| BackgroundCoronary heart disease（CHD）is one of the main causes of death in the world.The pathophysiological feature of coronary heart disease is the formation of atherosclerotic plaques.Repeated lipid deposition,inflammatory cell infiltration,calcification and internal bleeding in plaque lead to coronary artery stenosis,myocardial ischemia and intermittent or persistent angina.Bleeding and rupture of plaque can lead to thrombosis,myocardial infarction,stroke and other acute cardiovascular and cerebrovascular events.The risk of coronary heart disease is regulated by the interaction between genes and lifestyle factors.It is of great significance to explore the molecular mechanism that affects the occurrence and development of coronary heart disease and to find the potential therapeutic target of coronary heart disease.Circular RNAs（circRNAs）are a series of non-coding RNAs（ncRNAs）that generated from exons or introns cyclization of specific coding genes,which termed as the parental genes.The ring structure allows circRNAs to resist ribonuclease R（RNase R）and confers a longer half-life than that of linear RNAs.The same parent gene can form different circRNAs,which can be generated by single exon circRNA,intron circRNA or exon-intron circRNA.Emerging studies have revealed that circRNAs,especially cytoplasmic exonic circRNAs,function as endogenous RNAs sponges which can competitively bind to microRNAs（miRNAs）by completely or partially complementary base binding,thus suppressing miRNA activity and participating in post-transcriptional regulation of miRNA target genes,and thereby regulating cell proliferation,apoptosis,or other physiological activities in cancer,neurobiology and immune system.The impact of circRNA on CHD remains poorly characterized.Limited studies have found that circRNA is involved in the regulation of cardiac hypertrophy,heart failure and myocardial fibrosis by combining with miRNA.In CHD,circRNA is involved in regulating the proliferation and migration of vascular endothelial cells and smooth muscle cells,which may affect the development of atherosclerotic plaque.At present,based on a small number of peripheral blood samples of circRNA microarray detection,it was found that the expression profile of circRNA in patients with CHD was different from that in the control group.However,it has not been reported that highthroughput sequencing technology was used to study the expression profile of circRNA in peripheral blood samples of patients with CHD and to verify the different circRNA in human coronary tissues.Since circRNAs generate from their parental genes,whether circRNAs could regulate their parental genes through miRNAs sponge remain unknown in CHD.Objectives1.To explore the expression profile of circRNA in peripheral blood mononuclear cells of patients with CHD and find out the differential expressed circRNA.2.To determine the expression level of the dysregulated circRNA in human atherosclerotic coronary tissues.3.To explore the possible functional genes and enrichment pathways involved in the mechanism of CHD related parental gene-circRNA-miRNA-mRNA axis in the occurrence and development of CHD.Materials and methods1.Collection and treatment of CHD and control blood samplesWe included 100 participants who underwent diagnostic percutaneous coronary intervention（PCI）from 2016.10 to 2017.05 in Qilu Hospital of Shandong University.The participants were divided into two groups:Patients with coronary heart disease（70 cases,the maximum stenosis rate of trunk>50%）and the control group（30 cases,the maximum stenosis rate of trunk<50%）,excluding other basic diseases and other cardiovascular diseases.Peripheral blood mononuclear cells（PBMCs）were extracted.The information of age,gender,blood pressure,triglyceride,cholesterol,LDL-C,HDLC,diabetes history,smoking history,alcohol intake,pharmacological therapy history were recorded.2.Isolation and treatment of PBMCs circRNATotal RNA was isolated with Trizol.RNA concentration and quality were assessed using a NanoDrop ND1000 spectrophotometer（NanoDrop,Wilmington,DE,USA）.OD260/OD280 ratios between 1.8 and 2.1 were deemed acceptable.Then,sequencing libraries were generated from the rRNA-depleted and RNase R-digested RNAs using an NEBNext? UltraTM Directional RNA Library Prep Kit for Illumina?