Poor Collateralization in T2DM: Role of SLC4A10+ MAIT Cells

Abstract

Background: Type 2 diabetes is strongly associated with impaired collateralization, which increases the risk of cardiovascular complications, such as myocardial infarction and heart failure. This study explored the immune cell dynamics in patients with type 2 diabetes with chronic total occlusion and their impact on collateralization.

Methods: Peripheral blood mononuclear cells were extracted from patients with type 2 diabetes with chronic total occlusion, exhibiting either good or poor collateralization. Single-cell RNA sequencing was conducted to profile the quantitative and transcriptomic dynamics of immune cells in these 2 groups. Moreover, coculture experiments were executed, and ischemic models of the hindlimb and myocardium were induced in diabetic mice to corroborate the single-cell RNA sequencing findings. Additional validation was attained by conducting an analysis on a separate cohort of patients.

Results: Single-cell RNA sequencing of peripheral blood mononuclear cells identified elevated levels of mucosal-associated invariant T (MAIT) cells in patients with poor collateralization. In diabetic mice, inhibition of MAIT cell activation significantly improved angiogenesis under ischemic conditions. In vitro, MAIT cell-derived CCL3L1 (C-C motif chemokine ligand 3-like 1) drove macrophage polarization toward a proinflammatory phenotype through CCR5 (C-C chemokine receptor type 5) interaction. Furthermore, an independent patient cohort confirmed that elevated MAIT cell levels represent an independent risk factor for poor collateralization.

Conclusions: These findings highlight the critical role of MAIT cells in regulating collateralization in type 2 diabetes chronic total occlusion patients and propose circulating MAIT cell levels as a potential biomarker for predicting and intervening in poor collateralization.

Keywords: cardiovascular diseases; coronary artery disease; diabetes mellitus, type 2; heart failure; myocardial infarction.

Figure 1.

Single-cell landscape of peripheral blood cells from patients with type 2 diabetes (T2DM) with chronic total occlusion (CTO). A, Schematic overview of the design of this study. B, Uniform Manifold Approximation and Projection (UMAP) visualization of all identified cell clusters. Each dot represents a cell colored according to cell type. C, UMAP visualization of all identified cell clusters. Each dot represents a cell colored according to individual samples. D, UMAP visualization of all identified cell clusters. Each dot represents a cell colored according to sample group. E, Dot plots showing scaled expression of marker genes for different cell clusters. F, Gene Ontology enrichment based on differentially expressed marker genes determined the biological function of different cell lineages. G, Proportion of different cell clusters in patients with good collateralization (GC) and poor collateralization (PC; n=6 for each group). H, Proportion of differentially expressed genes (DEGs; GC vs PC; n=6 for each group) in different cell cluster. See also Figures S1 and S2; Table S1. BC indicates poor collateralization; DC, dendritic cells; FCM, flow cytometry; GC, good collateralization; NA, not significant; NK, natural killer cell; PBMC, peripheral blood mononuclear cells; PC, poor collateralization; and scRNA-seq, single-cell RNA sequencing.

Figure 2.

Single-cell RNA sequencing (scRNA-seq) revealed monocyte heterogeneity of peripheral blood cells from patients with type 2 diabetes (T2DM) with chronic total occlusion (CTO). A, Uniform Manifold Approximation and Projection (UMAP) visualization of monocyte subsets. Each dot represents a cell colored according to cell type. B, Violin plots showing scaled expression of marker genes for different monocyte subsets. C, UMAP visualization of all identified monocyte subsets. Each dot represents a cell colored according to individual samples (top) and sample group (bottom). D, Heat map showing gene set variation analysis (GSVA) analysis of different monocyte subsets. E, Bar plots showing gene set variation analysis explore the pathway enrichment of different monocyte subsets in patients with good collateralization (GC) and poor collateralization (PC). F, Proportion of different monocyte subsets in patients with GC and PC. G and H, Flow cytometry analysis of CD14_monocytes, CD14_CD16_monocytes, CD16_monocytes among CD45+cells from patients with GC (n=15) and PC (n=15). Data are represented as mean±SD. Statistical significances were determined by unpaired Student t test. See also Figures S3 and S4. IL indicates interleukin; MHC, major histocompatibility complex; NF-κB, nuclear factor-κB; SH2, Src-homology 2; and TNF, tumor necrosis factor.

Figure 3.

Single-cell RNA sequencing (scRNA-seq) revealed the heterogeneity of T cells from patients with type 2 diabetes (T2DM) with chronic total occlusion (CTO). A, Uniform Manifold Approximation and Projection (UMAP) visualization of T-cell subsets (N=18 824). Each dot represents a cell colored according to cell type. B, UMAP visualization of all identified T-cell subsets (N=18824). Each dot represents a cell colored according to individual samples (top) and sample groups (bottom). C, Dot plots showing scaled expression of marker genes for different cell clusters. D, Proportion of different T-cell subsets in patients with good collateralization (GC) and poor collateralization (PC; N=6 for each group). E, Heat map showing the activity of metabolic pathways of different T-cell subsets. F, Heat map showing Single-Cell Regulatory Network Inference and Clustering algorithm to dissect the transcriptional regulatory networks of different T-cell subsets governing coronary collateral formation in patients with T2DM with CTO. G, Disease contribution of distinct T-cell subsets. The radius is proportional to the contribution score. H, Functional enrichment analysis of the upregulated differentially expressed genes (DEGs) of mucosal-associated invariant T (MAIT) cells in patients with PC (N=6). I, Violon plot exhibiting score of cytokine production (top) and cytotoxicity (bottom) in MAIT cells. The score represents Z score of normalized mean expression level of selected gene signatures. Statistical significances were determined by unpaired Student t test. See also Figure S5.

