OCT (> 99% GC purity; Sigma-Aldrich, Saint Louis, MO, USA) was dissolved in hot ethanol, diluted with fetal bovine Serum (FBS)-free Dulbecco’s Modified Eagle Medium (DMEM) (Thermo Fisher Scientific, MA, USA), aliquoted and stored at -20°C. Purified Lipopolysaccharide (LPS, γ-irradiated, BioXtra, suitable for cell culture; Sigma-Aldrich) extracted from Escherichia coli 026:B6 was dissolved in sterile distilled water, aliquoted and stored at -20°C. Cells were counted with the Counting Kit-8 (Dojindo Molecular Technologies, Rockville, MD, USA). Cytotrace™ Green CMFDA dye, fluorescein isothiocyanate (FITC)-dextran and Image-iT™ Fixative Solutions and goat serum were purchased from Thermo Fisher Scientific. Fluor™ 568 carboxylic acid, succinyl ester was purchased from Molecular Probes (OR, USA). Total RNA was extracted using a Qiagen RNeasy^® Mini Kit (Qiagen, Hilden, Germany). Both iScript cDNA Synthesis Kit and iTaq Universal Probes Supermix were from Bio-Rad Laboratories (CA, USA). All antibodies used in this study are listed in Table 1.

Primary human aortic endothelial cells (HAECs) were purchased from ScienCell Research Laboratories (San Diego, CA, USA) and grown in Endothelial Cell Medium (ECM) (ScienCell Research Lab) supplemented with 5% FBS (Thermo Fisher Scientific), 1% of Endothelial Cell Growth Supplement and 1% of penicillin/streptomycin solution at 37°C in a humidified atmosphere of 95% air and 5% CO₂. The ECM was changed every other day until the cells reached confluence. Only HAECs at passages between 4–5 were used for all experiments (26, 27).

HAECs in the logarithmic growth phase were disassociated with trypsinization, centrifuged, and seeded into 96-well flat-bottom plates at a density of 5 × 10⁴ cells/mL. The cells were cultured in endothelial growth medium and incubated overnight at 37°C in a humidified atmosphere of 5% CO₂. To achieve cell cycle synchronization at the G0/G1 phase, the medium was replaced with serum-free DMEM and the cells were incubated for an additional overnight period. Following synchronization, cells were treated with various concentrations of OCT prepared in DMEM supplemented with 2% lipoprotein-depleted FBS (LD-FBS) and maintained at 37°C with 5% CO₂ for 48 h. LD-FBS was prepared from commercial FBS, using a traditional lipoprotein separation procedure (28). Briefly, FBS was adjusted to a density of 1.21 g/mL with potassium Bromide (KBr, Sigma-Aldrich) and subjected to ultracentrifugation at 330,000 × g for 48 h at 4°C. The upper lipoprotein-enriched fraction was discarded, and the bottom fraction was extensively dialyzed against PBS at 4°C to remove residual KBr and stored at -20 °C. After treatment, the medium in each well was replaced with 100 μL of fresh medium containing 10 μL of CCK-8 reagent (Dojindo) and incubated for 4 h at 37°C. Absorbance was measured at 420 nm using a microplate reader (Bio-Tek Synergy H1, Vermont, USA). Cell viability in the treated groups was expressed relative to the untreated control group. Data represent the mean from four independent experiments, with each experiment performed at least three times.