（NEB,USA）following the manufacturer’s recommendations.3.Construction of high throughput sequencing LibraryQualified RNA samples were fragmented and were then used for first-and secondstrand cDNA synthesis with random hexamer primers.The library was purified and then qualified using an Agilent Bioanalyzer 2100 system.4.Identification and quantification of circRNAsAfter cluster generation,the library preparations were sequenced on an Illumina HiSeq 2500 platform,and raw reads were generated.Raw reads were cleaned using the Trimmomatic program to acquire the clean reads.The clean reads were aligned to the reference human genome（hg38）with Bowtie2 and counting by HTseq in union mode.Untrimmed reads mapping contiguously to the genome were excluded.The remaining reads were used for the potential identification and characterisation of circRNAs through the findcirc pipeline.Briefly,anchors（extracted from both ends of the reads）that aligned in the reverse orientation（head to tail）indicated a back-spliced junction such as that observed in circRNAs.Differential expression analysis between the two groups was performed using DESeq2.The adjusted p-value（padj）is the p-value adjusted for multiple testing using Benjamini-Hochberg to estimate the false discovery rate（FDR）.circRNAs with a padj<0.05 and a fold change≥2.0 or≤0.5 were considered differentially expressed.5.Prediction of CHD-related parental gene-circRNA-miRNA-parental mRNA interactionsFirst,we selected 12 dysregulated exonic circRNAs whose parental genes had association with CHD through literature retrieval.We then used algorithms in miRNA prediction software（miRanda and TargetScan）to obtain candidate miRNAs.All the circRNA-miRNAs conformed to seed-sequence complementarity.Next,we selected the top 100 nodes from the candidate miRNAs whose targets included the 12 circRNA parental mRNAs.Then,we used Cytoscape 3.01 to establish a parental gene-circRNAmiRNA-parental mRNA network.6.Bioinformatic analysisGene ontology（GO）and Kyoto Encyclopedia of Genes and Genomes（KEGG）pathway analyses were performed to predict the biological functions of the circRNAs implicated in CHD.Hierarchical clustering of the top 100 differentially expressed circRNAs were performed in R heatmap.2.7.Histopathology staining of human coronary arteriesThe coronary arteries were dissected into arterial rings with or without plaques in 5-6 mm length（judged with a stereomicroscope）and before RNA isolation,2 mm of the arterial ring was dissected for H&E and oil red staining to match coronary arteries with atherosclerotic plaques with normal non-plaque arteries.8.Quantitative real-time polymerase chain reaction（qRT-PCR）of human coronary arteriesThe tissue RNA were extracted with Trizol and total RNA were removed with gDNA,rRNA and linear RNA.Next,cDNA was synthesised with a reverse transcriptase and circRNA expression was determined using qPCR.GAPDH was chosen as the control gene.9.Statistical analysisData are presented as the mean±SD or proportions when appropriate.Categorical variables were tested using the Chi-square test.Continuous variables were first tested by Kolmogorov-Smimov and Shapiro-Wilk tests to verify whether the data sets were normally distributed.If normally distributed,the data were analysed using two-tailed Student’s t-tests.If not,Mann-Whitney U tests was performed.Statistical analyses were performed using SPSS 25.0.P<0.05 was considered statistically significant.Results1.Characteristics of human CHD blood samplesA total of 70 CHD patients and 30 controls were selected based on the results of diagnostic coronary angiography.Stenosis of the left main coronary trunk or maximum stenosis in a major epicardial artery（%）in the CHD group was 79.21±17.40,whereas stenosis in the control group was 16.67±17.29.There were no significant differences in age,gender,diabetes