Stem Cell Pluripotency Genes Klf4 and Oct4 Regulate Complex SMC Phenotypic Changes Critical in Late-Stage Atherosclerotic Lesion Pathogenesis

Abstract

Background: Rupture and erosion of advanced atherosclerotic lesions with a resultant myocardial infarction or stroke are the leading worldwide cause of death. However, we have a limited understanding of the identity, origin, and function of many cells that make up late-stage atherosclerotic lesions, as well as the mechanisms by which they control plaque stability.

Methods: We conducted a comprehensive single-cell RNA sequencing of advanced human carotid endarterectomy samples and compared these with single-cell RNA sequencing from murine microdissected advanced atherosclerotic lesions with smooth muscle cell (SMC) and endothelial lineage tracing to survey all plaque cell types and rigorously determine their origin. We further used chromatin immunoprecipitation sequencing (ChIP-seq), bulk RNA sequencing, and an innovative dual lineage tracing mouse to understand the mechanism by which SMC phenotypic transitions affect lesion pathogenesis.

Results: We provide evidence that SMC-specific Klf4- versus Oct4-knockout showed virtually opposite genomic signatures, and their putative target genes play an important role regulating SMC phenotypic changes. Single-cell RNA sequencing revealed remarkable similarity of transcriptomic clusters between mouse and human lesions and extensive plasticity of SMC- and endothelial cell-derived cells including 7 distinct clusters, most negative for traditional markers. In particular, SMC contributed to a Myh11^-, Lgals3⁺ population with a chondrocyte-like gene signature that was markedly reduced with SMC-Klf4 knockout. We observed that SMCs that activate Lgals3 compose up to two thirds of all SMC in lesions. However, initial activation of Lgals3 in these cells does not represent conversion to a terminally differentiated state, but rather represents transition of these cells to a unique stem cell marker gene-positive, extracellular matrix-remodeling, "pioneer" cell phenotype that is the first to invest within lesions and subsequently gives rise to at least 3 other SMC phenotypes within advanced lesions, including Klf4-dependent osteogenic phenotypes likely to contribute to plaque calcification and plaque destabilization.

Conclusions: Taken together, these results provide evidence that SMC-derived cells within advanced mouse and human atherosclerotic lesions exhibit far greater phenotypic plasticity than generally believed, with Klf4 regulating transition to multiple phenotypes including Lgals3⁺ osteogenic cells likely to be detrimental for late-stage atherosclerosis plaque pathogenesis.

Keywords: coronary disease; single nucleotide polymorphisms; smooth muscle cell; smooth muscle differentiation; smooth muscle progenitor cell.

Figure 1.

Bulk RNA-seq and SMC^Oct4 and SMC^Klf4 ChIP-seq analysis of advanced BCA lesion areas from SMC-specific Klf4 vs Oct4 knockout apoE^-/- mice showed virtually opposite genomic signatures and identify candidate SMC Klf4 or Oct4 target genes that may play an important role in late-stage atherosclerotic lesion pathogenesis. RNA-seq was performed on the BCA regions from 18-week Western diet–fed SMC^Klf4-Δ/Δ vs SMC^Klf4-WT/WT and SMC^Oct4-Δ/Δ vs SMC^Oct4-WT/WT mice. A, GSEA analysis showing MSigDB Hallmark pathways from comparing KO vs WT of both lines are shown. B, Venn diagram showing the common pathways between downregulated SMC^Klf4-Δ/Δ vs SMC^Klf4-WT/WT and upregulated in SMC^Oct4-Δ/Δ vs SMC^Oct4-WT/WT. C, Venn diagram showing the common pathways between upregulated SMC^Klf4-Δ/Δ vs SMC^Klf4-WT/WT and downregulated in SMC^Oct4-Δ/Δ vs SMC^Oct4-WT/WT. D, Bar graph showing the top 8 common pathways from B and all common pathways in C differentially regulated pathways between SMC^Klf4-Δ/Δ vs SMC^Klf4-WT/WT and SMC^Oct4-Δ/Δ vs SMC^Oct4-WT/WT analyses. E, ChIP-seq was done on the BCA regions from 18-week Western diet–fed SMC^Klf4-Δ/Δ vs SMC^Klf4-WT/WT and SMC^Oct4-Δ/Δ vs SMC^Oct4-WT/WT mice (n=15). The Venn diagram shows overlap of putative “closest gene” binding partners of Klf4 and Oct4 in SMC (ie, those binding partners that are lost in the SMC^Klf4-Δ/Δ compared with SMC^Klf4-WT/WT) that are also human CAD GWAS loci. P value of gene overlap <0.01. F, Table showing genes present in the Venn diagram in E. BCA indicates brachiocephalic artery; CAD, coronary artery disease; ChIP-seq, chromatin immunoprecipitation sequencing; FDR, false discovery rate; GSEA, gene set enrichment analysis; GWAS, genome-wide association study; KO, knockout; RNA-seq, RNA sequencing; SMC, smooth muscle cell; and WT, wild-type.

