Disease- and sex-specific differences in patients with heart valve disease: a proteome study

Figures

• 

  Figure 1.. Study design. (A) Aortic stenosis results in pressure overload and leads to concentric cardiac hypertrophy, whereas mitral valve regurgitation causes volume overload and leads to eccentric hypertrophy. (B) Proteins were extracted from biopsies of the left ventricle of 41 patients with aortic stenosis (female n = 21; male n = 20), 17 patients with mitral valve regurgitation (female n = 5; male n = 12), and 17 healthy control hearts (CON) (female n = 8; male n = 9). Protein extracts were digested to peptides using trypsin and LysC and analyzed by high-resolution tandem mass spectrometry (LC–MS/MS). A deep reference proteome, acquired by analyzing a high-pH prefractionated peptide mixture of all samples, was used for MS1 matching. RAW data were processed with MaxQuant, and LFQ intensities were used for disease- and sex-specific protein expression analyses. Integrative analyses were performed on proteomic data and clinical imaging (cardiac magnetic resonance imaging) phenotypes.

• 

  Figure 2.. Imaging parameters. Overview of left ventricular clinical parameters from magnetic resonance imaging stratified by conditions (AS/MR). AS, aortic valve stenosis; EDVi, end-diastolic volume indexed to body surface area; MR, mitral valve regurgitation. Statistical comparison was performed with a two-sided, two-sample Wilcoxon rank test.

• 

  Figure 3.. Sex-stratified imaging parameters. Overview of clinical parameters from magnetic resonance imaging stratified by sex in AS and MR patients. End-diastolic volume, ejection fraction, fibrous tissue content, and myocardial mass of the left ventricle are described, and the mean pressure gradient across the aortic valve. AS, aortic valve stenosis; EDVi, end-diastolic volume indexed to body surface area; MR, mitral valve regurgitation. Group-wise statistical comparison was performed with a two-sided, two-sample Wilcoxon rank test.

• 

  Figure S1.. Heart proteome coverage and distribution. (A) Number of quantified proteins per sample. (B) Cumulative protein intensities in % ordered across the respective protein rank.

• 

  Figure 4.. Heart proteome distribution. (A, B) Principal component analysis of protein measurements displaying (A) all three conditions and sex assignment and (B) moderate (2) and severe (3) mitral valve insufficiency grades within the patient cohort of MR. AS, aortic valve stenosis; CON, healthy control hearts; MR, mitral valve regurgitation.

• 

  Figure 5.. Quantitative analyses of disease- and sex-specific differences in protein abundance. (A, B, C) Volcano plots denoting the log2 fold change and the negative log10 of the Benjamini–Hochberg-adjusted P-values for the comparisons between AS versus CON (A), MR versus CON (B), and AS versus MR (C). Differentially expressed proteins (adjusted P-value < 0.05) are colored in yellow (AS) or green (MR) and red (up-regulated in AS compared with MR) and orange (down-regulated in AS compared with MR). (D) Scatter plot of log2-fold changes in AS and MR versus CON. Colors denote changes that are shared (pale green) or divergent in the comparison against CON (black) or the specific one (yellow for AS; green for MR); that is, the change against CON is also significant against the respective other condition. Statistical analysis was performed using R including limma linear models and the Benjamini–Hochberg multiple testing correction. (E, F) Pie chart of GO term enrichment analysis. The merged set of enriched terms from up- and down-regulation is summarized into redundant clusters by REVIGO. (E, F) Proportions are based on the GOA frequency of the cluster representative terms, which are assigned to…

