CITE-Seq Hits Vascular Medicine

Atherosclerosis is the most common underlying pathology of coronary artery disease, peripheral artery disease, and cerebrovascular disease. Recent approaches including clinical interventional studies, in vivo imaging, lineage tracking, knockout mouse studies, and advanced mRNA sequencing techniques have drawn more attention to the critical role of immune cells, including T cells and macrophages, in the development of atherosclerosis. Previously, the immune cell composition of tissues was determined by flow cytometry and immunohistochemistry. The more recent development of massively parallel single-cell RNA sequencing (scRNA-seq) and mass cytometry (CyTOF) demonstrated the methodologic shortcomings of classical flow cytometry and gave us new insights into immune cell subsets in a variety of diseases including atherosclerosis. As suggested by its name, scRNA-seq yields single-cell transcriptomes, which has uncovered the heterogeneity of immune cell populations. Recent studies utilizing scRNA-seq have given us a better understanding of aortic cells and particularly leukocytes (1–3). Another method for inferring human plaque immune composition is by using deconvolution methods on bulk RNA-seq data from atherosclerotic tissue and CyTOF analysis of an individual plaque. However, conventional bulk RNA-seq approaches are limited, because they average gene expression patterns across whole tissues. On the other hand, single-cell technologies allow precise measurements of individual cell phenotypical and functional variations that have the potential to improve the understanding of human atherosclerotic vascular disease. In CyTOF, single-cell suspensions are stained with an antibody panel to detect cellular antigens. Unlike flow cytometry, antibodies for CyTOF are conjugated with rare earth metal isotopes, which allows for the detection of up to 40 antigens. CyTOF allows for the simultaneous detection of extracellular, intracellular, and phosphorylated antigens. Recent studies have also used CyTOF to study immune cell composition in atherosclerosis (1, 4).

Recently, Fernandez et al. (5) used an even more advanced approach to study T cell heterogeneity in human atherosclerosis. Leukocytes from human plaques obtained by endarterectomy and peripheral blood mononuclear cells (PBMCs) were labeled by barcoding and analyzed by Cellular Indexing of Transcriptional Epitope Sequencing (CITE-Seq), which combines single-cell transcriptomics with oligonucleotide barcoded antibodies and thus integrates protein with mRNA information (5). This study reported data from 23 human samples stratified as asymptomatic (n = 14) and symptomatic (n = 9). Symptomatic patients were defined as having had a cerebral ischemic event. No healthy donor PBMCs were analyzed, which can be one of the limitations of this study. CITE-Seq was validated by applying CyTOF in the same samples. CyTOF added an additional dimension, because CyTOF, but not CITE-Seq, allows for the detection of intracellular antigens.

Fernandez et al. reported that T cells accounted for the majority of immune cells, especially in plaques obtained from endarterectomy. Moreover, CD8⁺ T cells were enriched and more abundant than CD4⁺ T cells in plaque compared to blood. Unique differences and highly specialized functions of plaque T cells were also identified. T cells were not only prominent in plaques, but they were more activated, differentiated, and exhausted compared to their blood counterparts; whereas T cells from the blood expressed genes associated with cytokine inhibition, RNA synthesis, and metabolic reprogramming of circulating T cells. Plaque T cells expressed high levels of programed cell death protein-1 (PD-1), a marker of T cell exhaustion that is upregulated on T cell activation. This was confirmed at the transcriptional level by an exhaustion gene expression signature similar to that of exhausted T cells in the tumor microenvironment. The coexistence of exhausted and activated T cell subsets within the same plaque suggested that highly activated T cells may initiate an exhaustion reprogramming (the progressive loss of T cell function), possibly sustained by chronic plaque inflammation. Because symptomatic plaques are defined as lesions following acute cerebrovascular (CV) events in this paper, the authors mentioned that it cannot be excluded that T cell exhaustion is initiated by T cell overactivation triggered by plaque disruption. Interestingly, single-cell transcriptomics highlighted distinctive alterations of both plaque CD4⁺ and CD8⁺ T cells associated with recent CV events. In symptomatic patients, both T cell lineages presented gene expression signatures largely consistent with activation, differentiation and exhaustion. Conversely, T cells were mostly activated in plaques of asymptomatic patients.

