The interleukin-1 receptor type-1 in disturbed flow-induced endothelial mesenchymal activation

Introduction Atherosclerosis is a progressive disease that develops in areas of disturbed flow (d-flow). Progressive atherosclerosis is characterized by bulky plaques rich in mesenchymal cells and high-grade inflammation that can rupture leading to sudden cardiac death or acute myocardial infarction. In response to d-flow, endothelial cells acquire a mesenchymal phenotype through endothelial-to-mesenchymal transition (EndMT). However, the signaling intermediaries that link d-flow to EndMT are incompletely understood. Methods and Results In this study we found that in human atherosclerosis, cells expressing SNAI1 (Snail 1, EndMT transcription factor) were highly expressed within the endothelial cell (EC) layer and in the pre-necrotic areas in unstable lesions, whereas stable lesions did not show any SNAI1 positive cells, suggesting a role for EndMT in lesion instability. The interleukin-1 (IL-1), which signals through the type-I IL-1 receptor (IL-1R1), has been implicated in plaque instability and linked to EndMT formation in vitro. Interestingly, we observed an association between SNAI1 and IL-1R1 within ECs in the unstable lesions. To establish the causal relationship between EndMT and IL-1R1 expression, we next examined IL-1R1 levels in our Cre-lox endothelial-specific lineage tracing mice. IL-1R1 and Snail1 were highly expressed in ECs under atheroprone compared to athero-protective areas, and oscillatory shear stress (OSS) increased IL-1R1 protein and mRNA levels in vitro. Exposure of ECs to OSS resulted in loss of their EC markers and higher induction of EndMT markers. By contrast, genetic silencing of IL-1R1 significantly reduced the expression of EndMT markers and Snail1 nuclear translocation, suggesting a direct role for IL-1R1 in d-flow-induced EndMT. In vivo, re-analysis of scRNA-seq datasets in carotid artery exposed to d-flow confirmed the IL-1R1 upregulation among EndMT population, and in our partial carotid ligation model of d-flow, endothelial cell specific IL-1R1 KO significantly reduced SNAI1 expression. Discussion Global inhibition of IL-1 signaling in atherosclerosis as a therapeutic target has recently been tested in the completed CANTOS trial, with promising results. However, the data on IL-1R1 signaling in different vascular cell-types are inconsistent. Herein, we show endothelial IL-1R1 as a novel mechanosensitive receptor that couples d-flow to IL-1 signaling in EndMT.