HAECs were seeded into 12-well plates at a density of 5.0 × 10⁴ cells/mL and incubated overnight at 37°C in a humidified atmosphere containing 5% CO₂. The following day, HAECs were treated with varying concentrations of OCT (0, 1.25, 2.5, or 5μM) and incubated overnight under the same conditions. To induce inflammation, HAECs were stimulated with 100 ng/mL of LPS in 0.5% BSA-DMEM for 1, 2 or 4 h at 37°C. Control group was treated with just the vehicle. In parallel, THP-1 monocytes (ATCC, Manassas, VA, USA) were labeled with Cytotrace™ Green CMFDA dye (29). Briefly, THP-1 cells were collected by centrifugation and resuspended at 5 × 10⁵ cells per tube in 500 μL of working dye solution prepared in serum-free DMEM medium. Dye solution freshly prepared was added to cells and incubated at 37°C in a light-proof box for 30 min. After incubation, the labeled THP-1 cells were washed twice with pre-warmed DMEM to remove residual unbound dye. Labeled THP-1 cells were then resuspended in pre-warmed growth medium, added to the LPS-stimulated HAECs monolayers to achieve a final concentration of 1 × 10⁵ cells per well, and co-incubated for 1 h at 37°C. Non-adherent labeled THP-1 cells were removed by washing the wells twice with pre-warmed DMEM. The remaining adherent cells were fixed with Image-iT™ Fixative Solutions in phosphate-buffered solution (Thermo Fisher Scientific) for 15 min at room temperature. Wells were then washed, covered with PBS, and imaged using a Leica DMI 400B fluorescence microscope (Leica Microsystems, Wetzlar, Germany). The number of adherent THP-1 cells per field was quantified using ImageJ software (NIH, Bethesda, MD, USA).

HAECs were seeded in 6-well plates with 5 x 10⁵ cells per well and incubated at 37°C in a humidified atmosphere containing 5% CO₂ overnight. The OCT group were pretreated at 2.5 µM in ECM overnight and the other two groups incubated with ECM for the same time. All the HAECs were washed with prewarmed PBS and then treated with LPS at 100 ng/mL in DMEM supplemented with 0.5% BSA for 4 h at 37°C. Total RNA was extracted with a Qiagen RNeasy^® Mini Kit, according to the manufacturer’s instructions. Total RNA concentration was measured, using a Nanodrop (Thermo Fisher Scientific). To quantitate expression of selected genes by RT-qPCR, complementary DNA was first synthesized from 1000 ng of total RNA, using iScript cDNA Synthesis Kit (Bio-Rad). mRNA levels of proinflammatory cytokines (IL-6, IL-8, TNF-α, IL-1β) and chemokine (MCP-1), adhesion molecules (E-selectin, P-selectin, ICAM-1,VCAM-1), and cytoskeleton protein (VE-Cadherin, β-Catenin, and ZO-1) were determined by RT-qPCR with TaqMan Gene Expression Assays and iTaq Universal Probes Supermix (Bio-Rad),. GAPDH or 18S were used as an internal control. The primer kits used for RT-qPCR can be found in Table 2.

Libraries were constructed from 500–800 ng of total RNA, using the Illumina’s TruSeq Stranded Total RNA kit with Ribo-Zero (Illumina, San Diego, CA, USA). The fragment size of RNAseq libraries was verified using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and concentrations were determined using Qubit instrument (Thermo Fisher scientific). The libraries were loaded onto the Illumina NovaseqX (Illumina, San Diego, CA, USA) for 2 x 100 bp paired end read sequencing. The FASTQ files were generated using the bcl2fastq software. A total of ~2 million paired-end reads were generated with a base call quality of ≥Q30. Raw sequence reads in FASTQ format were trimmed using Trimmomatic (v0.39), and the filtered reads were aligned to the human reference genome (hg38) using STAR (v2.7.11b). Gene quantification was performed with RNA-Seq by Expectation-Maximization (RSEM), generating transcript-per-million (TPM) and fragments-per-kilobase-per-million mapped reads (FPKM). Differential gene expression (DGE) analysis was conducted using DESeq2. Pathway and functional enrichment analyses were performed using Ingenuity Pathway Analysis (IPA; QIAGEN, Venlo, Netherlands), using the MSigDB Hallmark gene sets to identify significantly enriched biological processes and canonical pathways.