• 

  Figure S2.. Comparison of protein quantification. (A, B, C) Comparison of our study with Brandenburg et al in terms of quantified proteins (A) and significantly different proteins between control and AS (B) and MR (C) versus Brandenburg (NF versus NEFHG and LEFHG and LEFLG and PLFLG) based on a ComBat-corrected joint linear model analysis. (B, C) Fisher’s exact test for association between (B) AS-CON and Brandenburg, P = 8.6 × 10−9 and (C) MR-CON and Brandenburg, P = 4.5 × 10−4. Statistical information concerning Fig S2: protein quantifications and sample annotations of the Brandenburg et al study were retrieved from the associated PRIDE submission https://www.ebi.ac.uk/pride/archive/projects/PXD019594 file 20190503_S_Brandenburg ID and Quantitation Results.xlsx. Missing values were imputed as for our dataset. Proteins were mapped based on UniProt Accession number (matching the first entry in case of multiple UniProt Accession numbers in a protein group), without valid value filter in our study, but only proteins quantified in both datasets were kept. To obtain compatible data, batch correction via ComBat (Johnson et al, 2007; PMID: 16632515) using the sva R package (v. 3.38.0) and setting…

• 

  Figure 6.. Disease- and sex-specific differences in abundance of proteins related to ECM composition. (A) Clustered heatmap showing the condition’s mean abundance of proteins belonging to ECM-related GO terms. Annotation bars denote significant changes in condition (I) and in sex (II—effect in sex MR; III—effect in sex AS). Proteins described in the text are labeled. (B) Median log2 of LFQ intensity of all proteins belonging to the ECM; P-values are calculated using a Wilcoxon rank test. (C) Comparison of fibrous tissue content as measured by MRI in AS and MR (C). P-values are calculated using a Wilcoxon rank test with one sample against the reference mean. Dots represent individual subjects. (D, E) Combined results (−log10-transformed P-value) from GO term enrichment analysis in AS and MR versus CON (D) shown in the direction of regulation they were found in and on proteins found increased in male or female AS only (E). (F) Comparison of fibrous tissue content as measured by MRI in AS stratified by sex. P-values are calculated using a Wilcoxon rank test with two samples (female versus male, denoted by brackets). AS, aortic valve stenosis; CON, healthy control hearts; MR, mitral valve…

• 

  Figure S3.. Proteins belonging to ECM. Log2-transformed LFQ intensity for selected proteins belonging to the ECM. All proteins shown are mentioned in the main text and are significant in at least one comparison. In case of ambiguous gene symbols, the majority protein ID is shown in brackets. Full protein IDs and P-values for all comparisons are given in Table S1. AS, aortic valve stenosis; CON, healthy control hearts; MR, mitral valve regurgitation.

• 

  Figure S4.. Full GO term enrichment results of the sex-stratified analysis. (A, B) Combined results of GO term enrichment analysis of proteins found to be differentially expressed in only one sex (female = pink; male = blue) in AS (A) and MR (B). The negative log10-transformed P-value is encoded by circle-size plots that are subdivided to show the direction of the change in protein expression (up, down) and the respective category assignments. False discovery rate in the gprofiler2 analysis was set to 5%.

• 

  Figure 7.. Disease- and sex-specific differences in abundance of proteins related to energy metabolism and mitochondria. (A) Clustered heatmap showing the condition’s mean abundance of proteins belonging to energy metabolism and mitochondria-related GO terms. Annotation bars denote significant changes in condition (I) and in sex (II—effect in sex MR; III—effect in sex AS). Proteins described in the text are labeled. (B) Combined results (−log10-transformed P-value) of GO term enrichment analysis in AS and MR versus CON reduced to representative terms. Bars are shown in the direction of regulation they were found in. (C) Summary of metabolic processes derived from enrichment results and selected key proteins. Colored arrows denote direction of change and the condition the change was found in. (D) Median log2 of LFQ intensity of all proteins with mitochondrial annotation. P-values are calculated using a Wilcoxon rank test. (E) Comparison of ejection fraction (in %) as measured by magnetic resonance imaging in AS and MR. P-values are calculated using a Wilcoxon rank test with two samples (AS versus MR, denoted by brackets) and one sample against the reference mean (no bracket). The reference…

• 

  Figure S5.. Proteins belonging to metabolism. Log2-transformed LFQ intensity for single proteins belonging to the metabolism. All proteins shown are mentioned in the main text and are significant in at least one comparison. Full protein IDs and P-values for all comparisons are given in Table S1. AS, aortic valve stenosis; CON, healthy control hearts; MR, mitral valve regurgitation.