An interplay of non-coding RNAs regulates CDH13 expression and affects endothelial function and coronary artery disease risk

An interplay of non-coding RNAs regulates CDH13 expression and affects endothelial function and coronary artery disease risk | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article An interplay of non-coding RNAs regulates CDH13 expression and affects endothelial function and coronary artery disease risk Shuangyue Li, Xiaoning Song, Anastasiia Diagel, Ling Li, Aldo Moggio, and 26 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7333062/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Many common diseases have a polygenic architecture. The responsible alleles are thought to mediate risk by disturbing gene regulation in most cases, however, the precise mechanisms have been elucidated only for a few. Here, we investigated the 16q23.3 genomic locus, which genome-wide significantly associates with coronary artery disease, a globally leading cause of death caused by accumulation of lipid-rich inflammatory plaques in the arterial wall. The locus harbors CDH13, whose mRNA and protein we found to be suppressed in atherosclerotic human and mouse arteries. Loss-of-function(LoF) variants of CDH13 were associated with detrimental cardiovascular phenotypes in the UK Biobank. Its knock-out increased plaque-sizes in Cdh13 -/- / Apoe -/- mice compared to Apoe -/- mice on a Western diet. After establishing an atheroprotective role of CDH13 , we studied its regulation. Integration of population genomic and transcriptomic datasets by GWAS-eQTL colocalization analysis identified CDH13 and four long non-coding RNAs (lncRNAs) as candidate causal genes at the 16q23.3 locus. dCas13-mediated RNA immunoprecipitation revealed that the lncRNA CDH13-AS2 binds to CDH13 mRNA in human endothelial cells (ECs). Its CRISPR/Cas9-based knockout in ECs was atherogenic, whereas dCas9-based transcriptional activation (CRISPRa) of CDH13-AS2 was atheroprotective; effects that were found to be mediated by the stability of CDH13 mRNA. To further understand how the CDH13-AS2 protects the mRNA we searched in silico and screened in vitro for microRNAs (miRNAs) that bind to CDH13 3’UTR. Indeed, four miRNAs, miR-19b-3p, miR-125b-2-3p, miR-433-3p, and miR-7b-5p, were found experimentally to accelerate CDH13 mRNA degradation, an effect that was neutralized by CRISPRa of CDH13-AS2 . Taken together, our study demonstrates an interplay of miRNAs, lncRNAs, and mRNA, which modulates the abundance of an atheroprotective protein in endothelial cells, which may offer a new therapeutic target for coronary artery disease. Health sciences/Cardiology/Cardiovascular biology/Cardiovascular genetics Biological sciences/Molecular biology/Non-coding RNAs Coronary artery disease (CAD) long non-coding RNA (lncRNA) microRNA (miRNA) GWAS-eQTL colocalization analysis dRfxCas13d CRISPR/Cas9 CRISPR activation (CRISPRa) 3’ untranslated region (3’ UTR) RNA stability and RNA interference Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Coronary artery disease (CAD) is a genetically-mediated and often devastating common disease. Genome-wide association studies (GWAS) identified hundreds of risk alleles that cumulatively shape a population-wide disposition for atherosclerosis 1-3 , i.e. the build-up of lipid-rich inflammatory plaques in arterial vessels, which ultimately may occlude and cause myocardial infarction. CAD is also a prototypic polygenic disease that is primarily precipitated by disturbed gene expression rather than by structural changes of the affected proteins 1, 4, 5 . However, the molecular mechanisms that link CAD risk alleles with altered gene expression have been elaborated in only a few exceptional cases 6-8 . In human biology, a large fraction of key regulators of tissue-specific gene expression are non-coding RNAs (ncRNAs), such as long non-coding RNA (lncRNA) and microRNA (miRNA). Such RNAs regulate the expression, splicing, stability, or translation of the cognate protein-coding genes either by interacting directly with the mRNA or indirectly via binding to other ncRNAs 9-14 . Therapeutically, the modulating effects of ncRNAs can be harnessed by either silencing or activating genes 15, 16 . For instance, a phase 2 clinical trial showed that inhibition of miR-132 by an antisense oligonucleotide may prevent cardiac remodeling post-myocardial infarction 17 . Several lncRNA coding genes mapped to CAD-GWAS loci 3 , and the most significant CAD locus is mapped to the lncRNA coding gene, ANRIL 16 . Since 2017, the 16q23.3 locus has been associated with CAD by GWAS and repeatedly replicated by studies with ever-growing sample sizes and ethnic diversity 1, 2, 18 19 . Yet, the functional implications of this locus remains unclear. Our exploration revealed that the 16q23.3 locus, residing in the CDH13 (cadherin 13, T cadherin) regulates both the protein-coding and lncRNA-coding genes. Both our mouse and human data indicated that genetic loss of CDH13 is atherogenic, which aligned with the observation that lower CDH13 expression was found in arterial tissue from patients and mice with atherosclerosis. In a search for potential therapeutic RNA targets, we experimentally identified a lncRNA and four miRNAs to participate in the regulation of CDH13 mRNA. A series of CRISPR/Cas9-based knockdown and dCas9-based transcriptional activation (CRISPRa) experiments in human endothelial cells clarified the role of interacting ncRNAs in regulating CDH13 mRNA and consolidated the notion of an atheroprotective role of CDH13. Results Five candidate causal genes are identified at the 16q23.3 locus The CAD association signals at the 16q23.3 locus reside within the intragenic region of the protein-coding gene, CDH13 (cadherin 13, T cadherin). To systematically map candidate causal genes at this locus, we conducted a colocalization analysis using the CAD GWAS signal at the 16q23.3 locus and expression quantitative trait locus (eQTL) datasets from disease-relevant tissues (Method, Figure.1a and 1b). The GWAS association signal was from the latest summary genetic statistics of the CARDIoGRAMplusC4D Consortium 1 . The tissue types explored for eQTL analyses included arteries, adipose tissues, liver, blood, and skeletal muscle of ~ 600 individuals from the Stockholm-Tartu Atherosclerosis Reverse Networks Engineering Task (STARNET) project 20, 21 (Method, Figure.1a). The results colocalized the CAD GWAS signal at the 16q23.3 locus with an eQTL signal specific to arterial tissues pointing to CDH13 and four lncRNAs ( CDH13-AS1 , CDH13-AS2 , CEDORA, and CTD-3253I12.1 ) to be candidate causal genes for CAD (Method, Figure. 1b, Extended Data Figure 1a, supplementary Table 1). Namely, their expressions in arterial tissues likely mitigated the risks of CAD diseases. The loss of CDH13 promotes the development of atherosclerosis in humans and mice. To understand the underlying mechanism of CAD related to the 16q23.3 locus, a critical step is to elucidate the functionality of the protein-coding gene, CDH13 . We first analyzed the animal model for atherosclerosis, in the Apoe -/- mouse, theCdh13 protein levels gradually declined in the aorta after 4, 8, and 12 weeks of Western diet treatment (Figure. 1c, Extended Data Figure 1b). Likewise, the bulk RNA sequencing of atherosclerosis plaques from patients undergoing carotid endarterectomy (Munich Vascular Biobank) 22, 23 revealed in advanced plaques (n=145) less CDH13 expression than in early plaques (n=57) (Method, Figure. 1d, supplementary Table 2). Thus, we observed reduced arterial expression of CDH13 during the development of atherosclerosis in both humans and mice, suggesting a protective role of CDH13. To confirm this hypothesis, we generated Cdh13 -/- gene knockout (KO) mice on the atherogenic background by crossing Cdh13 -/- with Apoe -/- mice to obtain the Cdh13 -/- / Apoe -/- mice (Extended Data Figure 1c and 1d). Compared to the Apoe -/- mice, the Cdh13 -/- / Apoe -/- mice on an eight-week Western diet (WD)(Extended Data Figure 1e) had increased atherosclerosis lesion areas in the aortic root and arch (Method, Figure. 1e, Extended Data Figure 1f), reinforcing CDH13 to be atheroprotective. To further assess the roles of CDH13 , we investigated the effect of its loss-of-function (LoF) variants on CAD and other 19 vascular-related diseases or traits using phenotype and exome sequencing data from the 470,000 UK Biobank participants (Method, Supplementary Table 3 and 4). LoF variants of CDH13 were found in 272 participants who showed a trend of increased CAD incidence (β = 0.335, P = 0.184). LoF variants were significantly associated with atherogenic traits, such as increased arterial stiffness index (β = 2.450, P=3.92e-5), serum C-reactive protein level (β = 0.654, P=1.58e-2), blood leukocyte and lymphocyte counts (β=0.427 and 0.281, P=7.16e-4 and 5.01e-5, respectively), but decreased serum adiponectin level (β = -0.349, P=4.53e-2) (Figure 1f, Supplementary Table 3 and 4). Together, the results from mice and humans suggest that CDH13 is a causal gene at the 16q23.3 locus (Figure 1), with higher expression appearing to be protective against CAD. All five candidate causal genes are expressed in human endothelial cells Beyond CDH13 , our eQTL analysis also prioritized four lncRNAs as candidate causal genes at the 16q23.3 locus (Figure. 1b). To identify relevant cell type(s) that express the respective RNAs we analyzed the publicly available single-cell (sc) RNA-seq dataset of proximal-to-mid right coronary artery (RCA) from four patients who underwent heart transplantation(Methods) 24 . Our analysis identified 14 major cell populations in this dataset(Method, Figure. 2a and 2b), and CDH13 was highly expressed in vascular muscle cells (VSMCs), fibromyocytes, and endothelial cells (ECs), especially arterial ECs (Method, Figure. 2c to 2f). Unfortunately, none of the four lncRNAs were identified in the scRNAseq datasets due to the low-expression nature of lncRNAs. Thus, we did qPCR experiments for the four lncRNAs in cardiovascular cell types, including human coronary artery ECs, VSMCs, fibroblasts (FBs), monocytes (the THP1 cell line), macrophages (THP1-differentiated), and T cells (the Jurkat cell line). As a result, EC was the only cell type expressing all four lncRNAs with a relatively high level (Figure. 2g, Extended Data Figure 2a to 2d). Therefore, our subsequent investigations focused on ECs to study the interaction between lncRNA and CDH13 . LncRNA interacts with CDH13 mRNA in human endothelial cells Given that lncRNAs often regulate gene expression in cis , we explored whether the four lncRNAs could affect CDH13 expression. We first tested whether the four lncRNAs could directly bind to on the CDH13 mRNA by conducting dRfxCas13d-based RNA-immunoprecipitation (RIP) in human umbilical vein endothelial cells (HUVEC). The dRfxCas13d construct was fused with an HA tag, allowing the anti-HA magnetic beads to pull down and enrich the binding RNAs 25 (Figure. 3a). Five RNA-targeting guide RNAs (rgRNAs) of CDH13 were used to pull down CDH13 mRNA and its potential binders. The scrambled rgRNAs were used as controls (supplementary Table 6). Using the dRfxCas13d/rgRNA- CDH13 system, we detected the significant binding of CDH13-AS2 on CDH13 mRNA, but not the other three lncRNAs (Figure. 3b). This might suggest the direct effects of CDH13-AS2 on CDH13 . To confirm this, we further investigated the interaction between CDH13-AS2 and CDH13 mRNA in HEK.293T cells, where neither transcript was expressed to minimize confounding factors. We first specifically activated the expression of CDH13-AS2 and CDH13 by CRISPR-mediated transcriptional activation (CRISPRa) 26 (Figure. 5a). The sequence of dgRNAs was designed based on the promoter or enhancer of the gene target (supplementary Table 8). Five dgRNAs were used to activate the expression of CDH13-AS2 , three and two dgRNAs, respectively, targeting the promoter (ENCODE ID, E1832741) and the enhancer (ENCODE ID, E1832742) of the lncRNA coding gene (Extended Data Figure 3a). We tested the effectiveness of the five dgRNAs individually to identify a highly potent dgRNA. All five dgRNAs significantly activated the expression of CDH13-AS2 with high efficiency, and dg1_ CDH13-AS2 induced the strongest activation and was therefore used for the following experiments (Extended Data Figure 3c). To activate the expression of CDH13 , we designed eight dgRNAs for this gene, dg 1 - 4 for the promoter (ENCODE ID, E1832187) and dg 4 - 8 for the enhancer (ENCODE ID, E1832188), all near the transcription start site (TSS) or exon 1 (Extended Data Figure 3b). Four dgRNAs, dg2, 4, 7, and 8, drove expression activation of CDH13 , and dg4_ CDH13 showed the highest efficiency (Extended Data Figure 3d). In the HEK. 293T cells, we successfully induced high expression of both CDH13-AS2 and CDH13 using the corresponding CRISPRa (Figure 3c, Extended Data Figure 3c, 3d). In these cells, we transfected the plasmid system of dRfxCas13d/rgRNA- CDH13 -based RIP, which was able to pull down the CDH13-AS2 (Figure 3c). Likewise, we designed the dRfxCas13d/rgRNA- CDH13-AS2 mediated RIP as previously, which successfully precipitated CDH13 mRNA (Figure 3c and 3d, supplementary Table 6). These results were in line with our in silico experiments, in which we predicted the lncRNA-mRNA binding of CDH13, respectively, with the four lncRNAs using the LncRRIsearch webtool (supplementary Table 5). Among the four, we predicted CDH13-AS2 as the only positive binder via reverse-complementary to the 3’UTR of CDH13 mRNA. The local base-pairing interaction energy was -113.58 kcal/mol, representing the strongest interaction among the 100 anticipated bindings (Extended Data Figure 3e, 3f and supplementary Table 5). These experiments consolidated our findings on the interaction between CDH13-AS2 and CDH13 . CDH13-AS2 positively regulates CDH13 expression and EC functions To explore whether CDH13-AS2 could affect CDH13 expression and CAD-relevant cellular phenotypes, we knocked out CDH13-AS2 ( CDH13-AS2 -KO) using the dual-CRISPR/Cas9 targeting strategy (Figure. 4a). The third generation of lentiviral system was used to deliver the sgRNAs and spCas9 transgenes (supplementary Table 7). The dual-CRISPR strategy excised a 34bp of the shared exon (ENSE00002602225) of CDH13-AS2 major transcripts 27 (Figure. 4a, Extended Data Figure 4a). Amplification PCR and qPCR on the KO lines showed a successful knockdown of CDH13-AS2 expression (Figure. 4b). Likewise, we observed reduced CDH13 mRNA and protein levels were in CDH13-AS2 -KO HUVECs compared to control cells (Figure. 4b and 4c). Furthermore, we assessed functions related to EC fitness, including apoptosis, migration, proliferation, and immune cell adhesion, in the CDH13-AS2 -KO HUVECs. To induce apoptosis, gene-edited cells were treated with 100 ng/ml LPS, and after the treatment, apoptosis-triggered fluorescence changes were detected every 10 or 12 hours until 72 hours. After 24 hours of treatment, CDH13-AS2 -KO cells showed stronger apoptosis compared to control cells at each time point of measurement (Figure. 4d). To explore the migration of the gene-edited cells, we conducted the wound healing assay, and imaged cell migration every 24 hours until 72 hours. After wounding, CDH13-AS2 -KO cells displayed slower migration into the wounded area at three time points (24, 48, and 72 hours), indicated by a lower percentage of cell coverage in the wounded area (Figure. 4e). The proliferation of CDH13-AS2 -KO HUVECs was assayed by BrdU incorporation for 16 hours followed by flow cytometry analysis. The result showed decreased proliferation in the CDH13-AS2 -KO cells (Figure. 4f). Increased EC apoptosis and reduced EC migration and proliferation could decelerate the healing of endovascular lesions and promote atherosclerosis. To probe the immune cell adhesion, we labeled THP1 monocytes with calcein, a live cell-permeant dye, and added the labeled cells onto CDH13-AS2 -KO and control HUVEC cells. We observed increased monocyte adhesion on the CDH13-AS2 -KO cells (Figure. 4g, Extended Data Figure 4b), which might contribute to increased immune cells in atherosclerosis plaques. To test whether CDH13-AS2 overexpression could lead to the opposite effect, we employed CRISPRa experiments in HUVECs as previously 26 (Figure. 5a, Extended Data Figure 3c to 3f, Extended Data Figure 5a). Converse to the phenotypes in CDH13-AS2 -KO HUVECs (Figure. 4), HUVECs with CDH13-AS2 -CRISPRa had higher CDH13 expression increased migration and proliferation, and decreased apoptosis and monocyte adhesion (Figure. 5b-5g, Extended Data Figure 5b). Thus, the enhanced level of CDH13-AS2 increased CDH13 expression and led to atheroprotective endothelial phenotypes, which might mitigate the risk of CAD. CDH13-AS2 stabilizes CDH13 mRNA LncRNAs were shown to stabilize other RNAs 28, 29 , which likewise appeared to be feasible for CDH13-AS2 and CDH13 mRNA given their positive correlation. To test this hypothesis, we use a time-course CRISPRa experiment and an RNA stability assay. We conducted these experiments in HEK. 293T cell line again to minimize potential confounding effects (Extended Data Figure 3c to 3f). We conducted time-course experiments to observe RNA CRISPRa using plasmids encoding dg1_ CDH13-AS2 /dCas9. 24 hours after the transfection, we observed an approximately four-fold increase of CDH13-AS2 , and the high expression level was maintained for 120 hours (Figure. 6a). At 144 hours, we observed a slight RNA decay of CDH13-AS2 (Figure. 6a). The experiment showed good RNA stability of CDH13-AS2 . Time course experiments of the dg4_ CDH13 /dCas9-based CRISPRa indicated that the activation of CDH13 peaked at 72 hours after transfection (Figure. 6b). At 96 hours, we observed the mRNA decay of CDH13 , which was 48 hours earlier than CDH13-AS2 (Figure. 6a and 6b). However, additional CRISPRa of the CDH13-AS2 in the same cells shifted the peak expression level of CDH13 from 72 to 96 hours, postponing the mRNA decay by 24 hours (Figure. 6c). At the endpoint of the experiments (144 hours), cells with the additional activation of CDH13-AS2 still showed higher CDH13 expression compared to without (Figure. 6d). The results suggest that CDH13-AS2 can stabilize the expression of CDH13 . We further confirm this by testing whether CDH13-AS2 could prevent CDH13 mRNA decay after transcription blocking. In CDH13 and CDH13-AS2 CRISPRa experiments, after 48 hours of plasmid transfection, actinomycin D was used to block the gene transcription. After 6 hours of actinomycin D treatment, less RNA decay was observed for CDH13-AS2 than CDH13 mRNA, ~ 60% vs ~ 80% (Figure. 6e). Comparing the CDH13 single vs CDH13 & CDH13-AS2 dual CRISPRa, CDH13-AS2 overexpression was able to reduce CDH13 mRNA decay by ~ 30% (Figure. 6f). However, CDH13 overexpression did not affect CDH13-AS2 , comparing the CDH13-AS2 single vs CDH13 & CDH13-AS2 dual CRISPRa (Figure. 6g).Our data indicated CDH13-AS2 as a stabilizer for CDH13 mRNA. CDH13-AS2 competes with miRNAs to stabilize CDH13 mRNA A ~1.6kb long 3’UTR was annotated for the CDH13 coding gene by the Ensembl genome database (GRCh37.p13), and CDH13-AS2 was predicted to bind to CDH13 3’UTR (supplementary Table 5, Extended Data Figure 3e and 3f, Extended Data Figure 6). Given that miRNAs can regulate gene expression by binding UTRs to trigger RNA decay 30 , we, therefore, tested whether there are miRNA(s) at the 3’UTR of CDH13 mRNA. We first predicted miRNA binding on the CDH13 3’UTR using four databases, miRTarBase 31 , miRWalk 32 , scanMiRApp 33 , and TargetScan 34, 35 . The predicted miRNAs were further selected based on their expression level in ECs 36 (supplementary Table 9). 63 predicted miRNAs with reads per million >20 were selected and further prioritized based on 1) being conserved between human and mouse, 2) experimental hints (RNA-seq), or 3) suggested by ≥ two prediction databases (supplementary Table 9). Finally, 11 miRNAs (supplementary Table 10), including miR-30a-5p, miR-181a-5p, miR-155-5p, miR-19b-3p, miR-125b-2-3p, miR-34a-5p, let-7b-5p, let-7b-3p, miR-125a-3p, miR-485-3p, and miR-433-3p were finalized for further investigation. The corresponding 11 miRNA mimics and a scrambled control mimic were purchased commercially (Qiagen) (supplementary Table 11). A 1323bp of 3’UTR covering all the predicted binding sites of the 11 miRNAs was synthesized and cloned into a dual luciferase plasmid (psiCHECK™-2|RNAi Assay, Promega) to generate the psiCHECK-UTR1323 construct. The scrambled control (CTR mimic) and the 11 miRNA mimics were respectively co-transfected with the psiCHECK-UTR1323 construct to functionally analyze potential miRNA binding sites on CDH13 3’UTR by an RNA interference assay. The assay is based on the Renilla/Firefly luminescence signals. Both luminescences were encoded on the psiCHECK backbone, the former was fused with the 3’UTR sequence, and the latter was on the same plasmid as the internal control (Figure. 7a). After co-transfection of the 11 miRNAs, respectively, with the psiCHECK-UTR1323 construct for 48 hours, both Renilla and Firefly luminescence were measured. In comparison to the CTR mimic, five miRNAs, miR-125b-2-3p, miR-19b-3p, miR-181a-5p, miR-433-3p, and let-7b-5p triggered Renilla luminescence reduction relative to the Firefly signal (Figure. 7b). demonstrating their effects on triggering RNA decay at CDH13 3’UTR. To probe whether CDH13-AS2 could block the interaction of the five miRNAs with CDH13 3’UTR, we co-transfected control or CDH13-AS2 CRISPRa plasmids, the psiCHECK-UTR1323 and correspondingly the five miRNAs in the HEK. 293T cells for 48 hours (Figure. 7c). Compared to the control, cells with CDH13-AS2 _CRISPRa successfully restored the reduced luciferase activity caused by four of the five miRNAs, except for miR-181a-5p (Figure. 