Figure 2.

scRNA-seq analysis of healthy aorta and advanced BCA lesions identified 14 distinct cell clusters including 7 predominately of SMC origin. A, Cells were harvested for scRNA-seq from microdissected advanced BCA lesions from SMC lineage-traced (Myh11-Cre^ERT2 Rosa-eYFP apoE^-/-) or endothelial lineage-traced (Cdh5-Cre^ERT2 Rosa-eYFP apoE^-/-) 18-week Western diet–fed mice, as well as healthy 8-week-old apoE^-/- SMC lineage-traced mouse aorta as a control. In aggregate, these data are the results of 7 experiments, 19 Chromium 10X libraries, and 23 mice. Results are shown as a UMAP colored by cluster. B, Dot plot of traditional markers used for cell identity. Cand D, Feature plots showing gene expression of Myh11, Acta2 (C) and Cdh5 and Pecam1 (D), as representative genes of classical markers for SMC and endothelial cells, respectively. Myh11 was detected only in clusters 1 to 2, whereas Cdh5 was detected exclusively in cluster 8. E, Dot plots and feature plots of selected genes representing 3 SMC phenotypic groups, stem/inflammatory, ECM remodeling, and osteogenic. Further plots and pathway analysis justifying the groups can be found in Figures V and VI in the Data Supplement. FandG, UMAP of mouse cells showing (F) Myh11-Cre^ERT2 Rosa-eYFP apoE^-/- (black dots) or (G) Cdh5-Cre^ERT2 Rosa-eYFP apoE^-/-lineage-traced cells from BCA plaques of mice fed a Western diet for 18 weeks (black dots). BCA indicates brachiocephalic artery; scRNA-seq, single-cell RNA sequencing; SMC, smooth muscle cell; and UMAP, uniform manifold approximation and projection.

Figure 3.

scRNA-seq of advanced BCA lesions from SMC-specific Klf4 knockout vs wildtype littermate control mice identified multiple Klf4-dependent clusters. A, Cells from advanced BCA lesions were isolated from SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ mice after 18 weeks of Western diet and submitted for scRNA-seq. UMAP analysis shows cells from SMC^Klf4-WT/WT on the left and from SMC^Klf4-Δ/Δ cells on the right. The analysis includes 5469 cells from 2 different experiments (Chromium 10x runs) and 11 animals. B, Percentage of SMC-lineage traced cells (eYFP⁺ cells sorted from BCA plaque) in each UMAP cluster. C, Percentage of non-SMC lineage-tracing cells (unsorted cells) in each UMAP cluster from each respective group described in A. D–G, BCA lesions from SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ mice were stained for selected immunofluorescent markers from scRNA-seq clusters. D, Costaining for TRPV4 and S100b (markers present in cluster 7) with eYFP in SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ animals. Images are based on confocal microscopy z-stacks of 1 µm thickness on 10-µm sections of BCA lesions. The top left panel shows a maximum intensity projection ×20 zoom. A close-up view of eYFP⁺TRPV4⁺S100b⁺ cells is shown in a 1-µm individual z-stack in a WT animal. E, Quantification of the frequency of triple-positive (eYFP⁺ TRPV4⁺ S100B⁺) cells as a percent of total eYFP⁺ SMC-derived cells in SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ (n=6–8, error bars show SEM). F, Results of costaining for eYFP and Phactr1 (a marker of cluster 3) in advanced BCA lesions from SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ animals. The top panels show ×20 images. The bottom panels show a close-up image of a BCA lesion from SMC^Klf4-Δ/Δ animal. G, Quantification of the frequency of double-positive (eYFP⁺ Phactr1⁺) cells as a percent of total eYFP⁺ SMCs in advanced BCA lesions from SMC^Klf4-WT/WT and SMC^Klf4-Δ/Δ mice (n=6, error bars show SEM). BCA indicates brachiocephalic artery; scRNA-seq, single-cell RNA sequencing; SMC, smooth muscle cell; UMAP, uniform manifold approximation and projection; and WT, wild-type.

Figure 4.