Figure 4.

The effects of inhibiting mucosal-associated invariant T (MAIT) cell activity on blood flow and angiogenesis in diabetic mice with hindlimb or myocardial ischemia. A, Laser Doppler perfusion imaging assessments were performed on days 0 and 21 after femoral artery ligation. Perfusion recovery was measured by quantifying microvascular blood flow and represented as the ratio of flow in the right (ischemic) to the left (nonischemic) hindlimb (n=8 mice/group). B, Representative hematoxylin and eosin (H&E) images of muscles harvested at the indicated time with vehicle group and Ac-6-FP group. Quantification of the regenerating area at different group (n=6 mice/group, white scale bar=50 µm, gray scale bar=20 µm). C, Representative Masson’s trichrome staining images of muscles harvested at the indicated time with vehicle group and Ac-6-FP group. Quantification of the fibrosis area at different group (n=7 mice/group, white scale bar=50 µm, gray scale bar=20 µm). D, The density of capillaries (CD31+) and arterioles (α-SMA [α smooth muscle actin]-SMA+) in the ischemic gastrocnemius muscle were examined by immunofluorescence on day 7 postsurgery (n=6 or 7 mice/group, scale bar=50 µm). E, Echocardiographic analysis of diabetic mice submitted to myocardial ischemia, treated with vehicle and Ac-6-FP at 4 weeks after coronary artery ligation (n=6 mice/group). F, Angiogenesis analysis of the ischemic myocardium in Vehicle or Ac-6-FP–treated mice, as examined by CD31, α-SMA antibody and diisopropanolamine (DAPI; n=6 mice/group, scale bar=50 µm). G, Masson staining of myocardium in Vehicle and Ac-6-FP–treated mice (n=6 mice/group, scale bar=1000 µm). Data are presented as mean±SD. Statistical significance for paired comparisons was performed by Student t test (A, B, C, D, F, and G). Data were analyzed using 2-way ANOVA with Tukey post hoc tests for group comparisons (E). Ac-6-FP indicates acetyl-6-formylpterin; DAPI, 4′,6-diamidino-2-phenylindole; HPF, high power field; LV, left ventricular; LVEF, left ventricular ejection fraction; and LVFS, left ventricular fractional shortening.

Figure 5.

Mucosal-associated invariant T (MAIT) cells skew macrophages to a proinflammatory phenotype. A, Heat map showing summary of the interaction number of T cells and monocyte subsets. B and C, The interaction number of MAIT cells and monocyte subsets in patients with good collateralization (GC) and poor collateralization (PC). D, Volcano plot illustrating the differential genes between macrophage cocultured with MAIT cells, from the peripheral blood of patients with type 2 diabetes (T2DM) with chronic total occlusion (CTO), both with GC and PC. E, Functional enrichment analysis of the differentially expressed genes (DEGs) in macrophage cocultured with MAIT cells from the peripheral blood of patients with T2DM with CTO. F, Gene set variation analysis plot showing significant enrichment of pathways in macrophage cocultured with MAIT cells from the peripheral blood of patients with T2DM with CTO, both with GC and PC. G and H, Frequencies of CD86+ macrophage in the context of cocultured with MAIT cells from patients with T2DM with CTO were analyzed by flow cytometry (n=4 for each group). I and J, Representative images of immunofluorescent staining of iNOS (inducible NO synthase) and quantification of the iNOS mean intensity in macrophage cocultured with MAIT cells from patients with T2DM with CTO (n=4 for each group, white scale bar=50 µm, gray scale bar=10 µm). K, The THP-1–induced macrophages were cocultured with MAIT cells from patients with T2DM with CTO for 24 hours. The mRNA was extracted for testing NOS2 (NO synthase 2), CD86, IFNG (interferon γ), and TNF (tumor necrosis factor; n=4 for each group). L, Flow cytometry analyzed the frequency of CD86-positive macrophages in cocultures of MAIT cells and macrophages, after treatment with Ac-6-FP or a control vehicle for 24 hours (n=4 for each group). M and N, Representative images of immunofluorescent staining of iNOS and quantification of the iNOS mean intensity in cocultures of MAIT cells and macrophages, after treatment with Ac-6-FP (10 μM) or a control vehicle for 24 hours (n=4 for each group, white scale bar=50 µm, gray scale bar=10 µm). O, The THP-1 induced macrophage were cocultured with MAIT cells, after treatment with Ac-6-FP or a control vehicle for 24 hours. The mRNA was extracted for testing NOS2, CD86, IFNG, and TNF (n=4 for each group). P, proinflammatory macrophages marker fluorescent staining of myocardium in vehicle and Ac-6-FP–treated mice (n=6 mice/group). Q, Flow cytometry analyzed the frequency of CD86-positive macrophages in ischemic myocardium in vehicle and Ac-6-FP–treated mice (n=6 mice/group). R, Flow cytometry analyzed the frequency of CD86-positive macrophages in ischemic calf in vehicle and Ac-6-FP–treated mice (n=6 mice/group). Statistical significance for paired comparisons was performed by Student t test. Data are presented as mean±SD. Statistical significances were determined by unpaired Student t test (G, H, J, L, and K). See also Figures S6 and S7 and Table S2. DAPI indicates 4′,6-diamidino-2-phenylindole; FACS, fluorescence-activated cell sorting; GO, Gene Ontology; and NF-κB, nuclear factor-κB.

