BAG3 genetic variant is associated with heart failure with preserved ejection fraction phenotypes

Figures

• 

  Fig 1. Identification of BAG3 gene polymorphisms. A) Diagram depicting the genomic locations of polymorphisms: rs2240777, rs2234962 and rs196326. Known allele frequencies of each SNP according to Genome1000 [12] are shown in the adjacent bar graphs. Arrows indicate the primers used for sequence analysis. B) Representative electropherograms indicating the presence of rs2240777 and C) rs196326 polymorphisms in our cohort. D) However, we only detected the T allele (reference allele) of rs2234962 SNP in our cohort. E) Analysis of BAG3 mRNA expressions according to rs196326 SNP (CC,n = 12; CG,n = 36; GG, n = 13) and F) based on rs2240777 polymorphism (AA, n = 29; AG/GG, n = 32). Data are presented as mean ± SEM. Statistical test used: E) One way ANOVA followed by multiple comparisons, F) Student’s t-test.

• 

  Fig 2. Echocardiography indices of patients with HFpEF according to rs196326 polymorphism. Echocardiography analyses were performed in HFpEF patients to assess A) Left ventricular mass index (LVMi), B) Relative wall thickness (RWT), C) Ejection fraction (EF), D) E/A ratio, and E) Mitral valve early velocity (MV E Vel), comparing individuals with the CC (n = 6), CG (n = 24) and GG (n = 6) genotype of the rs196326 SNP. F) Systolic blood pressure was also examined comparing patients with rs196326 genotypes. Data are presented as mean ± SEM. Statistical tests used: A,B,C,E,F, One-way ANOVA followed by multiple comparisons test; D, Kruskal-Wallis test followed by multiple comparisons.

• 

  Fig 3. Analysis of serum biomarkers in HFpEF patients according to rs196326 genotype. Serum biomarkers were examined using ELISA in patients with HFpEF, including A) Heart failure marker NT-proBNP, B) Autophagy biomarker Beclin-1, C) Caspase-3 (marker of apoptosis and metabolic stress), and D) GRP78 (biomarker of ER stress). Comparisons were performed according to rs196326 genotypes (CC,n = 6; CG,n = 24; GG,n = 6). Data are presented as mean ± SEM. Statistical test used: A,B,C,D, Kruskal-Wallis test followed by multiple comparisons.

• 

  Fig 4. Echocardiography analyses comparing HFpEF patients with different rs2240777 genotypes. Analysis of cardiac morphology and functions was performed to compare several echocardiography parameters including the A) Left ventricular mass index (LVMi), B) Relative wall thickness (RWT), C) Ejection fraction (EF), D) E/A ratio, and E) Mitral valve early velocity (MV E Vel), between patients with rs2240777 genotypes: AA (n = 16) and AG/GG (n = 20). F) Systolic blood pressure was also examined comparing patients with the rs2240777 AA vs AG/GG genotypes. Data are presented as mean ± SEM. Statistical tests used: A,C,F, Student’s t-test; B,D,E, Mann-Whitney U test.

• 

  Fig 5. Levels of clinical and pathophysiological biomarkers according to rs2240777 genotypes. Biomarkers of A) Heart failure and myocardial stretch (NT-proBNP), B) Autophagy (Beclin-1), C) Apoptosis and metabolic stress (caspase-3), and D) Endoplasmic reticulum (ER) stress (GRP78) were analysed and compared between HFpEF patients with the rs2240777 AA genotype (n = 16) and AG or GG genotypes (n = 20). Data are presented as mean ± SEM. Statistical tests used: A,B,C,D, Mann-Whitney U test.

• 

  Fig 6. Computational analysis to predict the molecular processes affected by the rs196326 SNP. A) Data from SpliceAI describing that the rs196326 genetic variation does not significantly alter splicing donors or acceptors near the polymorphic site. B) SpliceAid2 algorithm was used to predict the binding of intronic splicing enhancers (ISE) and intronic splicing silencers (ISS) at the polymorphic site. The data indicate that rs196326 C/G variations may change the likelihood of binding to different ISE/ISS, from binding with hnRNP H1/H2 (C allele) to hnRNP A1 (G allele). C) The rs196326 SNP may alter the binding to transcription factors/enhancers. By using Ciiider it was predicted that the reference sequence carrying C allele favours binding to Nkx2-5, whereas the alternate G allele favours binding to CREB. The RNA genomic data shows that this sequence is adjacent to the transcribed BAG3 antisense RNA region suggesting a possible alteration in BAG3 gene expression regulation.