Fernandez et al. also performed T cell receptor (TCR) sequencing. An independent meta-cluster (MC) analysis on plaque T cells identified an MC of CD69⁺CCR5⁺PD1^intCD127⁻CD8⁺ T cells whose frequencies correlated with TCR clonality in plaques, suggesting that antigen-specific T cell clones may actively expand in the plaque. These observations support the notion that specific antigens drive the autoimmune response in atherosclerosis. Among the candidates that may serve as T cell-activating antigens, low density lipoprotein (LDL) and its core protein apolipoprotein B (ApoB) show the strongest clinical correlation with atherosclerosis in humans. However, almost all published studies, including this paper on CD8⁺ and CD4⁺ T cells in atherosclerosis, do not consider the antigen specificity of these T cells. This is a limitation, as most T cell functions are dependent on antigen engagement. However, it is technically difficult to determine the antigen specificity of T cells in any disease including atherosclerosis.

It is important to note that scRNA-seq, CyTOF, and CITE-seq require enzymatic tissue digestion and mechanical dissociation to produce a single-cell suspension, which have their disadvantages. Specifically, the enzymes used for tissue digestion may change the expression of cell surface molecules by proteolytic cleavage, the single-cell transcriptomes can be altered by cellular processes during digestion and isolation, and different cell types can be affected more or less by the isolation procedures, which skew the proportion of cell types. T cells, which are relatively small, round and mechanically robust, usually survive better than other immune cells, including macrophages or dendritic cells, which are large, branched and fragile. Moreover, to create robust assays, these kinds of methods require proper quality controls at the pre- and postsequencing levels as well as batch correction. In this paper, CyTOF data and surface marker expressions in CITE-seq were analyzed using Cytobank, one of the many existing tools to analyze cell surface markers from single-cell data. Single-cell data are commonly affected by technical artifacts known as “doublets,” which refers to 2 cells that were encapsulated in the same droplet. Doublets between PBMCs and plaques, which mean to be cells having both PBMC and plaque barcodes, were removed. However, this method cannot detect PBMC–PBMC or plaque–plaque doublets. Therefore, these doublets remain in the data set, possibly yielding confusing transcriptomes.

Cells were clustered using the state-of-the-art Louvain algorithm based on expression of 21 investigator-selected markers. Immune cell types were determined based on the expression of canonical markers and were thus gated directly from the viSNE plot. For CITE-seq, the Cell Ranger (from 10x Genomics) analysis pipeline was used, which processes scRNA-seq output to align reads, generates feature-barcode matrices and performs clustering and gene expression analysis. The output contained transcriptomes of 3573 cells with 50 701 mean reads per cell. For the scRNA-seq, a total of 7169 cells were analyzed and the outputs had an average of 111 670 mean reads per cell.

Single-cell data can be affected by batch effects, which refers to differences between data sets generated at different times or by different technicians. Other factors such as reagent quality can also contribute to batch effects. Batch correction of scRNA-seq data in this paper was performed using mutual nearest neighbors, a recently used method.

In summary, Fernandez et al. reported the first comprehensive paper about immune cell heterogeneity, especially T cells, involved in human plaques as well as PBMCs using CITE-seq, a new methodology. Previous knowledge of T cell function in atherosclerosis relied on findings from experimental models. This paper gives new insights into the leukocyte infiltrate in human atherosclerotic plaque. The results are relevant to medicine, because targeting immune cells and their specific cytokines can be an attractive pathway to developing new preventative and therapeutic approaches.

Nonstandard Abbreviations

scRNA-seq, single-cell RNA sequencing; PBMCs, peripheral blood mononuclear cells; CITE-Seq, Cellular Indexing of Transcriptional Epitope Sequencing; PD-1, Programmed cell death 1; CV, cerebrovascular; TCR, T cell receptor; MC, meta cluster; CCR5, C-C motif chemokine receptor 5; LDL, low density lipoprotein; ApoB, apolipoprotein B

Author Contributions

All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.

Authors’ Disclosures or Potential Conflicts of Interest

Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:

Employment or Leadership

None declared.

Consultant or Advisory Role

None declared.

Stock Ownership

None declared.

Honoraria

None declared.

Research Funding

R. Saigusa, a postdoctoral fellowship from Japan Society for the Promotion of Science; K. Ley, National Institutes of Health, HL136275, HL140976, HL145241, HL148094.

Expert Testimony

None declared.

Patents

None declared.