history,smoking history,alcohol intake,blood pressure,or triglycerides（TG）,cholesterol,LDL-C,or HDL-C levels between CHD patients and controls.2.Characteristics of human arteriesThe coronary arteries were acquired from the Organ Transplantation and Donation Department at Qilu Hospital of Shandong University.Six coronary artery donors died from brain injuries and had no other known diseases.3.PBMC circRNAs in CHD patients and controlsHigh-throughput sequencing of the 100 PBMC samples showed 24364 circRNAs in both the CHD patients and controls,whereas 22568 circRNAs were expressed in only one of the groups.Of the circRNAs,59%of reads were from exons,14.2%from introns,and 19.5%were sense overlapped.2283 circRNAs were significantly downregulated and 85 were significantly upregulated in the CHD patients.The distributions of the aberrant human circRNAs revealed that they were transcribed from all chromosomes except chromosome Y.Hierarchical clustering indicated the top 70 downregulated and 30 upregulated circRNAs differentially expressed in the two groups.4.Dysregulated circRNAs in various pathways that regulate metabolism,protein modification,and protein binding in CHDWe investigated the parental genes of differentially expressed circRNAs by GO and KEGG analyses.The 70 downregulated and 30 upregulated circRNAs were analysed separately.In the BP part of GO analysis,various biological processes such as organelle organisation,cellular protein metabolic,and protein modification were enriched.The CC results showed that the downregulated circRNAs origin genes enriched in intracellular part and nucleoplasm.The most enriched GO terms in the MF were protein binding.The downregulated circRNAs could also take part in various pathways such as MAPK,FoxO,T cell receptor,and B cell receptor signalling pathways.5.Construction of a CHD related gene-circRNA-miRNA-mRNA networkOf the 100 circRNA candidates,we focused on parental genes and found 12 that were tightly associated with CHD.With these 12 dysregulated gene-circRNAs and their predicted crosstalk miRNAs,a gene-circRNA-miRNA-mRNA network was constructed.Further,we selected the top five related miRNAs for each predicted circRNA.Using miRanda and TargetScan,we then identified the top 20 target genes for each of these miRNAs for KEGG pathway analyses.Many cardiovascular diseaserelated pathways,including the p53 signalling pathway,adrenergic signalling pathway in cardiomyocytes,vascular smooth muscle contraction signalling pathway,and Ras signalling pathway,were associated with these genes.These results indicated that the 12 circRNAs are closely associated with gene regulation and cardiovascular pathophysiology.6.qRT-PCR validation of eight selected circRNAs from human coronary arteries Of the 12 circRNAs that were found to be closely associated with CHD,eight circRNA-specific primers were designed and validated by PCR and sequencing.Every circRNA was tested in six matched samples.The dysregulation of the six circRNAs in plaque tissues agreed with the results of sequencing in PBMCs samples.7.Three significantly downregulated circRNAs were involved in fatty acid biosynthesis and metabolism in cardiovascular diseaseOf the qRT-PCR validated circRNAs,there were three downregulated circRNAs（hsacirc0066187,hsacirc0030042,and hsacirc0003915）that might sponge two shared predicted miRNAs（hsa-miR-30c-1-3p and hsa-miR-30c-2-3p）.According to the KEGG analysis,the target genes of these two miRNAs were enriched in genes involved in fatty acid biosynthesis,fatty acid metabolism,and TGF-β pathways,which may affect CHD pathogenesis.Conclusions1.CircRNA are differentially expressed in CHD.2.Of the 8 selected circRNAs validated in atherosclerotic coronary arteries,6 circRNAs coincided with the result of sequencing in PBMCs..3.CHD related parental gene-circRNA-miRNA-parental mRNA axis may have the potential to play vital role in circRNA regulating CHD.BackgroundFormation and instability of atherosclerotic plaques are the common pathophysiologic basis of atherosclerosis（AS）.Through binding with LOX-1 and SRB1 