The production of cytokines in cell culture supernatants of HAECs in response to LPS stimulation was quantified using enzyme-linked immunosorbent assay (ELISA). Interleukin-6 (IL-6), interleukin-8 (IL-8), and monocyte chemoattractant protein-1 (MCP-1) were measured using uncoated ELISA kits (Thermo Fisher Scientific). Tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β) levels were assessed using precoated ELISA kits (R&D Systems, MN, USA). HAECs were seeded into 12-well plates at a density of 5.0 × 10⁴ cells/mL and incubated overnight at 37°C in a humidified atmosphere containing 5% CO₂. The OCT group were pretreated at 2.5 µM in ECM overnight. All the cells were washed with prewarmed PBS and then treated with LPS at 100 ng/mL in DMEM supplemented with 0.5% BSA for 4 h at 37°C. The medium was collected and measured with the ELISA kits. The color signal was measured microplate reader (Bio-Tek Synergy H1, Vermont, USA).

Subconfluent HAECs grown in 6-well plates after treatment with OCT were stimulated with 100 ng/mL LPS for 4 h at 37°C. Cells were then washed once with prewarmed phosphate-buffered saline (PBS), then detached using enzyme-free Cell Dissociation Buffer (Thermo Fisher Scientific). Detached cells were collected and washed with fluorescence-activated cell sorting (FACS) buffer, consisting of 5 mmol/L EDTA, 0.1% sodium azide (NaN₃), and 1% fetal calf serum (Thermofisher Scientific) in Dulbecco’s PBS. Cells were incubated on ice for 1 h with fluorescently labeled monoclonal antibodies specific for human E-selectin, vascular cell adhesion molecule-1 (VCAM-1), and intercellular adhesion molecule-1 (ICAM-1) (Biolegend, USA) in a dark room. After antibody incubation, cells were washed twice with cold FACS buffer and analyzed for protein expression using flow cytometry in a BD LSRFortessa (BD Biosciences, Franklin Lakes, NJ, USA). Typically, the number of counted events was at least 10,000. Fluorescence intensities were recorded and analyzed to assess surface expression of adhesion molecules with FlowJo (BD Bioscience, OR, USA).

To assess endothelial monolayer integrity and permeability, cells were grown in 24-well Transwell^® inserts with 0.4 μm pore polyester membranes pre-coated with Matrigel (Thermo Fisher Scientific) (30, 31). In brief, 2 × 10⁵ HAECs per 100 μL in ECM were carefully seeded into the upper chamber of each Transwell^® insert, with 1.5 mL of ECM in the lower chamber, and then incubated at 37°C overnight. For the OCT group, cells were treated with 2.5 μM OCT in ECM overnight. After cells were treated with 100 ng/mL LPS in DMEM with 0.5% BSA for 4 h, 150 μL of DMEM containing 10 μg/mL fluorescein isothiocyanate (FITC)-dextran (40 kDa) was added to the upper chamber of each insert. Inserts were transferred to a new 24-well plate receiver, with each lower chamber containing 1 mL of fresh DMEM. The plate was incubated at room temperature in the dark to allow diffusion of FITC-dextran through the cell monolayer. At 10, 20 and 30 min post-incubation, 100 μL of medium was collected from each receiver well and placed into a black opaque 96-well plate for fluorescence analysis. Fluorescence intensity of FITC-dextran in the collected samples was measured using a microplate spectrofluorometer (excitation at 485 nm; emission at 535 nm) (Bio-Tek Synergy H1, Vermont, USA).

