Influence of Shear Stress, Inflammation and BRD4 Inhibition on Human Endothelial Cells: A Holistic Proteomic Approach

Abstract

Atherosclerosis is an important risk factor in the development of cardiovascular diseases. In addition to increased plasma lipid concentrations, irregular/oscillatory shear stress and inflammatory processes trigger atherosclerosis. Inhibitors of the transcription modulatory bromo- and extra-terminal domain (BET) protein family (BETi) could offer a possible therapeutic approach due to their epigenetic mechanism and anti-inflammatory properties. In this study, the influence of laminar shear stress, inflammation and BETi treatment on human endothelial cells was investigated using global protein expression profiling by ion mobility separation-enhanced data independent acquisition mass spectrometry (IMS-DIA-MS). For this purpose, primary human umbilical cord derived vascular endothelial cells were treated with TNFα to mimic inflammation and exposed to laminar shear stress in the presence or absence of the BRD4 inhibitor JQ1. IMS-DIA-MS detected over 4037 proteins expressed in endothelial cells. Inflammation, shear stress and BETi led to pronounced changes in protein expression patterns with JQ1 having the greatest effect. To our knowledge, this is the first proteomics study on primary endothelial cells, which provides an extensive database for the effects of shear stress, inflammation and BETi on the endothelial proteome.

Keywords: BET Inhibitor; BRD4; DIA-MS; HUVEC; JQ1; atherosclerosis; endothelial; proteomic; shear stress.

Figure 1

Experimental Design. Eight different treatment groups for human umbilical cord derived vascular endothelial cells (HUVEC) were defined with four biological replicates each. Four groups were subjected to 25 dyn/cm² laminar shear stress (SS) for 24 h. Two were treated as controls, two were exposed to JQ1 (inhibitor of the transcription modulatory bromo- and extra-terminal domain protein BRD4) for the first 20 h, two were exposed to 25 ng/mL TNFα for the last 4 h, two were exposed to 20 h of 500 nM JQ1 and subsequently 4 h of 25 ng/mL TNFα. Proteins were extracted, endoproteolytically digested and analyzed by IMS-DIA-MS, detecting 4037 protein groups. After consistency filtering, 3316 protein groups were submitted to further statistical and functional analysis.

Figure 2

Venn diagrams, principal component analysis and a heatmap of discovered statistically significant differential proteins comparing groups. (A) Treatment groups were compared to their respective control with or without shear stress (SS), numbers of distinctive or overlapping significantly differential proteins are annotated. A comparison of all treatment groups with SS compared to all groups without SS was also performed, numbers of distinctive or overlapping significantly differential proteins are annotated. (B) Principal component analysis (PCA) of all treatment groups was performed. The first 2 principal components are shown. The treatment groups show distinct clustering, largely dependent on the application of JQ1. (C) Heat map of the first 200 differentially expressed proteins discovered by ANOVA-like testing over all treatment groups. The treatment groups separate distinctively according to the received treatment. While JQ1 has the greatest effect, some proteins are strongly affected by the other treatment modalities as well. The blue box marks a group of proteins that separate well between all treatment groups.

Figure 3

Top 16 significantly differentially expressed proteins between all treatment groups. For each of the proteins the boxplots show protein expression on a log2−scale across all treatment groups. ANOVA-like testing was performed. Na⁺/H⁺ exchange regulatory cofactor 2 (SLC9A3R2), ferritin heavy chain (FTH1), fibronectin (FN1), ferritin light chain (FTL), eukaryotic translation initiation factor 5 (EIF5), heme oxygenase 1 (HMOX1), prolyl 4-hydroxylase subunit alpha-1 (P4HA1), neuropilin-1 (NRP1), tumor necrosis factor ligand superfamily member 4 (TNFSF4), hemoglobin subunit alpha (HBA1), serpin H1 (SERPINH1), receptor-type tyrosine-protein phosphatase beta (PTPRB), probable glutathione peroxidase 8 (GPX8), Ras-related protein Rab-5B (RAB5B), sorbin and SH3 domain-containing protein 2 (SORBS2), sequestosome-1 (SQSTM1).

Figure 4

Volcano plot of all tested proteins under SS or JQ1. (A) The volcano plot shows for all proteins the results of the contrast test between control with and without shear stress (SS). The x-axis shows the effect size (as log2 fold change), the y-axis shows significance (as −log10 (p−value)). The significant differentially expressed proteins are marked by color. Red proteins are upregulated in SS and blue proteins are downregulated. (B) The volcano plot shows the differential proteins in the contrast test between control and the JQ1 treatment group without SS. The significant differentially expressed proteins are marked by color. Red proteins are upregulated in JQ1, and blue proteins are downregulated.

Figure 5

Volcano plot of all tested proteins under JQ1 + TNFα and SS compared to TNFα and SS and heat maps of the results from a gene set variation analysis using GO-terms of the treatment groups. (A) The volcano plot shows for all proteins the results of the contrast test between the TNFα treatment group under shear stress (SS) and the TNFα + JQ1 treatment group under SS. The x-axis shows the effect size (as log2 fold change), the y-axis shows significance (as -log10 (p-value)). The significant differentially expressed proteins are marked by color. Red proteins are upregulated in TNFα + JQ1 and blue proteins are downregulated. (B) The central part of the heat map displays the gene set variation scores of the top 15 overrepresented gene ontology (GO) terms comparing control with JQ1 treatment without SS. The GO terms (rows) are ordered according to significance (the right panel shows p-values). The samples (columns) are ordered according to hierarchical clustering using the gene set variation scores. The top panel shows the clustering via a dendrogram, together with sample metadata. The groups separate well. The three significant GO terms “regulated exocytosis”, “response to wounding” and “cell activation” are downregulated under JQ1. (C) Comparing TNFα with TNFα + JQ1 under SS the top 15 significantly influenced GO terms are related to cell cycle, transcription, and catabolic state are enriched. The groups separate well. (D) Comparing all treatment groups together with and without SS, the enrichment of the respective GO-terms is most pronounced in groups treated with TNFα and JQ1 as well as TNFα irrespective of SS. These groups separate well. The control groups and the groups treated just with JQ1 show less distinctive enrichment.