7d through 7h). The results demonstrated that CDH13- AS2 can inhibit miRNA-triggered CDH13 mRNA degradation and therefore stabilize its expression. The lncRNA-miRNA-mRNA interplay might thus be harnessed to develop RNA-based therapies for CAD. Discussion GWAS have reproducibly associated the 16q23.3 locus with the risk of CAD, but the underlying mechanisms remained unclear 1, 2, 18, 19 . Here, we identified CDH13 and four lncRNAs as causal genes at the locus. In the UK Biobank, carriers of LoF alleles displayed increased risk for atherogenic traits. Colocalization analysis indicated CDH13 contributing to CAD via its roles in the arterial wall(Figure 1a and 1b, Extended Data Figure 1a, supplement Table 1, 3 and 4). Moreover, advanced atherosclerotic plaques in the carotid artery displayed lower CDH13 expression compared to early plaques(Figure 1d). Cdh13 protein levels gradually declined during the development of atherosclerosis in Apoe -/- mice on a Western diet(Figure 1c). Cdh13 -/- /Apoe -/- mice on a Western diet developed larger atherosclerotic lesions compared to Apoe -/- mice(Figure 1e). In line with our findings, Fujishima et al. reported enhanced neointima proliferation and atherosclerosisin Cdh13 -/- / Apoe -/- mice 37 . Thus, results from both humans and mice indicate that lower CDH13 expression aggravates atherosclerosis or – vice versa – CDH13 is atheroprotective. Our GWAS-eQTL colocalization analysis also indicated four lncRNAs as candidate causal genes at the 16q23.3 locus(Figure 1b, supplement Table 1). We experimentally validated the in-silico prediction(supplement Table 5) that one of the lncRNAs, CDH13-AS2 , binds to the CDH13 3’UTR(Figure 3b to 3d). CRISPR-based knockdown and activation of CDH13-AS2 indicated a protective role of this antisense RNA(Figure 4 and 5), as it prevented CDH13 mRNA decay and enhanced its beneficial effects in EC(Figure 6). We further demonstrated that four miRNAs were involved in the decay of CDH13 mRNA. RNA interference and CRISPRa experiments revealed that the miRNA-triggered decay of CDH13 mRNA was by ameliorated by CDH13-AS2 (Figure 7). Albeit more than 90% of GWAS signals map to non-coding regions only a few loci were found to act on CAD through ncRNA functions, such as MIAT at the 22q12.1 locus and ANRIL at the 9p21 locus 16, 38 , which is still the strongest CAD risk locus known so far 39 . At the 16q23.3 locus, we found that the lncRNA CDH13-AS2 is one of the causal genes as it up-regulates the protein-coding gene CDH13 in a cis manner by competing with the binding of miRNAs. LncRNAs are commonly reported to bind to miRNAs, acting as sponges, to prevent mRNA degradation 40, 41 . A classic regulatory mechanism of miRNAs involves binding to the 3′ UTR of a mRNA, leading to its degradation or translational repression 12, 42-44 . Here we showed that CDH13-AS2 competed with the binding of four miRNAs (miR-125b-2-3p, miR-19b-3p, miR-433-3p, and let-7b-5p)(Figure 7d to 7h). Similar interactions involving miRNAs have been reported for cancer and Alzheimer's disease 40, 45 . While current drugs for treatment of CAD uniformly address genes that increase CAD risk via lipid levels 1-3 , i.e. PCSK9 46 , LPA 47 , ANGPTL3 48 , and APOC3 49 , endogenous atheroprotective mechanisms are less well explored therapeutically. The potential of an atheroprotective milieu is illustrated by the internal mammary artery, which – even when exposed to high lipid levels or strong genetic disposition – resists plaque formation 50, 51 . In such a sense, our findings could be harnessed to design therapeutic strategies for CAD, going beyond lipid lowering, by increasing CDH13 expression, thereby enhancing arterial resilience to the disease. From a therapeutic perspective, it may be challenging to interfere with the regulation of a widely expressed gene . In this respect, tissue or cell-specific regulation may be beneficial. To some extent, it is the case of CDH13 given its expression in arteries limited to a few mesoderm-derived cell types (Figure. 2a-2d) and the CAD risk at the 16q23.3 locus is mediated by CDH13 and lncRNAs exclusively in artery tissues. Indeed, endothelial cells demonstrated improved function upon activation via CDH13-AS2. The lncRNA-miRNA-mRNA interplay might inspire intravascular delivery of CDH13 –miRNA target site blockers, antagomirs or anti-miRs to increase the expression of the gene, which protects coronary arteries from developing CAD. Target site blockers, anti-miR and alike, are already being explored as a therapeutic strategy for cardiovascular and cerebrovascular diseases as they can penetrate the arterial wall 15, 52-54 . We are aware of the limitations of our study. First, although we were able to prioritize CDH13 and the four lncRNAs as candidate causal genes for CAD by GWAS-eQTL colocalization analysis, no data were available to directly investigate the genetic link of the four miRNAs with CAD. Indeed, currently available population transcriptomes do not include sequences of such short RNAs. Second, our investigation provides mechanistic insights rather than specific therapeutics to increase CDH13 expression, which may be challenging given that RNA delivery specific to ECs was shown to have low efficiency. However, RNA-based therapy, such as target site blockers or anti-mirs, showed good penetration into artery walls, which are already being explored for cardiovascular diseases 15, 52-54 . Additionally, further technological advancements are anticipated to enhance cell-specific RNA delivery. Third, CDH13 was also expressed in VSMCs (Figure. 2a through 2f), which is worthy of an in-depth investigation but was not explored here. We rather focused on ECs that displayed high expression levels of CDH13 as well as lncRNAs and miRNAs involved in its regulation. Taken together, a lncRNA-miRNA-mRNA interplay affects the expression of CDH13 , which resides as the only coding gene at a GWAS locus for CAD. Multiple lines of evidence indicate that the GWAS signal is mediated by the protective effects of CDH13 on the arterial wall. The RNA interplay may be explored therapeutically by mimicking the effect of CDH13-AS2 tointerfere with the miRNA-mediated degradation of CDH13 mRNA at its long 3’UTR. Methods Colocalization analysis of the GWAS and eQTL signals For colocalization analysis at the 16q23.3 locus, we used full-genome GWASs dataset of CAD, respectively, from CARDIoGRAMplusC4D and expression quantitative trait loci (eQTL) data of five tissue types of ~ 600 individuals from the STARNET (Stockholm-Tartu Atherosclerosis Reverse Networks Engineering Task) study 1, 55, 56 (Figure.1c). The five tissue types included blood(n = 560), artery (atherosclerotic aortic root, n = 539 and free internal mammary artery, n = 553)), adipose tissue (subcutaneous, n = 534 and visceral abdominal, n = 534), skeletal muscle (n = 534), and liver (n = 546). Initially, we overlapped GWAS and eQTL summary statistics, utilizing a GWAS p-value threshold of 5e - 8 and an eQTL p-value cutoff of 0.01 to select SNPs for further testing. To estimate colocalization, we ran the coloc.abf function implemented in the coloc R package 57 that uses an approximate Bayes factor to estimate the posterior probabilities between a given GWAS and eQTL signal (supplementary Table 1). Significant colocalizations were defined with PP4 ≥ 0.70, indicating a common causal variant between the GWAS and eQTL data. Differential expression analysis of CDH13 using bulk RNA-seq data of patients RNA sequencing data of human carotid artery plaques 57 early and 145 advanced plaques of the human carotid arteries were harvested during carotid artery endarterectomy (CEA) surgery, transported to the laboratory, and snap-frozen. CEA was performed due to advanced atherosclerotic lesion formation and stenosis in the carotid arteries. The patients’ characteristics are summarized in supplementary Table 2. The tissue handling,RNA extraction, and sequencing were as previously 58 . Bulk RNA sequencing and quality control (QC) were performed as described 23 . Raw read counts were normalized with the trimmed mean of M-values (TMM) and transformed with voom, resulting in log2-counts per million with associated precision weights, which were then used for differential expression analysis. Rare variant association analysis in the UK Biobank The rare variant association analysis for CDH13 was conducted using data from the UK Biobank (https://www.ukbiobank.ac.uk/) under the project 25214. We used whole-exome sequencing (WES) data of 470,000 participants from the UK Biobank (The final release of population-level exome OQFE variants) 59 . The UK Biobank annotation helper file was used to obtain the list of rare variants within the CDH13 gene region. Variant annotation was derived from the snpEff tool using the Ensembl v85 gene definitions to determine their functional impact on transcripts and genes. Loss-of-function variants (LoF) included stop codon-introducing or splice site-disrupting SNPs, insertion/deletion variants predicted to disrupt a transcript's reading frame, or larger deletions removing either the first exon or more than 50% of the protein-coding sequence of the transcript. We collected phenotypic data for 6 binary and 14 quantitative traits, including blood biomarkers, vascular health indicators, and immune cell parameters. The phenotypes were defined using the respective Field IDs, ICD-10 and ICD-9 codes obtained from the primary care, OPCS-4 diagnoses, and self-reported codes. When multiple measurements of the same phenotype were available for an individual, we utilized the measurements ofthe first visit. Age, sex, and 10 genotype principal components were gathered as covariates. The data was accessed through the UK Biobank Research Analysis Platform (RAP). The association analysis was conducted on 500K WES data following the UKB tutorial ‘Burden Testing with regenie Using WES Annotation Files’ (https://dnanexus.gitbook.io/uk-biobank-rap/science-corner/using-regenie-to-generate-variant-masks#annotation-file). We used the genome-wide regression approach implemented in REGENIE 60 . Computationally efficient whole-genome regression for quantitative and binary traits. Nat Genet 53, 1097-1103. 10.1038/s41588-021-00870-7), available through the Swiss-Army-Knife tool library in RAP. Variants with an allele frequency below 1% were classified as rare. For association analysis, we used M1 variant masks, which LoF variants. To aggregate the effects of rare variants across the CDH13 region, a burden test was applied. This approach consolidates the variants from our defined mask into a unified burden mask and evaluates it as a single genotype to produce association statistics. To assess associations with binary phenotypes, a logistic regression model was used, and for quantitative traits, linear regression. The output of the burden test included P-value, estimates of beta (effect size), and the standard error. Single-cell RNA sequencing (scRNA-seq) analysis To characterize coronary artery expression of candidate genes at the 16q23.3 locus, we utilized the publicly available single-cell RNA sequencing (scRNA-seq) dataset GSE131778(Figure. 2a to 2f). This dataset includes single-cell transcriptomic data obtained from eight coronary artery samples derived from four patients who underwent heart transplantation. The samples were collected from the proximal-to-mid right coronary artery (RCA) after removing debris and selecting viable cells through fluorescence-activated cell sorting (FACS). The cells were processed using the 10X Genomics Chromium platform with 3' chemistry reagents (version 2). The scRNA-seq analysis was performed following established protocols 24, 61-63 . Briefly, the scRNA-seq expression matrix was analyzed using the R package "Seurat." Initially, gene expression levels were normalized using the "NormalizeData" function. Next, the "FindVariableFeatures" function was employed to identify the top 2,000 highly variable genes (HVGs) for further analysis. To reduce dimensionality, principal component analysis (PCA) was performed using the "RunPCA" function. Batch effects were corrected using the "Harmony" package, ensuring that the downstream analyses were not biased by inter-sample variability. To identify cell populations, the "FindNeighbors" function was applied to compute the k-nearest neighbors for each cell, followed by the "FindClusters" function to determine optimal clustering. The clustering resolution parameter was set to 0.5 to balance the granularity of cluster identification. To visualize the identified clusters, uniform manifold approximation and projection (UMAP) were utilized. Cluster annotation was conducted by referencing marker genes documented in the original study associated with this dataset 23 . Additionally, the "FeaturePlot" and "DotPlot" functions were utilized to specifically visualize the gene expression across various cell types within the dataset. Mouse studies All mouse experiments were performed according to the regulations of German legislation on animal protection and were approved by the local animal care committee (District Government of Upper Bavaria, GZ: ROB-55.2-2532.Vet_02-18-177). The mice were bred and aged in the German Heart Centre Munich vivarium under standard conditions, following a 12-hour light/dark cycle with free access to food/water, maintaining temperature and humidity. Apolipoprotein E and Cadherin 13 double knockout ( Cdh13 -/- /Apoe -/- ) mice were generated through crossbreeding Cdh13 +/- (purchased from The Jackson Laboratories, Bar Harbor, USA; Cdh13tm1Brns/J , subsequently termed Cdh13 -/ - ) with Apoe tm1Unc (purchased from The Jackson Laboratories, Bar Harbor, USA; subsequently termed Apoe -/- ) mice over five generations. Experimental ages ranged from 12 weeks. Both males and females were included in the experiments, ensuring a balanced distribution of sex across treatment and genotype conditions. The mice were fed a Western Diet (MD88137 Adjusted Calories diet, 42% from fat, Harlan) for 4, 8, or 12 weeks to evaluate the progression of atherosclerosis. Both genotype groups were closely matched in terms of age. Experimental animals that died during rearing, suffered from other diseases, or failed to meet the scoring criteria (supplementary Table 12) were excluded. All remaining mice that met the scoring criteria were included for further experiments and statistical analysis. Histology To evaluate aortic pathology, mice of both sexes in a balanced ratio were sacrificed at predetermined time points, as previously described (Extended Data Figure 1b and 1e). Euthanasia was performed using gaseous isoflurane, followed by transcardiac perfusion with 0.9% saline until complete replacement of circulating blood was achieved. The aorta and heart were exposed by carefully removing the surrounding tissues and subsequently fixed in 4% paraformaldehyde in PBS at 4 °C for 24 hours. The aortic roots were embedded in molds using an optimal cutting temperature (OCT) compound (Sakura Finetek, Tokyo, Japan) and snap-frozen on dry ice. Frozen tissue blocks were sectioned into 5 μm slices and mounted onto microscope slides for further analysis 64 . Enzyme-Linked Immunosorbent Assay Mice arteries’ Cdh13 expression was assessed using a commercially available ELISA kit (abx518770, Abbexa) according to the manufacturer’s protocol. Aortic tissue, spanning from the aortic arch to the iliac artery, was excised, homogenized in 100 µL PBS using a tissue grinder, and centrifuged at maximum speed for 15 minutes at 4°C. The supernatant was collected for protein concentration analysis using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific) and subsequent ELISA assay. All procedures were carried out according to the manufacturer's protocol. A total of 100 µL of each standard, test sample, and control (zero) were added to wells on a pre-coated plate, followed by a 1-hour incubation at 37°C. Subsequently, the plate was washed three times with 1X Wash Buffer, and 100 µL of Detection Reagent B working solution was added to each well. After a 30-minute incubation at 37°C, the solution was discarded, and the plate was washed five times. Next, 90 µL of TMB Substrate was added to each well, and the plate was resealed and incubated for 10–20 minutes at 37°C, protected from light. Lastly, 50 µL of Stop Solution was added to each well, and the absorbance was promptly measured at 450 nm. Oil Red O staining The aortic root were imaged post-Oil Red O (ORO) staining to assess plaque burden. Tissues were briefly washed with PBS, then incubated at 37°C for 30 minutes in 3 mg/mL ORO solution (Sigma-Aldrich, O0625) in 60% isopropanol. Excess dye was removed with 60% isopropanol. Aortic arches were opened, pinned on a black pad, and imaged using a Stemi 2000-C microscope with an Axiocam ERc 5s camera and ZEN 2.3 Blue software. Plaque area at the aortic root was quantified by measuring at the maximum cross-sectional area. Aortic root sections were photographed after hematoxylin restaining of nuclei. ORO-positive areas were quantified using ImageJ, and lesion area percentages were calculated to evaluate atherosclerosis severity 64 . Cell cultures and passaging HEK 293T cells (ATCC, USA) were cultured in high-glucose Dulbecco's Modified Eagle Medium (DMEM; Gibco BRL, Grand Island, USA) supplemented with 10% fetal bovine serum (FBS; Gibco BRL) (supplementary Table 14) . The primary human umbilical vein endothelial cells (HUVEC; PromoCell, pooled donors) were cultured in Endothelial Cell Growth Medium MV 2 (C-22022) supplemented with a complete supplement mix. All cells were maintained at 37°C in a humidified atmosphere with 5% CO 2 . The culture medium was refreshed every two days, and cells (HEK.293T, HUVEC) were passaged upon reaching 90% confluency. Primary artery cells were handled using the DetachKit (PromoCell) following the manufacturer's protocols (supplementary Table 14). The seeding density for HUVECs was maintained between 5,000 and 10,000 cells per cm 2 . THP-1 monocytes (ATCC, USA) were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum (FBS; Gibco BRL) and 1% penicillin-streptomycin. Cells were kept at a density of 2 × 10 5 to 1 × 10 6 cells/mL to ensure optimal growth and viability. Cells were split every 2 - 3 days by diluting them with fresh culture medium. To induce differentiation of THP1 cells into macrophage-like cells, cells were treated with Phorbol 12-myristate 13-acetate (PMA) under a concentration of 100 ng/mL, 72 hours to trigger differentiation. CRISPR and DNA Cloning Plasmids used in this study, including lentiCRISPRv2 (CRISPR knockout), lentiSAMv2 (CRISPR activation), pXR003:CasRx gRNA cloning backbone (dRfxCas13d-based RIP), and psiCHECK™-2 CDH13 3' UTR (RNA interference assay), were commercially purchased, expanded in LB medium with antibiotics, and extracted using the PureYield™ Plasmid Mini/Max-prep System (Promega, A1223 /A2392). Linearization of lentiCRISPRv2 and lentiSAMv2 was performed using Esp3I (Thermo Scientific, FD0454), pXR003:CasRx by BbsI (FD1014), and psiCHECK™-2 by XhoI (FD 0694) and NotI (FD0593). Target fragments were purified via agarose gel electrophoresis and the QIAquick Gel Extraction Kit (28704). rgRNA, sgRNA, and dgRNA oligos were phosphorylated with T4 PNK (NEB M0201S) at 37°C for 30 minutes and respectively ligated with pXR003:CasRx, lentiCRISPRv2, and lentiSAMv2 using Quick Ligase (NEB, M2200S). The synthetic CDH13 3' UTR (1323bp) was cloned into psiCHECK™-2 vector following the same procedures. Recombinant plasmids were transformed into Stbl3 bacteria, expanded, and screened on selective agar plates for positive clones following standard protocols. Cell transfection, lentivirus packaging, and cell infection Plasmid encoding genes of interest or guide RNAs (supplementary Table 6 ,- 8 and 15) were transfected into cell lines using FuGENE® HD (Promega, E5911), following the manufacturer’s protocol. Briefly, cells were seeded to reach 70-80% confluency on the day of transfection. Plasmid DNA, Opti-MEM™ I Reduced Serum Medium (Gibco), and FuGENE® HD were mixed in a ratio corresponding to the seeding surface area. After a 30-minute incubation at room temperature, the transfection complex was gradually added to the cells in the well plate. Transfected cells were incubated in a humidified incubator at 37°C with 5% CO 2 . The time to harvest cells depends on different experiments. To generate lentiviral particles, psPAX2 (Addgene-12260), pCMV-VSV-G (Addgene-8454), and guide RNA-expressing plasmids (supplementary Table 6 and 7) were co-transfected into HEK. 293T cells using the above transfection protocol. After 72 hours, the virus-containing supernatant was harvested, transferred to polypropylene storage tubes, and centrifuged at 2000 × g for 5 minutes to remove residual packaging cells. The clarified supernatant was filtered through a 0.45 µm PES filter, aliquoted, snap-frozen in liquid nitrogen, and stored at -80°C to preserve viral titer. For infection, HUCEV cells were incubated with the collected viral supernatant, diluted 1:1 with fresh culture medium, and supplemented with 10 µg/mL polybrene (1:1000 dilution) to enhance transduction efficiency. Cells were maintained in viral-containing medium for 48 hours before further processing. CRISPR/Cas9-based gene knockdown To knock out CDH13-AS2 using the CRISPR/Cas9 system, we utilized two sgRNAs and spCas9 to excise a 34bp of the shared exon (ENSE00002602225) (Figure. 4a). One lentiviral vector encoded by a sgRNA, spCas9, and a puromycin resistance cassette was used to pack the virus for gene targeting in cells. HUVEC cells were routinely passaged 48 hours after viral infection and subjected to positive selection after 7 days. Puromycin (1 ug/mL) selection was employed to eliminate negative cells after virus infection 27 . CRISPR/Cas9-based transcriptional activation (CRISPRa) To increase the expression of RNAs, we employed CRISPR-mediated transcriptional activation (CRISPRa) 26 . The CRISPRa system includes two components, 1) the spCas9 lacking nuclease activity and fused with a transcription activator VP64 (dCas9-VP64) and 2) a short version of sgRNA (14-15 bp, dgRNA) tagged with hairpin aptamer (MS2) specifically recruiting transcriptional activation complex, MCP:P65:HSF1 (MPH) (Figure. 