Figure 6.

Mucosal-associated invariant T (MAIT) cell–derived CCL3L1 (C-C motif chemokine ligand 3-like 1)-induced proinflammatory phenotype of macrophages. A, Volcano plot illustrating differentially expressed genes across distinct cellular subpopulations, with red markers highlighting MAIT cell–specific genes that also function as secreted factors. B, ELISA measurement of CCL3L1 protein levels in plasma from patients with poor collateralization (PC, n=45) and good collateralization (GC, n=59). Data are presented as mean±SD, with statistical significance determined by Student t test. C, Real-time polymerase chain reaction (PCR) analysis of inflammatory marker expression (CD86, TNF [tumor necrosis factor], NOS2 [NO synthase 2], IFNG [interferon γ]) in THP-1–derived macrophages treated with 100 ng/mL rmCCL3L1 for 24 hours. Data are presented as mean±SD, with statistical significance determined by Student t test. D, Immunofluorescence analysis of iNOS (inducible NO synthase) expression in THP-1–derived macrophages treated with rmCCL3L1. Data are presented as mean fluorescence intensity (MFI) ±SD. Statistical significance was determined using Student t test. E, Flow cytometry analysis of CD86 and CD11b protein expression on the surface of THP-1–derived macrophages treated with rmCCL3L1. Statistical analysis was performed on the percentage of cells positive for each marker, with significance determined by Student t test. F, Schematic diagram of a coculture experiment between MAIT cells and THP-1–derived macrophages, conducted in the presence of either 10 µg/mL anti-human CCL3L1 neutralizing antibody (CCL3L1-neutralizing antibody group) or 10 µg/mL isotype control antibody (IgG, isotype control group). G, Immunofluorescence staining of iNOS (Nos2) in THP-1–derived macrophages cocultured with MAIT cells from patients with PC and GC, with or without neutralization of CCL3L1. Representative images are shown, with quantification presented as mean fluorescence intensity (MFI) ±SD, with statistical significance determined by Student t test. H, Flow cytometry analysis of CD86 and CD11b protein expression on the surface of THP-1–derived macrophages cocultured with MAIT cells from patients with PC and GC, with or without neutralization of CCL3L1. Statistical analysis was performed on the percentage of cells positive for each marker, with significance determined by Student t test. I, real-time PCR analysis of inflammatory marker expression (CD86, TNF, Nos2, and IFNG) in THP-1–derived macrophages cocultured with MAIT cells from patients with PC and GC, with or without neutralization of CCL3L1. Data are presented as mean±SD, with statistical significance determined by Student t test. J, Real-time PCR analysis of inflammatory marker expression (CD86, TNF, Nos2, IFNG) in THP-1–derived macrophages treated with rmCCL3L1 and transfected with siRNA targeting CCR3, CCR5, or a nontargeting vehicle. Data are presented as mean±SD, with statistical significance determined by 1-way ANOVA. See also Figure S8. DAPI indicates 4′,6-diamidino-2-phenylindole.

Figure 7.

Increased circulating mucosal-associated invariant T (MAIT) cells related to poor collateralization formation. A and B, Flow cytometry analysis of CD3+CD4+ cells among live cells from patients with good collateralization (GC; n=59) and poor (PC; n=45) collateralization. Statistical significance for paired comparisons was performed by Student t test. C and D, Flow cytometry analysis of CD3+CD8+ cells among live cells from patients with GC and PC. Statistical significance for paired comparisons was performed by Student t test. E and F, Flow cytometry analysis of MAIT cells among CD3+ cells from patients with GC and PC. Statistical significance for paired comparisons was performed by Student t test. G, The proportion of poor collateralization increased stepwise with the increase of that of MAIT cells. H, Restricted cubic splines plot detecting the association between the proportion of MAIT cells and poor collateralization. I, Receiver operating characteristic curve analysis of the proportion of MAIT cells for diagnosing poor collateralization. J, Subgroup analyses of the proportion of MAIT cells for diagnosing poor collateralization. Statistical significance for paired comparisons was performed by Student t test. Data are presented as mean±SD. See also Figures S9 and S10. AUC indicates area under the curve; BMI, body mass index; GFR, glomerular filtration rate; HbA1c, glycated hemoglobin A1c; and OR, odds ratio.