A similar Transwell^® insert method was also used to measure the effects of OCT on LDL leakage after HAECs were treated by LPS. Fluorescently labeled LDL was prepared as previously described (32). Briefly, Human LDL (d: 1.019 – 1.064 g/mL) was isolated from plasma from healthy donors by preparative sequential KBr density ultracentrifugation (330,000 x g) followed by extensive dialysis at 4°C against PBS to remove KBr (28). Pure LDL (6 mg of protein) was then incubated with Alexa Fluor™ 568 carboxylic acid, succinyl ester (Molecular Probes, Eugene, OR) for 1 h at room temperature in the dark. Free, unbound dye was separated from labeled LDL by preparative fast protein liquid chromatography (FPLC) using a Hi-Trap Desalting column (GE Healthcare, Chicago, IL, USA). To assess HAECs integrity, 150 μL of DMEM containing 100 μg/mL Alexa Fluor™ 568-labelled LDL, was added to the upper chamber of each insert after LPS stimulation. The samples were collected as described above and measured in a microplate spectrofluorometer (excitation at 578 nm; emission at 602 nm).

For confocal imaging, 5 x 10⁴ HAECs were cultured in 35 mm Dish with 20 mm glass bottom Well (Cellvis, CA, USA). Following LPS treatment, cells were washed once with prewarmed PBS and fixed with Image-iT™ Fixative Solutions in PBS for 15 min at room temperature, then permeabilized with 0.1% Triton X-100 in PBS for 10 min. After blocking with 10% heated goat serum (Thermo Fisher Scientific) in PBS for 1 h, cells were incubated with primary antibodies targeting the proteins of interest (Table 1) overnight at 4°C. On the next day, cells were washed and incubated with species-appropriate fluorescent secondary antibodies (1:2000 dilution) and phalloidin labeling probes (Alexa Fluor™ 594 or Alexa Fluor™ 488, 1:400 dilution) for 1 h at room temperature in the light-proof dark box. Nuclei were counterstained with To-pro-3 (Thermo Fisher Scientific) for 10 min and examined using a Leica TCS SP5 confocal fluorescent microscope (Leica Microsystems, Wetzlar, Germany). Image acquisition settings were kept constant across all experimental conditions. Quantification of Fluorescence images were analyzed and quantified using ImageJ software.

Statistical analyses among the three groups were performed using GraphPad Prism version 9.5.1 (GraphPad Software Inc. La Jolla, CA, USA). Data are expressed as the mean ± SD, and analyzed by a one-way ANOVA followed by a Tukey post hoc test where appropriate. p < 0.05 was considered statistically significant. Each experiment was repeated at least 3 times under the same conditions.

OCT treatment of HAECs at concentrations ranging from 0.5 to 10 μM did not result in any apparent cytotoxicity (Supplementary Figure S1A). We also tested different doses of LPS ranging from 10 to 10,000 ng/mL for 2, 4, and 6 h to evaluate HAEC viability. LPS at 100 ng/mL for 4 h only elicited a mild inflammatory response without inducing significant cytotoxicity (Supplementary Figure S1B). We, therefore, used 100 ng/mL of LPS for 4 h to examine the possible protective effect of OCT on HAECs in subsequent studies. In a time-course study, LPS stimulation led to a time-dependent increase in THP-1 cells adhesion to monolayer of HAECs (Figure 1). Adhesion of THP1 to HAECs stimulatedby LPS was significantly increased by 6.9-fold at 1 h, 15.6-fold at 2 h, and 25.3-fold at 4 h compared to unstimulated control group. However, when cells were pretreated with OCT overnight, LPS-induced THP-1 adhesion to HAECs was markedly reduced in a dose- and time-dependent manner (p < 0.05). 1.25 μM of OCT pretreatment reduced LPS-stimulated monocyte adhesion by 71% at 2 h (p < 0.05), and by 77% at 4 h (p < 0.05). 2,5 μM of OCT pretreatment reduced LPS-stimulated monocyte adhesion by 50% at 1 h (p < 0.05), 85% at 2 h (p<0.05), and 91% at 4 h (p < 0.05). 5 μM of OCT treatment significantly reduced LPS-stimulated monocyte adhesion by 72% at 1 h (p < 0.05), 85% at 2 h (p < 0.05), and 91% at 4 h (p < 0.05). Both 2.5 µM and 5 µM of OCT achieved maximal reduction of monocyte adhesion, at concentrations that blocked ~90% of THP-1 adhesion to HAECs after 4 h of LPS stimulation. We, therefore, selected 2.5 µM of OCT and 4 h treatment of LPS for subsequent experiments.

