T2D Research

Zhao et al. 2025 Single-Cell PBMCs in Type 2 Diabetes

Properties2
tagsreference, type-2-diabetes, pbmc, immunology, transcriptomics
aliasesZhao et al. 2025 Single-Cell PBMCs in Type 2 Diabetes

5 min read

Zhao et al. 2025 Single-Cell PBMCs in Type 2 Diabetes

Zhao and Fang used single-cell PBMC profiling to compare a very small Chinese T2D cohort with healthy controls and reported T-cell and monocyte transcriptional changes involving TNF/NF-kB, interferon-gamma, T-cell receptor, and chemokine signaling.

Citation

  • Full citation: Zhao J, Fang Z. Single-cell RNA sequencing reveals the dysfunctional characteristics of PBMCs in patients with type 2 diabetes mellitus. Frontiers in Immunology. 2025;15:1501660. 1
  • DOI: 10.3389/fimmu.2024.1501660. 1
  • Publication date: 2025-01-23. 1
  • Study type: prospective controlled clinical trial registered as ChiCTR2100049613. 1
  • Public data: sequencing data deposited under GSE255566. 1

Research Question

Zhao and Fang asked how immune cells function in peripheral blood from T2D patients and whether PBMC scRNA-seq could identify immune mechanisms or candidate biomarkers beyond glycemic clinical indicators. 1

Cohort And Ancestry

  • The study included 3 healthy participants and 3 T2D patients from The First Affiliated Hospital of Anhui University of Chinese Medicine in China. 1
  • T2D inclusion criteria included HbA1c <= 7.5%, disease duration <= 3 years, age 18-70 years, and no insulin therapy requirement. 1
  • Healthy and T2DM groups differed significantly in BMI, fasting blood glucose, and HbA1c, while age, sex, and disease course were not significantly different. 1
  • The paper does not report genetic ancestry inference or ancestry-stratified immune analysis, so it should support general PBMC/T2D immune context rather than direct ancestry claims.

PBMC Assay

  • PBMCs were isolated from whole blood by Ficoll density gradient centrifugation, and 10x Genomics Chromium 3-prime scRNA-seq was performed by BGI Shenzhen. 1
  • Cell Ranger v5.0.1 generated raw expression matrices, and Seurat v3.2.0 was used for downstream filtering, clustering, UMAP visualization, marker detection, and differential expression. 1
  • Marker genes per cluster were identified with FindAllMarkers (logFC > 0.25, minPct > 0.1, Padj ≤ 0.05). DEGs between T2DM and healthy groups were identified with FindMarkers using the same thresholds. 1 The paper does not specify which statistical test Seurat was configured to use; the Seurat v3.2.0 default is the Wilcoxon rank-sum test. 2
  • Volcano plots used relaxed thresholds (logFC > 0.2, p < 0.05) for display. 1
  • The analysis retained 13,591 cells and identified 20,092 genes across 6 samples. 1
  • Four broad PBMC cell types were annotated: B cells, T cells, monocytes, and NK cells. 1

Findings Relevant To The Paper

Differential Expression

  • At the whole-PBMC level, 3188 marker genes were identified across groups, with 11 genes associated with T2DM status (PKM, MAPK1, MAPK3, PIK3R1, HK1, HK3, INSR, PIK3CD, SOCS1, IRS2, TNF). Enriched pathways included NF-kB, HIF-1, and TNF signaling; GSEA emphasized TNF signaling. 1
  • T cells had 119 DEGs (58 up, 61 down) in T2DM versus healthy controls. GO enrichment included leukocyte chemotaxis, T-cell receptor signaling, immune response-regulating signaling, and TNFA signaling via NF-kB. 1
  • Monocytes had 175 DEGs (51 up, 124 down). GO enrichment included cytokine activity and binding, S100 protein binding, and HALLMARK_INTERFERON_GAMMA_RESPONSE. 1
  • T-cell WGCNA: the brown module (core gene TNFRSF1A) was associated with FBG (fasting blood glucose), FINS (fasting insulin), and HOMA-IR; CLEC2B, B2M, and MALAT1 were shared between the brown module and the baseline DEG list. 1
  • Monocyte WGCNA: the red module was associated with FBG and enriched for the chemokine signaling pathway. 1
  • HLA-DQA2 and CXCL8 were common DEGs between monocytes and T cells. 1
  • The study interprets T cells and monocytes as central PBMC immune compartments in T2D, but the sample size is too small for strong generalizable directionality claims.

Differential Abundance

  • Four broad cell types were annotated (B cells, T cells, monocytes, NK cells); cell-type percentages were reported in Figure 3C but not as numerical values in the text. 1
  • T-cell reclustering revealed two CD4+ clusters (3127 cells) and three CD8+ clusters (3678 cells). CD4+ T cells in T2DM patients tended toward memory and naive states; CD8+ T cells tended toward effector and memory states. 1
  • Monocyte reclustering identified dendritic cells, CD14+ monocytes, and FCGR3A+ monocytes. 1
  • T cells from T2DM participants were more likely to be in G1, S, and G2M cell-cycle phases than T cells from healthy participants, interpreted as active proliferation. 1
  • Monocytes from T2DM participants also showed elevated G1/S/G2M fractions relative to healthy controls. 1

Clinical Correlations (blood glucose, insulin, HbA1c)

  • T-cell DEGs: RPL27, TXN1P, and RPL37 expression negatively correlated with HbA1c; MNDA negatively correlated with FBG; DDX5 positively correlated with FBG; GIMAP7 positively correlated with HOMA-IR. 1
  • Monocyte DEGs: CLEC7A, SIGLEC14, and AC018755.4 negatively correlated with HbA1c; VSTM1 negatively correlated with FINS and HOMA-IR. 1
  • The monocyte red WGCNA module was negatively correlated with FBG. 1

Ancestry-Relevant Interpretation

  • Zhao and Fang provides additional single-cell evidence that PBMC immune-cell states in T2D involve inflammatory signaling pathways, strengthening the background rationale for PBMC-focused immune analysis.
  • The source does not test whether PBMC immune features vary by ancestry, does not genotype participants, and uses a single-center Chinese cohort, so it cannot directly support ancestry-associated PBMC claims.
  • For the manuscript, this paper is most useful as a small-cohort external PBMC scRNA-seq comparator for pathway-level themes such as TNF/NF-kB, interferon-gamma, T-cell receptor, and chemokine signaling.

Limitations

  • The authors state that the sample size is relatively small and that bias may arise from disease severity, large age differences, and the single-center design. 1
  • The study calls for multicenter validation with larger sample sizes and correlation analysis with pancreatic samples from T2D patients. 1
  • The source is underpowered for stable cell-proportion estimates and should not be weighted the same as larger single-cell studies.

Sources

  • _raw/ingested/Zhao2025_singlecell.pdf

Footnotes

  1. extracted; from Zhao 2025 ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8 ↩9 ↩10 ↩11 ↩12 ↩13 ↩14 ↩15 ↩16 ↩17 ↩18 ↩19 ↩20 ↩21 ↩22 ↩23 ↩24 ↩25 ↩26 ↩27 ↩28 ↩29 ↩30 ↩31

  2. inferred; Seurat v3.2.0 default test is Wilcoxon rank-sum; the paper does not state which test was configured ↩

Graph View

Backlinks