Introduction
Progressive unstable atherosclerosis is characterized by a thin fibrous cap, low smooth muscle and high mesenchymal cell content, with a high grade of inflammation, and is the leading cause of cardiovascular disease-related death (1). Despite decades of research on atherosclerosis, the molecular mechanism(s) of progressive unstable atheroma are only partially understood. Human studies have suggested that the cellular composition of plaques is a predictor of plaque instability (2). A sub-endothelial cell population in unstable lesions with mesenchymal features is usually associated with the incidence of atheroma in human aortas (3). Mouse studies using endothelial-specific lineage tracing and single cell RNA sequencing (scRNA-seq) systems demonstrate that endothelial cells (ECs) can undergo mesenchymal transition (EndMT) and form a substantial number of these sub-endothelial cells (4)(5)(6)(7). EndMT, which develops in areas of disturbed blood flow (d-flow) (6), has been implicated in the progression of atherosclerosis (8) where EndMT plays an important role in plaque destabilization by promoting local inflammation (8,9). Despite this, the molecular mechanism by which EndMT develops under d-flow areas remains incompletely understood.
The interleukin-1 (IL-1) signaling pathway has been shown to play a role in the severity of atherosclerosis progression and EndMT formation (10). IL-1 is therefore a suitable therapeutic target for modulation of EndMT that enhances atherosclerosis stability. The major clinically concluded CANTOS study tested the therapeutic effect of the anti-IL-1β antibody, Canakinumab, in a group of patients after myocardial infarction, and showed 31% reduction in cardiovascular and all-cause mortality in certain, but not all, groups of patients (11). While global inhibition of IL-1 signaling in this study showed some promising results, a number of patients developed off-target immunosuppression effects (12), limiting the global targeting of IL-1 in atherosclerosis. Another limitation in the field is that despite extensive studies, there is no clear principal cell type responsible for IL-1 signaling in human progressive atherosclerosis.
The term Interleukin-1 (IL-1) refers to 2 cytokines, IL-1α and IL-1β, which signal exclusively through a common receptor, IL-1 receptor type I (IL-1RI) (13). IL-1 is implicated in several aspects of plaque instability (14,15) and IL-1 signaling is linked to EndMT formation in vitro (16). Global inhibition of IL-1 in atherosclerosis has been extensively investigated (12,17). However, the data on IL-1R1 is inconsistent. In mice, either global or myeloid specific deletion of IL-1R1 reduces atherosclerosis (18), whereas cell specific deletion of IL-1R1 in vascular smooth muscle cells increases atherosclerosis and make lesions more unstable (19)(20)(21). However, in humans with coronary atherosclerotic disease it is still unclear where and at which stage of the disease IL-1R1 is expressed.
Previously we and others have shown that in human coronary atherosclerotic plaques IL-1 is predominantly expressed within the endothelium (22)(23)(24), suggesting a role for endothelial IL-1 signaling in the pathogenesis of atherosclerosis. Our published data demonstrated enhanced signaling activation of interleukin-1 receptor predominantly within the endothelium of progressive human atherosclerotic plaques (22,24). In addition, we showed that this activation is induced by d-flow in vitro and in vivo (25). Despite all of that, the signaling roles of IL-1R1 in the endothelium have yet to be extensively investigated.
Taking into account the important role of EndMT in progressive human atherosclerosis and IL-1R1 signaling roles, in this study, we directly assessed IL-1R1 expression and linked it to EndMT in human atherosclerotic plaques. Furthermore, we utilized our endothelial specific IL-1R1 KO mice to determine the endothelial roles of IL-1R1 in d-flow-induced EndMT. Our study should aid in developing future strategies for targeting the IL-1R1 within the endothelium to limit atherosclerosis progression and maintain lesion stability.
Mouse aortic endothelial cells from IL-1R1 fl/fl were isolated as previously described (25). Purity of endothelial cell isolation was assessed with double immunostaining for CD31 (Anti-CD31, abcam, ab9498) and smooth muscle actin (Anti-Actin SMA-FITC antibody CLONE 1a4, Sigma, F3777). Cell transformation was conducted in LSU Molecular-sub-core. The cells were divided into two 10 cm 2 plates and at 70% confluence the cells were treated with either Ad-CMV-GFP (VECTOR BIOLABs, cat. # 1060, 1 × 10 8 PFU/ml) as wild-type (WT) controls, or Ad-GFP-2A-iCre (VECTOR BIOLABS, cat. # 17772, 2 × 10 8 PFU/ml) as IL-1R1 knockout (IL-1R1 KO) cells. The transduction was induced for 72 h. and the GFP positive cells were sorted using flow cytometry at the LSU Research Core Facility. The efficiency of IL-1R1 deletion was assessed by measuring the IL-1R1 mRNA expression using qRT-PCR.

Oscillatory shear stress experiments
Oscillatory shear stress (OSS, 1 dynes/cm 2 with 1 s switching time; flow rate 2.38 ml/min) was induced by the Ibidi parallel plate flow system as previously described (26). The cells at a seeding density of 1 × 10 6 were plated in channel slides (µ-slide I 0.6 luer; 150 µl per slide) to allow to adhere for 2 h. in 1% (v/v) MCDB131 media. Confluent slides were placed under the flow, Perfusion Set RED, using Ibidi pump system (IBIDI USA, INC) for 24 h.
The male sex is at higher risk for atherosclerotic heart disease (27). Therefore, only male mice were used in the current study. Male mice at 8-10-weeks of age of IL-1R1 iEC−KO and their littermate controls were injected intraperitoneally with tamoxifen (Sigma, cat. # T-5648) in peanut oil (0.1 ml at concentration of 45.5 mg/kg ∼1 mg/mouse) daily for 5 consecutive days. The mice (number in the figure legends) were allowed to recover for 2 weeks before the surgery was conducted. To induce d-flow, partial carotid ligation was performed as previously shown (25,28). Briefly, the anesthesia was induced, and the frontal area of the mouse neck was cleaned and opened by a midline incision. The left carotid artery was dissected by blunt dissection and the three branches (internal carotid, external carotid, and occipital branch) were ligated with 0-6 silk sutures, leaving the superior thyroid branch intact. The skin incisions were sutured with 0-5 sutures and the mice were hydrated with 0.5% normal saline (1 ml) and analgesics were administered subcutaneously using Carprofen (5 mg/kg; Ceva, USA). 2-week post-carotid ligation, mice were euthanized and left, and right carotid arteries were collected for RNA analysis as shown (28). The inducible endothelial deletion of IL-1R1 were verified in mRNAs isolated from carotid intima using TRizol flush method and that is after tamoxifen injection, and we measured the intimal expressions of IL-1R1, CD31, SNAI1 mRNA as we previously described (25,28).

