Introduction: 
The unfolded protein response of the endoplasmic reticulum (UPR^ER) is highly activated in cardiovascular diseases and aims to restore homeostasis by improving protein folding. An analogue cytoprotective mitochondrial unfolded protein response (UPR^Mito) regulated by an orthologue of the activating transcription factor 5 (ATF5) was described in model organism caenorhabditis elegans. However, in humans it is not clear if there is a distinct UPR^Mito, and if there is an activation of an UPR^Mito during stress induced by cardiovascular diseases.

The endothelium is subjected to high levels of mitochondrial stress, e.g. in atherosclerosis. Aim of our study is to provide a systematic characterization of the UPR^Mito in human coronary artery endothelial cells (HCAEC).

Methods and Results: 
HCAEC were treated with 0 to 1000 µM of Nicotinamide Riboside (NR), a NAD+ precursor that activates UPR^Mito without inducing cellular stress, for 24 hours. Viability was measured using an alamarblue^® assay. NR treatment did indeed induce no toxicity and improved cell viability at a concentration of 100µM NR. Quantitative PCR confirmed a dose-dependent upregulation of UPR^Mito markers. The mitochondrial chaperones HSP10 and HSP60 were only upregulated upon 100µM NR (HSP10: 1.3-fold, HSP60: 2.0-fold), while ATF5 is already upregulated upon lower concentrations of NR (50µM: 2.1-fold, 100µM: 2.5-fold). Additionally, we measured expression of ATF4, which is discussed to regulate an integrated stress response in mammals by activating both the UPR^Mito and UPR^ER. Intriguingly, ATF4 upregulation upon NR treatment was even more pronounced than ATF5 upregulation (e.g., 50µM: 3.8-fold). We further analyzed GTEX-Data and found ATF5 to be almost exclusively expressed in liver tissue, while ATF4 shows ubiquitous expression. Among 54 analyzed different human tissues, aortic tissue and coronary arterial tissue are among the five tissues with the highest ATF4 expression.

Bioinformatic analysis of available RNA-seq-data of human carotid atherosclerosis plaques revealed that the two transcription factors are differentially regulated. ATF4 is consistently downregulated in unstable plaque, while ATF5 is upregulated. Moreover, ATF4 is in both conditions highly more abundant than ATF5. Accordingly, in silico analyses of RNA-seq of endothelial cells incubated under atherosclerotic conditions (oxLDL + high glucose) showed a downregulation of ATF4 and upregulation of ATF5. Interestingly, genes previously described to be involved in the UPR^Mito exhibited a similar expression pattern as ATF4.

To further elucidate these findings, we will perform transcriptomic analyses of HCAEC treated with atherosclerotic stimuli (IL-1ß, oxLDL), UPR^ER inductors (Thapsigargin, Tunicamycin) and stressors of mitochondrial proteostasis (Oligomycin, MitoBloCK-6), to analyze, if ATF5- or ATF4-related pathways are dysregulated. Specific inhibitors and UPR^Mito activators will be used to investigate the significance of ATF4 and ATF5 on endothelial cell function (proliferation, migration, apoptosis, monocyte adhesion). 

Conclusion: 
The UPR^Mito is dysregulated during the pathogenesis of atherosclerosis in HCAEC. Further findings are required to elucidate, if the UPR^Mito is activated separately and specifically by ATF5 or as part of an integrated cellular stress response by ATF4. A deeper understanding of these stress responses is crucial for the identification of novel therapeutic targets in atherosclerosis.