Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Sep 29;142(13):1279-1293.
doi: 10.1161/CIRCULATIONAHA.119.042863. Epub 2020 Jul 24.

Pathogenic Autoimmunity in Atherosclerosis Evolves From Initially Protective Apolipoprotein B100-Reactive CD4+ T-Regulatory Cells

Affiliations

Pathogenic Autoimmunity in Atherosclerosis Evolves From Initially Protective Apolipoprotein B100-Reactive CD4+ T-Regulatory Cells

Dennis Wolf et al. Circulation. .

Abstract

Background: Throughout the inflammatory response that accompanies atherosclerosis, autoreactive CD4+ T-helper cells accumulate in the atherosclerotic plaque. Apolipoprotein B100 (apoB), the core protein of low-density lipoprotein, is an autoantigen that drives the generation of pathogenic T-helper type 1 (TH1) cells with proinflammatory cytokine secretion. Clinical data suggest the existence of apoB-specific CD4+ T cells with an atheroprotective, regulatory T cell (Treg) phenotype in healthy individuals. Yet, the function of apoB-reactive Tregs and their relationship with pathogenic TH1 cells remain unknown.

Methods: To interrogate the function of autoreactive CD4+ T cells in atherosclerosis, we used a novel tetramer of major histocompatibility complex II to track T cells reactive to the mouse self-peptide apo B978-993 (apoB+) at the single-cell level.

Results: We found that apoB+ T cells build an oligoclonal population in lymph nodes of healthy mice that exhibit a Treg-like transcriptome, although only 21% of all apoB+ T cells expressed the Treg transcription factor FoxP3 (Forkhead Box P3) protein as detected by flow cytometry. In single-cell RNA sequencing, apoB+ T cells formed several clusters with mixed TH signatures that suggested overlapping multilineage phenotypes with pro- and anti-inflammatory transcripts of TH1, T helper cell type 2 (TH2), and T helper cell type 17 (TH17), and of follicular-helper T cells. ApoB+ T cells were increased in mice and humans with atherosclerosis and progressively converted into pathogenic TH1/TH17-like cells with proinflammatory properties and only a residual Treg transcriptome. Plaque T cells that expanded during progression of atherosclerosis consistently showed a mixed TH1/TH17 phenotype in single-cell RNA sequencing. In addition, we observed a loss of FoxP3 in a fraction of apoB+ Tregs in lineage tracing of hyperlipidemic Apoe-/- mice. In adoptive transfer experiments, converting apoB+ Tregs failed to protect from atherosclerosis.

Conclusions: Our results demonstrate an unexpected mixed phenotype of apoB-reactive autoimmune T cells in atherosclerosis and suggest an initially protective autoimmune response against apoB with a progressive derangement in clinical disease. These findings identify apoB autoreactive Tregs as a novel cellular target in atherosclerosis.

