Background
The vascular endothelium consists of endothelial cells (ECs) with important biological functions, and their impairment is associated with various pathologies. ECs vary based on tissue origin and gene expression, while their functionality depends on calcium (Ca2+) signaling. In tumors, disruption of Ca2+ homeostasis after calcium electroporation (CaEP) has been shown to elicit an enhanced antitumor effect with only a minimal effect on normal tissue. The difference in response to CaEP was observed not only between cancer and normal cells but also between different endothelial cell lines. Although several vascular EC models have been developed, there is a lack of understanding regarding the molecular basis that could help explain different responses between tumor and normal tissue to CaEP. Therefore, our study aimed to determine the effect of CaEP on the established immortalized human endothelial cell lines EA.hy926 and HMEC-1 in terms of the cytoskeleton, Ca2+ kinetics and differences in gene expression involved in the regulation of Ca2+ signaling and homeostasis.
Methods
Optimization of electroporation parameters was performed to achieve the highest permeabilization of EA.hy926 and HMEC-1 cells with minimal effect on cell survival. Optimized pulse parameters (8 square-wave electric pulses, 1000 V/cm, 100 µs, 1 Hz) were used for CaEP of EA.hy926 and HMEC-1 cells in the presence of increasing Ca2+ concentrations (0 mM (control (Ctrl)), 0.5 mM, 1 mM, 2 mM and 3 mM CaCl2). The viability of cells after CaEP was determined using the Presto Blue assay, while the effect of CaEP on the cytoskeleton of EA.hy926 and HMEC-1 cells was determined by immunofluorescence staining of actin filaments (F-actin), microtubules (α-tubulin) and cell‒cell junctions (VE-cadherin). To determine the differences between EA.hy926 and HMEC-1 cells in the regulation of intracellular free Ca2+ concentration ([Ca2+]i), spectrofluorometric Ca2+ kinetic measurements were performed in cells preloaded with Fura-2-AM and exposed to ionomycin, thapsigargin, ATP, bradykinin, angiotensin II, acetylcholine, LaCl3 and GdCl3 individually or in combination. Molecular differences between EA.hy926 and HMEC-1 cells were determined through transcriptomic profiling of differentially expressed genes and molecular pathways involved in the regulation of [Ca2+]i and Ca2+ signaling via RNA sequencing (RNA-seq).
Results
In the presence of increasing Ca2+ concentrations, EA.hy926 cells exhibited higher susceptibility to CaEP with lower survival than HMEC-1 cells. The sensitivity of EA.hy926 cells to a large increase in [Ca2+]i after CaEP exposure was further confirmed by immunofluorescence staining, which showed morphologically altered structures of actin filaments and microtubules as well as cell‒cell junctions. Moreover, significantly lower mean intensities of cytoskeleton structures in treated EA.hy926 cells were observed in a time- and Ca2+ concentration-dependent manner. Fluorometric Ca2+ kinetic measurements in EC cells preloaded with Fura-2-AM showed an increase in the fluorescence (F340/F380) ratio, indicating a significant rise in [Ca2+]i in EA.hy926 cells compared with HMEC-1 cells after exposure to flow of buffer and agonists of G protein coupled receptor (GPCR)-dependent response, bradykinin and angiotensin II. In HMEC-1 cells, significantly higher changes in [Ca2+]i compared to EA.hy926 cells were observed after exposure to ionomycin, while exposure to thapsigargin, ATP and acetylcholine induced a similar response in both cell lines. ATP without the presence of Ca2+ induced a significantly higher rise in [Ca2+]i in EA.hy926 cells, suggesting that Ca2+ influx is mediated by metabotropic P2Y receptors as well as from the ER via activation of ionotropic purinergic P2X receptors. RNA-seq analysis showed a significant difference in the expression of cytoskeleton- and Ca2+-related genes between EA.hy926 and HMEC-1 cells. Among differentially expressed genes (DEGs) related to cytoskeleton ICAM2, MYH3 and PECAM1 were the top three significantly upregulated genes in EA.hy926 cells; however, most genes related to actin filaments, microtubules and VE-cadherin junctions were downregulated in EA.hy926 cells compared with HMEC-1 cells. TRPM6, CACNG7, and TRPM2 were found to be the top upregulated genes, while TRPV4, PIEZO2 and TRPV2 were the top three downregulated Ca2+-related genes in EA.hy926 cells compared to HMEC-1 cells. Among genes involved in Ca2+ influx, the EA.hy926 cell line showed significantly higher expression of ORAI2, TRPC1, TRPM2, CNGA3 and TRPM6 and significantly lower expression of TRPV4 and TRPC4 than HMEC-1 cells. KEGG analysis of the Ca2+ signaling pathway showed significant upregulation of genes related to Ca2+ import into the cytoplasm (ORAI, CACNA1A, IP3R) and significant downregulation of genes involved in Ca2+ export from the cytoplasm (NCX, MCU, and SERCA) in EA.hy926 cells compared to HMEC-1 cells.
Conclusions
Our findings show significant differences in the response to CaEP and in the regulation of [Ca2+]i between the vascular endothelial cell lines EA.hy926 and HMEC-1, which are primarily due to their distinct transcriptomic profiles. Compared to HMEC-1 cells, the EA.hy926 cell line is more susceptible and sensitive to changes in [Ca2+]i due to overexpression of Ca2+-related genes and inability to alleviate the changes in [Ca2+]i, which was confirmed by immunofluorescence staining and Ca2+ kinetic assays. In addition, our study provides a bioinformatic basis for the selection of the EC model depending on the objective of the research.