receptors on the endothelial cell membrane,ox-LDL activates inflammatory processes,foam cell formation and apoptosis,and promotes the development of AS.Oxidised low-density lipoprotein（ox-LDL）stimulates a variety of atherogenic pathways involving inflammatory response,foam cell formation,and cell apoptosis.The high dose ox-LDL also triggers abnormal autophagy with overexpression of beclin1,activation of autophagy,and blockage of lysosomal degradation.Although autophagy is thought to be a degradation process of cellular components modified by oxidised lipids,the subsequent death and loss of endothelial cells may lead to instability of advanced plaque,increased platelet adhesion,and accelerated thrombosis.Therefore,regulation of the abnormal autophagy in vascular endothelial cells induced by ox-LDL may be a potential target for plaque stabilization.Beclinl is a key initial regulator in autophagy.It has been reported that beclin1 involved in the phosphatidylserine（PS）exposure process required for dead-cell clearance during embryonic development and ox-LDL-induced abnormal endothelial cell autophagy.Forkhead box O1（FOXO1）,a transcription factor involved in a series of intracellular functions,including autophagy,oxidative stress,mitochondrial dysfunction,and apoptosis,is a shared effector of multiple atherogenic pathways in endothelial cells.It inhibits autophagosome-lysosome fusion,leading to endothelial autophagic apoptosis in diabetes.In previous,we found that the expression of circ0030042（hsacirc0030042）was significantly down-regulated in patients with CHD by circRNA sequencing of PBMCs and validated in human coronary arteries.The circBase database showed that hsacirc0030042 was 1352nt in length and generated from the second exon of gene FOXO1.In HUVEC cells,circ0030042 was highly sequenced,while mouse FOXO1 derived mmucirc0010680 was had high homology with circ0030042,suggesting high conservation and tissue specificity of this circRNA.Whether circ0030042 is involved in the regulation of endothelial function is unknown.It is known that circRNA can play important role in transcriptional and post transcriptional RNA regulation as endogenous competitive RNA（ceRNA）.Through software analysis,we found that circ0030042 may contain the binding sites of Ago2,eIF4A3 and HuR.Whether circ0030042 has the function of ceRNA and whether it is involved in the regulation of abnormal autophagy induced by ox-LDL through regulating the parental gene FOXO1 or other autophagy related gene needs further study.Objectives1.To clarify the expression of circ0030042 in ox-LDL induced HUVECs;2.To investigate whether circ0030042 is involved in the regulation of ox-LDL induced HUVECs autophagy;3.To explore the mechanism of circ0030042 regulating the autophagy related proteins,beclin1 and FOXO1.Materials and methods1.HUVECs culture and ox-LDL interventionHUVECs were extracted from the umbilical cord of healthy pregnant women,cultured at 37℃,5%CO2,and passage the next day.The cells of 4-8 passage were exposed to 0,3,6,12,24h 100ug/ml ox-LDL.The expression of circ0030042 were detected using qPCR.The expression of FOXO1、beclinl、p62 and LC3B were detected by western blot.2.Construction of circ0030042-lentivirus stable overexpression HUVECsTo achieve lentivirus-mediated hsacirc0030042 overexpression,the lentivirus vector pLO-ciR and pLO-ciR were transfected into HUVECs respectively at MOI 10.The cycle junction site was validated by sequencing after PCR.The stable hsacirc0030042 over-expression HUVECs（c0030042）and pLO-ciR HUVECs（circN.C）were obtained by puromycin 1ug/ml selection.qPCR was used to test the expression of hsacirc0030042 with specific divergent primers.3.Construction of eIF4A3-lentivirus stable overexpression HUVECsTo achieve lentivirus-mediated human eIF4A3 overexpression,the lentiviral HBLV-h-eIF4A3-GFP-PURO and HBLV-GFP-PURO transfected into HUVECs respectively at MOI 10.The stable eIF4A3 overexpression HUVECs（h-eIF4A3-GFP）and GFP alone HUVECs（GFP-N.C）were obtained by puromycin lug/ml selection.Western blot was used to test the overexpression