Expression Levels of the Tnni3k Gene in the Heart Are Highly Associated with Cardiac and Glucose Metabolism-Related Phenotypes and Functional Pathways

Figures

• 

  Figure 1. Overall flow of the work. WGS: whole genome sequencing; SNP: single nucleotide polymorphism; qRT-PCR: quantitative real-time reverse transcription polymerase chain reaction.

• 

  Figure 2. Tnni3k expression in mouse and human tissues. (A) Expression of Tnni3k in the heart tissue of BXD mice and (B) between B and D genotypes. The x-axis indicates the BXD strains or B and D alleles, and the y-axis denotes the RMA (robust multichip average) normalized log2 expression levels. Expression of Tnni3k across various adult tissues of (C) human and (D) mouse based on RNA sequencing data obtained from National Center for Biotechnology Information (NCBI) gene database. The x-axis indicates the expression levels in reads per kilobase per million mapped reads (RPKM) and the y-axis shows the adult tissue names. *** p < 0.0001, indicating a significant difference between B and D alleles.

• 

  Figure 3. eQTL mapping of Tnni3k in the BXD strains. (A) Manhattan plots showing the genetic locus on chromosome 3 in murine whole genome that regulates expression of Tnni3k. (B). Manhattan plots on chromosome (Chr) 3 showing the detailed location of eQTL at 154.78 megabases. Approximately 7300 informative SNP genotype markers across the BXD mouse genome were used for calculating the association with Tnni3k mRNA expression variation using likelihood ratio statistics (LRS). The x-axis denotes the chromosomal position in megabases on the mouse genome or Chr 3, and y-axis indicates the peak LRS score. The pink and grey horizontal lines indicate significant and suggestive LRS, respectively. The purple triangle on the x-axis indicates the genomic position of Tnni3k (Chr 3: 154.786). LRS is shown by blue lines, and additive effects are shown by red and green lines.

• 

  Figure 4. Pathway enrichment analysis of Tnni3k-correlated genes. (A) Top 5 significantly enriched (FDR p-value < 0.02) pathways including diabetic cardiomyopathy, adrenergic signaling in cardiomyocytes, insulin resistance, TGFβ (transforming growth factor beta), and AGE-RAGE signaling in diabetic complications. (B) Other significantly enriched (p-value < 0.05) cardiac and glucose metabolism-related pathways. The x-axis indicates the gene ratio/count, and the y-axis represents the KEGG pathways. (C) Genetically engineered alleles, transgenes, or QTL variants were retrieved from the Mouse Genome Informatics (MGI) database (http://www.informatics.jax.org/allele, accessed on 24 December 2021) with keywords “glucose metabolism” or “cardiovascular”. The Venn diagram shows the number of overlapping Tnni3k-correlated genes involved in “glucose metabolism” and “cardiovascular” related functions.

• 

  Figure 5. Protein–protein interaction network of Tnni3k-correlated genes involved in top key pathways. The PPI network was constructed using Tnni3k and its correlated genes involved in the top five pathways identified. Node color (red) intensity represents increasing degree. Triangle nodes represent proteins directly interacting with TNNI3K. The interactions supported by ‘experiments’ evidence in the STRING database are represented with blue dotted lines.

• 

  Figure 6. Expression variation of important pathway genes between B and D genotypes. (A) The heatmap of genes involved in top five Tnni3k-associated pathways constructed using ClustVis tool (https://biit.cs.ut.ee/clustvis/, accessed on 30 December 2021). RMA normalized intensity values across the mice were considered. Each row represents a gene name, and each column indicates a BXD strain grouped according to control B with wild-type T659 Tnni3k with no splicing (inherent from parental B6 mouse) or D alleles carrying a compound T659I and splicing (inherent from parental D2 mouse). Genes significantly different between the two genotypes are indicated with an asterisk (* p < 0.05, ** p < 0.01, *** p < 0.001). (B) Results of qRT-PCR analysis of myocardial genes that directly interact with Tinn3k and have significant expression differences between B and D alleles. BXD strains with wild-type Tnni3k represent a B allele (white columns) and those with compound Tnni3k variants represent a D allele (black columns). The error bars show standard deviation (SD) values. * p < 0.05, ** p < 0.01, *** p < 0.001.

• 

  Figure 7. Scatter plots of the correlations between Tnni3k expression and echocardiography phenotypes in BXD mice of Tnni3k B or D genotypes. (A) Correlation between Tnni3k expression and echocardiography parameters. Tnni3k expression (x-axis) is significantly correlated with heart rate (left panel, y-axis, beats/min) and thickness of left ventricular posterior wall (LVPW) at the end-diastole (d, right panel, y-axis, mm) in BXD strains (N = 5 mice/strain, 25 strains). The Pearson correlation coefficient (r) was used to determine the relationship. The r- and p-values are indicated in each graph. The gene expression level is log2-transformed. (B,C) Echocardiography parameters between B and D genotypes of BXD mice. Y-axis indicates heart rate ((B), left, beat/min), LVPW thickness ((B), right, mm), LV volume at the end-systole (s, (C), left, μL), and ejection fraction (EF, (C), right, %) in BXD strains of B (25 strains) and D (16 strains) genotypes (N = 5 mice/strain). * p < 0.05, indicating a significant difference between B and D alleles. (D) Correlation between Tnni3k expression and electrocardiography (ECG) tracing parameters. Tnni3k expression (x-axis) is significantly positively…

• 

  Figure 8. Scatter plots of the correlations between Tnni3k expression and glucose metabolism-related phenotypes in BXD mice. (Left panel) Tnni3k expression (x-axis) is significantly negatively correlated with blood glucose level (y-axis) in BXD mice (N = 5 mice/strain, 30 strains). (Right panel) Tnni3k expression (x-axis) is significantly positively correlated with insulin response during oral glucose tolerance test (y-axis) in BXDs (N = 5 mice/strain, 42 strains). The Pearson correlation coefficient (r) was used to determine the relationship. The r and p-values are indicated in each graph. The gene expression level is log2-transformed.