Keywords: T-lymphocytes; T-lymphocytes, regulatory; apolipoprotein B-100; atherosclerosis; autoimmunity.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
A population of oligoclonal apoB-reactive CD4+ T cells (apoB+) resides in lymph nodes of healthy C57Bl/6 mice. A, MHC-II peptide binding map of mouse apoB. The binding affinity of apoB peptides spanning the whole apoB sequence was predicted in silico and expressed as percentile relative to all other predicted values (gray, left y axis). Measured binding affinity of peptides (right y axis) in a competitive binding assay is shown in white. Peptides with proven relevance in the Apoe–/– mouse model of atherosclerosis are shown in green. B, Design of the apo B978-993:MHC tetramer. APC and PE were used in separate batches of tetramers. C, Confocal microscopy of mouse lymph node CD4+ T cells after incubation with PE- and APC-labeled apoB:MHC tetramers and anti-TCR-β-FITC. D, Gating strategy in flow cytometry to identify apoB+ CD4+ T cells with a positive signal for apoB:MHC tetramers in both colors (PE and APC). Percentage of apoB+ cells among all CD4+ T cells indicated in the graph (red). CD4+ T cells negative for the apoB:MHC tetramer were termed apoBneg. E, Eight-week-old wild-type C57BL6/J or MHC-II–mismatched BALBc mice immunized with mouse apo B978-993 in adjuvant CFA subcutaneously. The adjuvant CFA alone (adjuvant) or saline (none) served as controls (n=4–8 mice per group). F, Expression of the TCR-signaling intermediate Nur77 in all or CD44+ apoBneg and apoB+ cells from Nur77-GFP reporter mice in response to vaccination with apo B978-993 (n=3 mice per group). G, Parallel DNA sequencing of the TCR β-chain. Percentage of the 10 most abundant unique TCR-clonotypes (coded in colors) within all apoBneg and apoB+ cells from 3 pooled mice per group. H, Fractional TCRβ V-chain usage among all apoBneg and apoB+ cells.; All bars indicate mean values per group. **P<0.01, ****P<0.001 by ANOVA with Dunnett multiple comparison test (E) and an unpaired, 2-sided t test (F). Representative pictures shown in C and D. apoB indicates apolipoprotein B; APC, allophycocyanin; CFA, complete Freund's adjuvant; FITC, fluorescein isothiocyanate; FSC, forward scatter; GFP, green fluorescent protein; IDL, intermediate-density lipoprotein; L/D, live/dead viability stain; LDL, low-density lipoprotein; LDLR, low-density lipoprotein receptor; Lin., lineage-defining antibodies against CD19/B220/CD11b/CD11c/Nk1.1/TER-119/CD8; MFI, mean fluorescent intensity; MHC-II, major histocompatibility complex II; PE, phycoerythrin; SSC, side scatter; TCR, T-cell receptor; and VLDL, very low density lipoprotein.
Figure 2.
Figure 2.
Lymph node apoB-reactive T cells acquire a proliferating and memory phenotype in atherosclerosis-prone Apoe–/– mice. A, The fraction of antigen-unexperienced Tnaive (CD44 CD62L+, white), antigen-experienced T-effector memory (TEM, CD44+CD62L, black), and T central memory (TCM, CD44+CD62L+, gray) cells among all antigen-specific T cells of the specificity shown (apoB+ or MCMV-specific M25+) was quantified in flow cytometry and expressed as percent of all cells. T cells were isolated from 8-week-old wild-type (WT) or Apoe/–mice (n=5 mice per group). M25+ indicates T cells binding a MCMV:MHC-II tetramer. If indicated, Apoe–/– mice were infected with MCMV to induce antigen-specific T cells (n=5 mice). B, Number of apoB+ T cells in pooled lymph nodes of 4- to 8-week-old WT and atherosclerosis-prone Apoe–/– mice on a chow diet (CD). Numbers expressed as total numbers per mouse (n=10 mice per group). C, Four-week-old, female Apoe–/– mice consumed a Western diet (WD) for 4 weeks. BrDU was incorporated in the drinking water for an additional 9 days (n=3–8 mice per group). D, Proliferating CD4+ T cells (% all CD4+ T cells or CD44+ CD62L T cells) were identified by binding of an anti-BrDU FITC antibody in flow cytometry. Representative gating for all cells is shown in D. E, Quantification of proliferating T cells (BrdU+) within indicated populations. F, Quantification of the mean fluorescence intensity of a Ki-67-BV786 antibody. All bars indicate mean values per group. **P<0.01, ***P<0.001 by an unpaired, 2-sided t test (B, E, and F). BrdU indicates bromodeoxyuridine; FITC, fluorescein isothiocyanate; MCMV, murine cytomegalovirus; MFI, mean fluorescent intensity; MHC-II, major histocompatibility complex II; and WT, wild-type.
Figure 3.
Figure 3.
ApoB-reactive CD4+ T cells simultaneously express cellular markers and transcripts of regulatory and TH1, TH17, and TFH cells. A and B, Expression of the TH-lineage–defining transcription factors FoxP3, RORγT, GATA3, T-bet, and Bcl-6 (TFH) by flow cytometry. Lymph node CD4+ T cells from 8-week-old Apoe–/– mice (n=5 per group) were stained with apoB:MHC. C, Expression of TFs shown as frequency distribution with RORγT in the inner and FoxP3 in the outer circle. D, Gene-set enrichment of a Treg gene signature in apoB+ versus apoBneg non-Treg genes. Gene expression was quantified in RNA sequencing of sorted CD4+ T cells from pooled lymph nodes of 3 replicates of 8-week-old female Apoe–/– mice on a chow diet (each pooled from lymph nodes of 10 mice): apoBnegCD25+FR4+ (apoBneg Treg), apoBneg non-Treg (apoBnegCD25-FR4-), and apoB+ (10 mice pooled per replicate). Gene expression is displayed as log10(RPKM). E, Gene expression of TH-defining genes displayed in a heat map as log10(RPKM) and column min.-max. expression. F, Cytokine expression (% cytokine+ among all T cells) in intracellular flow cytometry from 3 to 5 mice per group. G, scRNAseq of sorted apoBneg and apoB+ cells from 3 individual mice. Single-cell transcriptomes were analyzed with a dimensionality reduction algorithm to identify groups of cells with similar gene expression. H, Average cluster gene expression (transcripts/cell) is displayed as heat map with column min.-max. expression (transcripts per cluster). All bars indicate mean values per group. *P<0.05, ***P<0.001, ****P<0.001 by an unpaired, 2-sided t test. Representative pictures are shown in B and D. 10x Gen. indicates 10x Genomics; apoB, apolipoprotein B; ES, enrichment score; FACS, fluorescence-activated cell sorting; FoxP3, Forkhead Box P3; IL-10, interleukin-10; IL-17, interleukin-17; IFN-γ, interferon γ; MHC, major histocompatibility complex; min.-max., minimum-maximum; Mix., clusters that are equally populated by apoBneg and apoB+ T cells; RORγT, RAR-related orphan receptor γ; RPKM, reads per kilobase per million; scRNAseq, single-cell RNA sequencing; Seq., sequencing; T-bet, T-box expressed in T cells; TFs, transcription factors; TH, T-helper cell type; Treg, regulatory T cell; and tSNE, t-distributed stochastic neighbor embedding.