Western blotting
Gel electrophoresis was performed on denatured protein samples using the Invitrogen TM Mini Gel Tank and NuPAGE TM 4%-12%, Bis-Tris, 1.0-1.5 mm, Mini Protein Gels, as previously described (29). Samples of 25 µl protein lysates were loaded per well and allowed to run in MOPS SDS 1× running buffer for 45 min at 200 V. Wet transfer was performed in 1× Invitrogen TM Blot TM Transfer Buffer for 1 h. at 20 V. Once the proteins were transferred to the PVDF membranes (Bio-Rad, Cat. #1620177), the membranes were blocked in 5% (w/v) non-fat dry milk in 0.01% (v/v) TBST (Tween-Tris-buffered saline) for 1 h. The membranes were incubated with the desired primary antibodies (Supplementary Table SI), diluted as appropriate in 1% (w/v) BSA/PBS (Bovine serum albumin, Sigma, Cat. # A2153) overnight at 4°C with gentle agitation. The membranes were then incubated with the secondary antibodies (Peroxidase Affinity Pure Goat Anti-mouse or Anti-Rabbit IgG, Jackson ImmunoResearch, Cat. # 115-035-003) for 1 h.
Detection was performed by chemiluminescence using SuperSignal TM Chemiluminescence substrate (Cat. # 34580) and scanned using the Bio-Rad ChemiDoc TM MP Imaging System. For quantification, protein densitometry data were quantified using image j software. Fold changes in the protein expressions were measured relative to the loading control and to the static condition as previously demonstrated (25,28).

Immunohistochemistry
Human coronary artery samples were collected in the pathology department at LSU Health-Shreveport. The collection was approved by the Local LSU Health Research Ethics Committee. The samples were stored in 10% v/v formalin until processing. The tissue samples were Paraffin-embedded and sectioned at 5 µm on slides. Sections of human coronary atherosclerotic plaques were blindly stained and scored for lesion stability criteria by a pathologist at LSU Health-Shreveport.

Immunocytochemistry
For cellular immunostaining, the protocol used was adapted from (28) with slight modifications. Briefly, the cells were fixed in 3.7% (v/v) paraformaldehyde for 30 min followed by frequent rinsing in 1× PBS. The permeabilization was performed using 0.01% Triton-X100 in PBS for 10 min, and the slides were blocked in 10% horse serum/1% BSA solution for 1 h. at room temperature. Cells were incubated with primary antibodies at 4°C overnight and the primary antibodies used were anti-CD31 (abcam, ab9498, 1:100 dilution), and anti-N cadherin (abcam, ab18203, 1:100 dilution). After subsequent washing in 0.01% v/v TBST, the cells were incubated with Alexa Fluor 488 phalloidin (A12379, 1:400 dilution) mixed in 10% horse serum/1% BSA solution with secondary antibodies (1:1,000 dilution each; Donkey anti-mouse Alexa Fluor 555-A32773, Donkey anti-rabbit Alexa Fluor 647-A31573) at room temperature for 1 h. Cells were rinsed thoroughly with 1× PBS and then slides were mounted in Prolong TM Glass Antifade Mountant containing NucBlue (P36985). Images were acquired using an EVOS M5000 microscope.

Real-Time PCR
mRNA was extracted and isolated using RNeasy Mini Kit (Qiagen, Cat. # 74104) according to the manufacturer's instructions. Reverse transcription and real-time PCR were conducted as previously demonstrated (25). Ct values of the target genes were normalized upon the Ct values of the housekeeping genes (RPL13A and GAPDH). All PCR analysis was conducted in duplicates and the data are from 4 independent experiments. Primer sequences for the human target genes used as following:

Statistical analysis
Data is analyzed as mean ± standard error of the mean (SEM) using GraphPad prism software (Version 9.5, GraphPad, San Diego, CA). For multiple comparisons, data were analyzed by 1-Way ANOVA followed by Tukey's post-test to compare data with one independent variable, whereas for the multiple comparisons of data with two independent variables we used 2-Way ANOVA and Bonferroni's post-test. An unpaired student's t-test was used to compare two different datasets. Statistical significance was achieved when p < 0.05.

