| ObjectiveThe first aim of this study was to examine the correlation between long-chain non-coding RNA H19(lnc RNA H19)and Zhuang nationality schizophrenia,and to construct an lnc RNA H19-micro RNA(mi RNA)-tyrosine metabolism pathway(TMP)regulatory network which associated with schizophrenia.The second aim was to examine the correlation between Zhuang nationality schizophrenia and candidate mi RNAs in the regulatory network.At the same time,the diagnostic value of these candidate mi RNAs were evaluated.The third aim was to examine the effect of lnc RNA H19 on mi RNA,target gene,protein and cell functions in the axis to clarify the mechanism of lnc RNA H19 regulating on TMP by targeting mi RNAs and thus affecting the occurrence of Zhuang nationality schizophrenia.MethodsPart ?: Correlation between lnc RNA H19 and schizophrenia in Zhuang nationality population was studied,and lnc RNA H19-mi RNAs-TMP regulatory network was constructed.First,a case control study was conducted in Zhuang nationality population in Guangxi,China.The lnc RNA H19 expression level was detected in 50 patients with schizophrenia and 46 healthy controls using real-time quantitative polymerase chain reaction(q RT-PCR).The correlation between lnc RNA H19 and schizophrenia in Zhuang nationality population was analyzed using SPSS20.0 software.Second,the target mi RNAs of the TMP genes COMT,DBH,ADH1 B,and TH were analyzed through the mi Records database.Target mi RNAs that were predicted at least via three prediction software were selected.Finally,the bonding of the target mi RNAs and lnc RNA H19 were scored using the DIANA-Lnc Base v2 database.Scores above 0.5 were left to construct the lnc RNA H19-mi RNAs-TMP regulatory network using the Cytoscape software.The mi RNAs(micro RNA)in this regulatory network were used as candidate mi RNAs in the follow-up study.Part II: Candidate mi RNAs' s diagnostic value and association between candidate mi RNAs and schizophrenia in Zhuang nationality population were studied.First,candidate mi RNAs were detected in 50 patients with Zhuang nationality schizophrenia and 46 healthy controls using q RT-PCR.The relationship between candidate mi RNA and schizophrenia was analyzed using SPSS 20.0 software.Receiver operative characteristic curve(ROC),sensitivity,specificity,and Yoden index were analyzed simultaneously.The area under the curve(AUC)was compared using Z test.Second,data were divided into training and test samples at a ratio of 2:1.Machine learning methods,such as Decision Trees(DT),Support Vector Machines(SVM),Random Forest(RF),XGBoost,and Artificial Neural Network(ANN),were used to establish the mi RNA biomarker diagnostic model for schizophrenia.Part III: lnc RNA H19's cellular function in regulating tyrosine metabolism pathway by affecting mi RNAs was studied.The binding of lnc RNA H19 and mi RNAs and that of mi RNAs and tyrosine metabolism target genes were analyzed through the micro RNA.org database.The optimal lnc RNA H19-mi RNA-m RNA axis was selected for cell experiment verification.The small interfering RNA for lnc RNA H19(si RNA H19)and control plasma were constructed and transfected in SH-SY5 Y by the liposome method.Western blot,q RT-PCR,cell proliferation or toxicity test(CCK-8),and flow cytometry were used to detect the effects of lnc RNA H19 on mi RNA expression,downstream gene expression,and cellular function.ResultsPart ?: Correlation between long noncoding RNA?H19 and schizophrenia in Zhuang nationality population was studied,and lnc RNA H19-mi RNAs-TMP regulatory network was constructed.First,the 50 cases of Zhuang nationality schizophrenia and 46 healthy controls did not show gender difference(P>0.05).The participants in the control group were older than those in the case group and the difference was statistically significant(P<0.05).The lnc RNA H19 expression in the schizophrenia group was higher than that of the healthy control group,with statistically significant difference(t=2.374,P=0.025),thereby suggesting that lnc RNA H19 may be correlated with schizophrenia in Zhuang nationality.Second,on the basis of the analysis using mi Records,there were 160 target mi RNAs of the TMP genes that were predicted at least via three prediction software.Finally,eight mi RNAs were regulated by lnc RNA H19 and affected the tyrosine metabolism pathway on the basis of the analysis using DIANA-Lnc Base v2 database.The mi RNAs included hsa-mi R-301 a,hsa-mi R-412,hsa-mi R-625,hsa-mi R-361-3p,hsa-mi R-214,hsa-mi R-922,hsa-mi R-338-3p,and hsa-mi R-505.Cytoscape software was used to draw the lnc RNA H19-mi RNAs-TMP regulatory network on the basis of the analysis results.Part II: Candidate mi RNAs' diagnostic value and association between