A large fraction of Myh11-tdTomato⁺ SMCs phenotypically transition through an Lgals3 activation step and accumulate within advanced BCA lesions of Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- mice. A, Experimental design; SMC^{Dual Lineage} (Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/-) mice were injected with tamoxifen at 6 to 8 weeks of age and subsequently placed on Western diet for 18 weeks to induce advanced atherosclerosis. Video of mouse model is available at https://www.cvrc.virginia.edu/Owens/educationaltopics.html. B, Sections of the BCA at the origin from the aorta in an SMC^{Dual Lineage} mouse, immunostained for GFP, tdTomato, and Lgals3, were imaged with confocal microscopy z-stacks of 1 µm thickness. Picture at right shows the maximum-intensity projection image of the 10-µm tissue thickness, and the close-up view shows eGFP⁺ Lgals3⁺ cells within a 1-µm individual z-stack. C, Quantification of cells positive for eGFP or Lgals3 by single-cell counting in 1-µm z-stacks of advanced BCA lesions from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE ^-/- mice (n=6, error bars show SEM). D, Flow-cytometric evaluation of cells freshly isolated from microdissected BCA lesions from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- BCA mice (n=4). These analyses included gating on live, single cells with fluorescence-minus-1 controls. Results show clear resolution of tdTomato⁺, eGFP⁺, and dual tdTomato⁺ eGFP⁺ (transition state) cells with normalization to the total number of lesion cells in the left panel or lineage-traced SMC in the right panel. E, Immunostaining of BCA lesions from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- 18-week Western diet–fed mice for GFP, tdTomato, and Myh11 reveals rare tdTomato⁺GFP⁺ cells in the fibrous cap (open arrows), overlying GFP⁺-only cells (filled arrows). F, Quantification of the proportion of cells in the fibrous cap of advanced BCA lesions that were tdTomato⁺, GFP⁺, or double-positive as a percent of total DAPI-stained fibrous cap cells. The fibrous cap was defined as the 30-µm lesion area underlying the lumen (n=6, error bars show SEM; the P value refers to unpaired t test between tdTomato⁺ and GFP⁺ cells). G, Experimental design; Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- mice were injected with tamoxifen between 6 and 8 weeks of age and put on Western diet for 10 weeks to induce early atherosclerotic lesions. Ten animals were sampled at 4 locations in the BCA and imaged to capture early-stage lesions with <50 SMCs based on counting of lineage-traced SMC-derived cells in the entire lesion. H, Immunostaining for GFP, tdTomato, and Lgals3 by confocal microscopy as in A. I, Quantification of GFP⁺ and tdTomato⁺ cells in lesions of <50 lineage-traced SMCs total (n=8, error bars show SEM, P<0.01 by unpaired t test), J, SMCs isolated from Myh11 lineage tracing mice were treated with siLgals3 or a scrambled control siNeg, serum starved, and scratched with a pipet tip in 10% serum media + 10 µg/mL PDGF-BB. Light microscopic images of 3 points along the scratch were each taken at 0, 24, and 48 hours, and cells within the scratch area defined at time 0 hours were quantified (n=6, error bars show SEM). BCA indicates brachiocephalic artery; GFP, green fluorescent protein; PDGF, platelet-derived growth factor; siLgals3, small interfering RNA to Lgals3; and SMC, smooth muscle cell.

Figure 5.

Myh11-expressing differentiated SMC that transition through an Lgals3 activation state take on multiple phenotypes. A, scRNA-seq was performed on BCA plaque cells isolated from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- mice fed a Western diet for 18 weeks, as well as eGFP⁺ and tdTomato⁺ cells sorted from plaque or whole atherosclerotic aorta. The results of UMAP analyses are shown colored by the library of origin. Results are from 2143 cells from 3 separate Chromium 10x runs and 6 animals. B, UMAP of analysis based on only cells from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- experiments showing clear separation of tdTomato⁺ and GFP⁺ (right side). C, Quantification of unsorted, GFP⁺, and tdTomato⁺ cells in each cluster in A. D and E, Atherosclerotic lesions from Myh11-Dre^ERT2 Lgals3-Cre Rosa-tdTomato-eGFP apoE^-/- were immunostained for major markers of GFP⁺ clusters from our scRNA-seq analyses and Acta2. Frozen sections (10 µm) of renal arteries were used for VCAM1 and Sca1 antibody staining, along with tdTomato signal (endogenous) and a GFP antibody. Paraffin sections (10 µm) of BCA were used for Acta2 and Sox9. Because of limited available antibodies, Sox9 required staining for tdTomato and GFP on sequential sections, which were used for quantification in D. Sox9 and GFP are shown as the representative image in E. Sox9 positivity was not found on SMCs outside the most advanced BCA lesions, including renal and aortic root sections. BCA indicates brachiocephalic artery; GFP, green fluorescent protein; scRNA-seq, single-cell RNA sequencing; SMC, smooth muscle cell; and UMAP, uniform manifold approximation and projection.

Figure 6.

Human scRNA-seq shows similar groups of fibrous cap and osteogenic SMC in advanced lesions. A, UMAP analysis of scRNA-seq results from human carotid endarterectomy samples, colored by cluster (left) or by origin (right) against mouse scRNA-seq data from Figure 2. B, Human coronary artery lesions graded as stable or ruptured were immunostained for Acta2 and Phactr1, a marker of cluster 1 to 3 SMCs identified in our scRNA-seq analyses of advanced BCA lesions. Images of whole immunostained human coronary lesions were obtained by taking a tile scan of ×20 zoom confocal microscopy z-stacks of full 10-µm thickness. C, Human coronary artery lesions graded as stable or ruptured were immunostained for Sox9, TRPV4, and S100b, which were markers of osteogenic SMCs identified in our scRNA-seq analyses of mouse advanced BCA lesions, imaged as described in B. BCA indicates brachiocephalic artery; scRNA-seq, single-cell RNA sequencing; SMC, smooth muscle cell; and UMAP, uniform manifold approximation and projection.