5a). The sequence of dgRNAs was designed based on the promoter or enhancer of the target gene (Extended Data Figure 3c to 3f). HUVEC cells were routinely passaged 48 hours after viral infection and subjected to positive selection after 7 days. Blasticidin (1 ug/mL) and Hygromycin (0.5 ug/mL) were employed to eliminate negative cells after virus infection. dRfxCas13d-based RNA immunoprecipitation We performed RIP using the CRISPR system based on the RNA-targeting nuclease, Cas13. Several Cas13 orthologs have been found to recognize and target RNAs in mammalian cells, including LwaCas13a (139 kDa), PspCas13b (128 kDa), and RfxCas13d (112 kDa). We selected the enzymatically dead version of the smallest Cas13, dRfxCas13d (dCas13d) 25 , to explore the potential binding interaction among the four antisense lncRNAs and CDH13 . The dCas13d construct was fused with a HA tag and a double-stranded RNA binding domain (dsRBD) stabilizing the dCas13d / rgRNA / target complex. The anti-HA magnetic beads were employed to pull down and enrich the binding RNAs and proteins. Five rgRNAs of CDH13 were used to pull down CDH13 mRNA and its potential binders (supplementary Table 6). The scrambled rgRNAs were used as the control(supplementary Table 6) 25 . RNA immunoprecipitation HUVEC cells (1.5 × 10 7 ) were washed with PBS and lysed in a buffer containing 50 mM Tris-HCl, 100 mM NaCl, 1% NP-40, 0.1% SDS, and 0.5% sodium deoxycholate at pH 7.4, supplemented with a Protease Inhibitor Cocktail Set III (EMD Millipore). Five percent of each lysate was reserved for input RNA preparation, while the remaining lysates were divided into two aliquots for immunoprecipitation. Immunoprecipitation was performed using 10 µg of either a specific antibody (CST, HA-Tag-C29F4, Rabbit mAb #3724) or control IgG (CST, Normal Rabbit IgG #2729), followed by incubation at 4°C for 8 hours. The immunoprecipitants were washed six times with a buffer containing 5 mM Tris-HCl, 150 mM NaCl, and 0.1% Triton X-100. Bound RNA fragments were isolated and purified using QIAzol Lysis Reagent (Qiagen, Cat. No. 79306) following the manufacturer's protocols. The RNA underwent reverse transcription using SuperScript III with a random primer mix (Thermo Fisher, 12574026). The cDNA obtained was amplified using PCR for the target transcripts. Nucleic acid isolation For DNA extraction from cultured cells, the DNeasy Blood & Tissue Kit (Qiagen,Cat. No. 69504) was used according to the manufacturer’s instructions. For the plasmid DNA from E.coli , the PureYield™ Plasmid Mini/Max-prep System, A1223 or A2392 was used following the manufacturer’s instructions. Total RNA was extracted from cultured cells using QIAzol Lysis Reagent (Qiagen, Cat. No. 79306) following a modified Chomczynski and Sacchi protocol (1993). Cells were lysed with QIAzol and harvested using a cell scraper, then transferred to 1.5 mL microcentrifuge tubes. Following an incubation at room temperature, the lysates were vortexed, followed by the addition of 120 µL DNase-/RNase-free water and 100 µL chloroform. Samples were mixed by inversion, incubated at room temperature for another 10 minutes, and centrifuged at 13,000 rpm for 10 minutes at 4°C to separate phases. The aqueous phase was transferred to fresh tubes, combined with ice-cold isopropanol and 1.5 µL GlycoBlue™ coprecipitant (Thermo Fisher Scientific), vortexed, and incubated at -20°C for at least 30 minutes. RNA was pelleted by centrifugation at 14,000 rpm for 30 minutes at 4°C, washed twice with 75% ethanol, and centrifuged 10minutes at same conditions. Pellets were air-dried and resuspended in DNase-/RNase-free water. DNA/RNA quality and concentration were determined using a NanoQuant Plate™ (Tecan) and an Infinite M200Pro spectrophotometer (Tecan), based on the 260/280 nm absorbance ratio. Resuspended RNA was either used immediately or stored at - 80°C. Polymerase chain reaction (PCR) Primers (17 - 30 bp) for PCR were designed using PrimerBank or Primer3 and synthesized by Eurofins Genomics (Ebersberg, Germany) (supplementary Table 16). Stock primers (100 µM) were diluted to 10 µM working solutions with DNase-/RNase-free water and stored at 4°C. cDNA was synthesized from mRNA using Maxima™ H Minus cDNA Synthesis MasterMix (Thermo Fisher Scientific) and stored at -20°C. Amplification PCR used Q5 2 x Master Mix (New England Biolabs) with primers (supplementary Table 16) and 750 ng of cDNA or 150 ng of genomic DNA. PCR products were resolved on 1 - 3% agarose gels depending on the product size, visualized under UV, and documented with Amersham ImageQuant 800 (Cytiva). qPCR was performed with PerfeCTa SYBR® Green FastMix (Quantabio) on the ViiA 7 Real-Time PCR System (Applied Biosystems) using ~100 ng cDNA and primers (supplementary Table 16). Reactions were run in triplicate, normalized to housekeeping genes (ΔCt), and analyzed by the ΔΔCt method to calculate fold changes (2^-ΔΔCt) relative to controls. Protein isolation and Western blot analysis All procedures were performed on ice unless stated otherwise. Cells were lysed in 1x RIPA buffer (NEB, 9806S) with HALT™ protease inhibitor (1:100, Thermo Fisher Scientific), incubated at 4°C for 30 minutes, and centrifuged at ~13,000 rpm for 30 minutes. Supernatants were collected, and protein concentrations were measured using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific) with absorbance at 562 nm (Infinite M200Pro spectrophotometer - Tecan). For Western blotting, 30 µg of protein was mixed with Laemmli buffer, boiled at 95°C, and resolved on 4 - 20% gradient gels (Bio-Rad) at 120 V. Proteins were transferred to PVDF membranes, blocked with 5% BSA, and incubated overnight with primary antibodies (supplementary Table 13). After washing, membranes were incubated with HRP-conjugated secondary antibodies (supplementary Table 13) and visualized using SuperSignal™ West Dura substrate on the Amersham ImageQuant 800 system (Cytiva). Wound healing migration assay To perform the HUVEC wound healing migration assay, sterile forceps were used to position inserts (ibidi, No. 80369) into the plate wells, ensuring consistent alignment of the “wound field”. The cell suspension was prepared at a concentration of 0.5 x 10 6 cells/mL in medium, and 300 µL of this suspension was carefully added to each well without disturbing the inserts. The plate was incubated overnight in a cell culture incubator to facilitate the formation of a confluent monolayer. To start the assay, inserts were gently removed using sterile forceps. The media was aspirated, and the wells were washed with fresh media to eliminate dead cells and debris. Following washing, medium containing 2.5 µg/mL of mitomycin C was added to inhibit cell proliferation, ensuring wound closure was driven by cell migration. The wells were observed under a light microscope, and additional washing was conducted if necessary. Cells were subsequently incubated, and wound closure was monitored at 24-, 48-, and 72-hours post-insert removal using light microscopy. ImageJ software was used to measure the percentage of closure or the rate of cell migration into the wound field. BrdU proliferation assay For the cell proliferation assay, immunofluorescent staining of incorporated bromodeoxyuridine (BrdU) was performed using the BD Pharmingen™ APC BrdU Flow Kit (Catalog No. 557892), followed by the manufacturer’s protocol. To label cells with BrdU, 10 μL of a 1 mM BrdU solution was added per mL of culture medium, ensuring a cell density of no more than 1 x 10 6 cells/mL to maintain normal cell cycling. After a 15-hour incubation, BrdU-plused cells were detached following washing and centrifugation at 200 g for 5 minutes. The cells were fixed and permeabilized on ice for 30 minutes using BD Cytofix/Cytoperm Buffer, followed by further permeabilization with BD Cytoperm Permeabilization Buffer Plus and refixed for 10 minutes. To detect incorporated BrdU, the cells were treated with DNase (300 µg/mL) and then incubated at 37 °C for 1 hour. After washing, the cells were resuspended in 50 µL of BD Perm/Wash Buffer containing diluted fluorescent anti-BrdU and incubated at 4°C for 20 minutes. The analysis of stained cells was conducted using a BD LSRFortessa flow cytometer (BD Biosciences) at a low flow rate to achieve optimal resolution. Gating was conducted for viable cells (forward scatter area [FSC-A] versus side scatter area [SSC-A]) and single cells (FSC-A versus forward scatter width [FSC-W], SSC-A versus side scatter width). Data analysis and statistical plotting were performed using FlowJo software v.9.9.6 (FlowJo LLC). Adhesion assay HUVEC endothelial cells were seeded at 5,000 cells/cm² on gelatin-coated 6-well or 12-well plates and cultured for 48 hours to form a confluent monolayer. THP1 monocytes were labeled with calcein dye (Invitrogen™, C1430) at a 1:1000 dilution, prepared at a concentration of 1.0 x 10 6 cells/mL in serum-free RPMI-1640 medium, and incubated at 37°C for 60 minutes. After centrifugation, the supernatant was extracted, and the THP1 cells underwent two washes with serum-free RPMI-1640 before being resuspended at the same concentration for application onto HUVECs. The HUVEC endothelial cell medium was substituted with serum-free RPMI-1640, and 1 mL of the stained monocyte suspension was introduced into each well with the endothelial monolayer. Co-cultures were incubated for 180 minutes under standard conditions. After incubation, non-adherent leukocytes were removed by washing the wells three times with 2 mL of PBS. After the final wash, 500 μL of 1 X RIPA Lysis Buffer was added to each well for a 5-minute incubation at room temperature. After 14,000 X g centrifugation, 100 μL of lysate was then transferred to a 96-well fluorescence-compatible plate. Fluorescence was measured using the plate reader (Infinite M200Pro spectrophotometer - Tecan) at excitation/emission wavelengths of 480nm/520nm. A standard curve was generated based on a gradient of stained THP1 cells, and the number of adhered cells was calculated based on the fluorescence intensity, i-Control software was used for data collection (Tecan). For adhesion confocal microscopy images, endothelial cells were double-stained with Phalloidin (Invitrogen™, A12381) and DAPI, and images were captured after monocyte addition, incubation, and washing. Apoptosis assay Apoptosis assay for gene-edited HUVECs was conducted using the RealTime-Glo™ Annexin V Apoptosis kit (Promega, JA1000) according to the manufacturer’s protocol. This live-cell, non-lytic, real-time assay detects phosphatidylserine (PS) exposure on the outer leaflet of the cell membrane during apoptosis. HUVEC cells were plated at a density of 5,000 cells/cm² on gelatin-coated 96-well plates, incubated for 48 hours to establish a monolayer, and then treated with 100 ng/mL LPS. A 2X Detection Reagent was created by diluting each component 500-fold in the complete cell culture medium. An equal volume (100 µL) of the 2X Detection Reagent was added to each well. The assay was incubated for various time intervals, and apoptosis was assessed by measuring relative fluorescence units (RFU) using green fluorescence with excitation at 485 nm and emission at 525 - 530 nm (Infinite M200Pro spectrophotometer - Tecan). Fluorescence units (RFU) were assessed using M200Pro (Tecan) and i-Control software (Tecan). RNA stability assay For the RNA stability assay, we treated cells with 10 μg/mL actinomycin D, a transcription inhibitor. The negative control (mock) was a medium with 10 μl/mL DMSO. After 48 hours of CRISPRa in cells, the culture medium was replaced with 10 μg/mL actinomycin D medium or mock medium and incubated at 37°C with 5% CO 2 . RNA samples were obtained after 0, 3, and 6 hours of treatment for RNA degradation comparison by qPCR. RNA interference assay RNA interference assay was used for assessing potential microRNA binding on CDH13 3’UTR and conducted using the Dual-Luciferase™ Reporter Assay System (Promega, E1910) following the manufacturer’s instructions. The relevant 3’UTR region of CDH13 consists of 1323 bp. The 1323bp sequence (supplementary Table 17) flanking the Xhol and Notl cutting sites was synthesized via GenScript Biotech (Netherlands) B.V. and inserted into the psiCHECK™-2 Vector backbone (Promega, C8021) via restriction enzyme cloning. For the assay, HEK.293T cells were seeded in 12-well plates and incubated at 37°C with 5% CO2. Upon reaching 70-80% confluency, cells were co-transfected in triplicate with 600 ng of the plasmid DNA construct and 50 nmol of microRNA. Following 48 hours, the cells were cooled on ice, rinsed once with 500 μL ice-cold 1× PBS, and then lysed with 50 μL 1× Passive Lysis Buffer (Promega, E1910). After centrifugation, 10 μL of the lysate was transferred to a white 96-well plate for a dual luciferase assay following the manufacturer’s protocol. Luciferase activities were assessed using M200Pro (Tecan) and i-control software (Tecan). The luciferase activity was quantified as the ratio of Renilla to Firefly luciferase activity and reported in arbitrary light units. Statistical analysis The data of biological experiments were obtained from a minimum of three independent biological replicates. Quantitative results are expressed as the mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism, applying unpaired t-tests, one-way ANOVA with multiple comparisons, or two-way ANOVA for factorial analyses, as appropriate. Statistical significance was defined as *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001. Description of sample size (number of LoF mutation carriers, non-carriers, total sample size) and statistical tests performed for genetic association analysis, can be found in the corresponding sections of the main text, Method details, and Supplementary Tables. Data and code availability This paper analyzes existing data of single-cell RNA-seq data from Gene Expression Omnibus (GEO), dataset GSE131778, GSE247238, and GSE247238 . Original western blot images will be deposited at the required platform and publicly available as of the date of publication. Microscopy data reported in this paper will be shared by the lead contact upon request. All original code will be deposited at the required platform and will be publicly available as of the date of publication. Any additional information required to reanalyze the data reported in this paper will be available from the lead contact upon request. Declarations Data and code availability This paper analyzes existing data of single-cell RNA-seq data from Gene Expression Omnibus (GEO), dataset GSE131778, GSE247238, and GSE247238 . Original western blot images will be deposited at the required platform and publicly available as of the date of publication. Microscopy data reported in this paper will be shared by the lead contact upon request. All original code will be deposited at the required platform and will be publicly available as of the date of publication. Any additional information required to reanalyze the data reported in this paper will be available from the lead contact upon request. Author contributions Z.C., S.L., and H.S. conceptualized the study, designed the experiments, supervised and led the trainees, and co-wrote the manuscript. Z.C. and S.L. conducted cellular and molecular experiments, analyzed data, and visualized data. S.L. and X.S. performed animal experiments and analyzed data. A.D. conducted genetic analyses. L.L. contributed to GWAS and colocalization analysis. A.M. led trainees and contributed to flow cytometry and FACS data analysis. L.M. provided human carotid artery data and Z.L. conducted the corresponding analysis. Y.C. conducted single-cell data analysis. T.D. led trainees and designed mouse experiments. M.L. contributed to animal tissue harvest. R.S. conducted flow cytometry and analyzed FACS data. C.L. and A.R. cloned the CRISPR constructs. L.F.B. quantified and analyzed immunostaining data. N.B. and M.C. explored the SMC-specific eQTLs at the 16q23.3 locus. C. E. R and, A. J. L explored the EC-specific eQTLs at the 16q23.3 locus. H.S. R.B. and T.K. helped with data integration and edited the manuscript. M.S., J.K., N.K., A.S.and M.V.S. advised the methods of statistical analyses and edited the manuscript. L.M. and J.L.M.B. provided the STARNET dataset. M.N.J provided the materials, supervised the microRNA-related experiments, and edited the manuscript. Acknowledgments We acknowledge UK Biobank (project code 25214), and CARDIoGRAMplusC4D Consortium, and the Common Metabolic Diseases Knowledge Portal for providing research data. We thank the animal facility of the German Heart Center Munich and the animal welfare officer, Dr. Susanne Naumann for supporting our mouse experiments. We thank Dr. Zhiyuan Wu who explored RNA sequencing data of atherosclerosis plaques from patients undergoing carotid endarterectomy for this project. We thank Christopher Wolf, an experiment assistant, who supported our experiment. Sources of Funding We appreciate the major funding for the current projects, including Sonderforschungsbereich SFB TRR 267 (DFG, 403584255, project B05_Z.C. & H.S.), the German Research Foundation (DFG 510049865), Corona-Stiftung Nachwuchsforschungsgruppe (Junior Research Group Grant), and the German Centre for Cardiovascular Research (DZHK) (“Förderkennzeichen”, ID: 81X3600510) (Z.C.). The work was further supported by the German Federal Ministry of Education and Research (BMBF) within the scheme of target validation (BlockCAD: 16GW0198K), COMMITMENT (01ZX1904A), e:Med research and funding concept (AbCD-Net: 01ZX1706C) (H.S.). As a Co-applicant of the British Heart Foundation (BHF)/DZHK-collaboration (DZHK-BHF: 81X2600522) and the Leducq Foundation for Cardiovascular Research (PlaqOmics: 18CVD02), we gratefully acknowledge their funding (H.S.). Further, we recognize the support of the Bavarian State Ministry of Health and Care which funded this work with DigiMed Bayern (grant No: DMB-1805–0001) within its Masterplan “Bayern Digital II” and of the German Federal Ministry of Economics and Energy in its scheme of ModulMax (grant No: ZF4590201BA8) (H.S.). Finally, we thank the support from National Institutes of Health (NIH) grants DK136405 and DK117850 (A.J.L.), NIH grant R01HL156120 (M.C.), and an American Heart Association Established Investigator Award 24EIA1258067 (M.C.). Disclosures NA. The authors declare no competing interests. Correspondence and requests for materials Further information and requests for resources and reagents should be directed to and will be fulfilled by Zhifen Chen ( [email protected] ) and Shuangyue Li ( [email protected] ). References Aragam, K.G. et al. Discovery and systematic characterization of risk variants and genes for coronary artery disease in over a million participants. Nat Genet 54 , 1803-1815 (2022). Tcheandjieu, C. et al. Large-scale genome-wide association study of coronary artery disease in genetically diverse populations. Nat Med 28 , 1679-1692 (2022). Chen, Z. & Schunkert, H. Genetics of coronary artery disease in the post-GWAS era. J Intern Med 290 , 980-992 (2021). Musunuru, K. & Kathiresan, S. Genetics of Common, Complex Coronary Artery Disease. Cell 177 , 132-145 (2019). Hodonsky, C.J. et al. Multi-ancestry genetic analysis of gene regulation in coronary arteries prioritizes disease risk loci. Cell Genom 4 , 100465 (2024). Kim, H.J. et al. Molecular mechanisms of coronary artery disease risk at the PDGFD locus. Nat Commun 14 , 847 (2023). Cui, G. et al. A functional variant in APOA5/A4/C3/A1 gene cluster contributes to elevated triglycerides and severity of CAD by interfering with microRNA 3201 binding efficiency. J Am Coll Cardiol 64 , 267-277 (2014). Cheng, P. et al. Smad3 regulates smooth muscle cell fate and mediates adverse remodeling and calcification of the atherosclerotic plaque. Nat Cardiovasc Res 1 , 322-333 (2022). de Goede, O.M. et al. Population-scale tissue transcriptomics maps long non-coding RNAs to complex disease. Cell 184 , 2633-2648 e2619 (2021). Werner, A., Kanhere, A., Wahlestedt, C. & Mattick, J.S. Natural antisense transcripts as versatile regulators of gene expression. Nat Rev Genet 25 , 730-744 (2024). Farina, F.M., Weber, C. & Santovito, D. The emerging landscape of non-conventional RNA functions in atherosclerosis. Atherosclerosis 374 , 74-86 (2023). Faghihi, M.A. et al. Evidence for natural antisense transcript-mediated inhibition of microRNA function. Genome Biol 11 , R56 (2010). Shang, R., Lee, S., Senavirathne, G. & Lai, E.C. microRNAs in action: biogenesis, function and regulation. Nat Rev Genet 24 , 816-833 (2023). Chen, L.L. & Kim, V.N. Small and long non-coding RNAs: Past, present, and future. Cell 187 , 6451-6485 (2024). Dzau, V.J. & Hodgkinson, C.P. RNA Therapeutics for the Cardiovascular System. Circulation 149 , 707-716 (2024). Nemeth, K., Bayraktar, R., Ferracin, M. & Calin, G.A. Non-coding RNAs in disease: from mechanisms to therapeutics. Nat Rev Genet 25 , 211-232 (2024). Bauersachs, J. et al. Efficacy and safety of CDR132L in patients with reduced left ventricular ejection fraction after myocardial infarction: Rationale and design of the HF-REVERT trial. Eur J Heart Fail 26 , 674-682 (2024). Nelson, C.P. et al. Association analyses based on false discovery rate implicate new loci for coronary artery disease. Nat Genet 49 , 1385-1391 (2017). van der Harst, P. & Verweij, N. Identification of 64 Novel Genetic Loci Provides an Expanded View on the Genetic Architecture of Coronary Artery Disease. Circ Res 122 , 433-443 (2018). Li, L. et al. Transcriptome-wide association study of coronary artery disease identifies novel susceptibility genes. Basic Res Cardiol 117 , 6 (2022). Koplev, S. et al. A mechanistic framework for cardiometabolic and coronary artery diseases. Nat Cardiovasc Res 1 , 85-100 (2022). Fasolo, F. et al. Long Noncoding RNA MIAT Controls Advanced Atherosclerotic Lesion Formation and Plaque Destabilization. Circulation 144 , 1567-1583 (2021). Paloschi, V. et al. Utilization of an Artery-on-a-Chip to Unravel Novel Regulators and Therapeutic Targets in Vascular Diseases. Adv Healthc Mater 13 , e2302907 (2024). Wirka, R.C. et al. Atheroprotective roles of smooth muscle cell phenotypic modulation and the TCF21 disease gene as revealed by single-cell analysis. Nat Med 25 , 1280-1289 (2019). Han, S. et al. RNA-protein interaction mapping via MS2- or Cas13-based APEX targeting. Proc Natl Acad Sci U S A 117 , 22068-22079 (2020). Liao, H.K. et al. In Vivo Target Gene Activation via CRISPR/Cas9-Mediated Trans-epigenetic Modulation. Cell 171 , 1495-1507 e1415 (2017). Chen, Z. et al. Functional Screening of Candidate Causal Genes for Insulin Resistance in Human Preadipocytes and Adipocytes. Circ Res 126 , 330-346 (2020). Wang, J. et al. LncRNA PTTG3P promotes tumorigenesis and metastasis of NSCLC by binding with ILF3 to maintain mRNA stability and form a positive feedback loop with E2F1. Int J Biol Sci 19 , 4291-4310 (2023). Wang, Y. et al. LncRNA LINRIS stabilizes IGF2BP2 and promotes the aerobic glycolysis in colorectal cancer. Mol Cancer 18 , 174 (2019). Liu, G., Zhang, R., Xu, J., Wu, C.I. & Lu, X. Functional conservation of both CDS- and 3'-UTR-located microRNA binding sites between species. Mol Biol Evol 32 , 623-628 (2015). Huang, H.Y. et al. miRTarBase 2020: updates to the experimentally validated microRNA-target interaction database. Nucleic Acids Res 48 , D148-D154 (2020). Sticht, C., De La Torre, C., Parveen, A. & Gretz, N. miRWalk: An online resource for prediction of microRNA binding sites. PLoS One 13 , e0206239 (2018). Soutschek, M., Gross, F., Schratt, G. & Germain, P.L. scanMiR: a biochemically based toolkit for versatile and efficient microRNA target prediction. Bioinformatics 38 , 2466-2473 (2022). McGeary, S.E. et al. The biochemical basis of microRNA targeting efficacy. Science 366 (2019). Agarwal, V., Bell, G.W., Nam, J.W. & Bartel, D.P. Predicting effective microRNA target sites in mammalian mRNAs. Elife 4 (2015). Kavakiotis, I., Alexiou, A., Tastsoglou, S., Vlachos, I.S. & Hatzigeorgiou, A.G. DIANA-miTED: a microRNA tissue expression database. Nucleic Acids Res 50 , D1055-D1061 (2022). Fujishima, Y. et al. Adiponectin association with T-cadherin protects against neointima proliferation and atherosclerosis. FASEB J 31 , 1571-1583 (2017). Ishii, N. et al. Identification of a novel non-coding RNA, MIAT, that confers risk of myocardial infarction. J Hum Genet 51 , 1087-1099 (2006). Samani, N.J. et al. Genomewide association analysis of coronary artery disease. N Engl J Med 357 , 443-453 (2007). Singh, R. et al. The hidden players: Shedding light on the significance of post-translational modifications and miRNAs in Alzheimer's disease development. Ageing Res Rev 90 , 102002 (2023). Wong, C.H. et al. CircRTN4 promotes pancreatic cancer progression through a novel CircRNA-miRNA-lncRNA pathway and stabilizing epithelial-mesenchymal transition protein. Mol Cancer 21 , 10 (2022). Jonas, S. & Izaurralde, E. Towards a molecular understanding of microRNA-mediated gene silencing. Nat Rev Genet 16 , 421-433 (2015). Wu, L., Fan, J. & Belasco, J.G. MicroRNAs direct rapid deadenylation of mRNA. Proc Natl Acad Sci U S A 103 , 4034-4039 (2006). Peter, M.E. Targeting of mRNAs by multiple miRNAs: the next step. Oncogene 29 , 2161-2164 (2010). Zhu, Y. et al. LncRNA MIR200CHG inhibits EMT in gastric cancer by stabilizing miR-200c from target-directed miRNA degradation. Nat Commun 14 , 8141 (2023). Ridker, P.M. et al. Lipid-Reduction Variability and Antidrug-Antibody Formation with Bococizumab. N Engl J Med 376 , 1517-1526 (2017). Koren, M.J. et al. Preclinical development and phase 1 trial of a novel siRNA targeting lipoprotein(a). Nat Med 28 , 96-103 (2022). Rosenson, R.S. et al. Evinacumab in Patients with Refractory Hypercholesterolemia. N Engl J Med 383 , 2307-2319 (2020). Ballantyne, C.M. et al. Plozasiran, an RNA Interference Agent Targeting APOC3, for Mixed Hyperlipidemia. N Engl J Med 391 , 899-912 (2024). Kraler, S. et al. Resilience of the Internal Mammary Artery to Atherogenesis: Shifting From Risk to Resistance to Address Unmet Needs. Arterioscler Thromb Vasc Biol 41 , 2237-2251 (2021). Loop, F.D. Internal-thoracic-artery grafts. Biologically better coronary arteries. N Engl J Med 334 , 263-265 (1996). Lucas, T., Bonauer, A. & Dimmeler, S. RNA Therapeutics in Cardiovascular Disease. Circ Res 123 , 205-220 (2018). Ballantyne, M.D., McDonald, R.A. & Baker, A.H. lncRNA/MicroRNA interactions in the vasculature. Clin Pharmacol Ther 99 , 494-501 (2016). Caporali, A. et al. Non-coding RNAs as therapeutic targets and biomarkers in ischaemic heart disease. Nat Rev Cardiol 21 , 556-573 (2024). Mishra, A. et al. Stroke genetics informs drug discovery and risk prediction across ancestries. Nature 611 , 115-123 (2022). Franzen, O. et al. Cardiometabolic risk loci share downstream cis- and trans-gene regulation across tissues and diseases. Science 353 , 827-830 (2016). Giambartolomei, C. et al. