The Role And Predictive Value Of MicroRNA-149 In The Development Of Pulmonary Hypertension Due To Left Heart Diseases

Backgrounds:Pulmonary hypertension(PH)is the main cause of right heart failure and sudden cardiac death.According to the etiology and pathogenesis,PH is divided into five categories.Pulmonary hypertension due to left heart diseases(PH-LHD)belongs to the second category,which is the most common PH type.Increased left ventricular load,which is caused by left heart diseases(LHD)including heart failure and valvular heart disease,enhances the pulmonary venous pressure,subsequently leading to pulmonary vascular remodeling and PH.However,more evidence showed that PH and pulmonary vascular remodeling would continue to development after the initiation of PH,even if the cause which increased the left ventricular load was cured,suggesting that multiple mechanisms might contribute to the maintaining of PH-LHD.Patients with PH-LHD usually have poor prognosis.Moreover,because of the non-specific clinical symptoms including shortness of breath,fatigue,decreased activity and so on,the early detection of PH-LHD is still difficult up to date.PH is a concept of hemodynamics.Right heart catheter(RHC)is the golden standard for the diagnosis of PH-LHD.However,RHC is not a routine examination for patients with LHD in current clinical practice,which hindered the diagnosis of PH.Echocardiography is currently popularly used in detecting cardiac remodeling and function in patients with LHD.Echocardiography can be used to estimate the pulmonary artery systolic pressure(PASP).But the accuracy is uncertain especially for those with mild elevation of PH.At the same time,the treatment strategy is limited for PH-LHD.Target strategies for IPAH including Prostacyclin,inhaled NO and others could temporarily reduce the pulmonary artery pressure,but cannot reverse its progress or improve its long-term prognosis.Therefore,it is important to study the mechanisms of PH-LHD.Biomarkers are potentially hopeful in the early detection of PH.It was reported that biomarkers such as IL-6,bilirubin,uric acid,etc.are correlated with the prognosis of PH.However,these bio markers are nonspecific in the early diagnosis.At present,brain natriuretic peptide(BNP)and BNP precursor are commonly used biomarkers.However,they are easily affected by the patient’s renal function.Circulating microRNAs,which exist in body fluids,have become a new research focus.The level of circulating microRNAs changes with different pathological conditions.It can be used as a biomarker in the diagnosis including cancer,cardiovascular disease and diabetes.However,it remains largely unknown regarding the changes of microRNAs in the study of PH-LHD.As an evolutionarily conserved endogenous non-coding RNA,microRNAs plays a role in regulating the expression of target genes at post-transcriptional levels.Circulating microRNAs can not only be used as biomarkers associated with disease,but also transmit signals between cells and participate in the development of PH.For example,activation of mi R-143 regulates the crosstalk between smooth muscle and endothelial cell,and plays a key role in the development of hypoxia-induced PH.Specific microRNAs,such as miR-21,mi R-27 a,miR-17-92 cluster,miR-124 cluster,also play important roles in the development of PH by regulating the pulmonary artery endothelial cell(PAEC)and pulmonary artery smooth muscle cell(PASMC).However,there is less research about the role of microRNAs in the development of PH-LHD.As a sensitive and high-throughput assay,microRNA microarrays can quickly screen for changes in all known microRNA expression in different samples.Bioinformatics is often utilized for analyzing screened microRNAs,which can be used for data mining,functional annotation and pathway analysis of genes related to disease.Animal models with left heart overload is commonly used for the study of PH-LHD.The target genes associated with PH-LHD can be screened in the lung tissue of the model.Further bioinformatics analysis is used to obtain the key microRNAs and the key pathways,so as to provide a theoretical basis for PH-LHD diagnosis and treatment.Objectives:To study the effects of left ventricular overload on pulmonary hemodynamics and pulmonary vascular remodeling,the rat models with transverse aortic constriction(TAC)were established.With the TAC rat model,microRNA microarray was used to screen the differential expression of microRNAs.The target genes and signal pathways were then predicted by the miR-Path tool.Real-time quantitative PCR was