To evaluate the anti-inflammatory effect of the OCT in HAECs, the mRNA expression and secreted pro-inflammatory cytokines were analyzed using RT-qPCR and ELISA, respectively. LPS treatment significantly increased mRNA levels of IL-1β, IL-6, IL-8, MCP-1, and TNF-α by 5 to 10-fold compared with the untreated control group (p < 0.05). However, OCT pretreatment reduced the mRNA expression of all these proinflammatory cytokines to near basal levels in HAECs stimulated with LPS (p < 0.05) (Figure 2A). ELISA analysis further confirmed the anti-inflammatory effect of OCT. LPS stimulation markedly increased the concentrations of IL-1β, IL-6, IL-8, MCP-1, and TNF-α in the cell culture supernatant compared with control (p < 0.05), whereas OCT pretreatment significantly decreased the release of these cytokines compared with LPS treatment alone (p < 0.05) (Figure 2C). To investigate the potential mechanism by which OCT influences LPS-induced inflammation, we also evaluated mRNA expression of several key genes in the TLR4-mediated NF-κB signaling pathway. LPS upregulated TLR4, MYD88, TIRAP, TRAF6, and IRAK1 compared with control (p < 0.05), but this response was suppressed by OCT pretreatment (p < 0.05) (Figure 2B).

To explore possible mechanism underlying decreased monocyte adhesion to LPS-stimulated HAECs due to OCT pretreatment, we analyzed the expressions of several adhesion molecules. LPS stimulation significantly increased the mRNA levels of VCAM-1, ICAM-1, SELE, and SELP compared with the control group (p < 0.05), but OCT pretreatment significantly attenuated their upregulation from LPS treatment (p < 0.05) (Figure 3A). Flow cytometry analysis confirmed that LPS markedly increased the surface expression of VCAM-1, ICAM-1 and E-selectin (p < 0.05), whereas OCT-treated HAECs largely reversed the LPS-induced expression increase and maintained the protein levels of these adhesion proteins near control levels (p < 0.05) (Figure 3B). In addition, by confocal microscopy LPS treatment resulted in strong fluorescence signals corresponding to all four adhesion molecules around endothelial cells, as is evidenced by the significant shift by 3 to 10-fold increase in the mean fluorescent intensity relative to non-stimulated controls (p < 0.05). In contrast, OCT-pretreated cells displayed a markedly weaker fluorescence signal for these adhesion proteins (p < 0.05) (Figure 3C; Supplementary Figures S2-S5).

To examine the protective effect of the OCT on LPS-induced endothelial monolayer barrier disruption, we performed two different assays of endothelial function. LPS stimulation significantly increased the passage of FITC-dextran across the endothelial monolayer at all measured time points compared with control (p < 0.05). Pretreatment with the OCT, however, markedly reduced FITC-dextran infiltration through the monolayer HAECs at each time point compared with LPS treatment alone (p < 0.05) (Figure 4A). In a fluorescent labelled LDL leakage assay, LPS stimulation enhanced the leakage of LDL particles across the endothelial monolayer by approximately 3-fold compared with control (p < 0.05). OCT pretreatment, however, significantly attenuated LPS-induced LDL leakage at all the tested time points (p < 0.05), thus maintaining the endotheial barrier function similar to the control cells (Figure 4B).