IL-1R1 expression is associated with EndMT in human unstable atheroma
Previously we and others have shown that IL-1 in human coronary atherosclerotic plaques is predominantly expressed within the endothelium (22)(23)(24), demonstrating a role for endothelial IL-1 signaling in the pathogenesis of atherosclerosis. However, the data on its receptor, the IL-1R1, its site of expression in human atherosclerotic plaques are still unclear. Therefore, we analyzed sections of human coronary atherosclerotic lesions to detect the cell type-specific expressing IL-1R1. To further explore whether the expression of IL-1R1 is altered by the advanced stages of atherosclerosis, we categorized the human progressive atherosclerotic plaques into two grades, either IV or V as per the AHA recommendations (34). These lesions were further categorized into very stable, moderately stable, and unstable lesions depending on the stability characteristics that are established by (1) using extensive histological landmarks and blindly assessed and graded by a pathologist at Louisiana State University Health Sciences Center at Shreveport ( Figure 1A).
To investigate IL-1R1 expression between the three phenotypically different stages of human atheroma, the lesions were stained for IL-1R1, Snail1 (EndMT transcription factor), and von-Willebrand factor: vWF (Endothelial marker) ( Figures 1B-E). Interestingly, lesions that are very stable show predominant expression of IL-1R1 within the endothelium and hardly any Snail1 expression, suggesting that EndMT is not happening in the stable human atherosclerosis ( Figure 1B). Surprisingly, we observed an increased expression of IL-1R1 mainly in a sub-endothelial population that are co-expressing Snail1 and vWF ( Figures 1C-E). The increase in Snail1and IL-1R1 proteins were significantly higher in unstable lesions compared to moderately stable lesions ( Figures 1D,E) Furthermore the association between IL-1R1 and Snail1 expressions in unstable lesions was very significant (p = 0.0032) ( Figure 1F). Collectively, the data demonstrates that IL-1R1 is expressed predominantly in endothelial cells in stable advanced human atherosclerotic plaques, however, the expression of IL-1R1 is increased in EndMT in unstable human atherosclerosis.
To confirm the causal association between IL-1R1 and EndMT in human unstable atherosclerosis, we first re-assessed the relative expression of active IL-1R1 and EndMT genes in macroscopically intact tissue vs. atheroma tissue from an online dataset (GSE43292) (35). The selected dataset was used because it provides a direct assessment to the list of genes that are upregulated in the intimal area of human unstable lesions and compared them with a macroscopically intact area from the same patient. In the original study, the endarterectomy specimens were characterized histologically according to the AHA and most atheroma samples were presented at stage IV and V lesions, and the macroscopically considered as "intact tissue" was almost exclusively composed of stage I and II lesions (35).
The groups of genes that are differentially expressed between either the macroscopically intact (areas surrounding the atheroma from the same patient samples) or atheroma tissue (from carotid endarterectomy samples) of n = 32 human subjects each were re-analyzed for genes that are involved in IL-1R1 activation mainly IRAK1/4 (Interleukin-1 Receptor Associated Kinase 1/4) and NFKB1 (nuclear factor kappa B1), and EndMT transcriptional factors SNAI1 (Snail1) and SNAI2 (Snail2), which were further blotted using a heat map (Figure 2A).
Interestingly, we examined the relative expression IL-1R1, IRAK1, NFKB and SNAI1) and found a significant differences in these genes' levels with a marked increase in the atheroma compared to intact tissues ( Figures 2B-E). Collectively, these findings suggest that the EndMT and active IL-1R1 are highly expressed in human severe atherosclerotic lesions as compared to intact tissues, and may suggest that the regional expressions of IL-1R1 and EndMT is induced by local factors rather than systemic effects.

