| Objective: Nonalcoholic fatty liver disease(NAFLD)is one of the most common chronic and metabolic liver diseases with a more than 25.00% global prevalence rate in 2018.NAFLD is characterized by excessive triglycerides accumulation and is a general term for a series of liver diseases that affect people who consume small amounts of alcohol.It can progress to inflammatory nonalcoholic steatohepatitis,fibrosis,cirrhosis,and even liver cancer.In addition,NAFLD is also closely related to metabolic syndrome,type 2 diabetes,insulin resistance,obesity,cardiovascular disease and other chronic metabolic diseases.Recently,environmental metals pollution became more and more serious.Therefore,its harmful effects on human health have become the focus of environmental and health research.Exposure to multiple toxic metals and excess or deficiency of essential trace elements can cause health damage.Epidemiological studies suggested that low levels of copper,iron,selenium,magnesium,zinc and high levels of arsenic,calcium,cadmium and lead may be independent risk factors of NAFLD.However,previous studies have focused on the health damaging effects of single metal,and environmental metals are usually in the form of co-exposure,particularly the existence of interactions between metals.Thus,it is necessary and important to clarify the damage effects of multi-metal co-exposure on NAFLD.Additionally,researchers considered that excessive oxidative stress and inflammasome activation are closely related to the occurrence and development of NAFLD.It was reported that the concentrations of barium,chromium,cadmium,molybdenum,vanadium and zinc were positively correlated with oxidative stress.As part of innate immune,complement system plays an important role in the process of anti-oxidative stress and anti-inflammatory reaction.Previous studies found that complement C3 and C4 were activated in liver biopsies from NAFLD patients,C3 promotes the accumulation of triglycerides in the liver and paralleled the degree of liver injury.However,to date,epidemiological studies focusing on the role of multiple metals co-exposure and complement system in the pathogenesis of NAFLD,are still scarce in the current literature.Aimed to provide clues for the identifying independent risk factors and searching for predictive biomarkers of NAFLD risk,and to provide evidence for further clarifying the mechanism of NAFLD development and identifying high-risk individuals of NAFLD,we use pollutants co-exposure model and Mendelian randomization(MR)analysis to explore the causal relationships of serum 22 metals,C3 and C4 with NAFLD risk in the Fangchenggang Area Male Health and Examination Survey(FAMHES)cohort.Methods:In the first part of this study,(1)Using cross-sectional and longitudinal analysis based on hospital physical examination,the individuals were recruited from the baseline survey in 2009 and follow-up stage in 2013 of the FAMHES cohort.A total of 1,594 participants(330 with NAFLD,1,264 without NAFLD)were enrolled in the final cross-sectional analysis,and a total of 566 men(144 with incident NAFLD,422 without NAFLD)participated in the follow-up exam.(2)Several types of data were collected in this study including questionnaire data obtained from in-person interviews,clinical data from physical examination and laboratory tests.In-person interviews were conducted by trained interviewers using a standardized and structured questionnaire that included questions pertaining to prior use of alcohol and tobacco,medical history,family history of hepatic disease,as well as other demographic,socioeconomic,and lifestyle factors.Abdominal ultrasound was used to diagnose liver steatosis.Inductively coupled plasma mass spectrometer was used to measure 22 serum metals concentrations.(3)The binary logistic regression models were used to calculate odds ratios(ORs)and 95% confidence intervals(CIs)at baseline survey stage,and the cox regression models were used to estimate hazard ratios(HRs)and 95% CI at follow-up stage.In the first part of this study,we used two methods of Mann-Whitney U test and least absolute shrinkage and selection operator(LASSO)regression model to construct multiple-metals model.We used metals score to estimate the joint effects of co-exposure metals.Restricted cubic splines were used to evaluate overall associations between serum metals and NAFLD risk and to determine whether there were nonlinear exposure-response relationships.Stratified analyses were performed by selected metals and corresponding dichotomous variables for baseline characteristics,including age(? 40 or > 40 years),marital status(yes or no),smoking(yes or no),drinking(yes or no),Met S(yes or no)and IR(yes or no).In the second part of this study,(1)Using the same design and research objects as in the first part,a total of 1,600 participants(328 with NAFLD,1,272 without NAFLD)were enrolled in the final analysis,and a total of 572 men(146 with incident NAFLD,426 without NAFLD)participated in the follow-up exam.(2)Serum C3 and C4 levels were measured with a Hitachi 7600 autoanalyzer.