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet 10 , e1004383 (2014). Wang, Y. et al. Myeloid cannabinoid CB1 receptor deletion confers atheroprotection in male mice by reducing macrophage proliferation in a sex-dependent manner. Cardiovasc Res 120 , 1411-1426 (2024). Backman, J.D. et al. Exome sequencing and analysis of 454,787 UK Biobank participants. Nature 599 , 628-634 (2021). Mbatchou, J. et al. Computationally efficient whole-genome regression for quantitative and binary traits. Nat Genet 53 , 1097-1103 (2021). Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol 36 , 411-420 (2018). Stuart, T. et al. Comprehensive Integration of Single-Cell Data. Cell 177 , 1888-1902 e1821 (2019). Becht, E. et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat Biotechnol (2018). Sharifi, M.A. et al. ADAMTS-7 Modulates Atherosclerotic Plaque Formation by Degradation of TIMP-1. Circ Res 133 , 674-686 (2023). Additional Declarations There is NO Competing Interest. Supplementary Files ExDFNCRRNAsat16q23Li.pdf supplement figure from 1 to 6 supplementtableNCRRNAsat16q23Li.pdf supplement table from 1 to 17 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7333062","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":502178270,"identity":"2e54bead-f2b6-462e-9412-8f13782aceac","order_by":0,"name":"Shuangyue Li","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Shuangyue","middleName":"","lastName":"Li","suffix":""},{"id":502178271,"identity":"688ff85a-8f7f-45ef-8f76-9a1df6aff2fd","order_by":1,"name":"Xiaoning Song","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Xiaoning","middleName":"","lastName":"Song","suffix":""},{"id":502178272,"identity":"67cc8bd1-b800-47a3-b140-f200037a75d9","order_by":2,"name":"Anastasiia Diagel","email":"","orcid":"https://orcid.org/0009-0001-4512-9874","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Anastasiia","middleName":"","lastName":"Diagel","suffix":""},{"id":502178273,"identity":"29762502-c276-48bf-b51e-b326b56c149d","order_by":3,"name":"Ling Li","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany\tTUM School of Computation, Information and Technology (CIT), Technical University Munich, 85748 Garching bei München, Germany","correspondingAuthor":false,"prefix":"","firstName":"Ling","middleName":"","lastName":"Li","suffix":""},{"id":502178274,"identity":"e3d40f4a-b8b8-47af-80ae-c02132373f70","order_by":4,"name":"Aldo Moggio","email":"","orcid":"https://orcid.org/0000-0001-6513-5406","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich","correspondingAuthor":false,"prefix":"","firstName":"Aldo","middleName":"","lastName":"Moggio","suffix":""},{"id":502178276,"identity":"627af6ad-3735-43a6-9891-0529caa1ff7a","order_by":5,"name":"Yifan Chen","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Yifan","middleName":"","lastName":"Chen","suffix":""},{"id":502178275,"identity":"565eb360-b908-4b9d-b2bf-8d0d848570e1","order_by":6,"name":"Zhaolong Li","email":"","orcid":"","institution":"Molecular Vascular Medicine, Klinikum rechts der Isar - Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Zhaolong","middleName":"","lastName":"Li","suffix":""},{"id":502178277,"identity":"8e42d0fd-2926-4465-bf24-7951f1d2bc60","order_by":7,"name":"Tan Dang","email":"","orcid":"https://orcid.org/0009-0005-1770-5765","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Tan","middleName":"","lastName":"Dang","suffix":""},{"id":502178278,"identity":"f0215c1e-a0d6-401e-8164-4d7dfb6c5826","order_by":8,"name":"Miaomiao Li","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Miaomiao","middleName":"","lastName":"Li","suffix":""},{"id":502178279,"identity":"aad39727-cc41-4830-a1b0-d6f66acef5e3","order_by":9,"name":"Rui Shen","email":"","orcid":"https://orcid.org/0009-0000-8807-9523","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Shen","suffix":""},{"id":502178280,"identity":"52264cd8-b83b-4077-81a4-bdf13c6fedbb","order_by":10,"name":"Angela Ma","email":"","orcid":"","institution":"Department of Genetics \u0026 Genomic Sciences, Institute of Genomics and Multiscale Biology, Icahn School of Medicine at Mount Sinai","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Ma","suffix":""},{"id":502178281,"identity":"e5cd2060-f1c2-42f9-9bc6-865aa117fbf5","order_by":11,"name":"Marius Schwab","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich","correspondingAuthor":false,"prefix":"","firstName":"Marius","middleName":"","lastName":"Schwab","suffix":""},{"id":502178282,"identity":"7c529edd-83c0-4643-96b3-a86622390db1","order_by":12,"name":"Nicolas Barbera","email":"","orcid":"","institution":"Center for Public Health Genomics, University of Virginia","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Barbera","suffix":""},{"id":502178283,"identity":"e8c52aeb-f0a9-4bc1-9c91-60639f39949e","order_by":13,"name":"Constanze Lehertshuber","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Constanze","middleName":"","lastName":"Lehertshuber","suffix":""},{"id":502178284,"identity":"f066f626-1da1-4ca3-b546-51a844889357","order_by":14,"name":"Amos Romer","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Amos","middleName":"","lastName":"Romer","suffix":""},{"id":502178285,"identity":"b19373eb-6ed5-4baf-be80-b53ba86dff3d","order_by":15,"name":"Luigi Filippo Brizzi","email":"","orcid":"https://orcid.org/0009-0007-5917-1057","institution":"Department of Cardiac-Thoracic-Vascular Sciences and Public Health, University of Padova, Padova, Italy.","correspondingAuthor":false,"prefix":"","firstName":"Luigi","middleName":"Filippo","lastName":"Brizzi","suffix":""},{"id":502178286,"identity":"5116b9c8-3b81-4250-8077-23bba3ec3a3f","order_by":16,"name":"Johannes Krefting","email":"","orcid":"https://orcid.org/0000-0002-7755-0954","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Johannes","middleName":"","lastName":"Krefting","suffix":""},{"id":502178287,"identity":"3042de7d-4c75-47f3-81ad-75b14da94d54","order_by":17,"name":"Nils Krüger","email":"","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Nils","middleName":"","lastName":"Krüger","suffix":""},{"id":502178288,"identity":"228c30a0-8fd9-49c4-a9aa-8f16f093ea11","order_by":18,"name":"Hendrik Sager","email":"","orcid":"https://orcid.org/0000-0002-0197-6553","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Hendrik","middleName":"","lastName":"Sager","suffix":""},{"id":502178289,"identity":"31475c48-f97b-4ccf-80d1-c5bcede3b189","order_by":19,"name":"Reinier Boon","email":"","orcid":"https://orcid.org/0000-0001-7944-9748","institution":"Goethe University Hospital, Institute for Cardiovascular Regeneration","correspondingAuthor":false,"prefix":"","firstName":"Reinier","middleName":"","lastName":"Boon","suffix":""},{"id":502178290,"identity":"94bd33f8-0794-4b0f-b349-165ae17ff4ce","order_by":20,"name":"Mete Civelek","email":"","orcid":"","institution":"Center for Public Health Genomics, University of Virginia\tDepartment of Biomedical Engineering, University of Virginia","correspondingAuthor":false,"prefix":"","firstName":"Mete","middleName":"","lastName":"Civelek","suffix":""},{"id":502178291,"identity":"b0f8ff36-c510-425e-a7d3-82f2a2d9edc7","order_by":21,"name":"Casey Romanoski","email":"","orcid":"","institution":"Department of Cellular \u0026 Molecular Medicine, The University of Arizona","correspondingAuthor":false,"prefix":"","firstName":"Casey","middleName":"","lastName":"Romanoski","suffix":""},{"id":502178292,"identity":"52d24791-c671-428e-9f1a-a3e9be357784","order_by":22,"name":"Aldons Lusis","email":"","orcid":"https://orcid.org/0000-0001-9013-0228","institution":"David Geffen School of Medicine, University of California","correspondingAuthor":false,"prefix":"","firstName":"Aldons","middleName":"","lastName":"Lusis","suffix":""},{"id":502178295,"identity":"539fb1bf-d6f1-4610-b884-146e2383aad9","order_by":23,"name":"Andreas Schober","email":"","orcid":"","institution":"Institute for Cardiovascular Prevention (IPEK), Ludwig-Maximilians-Universität München\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Schober","suffix":""},{"id":502178293,"identity":"1bbf7cd2-bc46-4d6b-a529-99c383697440","order_by":24,"name":"Thorsten Kessler","email":"","orcid":"https://orcid.org/0000-0003-3326-1621","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Thorsten","middleName":"","lastName":"Kessler","suffix":""},{"id":502178296,"identity":"5471d302-70ad-43a8-8a5b-a734ed4a19a1","order_by":25,"name":"Moritz von Scheidt","email":"","orcid":"https://orcid.org/0000-0001-7159-8271","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Moritz","middleName":"","lastName":"von Scheidt","suffix":""},{"id":502178297,"identity":"0e35b132-fdf9-4b60-868c-d04b625434b8","order_by":26,"name":"Johan Björkegren","email":"","orcid":"","institution":"Department of Medicine, Huddinge, Karolinska Institutet, Karolinska Universitetssjukhuset\tDepartment of Genetics \u0026 Genomic Sciences, Institute of Genomics and Multiscale Biology, Icahn School of Medicine at Mount Sinai","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"","lastName":"Björkegren","suffix":""},{"id":502178294,"identity":"912c0c64-b5e0-4005-83ef-5cac412946be","order_by":27,"name":"Lars Maegdefessel","email":"","orcid":"https://orcid.org/0000-0001-5228-2634","institution":"Molecular Vascular Medicine, Klinikum rechts der Isar - Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Lars","middleName":"","lastName":"Maegdefessel","suffix":""},{"id":502178298,"identity":"d9151c9c-081c-4ed2-bb39-609bab5af3c1","order_by":28,"name":"Maliheh Nazari-Jahantigh","email":"","orcid":"","institution":"Institute for Cardiovascular Prevention (IPEK), Ludwig-Maximilians-Universität München\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Maliheh","middleName":"","lastName":"Nazari-Jahantigh","suffix":""},{"id":502178299,"identity":"35871e3b-989b-4acc-8291-d2cd74986a12","order_by":29,"name":"Heribert Schunkert","email":"","orcid":"https://orcid.org/0000-0001-6428-3001","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":false,"prefix":"","firstName":"Heribert","middleName":"","lastName":"Schunkert","suffix":""},{"id":502178269,"identity":"049f5d8f-372c-4c0e-bb2a-ca48bc552904","order_by":30,"name":"Zhifen Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYDACCcaGDyDagIGB8QFY5ABhLY0zoFqYDQ4Qp4WBEaaFTYIoLfKzmxsbPu5gkDdn7z1W/bFtGwPf8Qb8WhjnHGxsnHmGwXBnz7m0GwfbbjNIniFgDbNEYvtj3jaGBIMbOWZgLQY3EvBrYZNIbGz+C9Jy/41ZAVjL/Qf4tfCAtDCCbeExY4DYgl8HgwRQS2Nvm4ThhjM5xhJnzt3mkTxDwGHyM9IfNvxss5E3OH7G8ENF2W05vuMHCFgDtQzhUqLUj4JRMApGwSjADwBsu0s3hfafMAAAAABJRU5ErkJggg==","orcid":"","institution":"Department of Cardiology, German Heart Center, TUM University Hospital, TUM School of Medicine and Health, Technical University Munich\tDeutsches Zentrum für Herz- und Kreislaufforschung (DZHK), partner site Munich Heart Alliance (MHA), Munich, Germany","correspondingAuthor":true,"prefix":"","firstName":"Zhifen","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-08-09 10:00:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7333062/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7333062/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89463983,"identity":"1e161da0-d924-4dc1-a1fd-b668e902e4b0","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1576013,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li1.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/4fa2badcc3a96794ceea5493.png"},{"id":89463986,"identity":"b8f8bf95-87c1-484f-b27b-3d89c9cf3f5f","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1315320,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li2.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/8790227695d3d0da7e7ea639.png"},{"id":89463987,"identity":"a486f797-2f61-46d3-851f-d0648382b7f9","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":786824,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li3.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/5425ba383b6ec97c69437bac.png"},{"id":89464750,"identity":"1f1ea1f1-81cf-481a-80c3-0add2313eb36","added_by":"auto","created_at":"2025-08-20 08:23:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1843847,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li4.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/5f4aa1d471d1865676ea924a.png"},{"id":89463984,"identity":"fcdedbdd-75e5-4bac-abc5-af712f04a386","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2010400,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li5.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/4a9f7478cbd710f2f79ca8d0.png"},{"id":89465669,"identity":"a7baa340-f0c7-4263-977b-e36e5dcd3d0d","added_by":"auto","created_at":"2025-08-20 08:31:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1042337,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li6.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/7521d9a2cce1faef604f3f67.png"},{"id":89463988,"identity":"20ec61cc-3bb8-4e83-8d98-410af3d9c7c4","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":981728,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"FigureNCRRNAsat16q23Li7.png","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/369ca434165275dd1405b33c.png"},{"id":95000467,"identity":"70c7208f-c84d-4383-bea1-85b62515996e","added_by":"auto","created_at":"2025-11-03 08:57:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10783834,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/cffd1bc6-ca0a-43df-b6a0-e8f62eddf909.pdf"},{"id":89464755,"identity":"85b2b010-5773-4a5d-8a43-c3088ef24e03","added_by":"auto","created_at":"2025-08-20 08:23:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6292041,"visible":true,"origin":"","legend":"supplement figure from 1 to 6","description":"","filename":"ExDFNCRRNAsat16q23Li.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/67cf92865ae6d73bdb63118f.pdf"},{"id":89463990,"identity":"1d46dfe3-cd63-4faf-9423-b4eb78af5a74","added_by":"auto","created_at":"2025-08-20 08:15:37","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":11075502,"visible":true,"origin":"","legend":"supplement table from 1 to 17","description":"","filename":"supplementtableNCRRNAsat16q23Li.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7333062/v1/73cb273a8f9604c048679574.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"\u003cp\u003eAn interplay of non-coding RNAs regulates \u003cem\u003eCDH13\u003c/em\u003e expression and affects endothelial function and coronary artery disease risk\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCoronary artery disease (CAD) is a genetically-mediated and often devastating common disease. Genome-wide association studies (GWAS) identified hundreds of risk alleles that cumulatively shape a population-wide disposition for atherosclerosis \u003csup\u003e1-3\u003c/sup\u003e, i.e. the build-up of lipid-rich inflammatory plaques in arterial vessels, which ultimately may occlude and cause myocardial infarction. CAD is also a prototypic polygenic disease that is primarily precipitated by disturbed gene expression rather than by structural changes of the affected proteins\u003csup\u003e1, 4, 5\u003c/sup\u003e. However, the molecular mechanisms that link CAD risk alleles with altered gene expression have been elaborated in only a few exceptional cases \u003csup\u003e6-8\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIn human biology, a large fraction of\u003c/strong\u003e key regulators of \u003cstrong\u003etissue-specific gene expression are\u0026nbsp;\u003c/strong\u003enon-coding RNAs (ncRNAs), such as long non-coding RNA (lncRNA) and microRNA (miRNA). Such RNAs regulate the expression, splicing, stability, or translation of the cognate protein-coding genes either by interacting directly with the mRNA or indirectly via binding to other ncRNAs\u003csup\u003e9-14\u003c/sup\u003e. Therapeutically, the modulating effects of ncRNAs can be harnessed by either silencing or activating genes\u003csup\u003e15, 16\u003c/sup\u003e. For instance, a phase 2 clinical trial showed that inhibition of miR-132 by an antisense oligonucleotide may prevent cardiac remodeling post-myocardial infarction\u003csup\u003e17\u003c/sup\u003e. Several lncRNA coding genes mapped to CAD-GWAS loci \u003csup\u003e3\u003c/sup\u003e, and the most significant CAD locus is mapped to the lncRNA coding gene, \u003cem\u003eANRIL\u003c/em\u003e\u003csup\u003e16\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eSince 2017, the\u0026nbsp;\u003cem\u003e16q23.3\u003c/em\u003e locus has been associated with CAD by GWAS and repeatedly replicated by studies with ever-growing sample sizes and ethnic diversity\u003csup\u003e1, 2, 18\u003c/sup\u003e \u003csup\u003e19\u003c/sup\u003e. Yet, the functional implications of this locus remains unclear.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur exploration revealed that the\u0026nbsp;\u003cem\u003e16q23.3\u003c/em\u003e locus, residing in the\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e(cadherin 13, T cadherin) regulates both the protein-coding and lncRNA-coding genes. Both our mouse and human data indicated that genetic loss of \u003cem\u003eCDH13\u003c/em\u003e is atherogenic, which aligned with the observation that lower\u003cem\u003e\u0026nbsp;CDH13\u003c/em\u003e expression was found in arterial tissue from patients and mice with atherosclerosis. In a search for potential therapeutic RNA targets, we experimentally identified a lncRNA and four miRNAs to participate in the regulation of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA. A series of CRISPR/Cas9-based knockdown and dCas9-based transcriptional activation (CRISPRa) experiments in human endothelial cells clarified the role of interacting ncRNAs in regulating \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA and consolidated the notion of an atheroprotective role of \u003cem\u003eCDH13.\u003c/em\u003e\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eFive candidate causal genes are identified at the \u003cem\u003e16q23.3\u003c/em\u003e locus\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe CAD association signals at the \u003cem\u003e16q23.3\u0026nbsp;\u003c/em\u003elocus reside within the intragenic region of the protein-coding gene, \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003e(cadherin 13, T cadherin). To systematically map candidate causal genes at this locus, we conducted a colocalization analysis using the CAD GWAS signal at the \u003cem\u003e16q23.3\u0026nbsp;\u003c/em\u003elocus and expression quantitative trait locus (eQTL) datasets from disease-relevant tissues (Method, Figure.1a and 1b). The GWAS association signal was from the latest summary genetic statistics of the CARDIoGRAMplusC4D Consortium\u003csup\u003e1\u003c/sup\u003e. The tissue types explored for eQTL analyses included arteries, adipose tissues, liver, blood, and skeletal muscle of ~ 600 individuals from the Stockholm-Tartu Atherosclerosis Reverse Networks Engineering Task (STARNET) project\u003csup\u003e20, 21\u003c/sup\u003e (Method, Figure.1a). The results colocalized the CAD GWAS signal at the \u003cem\u003e16q23.3\u003c/em\u003e locus with an eQTL signal specific to arterial tissues pointing to \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eand four lncRNAs (\u003cem\u003eCDH13-AS1\u003c/em\u003e, \u003cem\u003eCDH13-AS2\u003c/em\u003e, \u003cem\u003eCEDORA,\u003c/em\u003e and \u003cem\u003eCTD-3253I12.1\u003c/em\u003e) to be candidate causal genes for CAD (Method, Figure. 1b, Extended Data Figure 1a, supplementary Table 1). Namely, their expressions in arterial tissues likely mitigated the risks of CAD diseases.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe loss of \u003cem\u003eCDH13\u003c/em\u003e promotes the development of atherosclerosis in humans and mice.