IL-1R1 expression is associated with EndMT under areas of disturbed blood flow
Atherosclerosis develops at sites of local endothelial cell (EC) activation, where a shift in endothelial cell phenotype in response to d-flow occurs (6,36). The molecular mechanism by which ECs switch their phenotypes under d-flow areas is incompletely understood. D-flow (a low amplitude and a multi-directional force) (37) is present in certain parts of the arterial tree, such as the lesser curvature of the aortic arch, whereas the atheroprotective high amplitude and unidirectional (laminar) flow is present in other areas, such as the greater curvature of aortic arch (38) (Figures 3A-D). EndMT has been shown to be induced by d-flow at the atheroprone areas (36,39). Having shown that IL-1R1 expression is locally increased in human unstable atherosclerosis and that is associated with EndMT, we first asked if IL-1R1 is expressed under areas of d-flow, and if that is associated with EndMT. As expected, in mice with an atherosclerosis phenotype, IL-1R1 expression is predominant in the atheroprone areas as assessed by immunofluorescence and no IL-1R1 staining was detected in the athero-protective area of the aortic arch ( Figures 3B,C).
To establish the relationship between EndMT and IL-1R1 expression we stained the aortic arch of our EC-specific lineage tracing mice ( Figure 3D). The endothelial specific expression of IL-1R1 under d-flow areas was confirmed by a co-localization with CD31 and that coincides with Snail1 expression, only in the atheroprone areas ( Figure 3D), confirming the association between IL-1R1 and d-flow-induced EndMT. Furthermore, we examined the effect of laminar (LSS) and oscillatory shear stress (OSS) on IL-1R1 and SNAI1 mRNA expressions in vitro in HAOECs. Compared to static condition, LSS did not significantly alter the mRNA levels of IL-1R1 nor SNAI1 (Snail1) in our in vitro flow system. Interestingly, OSS significantly increased IL-1R1 and SNAI1 mRNA expressions compared to static conditions ( Figures 3E,F). We confirmed the changes in IL-1R1 at protein levels in vitro and relative to the athero-protective LSS, the atheroprone OSS significantly increased IL-1R1 protein expression with p < 0.01 (Supplementary Figure S1).
These findings further demonstrate that IL-1R1/Snail1 are induced by OSS and supports the potential role that IL-1R1 might play in d-flow-induced EndMT.

IL-1R1 knockdown prevents OSSinduced EndMT in vitro
Recent studies have demonstrated that d-flow promotes EndMT formation (36,40,41). However, the signaling intermediaries that link the mechanical d-flow to EndMT are still unclear. During EndMT, endothelial cells express mesenchymal markers such as N-cadherin, alpha smooth muscle actin (α-SMA), vimentin, in addition to the induction of the transcriptional factor Snail1. These mesenchymal markers drive the phenotypic changes in ECs that involve loss of EC markers such as VE-cadherin and CD31 by a Snail1 gene upregulation, and subsequent loss of cell-cell contacts ( Figure 4A) (8,(42)(43)(44).
To determine if IL-1R1 is involved in induction of d-flowinduced EndMT, we first examined silencing of IL-1R1 using SMARTpool siRNA to IL-1R1 (IL-1R1 siRNA) in vitro in HAOECs and then the cells were subjected to OSS using the Ibidi flow system for 24 h (Figures 4B-D). EndMT induction by Frontiers in Cardiovascular Medicine the athero-prone OSS was assessed using a combination of immunofluorescent staining. Compared to mock control, IL-1R1 siRNA treated cells showed a remarkable reduction in Ncadherin (mesenchymal cell marker) and an increase in the endothelial CD31 (endothelial cell marker) expression after OSS exposure ( Figure 4B). Consistent with this, OSS increased Snail1 nuclear translocation in mock cells, which was significantly decreased in IL-1R1-deleted ECs ( Figures 4C,D).
To assess the protein and mRNA levels of endothelial and EndMT markers by OSS, we measured proteins using Western blot and mRNA levels using qRT-PCR ( Figure 5) in mock vs. IL-1R1 deleted ECs. The transfection efficiency of IL-1R1 siRNA in HAOECs was confirmed ( Figure 5A). Consistent with our immunofluorescent staining, OSS induced an upregulation in Ncadherin and α-SMA protein and mRNA levels ( Figures 5B-D) and vimentin mRNA expression ( Figure 5D) in the mock control, whereas IL-1R1 siRNA pre-treatment in ECs showed a significant decrease in the induction of these markers even after the OSS exposure ( Figures 5B-D). Furthermore, OSS induced a downregulation in endothelial markers, which was prevented in IL-1R1 siRNA cells (Figures 5E,F). Likewise, in mouse IL-1R1 KO ECs that were exposed to OSS, the induction in the EndMT marker was remarkably decreased compared to WT cells (Supplementary Figure S2).
Collectively, the data suggests that IL-1R1 is implicated in an upstream signaling to the mesenchymal phenotype in HAOECs when exposed to the athero-prone oscillatory flow.

