CDKN1A, CDKN2A): Upregulation of p21 (CDKN1A) and p16 (CDKN2A) in type 2 diabetic Beta cells indicates active cell cycle arrest and senescence, contributing to progressive Beta cell loss. Evidence: Differential gene expression analysis shows significant upregulation of CDKN1A and CDKN2A in type 2 diabetic Beta cells (Section 11). GSA shows apoptosis and cellular senescence pathways enriched in Beta cells (Section 12). Validation: Administer senolytic or senomorphic compounds to diabetic animal models to determine if they can selectively remove senescent Beta cells, improve Beta cell mass, reduce inflammation, and enhance insulin secretion and glucose control.

Follow-up validation ideas:

1. Beta cell senescence validation: Conduct immunofluorescence or single-cell RNA FISH on human pancreatic biopsies to co-localize insulin with CDKN1A, CDKN2A, and GADD45G proteins or mRNA in diabetic islets. Use functional assays on isolated human islets to assess proliferation rates and senescence-associated secretory phenotype (SASP) under glucotoxic/lipotoxic conditions, and test the efficacy of senolytic/senomorphic compounds.
2. Macrophage polarization and metabolism validation: Use multi-parameter flow cytometry or imaging mass cytometry (IMC) on dissociated pancreatic tissue to quantify M1, M2A, M2B, M2C, M2D macrophage subsets and their expression of surface markers like SLC2A1, SUCNR1, and SCARB1 in prediabetic/diabetic vs. non-diabetic samples. Perform in vitro experiments with human primary macrophages to assess the impact of glucose/lipid overload on their polarization, glycolysis (e.g., Seahorse assay), and inflammatory cytokine secretion, with and without pharmacological modulators of SLC2A1 or SUCNR1.
3. Pancreatic stellate cell activation and fibrotic pathways: Employ spatial transcriptomics or high-resolution multiplex immunofluorescence to map the localization and activation status of pancreatic stellate cells and their expression of TGFBR2/3 and TNFRSF12A in relation to fibrotic markers (collagens, alpha-SMA) and islet cells in diabetic pancreatic sections. Perform in vitro assays using primary human pancreatic stellate cells to test the effects of TGFB1 and TNFSF12 on their activation, proliferation, and ECM production, and evaluate the therapeutic potential of pathway inhibitors.
4. Acinar-Ductal communication disruption: Use 3D co-culture models of human acinar and ductal cells to investigate the impact of diabetic stressors on TGFB1-TGFbeta_receptor2 interaction and its downstream effects on cellular function. Validate the presence or absence of this specific interaction at the protein level using proximity ligation assays or co-immunoprecipitation on pancreatic tissue.
5. Lipid metabolism in pancreatic cells: Use lipidomics on sorted pancreatic cell types (stellate cells, macrophages, beta cells) from diabetic and non-diabetic individuals to confirm dysregulation of desmosterol and related lipid species. Investigate the functional impact of LXR agonists/antagonists on lipid handling, inflammation, and fibrotic responses in primary pancreatic cell cultures.

Limitations:

This report is based on single-cell RNA sequencing data, which provides high-resolution insights into cellular heterogeneity and transcriptional states. However, the nature of dissociated single-cell data inherently limits the ability to directly infer the spatial organization and direct physical interactions of cells in vivo. While cell-cell interaction analyses provide probabilistic predictions, experimental validation is crucial to confirm these interactions and their functional consequences. Proportional changes in cell types should be interpreted with caution as they do not directly equate to absolute cell number changes without additional quantification. Furthermore, the findings represent correlative associations with disease states, and direct causality cannot be definitively established from this dataset alone. Future studies incorporating spatial transcriptomics, in situ functional assays, and longitudinal studies in larger, diverse cohorts are needed to further validate and extend these findings.

15. Query List

1. Show UMAPs with condition, sample, major cell type, minor cell type, celltype_subset in 2 columns and save.
2. Show expression of genes CD3D CD4 CD8A CD79A MS4A1 MZB1 CD14 LYZ FBLN1 NOTCH3 EPCAM MUC1 CD34 on UMAP, along with minor cell type annotation. Set ncols=4 and save.
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.
5. Show subset population bar plot for macrophages and save.
6. Show box plot of statistically significant differences in macrophage subset populations between conditions, and save. Determine 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. Select genes related to immune checkpoint pathway and cell cycle pathway. Show cell-cell interactions for these genes and save.
9. Find statistically significant differences in cell-cell interactions between conditions for major immune and stromal cells and show them as a dot plot, then save. Set max_n_items_per_group = 25.
10. Extract condition-specific markers for Macrophage and show as a dot plot, then save. Include only surfaceome markers, up to 50 per condition.
11. Select genes related to Cell cycle pathway. For Beta cell, Alpha cell, Ductal cell, Macrophage, find statistically significant differences in expression between conditions and show them as box plots, then save. Set max_n_items_to_plot = 24, and ncols appropriately so that the aspect ratio is approximately 2x3.
12. Show Gene ontology (GSA) analysis results for Ductal cell, Acinar cell, Beta cell as bar plots and save.
13. Show Gene set enrichment analysis results for Beta cell, Alpha cell, Ductal cell, Macrophage as a dot plot and save. Use RdBu_r colormap and set n_pws_to_show = 80.