Risk Prediction Study Of Urinary Cadmium Levels And M~6A Modifying Enzymes On Elevated Urinary N-acetyl-β-D-glucosidase

Objectives:1.To investigate the association between urinary cadmium levels and urinary NAG levels.2.To analyze the relationship between urinary cadmium levels,m~6A-modifying enzyme mRNA levels,and urinary NAG levels.3.To reveal the role of urinary cadmium levels and m~6A modifying enzyme mRNA levels in predicting the risk of elevated urinary NAG levels.Methods:1.Randomly selected residents of two communities in a city in North China as study subjects,and according to common clinical standards,subjects with urinary NAG≥10.7U/g Cr were selected as the case group(higher UNAG group)and the rest as the control group through a case-control study.A face-to-face questionnaire was used to collect basic information about the study participants and their health status,and collecting fasting peripheral venous blood and mid-morning urine from the study participants.2.Inductively coupled plasma mass spectrometry(ICP-MS)was used to measure urinary cadmium level;using sarcosine oxidase to measure urinary creatinine(UCr);urinary N-acetyl-β-D-glucosidase(UNAG)was measured by p-nitrophenol;RNA was extracted from peripheral blood leukocytes and reverse transcribed to c DNA.And then using Real-time quantitative PCR detected the expression levels of m~6A modifying enzymes:methyltransferase METTL3,METTL14,de-methyltransferase FTO and ALKBH5.3.Using Epi Data 3.1 to build the database,and analyzing the data using nonparametric test(Mann-Whitney U test,Kruskal-Wallis test),χ~2test,and correlation test in SPSS 26.0software,and risk prediction analysis was performed using R language.Constructing a multivariate logistic regression model to select the independent risk/protective factors influencing the elevated UNAG level,and building a Nomogram risk prediction model.Consistent internal validation of the constructed model was performed using bootstrap 100times repeated sampling method,calibration curve.Using receiver operating characteristic curves(ROC)as well as decision curve analysis(DCA)to assess the value of the model.Results:1.This study had 514 participants,243 in the higher UNAG group,271 in the control group.The average age of the higher UNAG group was 69.16±7.81 years old,compared to64.88±10.17 years in the control group.Age and smoking were statistically different between the two groups(P<0.05).In the higher UNAG group the urinary cadmium was1.53μg/g Cr and 1.11μg/g Cr for the control group,in the higher UNAG group the urinary cadmium was higher(Z=-2.484,P<0.05).2.The relationship between urinary cadmium levels and UNAG levels was analyzed.The results showed that in the hype-UNAG group,the urinary cadmium levels of 60-70 years old,male,BMI<24Kg/m~2,high school or specialist education,household per capita monthly income of 3000-5000 yuan,and smoking people were higher than those of the control group(all P<0.05).In the total population,UNAG level increased with the increase of urinary cadmium level(P=0.012).Further correlation analysis showed that the level of urinary cadmium was positively correlated with age(r=0.37),gender(r=0.65),UNAG(r=0.42),and negatively correlated with BMI(r=-0.70),education level(r=-0.30),and family monthly income per capita(r=-0.52).3.Analysis of the relationship between mRNA levels of m~6A modifying enzymes and UNAG levels showed that the expression level of de-methyltransferase FTO was downregulated in the higher UNAG group(P<0.05),while there were no statistically significant differences between the two groups in other(de)methyltransferases.In the higher UNAG group,methyltransferase METTL3 was significantly different among different age groups(P=0.028),and the expression levels of methyltransferase METTL14and de-methyltransferase ALKBH5 increased with increasing age(P<0.05),the expression level of de-methyltransferase ALKBH5 had a statistically significant difference in the group with or without alcohol consumption(P=0.016).In the control group,the expression level of de-methyltransferase FTO increased with increasing age(P=0.017).4.The relationship between urinary cadmium levels and mRNA levels of m~6A modifying enzymes was analyzed by grouping urinary cadmium levels into three quantiles.It was found that the difference of methyltransferase METTL14 in the higher UNAG group was statistically significant among different urinary cadmium levels(P=0.045),while the expression level of de-methyltransferase FTO increased with the increase of urinary cadmium level(P=0.022).In the control group,there were statistically significant differences in the expression levels of methyltransferase METTL14,de-methyltransferase FTO and ALKBH5 among different levels of urinary cadmium(P<0.05),and ALKBH5expression level increased with increasing urinary cadmium level(P=0.001).5.The correlation analysis between urinary cadmium levels,m~6A modifying enzymes mRNA levels and UNAG showed that methyltransferase METTL14 was positively correlated with METTL3,urinary cadmium levels and UNAG levels;methyltransferase METTL3 was positively correlated with urinary cadmium levels and ALKBH5;de-methyltransferase FTO was negatively correlated with METTL14,METTL3,ALKBH5,and UNAG levels.6.Application of a Nomogram risk prediction model to analyze the risk prediction effect of urinary cadmium level and m~6A modifying enzymes on elevated UNAG level,the results showed that the risk of elevated UNAG level was higher when urinary cadmium level was higher,FTO expression level was lower,and age was higher.Further internal validation with 100 repetitions using bootstrap method and the model fit was good by the Hosmer-Lemeshow test with P=0.472>0.10.And calibration curve showed that the prediction rate of Nomogram model was close to the actual incidence rate of Y=X,indicating that the Nomogram model constructed in this study had high accuracy and good internal validation results.7.ROC as well as DCA were used to analyze the potential value of Nomogram,and the results displayed that the area under the curve of ROC was 0.687(95%CI:0.637,0.728),which was greater than 0.5,indicating that the accuracy of the Nomogram model evaluated by ROC was good.And in this study,with 68.7%certainty,the risk of elevated UNAG level was higher when urinary cadmium level was higher,FTO expression level was lower,and age was higher.In addition,through the DCA curve,it was found that the Nomogram model predicted the highest net benefit of the model curve when the threshold range of occurrence of elevated UNAG level was between 0.21 and 0.85,i.e.,the applicability of the model was better when urinary cadmium level,de-methyltransferase FTO,and age were used as predictor variables for elevated UNAG,and the value of clinical application was higher.Conclusions:1.Urinary cadmium level was higher in the higher UNAG group,and UNAG level increased with higher urinary cadmium level.2.The de-methyltransferase FTO mRNA expression level was down-regulated in the higher UNAG group,which was negatively correlated with the UNAG level.3.Urinary cadmium level,de-methyltransferase FTO and age could predict the risk of elevated UNAG level,and the prediction model constructed in this study was accurate and had high clinical application value;the risk of elevated UNAG level was higher when urinary cadmium level was higher,de-methyltransferase FTO mRNA expression level was lower and age was older.