Transcriptional Profiling of Normal, Stenotic, and Regurgitant Human Aortic Valves

Abstract

The genetic mechanisms underlying aortic stenosis (AS) and aortic insufficiency (AI) disease progression remain unclear. We hypothesized that normal aortic valves and those with AS or AI all exhibit unique transcriptional profiles. Normal control (NC) aortic valves were collected from non-matched donor hearts that were otherwise acceptable for transplantation (n = 5). Valves with AS or AI (n = 5, each) were collected from patients undergoing surgical aortic valve replacement. High-throughput sequencing of total RNA revealed 6438 differentially expressed genes (DEGs) for AS vs. NC, 4994 DEGs for AI vs. NC, and 2771 DEGs for AS vs. AI. Among 21 DEGs of interest, APCDD1L, CDH6, COL10A1, HBB, IBSP, KRT14, PLEKHS1, PRSS35, and TDO2 were upregulated in both AS and AI compared to NC, whereas ALDH1L1, EPHB1, GPX3, HIF3A, and KCNT1 were downregulated in both AS and AI (p < 0.05). COL11A1, H19, HIF1A, KCNJ6, PRND, and SPP1 were upregulated only in AS, and NPY was downregulated only in AS (p < 0.05). The functional network for AS clustered around ion regulation, immune regulation, and lipid homeostasis, and that for AI clustered around ERK1/2 regulation. Overall, we report transcriptional profiling data for normal human aortic valves from non-matched donor hearts that were acceptable for transplantation and demonstrated that valves with AS and AI possess unique genetic signatures. These data create a roadmap for the development of novel therapeutics to treat AS and AI.

Keywords: RNA sequencing; aortic insufficiency; aortic stenosis; transcriptional profiling.

Figure 1

Differentially expressed genes in aortic valves after RNA Sequencing. (a) Venn diagram of differentially expressed genes (DEGs) in valves with aortic stenosis (AS), aortic insufficiency (AI), and normal controls (NC). (b) Volcano plots representing the distribution of gene expression in AS, AI, and NC aortic valves. The gene expression of AI vs. NC, AS vs. NC, and AS vs. AI are represented relative to statistical significance values (p < 0.05). The x-axis represents relative fold change, while the y-axis represents statistical significance. Each point represents a gene and its expression. (c) Cluster analysis of DEGs in the AS, AI, and NC groups. Increased expression is represented in red while decreased expression is represented in blue. (d) The top 30 upregulated and downregulated genes identified by RNA sequencing are identified for AS vs. NC and AI vs. NC. Gene identification and gene names are listed with the fold change and adjusted p-value (p-adj). Red represents upregulated genes relative to NC and blue represents downregulated genes relative to NC.

Figure 2

Confirmation of selected differentially expressed genes using quantitative polymerase chain reaction. (a) Quantitative polymerase chain reaction (qPCR) was performed for the differentially expressed genes (DEGs) with the greatest expression differences between valves with aortic stenosis (AS) or aortic insufficiency (AI) compared to the normal controls (NC), as well as for selected genes which were identified as DEGs in previous studies. Data are presented as mean ± standard error. * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001. (b) Gene expression of selected housekeeping genes ATP5F1, CYC1, and RPL32, showing no difference in expression when comparing AS, AI, and NC valves. Data are presented as mean ± standard error.

Figure 3

Functional network of genes involved in aortic stenosis. Each node represents a gene ontology (GO) term that has a significant number of differentially expressed genes. Edges represent known associations between GO terms. GO terms are grouped together by similarity and assigned a representative group name. Major clusters for aortic stenosis were ion regulation, immune system response, blood pressure regulation, and lipid homeostasis.

Figure 4

Functional network of genes involved in aortic insufficiency. Each node represents a gene ontology (GO) term that has a significant number of differentially expressed genes. Edges represent known associations between GO terms. GO terms are grouped together by similarity and assigned a representative group name. The major system for aortic insufficiency involved the ERK1 and ERK2 signaling cascade.

Figure 5

Protein-protein interaction model of valves with aortic stenosis. (a) Protein-protein interaction model of differentially expressed genes (DEGs) from RNA sequencing analysis of aortic stenosis (AS) versus normal control (NC) valves. DEGs found in AS versus NC alone with >3-fold change were included. The STRING database was used to generate gene relationships. STRING scores <0.7 were excluded. Edges are weighted to indicate STRING score. A list of high-impact genes was generated from these interactions. (b) High-impact genes for AS were defined as having (1) 15 or more interactions; (2) >5-fold absolute change; (3) connectivity to surrounding clusters.

Figure 6

Protein-protein interaction model of valves with aortic insufficiency. (a) Protein-protein interaction model of differentially expressed genes (DEGs) from RNA sequencing analysis of aortic insufficiency (AI) versus normal control (NC) valves. DEGs found in AI versus NC alone with >1.5-fold change were included. The STRING database was used to generate gene relationships. STRING scores <0.7 were excluded. Edges are weighted to indicate STRING score. A list of high-impact genes was generated from these interactions. (b) High-impact genes for AI were defined as (1) >5 interactions; (2) >2-fold absolute change; (3) connectivity to surrounding cluster.