Single-Cell Transcriptomics Identifies Selective Lineage-Specific Regulation of Genes in Aortic Smooth Muscle Cells in Mice

Abstract

Background: Smooth muscle cells (SMCs) of the proximal thoracic aorta are derived from second heart field (SHF) and cardiac neural crest (CNC) lineages. Recent studies, both in vitro and in vivo, have implied relevance of lineage-specific SMC functions in the pathophysiology of thoracic aortic diseases; however, whether 2 lineage-derived SMCs have any predisposed transcriptional differences in the control aorta remains unexplored.

Methods: Single-cell RNA sequencing and single-nucleus assay for transposase-accessible chromatin sequencing were performed on isolated cells from the aortic root and ascending aortas of 14-week-old SHF-traced (Mef2c-Cre^+/0-Yfp^+/0) and CNC-traced (Wnt1-Cre^+/0-Yfp^+/0) male mice. RNA in situ hybridization was performed for spatial expression of selected differentially expressed genes (DEGs) of both lineages.

Results: Lineage stratification of SMCs in the proximal thoracic aorta was identified using antibody-based immunofluorescence staining. Single-cell RNA sequencing recognized 12 consistently upregulated DEGs (Des, Tnnt2, Hand2os1, Psd, Gpc3, Meis2, Dcn, Gm34030, Palld, Nrtn, Lum, and Cfh) in SHF-derived SMCs and 9 consistently upregulated DEGs (Ccn5, Ccdc42, Tes, Eln, Aebp1, Galnt6, Ccn2, Aopep, and Wtip) in CNC-derived SMCs. RNA in situ hybridization validated upregulated expressions of selective SHF-specific DEGs at the aortic root. We found SHF-derived SMCs contain a distinct, large subpopulation of SMCs that is enriched with Des and Tnnt2 expressions. Single-nucleus assay for transposase-accessible chromatin analysis further confirmed higher chromosomal accessibility for upregulated DEGs of SHF-derived SMCs.

Conclusions: The present study recognizes the presence of limited but distinct transcriptomic differences between CNC-derived and SHF-derived SMCs in the control proximal thoracic aorta.

Keywords: aorta; chromatin; gene expression profiling; neural crest.

Figure 1:. Lineage-specific localization of cells in the proximal aorta.

(A) SHF- and CNC-tagged mice were generated by breeding individual lineage-specific Cre mice with Yfp-reporter mice. (B) Selected locations for tissue sections from proximal aortas of 14-week-old Mef2c-Cre^+/0 Yfp^+/0 and Wnt1-Cre^+/0 Yfp^+/0 mice (n =6 mice/lineage; 3 locations per mouse at aortic root, proximal and distal ascending aorta). (C) Representative immunofluorescence images for YFP staining in indicated sections of both lineages. Size bars: 150 μm (low magnification) and 75 μm (high magnification); dashed lines in proximal ascending aortas indicate middle of the media. (D) Quantification of YFP+ medial area per section as a percentage of total medial area, for both lineages. Two-way ANOVA and Bonferroni’s multiple comparisons tests were used for statistical analysis. (E) Graphs represent the percentage of YFP+ medial area that overlaps with either the luminal or abluminal half of the media in only proximal ascending aortic sections. Unpaired two-tailed Student t tests were used. (D-E) Each data point represents an individual mouse. N=6/lineage. Data presented as Mean±SD. SHF second heart field; CNC cardiac neural crest.

Figure 2:. Cell populations identified from single-cell RNA sequencing of lineage-traced wild type proximal aorta.

(A) Experimental workflow for scRNA sequencing. SHF- or CNC-lineage traced proximal aortic tissues (aortic root to distal ascending aorta, indicated in red dashed lines) were pooled from 2 mice, and digested to prepare single-cell suspension. Cells were sorted based on viability and YFP+ & –ve signaling and single cell (RNA) libraries were prepared. The experiment was repeated twice with new set of aortic samples for each lineage-tagged (Mef2c-Cre^+/0 Yfp^+/0 or Wnt1-Cre^+/0 Yfp^+/0) mice. (B) UMAP shows unsupervised clustering of all cells from proximal aortas. Inset pie chart indicates percent population of the major cell types. (C) Dot plot for highest specificity gene markers of major cell clusters using DEG analysis with Wilcoxon rank sum. Dot size represents percent of cells expressing denoted gene, color represents average normalized expression level.

