**Background:** Endothelial cells (ECs) line the inner wall of blood and lymphatic vessels and play critical roles in vascular homeostasis, angiogenesis, and various pathophysiological processes such as atherosclerosis, lymphedema, and cancer metastasis. Circular RNAs (circRNAs) are a class of covalently closed RNA molecules that are stable, cell-type-specific, and can regulate gene expression. While circRNAs have been studied in many cell types, a systematic characterization of circRNA signatures in ECs has been lacking. This study aimed to explore the circRNA expression landscape in different EC types to identify potential new molecular signatures and EC-specific markers.
**Methods:** The researchers analyzed publicly available RNA sequencing data from primary cultures of human endothelial cells: Human Umbilical Vein Endothelial Cells (HUVECs), Human Artery Endothelial Cells (HUAECs), Human Coronary Artery Endothelial Cells (HCAECs), and Human Dermal Lymphatic Endothelial Cells (HDLECs). Non-endothelial controls included Smooth Muscle Cells from Umbilical Artery (SMC-UAs) and Stromal Vascular Fraction from Superficial Adipose Tissue (SAT-SVFs). CircRNAs were identified de novo using the CIRI2 algorithm, which detects backspliced junctions. Only circRNAs with 10 or more reads were included. Expression profiles were compared using Pearson correlation, variance analysis, and Venn diagrams. Validation was performed by RT-qPCR and RNase R treatment for selected circRNAs.
**Key Results:** A total of 4713 distinct circRNAs were identified across all cell types, with 2254 (47.82%) exclusive to a particular cell type. The number of detected circRNAs varied: 207 for HCAECs, 259 for SAT-SVFs, 483 for HUVECs, 706 for HDLECs, 1570 for SMC-UAs, and 2685 for HUAECs. The majority (95.04%) were exonic circRNAs. Circular RNAs represented 19.48% of total transcripts in HUAECs, 6.05% in HDLECs, 4.92% in HUVECs, and 2.96% in HCAECs. The circular-to-linear ratio (CLR) varied significantly between cell types (p < 0.05), with HCAECs showing the highest average CLR (62%) and HDLECs the lowest (14%). Pearson correlation analysis showed that circRNA expression profiles were more distinct between EC types than mRNA profiles; for example, HDLECs had 77% circRNA expression correlation with HUVECs but only 23% for matched linear RNAs. Four circRNAs were common to all ECs and absent from non-ECs: circCARD6, circPLXNA2, circCASC15, and circEPHB4. These are associated with genes involved in endothelial migration and cancer progression. RT-qPCR validated 11 out of 17 selected circRNAs (65% validation), and RNase R treatment confirmed their circular nature.
**Clinical Implications:** This study reveals that circRNAs exhibit a strikingly unique expression profile specific to individual EC types, making them superior to linear RNAs for distinguishing between endothelial cell types. The identification of EC-specific circRNAs, particularly the four common ones (circCARD6, circPLXNA2, circCASC15, circEPHB4), provides potential new biomarkers for vascular health and disease. Because ECs are in direct contact with blood and lymph, and circRNAs are stable and can be secreted in exosomes, these circRNAs could be used for non-invasive diagnosis or monitoring of vascular pathologies, cardiovascular diseases, and cancers. They may also serve as therapeutic targets; for example, circRNAs involved in cancer-related pathways could be targeted to inhibit tumor growth and metastasis. However, further research and validation are needed to establish clinical significance.