further used to validate the changes of microRNAs in lung tissue samples and serum of rat TAC models,and the correlations between the changes of circulating microRNAs and hemodynamics was analyzed.Finally,the correlations of circulating level of screened microRNA with PH was studied in LHD patients.Methods:Part Ⅰ:1.Animal models:Forty healthy male SD rats were randomly divided into Sham group(n=20)and TAC group(n=20).The aorta arch was constricted with the 16 G needle in the TAC group.Sham group does not ligate the aortic arch.2.Verification of the models:1)the general data;2)the changes of cardiac structure and function examined by cardiac ultrasonography;3)the hemodynamic changes including pulmonary circulation,right ventricular system and left ventricular system were measured by RHC;4)pathological changes.Part Ⅱ:1.MicroRNA microarray was used to detect differential expression of microRNAs in TAC and Sham rat groups.The target genes of the differentially expressed microRNA were predicted by mi R-Path v.3,then enrichment of target genes was obtained to predict PH-LHD related signal pathway.2.Real-time quantitative PCR was used to detect differentially expressed microRNA level in lung samples by using SYBR Green fluorescent dye method.The correlation between the micro RNA level and the hemodynamic parameters was also analyzedPart Ⅲ:1.A total of 80 cases of LHD patients were enrolled according to the criteria,who were admitted to the Second Affiliated Hospital of the Third Military Medical University from January,2014 to November,2016.According to 2009 ESC/ERS guidelines,PASP≥50mmHg was set as the basis for diagnosis of PH.The patients with LHD were divided into 2 groups: PH-LHD group,n=34 and LHD group,n=46.The clinical data were collected,including age,sex,height,weight,smoking history,blood pressure,heart rate,NYHA cardiac function classification,data of cardiac ultrasound and BNP.Levels of microRNA in the serum of all subjects were measured by real-time quantitative PCR,and their correlation with clinical data was analyzed.2.Statistical analysis was performed by using SPSS 20.0 software.The measurement data are expressed as the median and the quartile,and the count data are expressed as the number and percentage;Mann-Whitney U nonparametric test was used to analyze the differences between the two groups.Using Spearman correlation coefficient to evaluate the correlation between miR-149 level and clinical parameters.Logistic regression was used to analyze single factor and multivariate analysis,and the ratio was 95% confidence interval.Receiver operating characteristic(ROC)was used to evaluate the area under the curve(AUC),the sensitivity,specificity of related indicators in predicting PH;thus determining the best combination of biomarkers in PH-LHD prediction.Result:Part Ⅰ:1.Three rats in the TAC group,and 9 rats in the Sham group died due to anesthesia,pneumothorax and aortic rupture.During the 8 weeks rearing process,the weight of the TAC group was decreased compared with that in the Sham group.The mental state of the rats in the TAC group was poor,and 7 rats died of breeding;10 TAC models and 11 Sham models were constructed finally.Further weighing measurements showed that the right ventricular hypertrophy index(RVHI)increased(23.46±0.77% vs 27.40±2.68%;P=0.010)in the TAC group compared with the Sham group.2.At the 8th week,cardiac ultrasound was examined and the results are as follows: Compared with the Sham group(n=11),the diameter of pulmonary artery in the TAC group(n=10)was increased: 2.61(2.50,3.31)mm vs 2.97(2.75,3.25)mm(p<0.05);the left ventricular ejection fraction was decreased :84.99(79.53,88.46)% vs 69.75(56.57,82.61)%(p<0.05);the left ventricular mass was increased: 573.27(465.68,677.54)mg vs 680.77(588.27,836.55)mg(p<0.05);the diastolic left ventricular volume was increased: 160.40(125.04,187.51)ul vs 201.86(162.47,217.21)ul(p<0.05).3.At the 8th week,the RHC was examined and the results are as follows: Compared with the Sham group(n=11),the mean pulmonary arterial pressure(mPAP)in the TAC group(n=10)was increased(16.22±2.23 mmHg vs 22.13±2.74mmHg;p <0.001),90% of them was between 19-24 mmHg,which belonged to the borderline pulmonary hypertension(BPH)and 10% of them reached 25 mmHg,meaning the formation of PH.Moreover,the left ventricular systolic pressure(LVSP)(140.04±28.01 mmHg vs 172.22±29.17mmHg;p<0.001)and the right ventricular systolic pressure(RVSP)(26.06±2.74 mmHg vs 29.12±2.08mmHg;p=0.031)were increased;the LV max(dp/dt)(12278.66±13009.28mmHg/s vs 9124.76±838.36 mmHg/s;p<0.001)and