of eIF4A3.4.Transfection of stubRFP-sensGFP-LC3 adenovirusTo observe the autophagy of HUVECs,stubRFP-sensGFP-LC3 adenovirus（Genechem,Shanghai,China）were transfected into cells upon reaching 40-50%confluence according to the manufacturer’s instructions.Following transfection for 72 h and the following treatment,the autophagosomes were photographed using confocal laser scanning microscopy.5.siRNA and RNA interferenceUpon reaching 40-50%confluence,cells were transfected with specific siRNA or negative control siRNA using Lipofectamine 3000 in Opti-MEM.After 6 hours of transfection,the medium was replaced with complete ECM,and the cells were cultured for an additional 24h-48h for additional testing.6.Transfection of circ-delete plasmidUsing the catRAPID and Vienna RNA algorithm for circRNA secondary structure and RNA-protein interaction,hsacirc0030042-eIF4A3 binding was predicted to be mediated through three major RNA regions（180-227,812-867,and 1082-1137 nucleotides）.With the plasmid used in pLCDH-c0030042,we constructed special hsacirc0030042 overexpression plasmid in which three main binding sites of hsacirc0030042 to eIF4A3 were deleted（circ-delete）.The cycle junction was the same as pLCDH-c0030042.The circ-delete plasmids were transfected into 293A cells or HUVECs using Lipofectamine 3000.7.Northern blotNorthern blot was performed as others described.In brief,the samples were run on a 1%formaldehyde-polyacrylamide-urea gel,transferred to positively charged Hybond N+membranes（Amersham）followed by cross-linking through UV irradiation.The membranes were subjected to hybridization with 100 pmol 3’-digoxigenin（DIG）labeled probe for circ0030042 overnight at 50℃.DIG-labeled GAPDH probe was used as control.8.Western blotCells were lysed using RIPA or other lysis buffer to acquire total protein or nuclear/cytoplasmic protein.Equal amounts of proteins were separated through SDSPAGE gels.Then,proteins were transferred to methanol-activated polyvinylidene fluoride membranes,and incubated with primary antibodies overnight at 4℃.The membranes were incubated with secondary antibodies the next day for 1 h at room temperature.Blots were imaged with a LAS-4000 luminescent image analyser.Protein expression was quantified using Adobe Photoshop CS6,normalized to GAPDH expression in each sample,and expressed as a percentage of the control.9.RNA and gDNA extraction and quantitative real-time PCRTotal RNA was extracted from HUVECs using TRIzol.The RNA was treated with rRNA depletion or linear RNA depletion as required.The product was reversetranscribed into cDNA using a TAKARA PrimeScript RT Reagent Kit.The cDNA was subjected to qPCR using SYBR Green for the relative quantification of mRNA expression.The hsacirc0030042 divergent primers were used and GAPDH was used to normalize mRNA levels.The products were detected by agarose gel electrophoresis as needed.Total gDNA was extracted from HUVECs using Genomic DNA Extraction Kit（TAKARA）according to the procedure.The hsacirc0030042 convergent primers and divergent primers were used.The homo-GAPDH convergent primers and divergent primers were applied as control.The products were detected by agarose gel electrophoresis.10.Fluorescence in situ hybridization（FISH）HUVECs were fixed and pretreated with 4%paraformaldehyde and PBS with 0.5%Triton X-100.Then,h-circ0030042 Probe Mix and h-18s rRNAFISH Probe Mix（Ribo）were applied separately to hybridize with the samples at 37℃ overnight.Finally,with DAPI stained nucleus,the slides were photographed using confocal laser scanning microscopy to situate hsacirc0030042 in HUVECs.11.RNA Immunoprecipitation（RIP）RNA immunoprecipitation was performed using eIF4A3,Ago2,and HuR specific antibody.IgG antibody was used as negative control.The immunoprecipitation was performed using Magna RIP kit（Millipore）according to manufacturer’s instructions.In brief,HUVECs were lysed in lysis buffer with proteinase inhibitor cocktail and RNase inhibitor.The lysate was mixed with antibody coupled magnetic beads and left under rotation overnight at 4℃.Beads-antibody complex