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo understand the underlying mechanism of CAD related to the \u003cem\u003e16q23.3\u003c/em\u003e locus, a critical step is to elucidate the functionality of the protein-coding gene, \u003cem\u003eCDH13\u003c/em\u003e. We first analyzed the animal model for atherosclerosis, in the \u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mouse, theCdh13 protein levels gradually declined in the aorta after 4, 8, and 12 weeks of Western diet treatment (Figure. 1c, Extended Data Figure 1b). Likewise, the bulk RNA sequencing of atherosclerosis plaques from patients undergoing carotid endarterectomy (Munich Vascular Biobank)\u003csup\u003e22, 23\u003c/sup\u003e revealed in advanced plaques (n=145) less \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eexpression than in early plaques (n=57) (Method, Figure. 1d, supplementary Table 2). Thus, we observed reduced arterial expression of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eduring the development of atherosclerosis in both humans and mice, suggesting a protective role of CDH13. To confirm this hypothesis, we generated \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e gene knockout (KO) mice on the atherogenic background by crossing \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e with \u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice to obtain the \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e/\u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice (Extended Data Figure 1c and 1d). Compared to the \u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice, the \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e/\u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice on an eight-week Western diet (WD)(Extended Data Figure 1e) had increased atherosclerosis lesion areas in the aortic root and arch (Method, Figure. 1e, Extended Data Figure 1f), reinforcing \u003cem\u003eCDH13\u003c/em\u003e to be atheroprotective. To further assess the roles of\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e, we investigated the effect of its loss-of-function (LoF) variants on CAD and other 19 vascular-related diseases or traits using phenotype and exome sequencing data from the 470,000 UK Biobank participants (Method, Supplementary Table 3 and 4). LoF variants of \u003cem\u003eCDH13\u003c/em\u003e were found in 272 participants who showed a trend of increased CAD incidence (β = 0.335, P = 0.184). LoF variants were significantly associated with atherogenic traits, such as increased arterial stiffness index (β = 2.450, P=3.92e-5), serum C-reactive protein level (β = 0.654, P=1.58e-2), blood leukocyte and lymphocyte counts (β=0.427 and 0.281, P=7.16e-4 and 5.01e-5, respectively), but decreased serum adiponectin level (β = -0.349, P=4.53e-2) (Figure 1f, Supplementary Table 3 and 4). Together, the results from mice and humans suggest that \u003cem\u003eCDH13\u003c/em\u003e is a causal gene at the \u003cem\u003e16q23.3\u003c/em\u003e locus (Figure 1), with higher expression appearing to be protective against CAD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAll five candidate causal genes are expressed in human endothelial cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBeyond \u003cem\u003eCDH13\u003c/em\u003e, our eQTL analysis also prioritized four lncRNAs as candidate causal genes at the \u003cem\u003e16q23.3\u003c/em\u003e locus (Figure. 1b). To identify relevant cell type(s) that express the respective RNAs we analyzed the publicly available single-cell (sc) RNA-seq dataset of proximal-to-mid right coronary artery (RCA) from four patients who underwent heart transplantation(Methods)\u003csup\u003e24\u003c/sup\u003e. Our analysis identified 14 major cell populations in this dataset(Method, Figure. 2a and 2b), and \u003cem\u003eCDH13\u003c/em\u003e was highly expressed in vascular muscle cells (VSMCs), fibromyocytes, and endothelial cells (ECs), especially arterial ECs (Method, Figure. 2c to 2f). Unfortunately, none of the four lncRNAs were identified in the scRNAseq datasets due to the low-expression nature of lncRNAs. Thus, we did qPCR experiments for the four lncRNAs in cardiovascular cell types, including human coronary artery ECs, VSMCs, fibroblasts (FBs), monocytes (the THP1 cell line), macrophages (THP1-differentiated), and T cells (the Jurkat cell line). As a result, EC was the only cell type expressing all four lncRNAs with a relatively high level (Figure. 2g, Extended Data Figure 2a to 2d). Therefore, our subsequent investigations focused on ECs to study the interaction between lncRNA and\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLncRNA interacts with\u003cem\u003e\u0026nbsp;CDH13\u0026nbsp;\u003c/em\u003emRNA in human endothelial cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven that lncRNAs often regulate gene expression \u003cem\u003ein cis\u003c/em\u003e, we explored whether the four lncRNAs could affect \u003cem\u003eCDH13\u003c/em\u003e expression. We first tested whether the four lncRNAs could directly bind to on the \u003cem\u003eCDH13\u003c/em\u003e mRNA by conducting dRfxCas13d-based RNA-immunoprecipitation (RIP) in human umbilical vein endothelial cells (HUVEC). The dRfxCas13d construct was fused with an HA tag, allowing the anti-HA magnetic beads to pull down and enrich the binding RNAs\u003csup\u003e25\u003c/sup\u003e\u0026nbsp; (Figure. 3a). Five RNA-targeting guide RNAs (rgRNAs) of \u003cem\u003eCDH13\u003c/em\u003e were used to pull down \u003cem\u003eCDH13\u003c/em\u003e mRNA and its potential binders. The scrambled rgRNAs were used as controls (supplementary Table 6). Using the dRfxCas13d/rgRNA-\u003cem\u003eCDH13\u003c/em\u003e system, we detected the significant binding of \u003cem\u003eCDH13-AS2\u003c/em\u003e on \u003cem\u003eCDH13\u003c/em\u003e mRNA, but not the other three lncRNAs (Figure. 3b). This might suggest the direct effects of \u003cem\u003eCDH13-AS2\u003c/em\u003e on \u003cem\u003eCDH13\u003c/em\u003e. To confirm this, we further investigated the interaction between \u003cem\u003eCDH13-AS2\u003c/em\u003e and \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA in HEK.293T cells, where neither transcript was expressed to minimize confounding factors. We first specifically activated the expression of \u003cem\u003eCDH13-AS2\u003c/em\u003e and \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eby CRISPR-mediated transcriptional activation (CRISPRa)\u003csup\u003e26\u003c/sup\u003e (Figure. 5a). The sequence of dgRNAs was designed based on the promoter or enhancer of the gene target (supplementary Table 8). Five dgRNAs were used to activate the expression of \u003cem\u003eCDH13-AS2\u003c/em\u003e, three and two dgRNAs, respectively, targeting the promoter (ENCODE ID, E1832741) and the enhancer (ENCODE ID, E1832742) of the lncRNA coding gene (Extended Data Figure 3a). We tested the effectiveness of the five dgRNAs individually to identify a highly potent dgRNA. All five dgRNAs significantly activated the expression of \u003cem\u003eCDH13-AS2\u003c/em\u003e with high efficiency, and dg1_\u003cem\u003eCDH13-AS2\u003c/em\u003e induced the strongest activation and was therefore used for the following experiments (Extended Data Figure 3c). To activate the expression of \u003cem\u003eCDH13\u003c/em\u003e, we designed eight dgRNAs for this gene, dg 1 - 4 for the promoter (ENCODE ID, E1832187) and dg 4 - 8 for the enhancer (ENCODE ID, E1832188), all near the transcription start site (TSS) or exon 1 (Extended Data Figure 3b). Four dgRNAs, dg2, 4, 7, and 8, drove expression activation of \u003cem\u003eCDH13\u003c/em\u003e, and dg4_\u003cem\u003eCDH13\u003c/em\u003e showed the highest efficiency (Extended Data Figure 3d). In the HEK. 293T cells, we successfully induced high expression of both \u003cem\u003eCDH13-AS2\u003c/em\u003e and \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eusing the corresponding CRISPRa (Figure 3c, Extended Data Figure 3c, 3d). In these cells, we transfected the plasmid system of dRfxCas13d/rgRNA-\u003cem\u003eCDH13\u003c/em\u003e-based RIP, which was able to pull down the \u003cem\u003eCDH13-AS2\u0026nbsp;\u003c/em\u003e(Figure 3c). Likewise, we designed the dRfxCas13d/rgRNA-\u003cem\u003eCDH13-AS2\u003c/em\u003e mediated RIP as previously, which successfully precipitated \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA (Figure 3c and 3d, supplementary Table 6). These results were in line with our \u003cem\u003ein silico\u003c/em\u003e experiments, in which we predicted the lncRNA-mRNA binding of \u003cem\u003eCDH13,\u003c/em\u003e respectively, with the four lncRNAs using the LncRRIsearch webtool (supplementary Table 5). Among the four, we predicted \u003cem\u003eCDH13-AS2\u003c/em\u003e as the only positive binder via reverse-complementary to the 3’UTR of \u003cem\u003eCDH13\u003c/em\u003e mRNA. The local base-pairing interaction energy was -113.58 kcal/mol, representing the strongest interaction among the 100 anticipated bindings (Extended Data Figure 3e, 3f and supplementary Table 5). \u0026nbsp;These experiments consolidated our findings on the interaction between \u003cem\u003eCDH13-AS2\u003c/em\u003e and\u003cem\u003e\u0026nbsp;CDH13\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCDH13-AS2\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;positively regulates\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u003cem\u003eCDH13\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;expression and EC functions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore whether\u0026nbsp;\u003cem\u003eCDH13-AS2\u003c/em\u003e could affect\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e expression and CAD-relevant cellular phenotypes, we knocked out\u0026nbsp;\u003cem\u003eCDH13-AS2\u003c/em\u003e (\u003cem\u003eCDH13-AS2\u003c/em\u003e-KO) using the dual-CRISPR/Cas9 targeting strategy (Figure. 4a). The third generation of lentiviral system was used to deliver the sgRNAs and spCas9 transgenes (supplementary Table 7). The dual-CRISPR strategy excised a 34bp of the shared exon (ENSE00002602225) of \u003cem\u003eCDH13-AS2\u003c/em\u003e major transcripts\u003csup\u003e27\u003c/sup\u003e (Figure. 4a, Extended Data Figure 4a). Amplification PCR and qPCR on the KO lines showed a successful knockdown of\u0026nbsp;\u003cem\u003eCDH13-AS2\u003c/em\u003e expression (Figure. 4b). Likewise, we observed reduced \u003cem\u003eCDH13\u003c/em\u003e mRNA and protein levels were in \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO HUVECs compared to control cells (Figure. 4b and 4c). Furthermore, we assessed functions related to EC fitness, including apoptosis, migration, proliferation, and immune cell adhesion, in the \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO HUVECs. To induce apoptosis, gene-edited cells were treated with 100 ng/ml LPS, and after the treatment, apoptosis-triggered fluorescence changes were detected every 10 or 12 hours until 72 hours. After 24 hours of treatment, \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO cells showed stronger apoptosis compared to control cells at each time point of measurement (Figure. 4d). To explore the migration of the gene-edited cells, we conducted the wound healing assay, and imaged cell migration every 24 hours until 72 hours. After wounding, \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO cells displayed slower migration into the wounded area at three time points (24, 48, and 72 hours), indicated by a lower percentage of cell coverage in the wounded area (Figure. 4e). The proliferation of \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO HUVECs was assayed by BrdU incorporation for 16 hours followed by flow cytometry analysis. The result showed decreased proliferation in the \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO cells (Figure. 4f). Increased EC apoptosis and reduced EC migration and proliferation could decelerate the healing of endovascular lesions and promote atherosclerosis. To probe the immune cell adhesion, we labeled THP1 monocytes with calcein, a live cell-permeant dye, and added the labeled cells onto \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO and control HUVEC cells. We observed increased monocyte adhesion on the \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO cells (Figure. 4g,\u0026nbsp;Extended Data Figure 4b), which might contribute to increased immune cells in atherosclerosis plaques.\u003c/p\u003e\n\u003cp\u003eTo test whether \u003cem\u003eCDH13-AS2\u003c/em\u003e overexpression could lead to the opposite effect, we employed CRISPRa experiments in HUVECs as previously\u003csup\u003e26\u003c/sup\u003e (Figure. 5a, Extended Data Figure 3c to 3f, Extended Data Figure 5a). Converse to the phenotypes in \u003cem\u003eCDH13-AS2\u003c/em\u003e-KO HUVECs (Figure. 4), HUVECs with \u003cem\u003eCDH13-AS2\u003c/em\u003e-CRISPRa had higher \u003cem\u003eCDH13\u003c/em\u003e expression increased migration and proliferation, and decreased apoptosis and monocyte adhesion (Figure. 5b-5g, Extended Data Figure 5b). Thus, the enhanced level of \u003cem\u003eCDH13-AS2\u003c/em\u003e increased \u003cem\u003eCDH13\u003c/em\u003e expression and led to atheroprotective endothelial phenotypes, which might mitigate the risk of CAD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCDH13-AS2\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;stabilizes \u003cem\u003eCDH13\u003c/em\u003e mRNA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLncRNAs were shown to stabilize other RNAs\u003csup\u003e28, 29\u003c/sup\u003e, which likewise appeared to be feasible for\u0026nbsp;\u003cem\u003eCDH13-AS2\u003c/em\u003e and\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e mRNA given their positive correlation. To test this hypothesis, we use a time-course CRISPRa experiment and an RNA stability assay. We conducted these experiments in HEK. 293T cell line again to minimize potential confounding effects (Extended Data Figure 3c to 3f). We conducted time-course experiments to observe RNA CRISPRa using plasmids encoding dg1_\u003cem\u003eCDH13-AS2\u003c/em\u003e/dCas9. 24 hours\u0026nbsp;after the transfection, we observed an approximately four-fold increase of \u003cem\u003eCDH13-AS2\u003c/em\u003e, and the high expression level was maintained for 120 hours (Figure. 6a). At 144 hours, we observed a slight RNA decay of \u003cem\u003eCDH13-AS2\u003c/em\u003e (Figure. 6a). The experiment showed good RNA stability of \u003cem\u003eCDH13-AS2\u003c/em\u003e. Time course experiments of the dg4_\u003cem\u003eCDH13\u003c/em\u003e/dCas9-based CRISPRa indicated that the activation of \u003cem\u003eCDH13\u003c/em\u003e peaked at 72 hours after transfection (Figure. 6b). At 96 hours, we observed the mRNA decay of \u003cem\u003eCDH13\u003c/em\u003e, which was 48 hours earlier than \u003cem\u003eCDH13-AS2\u003c/em\u003e (Figure. 6a and 6b). However, additional CRISPRa of the \u003cem\u003eCDH13-AS2\u003c/em\u003e in the same cells shifted the peak expression level of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003efrom 72 to 96 hours, postponing the mRNA decay by 24 hours (Figure. 6c). At the endpoint of the experiments (144 hours), cells with the additional activation of \u003cem\u003eCDH13-AS2\u003c/em\u003e still showed higher \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eexpression compared to without (Figure. 6d). The results suggest that \u003cem\u003eCDH13-AS2\u003c/em\u003e can stabilize the expression of \u003cem\u003eCDH13\u003c/em\u003e. We further confirm this by testing whether \u003cem\u003eCDH13-AS2\u003c/em\u003e could prevent \u003cem\u003eCDH13\u003c/em\u003e mRNA decay after transcription blocking. In \u003cem\u003eCDH13\u003c/em\u003e and \u003cem\u003eCDH13-AS2\u003c/em\u003e CRISPRa experiments, after 48 hours of plasmid transfection, actinomycin D was used to block the gene transcription. After 6 hours of actinomycin D treatment, less RNA decay was observed for \u003cem\u003eCDH13-AS2\u003c/em\u003e than \u003cem\u003eCDH13\u003c/em\u003e mRNA, ~ 60% vs ~ 80% (Figure. 6e). Comparing the \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003esingle vs \u003cem\u003eCDH13\u003c/em\u003e\u0026amp;\u003cem\u003eCDH13-AS2\u003c/em\u003e dual CRISPRa, \u003cem\u003eCDH13-AS2\u003c/em\u003e overexpression was able to reduce \u003cem\u003eCDH13\u003c/em\u003e mRNA decay by ~ 30% (Figure. 6f). However, \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eoverexpression did not affect \u003cem\u003eCDH13-AS2\u003c/em\u003e, comparing the \u003cem\u003eCDH13-AS2\u0026nbsp;\u003c/em\u003esingle vs \u003cem\u003eCDH13\u003c/em\u003e\u0026amp;\u003cem\u003eCDH13-AS2\u003c/em\u003e dual CRISPRa (Figure. 6g).Our data indicated \u003cem\u003eCDH13-AS2\u003c/em\u003e as a stabilizer for \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCDH13-AS2\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;competes with miRNAs to stabilize\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u003cem\u003eCDH13\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;mRNA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA ~1.6kb long 3’UTR was annotated for the \u003cem\u003eCDH13\u003c/em\u003e coding gene by\u0026nbsp;the\u0026nbsp;Ensembl genome database (GRCh37.p13), and \u003cem\u003eCDH13-AS2\u003c/em\u003e was predicted to bind to \u003cem\u003eCDH13\u003c/em\u003e 3’UTR (supplementary Table 5, Extended Data Figure 3e and 3f, Extended Data Figure 6). Given that miRNAs can regulate gene expression by binding UTRs to trigger RNA decay\u003csup\u003e30\u003c/sup\u003e, we, therefore, tested whether there are miRNA(s) at the 3’UTR of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA. We first predicted miRNA binding on the \u003cem\u003eCDH13\u003c/em\u003e 3’UTR using four databases, miRTarBase\u003csup\u003e31\u003c/sup\u003e, miRWalk\u003csup\u003e32\u003c/sup\u003e, scanMiRApp\u003csup\u003e33\u003c/sup\u003e, and TargetScan\u003csup\u003e34, 35\u003c/sup\u003e. The predicted miRNAs were further selected based on their expression level in ECs\u003csup\u003e36\u003c/sup\u003e (supplementary Table 9).\u0026nbsp;63 predicted miRNAs with reads per million \u0026gt;20 were selected and further prioritized based on 1) being conserved between human and mouse, 2) experimental hints (RNA-seq), or 3) suggested by ≥ two prediction databases (supplementary Table 9). Finally, 11 miRNAs (supplementary Table 10), including miR-30a-5p, miR-181a-5p, miR-155-5p, miR-19b-3p, miR-125b-2-3p, miR-34a-5p, let-7b-5p, let-7b-3p, miR-125a-3p, miR-485-3p, and miR-433-3p were finalized for further investigation. The corresponding 11 miRNA mimics and a scrambled control mimic were purchased commercially (Qiagen) (supplementary Table 11). A 1323bp of 3’UTR covering all the predicted binding sites of the 11 miRNAs was synthesized and cloned into a dual luciferase plasmid (psiCHECK™-2|RNAi Assay, Promega) to generate the psiCHECK-UTR1323 construct. The scrambled control (CTR mimic) and the 11 miRNA mimics were respectively co-transfected with the psiCHECK-UTR1323 construct to functionally analyze potential miRNA binding sites on \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003e3’UTR by an RNA interference assay. The assay is based on the Renilla/Firefly luminescence signals. Both luminescences were encoded on the psiCHECK backbone, the former was fused with the 3’UTR sequence, and the latter was on the same plasmid as the internal control (Figure. 7a). After co-transfection of the 11 miRNAs, respectively, with the psiCHECK-UTR1323 construct for 48 hours, both Renilla and Firefly luminescence were measured. In comparison to the CTR mimic, five miRNAs, miR-125b-2-3p, miR-19b-3p, miR-181a-5p, miR-433-3p, and let-7b-5p triggered Renilla luminescence reduction relative to the Firefly signal (Figure. 7b). demonstrating their effects on triggering RNA decay at \u003cem\u003eCDH13\u003c/em\u003e 3’UTR. To probe whether \u003cem\u003eCDH13-AS2\u003c/em\u003e could block the interaction of the five miRNAs with \u003cem\u003eCDH13\u003c/em\u003e 3’UTR, we co-transfected control or \u003cem\u003eCDH13-AS2\u003c/em\u003e CRISPRa plasmids, the psiCHECK-UTR1323 and correspondingly the five miRNAs in the HEK. 293T cells for 48 hours (Figure. 7c). Compared to the control, cells with \u003cem\u003eCDH13-AS2\u003c/em\u003e_CRISPRa successfully restored the reduced luciferase activity caused by four of the five miRNAs, except for miR-181a-5p (Figure. 7d through 7h). The results demonstrated that \u003cem\u003eCDH13-\u003c/em\u003eAS2 can inhibit miRNA-triggered \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA degradation and therefore stabilize its expression. The lncRNA-miRNA-mRNA interplay might thus be harnessed to develop RNA-based therapies for CAD.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eGWAS have reproducibly associated the \u003cem\u003e16q23.3\u003c/em\u003e locus with the risk of CAD, but the underlying mechanisms remained unclear\u003csup\u003e1, 2, 18, 19\u003c/sup\u003e. Here, we identified \u003cem\u003eCDH13\u003c/em\u003e and four lncRNAs as causal genes at the locus. In the UK Biobank, carriers of LoF alleles displayed increased risk for atherogenic traits. Colocalization analysis indicated \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003econtributing to CAD via its roles in the arterial wall(Figure 1a and 1b, Extended Data Figure 1a, supplement Table 1, 3 and 4). Moreover, advanced atherosclerotic plaques in the carotid artery displayed lower \u003cem\u003eCDH13\u003c/em\u003e expression compared to early plaques(Figure 1d). Cdh13 protein levels gradually declined during the development of atherosclerosis in \u003cem\u003eApoe\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e mice on a Western diet(Figure 1c). \u003cem\u003eCdh13\u003csup\u003e-/-\u003c/sup\u003e/Apoe\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e mice on a Western diet developed larger atherosclerotic lesions compared to \u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice(Figure 1e). In line with our findings, Fujishima \u003cem\u003eet al.\u003c/em\u003e reported enhanced neointima proliferation and atherosclerosisin \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e/\u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice\u003csup\u003e37\u003c/sup\u003e. Thus, results from both humans and mice indicate that lower \u003cem\u003eCDH13\u003c/em\u003e expression aggravates atherosclerosis or – \u003cem\u003evice versa\u003c/em\u003e – \u003cem\u003eCDH13\u003c/em\u003e is atheroprotective.