candidate mi RNAs and schizophrenia in Zhuang nationality were studied.First,for the mi RNA expression level,hsa-mi R-922 was not detected in the samples.The has-mir-625(Z=-4.926,P<0.001),has-mir-361-3p(Z=-4.740,P<0.001),hsa-mi R-505(Z=-4.703,P<0.001)expression levels between Zhuang schizophrenia and controls showed statistically significant difference(all P<0.05),and the expression levels of these mi RNAs were lower in case group than that in control group.The hsa-mi R-301 a,hsa-mi R-412,hsa-mi R-214,and hsa-mi R-338-3p expression levels were not statistically different between patients with schizophrenia and controls(all P>0.05).Second,the ROC analysis results for has-mir-625 included the AUC value of 0.816,95% confidence interval(CI)of [0.721,0.911],sensitivity of 0.90,specificity of 0.71,and Youden index of 0.61.The ROC analysis results for has-mir-361-3p included the AUC value of 0.804,95% CI of [0.708,0.901],sensitivity of 0.83,specificity of 0.74,and Youden index of 0.56.The ROC analysis results for hsa-mi R-505 included the AUC value of 0.802,95% CI of[0.704,0.899],sensitivity of 0.85,specificity of 0.64,and Yoden index of 0.49.Comparison of the AUC values of the three mi RNAs showed that the difference between each two pairs was not statistically significant(all P>0.05).Third,the diagnostic model results constructed by machine learning methods for the XGBoost model included the AUC value of 0.928,95% CI of[0.807,1.000],sensitivity of 0.87 specificity of 1.00,and Youden index of 0.87.The diagnostic model results for the RF model included the AUC value of 0.928,95% CI of [0.829,1.000],sensitivity of 0.93,specificity of 0.92,and Youden index of 0.85.The diagnostic model results for the DT model included the AUC value of 0.921,95% CI of [0.823 1.000],sensitivity of 0.87,specificity of 0.92,and Youden index of 0.79.The diagnostic model results for the SVM model included the AUC value of 0.954,95% CI of [0.877,1.000],sensitivity of 0.93,specificity of 0.77,and Youden index of 0.70.The diagnostic model results for the ANN model included the AUC value of 0.949,95% CI of [0.868 1.000],sensitivity of 0.86,specificity of 0.85,and Youden index of 0.71.The AUC values of the five models were compared,and no statistically significant difference was found among models(all P> 0.05).However,the AUC values of SVM model and ANN model were higher than that of a single mi RNA,and the differences were statistically significant(both P <0.05).Part III: lnc RNA H19's cellular function in regulating tyrosine metabolism pathway by affecting mi RNAs was studied.The lnc RNA H19-has-mir-505 or has-mir-625-DBH axis was selected for cellular function experiment on the basis of the results of the population study and further bioinformatics analysis.First,the q RT-PCR results showed that the has-mir-625 expression level in the si RNA H19 group was higher than that in the control group(P<0.05).The has-mir-505 and DBH m RNA expression levels were not statistically different between the two groups(P>0.05).Second,48 and 72 h after the transfer of si RNA H19,the CCK8 results showed that the cell proliferation of the si RNA H19 group was higher than that of the control group(both P<0.05).Third,Western blot results showed that the DBH protein expression level in the si RNA H19 group was higher than that in the control group,whereas flow cytometry experiments showed that si RNA H19 had no effect on cell apoptosis.The results suggest that lnc RNA H19 has an inhibitory effect on DBH protein expression,but it is not produced by competing with mi RNA for inhibition.ConclusionsLnc RNA H19 was associated with schizophrenia in Zhuang nationality population.Such an association may affect the development of this disease by affecting the DBH regulating the tyrosine metabolism pathways.The mi RNAs has-mir-625,has-mir-361-3p,and hsa-mi R-505 were associated with schizophrenia in Zhuang nationality population and are expected to become potential diagnostic biomarkers for this disease in this population.Multiple mi RNAs can be fitted by machine learning methods to build a schizophrenia diagnosis model for Zhuang nationality population.Diagnosis accuracy was improved compared with that obtained by using a single mi RNA.There is a co-expression relationship between lnc RNA H19 and has-mir-625,which is not consistent with the traditional ce RNA mechanism.The possible mechanism is that lnc RNA H19 may act as a precursor or host of has-mir-625 to affect the transcription of it,thereby causing changes in DBH protein expression,regulating tyrosine metabolism pathways,and participating in the occurrence and development of schizophrenia.It needs further research to confirm. |
PDF Full Text Request