were subsequently washed thoroughly in RIP wash buffer.RNA purification and extraction used proteinase K digestion and phenol chloroform.With salt solution Ⅰ,salt solution Ⅱ and precipitate enhancer in ethanol,-80℃ for overnight,RNA were obtained for downstream RNA detection.12.Electron microscopy of autophagyThe treated cells were collected,1,000 rpm centrifuge 5 min,the supernatant was discarded,1 mL PBS resuspended,centrifuged 10 min,the supernatant was discarded,fixed with 2.5%glutaraldehyde,and then the the cells were embedded in spur resin after dehydration.Thin sections were cut on a Ultracut E microtome.Sectioned grids were stained with saturated solution of uranyl acetate and lead citrate.Sections were examined at 80 kV with a Hitachi transmission electron microscope.13.Indirect immunofluorescence assay and confocal microscopyUpon reaching 70-80%confluence,cells were fixed in 4%paraformaldehyde,permeabilized with PBS containing 0.5%Triton-X-100 for 15 min and blocked with PBS containing 5%bovine serum albumin for 1 h at room temperature.Then,samples were incubated with anti-eIF4A3 antibody（1:500,ab 180573）for 4℃ overnight,followed by incubation with anti-rabbit IgG Alexa Fluor 594-conjugated antibody for 1 h,and cells nuclei were visualized with DAPI.All fluorescence images were acquired on an Olympus confocal microscope.14.Flow cytometryPhosphotidylserine（PS）exposure of ox-LDL treated cells were analyzed using an Annexin V PE/7-AAD Apoptosis Detection Kit according to the protocol.Unstained control cells,cells stained with PE Annexin V only,and cells stained with 7-AAD only were used to set up compensation and to define quadrants.Apoptotic cells were examined within 1 h,and the percentage of PS positive cells（right quadrant）was measured using FlowJo software.15.Actinomycin D treatmentTo block transcription,cell culture medium was added with 10 ug/ml Actinomycin D in 0h,3h,6h,9h and 12h.After treatment with Actinomycin D for different time points,the remaining of mRNA was assessed using qRT-PCR.16.Statistical analysisData are presented as mean±SD.Statistical analyses were performed using SPSS 25.0.Continuous variables were first tested by Shapiro-Wilk tests to verify whether the data sets were normally distributed.If normally distributed,two-group comparisons were analysed using two-tailed Student’s t-test.If not,Mann-Whitney U tests was performed.One-way or two-way ANOVA were used for multiple comparisons.P<0.05 was considered statistically significant.Results1.circ0030042 information and human-mouse homology comparison in circBaseThe circBase database showed that hsacirc0030042 was 1352nt in length and generated from the second exon of gene FOXO1.In HUVEC cells,circ0030042 was highly sequenced,while mouse FOXO1 derived mmucirc0010680 was had 89%high homology with circ0030042.2.The divergent primer was verified and the distribution of circ0030042 in HUVECsThe 3-free RNA（RT to cDNA）and gDNA of HUVECs were applied to agarose gel electrophoresis with circ0030042 and GAPDH divergent primer and convergent primer,respectively.The results showed that the divergent primer were specific and could be used for subsequent detection.In human aortic smooth muscle cells（HASMCs）,human monocytes（THP-1）and HUVECs,circ0030042 and FOXO1 were all abundant in HUVECs.The results of northern blot and FISH indicated that circ0030042 existed and mainly distributed in cytoplasm of HUVECs.3.circ0030042 combines with eIF4A3 in HUVECsRIP indicated that hsacirc0030042 could function as Ago2 and eIF4A3 sponge,and eIF4A3 was specifically enriched.Using the catRAPID and Vienna RNA algorithm for circRNA secondary structure and RNA-protein interaction,hsacirc0030042eIF4A3 binding was predicted to be mediated through three major RNA regions（180227,862-867,and 1082-1137 nucleotides）,and through the five domains（61-112,111-162,176-227,251-302,and 336-387 aa）of eIF4A3 protein,respectively.These five domains in eIF4A3 are highly homologous in humans and mice.4.ox-LDL could down-regulate the expression of circ0030042 in HUVECsHUVECs were treated