IL-1R1 knockdown prevents OSS-induced EndMT in vivo
To confirm that IL-1R1 is induced by d-flow in vivo and that this induction is associated with EndMT, we re-analyzed an online dataset (Bioproject #PRJNA646233) from EC-enriched single cell RNA seq analysis after a mouse partial carotid ligation surgery (6). The sc-RNA seq was comprised of transcript expression results of left carotid artery (LCA: exposed to the atheroprone d-flow) and right carotid artery (RCA: exposed to the athero-protective laminar flow) obtained at 2 days and 2 weeks post-ligation. We first re-blotted the EndMT population and compared their induction levels under different time points in both RCA and LCA ( Figure 6A). Our re-analysis confirmed the induction of EndMT cells in the ligated LCA compared to the unligated RCA after 2 weeks. Next, we retested the differential expression of IL-1R1 in the ligated LCA and un-ligated RCA among the EndMT population between 2 days and 2 weeks. Interestingly, we observed a significant increase in IL-1R1 expression in the ligated LCA (d-flow induced) compared to un-ligated RCA (laminar flow) after 2 weeks ( Figure 6B).
To determine if IL-1R1 directly contributes to EndMT induction by d-flow in vivo, we used our own mouse model of Control-iEC (VE-cadherin-CreERT2 tg/+ , ApoE −/− ) and IL-1R1 iEC−KO (VE-cadherin-CreERT2 tg/+ , IL-1R1 flox/flox , ApoE −/− ) mice. After tamoxifen injection, we confirmed the EndMT and IL-1R1 complex are highly expressed in human atheromatous lesions compared to intact tissue. (A) GSE43292 dataset was analyzed and (B) the heat map to SNAI1, IRAK1, NF-kB, IRAK4, IL-1R1 and SNAI2 was shown. (B-E) expression levels of the IL-1R, IRAK1, NFKB1 and SNAI1 genes from n = 32 individuals per group, the value (intensity) of each probe represents the RNA expression level. Data was analyzed as mean ± SEM, and statistical analysis was conducted by unpaired t test, *p < 0.05, **p < 0.01.  Figure S3). The experimental groups, Control-iEC and IL-1R1 iEC−KO were subjected to the partial carotid ligation surgery to induce d-flow in the LCA. EndMT gene SNAI1 and endothelial CD31 were measured following isolation of intimal mRNA by qRT-PCR after 2 weeks post-ligation ( Figures 6C,D). Consistent with our in vitro findings, the intimal mRNA levels of SNAI1 were remarkably increased by d-flow following the ligation of LCA in the control mice and that was significantly decreased in IL-1R1 iEC−KO mice ( Figure 6C). By contrast, d-flow induction in the ligated LCA reduced CD31 mRNA levels in control animals and that was significantly increased in IL-1R1 iEC−KO ( Figure 6D). Collectively, these results demonstrate that IL-1R1 inhibition in mice prevents d-flow-induced EndMT and maintain the EC phenotype.