(3)We performed a bidirectional MR analysis by using two C3 related SNPs(rs3753394 and rs3745567),eight C4 related SNPs(rs1052693,rs11575839,rs2075799,rs2857009,rs2071278,rs3763317,rs9276606,and rs241428)and seven NAFLD related SNPs(rs12137855,rs780094,rs4240624,rs1227756,rs738491,rs738409 and rs5764455)to construct genetic risk score as instrumental variable.Two-stage least squares regression(2SLS)was used to calculate the formal MR analyses to estimate the possible causal effect of C3 and C4 on NAFLD(and vice versa),respectively.Results:In the first part of this study,we found that:(1)In cross-sectional analysis,age,married,education level,metabolic syndrome and insulin resistance were significantly higher in the NAFLD group,compared to the individuals without NAFLD;the distribution of serum barium,cadmium,calcium,iron,lead,magnesium,manganese,molybdenum,nickel,tin,vanadium and zinc were statistical significant between the two groups(all P < 0.05).In longitudinal analysis,education level,metabolic syndrome and insulin resistance were significantly higher in the incident NAFLD group,compared to the individuals without NAFLD;the distribution of serum barium,lead,manganese,molybdenum,vanadium and zinc were statistical significant between the two groups(all P < 0.05).(2)Spearman rank correlation analysis showed that serum 21 metals were correlated with each other.(3)The Mann-Whitney U test selected six metals(barium,lead,manganese,molybdenum,vanadium and zinc)to construct multiple-metals model,while LASSO regression method selected twelve metals(barium,cobalt,copper,iron,manganese,molybdenum,rubidium,selenium,stannum,titanium,vanadium and zinc).(4)After adjustment in six-metal model,only depressed molybdenum and elevated zinc were significantly associated with higher NAFLD risk both in cross-sectional(adjusted ORMo = 0.41,95% CI: 0.26,0.65;adjusted ORZn = 3.60,95% CI: 2.24,5.79)(Quartile 4 vs.Quartile 1)and longitudinal analysis(adjusted HRMo = 0.51,95% CI: 0.31,0.84;adjusted HRZn = 2.02,95% CI: 1.16,3.53).In twelve-metal model,similarly significant results were still observed(adjusted ORMo = 0.50,95% CI: 0.31,0.81;adjusted ORZn = 4.46,95% CI: 2.69,7.40;adjusted HRMo = 0.47,95% CI: 0.27,0.80;adjusted HRZn = 2.15,95% CI: 1.19,3.87).(5)The joint effects of molybdenum and zinc also increased NAFLD risk.Per SD increased metals score was significantly associated with higher risk of NAFLD both in the cross-sectional analysis(crude OR = 1.64,95% CI: 1.46,1.85;adjusted OR = 1.50,95% CI: 1.30,1.72)and longitudinal analysis(crude HR = 1.28,95% CI: 1.10,1.48;adjusted HR = 1.17,95% CI: 1.00,1.36).(6)Dose-response relationships were U-shape of molybdenum(Poverall = 0.023,Pnonlinearity = 0.043)and positive linear of zinc(Poverall < 0.001,Pnonlinearity = 0.467)with NAFLD risk.(7)Stratification analysis showed that serum molybdenum and zinc concentrations were not consistent with NAFLD risk according to strata of age,marital status,drinking,smoking,metabolic syndrome and insulin resistance.In the second part of this study,we found that:(1)In cross-sectional analysis,age,married,education level,metabolic syndrome and insulin resistance were significantly higher in the NAFLD group,compared to the individuals without NAFLD;the median levels of serum C3 and C4 in the NAFLD group(1.29 g/L and 0.35 g/L,respectively)were higher than those in the non-NAFLD group(1.08 g/L and 0.32 g/L,respectively),and the distribution of serum C3 and C4 were statistical significant between the two groups(all P < 0.05).In longitudinal analysis,education level,metabolic syndrome and insulin resistance were significantly higher in the incident NAFLD group,compared to the individuals without NAFLD;the median levels of serum C3 and C4 in the incident NAFLD group(1.18 g/L and 0.34 g/L,respectively)were higher than those in the non-NAFLD group(1.06 g/L and 0.32 g/L,respectively),and the distribution of serum C3 and C4 were statistical significant between the two groups(all P < 0.05).(2)After adjustment,per SD increased C3 levels were significantly associated with higher risk of NAFLD(adjusted OR = 1.65,95% CI: 1.40,1.94)in the cross-sectional analysis but not with C4(adjusted OR = 1.04,95% CI: 0.89,1.21).Similar results were observed in the longitudinal analysis(adjusted HRC3 = 1.20,95% CI: 1.02,1.42;adjusted HRC4 = 1.10,95% CI: 0.94,1.28).(3)In MR analysis,no causal relationships were found for genetically determined C3 levels and NAFLD risk by using unweighted or weighted GRS_C3(both adjusted OR = 1.07,95% CI: 0.98,1.18).Conversely,the results of genetically determined NAFLD for C3 also were not significant(adjusted ?unweighted: 0.15,95% CI:-0.31,0.60;adjusted ?weighted: 0.15,95% CI:-0.36,0.66,respectively).Both directions showed no significant associations from C4 to NAFLD and from NAFLD to C4.Conclusions: Our results support that:(1)In serum 21 metals,serum molybdenum levels were inversely and U-shape associated with NAFLD risk in Chinese men using multiple-exposure approach,whereas serum zinc levels were positively linear association,and the joint effects of them also associated with increasing NAFLD risk,indicating serum low molybdenum and high zinc concentrations may be the independent risk factors of NAFLD.(2)No significant causal relationships between serum C3 and C4 levels and NAFLD risk from both directions by using MR approach,although conventional means indicated positive results,which indicated that the complement system may not play a role of NAFLD development.The results need to be validated in larger cohorts and other ethnicities or ethnic groups. |
PDF Full Text Request