with 100ug/ml ox-LDL at different times,and the expression level of circ0030042 was significantly decreased from 3h.5.Detection of stable circ0030042 overexpression efficiency in HUVECsWith puromycin selection,the fluorescent expression rate of HUVECs in the control group was over 95%.Sequencing of the PCR product proved that the cyclized junction site of the overexpressed circ0030042 was accurate,and the qRT-PCR result showed that the expression of circ0030042 was more than 40 times than that of the control group.6.circ0030042 inhibits ox-LDL induced abnormal autophagy in HUVECsWestern blot showed that protein levels of FOXO1,beclin1,and LC3B were elevated,with no quantitative change in eIF4A3 level.After transfection with stubRFPsensGFP-LC3 adenoviruses for 72 h,cells were exposed to 100 μg/ml ox-LDL for 0,3,and 24 h.Confocal microscopy showed that ox-LDL upregulated autophagosome puncta over time.The autophagosome puncta of c0030042 group were notable less than those of HUVEC（control）group.Flow cytometry results showed that circ0030042 overexpression led to a decreased proportion of Annexin V-positive cells after treatment with ox-LDL for 24 h.Protein levels of endogenous FOXO1,beclinl,and LC3B in oxLDL-induced c0030042 were lower than those in the circ-N.C group.Endogenous levels of FOXO1,beclin1,and LC3B in circ0030042 siRNA-transfected HUVECs（circ-siR）were higher than those of the negative control（N.C）.eIF4A3 showed no quantitative change.7.circ0030042 inhibits abnormal autophagy through obstructing the recruitment of eIF4A3 to beclin1 and FOXO1 mRNA.Western blot showed that with eIF4A3 depression,FOXO1,beclin1,and LC3B were significantly decreased in both the c0030042 and circ-N.C groups.Compared with circ-N.C,c0030042 group showed decreased formation of autophagosomes induced by ox-LDL and eIF4A3 depression could further decrease the autophagic vacuoles in both c0030042 and circ-N.C group.The c0030042 group transfected with eIF4A3 overexpression lentivirus（c0030042+h-eIF4A3-GFP）showed increased FOXO1 and beclin1 expression compared with the c0030042 group transfected with empty GFP lentivirus（c0030042+GFP-N.C）,similar to the trend in the circ-N.C group.Furthermore,using RIP on c0030042 and circ-N.C groups,we found that eIF4A3 and hsacirc0030042 were obviously upregulated in c0030042.Meanwhile,the enriched FOXO1 and beclinl mRNA on eIF4A3 in the c0030042 group was much less than that in the circ-N.C group,while,interestingly,apoptosis-related Bax and Bcl2 mRNA showed no significant change.Moreover,immunofluorescence and western blot of circN.C and c0030042 group showed increased eIF4A3 cytoplasmic localization in c0030042 group.To verify that FOXO1 and beclin1 suppression was caused by hsacirc0030042 competitive co-binding with eIF4A3,we constructed three-bindingsite deletion hsacirc0030042 overexpression plasmids（circ-delete）and transfected them into 293A cells.RIP with eIF4A3 antibody showed that eIF4A3-bound hsacirc0030042 was obviously decreased in circ-delete compared with that in the normal hsacirc0030042 overexpression group（c0030042）.In HUVECs,the circdelete group presented no significant change in FOXO1,beclin1,and LC3B protein expression compared with levels in the empty vector-transfected group（circ-N.C）.Since eIF4A3 activity mainly reflected by its target mRNA,to investigate hsacirc0030042 function on eIF4A3,we used Actinomycin D to block transcription for different time points and the remaining beclin1 and FOXO1 mRNA was assessed using qRT-PCR.As expected,eIF4A3-siRNA led to the decreased mRNA stability of beclin1 and FOXO1.And hsacirc0030042 also decreased beclin1 and FOXO1 mRNA stability and the rebound of beclin1 and FOXO1 mRNA stability occurred with upregulation of eIF4A3.Conclusions1.circ0030042 was inhibited by the stimulation of ox-LDL in HUVECs;2.circ0030042 suppresses ox-LDL induced abnormal autophagy in HUVECs through inhibiting the expression of beclin1 and FOXO1;3.circ0030042 inhibits beclin1 and FOXO1 expression through obstructing the recruitment of eIF4A3 to beclin1 and FOXO1 