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur GWAS-eQTL colocalization analysis also indicated four lncRNAs as candidate causal genes at the \u003cem\u003e16q23.3\u003c/em\u003e locus(Figure 1b, supplement Table 1). We experimentally validated the \u003cem\u003ein-silico\u003c/em\u003e prediction(supplement Table 5) that one of the lncRNAs, \u003cem\u003eCDH13-AS2\u003c/em\u003e, binds to the \u003cem\u003eCDH13\u003c/em\u003e 3’UTR(Figure 3b to 3d). CRISPR-based knockdown and activation of \u003cem\u003eCDH13-AS2\u003c/em\u003e indicated a protective role of this antisense RNA(Figure 4 and 5), as it prevented \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003emRNA decay and enhanced its beneficial effects in EC(Figure 6). We further demonstrated that four miRNAs were involved in the decay of \u003cem\u003eCDH13\u003c/em\u003e mRNA. RNA interference and CRISPRa experiments revealed that the miRNA-triggered decay of \u003cem\u003eCDH13\u003c/em\u003e mRNA was by ameliorated by \u003cem\u003eCDH13-AS2\u003c/em\u003e(Figure 7).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAlbeit more than 90% of GWAS signals map to non-coding regions only a few loci were found to act on CAD through ncRNA functions, such as MIAT at the \u003cem\u003e22q12.1\u003c/em\u003e locus and ANRIL at the\u003cem\u003e9p21\u0026nbsp;\u003c/em\u003elocus\u003csup\u003e16, 38\u003c/sup\u003e, which is still the strongest CAD risk locus known so far\u003csup\u003e39\u003c/sup\u003e. \u0026nbsp;At the \u003cem\u003e16q23.3\u003c/em\u003e locus, we found that the lncRNA \u003cem\u003eCDH13-AS2\u003c/em\u003e is one of the causal genes as it up-regulates the protein-coding gene \u003cem\u003eCDH13\u003c/em\u003e in a \u003cem\u003ecis\u003c/em\u003e manner by competing with the binding of miRNAs.\u003c/p\u003e\n\u003cp\u003eLncRNAs are commonly reported to bind to miRNAs, acting as sponges, to prevent mRNA degradation\u003csup\u003e40, 41\u003c/sup\u003e. A classic regulatory mechanism of miRNAs involves binding to the 3′ UTR of a mRNA, leading to its degradation or translational repression\u003csup\u003e12, 42-44\u003c/sup\u003e. Here we showed that \u003cem\u003eCDH13-AS2\u003c/em\u003e competed with the binding of four miRNAs (miR-125b-2-3p, miR-19b-3p, miR-433-3p, and let-7b-5p)(Figure 7d to 7h). Similar interactions involving miRNAs have been reported for cancer and Alzheimer's disease\u003csup\u003e40, 45\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile current drugs for treatment of CAD uniformly address genes that increase CAD risk via lipid levels\u003csup\u003e1-3\u003c/sup\u003e, i.e. \u003cem\u003ePCSK9\u003c/em\u003e\u003csup\u003e46\u003c/sup\u003e, \u003cem\u003eLPA\u0026nbsp;\u003c/em\u003e\u003csup\u003e47\u003c/sup\u003e, \u003cem\u003eANGPTL3\u0026nbsp;\u003c/em\u003e\u003csup\u003e48\u003c/sup\u003e, and \u003cem\u003eAPOC3\u0026nbsp;\u003c/em\u003e\u003csup\u003e49\u003c/sup\u003e, endogenous atheroprotective mechanisms are less well explored therapeutically. The potential of an atheroprotective milieu is illustrated by the internal mammary artery, which – even when exposed to high lipid levels or strong genetic disposition – resists plaque formation\u003csup\u003e50, 51\u003c/sup\u003e. In such a sense, our findings could be harnessed to design therapeutic strategies for CAD, going beyond lipid lowering, by increasing\u003cem\u003e\u0026nbsp;CDH13\u003c/em\u003e expression, thereby enhancing arterial resilience to the disease. From a therapeutic perspective, it may be challenging to interfere with the regulation of a widely expressed gene\u003cem\u003e.\u003c/em\u003e In this respect, tissue or cell-specific regulation may be beneficial. To some extent, it is \u0026nbsp;the case of \u003cem\u003eCDH13\u003c/em\u003e given its expression in arteries limited to a few mesoderm-derived cell types (Figure. 2a-2d) and the CAD risk at the\u0026nbsp;\u003cem\u003e16q23.3\u003c/em\u003e locus is mediated by \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eand lncRNAs exclusively in artery tissues. Indeed, endothelial cells demonstrated improved function upon activation via\u0026nbsp;\u003cem\u003eCDH13-AS2.\u003c/em\u003e The lncRNA-miRNA-mRNA interplay might inspire intravascular delivery of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003e–miRNA target site blockers, antagomirs or anti-miRs to increase the expression of the gene, which protects coronary arteries from developing CAD. Target site blockers, anti-miR and alike, are already being explored as a therapeutic strategy for cardiovascular and\u0026nbsp;cerebrovascular\u0026nbsp;diseases as they can penetrate the arterial wall\u003csup\u003e15, 52-54\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe are aware of the limitations of our study. First, although we were able to prioritize \u003cem\u003eCDH13\u003c/em\u003e and the four lncRNAs as candidate causal genes for CAD by GWAS-eQTL colocalization analysis, no data were available to directly investigate the genetic link of the four miRNAs with CAD. Indeed, currently available population transcriptomes do not include sequences of such short RNAs. Second, our investigation provides mechanistic insights rather than specific therapeutics to increase \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eexpression, which may be challenging given that RNA delivery specific to ECs was shown to have low efficiency. However, RNA-based therapy, such as target site blockers or anti-mirs, showed good penetration into artery walls, which are already being explored for cardiovascular diseases\u003csup\u003e15, 52-54\u003c/sup\u003e. Additionally, further technological advancements are anticipated to enhance cell-specific RNA delivery. Third, \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003ewas also expressed in VSMCs (Figure. 2a through 2f), which is worthy of an in-depth investigation but was not explored here. We rather focused on ECs that displayed high expression levels of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eas well as lncRNAs and miRNAs involved in its regulation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTaken together, a lncRNA-miRNA-mRNA interplay affects the expression of \u003cem\u003eCDH13\u003c/em\u003e, which resides as the only coding gene at a GWAS locus for CAD. Multiple lines of evidence indicate that the GWAS signal is mediated by the protective effects of \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003eon the arterial wall. The RNA interplay may be explored therapeutically by mimicking the effect of \u003cem\u003eCDH13-AS2\u003c/em\u003e tointerfere with the miRNA-mediated degradation of \u003cem\u003eCDH13\u003c/em\u003e mRNA at its long 3’UTR.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eColocalization analysis of the GWAS and eQTL signals\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor colocalization analysis at the \u003cem\u003e16q23.3\u003c/em\u003e locus, we used full-genome GWASs dataset of CAD, respectively, from CARDIoGRAMplusC4D and expression quantitative trait loci (eQTL) data of five tissue types of ~ 600 individuals \u0026nbsp;from the STARNET (Stockholm-Tartu Atherosclerosis Reverse Networks Engineering Task) study\u003csup\u003e1, 55, 56\u003c/sup\u003e (Figure.1c). The five tissue types included blood(n = 560), artery (atherosclerotic aortic root, n = 539 and free internal mammary artery, n = 553)), adipose tissue (subcutaneous, n = 534 and visceral abdominal, n = 534), skeletal muscle (n = 534), and liver (n = 546). Initially, we overlapped GWAS and eQTL summary statistics, utilizing a GWAS p-value threshold of 5e - 8 and an eQTL p-value cutoff of 0.01 to select SNPs for further testing. To estimate colocalization, we ran the coloc.abf function implemented in the coloc R package\u003csup\u003e57\u003c/sup\u003e that uses an approximate Bayes factor to estimate the posterior probabilities between a given GWAS and eQTL signal (supplementary Table 1). Significant colocalizations were defined with PP4 \u0026ge; 0.70, indicating a common causal variant between the GWAS and eQTL data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferential expression analysis of\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u003cem\u003eCDH13\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;using bulk RNA-seq data of patients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRNA sequencing data of human carotid artery plaques 57 early and 145 advanced plaques of the human carotid arteries were harvested during carotid artery endarterectomy (CEA) surgery, transported to the laboratory, and snap-frozen. CEA was performed due to advanced atherosclerotic lesion formation and stenosis in the carotid arteries. The patients\u0026rsquo; characteristics are summarized in supplementary Table 2. The tissue handling,RNA extraction, and sequencing were as previously\u003csup\u003e58\u003c/sup\u003e. Bulk RNA sequencing and quality control (QC) were performed as described\u003csup\u003e23\u003c/sup\u003e. Raw read counts were normalized with the trimmed mean of M-values (TMM) and transformed with voom, resulting in log2-counts per million with associated precision weights, which were then used for differential expression analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRare variant association analysis in the UK Biobank\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe rare variant association analysis for\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e was conducted using data from the UK Biobank (https://www.ukbiobank.ac.uk/) under the project 25214. We used whole-exome sequencing (WES) data of 470,000 participants from the UK Biobank (The final release of population-level exome OQFE variants)\u003csup\u003e59\u003c/sup\u003e. The UK Biobank annotation helper file was used to obtain the list of rare variants within the\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e gene region. Variant annotation was derived from the snpEff tool using the Ensembl v85 gene definitions to determine their functional impact on transcripts and genes. Loss-of-function variants (LoF) included stop codon-introducing or splice site-disrupting SNPs, insertion/deletion variants predicted to disrupt a transcript\u0026apos;s reading frame, or larger deletions removing either the first exon or more than 50% of the protein-coding sequence of the transcript. We collected phenotypic data for 6 binary and 14 quantitative traits, including blood biomarkers, vascular health indicators, and immune cell parameters. The phenotypes were defined using the respective Field IDs, ICD-10 and ICD-9 codes obtained from the primary care, OPCS-4 diagnoses, and self-reported codes. When multiple measurements of the same phenotype were available for an individual,\u0026nbsp;we utilized the measurements ofthe first visit. Age, sex, and 10 genotype principal components were gathered as covariates. The data was accessed through the UK Biobank Research Analysis Platform (RAP).\u003c/p\u003e\n\u003cp\u003eThe association analysis was conducted on 500K WES data following the UKB tutorial \u0026lsquo;Burden Testing with regenie Using WES Annotation Files\u0026rsquo; (https://dnanexus.gitbook.io/uk-biobank-rap/science-corner/using-regenie-to-generate-variant-masks#annotation-file). We used the genome-wide regression approach implemented in REGENIE\u003csup\u003e60\u003c/sup\u003e. Computationally efficient whole-genome regression for quantitative and binary traits. Nat Genet 53, 1097-1103. 10.1038/s41588-021-00870-7), available through the Swiss-Army-Knife tool library in RAP. Variants with an allele frequency below 1% were classified as rare. For association analysis, we used M1 variant masks, which LoF variants. To aggregate the effects of rare variants across the\u0026nbsp;\u003cem\u003eCDH13\u003c/em\u003e region, a burden test was applied. \u0026nbsp;This approach consolidates the variants from our defined mask into a unified burden mask and evaluates it as a single genotype to produce association statistics. To assess associations with binary phenotypes, a logistic regression model was used, and for quantitative traits, linear regression. The output of the burden test included P-value, estimates of beta (effect size), and the standard error.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle-cell RNA sequencing (scRNA-seq) analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo characterize coronary artery expression of candidate genes at the \u003cem\u003e16q23.3\u003c/em\u003e locus, we utilized the publicly available single-cell RNA sequencing (scRNA-seq) dataset GSE131778(Figure. 2a to 2f). This dataset includes single-cell transcriptomic data obtained from eight coronary artery samples derived from four patients who underwent heart transplantation. The samples were collected from the proximal-to-mid right coronary artery (RCA) after removing debris and selecting viable cells through fluorescence-activated cell sorting (FACS). The cells were processed using the 10X Genomics Chromium platform with 3\u0026apos; chemistry reagents (version 2). The scRNA-seq analysis was performed following established protocols\u003csup\u003e24, 61-63\u003c/sup\u003e. Briefly, the scRNA-seq expression matrix was analyzed using the R package \u0026quot;Seurat.\u0026quot; Initially, gene expression levels were normalized using the \u0026quot;NormalizeData\u0026quot; function. Next, the \u0026quot;FindVariableFeatures\u0026quot; function was employed to identify the top 2,000 highly variable genes (HVGs) for further analysis. To reduce dimensionality, principal component analysis (PCA) was performed using the \u0026quot;RunPCA\u0026quot; function. Batch effects were corrected using the \u0026quot;Harmony\u0026quot; package, ensuring that the downstream analyses were not biased by inter-sample variability. To identify cell populations, the \u0026quot;FindNeighbors\u0026quot; function was applied to compute the k-nearest neighbors for each cell, followed by the \u0026quot;FindClusters\u0026quot; function to determine optimal clustering. The clustering resolution parameter was set to 0.5 to balance the granularity of cluster identification. To visualize the identified clusters, uniform manifold approximation and projection (UMAP) were utilized. Cluster annotation was conducted by referencing marker genes documented in the original study associated with this dataset\u003csup\u003e23\u003c/sup\u003e. Additionally, the \u0026quot;FeaturePlot\u0026quot; and \u0026quot;DotPlot\u0026quot; functions were utilized to specifically visualize the gene expression across various cell types within the dataset.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMouse studies\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll mouse experiments were performed according to the regulations of German legislation on animal protection and were approved by the local animal care committee (District Government of Upper Bavaria, GZ: ROB-55.2-2532.Vet_02-18-177). The mice were bred and aged in the German Heart Centre Munich vivarium under standard conditions, following a 12-hour light/dark cycle with free access to food/water, maintaining temperature and humidity. Apolipoprotein E and Cadherin 13 double knockout (\u003cem\u003eCdh13\u003csup\u003e-/-\u003c/sup\u003e/Apoe\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e) mice were generated through crossbreeding \u003cem\u003eCdh13\u003csup\u003e+/-\u003c/sup\u003e\u003c/em\u003e (purchased from The Jackson Laboratories, Bar Harbor, USA; \u003cem\u003eCdh13tm1Brns/J\u003c/em\u003e, subsequently termed \u003cem\u003eCdh13\u003csup\u003e-/\u003c/sup\u003e\u003c/em\u003e\u003csup\u003e-\u003c/sup\u003e) with \u003cem\u003eApoe\u003csup\u003etm1Unc\u003c/sup\u003e\u003c/em\u003e (purchased from The Jackson Laboratories, Bar Harbor, USA; subsequently termed \u003cem\u003eApoe\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e) mice over five generations. Experimental ages ranged from 12 weeks. Both males and females were included in the experiments, ensuring a balanced distribution of sex across treatment and genotype conditions. The mice were fed a Western Diet (MD88137 Adjusted Calories diet, 42% from fat, Harlan) for 4, 8, or 12 weeks to evaluate the progression of atherosclerosis. Both genotype groups were closely matched in terms of age. Experimental animals that died during rearing, suffered from other diseases, or failed to meet the scoring criteria (supplementary Table 12) were excluded. All remaining mice that met the scoring criteria were included for further experiments and statistical analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHistology\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate aortic pathology, mice of both sexes in a balanced ratio were sacrificed at predetermined time points, as previously described (Extended Data Figure 1b and 1e). Euthanasia was performed using gaseous isoflurane, followed by transcardiac perfusion with 0.9% saline until complete replacement of circulating blood was achieved. The aorta and heart were exposed by carefully removing the surrounding tissues and subsequently fixed in 4% paraformaldehyde in PBS at 4 \u0026deg;C for 24 hours. The aortic roots were embedded in molds using an optimal cutting temperature (OCT) compound (Sakura Finetek, Tokyo, Japan) and snap-frozen on dry ice. Frozen tissue blocks were sectioned into 5 \u0026mu;m slices and mounted onto microscope slides for further analysis\u003csup\u003e64\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnzyme-Linked Immunosorbent Assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMice arteries\u0026rsquo; Cdh13 expression was assessed using a commercially available ELISA kit (abx518770, Abbexa) according to the manufacturer\u0026rsquo;s protocol. Aortic tissue, spanning from the aortic arch to the iliac artery, was excised, homogenized in 100 \u0026micro;L PBS using a tissue grinder, and centrifuged at maximum speed for 15 minutes at 4\u0026deg;C. The supernatant was collected for protein concentration analysis using the Pierce\u0026trade; BCA Protein Assay Kit (Thermo Fisher Scientific) and subsequent ELISA assay. All procedures were carried out according to the manufacturer\u0026apos;s protocol. A total of 100 \u0026micro;L of each standard, test sample, and control (zero) were added to wells on a pre-coated plate, followed by a 1-hour incubation at 37\u0026deg;C. Subsequently, the plate was washed three times with 1X Wash Buffer, and 100 \u0026micro;L of Detection Reagent B working solution was added to each well. After a 30-minute incubation at 37\u0026deg;C, the solution was discarded, and the plate was washed five times. Next, 90 \u0026micro;L of TMB Substrate was added to each well, and the plate was resealed and incubated for 10\u0026ndash;20 minutes at 37\u0026deg;C, protected from light. Lastly, 50 \u0026micro;L of Stop Solution was added to each well, and the absorbance was promptly measured at 450 nm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOil Red O staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe aortic root were imaged post-Oil Red O (ORO) staining to assess plaque burden. Tissues were briefly washed with PBS, then incubated at 37\u0026deg;C for 30 minutes in 3 mg/mL ORO solution (Sigma-Aldrich, O0625) in 60% isopropanol. Excess dye was removed with 60% isopropanol. Aortic arches were opened, pinned on a black pad, and imaged using a Stemi 2000-C microscope with an Axiocam ERc 5s camera and ZEN 2.3 Blue software. Plaque area at the aortic root was quantified by measuring at the maximum cross-sectional area. Aortic root sections were photographed after hematoxylin restaining of nuclei. ORO-positive areas were quantified using ImageJ, and lesion area percentages were calculated to evaluate atherosclerosis severity\u003csup\u003e64\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell cultures and passaging\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHEK 293T cells (ATCC, USA) were cultured in high-glucose Dulbecco\u0026apos;s Modified Eagle Medium (DMEM; Gibco BRL, Grand Island, USA) supplemented with 10% fetal bovine serum (FBS; Gibco BRL) (supplementary Table 14) . The primary human umbilical vein endothelial cells (HUVEC; PromoCell, pooled donors) were cultured in Endothelial Cell Growth Medium MV 2 (C-22022) supplemented with a complete supplement mix. All cells were maintained at 37\u0026deg;C in a humidified atmosphere with 5% CO\u003csub\u003e2\u003c/sub\u003e. The culture medium was refreshed every two days, and cells (HEK.293T, HUVEC) were passaged upon reaching 90% confluency.\u003c/p\u003e\n\u003cp\u003ePrimary artery cells were handled using the DetachKit (PromoCell) following the manufacturer\u0026apos;s protocols (supplementary Table 14). The seeding density for HUVECs was maintained between 5,000 and 10,000 cells per cm\u003csup\u003e2\u003c/sup\u003e. THP-1 monocytes (ATCC, USA) were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum (FBS; Gibco BRL) and 1% penicillin-streptomycin. Cells were kept at a density of 2 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e to 1 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e cells/mL to ensure optimal growth and viability. Cells were split every 2 - 3 days by diluting them with fresh culture medium. To induce differentiation of THP1 cells into macrophage-like cells, cells were treated with Phorbol 12-myristate 13-acetate (PMA) under a concentration of 100 ng/mL, 72 hours to trigger differentiation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCRISPR and DNA Cloning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePlasmids used in this study, including lentiCRISPRv2 (CRISPR knockout), lentiSAMv2 (CRISPR activation), pXR003:CasRx gRNA cloning backbone (dRfxCas13d-based RIP), and psiCHECK\u0026trade;-2 \u003cem\u003eCDH13\u003c/em\u003e 3\u0026apos; UTR (RNA interference assay), were commercially purchased, expanded in LB medium with antibiotics, and extracted using the PureYield\u0026trade; Plasmid Mini/Max-prep System (Promega, A1223 /A2392). Linearization of lentiCRISPRv2 and lentiSAMv2 was performed using Esp3I (Thermo Scientific, FD0454), pXR003:CasRx by BbsI (FD1014), and psiCHECK\u0026trade;-2 by XhoI (FD 0694) and NotI (FD0593). Target fragments were purified via agarose gel electrophoresis and the QIAquick Gel Extraction Kit (28704). rgRNA, sgRNA, and dgRNA oligos were phosphorylated with T4 PNK (NEB M0201S) at 37\u0026deg;C for 30 minutes and respectively ligated with pXR003:CasRx, lentiCRISPRv2, and lentiSAMv2 using Quick Ligase (NEB, M2200S). The synthetic \u003cem\u003eCDH13\u003c/em\u003e 3\u0026apos; UTR (1323bp) was cloned into psiCHECK\u0026trade;-2 vector following the same procedures. Recombinant plasmids were transformed into Stbl3 bacteria, expanded, and screened on selective agar plates for positive clones following standard protocols.