To further explore the influence of OCT on cytoskeletal and junctional proteins in HAECs under inflammatory conditions, we investigated the changes of endothelial adherens and tight junctions. RT-qPCR analysis showed that LPS downregulated the mRNA expression of junctional proteins, such as CDH5, CTNNB1, and TJP1, compared with control (p < 0.05) (Figure 4C), but the mRNA expressions of these genes was preserved after OCT pretreatment. Confocal imaging analysis further corroborated these findings. LPS-treated HAECs exhibited remarkably reduced fluorescence intensities for VE-cadherin, β-catenin, and ZO-1 compared with control (p < 0.05) (Figure 4D; Supplementary Figures S6-S8). LPS stimulation also resulted in ruptures around the cell border and the fluorescently labeled proteins were often redistributed toward the perinuclear region. In contrast, pretreatment with OCT significantly counteracted these LPS-induced alterations, preserving cytoskeletal organization and maintaining junctional protein expression near basal levels. Notably, F-actin fibers remained well organized and smoothly extended in a similar pattern to those of control cells. In addition, co-localization analysis of VE-cadherin and β-catenin further showed that LPS treatment caused substantial rupture of cell–cell contacts, whereas OCT-treated cells retained intact and continuous cell borders after LPS stimulation (Figure 4E).

Upon LPS stimulation, mRNA expressions of CTTN, VCL and TALIN were significantly upregulated compared with control (p < 0.05), but OCT pretreatment effectively suppressed their expression to levels observed in control cells (Figure 5A). Immunofluorescent imaging revealed that OCT pretreatment consistently decreased protein levels of CORTACIN, VINCULIN, and TALIN, and prevented their redistribution compared with LPS group (p < 0.05) (Figure 5B). In normal control cells, cortactin was primarily distributed evenly around nuclei in the perinuclear cytoplasm. Upon LPS stimulation, however, cortactin became strongly polarized toward the cell periphery, accumulating within thin, finger-like protrusions that extended into the extracellular matrix. Most cortactin colocalized with actin, redistributed along the cell border, and contributed to the formation of protrusive structures that guided directional cell migration (Supplementary Figure S9). Following LPS stimulation, both vinculin and talin exhibited dense clustering along the cell periphery, with especially sharp enrichment at protrusive edges where they formed distinct, match-head–like focal accumulations (Supplementary Figures S10, S11). Further analysis using co-localization of Talin and Vinculin immunofluorescence labeling revealed that talin and vinculin co-localized at the tips of protruding actin filaments within lamellipodial and filopodial extensions at the periphery of LPS-treated endothelial cells. In contrast, OCT-pretreated cells exhibited markedly reduced talin and vinculin fluorescence intensities, fewer membrane protrusions, and a more stable cellular morphology resembling control cells (Figure 5C).

To explore the molecular mechanism underlying the cytoprotective effect of OCT in LPS-treated HAECs, RNA-seq analysis was conducted to compare the gene expression profiles of LPS-treated HAECs with or without OCT pretreatment. Principle Component Analysis (PCA) were employed to discern the impact of OCT treatment on gene expression levels. PCA analysis (Figure 6A) revealed three well-distinguished groups, indicating different gene expression patterns between the control, LPS and LPS plus OCT groups. Volcano plots (Figure 6B) of differentially expressed genes (DEGs) are shown for the LPS treatment compared with control (Figure 6B, left panel) and LPS plus OCT treatment compared with LPS alone (Figure 6B, right panel), respectively. LPS treatment resulted in 5117 DEGs, with 3010 genes up-regulated and 2107 genes down-regulated. In LPS-stimulated cells, OCT treatment resulted in 497 DEGs, with 117 genes up-regulated and 380 gene down-regulated. Pathway enrichment analysis further clarified the functional roles of DEGs (Figure 6C). LPS stimulation upregulated pathways related to cytokine and chemokine activity, several basic metabolic processes, and cell integrity pathways, indicating that LPS, at least in part, damages endothelial cells by triggering inflammation (Figure 6C, left panel). OCT treatment, however, reversed most of the DEGs induced by LPS in endothelial cells, primarily by downregulating inflammatory responses, including TNFα/NF-κB signaling, interferon-γ signaling, and IL2-STAT5 signaling pathways. OCT pretreatment also upregulated several pathways involved in lipid metabolism, including bile acid metabolism and fatty acid metabolism (Figure 6C, right panel).