Decreased expression of ErbB2 on left ventricular epicardial cells in patients with diabetes mellitus

Abstract

We investigated the cell surface expression of ErbB receptors on left ventricular (LV) epicardial endothelial cells and CD105⁺ cells obtained from cardiac biopsies of patients undergoing coronary artery bypass grafting surgery (CABG). Endothelial cells and CD105⁺ non-endothelial cells were freshly isolated from LV epicardial biopsies obtained from 15 subjects with diabetes mellitus (DM) and 8 controls. The expression of ErbB receptors was examined using flow cytometry. We found that diabetes mellitus (DM) and high levels of hemoglobin A1C are associated with reduced expression of ErbB2. To determine if the expression of ErbB2 receptors is regulated by glucose levels, we examined the effect of high Glucose in human microvascular endothelial cells (HMEC-1) and CD105⁺ non-endothelial cells, using a novel flow cytometric approach to simultaneously determine the total level, cell surface expression, and phosphorylation of ErbB2. Incubation of cells in the presence of 25 mM d-glucose resulted in decreased cell surface but not total levels of ErbB2. The level of ErbB2 at the cell surface is controlled by disintegrin and metalloproteinase domain-containing protein 10 (ADAM10) that is expressed on LV epicardial cells. Inhibition of ADAM10 prevented the high glucose-dependent decrease in the cell surface expression of ErbB2. We suggest that high Glucose depresses ErbB receptor signaling in endothelial cells and cardiac progenitor cells via the promotion of ADAM10-dependent cleavage of ErbB2 at the cell surface, thus contributing to vascular dysfunction and adverse remodeling seen in diabetic patients.

Keywords: ADAM10; Diabetes mellitus; ErbB receptors; Hemoglobin A1C; LV epicardial cells.

Figure 1.. Endothelial cells and CD105⁺CD31⁻ cells are characterized by a comparable level of ErbB receptors.

A. Flow cytometric strategy to determine the expression of ErbB receptors: after enzymatic digestion, viable (left to right: first plot), singlet (second plot), CD45 negative (non-immune, third) cells were analyzed for markers of endothelial cells (CD31, EC) and CD105⁺CD31⁻ non-endothelial cells (CD105+ cells, fourth plot). The level of ErbB receptors was determined separately in subpopulations of endothelial cells and CD105 cells. B. Graphical representation of mean values of cardiac cell subpopulations obtained from epicardial biopsies. Total number of cells was 4.0 ± 0.5 x 10³ cell/mg tissue. The average weight of LV epicardial biopsy was 26 mg. n=23. C. No statistical differences were found in the cell surface expression of ErbB receptors between endothelial cells (EC) and CD105⁺CD31⁻ cells (CD105). The expression of ErbB receptors was determined after subtraction of the mean fluorescence intensity (MFI) of isotype-matched controls from the MFI of specific antibodies and represented as a delta mean fluorescence intensity (ΔMFI); n=23, Mann-Whitney test.

Figure 2.. Cell surface expression of ErbB2 is inversely correlated with DM and HbA1c level.

A. Heatmap correlation matrix of ErbB receptors expression, subject demographics, and inflammatory markers. Color represents the Spearman correlation coefficient. The value inside each box represents the P-value. B-C. ErbB1 and ErbB2 expression are decreased on endothelial cells in DM (n=15) subjects compared to non-DM (n=8). Mann-Whitney. D-F. Association of ErbB1 (D) and ErbB2 (E-F) cell surface expression on endothelial cells (D-E) and CD105⁺ cells (F) and HbA1c; n=23, Spearman correlation.

Figure 3.. High glucose decreases the ErbB2 expression at the cell surface.

A.Representative flow cytometric plots showing the total (after permeabilization) and cell surface (without permeabilization) levels of ErbB1 (middle plot) and ErbB2 (right plot) in HMEC-1 cells. Isotype matched antibodies (left plot) were used to define negative gates. B-C. Graphical representation of total and cell surface levels of ErbB1 (B) and ErbB2 (C) on HMEC-1 cells incubated in the presence of 5 mM (white bars) or 25 mM (black bars) D-glucose for the indicated time; n=9, two-way ANOVA, Tukey’s multiple comparisons test. D. Flow cytometric histograms demonstrating the effect of 30 ng/ml EGF alone (blue histogram) and combination of EGF and 300 ng/ml ErbB2 inhibitor, TAK-165 (mubritinib, red histogram) on phosphorylation of ErbB2 in HMEC-1 cells. E.Graphical representation of data from flow cytometric analysis of ErbB2 phosphorylation; n=5, one-way ANOVA. F. HMEC-1 cells were incubated in the presence of 5 mM or 25 mM D-glucose for 72 hrs and stimulated with 30 ng/ml of EGF for an hour. n=5, two-way ANOVA, Tukey’s multiple comparisons test. G. Cell surface expression of ErbB receptors on epicardial CD105 highly proliferative cells incubated in the presence of 5 mM or 25 mM D-glucose for 72 hrs, n=6, Unpaired t test. H-I. Effect of ErbB receptors ligands, 30 ng/ml EGF (H) and 30 ng/ml NRG-1 (I) on ErbB2 phosphorylation in epicardial CD105 highly proliferative cells; n=5, one-way ANOVA, Tukey’s multiple comparisons test.

Figure 4.. Inhibition of ADAM10 prevents the high glucose-dependent decrease in cell surface ErbB2 expression.

A. Expression of ADAM10 on HMEC-1 and CD105⁺ cells. B. Relationship between the cell surface expression of ADAM10 and ErbB2 in CD105⁺ cells from CABG patients; n=10. Pearson correlation analysis. C. GI 254023X (3 μM), an ADAM10 inhibitor, rescues ErbB2 expression at the cell surface of HMEC-1 and LV epicardial CD105+ cells incubated in the presence of high glucose; n=5, two-way ANOVA, Tukey’s multiple comparisons test. D. Representative plots showing the effect of 3 μM GI 254023X on 30 ng/ml EGF-induced ErbB2 phosphorylation in HMEC-1 cells incubated in medium containing 5 mM (left) or 25 mM (right) D-glucose. Black histogram – vehicle, red – EGF, blue - combination of EGF and GI 254023X. E. Graphical representation of flow cytometry data on the effect of GI 254023X on EGF-induced ErbB2 phosphorylation in HMEC-1. n=6, two-way ANOVA, Tukey’s multiple comparisons test.