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell transfection, lentivirus packaging, and cell infection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePlasmid encoding genes of interest or guide RNAs (supplementary Table 6 ,- 8 and 15) were transfected into cell lines using FuGENE\u0026reg; HD (Promega, E5911), following the manufacturer\u0026rsquo;s protocol. Briefly, cells were seeded to reach 70-80% confluency on the day of transfection. Plasmid DNA, Opti-MEM\u0026trade; I Reduced Serum Medium (Gibco), and FuGENE\u0026reg; HD were mixed in a ratio corresponding to the seeding surface area. After a 30-minute incubation at room temperature, the transfection complex was gradually added to the cells in the well plate. Transfected cells were incubated in a humidified incubator at 37\u0026deg;C with 5% CO\u003csub\u003e2\u003c/sub\u003e. The time to harvest cells depends on different experiments.\u003c/p\u003e\n\u003cp\u003eTo generate lentiviral particles, psPAX2 (Addgene-12260), pCMV-VSV-G (Addgene-8454), and guide RNA-expressing plasmids (supplementary Table 6 and 7) were co-transfected into HEK. 293T cells using the above transfection protocol. After 72 hours, the virus-containing supernatant was harvested, transferred to polypropylene storage tubes, and centrifuged at 2000 \u0026times; g for 5 minutes to remove residual packaging cells. The clarified supernatant was filtered through a 0.45 \u0026micro;m PES filter, aliquoted, snap-frozen in liquid nitrogen, and stored at -80\u0026deg;C to preserve viral titer. For infection, HUCEV cells were incubated with the collected viral supernatant, diluted 1:1 with fresh culture medium, and supplemented with 10 \u0026micro;g/mL polybrene (1:1000 dilution) to enhance transduction efficiency. Cells were maintained in viral-containing medium for 48 hours before further processing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCRISPR/Cas9-based gene knockdown\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo knock out \u003cem\u003eCDH13-AS2\u003c/em\u003e using the CRISPR/Cas9 system, we utilized two sgRNAs and spCas9 to excise a 34bp of the shared exon (ENSE00002602225) (Figure. 4a). One lentiviral vector encoded by a sgRNA, spCas9, and a puromycin resistance cassette was used to pack the virus for gene targeting in cells. HUVEC cells were routinely passaged 48 hours after viral infection and subjected to positive selection after 7 days. Puromycin (1 ug/mL) selection was employed to eliminate negative cells after virus infection\u003csup\u003e27\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCRISPR/Cas9-based transcriptional activation (CRISPRa)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo increase the expression of RNAs, we employed CRISPR-mediated transcriptional activation (CRISPRa)\u003csup\u003e26\u003c/sup\u003e. The CRISPRa system includes two components, 1) the spCas9 lacking nuclease activity and fused with a transcription activator VP64 (dCas9-VP64) and 2) a short version of sgRNA (14-15 bp, dgRNA) tagged with hairpin aptamer (MS2) specifically recruiting transcriptional activation complex, MCP:P65:HSF1 (MPH) (Figure. 5a). The sequence of dgRNAs was designed based on the promoter or enhancer of the target gene (Extended Data Figure 3c to 3f). HUVEC cells were routinely passaged 48 hours after viral infection and subjected to positive selection after 7 days. Blasticidin (1 ug/mL) and Hygromycin (0.5 ug/mL) were employed to eliminate negative cells after virus infection.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003edRfxCas13d-based RNA immunoprecipitation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe performed RIP using the CRISPR system based on the RNA-targeting nuclease, Cas13. Several Cas13 orthologs have been found to recognize and target RNAs in mammalian cells, including LwaCas13a (139 kDa), PspCas13b (128 kDa), and RfxCas13d (112 kDa). We selected the enzymatically dead version of the smallest Cas13, dRfxCas13d (dCas13d)\u003csup\u003e25\u003c/sup\u003e, to explore the potential binding interaction among the four antisense lncRNAs and \u003cem\u003eCDH13\u003c/em\u003e. The dCas13d construct was fused with a HA tag and a double-stranded RNA binding domain (dsRBD) stabilizing the dCas13d / rgRNA / target complex. The anti-HA magnetic beads were employed to pull down and enrich the binding RNAs and proteins. Five rgRNAs of \u003cem\u003eCDH13\u003c/em\u003e were used to pull down \u003cem\u003eCDH13\u003c/em\u003e mRNA and its potential binders (supplementary Table 6). The scrambled rgRNAs were used as the control(supplementary Table 6)\u003csup\u003e25\u003c/sup\u003e .\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA immunoprecipitation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHUVEC cells (1.5 \u0026times; 10\u003csup\u003e7\u003c/sup\u003e) were washed with PBS and lysed in a buffer containing 50 mM Tris-HCl, 100 mM NaCl, 1% NP-40, 0.1% SDS, and 0.5% sodium deoxycholate at pH 7.4, supplemented with a Protease Inhibitor Cocktail Set III (EMD Millipore). Five percent of each lysate was reserved for input RNA preparation, while the remaining lysates were divided into two aliquots for immunoprecipitation. Immunoprecipitation was performed using 10 \u0026micro;g of either a specific antibody (CST, HA-Tag-C29F4, Rabbit mAb #3724) or control IgG (CST, Normal Rabbit IgG #2729), followed by incubation at 4\u0026deg;C for 8 hours. The immunoprecipitants were washed six times with a buffer containing 5 mM Tris-HCl, 150 mM NaCl, and 0.1% Triton X-100. Bound RNA fragments were isolated and purified using QIAzol Lysis Reagent (Qiagen, Cat. No. 79306) following the manufacturer\u0026apos;s protocols. The RNA underwent reverse transcription using SuperScript III with a random primer mix (Thermo Fisher, 12574026). The cDNA obtained was amplified using PCR for the target transcripts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNucleic acid isolation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor DNA extraction from cultured cells, the DNeasy Blood \u0026amp; Tissue Kit (Qiagen,Cat. No. 69504) was used according to the manufacturer\u0026rsquo;s instructions. For the plasmid DNA from \u003cem\u003eE.coli\u003c/em\u003e, the PureYield\u0026trade; Plasmid Mini/Max-prep System, A1223 or A2392 was used following the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e\n\u003cp\u003eTotal RNA was extracted from cultured cells using QIAzol Lysis Reagent (Qiagen, Cat. No. 79306) following a modified Chomczynski and Sacchi protocol (1993). Cells were lysed with QIAzol and harvested using a cell scraper, then transferred to 1.5 mL microcentrifuge tubes. Following an incubation at room temperature, the lysates were vortexed, followed by the addition of 120 \u0026micro;L DNase-/RNase-free water and 100 \u0026micro;L chloroform. Samples were mixed by inversion, incubated at room temperature for another 10 minutes, and centrifuged at 13,000 rpm for 10 minutes at 4\u0026deg;C to separate phases. The aqueous phase was transferred to fresh tubes, combined with ice-cold isopropanol and 1.5 \u0026micro;L GlycoBlue\u0026trade; coprecipitant (Thermo Fisher Scientific), vortexed, and incubated at -20\u0026deg;C for at least 30 minutes. RNA was pelleted by centrifugation at 14,000 rpm for 30 minutes at 4\u0026deg;C, washed twice with 75% ethanol, and centrifuged 10minutes at same conditions. Pellets were air-dried and resuspended in DNase-/RNase-free water.\u003c/p\u003e\n\u003cp\u003eDNA/RNA quality and concentration were determined using a NanoQuant Plate\u0026trade; (Tecan) and an Infinite M200Pro spectrophotometer (Tecan), based on the 260/280 nm absorbance ratio. Resuspended RNA was either used immediately or stored at - 80\u0026deg;C.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePolymerase chain reaction (PCR)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrimers (17 - 30 bp) for PCR were designed using PrimerBank or Primer3 and synthesized by Eurofins Genomics (Ebersberg, Germany) (supplementary Table 16). Stock primers (100 \u0026micro;M) were diluted to 10 \u0026micro;M working solutions with DNase-/RNase-free water and stored at 4\u0026deg;C. cDNA was synthesized from mRNA using Maxima\u0026trade; H Minus cDNA Synthesis MasterMix (Thermo Fisher Scientific) and stored at -20\u0026deg;C. Amplification PCR used Q5 2 x Master Mix (New England Biolabs) with primers (supplementary Table 16) and 750 ng of cDNA or 150 ng of genomic DNA. PCR products were resolved on 1 - 3% agarose gels depending on the product size, visualized under UV, and documented with Amersham ImageQuant 800 (Cytiva). qPCR was performed with PerfeCTa SYBR\u0026reg; Green FastMix (Quantabio) on the ViiA 7 Real-Time PCR System (Applied Biosystems) using ~100 ng cDNA and primers (supplementary Table 16). Reactions were run in triplicate, normalized to housekeeping genes (\u0026Delta;Ct), and analyzed by the \u0026Delta;\u0026Delta;Ct method to calculate fold changes (2^-\u0026Delta;\u0026Delta;Ct) relative to controls.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eProtein isolation and Western blot analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll procedures were performed on ice unless stated otherwise. Cells were lysed in 1x RIPA buffer (NEB, 9806S) with HALT\u0026trade; protease inhibitor (1:100, Thermo Fisher Scientific), incubated at 4\u0026deg;C for 30 minutes, and centrifuged at ~13,000 rpm for 30 minutes. Supernatants were collected, and protein concentrations were measured using the Pierce\u0026trade; BCA Protein Assay Kit (Thermo Fisher Scientific) with absorbance at 562 nm (Infinite M200Pro spectrophotometer - Tecan). For Western blotting, 30 \u0026micro;g of protein was mixed with Laemmli buffer, boiled at 95\u0026deg;C, and resolved on 4 - 20% gradient gels (Bio-Rad) at 120 V. Proteins were transferred to PVDF membranes, blocked with 5% BSA, and incubated overnight with primary antibodies (supplementary Table 13). After washing, membranes were incubated with HRP-conjugated secondary antibodies (supplementary Table 13) and visualized using SuperSignal\u0026trade; West Dura substrate on the Amersham ImageQuant 800 system (Cytiva).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWound healing migration assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo perform the HUVEC wound healing migration assay, sterile forceps were used to position inserts (ibidi, No. 80369) into the plate wells, ensuring consistent alignment of the \u0026ldquo;wound field\u0026rdquo;. The cell suspension was prepared at a concentration of 0.5 x 10\u003csup\u003e6\u003c/sup\u003e cells/mL in medium, and 300 \u0026micro;L of this suspension was carefully added to each well without disturbing the inserts. The plate was incubated overnight in a cell culture incubator to facilitate the formation of a confluent monolayer. To start the assay, inserts were gently removed using sterile forceps. The media was aspirated, and the wells were washed with fresh media to eliminate dead cells and debris. Following washing, medium containing 2.5 \u0026micro;g/mL of mitomycin C was added to inhibit cell proliferation, ensuring wound closure was driven by cell migration. The wells were observed under a light microscope, and additional washing was conducted if necessary. Cells were subsequently incubated, and wound closure was monitored at 24-, 48-, and 72-hours post-insert removal using light microscopy. ImageJ software was used to measure the percentage of closure or the rate of cell migration into the wound field.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBrdU proliferation assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the cell proliferation assay, immunofluorescent staining of incorporated bromodeoxyuridine (BrdU) was performed using the BD Pharmingen\u0026trade; APC BrdU Flow Kit (Catalog No. 557892), followed by the manufacturer\u0026rsquo;s protocol. To label cells with BrdU, 10 \u0026mu;L of a 1 mM BrdU solution was added per mL of culture medium, ensuring a cell density of no more than 1 x 10\u003csup\u003e6\u003c/sup\u003e cells/mL to maintain normal cell cycling. After a 15-hour incubation, BrdU-plused cells were detached following washing and centrifugation at 200 g for 5 minutes. The cells were fixed and permeabilized on ice for 30 minutes using BD Cytofix/Cytoperm Buffer, followed by further permeabilization with BD Cytoperm Permeabilization Buffer Plus and refixed for 10 minutes. To detect incorporated BrdU, the cells were treated with DNase (300 \u0026micro;g/mL) and then incubated at 37 \u0026deg;C for 1 hour. After washing, the cells were resuspended in 50 \u0026micro;L of BD Perm/Wash Buffer containing diluted fluorescent anti-BrdU and incubated at 4\u0026deg;C for 20 minutes. The analysis of stained cells was conducted using a BD LSRFortessa flow cytometer (BD Biosciences) at a low flow rate to achieve optimal resolution. Gating was conducted for viable cells (forward scatter area [FSC-A] versus side scatter area [SSC-A]) and single cells (FSC-A versus forward scatter width [FSC-W], SSC-A versus side scatter width). Data analysis and statistical plotting were performed using FlowJo software v.9.9.6 (FlowJo LLC).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdhesion assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHUVEC endothelial cells were seeded at 5,000 cells/cm\u0026sup2; on gelatin-coated 6-well or 12-well plates and cultured for 48 hours to form a confluent monolayer. THP1 monocytes were labeled with calcein dye (Invitrogen\u0026trade;, C1430) at a 1:1000 dilution, prepared at a concentration of 1.0 x 10\u003csup\u003e6\u0026nbsp;\u003c/sup\u003ecells/mL in serum-free RPMI-1640 medium, and incubated at 37\u0026deg;C for 60 minutes. After centrifugation, the supernatant was extracted, and the THP1 cells underwent two washes with serum-free RPMI-1640 before being resuspended at the same concentration for application onto HUVECs. The HUVEC endothelial cell medium was substituted with serum-free RPMI-1640, and 1 mL of the stained monocyte suspension was introduced into each well with the endothelial monolayer. Co-cultures were incubated for 180 minutes under standard conditions. After incubation, non-adherent leukocytes were removed by washing the wells three times with 2 mL of PBS. After the final wash, 500 \u0026mu;L of 1 X RIPA Lysis Buffer was added to each well for a 5-minute incubation at room temperature. After 14,000 X g centrifugation, 100 \u0026mu;L of lysate was then transferred to a 96-well fluorescence-compatible plate. Fluorescence was measured using the plate reader (Infinite M200Pro spectrophotometer - Tecan) at excitation/emission wavelengths of 480nm/520nm. A standard curve was generated based on a gradient of stained THP1 cells, and the number of adhered cells was calculated based on the fluorescence intensity, i-Control software was used for data collection (Tecan). For adhesion confocal microscopy images, endothelial cells were double-stained with Phalloidin (Invitrogen\u0026trade;, A12381) and DAPI, and images were captured after monocyte addition, incubation, and washing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eApoptosis assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eApoptosis assay for gene-edited HUVECs was conducted using the RealTime-Glo\u0026trade; Annexin V Apoptosis kit (Promega, JA1000) according to the manufacturer\u0026rsquo;s protocol. This live-cell, non-lytic, real-time assay detects phosphatidylserine (PS) exposure on the outer leaflet of the cell membrane during apoptosis. HUVEC cells were plated at a density of 5,000 cells/cm\u0026sup2; on gelatin-coated 96-well plates, incubated for 48 hours to establish a monolayer, and then treated with 100 ng/mL LPS. A 2X Detection Reagent was created by diluting each component 500-fold in the complete cell culture medium. An equal volume (100 \u0026micro;L) of the 2X Detection Reagent was added to each well. The assay was incubated for various time intervals, and apoptosis was assessed by measuring relative fluorescence units (RFU) using green fluorescence with excitation at 485 nm and emission at 525 - 530 nm (Infinite M200Pro spectrophotometer - Tecan). Fluorescence units (RFU) were assessed using M200Pro (Tecan) and i-Control software (Tecan).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA stability assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the RNA stability assay, we treated cells with 10 \u0026mu;g/mL actinomycin D, a transcription inhibitor. The negative control (mock) was a medium with 10 \u0026mu;l/mL DMSO. After 48 hours of CRISPRa in cells, the culture medium was replaced with 10 \u0026mu;g/mL actinomycin D medium or mock medium and incubated at 37\u0026deg;C with 5% CO\u003csup\u003e2\u003c/sup\u003e. RNA samples were obtained after 0, 3, and 6 hours of treatment for RNA degradation comparison by qPCR.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA interference assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRNA interference assay was used for assessing potential microRNA binding on \u003cem\u003eCDH13\u0026nbsp;\u003c/em\u003e3\u0026rsquo;UTR and conducted using the Dual-Luciferase\u0026trade; Reporter Assay System (Promega, E1910) following the manufacturer\u0026rsquo;s instructions. The relevant 3\u0026rsquo;UTR region of \u003cem\u003eCDH13\u003c/em\u003e consists of 1323 bp. The 1323bp sequence (supplementary Table 17) flanking the Xhol and Notl cutting sites was synthesized via GenScript Biotech (Netherlands) B.V. and inserted into the psiCHECK\u0026trade;-2 Vector backbone (Promega, C8021) via restriction enzyme cloning. For the assay, HEK.293T cells were seeded in 12-well plates and incubated at 37\u0026deg;C with 5% CO2. Upon reaching 70-80% confluency, cells were co-transfected in triplicate with 600 ng of the plasmid DNA construct and 50 nmol of microRNA. Following 48 hours, the cells were cooled on ice, rinsed once with 500 \u0026mu;L ice-cold 1\u0026times; PBS, and then lysed with 50 \u0026mu;L 1\u0026times; Passive Lysis Buffer (Promega, E1910). After centrifugation, 10 \u0026mu;L of the lysate was transferred to a white 96-well plate for a dual luciferase assay following the manufacturer\u0026rsquo;s protocol. Luciferase activities were assessed using M200Pro (Tecan) and i-control software (Tecan). The luciferase activity was quantified as the ratio of Renilla to Firefly luciferase activity and reported in arbitrary light units.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data of biological experiments were obtained from a minimum of three independent biological replicates. Quantitative results are expressed as the mean \u0026plusmn; standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism, applying unpaired t-tests, one-way ANOVA with multiple comparisons, or two-way ANOVA for factorial analyses, as appropriate. Statistical significance was defined as *, P \u0026le; 0.05; **, P \u0026le; 0.01; ***, P \u0026le; 0.001; ****, P \u0026le; 0.0001.\u0026nbsp;Description of sample size (number of LoF mutation carriers, non-carriers, total sample size) and statistical tests performed for genetic association analysis, can be found in the corresponding sections of the main text, Method details, and Supplementary Tables.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and code availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis paper analyzes existing data of single-cell RNA-seq data from Gene Expression Omnibus (GEO), dataset GSE131778, GSE247238, and GSE247238 . Original western blot images will be deposited at the required platform and publicly available as of the date of publication. Microscopy data reported in this paper will be shared by the lead contact upon request. All original code will be deposited at the required platform and will be publicly available as of the date of publication. Any additional information required to reanalyze the data reported in this paper will be available from the lead contact upon request.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData and code availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis paper analyzes existing data of single-cell RNA-seq data from Gene Expression Omnibus (GEO), dataset GSE131778, GSE247238, and GSE247238 . Original western blot images will be deposited at the required platform and publicly available as of the date of publication. Microscopy data reported in this paper will be shared by the lead contact upon request. All original code will be deposited at the required platform and will be publicly available as of the date of publication. Any additional information required to reanalyze the data reported in this paper will be available from the lead contact upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZ.C., S.L., and H.S. conceptualized the study, designed the experiments, supervised and led the trainees, and co-wrote the manuscript. Z.C. and S.L. conducted cellular and molecular experiments, analyzed data, and visualized data. S.L. and X.S. performed animal experiments and analyzed data. A.D. conducted genetic analyses. L.L. contributed to GWAS and colocalization analysis. A.M. led trainees and contributed to flow cytometry and FACS data analysis. L.M. provided human carotid artery data and Z.L. conducted the corresponding analysis. Y.C. conducted single-cell data analysis. T.D. led trainees and designed mouse experiments. M.L. contributed to animal tissue harvest. R.S. conducted flow cytometry and analyzed FACS data. C.L. and A.R. cloned the CRISPR constructs. L.F.B. quantified and analyzed immunostaining data. N.B. and M.C. explored the SMC-specific eQTLs at the 16q23.3 locus. C. E. R and, A. J. L explored the EC-specific eQTLs at the \u003cem\u003e16q23.3\u003c/em\u003e locus. H.S. R.B. and T.K. helped with data integration and edited the manuscript. M.S., J.K., N.K., A.S.and M.V.S. advised the methods of statistical analyses and edited the manuscript. L.M. and J.L.M.B. provided the STARNET dataset. M.N.J provided the materials, supervised the microRNA-related experiments, and edited the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe acknowledge UK Biobank (project code 25214), and CARDIoGRAMplusC4D Consortium, and the Common Metabolic Diseases Knowledge Portal for providing research data. We thank the animal facility of the German Heart Center Munich and the animal welfare officer, Dr. Susanne Naumann for supporting our mouse experiments. We thank Dr. Zhiyuan Wu who explored RNA sequencing data of atherosclerosis plaques from patients undergoing carotid endarterectomy for this project. We thank Christopher Wolf, an experiment assistant, who supported our experiment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSources of Funding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe appreciate the major funding for the current projects, including Sonderforschungsbereich SFB TRR 267 (DFG, 403584255, project B05_Z.C. \u0026amp; H.S.), the German Research Foundation (DFG 510049865), Corona-Stiftung Nachwuchsforschungsgruppe (Junior Research Group Grant), and the German Centre for Cardiovascular Research (DZHK) (“Förderkennzeichen”, ID: 81X3600510) (Z.C.). The work was further supported by the German Federal Ministry of Education and Research (BMBF) within the scheme of target validation (BlockCAD: 16GW0198K), COMMITMENT (01ZX1904A), e:Med research and funding concept (AbCD-Net: 01ZX1706C) (H.S.). As a Co-applicant of the British Heart Foundation (BHF)/DZHK-collaboration (DZHK-BHF: 81X2600522) and the Leducq Foundation for Cardiovascular Research (PlaqOmics: 18CVD02), we gratefully acknowledge their funding (H.S.). Further, we recognize the support of the Bavarian State Ministry of Health and Care which funded this work with DigiMed Bayern (grant No: DMB-1805–0001) within its Masterplan “Bayern Digital II” and of the German Federal Ministry of Economics and Energy in its scheme of ModulMax (grant No: ZF4590201BA8) (H.S.). Finally, we thank the support from National Institutes of Health (NIH) grants DK136405 and DK117850 (A.J.L.), NIH grant R01HL156120 (M.C.), and an American Heart Association Established Investigator Award 24EIA1258067 (M.C.).