Figure 4.
Figure 4.
ApoB-reactive CD4+ T-regulatory cells acquire a proinflammatory TH1-TH17 phenotype in the course of atherosclerosis. A, Expression of the TH-lineage–defining transcription factors (TFs) T-bet, FoxP3, RORγT among apoB+ cells by intracellular flow cytometry (%) in five 8- and 20-week-old female Apoe–/– mice consuming chow diet (CD). All CD4+ T cells were gated as parent population. B, Changes in the expression of pathways assigned to transcriptional master regulators in 4- and 20-week-old female Apoe–/– mice on CD determined in bulk RNA sequencing of apoB+ cells and ingenuity pathways analysis (IPA) from 2 to 3 replicates with cells from 10 pooled mice each. A positive activation z-score indicated an upregulation of the pathway in 20-week-old animals. C, Gene set enrichment analysis (GSEA) of TH-gene signatures on apoB+ T cells from 20- versus 4-week-old mice. An false discovery rate of <0.2 is considered significant. D, Differentially expressed genes in apoB+ cells displayed in a heat map with minimum and maximum values in a heat map. Values were log10 normalized. E, T cells from Foxp3-YFP-Cre/ROSA26-RFP/Apoe–/–lineage tracker mice express a FoxP3-YFP-fusion protein exclusively in current Tregs, whereas the RFP transgene is permanently expressed in cells in which FoxP3 was ever expressed. RFP+YFPT cells were termed exTregs. F, Percentage of exTregs and current Tregs in aortic CD4+ T cells. G and H, Percentage of exTregs among apoB+ and apoBneg cells (%) in lymph nodes of female Apoe–/– Treg lineage tracer mice after a CD or Western diet for 20 weeks (n=3–8 mice per group). All bars indicate mean values per group. *P<0.05, **P<0.01, ***P<0.001, ****P<0.001 by an unpaired, 2-sided t test. Representative pictures are shown in H. apoB indicates apolipoprotein B; RFP, red fluorescent protein; RORγT, RAR-related orphan receptor γ; RPKM, reads per kilobase million; TH, T-helper type; Treg, regulatory T cell; and YFP, yellow fluorescent protein.
Figure 5.
Figure 5.
Aortic CD4+ T cells have mixed TH17/TH1/Treg transcriptomes in scRNAseq. A, Single-cell RNA-sequencing (scRNAseq) of aortic CD4+ T cells after cell isolation of aortic leukocytes from Apoe–/– mice (n=10) after a control chow diet (CD) or a Western diet (WD) for 12 weeks, flow sorting, drop sequencing, and filtering of Cd3e+Cd4+Cd8a- cellular events. B, Single-cell transcriptomes were analyzed with an unsupervised dimensionality reduction algorithm (tSNE) to identify groups of cells with similar gene expression. The 4 resulting clusters are shown in the Left, the origin (WD/CD) is overlaid on the tSNE plot (Right). C, Differentially expressed genes in the 4 clusters in a single-cell heat map. D, Fraction of the 4 found clusters among all aortic CD4+ T cells. E, Expression of TH-lineage–defining genes in the 4 clusters (average transcripts/cell) displayed as heat map with column min.-max. expression. F, Color-coded enrichment scores indicating the similarity of each single-cell transcriptome with genes overexpressed in apoB+ T cells was overlaid on the tSNE projection of aortic T cells. G, Cytokine secretion in explanted atherosclerotic aortas after stimulation with the apo B978-993 (p6) peptide or a control with the carrier dimethyl sulfoxide alone (n=6–9 per group). Cytokines were quantified in the supernatant with a cytometric bead array. Cytokine secretion shown as % of the unstimulated control. All bars indicate mean values per group. *P<0.05. apoB indicates apolipoprotein B; DE, differentially expressed; FACS, fluorescence-activated cell sorting; IL-5, interleukin-5; IL-10, interleukin-10; IL-17, interleukin-17; max., maximum; min., minimum; TH, T-helper type; TNF-α, tumor necrosis factor α; Treg, regulatory T cell; and tSNE, t-distributed stochastic neighbor embedding.
Figure 6.
Figure 6.
Transferred apoB+ Tregs fail to protect from atherosclerosis. A, Transfer of 2×105 CD45.1+ apoB+ or apoBneg Tregs, or no cells into 8-week-old, female CD45.2+ Apoe–/– mice (n=5–6 per group). B, After 12 weeks, lymph nodes were harvested and fractions of Tregs (CD25+FR4+) and Tconv (conventional T cells; remaining cells) were quantified among all CD45.1+ T cells in lymph nodes of CD45.2+ donor mice. C, Atherosclerotic lesions in the thoracic/abdominal aorta were visualized by a Sudan-IV stain (D) and quantified as area (% aortic area) or the number of distinct atherosclerotic lesions in the thoracic/suprarenal aorta per mouse (E). All bars indicate mean values per group. Statistical significance was tested by a 1-way ANOVA with the Holm-Sidak multiple comparison test (D and E). *P<0.05. Representative pictures shown in C. apoB indicates apolipoprotein B; and Treg, regulatory T cell.
Figure 7.
Figure 7.
Coronary atherosclerosis in humans is accompanied by a population of circulating apoB-reactive T cells that secrete TH1 and TH17 cytokines. A, Human apoB was screened in silico for peptide sequences with a high binding affinity to human MHC-II variants. The affinity of the 30 top-binding self-peptides was confirmed in a competitive MHC-II binding assay against radiolabeled cognate ligands. Heat map of the IC50 of peptides to MHC-II alleles. Band C, ApoB-reactive CD4+ T cells (CD3+CD4+CD8-L/D) were identified by the intracellular expression of CD40L in flow cytometry. Human whole-blood samples (Dand E) or density gradient–isolated, IL-2–activated peripheral blood mononuclear cells (F) from individuals without (n=11) or with (n=23) stenotic coronary artery disease (CAD) were coincubated with the pool of human apoB-peptides (apoB), the carrier dimethyl sulfoxide alone (control), or the positive control staphylococcal enterotoxin B (SEB) for 6 hours (Dand E) or 10 days (F). Cytokine expression in human T cells was expressed as % cytokine+ CD4+ T cells. All bars indicate mean values per group. Statistical significance was calculated by a Mann-Whitney U test (D through F). *P<0.05. apoB indicates apolipoprotein B; FSC, forward scatter; IFN-γ, interferon γ; IL-2, interleukin-2; IL-10, interleukin-10; IL-17, interleukin-17; L/D, live/dead viability stain; SSC, side scatter; and TNF-α, tumor necrosis factor α.