Discussion
Accumulative evidence has demonstrated the phenotypic transition of endothelium to EndMT in human atheroma, and new emerging evidence with the single cell RNA studies have been reported on the presence of EndMT in mouse atherosclerotic plaques (43)(44)(45). D-flow is one of the main triggers to EndMT formation (36,39), however, there is little evidence on the signaling intermediaries that link d-flow to EndMT formation. Herein, we described a novel role for endothelial IL-1R1 in d-flow-induced EndMT in vitro and in vivo. We found for the first time, that IL-1R1 is predominantly expressed in endothelium of human coronary stable atherosclerotic plaques. The endothelial expression of IL-1R1 is increased in unstable lesions and is associated with the EndMT transcriptional factor Snail1. In our endothelial cell-specific lineage tracing mice, we observed the predominant expression of Human aortic ECs (HAOECs) exposed to laminar shear stress (LSS 10 dynes/cm 2 ; mimicking in vivo athero-protective flow and oscillatory shear stress (OSS; 1 dynes/cm 2 ; mimicking in vitro atheroprone flow) for 24 h. and lysed and mRNA was quantified using qRT-PCR for IL-1R1 (E) and SNAI1 (F). Data are represented as mean ± SEM, and 1-way ANOVA and Tukey's post-test test, *p < 0.05, n = 4 independent experiments. IL-1R1 and Snail1 in areas exposed to d-flow. Inhibition of IL-1R1 resulted in down-regulation of EndMT markers after OSS exposure in vitro. Importantly, in our endothelial cell-specific IL-1R1 KO mice we confirmed the direct effect of IL-1R1 inhibition on d-flow-induced EndMT in the partial carotid ligation model. Endothelial-to-mesenchymal transition (EndMT) is a process whereby endothelial cells undergo a series of molecular events that lead to phenotypic changes towards mesenchymal cells (40). EndMT plays a fundamental role during development (46) but there is a mounting evidence now that suggests EndMT is also involved in atherosclerosis (36,39,43). EndMT is characterized by multiple morphological and physiological changes, including loss of endothelial cell polarity, and disruption of intercellular junctions (42), so they can migrate in plaques. EndMT has been identified in progressive atherosclerotic plaques, and the extent of EndMT observed in the human plaques strongly correlates with the severity of the disease (43), implying clinical relevance of EndMT in the pathogenesis of atherosclerosis. However, the field faces several uncertainties regarding at which stage of human atherosclerosis EndMT appears to occur, and how that contributes to lesion stability, especially in the advanced stages where patients presented in the clinic with the symptoms of acute coronary syndrome. Our data using a blinded staging system and following the AHA recommendations for plaque characteristics (1, 2) detected EndMT in unstable stages IV and V human coronary atherosclerosis, whereas we did not observe EndMT in the stable stages. This finding is important as it implies that EndMT is involved in plaque stability, however, further characterization studies on their roles are warranted.
Interleukin-1 (IL-1) refers to two closely related cytokines IL-1α and IL-1β (47). Of the two, the leaderless cytokine IL-1β has been extensively investigated, and it has been shown to be involved in every stage of atherosclerosis from early lesion formation to progressive late-stage lesions (48). Global targeting of IL-1β demonstrated a decrease in atherosclerosis burden in mice (49, 50). Subsequently, the major clinical trial of inhibiting IL-1β with an antibody, canakinumab, administered in patients with myocardial infarction was concluded in 2017 showing a major reduction in cardiovascular mortality in sub-sets of patients. In this study, the magnitude of inflammation achieved in patients with high IL-6 and CRP was directly related to the magnitude of the clinical benefits of using anti-IL-1 (51), confirming the role of inflammation in CAD. However, some of those patients developed immunosuppressive off-target effects (11), which could be mitigated with an antibiotic regime (52). While in atherosclerosis it could be beneficial to inhibit immune cell infiltrations as evident that treatment with anti-IL-1B can provide additional cardio-protective benefits by reducing leukocyte recruitment to the atherosclerotic plaque as well as reducing its size (53). Nevertheless, the off-target effects on innate immune response maybe due to the systemic rather than local inhibition of IL-1. Intriguingly, two different research groups showed that inhibiting of IL-1β signaling in vascular wall rather than systemic inhibition is responsible for the beneficial effect on atherosclerosis burden reduction. Global IL-1R1 deletion in mice had significantly less atheroma and seems to not be altered in bone marrow transfer experiments (19), suggesting that vascular wall inhibition rather than myeloid inhibition is responsible (54). However, within the vascular wall, the data on IL-1R1 are still conflicting as to which cell-type specific IL-1R1 could be inhibited. Smooth muscle specific deletion of IL-1R1 showed high proinflammation and more instability features, whereas macrophage-specific deletion of IL-1R1 demonstrated less inflammation and more stable characteristics (20). In patients, there is no clear evidence on which one of the vascular wall cells is responsible for IL-1R1 effects, and whether the signaling through the receptor is different in different cell-types. Our study is the first to demonstrate an enhanced IL-1R1 expression in human atherosclerosis stages IV and V, and that is predominantly within the endothelium. While multiple studies on IL-1R1 have been diverse and even controversial, our study found that IL-1R1 expression is increased in the endothelium and that coincides with the appearance of EndMT in unstable lesions. These results align with the previously reported findings illustrating the predominate expression of IL-1β within the endothelium in human coronary atherosclerosis (24,29), suggesting the critical signaling of IL-1 within the endothelium.
Moreover, we found that the IL-1R1 expression is enhanced under d-flow areas. Both disturbed blood flow (d-flow) and IL-1 have been implicated in EndMT (36,44). The signaling pathway that links d-flow and EndMT is still unclear. Our recent work HAOECs were transfected with siRNA targeting IL-1R1 or mock and exposed to oscillatory shear stress (OSS) for 24 h. Cells lysates were collected for proteins of IL-1R1 (A), N-Cadherin (B), α-SMA (C), or mRNA (D). Protein or mRNA levels of endothelial markers Ve-Cadherin (E), and CD31 (F) were quantified using Western blot or qRT-PCR. Data analyzed by 2-way ANOVA and Bonferroni's post-test; *p < 0.05, **p < 0.01, representative blots are from n = 4-5 independent experiments. has demonstrated that d-flow induces interleukin-1 receptor signaling kinase within human and mouse progressive atherosclerosis (25). Interestingly, our current study determines the dynamic interplay between disturbed flow and IL-1 signaling in EndMT, as we showed that deletion of IL-1R1 in vitro and in vivo significantly prevented d-flow-induced EndMT. EndMT is a spectrum of phenotypic changes in ECs and whether inhibition of IL-1R1 mediated EndMT in atherosclerosis is yet to be investigated and future studies from our laboratory will focus on dissecting the downstream pathway(s) involved.