Figure 3:. Transcriptomic analysis of aortic SMCs based on lineages.

(A) Distribution of SMCs based on CNC and SHF lineage. (B) Modified Vann diagram indicates numbers of upregulated differentially expressed genes (DEGs) that were identified by pairwise comparison of genes using Wilcoxon rank sum between CNC- and SHF-derived SMCs. Six colored petals represent separate analysis consisting of all SMCs (pooled Yfp+ and Yfp- SMCs), only Yfp+ SMCs (pooled), and for four individual single cell isolation experiments. The middle circle includes common DEGs that are shared by all six analyses. Only 12 consistently upregulated DEGs for SHF and 9 consistently upregulated DEGs for CNC-derived SMCs were identified. (C) Dot plot for common upregulated DEGs of SHF- and CNC-derived SMCs. Dot size represents percent of cells expressing denoted gene, color represents average normalized expression level. (D) Violin plot comparing expression of selective DEGs between CNC and SHF lineages. P value is <0.001 for all represented plots. (E) Volcano plot represents upregulated DEGs for CNC- and SHF-derived SMCs. (F) Gene ontology enrichment analysis for overrepresented pathways involved with CNC- or SHF-derived SMCs.

Figure 4:. Analysis of subpopulations for lineage-specific SMCs at the proximal aorta.

(A) SMC sub-cluster populations for all SMCs (left), CNC-derived SMCs (middle) and SHF-derived SMCs (right). (B) Distribution of SMC sub-clusters in both lineages as percent population. (C) Top 5 DEGs presented in dot-plot for identified sub-clusters of SMCs irrespective of lineages using DEG analysis with Wilcoxon rank sum. Dot size represents percent of cells expressing denoted gene, color represents average normalized expression level. (D) Violin plot comparing expressions of selective DEGs in five identified sub-populations of SMCs in the proximal aorta.

Figure 5:. Spatial expressions of selective lineage-specific DEGs by RNA in situ hybridization.

(A) Representative images for spatial expressions of Des (Desmin) at aortic root, proximal and distal regions of ascending aorta. Size bars: 350 μm (low magnification) and 75 μm (high magnification) (B) Quantification of Des expression at three indicated locations for each aorta. One-way ANOVA and Bonferroni’s multiple comparisons tests were used for statistical analysis. (C) Percentage of Des expression present either at the luminal or abluminal half of the media in proximal ascending aorta sections. Unpaired two-tailed Student t tests were used. (D) Representative images for spatial expressions of Tnnt2 (in blue dots, SHF-derived DEG) and Ccn5 (in red dots; CNC-derived DEG) at aortic root, proximal and distal ascending aorta sections. Scale bar: lower magnification 300 μm, and higher magnification 70 μm. (E) Quantification of RNA expressions at indicated three locations for each aorta. One-way ANOVA and Bonferroni’s multiple comparisons tests were used. (F) Percentage of DEG expression present either at the luminal or abluminal half of the media in proximal ascending aorta sections. N=10 mice/gene. Unpaired two-tailed Student t tests were used. Data presented as Mean±SD.

Figure 6:. Single-cell assay for transposase-accessible chromatin dataset.

(A) Unsupervised clustering of total snATAC data from lineage-traced cells of 14-weeks old Mef2c-Cre^+/0Yfp^+/0 and Wnt1-Cre^+/0Yfp^+/0 male mice. (B) Lineage-based unsupervised distribution of SMC-specific snATAC data. (C) Volcano plot depicting selected upregulated genes on differential chromosomal accessibility peaks for CNC (left)- or SHF (right)-derived SMCs. (D) Enriched DNA accessibility peaks for chromosomal regions of selective DEGs identified from CNC- and SHF-derived SMCs. (E) Enriched DNA accessibility peaks for chromosomal regions of Hand2, Twist1, Sox9 and Gata4. (F) Motif enrichment analysis of lineage-specific SMC chromosomal accessibility for top five predicted transcription factor binding sites. (G) Pathway enrichment analysis based on top differentially accessible regions of CNC- and SHF-derived SMCs.