References

    1. Wolf D, Ley K. Immunity and inflammation in atherosclerosis. Circ Res. 2019;124:315–327. doi: 10.1161/CIRCRESAHA.118.313591 - PMC - PubMed
    1. Kobiyama K, Ley K. Atherosclerosis. Circ Res. 2018;123:1118–1120. doi: 10.1161/CIRCRESAHA.118.313816 - PMC - PubMed
    1. Libby P, Lichtman AH, Hansson GK. Immune effector mechanisms implicated in atherosclerosis: from mice to humans. Immunity. 2013;38:1092–1104. doi: 10.1016/j.immuni.2013.06.009 - PMC - PubMed
    1. Jonasson L, Holm J, Skalli O, Bondjers G, Hansson GK. Regional accumulations of T cells, macrophages, and smooth muscle cells in the human atherosclerotic plaque. Arteriosclerosis. 1986;6:131–138. doi: 10.1161/01.atv.6.2.131 - PubMed
    1. Galkina E, Kadl A, Sanders J, Varughese D, Sarembock IJ, Ley K. Lymphocyte recruitment into the aortic wall before and during development of atherosclerosis is partially L-selectin dependent. J Exp Med. 2006;203:1273–1282. doi: 10.1084/jem.20052205 - PMC - PubMed

Publication types