Limitation of the study
Our study provides the first assessment to the mechanical response of endothelial IL-1R1 to disturbed flow and EndMT formation in vitro and in vivo. Our study has strengths as well as some limitations. Firstly, we performed experiments on ApoE KO mice to study endothelial-specific IL-1R1 deletion on disturbed flow-induced EndMT. Our first observations on the association between IL-1R1 and Snail1 in human unstable atherosclerosis where the lesions exhibit more complex and advanced phenotype than that to the mice. The use of mice to study human pathology has some inherent limitations, which should be considered before any conclusions can be drawn. Mice do not usually develop atherosclerosis, due to the high protective HDL-C in their plasma, therefore, by knocking out ApoE and feeding the mice on a high fat diet for 4 weeks, we assessed the earliest possible changes in the vasculatures. We are not expecting the lesions that are developed in our mouse model to be phenotypically similar to that in humans, especially with the advanced stage IV and V lesions. This could explain why the staining pattern in mice is substantially different than that in humans. Secondly, we assessed the mRNA expressions of Snail1, CD31, and IL-1R1 in the intimal samples of ligated LCA and un-ligated RCA after the partial carotid ligation surgery in the Control-iEC and IL-1R1 iEC−KO animals. It could be there is some contamination in the samples with the medial mRNAs. However, we used a standard method by TRIzol flush after tissue harvesting, and tested the purity of the isolation as established by our research group (25,28) and others (6, 55).

Clinical significance of the current study
Coronary artery disease (CAD) accounts for ∼600,000 deaths/ year in the USA, contributing to significant morbidities (56). Atherosclerosis, an underlying disease process of CAD, is characterized by abnormal accumulation of lipids and proinflammatory cells within the vascular wall (57) which can rupture, leading to sudden death or acute myocardial infarction  (58). While current treatment strategies, including percutaneous coronary intervention, are used to improve the quality of life, they fail in preventing the severity of disease (59), leading to an urgent need for uncovering novel and specific therapeutic targets to prevent atherosclerosis-induced mortality. Early clinical observations suggest that atherosclerosis occurs at the arterial sites of disturbed blood flow (d-flow) (60). In response to d-flow, endothelial cells (ECs) acquire a mesenchymal phenotype through endothelial-tomesenchymal transition (EndMT) (36,40). EndMT is implicated in the progression of atherosclerosis, where it leads to plaque destabilization by promoting local inflammation (8). Interleukin-1 (IL-1) is highly implicated in atherosclerosis progression (19)(20)(21)61) and development of EndMT (16). The therapeutic approach of blocking IL-1 using an antibody, Canakinumab, significantly reduced mortality rates in CAD patients (12,62) but it also induced systemic off-target effects due to low circulating leukocyte counts (12), limiting the global targeting of IL-1. Thus, although inhibition of IL-1 is a great strategy to prevent progression of atherosclerosis, we need novel therapeutic targets for specific inhibition of IL-1 within the vascular wall to avoid systemic offtarget effects on immune cells (47).
In conclusion, global inhibition of IL-1 in patients with underlying atherosclerosis has already been tested in the completed CANTOS study with some promising effects (11). However, the data on interleukin-1 receptor IL-1R1 in mice are still conflicting (20, 21). Herein with our results, we showed that endothelial IL-1R1 expression is implicated in unstable atheroma undergoing EndMT. Furthermore, our study illustrates a crosstalk between d-flow and IL-1R1 in endothelial cells and demonstrates a novel role for the endothelial IL-1R1 expression in EndMT formation after d-flow exposure, which represents a new direction for potential precision therapy by targeting IL-1R1 within the endothelium. This might lead to better strategies in managing progressive atherosclerosis and maintaining plaque stability in patients.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Ethics statement
The animal study was reviewed and approved by the institutional animal care and use committee (Louisiana State University Health Sciences Center at Shreveport, protocol # P-23-003 to MA).

Author contributions
EK, MP, MA originally conceived, performed experiments, analysed data, and wrote the manuscript. SC, XY performed data analysis for the scRNA seq and online dataset, MT captured the H&E images, and QH, SC, XY, SV, MT contributed to manuscript writing. MA supervised the project. All authors contributed to the article and approved the submitted version.