the RV max(dp/dt)(1892.47±272.69 mmHg/s vs 1491.78±261.20 mmHg/s;p=0.003)were decreased.4.At the 8th week,the pulmonary arteries and lung tissues were examined by Hematoxylin and eosin staining.Compared with the sham group,the percentage of pulmonary arterial membrane thickness in the TAC group was increased(27.41±2.38 vs 54.95±4.81,P=0.027).The pulmonary arterial wall in the Sham group was observed thinner,lumen smooth,no stenosis,TAC group pulmonary arterial wall was thickened,and some pulmonary arteries were obstructed,surrounding by inflammatory cell.Immunofluorescence staining showed that the pulmonary vascular wall of the TAC group was thickened and the lumen became narrowed.Part Ⅱ:1.A total of 385 microRNAs were analyzed by micro RNA microarray in the lung samples from the Sham group and the TAC group.The changes greater than 1.5 times with statistically difference were screened.The individual difference was excluded by gene cluster analysis.In the TAC group,the rno-miR-143-3p,rno-miR-181a-5p,rno-miR-125b-1-3p increased,while rno-miR-206-3p,rno-let-7b-5p,rno-mi R-200b-3p,rno-miR-21-5p,rno-let-7b-5p,rno-miR-149-3p were decreased.The target genes of differentially expressed microRNA were revealed by further KEGG signaling pathways analysis.These genes included Tnf,Chrd,Smad2,Smad7,Acvr1 c,Acvr2a,Rps6kb1,Smurf2,Bmp5,and Tgfbr1,which focused on the TGF-β signaling pathway.2.The change of mi R-149 level was further verified in rat lung tissue.The level of mi R-149 in the TAC group was down-regulated(1.196±0.227 vs 0.338±0.028;p=0.002).The levels of mi R-149 in the serum of TAC rats were measured and the changes were consistent with those in the lung tissue with the RQ values were(1.084±0.36 vs 0.355±0.036;p <0.001).mi R-149 level was negatively correlated with mPAP(r=-0.752,p<0.001)and RVHI(r=-0.643,p=0.002),but positively correlated with LVSP(r=0.739,p<0.001),RV Max(dp/dt)(r=-0.752,p<0.001)and LV Max(dp/dt)(r=0.728,p<0.001),.There was no correlation between miR-149 and RVSP(r=-0.309,p=0.173).Part Ⅲ:1.The circulating mi R-149 level in clinical cases was measured.The results showed that the RQ values of the LHD control group and the PH-LHD group were 1.00(0.73,1.81)vs 0.58(0.49,1.08)respectively.The expression of mi R-149 in PH-LHD group was significantly lower than that in LHD group(P=0.002).The correlation between the level of miR-149 and the clinical data was analyzed,and it was found that the level of miR-149 was negatively correlated with PASP(R=-0.36,p=0.002),left ventricle end diastolic diameter(LVDD)(r=-0.345,p=0.002),BNP(r=-0.306,p=0.006)and cardiac function,and positively correlated with LVEF(r=0.247,p=0.027).2.Logistic regression revealed that the increase in circulating mi R-149 would elevate the occurrence probability of PASP≥50 mmHg.The results of further receiver operating characteristic(ROC)analysis show that,circulating miR-149 could be a predictive indicator for PASP≥50,the area under curve(AUC)was 0.699(95% CI: 0.579 to 0.820,P=0.002)with the sensitivity at 0.500,specificity at 0.891 when the cut-off value was 0.52.With the combination of circulating miR-149,left atrial end diastolic diameter(LADD)and BNP,selected by forward stepwise binary logistic regression analyses,the predictive value for PASP ≥ 50 mmHg under ROC curve was 0.889(95% CI: 0.818 ~ 0.960,P<0.001)with the sensitivity at 0.882,specificity at 0.818 when the cut-off value was 0.41.Conclusion:1.TAC model,which caused the increase of left ventricular load and the impairment of left ventricular function,results in the mild elevation of mPAP and the impairment of right ventricular function in rats models;2.The expression of microRNAs in lung tissues changed with the increase of left ventricular load and impaired left ventricular function.The data showed that microRNAs including rno-miR-143-3p,rno-miR-181a-5p,and rno-mi R-125b-1-3p increased,while microRNAs including rno-miR-206-3p,rno-let-7b-5p,rno-miR-200b-3p,rno-miR-21-5p,rno-let-7b-5p,rno-miR-149-3p decreased.KEGG signal pathway analysis showed that the target genes of these microRNAs focused on TGF-β signaling pathways.3.The level of mi R-149 in lung tissue and serum of the TAC group was significantly decreased when compared with that in the Sham group.4.Compared with LHD patients,miR-149 levels were significantly decreased in patients with PH-LHD,and circulating miR-149 was an independent predictor of PH in patients with LHD.Combination of miR-149,LADD and BNP had an optimal predictive value for PH in patients with LHD,with a sensitivity of 0.882 and a specificity of 0.818.