mRNA.BackgroundAtherosclerosis（AS）is one of the most important arteriosclerotic vascular disease.AS is characterized by the progressive formation of arterial wall fibro-adipose lesions,causing myocardial infarction,stroke and peripheral arterial disease,which have high morbidity and high mortality worldwide.The risk factors for the development of AS include high fat,high cholesterol,high blood pressure,smoking,diabetes and immune inflammation.To study the cellular and molecular mechanisms under AS provide an important view for further understanding the relationship between all these risk factors and the occurrence and clinical manifestations of AS.The pathogenesis of AS can be divided into three stages:initiation,progression and complications.Under physiological conditions,vascular endothelial cells play a role in protecting blood vessels through barrier function,secreting cytokines,regulating vasoconstriction and relaxation,inhibiting platelet adhesion and so on.In the initial and progressive stage of AS lesions,oxidisized low-density lipoprotein（ox-LDL）aggregates under the intima and induce inflammation and immune response,causing endothelial cells dysfunction.On the one hand,monocytes adhere to the intima and transform into macrophages that are more intend to phagocytosis of lipid granules.Macrophages can swallow lipoprotein particles with scavenger receptors,and gradually deposit under the intima to form lipid nuclei.On the other hand,the proinflammatory factors secreted by endothelial cells act on the vascular smooth muscle cells,leading to the proliferation and migration of smooth muscle cells to the plaque area,with the phenotype changing from contraction type to secretion type,further promoting the secretion of inflammatory factors and inducing the progression of AS.In the complications stage of AS,the apoptosis of endothelial cell lead to the release of tissue factors,promote platelet adhesion and thrombosis,and cause acute coronary syndrome and stroke.Therefore,the regulation of endothelial cell function plays an important role in stabilizing plaque and reducing acute cardiovascular events.There has been no report on the circRNA regulating AS in vivo.In previous,we found that the expression of circ0030042 in the control group was significantly higher than that in the patients with CHD.We also found that circ0030042 could absorb eIF4A3,inhibite the abnormal autophagy of vascular endothelial cells caused by oxLDL,and reduce the autophagic apoptosis of endothelial cells.In this part,we will investigate whether circ0030042 could stabilize AS plaque by inhibiting autophagy and regulating endothelial cell function in AS in vivo.Objectives1.To investigate the effect of circ0030042 and eIF4A3 on the levels of plasma lipid and inflammation factors in ApoE-/-mice;2.To clarify the role of circ0030042 and eIF4A3 on the morphology and vulnerability of AS plaques in ApoE-/-mice;3.To observe the effects of circ0030042 and eIF4A3 on the autophagy related proteins beclin1 and FOXO1 in plaques,and on the endothelial function.Materials and methods1.Animal groupApoE-/-mice（n=100,8 weeks old,male）were fed with a western diet（i.e.,a highfat diet with 40%fat and 1.25%cholesterol）for 10 weeks)and randomly divided into four groups（n=25 each）:an empty lentivirus group（GFP-N.C）,a hsacirc0030042 lentivirus group（c0030042）,a human eIF4A3 lentivirus group（h-eIF4A3-GFP）,and a hsacirc0030042 and human eIF4A3 lentivirus co-transfection group（c0030042+heIF4A3）.A 200 μl suspension（2×108 TU/ml empty lentivirus or hsacirc0030042 lentivirus or human eIF4A3 lentivirus or mixed）was injected into each mouse through the tail vein.After 6 weeks of atherogenic food feeding,mice were subjected to artery plaque measurement.2.Preparation of blood samples and tissue frozen sectionWhen mouse to be euthanized was fully anaesthetized,the blood was collected,TG、TC、LDL-C、HDL-C and glucose was detected.Following perfusion with 0.9%saline,the heart,carotid artery and full length of the aorta were conserved in 4%formaldehyde or storag... |
PDF Full Text Request