Single-cell transcriptomics as a building block for determining mechanistic insight of abdominal aortic aneurysm formation

This editorial refers to ‘Single-cell RNA sequencing reveals the cellular heterogeneity of aneurysmal infrarenal abdominal aorta’ by G. Zhao et al., pp. 1402–1416.

Abdominal aortic aneurysms (AAAs) have a devastating effect on public health, but there are currently no medical approaches to attenuate progression of the disease. Consequently, there is an urgent medical need to determine molecular mechanisms of AAAs to facilitate development of novel pharmacological approaches. Numerous studies have demonstrated that AAA tissues contain many different cell types that potentially have multiple subtypes of the major categories. Determination of the detailed transcriptomes of these cells has the potential to assist in the definition of disease mechanisms.

Single-cell RNA sequencing (scRNAseq) is a contemporary and powerful technology to identify transcriptomic gene profiles at cellular levels. Comprehensive and unbiased gene profiling by scRNAseq leads to a better understanding of cell-specific molecular characteristics in physiological and pathophysiology conditions. Since the initial report of scRNAseq technique in 2009,¹ the use of this technique has increased dramatically. Improved analytical tools for scRNA data, including Seurat R analysis package,¹ have contributed to a greater application of scRNAseq. The experimental aneurysm field commonly uses three diverse modes of promoting AAA in mice. Previous studies have applied scRNAseq technology to define the cellular heterogeneity of AAA tissue in mice that was generated by AngII infusion² or periaortic exposure to calcium chloride.³ Zhao et al.⁴ performed scRNAseq for the determination of transcriptomic profiles in the cellular components of AAAs induced by elastase in mice to complete the spectrum of the most commonly used mouse models of AAA.⁵

In the study of Zhao et al.,⁴ scRNAseq was performed in control and elastase-induced AAA mouse samples harvested at two intervals (Day 7 and 14 after elastase exposure). To obtain sufficient numbers of cells, five infra-renal aortic tissues were pooled in each group. Consistent with previous reports in other AAA models, scRNAseq demonstrated cellular heterogeneity in both control and aneurysm samples. Aortic tissues were composed of multiple cell types including endothelial cells, smooth muscle cells, fibroblasts, and immune cells. Compared to control aortas, elastase-induced AAAs exhibited decreased numbers of smooth muscle cells and increased immune cells. Sub-clustering analyses of smooth muscle cells identified four sub-clusters in both control and AAA samples. One specific sub-cluster was increased significantly 14 days after elastase application, which was considered to be an advanced stage of AAAs. Gene characterization was performed on the smooth muscle cell sub-clusters. Contractile genes, such as Myh11 and Tagln, were decreased in smooth muscle cells of AAAs, while mRNA of several cytokines were increased, including Cxcl2 and Ccl2. Interestingly, the smooth muscle cell sub-cluster which was highly populated 14 days after initiating the disease had a high abundance of Thbs1 and Notch3 that exert a critical role in AAA formation.^6–9 In addition to smooth muscle cells, a unique sub-cluster was also observed in macrophages. Of note, interleukin 10, an anti-inflammatory cytokine, was highly abundant in this unique cluster. These results are consistent with phenotypic switching of smooth muscle cells and macrophages in AAA tissues, which provided novel insights into cell-specific mechanisms of AAA formation.

scRNAseq requires sufficient number of cells for the mRNA analysis to provide statistically robust data. Although aortic tissues are composed of multiple cell types, the number of cells in one mouse aorta, especially one aortic segment, is usually insufficient for scRNAseq with current technology. Therefore, it is common to pool multiple aortic tissues for scRNAseq in mice. In the study by Zhao et al.,⁴ five infra-renal aortas were pooled in each group. scRNAseq was performed using 1396 to 1737 cells and revealed an informative array of transcriptomes in smooth muscle cells and macrophages. One common constraint of this technique is that, partially due to expense, biological and technical replicates are not common in scRNAseq. Therefore, careful study design, including procedural standardization, is required to obtain an appropriate number of cells and meaningful interpretations in scRNAseq.

Bulk RNA sequencing has become a common, unbiased approach to provide insight into potential disease mechanisms. However, cell-specific information cannot be obtained by this method. Since scRNAseq profiles at a single cell level, the determination of gene abundance at this level is a great advantage of scRNAseq. Despite this advantage, a major shortcoming is the lack of spatial distributions of cellular information within the tissue. The positional relationship between cells and their tissues is important to understanding the disease process. Therefore, classical approaches, such as immunostaining and in situ hybridization, remain reasonable approaches to elucidate the spatial localization of target genes and protein detected by scRNAseq. In recent years, spatial transcriptomics technology has been developed,¹⁰ and the regional distribution of target genes may be determined by this novel technique. The combination of spatial transcriptome and scRNAseq would provide more informative insights into understanding cell-specific molecular mechanism of AAAs. The study by Zhao et al.⁴ revealed unique gene alterations in some sub-clusters of smooth muscle cells and macrophages. It would be interesting to elucidate the localization of these molecules in AAAs.

With the publication of Zhao et al.⁴ describing cellular transcriptome of elastase-provoked AAAs by scRNAseq, this form of analysis has been performed on the three most commonly used mouse models of AAAs.^2–4 All three studies have consistently demonstrated the heterogeneity of aortic cells in AAAs. scRNAseq has become an instructive approach to provide detailed insight into cellular heterogeneity. While important, the data remain descriptive. These data illuminate the need for developing technology that permits manipulation of specific cellular sub-clusters to enable a determination of the role of these cells on the aneurysmal process.

Conflict of interest: none declared.

Figure 1.

scRNAseq revealed cellular heterogeneity of elastase-induced AAAs. FBs, fibroblasts; ECs, endothelial cells; ICs, inflammatory cells; SMCs, smooth muscle cells.

Funding

The authors’ research work was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (R01HL133723) and the American Heart Association (18SFRN33960163).

 

The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.