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclosures\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNA. The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrespondence and requests for materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFurther information and requests for resources and reagents should be directed to and will be fulfilled by Zhifen Chen ([email protected]) and Shuangyue Li ([email protected]).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAragam, K.G.\u003cem\u003e et al.\u003c/em\u003e Discovery and systematic characterization of risk variants and genes for coronary artery disease in over a million participants. \u003cem\u003eNat Genet\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 1803-1815 (2022).\u003c/li\u003e\n\u003cli\u003eTcheandjieu, C.\u003cem\u003e et al.\u003c/em\u003e Large-scale genome-wide association study of coronary artery disease in genetically diverse populations. \u003cem\u003eNat Med\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1679-1692 (2022).\u003c/li\u003e\n\u003cli\u003eChen, Z. \u0026amp; Schunkert, H. Genetics of coronary artery disease in the post-GWAS era. \u003cem\u003eJ Intern Med\u003c/em\u003e \u003cstrong\u003e290\u003c/strong\u003e, 980-992 (2021).\u003c/li\u003e\n\u003cli\u003eMusunuru, K. \u0026amp; Kathiresan, S. Genetics of Common, Complex Coronary Artery Disease. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e177\u003c/strong\u003e, 132-145 (2019).\u003c/li\u003e\n\u003cli\u003eHodonsky, C.J.\u003cem\u003e et al.\u003c/em\u003e Multi-ancestry genetic analysis of gene regulation in coronary arteries prioritizes disease risk loci. \u003cem\u003eCell Genom\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 100465 (2024).\u003c/li\u003e\n\u003cli\u003eKim, H.J.\u003cem\u003e et al.\u003c/em\u003e Molecular mechanisms of coronary artery disease risk at the PDGFD locus. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 847 (2023).\u003c/li\u003e\n\u003cli\u003eCui, G.\u003cem\u003e et al.\u003c/em\u003e A functional variant in APOA5/A4/C3/A1 gene cluster contributes to elevated triglycerides and severity of CAD by interfering with microRNA 3201 binding efficiency. \u003cem\u003eJ Am Coll Cardiol\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 267-277 (2014).\u003c/li\u003e\n\u003cli\u003eCheng, P.\u003cem\u003e et al.\u003c/em\u003e Smad3 regulates smooth muscle cell fate and mediates adverse remodeling and calcification of the atherosclerotic plaque. \u003cem\u003eNat Cardiovasc Res\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 322-333 (2022).\u003c/li\u003e\n\u003cli\u003ede Goede, O.M.\u003cem\u003e et al.\u003c/em\u003e Population-scale tissue transcriptomics maps long non-coding RNAs to complex disease. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e184\u003c/strong\u003e, 2633-2648 e2619 (2021).\u003c/li\u003e\n\u003cli\u003eWerner, A., Kanhere, A., Wahlestedt, C. \u0026amp; Mattick, J.S. Natural antisense transcripts as versatile regulators of gene expression. \u003cem\u003eNat Rev Genet\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 730-744 (2024).\u003c/li\u003e\n\u003cli\u003eFarina, F.M., Weber, C. \u0026amp; Santovito, D. The emerging landscape of non-conventional RNA functions in atherosclerosis. \u003cem\u003eAtherosclerosis\u003c/em\u003e \u003cstrong\u003e374\u003c/strong\u003e, 74-86 (2023).\u003c/li\u003e\n\u003cli\u003eFaghihi, M.A.\u003cem\u003e et al.\u003c/em\u003e Evidence for natural antisense transcript-mediated inhibition of microRNA function. \u003cem\u003eGenome Biol\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, R56 (2010).\u003c/li\u003e\n\u003cli\u003eShang, R., Lee, S., Senavirathne, G. \u0026amp; Lai, E.C. microRNAs in action: biogenesis, function and regulation. \u003cem\u003eNat Rev Genet\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 816-833 (2023).\u003c/li\u003e\n\u003cli\u003eChen, L.L. \u0026amp; Kim, V.N. Small and long non-coding RNAs: Past, present, and future. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e187\u003c/strong\u003e, 6451-6485 (2024).\u003c/li\u003e\n\u003cli\u003eDzau, V.J. \u0026amp; Hodgkinson, C.P. RNA Therapeutics for the Cardiovascular System. \u003cem\u003eCirculation\u003c/em\u003e \u003cstrong\u003e149\u003c/strong\u003e, 707-716 (2024).\u003c/li\u003e\n\u003cli\u003eNemeth, K., Bayraktar, R., Ferracin, M. \u0026amp; Calin, G.A. Non-coding RNAs in disease: from mechanisms to therapeutics. \u003cem\u003eNat Rev Genet\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 211-232 (2024).\u003c/li\u003e\n\u003cli\u003eBauersachs, J.\u003cem\u003e et al.\u003c/em\u003e Efficacy and safety of CDR132L in patients with reduced left ventricular ejection fraction after myocardial infarction: Rationale and design of the HF-REVERT trial. \u003cem\u003eEur J Heart Fail\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 674-682 (2024).\u003c/li\u003e\n\u003cli\u003eNelson, C.P.\u003cem\u003e et al.\u003c/em\u003e Association analyses based on false discovery rate implicate new loci for coronary artery disease. \u003cem\u003eNat Genet\u003c/em\u003e \u003cstrong\u003e49\u003c/strong\u003e, 1385-1391 (2017).\u003c/li\u003e\n\u003cli\u003evan der Harst, P. \u0026amp; Verweij, N. Identification of 64 Novel Genetic Loci Provides an Expanded View on the Genetic Architecture of Coronary Artery Disease. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, 433-443 (2018).\u003c/li\u003e\n\u003cli\u003eLi, L.\u003cem\u003e et al.\u003c/em\u003e Transcriptome-wide association study of coronary artery disease identifies novel susceptibility genes. \u003cem\u003eBasic Res Cardiol\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 6 (2022).\u003c/li\u003e\n\u003cli\u003eKoplev, S.\u003cem\u003e et al.\u003c/em\u003e A mechanistic framework for cardiometabolic and coronary artery diseases. \u003cem\u003eNat Cardiovasc Res\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 85-100 (2022).\u003c/li\u003e\n\u003cli\u003eFasolo, F.\u003cem\u003e et al.\u003c/em\u003e Long Noncoding RNA MIAT Controls Advanced Atherosclerotic Lesion Formation and Plaque Destabilization. \u003cem\u003eCirculation\u003c/em\u003e \u003cstrong\u003e144\u003c/strong\u003e, 1567-1583 (2021).\u003c/li\u003e\n\u003cli\u003ePaloschi, V.\u003cem\u003e et al.\u003c/em\u003e Utilization of an Artery-on-a-Chip to Unravel Novel Regulators and Therapeutic Targets in Vascular Diseases. \u003cem\u003eAdv Healthc Mater\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e2302907 (2024).\u003c/li\u003e\n\u003cli\u003eWirka, R.C.\u003cem\u003e et al.\u003c/em\u003e Atheroprotective roles of smooth muscle cell phenotypic modulation and the TCF21 disease gene as revealed by single-cell analysis. \u003cem\u003eNat Med\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1280-1289 (2019).\u003c/li\u003e\n\u003cli\u003eHan, S.\u003cem\u003e et al.\u003c/em\u003e RNA-protein interaction mapping via MS2- or Cas13-based APEX targeting. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e117\u003c/strong\u003e, 22068-22079 (2020).\u003c/li\u003e\n\u003cli\u003eLiao, H.K.\u003cem\u003e et al.\u003c/em\u003e In Vivo Target Gene Activation via CRISPR/Cas9-Mediated Trans-epigenetic Modulation. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e171\u003c/strong\u003e, 1495-1507 e1415 (2017).\u003c/li\u003e\n\u003cli\u003eChen, Z.\u003cem\u003e et al.\u003c/em\u003e Functional Screening of Candidate Causal Genes for Insulin Resistance in Human Preadipocytes and Adipocytes. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e126\u003c/strong\u003e, 330-346 (2020).\u003c/li\u003e\n\u003cli\u003eWang, J.\u003cem\u003e et al.\u003c/em\u003e LncRNA PTTG3P promotes tumorigenesis and metastasis of NSCLC by binding with ILF3 to maintain mRNA stability and form a positive feedback loop with E2F1. \u003cem\u003eInt J Biol Sci\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 4291-4310 (2023).\u003c/li\u003e\n\u003cli\u003eWang, Y.\u003cem\u003e et al.\u003c/em\u003e LncRNA LINRIS stabilizes IGF2BP2 and promotes the aerobic glycolysis in colorectal cancer. \u003cem\u003eMol Cancer\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 174 (2019).\u003c/li\u003e\n\u003cli\u003eLiu, G., Zhang, R., Xu, J., Wu, C.I. \u0026amp; Lu, X. Functional conservation of both CDS- and 3'-UTR-located microRNA binding sites between species. \u003cem\u003eMol Biol Evol\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 623-628 (2015).\u003c/li\u003e\n\u003cli\u003eHuang, H.Y.\u003cem\u003e et al.\u003c/em\u003e miRTarBase 2020: updates to the experimentally validated microRNA-target interaction database. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e48\u003c/strong\u003e, D148-D154 (2020).\u003c/li\u003e\n\u003cli\u003eSticht, C., De La Torre, C., Parveen, A. \u0026amp; Gretz, N. miRWalk: An online resource for prediction of microRNA binding sites. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, e0206239 (2018).\u003c/li\u003e\n\u003cli\u003eSoutschek, M., Gross, F., Schratt, G. \u0026amp; Germain, P.L. scanMiR: a biochemically based toolkit for versatile and efficient microRNA target prediction. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 2466-2473 (2022).\u003c/li\u003e\n\u003cli\u003eMcGeary, S.E.\u003cem\u003e et al.\u003c/em\u003e The biochemical basis of microRNA targeting efficacy. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e366\u003c/strong\u003e (2019).\u003c/li\u003e\n\u003cli\u003eAgarwal, V., Bell, G.W., Nam, J.W. \u0026amp; Bartel, D.P. Predicting effective microRNA target sites in mammalian mRNAs. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e (2015).\u003c/li\u003e\n\u003cli\u003eKavakiotis, I., Alexiou, A., Tastsoglou, S., Vlachos, I.S. \u0026amp; Hatzigeorgiou, A.G. DIANA-miTED: a microRNA tissue expression database. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e50\u003c/strong\u003e, D1055-D1061 (2022).\u003c/li\u003e\n\u003cli\u003eFujishima, Y.\u003cem\u003e et al.\u003c/em\u003e Adiponectin association with T-cadherin protects against neointima proliferation and atherosclerosis. \u003cem\u003eFASEB J\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 1571-1583 (2017).\u003c/li\u003e\n\u003cli\u003eIshii, N.\u003cem\u003e et al.\u003c/em\u003e Identification of a novel non-coding RNA, MIAT, that confers risk of myocardial infarction. \u003cem\u003eJ Hum Genet\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 1087-1099 (2006).\u003c/li\u003e\n\u003cli\u003eSamani, N.J.\u003cem\u003e et al.\u003c/em\u003e Genomewide association analysis of coronary artery disease. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e357\u003c/strong\u003e, 443-453 (2007).\u003c/li\u003e\n\u003cli\u003eSingh, R.\u003cem\u003e et al.\u003c/em\u003e The hidden players: Shedding light on the significance of post-translational modifications and miRNAs in Alzheimer's disease development. \u003cem\u003eAgeing Res Rev\u003c/em\u003e \u003cstrong\u003e90\u003c/strong\u003e, 102002 (2023).\u003c/li\u003e\n\u003cli\u003eWong, C.H.\u003cem\u003e et al.\u003c/em\u003e CircRTN4 promotes pancreatic cancer progression through a novel CircRNA-miRNA-lncRNA pathway and stabilizing epithelial-mesenchymal transition protein. \u003cem\u003eMol Cancer\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 10 (2022).\u003c/li\u003e\n\u003cli\u003eJonas, S. \u0026amp; Izaurralde, E. Towards a molecular understanding of microRNA-mediated gene silencing. \u003cem\u003eNat Rev Genet\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 421-433 (2015).\u003c/li\u003e\n\u003cli\u003eWu, L., Fan, J. \u0026amp; Belasco, J.G. MicroRNAs direct rapid deadenylation of mRNA. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e103\u003c/strong\u003e, 4034-4039 (2006).\u003c/li\u003e\n\u003cli\u003ePeter, M.E. Targeting of mRNAs by multiple miRNAs: the next step. \u003cem\u003eOncogene\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 2161-2164 (2010).\u003c/li\u003e\n\u003cli\u003eZhu, Y.\u003cem\u003e et al.\u003c/em\u003e LncRNA MIR200CHG inhibits EMT in gastric cancer by stabilizing miR-200c from target-directed miRNA degradation. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 8141 (2023).\u003c/li\u003e\n\u003cli\u003eRidker, P.M.\u003cem\u003e et al.\u003c/em\u003e Lipid-Reduction Variability and Antidrug-Antibody Formation with Bococizumab. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e376\u003c/strong\u003e, 1517-1526 (2017).\u003c/li\u003e\n\u003cli\u003eKoren, M.J.\u003cem\u003e et al.\u003c/em\u003e Preclinical development and phase 1 trial of a novel siRNA targeting lipoprotein(a). \u003cem\u003eNat Med\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 96-103 (2022).\u003c/li\u003e\n\u003cli\u003eRosenson, R.S.\u003cem\u003e et al.\u003c/em\u003e Evinacumab in Patients with Refractory Hypercholesterolemia. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e383\u003c/strong\u003e, 2307-2319 (2020).\u003c/li\u003e\n\u003cli\u003eBallantyne, C.M.\u003cem\u003e et al.\u003c/em\u003e Plozasiran, an RNA Interference Agent Targeting APOC3, for Mixed Hyperlipidemia. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e391\u003c/strong\u003e, 899-912 (2024).\u003c/li\u003e\n\u003cli\u003eKraler, S.\u003cem\u003e et al.\u003c/em\u003e Resilience of the Internal Mammary Artery to Atherogenesis: Shifting From Risk to Resistance to Address Unmet Needs. \u003cem\u003eArterioscler Thromb Vasc Biol\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 2237-2251 (2021).\u003c/li\u003e\n\u003cli\u003eLoop, F.D. Internal-thoracic-artery grafts. Biologically better coronary arteries. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e334\u003c/strong\u003e, 263-265 (1996).\u003c/li\u003e\n\u003cli\u003eLucas, T., Bonauer, A. \u0026amp; Dimmeler, S. RNA Therapeutics in Cardiovascular Disease. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e123\u003c/strong\u003e, 205-220 (2018).\u003c/li\u003e\n\u003cli\u003eBallantyne, M.D., McDonald, R.A. \u0026amp; Baker, A.H. lncRNA/MicroRNA interactions in the vasculature. \u003cem\u003eClin Pharmacol Ther\u003c/em\u003e \u003cstrong\u003e99\u003c/strong\u003e, 494-501 (2016).\u003c/li\u003e\n\u003cli\u003eCaporali, A.\u003cem\u003e et al.\u003c/em\u003e Non-coding RNAs as therapeutic targets and biomarkers in ischaemic heart disease. \u003cem\u003eNat Rev Cardiol\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 556-573 (2024).\u003c/li\u003e\n\u003cli\u003eMishra, A.\u003cem\u003e et al.\u003c/em\u003e Stroke genetics informs drug discovery and risk prediction across ancestries. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e611\u003c/strong\u003e, 115-123 (2022).\u003c/li\u003e\n\u003cli\u003eFranzen, O.\u003cem\u003e et al.\u003c/em\u003e Cardiometabolic risk loci share downstream cis- and trans-gene regulation across tissues and diseases. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e353\u003c/strong\u003e, 827-830 (2016).\u003c/li\u003e\n\u003cli\u003eGiambartolomei, C.\u003cem\u003e et al.\u003c/em\u003e Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. \u003cem\u003ePLoS Genet\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e1004383 (2014).\u003c/li\u003e\n\u003cli\u003eWang, Y.\u003cem\u003e et al.\u003c/em\u003e Myeloid cannabinoid CB1 receptor deletion confers atheroprotection in male mice by reducing macrophage proliferation in a sex-dependent manner. \u003cem\u003eCardiovasc Res\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, 1411-1426 (2024).\u003c/li\u003e\n\u003cli\u003eBackman, J.D.\u003cem\u003e et al.\u003c/em\u003e Exome sequencing and analysis of 454,787 UK Biobank participants. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e599\u003c/strong\u003e, 628-634 (2021).\u003c/li\u003e\n\u003cli\u003eMbatchou, J.\u003cem\u003e et al.\u003c/em\u003e Computationally efficient whole-genome regression for quantitative and binary traits. \u003cem\u003eNat Genet\u003c/em\u003e \u003cstrong\u003e53\u003c/strong\u003e, 1097-1103 (2021).\u003c/li\u003e\n\u003cli\u003eButler, A., Hoffman, P., Smibert, P., Papalexi, E. \u0026amp; Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. \u003cem\u003eNat Biotechnol\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 411-420 (2018).\u003c/li\u003e\n\u003cli\u003eStuart, T.\u003cem\u003e et al.\u003c/em\u003e Comprehensive Integration of Single-Cell Data. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e177\u003c/strong\u003e, 1888-1902 e1821 (2019).\u003c/li\u003e\n\u003cli\u003eBecht, E.\u003cem\u003e et al.\u003c/em\u003e Dimensionality reduction for visualizing single-cell data using UMAP. \u003cem\u003eNat Biotechnol\u003c/em\u003e (2018).\u003c/li\u003e\n\u003cli\u003eSharifi, M.A.\u003cem\u003e et al.\u003c/em\u003e ADAMTS-7 Modulates Atherosclerotic Plaque Formation by Degradation of TIMP-1. \u003cem\u003eCirc Res\u003c/em\u003e \u003cstrong\u003e133\u003c/strong\u003e, 674-686 (2023).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Coronary artery disease (CAD), long non-coding RNA (lncRNA), microRNA (miRNA), GWAS-eQTL colocalization analysis, dRfxCas13d, CRISPR/Cas9, CRISPR activation (CRISPRa), 3’ untranslated region (3’ UTR), RNA stability, and RNA interference ","lastPublishedDoi":"10.21203/rs.3.rs-7333062/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7333062/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMany common diseases have a polygenic architecture. The responsible alleles are thought to mediate risk by disturbing gene regulation in most cases, however, the precise mechanisms have been elucidated only for a few. Here, we investigated the \u003cem\u003e16q23.3\u003c/em\u003e genomic locus, which genome-wide significantly associates with coronary artery disease, a globally leading cause of death caused by accumulation of lipid-rich inflammatory plaques in the arterial wall. The locus harbors \u003cem\u003eCDH13, \u003c/em\u003ewhose mRNA and protein we found to be suppressed in atherosclerotic human and mouse arteries. Loss-of-function(LoF) variants of \u003cem\u003eCDH13\u003c/em\u003e were associated with detrimental cardiovascular phenotypes in the UK Biobank. Its knock-out increased plaque-sizes in \u003cem\u003eCdh13\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/-\u003c/em\u003e\u003c/sup\u003e/\u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice compared to \u003cem\u003eApoe\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/-\u003c/em\u003e\u003c/sup\u003e mice on a Western diet. After establishing an atheroprotective role of\u003cem\u003e \u003c/em\u003eCDH13\u003cem\u003e,\u003c/em\u003e we studied its regulation. Integration of population genomic and transcriptomic datasets by GWAS-eQTL colocalization analysis identified \u003cem\u003eCDH13\u003c/em\u003e and four long non-coding RNAs (lncRNAs) as candidate causal genes at the \u003cem\u003e16q23.3\u003c/em\u003e locus. dCas13-mediated RNA immunoprecipitation revealed that the lncRNA \u003cem\u003eCDH13-AS2\u003c/em\u003e binds to \u003cem\u003eCDH13\u003c/em\u003e mRNA in human endothelial cells (ECs). Its CRISPR/Cas9-based knockout in ECs was atherogenic, whereas dCas9-based transcriptional activation (CRISPRa) of \u003cem\u003eCDH13-AS2\u003c/em\u003e was atheroprotective; effects that were found to be mediated by the stability of \u003cem\u003eCDH13\u003c/em\u003e mRNA. To further understand how the \u003cem\u003eCDH13-AS2\u003c/em\u003e protects the mRNA we searched \u003cem\u003ein silico\u003c/em\u003e and screened \u003cem\u003ein vitro\u003c/em\u003e for microRNAs (miRNAs) that bind to \u003cem\u003eCDH13\u003c/em\u003e 3’UTR. Indeed, four miRNAs, miR-19b-3p, miR-125b-2-3p, miR-433-3p, and miR-7b-5p, were found experimentally to accelerate \u003cem\u003eCDH13\u003c/em\u003e mRNA degradation, an effect that was neutralized by CRISPRa of \u003cem\u003eCDH13-AS2\u003c/em\u003e. Taken together, our study demonstrates an interplay of miRNAs, lncRNAs, and mRNA, which modulates the abundance of an atheroprotective protein in endothelial cells, which may offer a new therapeutic target for coronary artery disease.\u003c/p\u003e","manuscriptTitle":"An interplay of non-coding RNAs regulates CDH13 expression and affects endothelial function and coronary artery disease risk","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-20 08:15:32","doi":"10.21203/rs.3.rs-7333062/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-cardiovascular-research","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natcardiovascres","sideBox":"Learn more about [Nature Cardiovascular Research](https://www.nature.com/natcardiovascres/)","snPcode":"","submissionUrl":"https://mts-natcardiovascres.nature.com/cgi-bin/main.plex","title":"Nature Cardiovascular Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d0c33b90-82f5-4fde-8f9c-265cc2a5df09","owner":[],"postedDate":"August 20th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":53425633,"name":"Health sciences/Cardiology/Cardiovascular biology/Cardiovascular genetics"},{"id":53425634,"name":"Biological sciences/Molecular biology/Non-coding RNAs"}],"tags":[],"updatedAt":"2025-10-06T15:08:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-20 08:15:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7333062","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7333062","identity":"rs-7333062","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}