| Inflammation is a defensive response of the body to harmful external stimuli.Local acute inflammation could be quickly recovered,while low-level chronic inflammation could persist for a long time and induce adverse effects on human health.Hypersensitive C-reactive protein(hsCRP)is one of the well-established sensitive indicators for the detection of chronic low-grade inflammation in the body.Elevated hsCRP levels can not only predict the risk of cardiovascular disease(CVD),but also related to the increased risks of the morbidity and mortality of chronic diseases including diabetes,cancer,chronic kidney disease,and neurodegenerative disease.Therefore,investigating the environmental factors and mechanisms that affect hsCRP levels has important public health implications for the prevention of these diseases.The levels of hsCRP were associated with various environmental factors including smoking,dietary habit,and air pollution,and metals were important components of these environmental factors.As everyone knows,metal elements are widely distributed in the environment and enter the human body mainly through the respiratory tract and digestive tract.Metal exposure may alter hsCRP levels and affect human health.Several studies have investigated the associations of internal concentrations of single metal with hsCRP levels including cadmium(Cd),lead(Pb),manganese(Mn),copper(Cu),zinc(Zn),and selenium(Se).However,the studies of single metal exposure did not conform to the objective reality of multi-metal exposure in the populations,and they did not consider the influence of metal-to-metal interaction on hsCRP levels,which had certain limitations.In addition,the potential mechanisms of the associations between metals and hsCRP levels remain largely unclear.DNA methylation is an epigenetic modification regulated by genes and environmental exposures,which plays an important biological function by regulating the targeted gene transcription.Metal exposure may lead to changes in hsCRP levels by altering the methylation levels of cytosine-phosphate-guanines(Cp Gs)and m RNA levels of specific genes.However,limited studies have explored the DNA methylation changes associated with metal exposure at the genome-wide level.Furthermore,it is lack of studies to examine the associations between plasma metal concentrations-related DNA methylation alterations and hsCRP levels.Based on the above background,the main purpose of this paper was as follows:(1)to investigate the associations of 17 plasma metal concentrations and the interactions between metals with hsCRP levels;(2)to investigate the DNA methylation profiles associated with plasma metal concentrations at the genome-wide level,and explore the associations between metal-related DNA methylation profiles and hsCRP levels;(3)to explore the possible mechanisms of associations between plasma metals and hsCRP levels by evaluating the relationships of metal-related Cp Gs with the m RNA levels of the targeted genes,and pathways enriched by genes where metal-related Cp Gs were located.This dissertation consists of the following two parts:Part I Associations of plasma metal concentrations with hypersensitive C-reactive protein levelsObjectives:Metal exposure may affect the level of chronic inflammation in the body,and participate in the occurrence and development of various diseases.hsCRP was well-acknowledged as the most sensitive biomarker of chronic low-grade inflammation.However,it was lack of studies evaluating the associations between multiple metal exposures and hsCRP levels.Therefore,we aimed to explore the associations of multiple plasma metals and interactions between metals with hsCRP levels.Methods:We conducted a cross-sectional study in participants from the baseline of the coronary heart disease(CHD)nested-case control populations within the Dongfeng-Tongji cohort(DFTJ cohort)(n=2882).We used inductively coupled plasma-mass spectrometry(ICP-MS)to measure 23 plasma metals concentrations and used the latex immunoturbidimetric method to determine serum hsCRP levels of the participants at baseline.After excluding metals with detection rate below 50%and not suitable for using plasma concentration as an internal exposure marker,generalized linear model was applied to evaluate the associations between the remaining 17 plasma metal levels and hsCRP levels,with adjustment for age,gender,smoking status,drinking status,body mass index(BMI),exercise,sleep duration at night,education level,hypertension,diabetes,hyperlipidemia,estimated glomerular filtration rate(e GFR),and incident CHD status.Given the correlations between plasma metal concentrations,we simultaneously included17 plasma metals in the LASSO penalty regression model to select the plasma metals that were significantly associated with hsCRP levels in order to reduce false positive results.The LASSO penalty regression model was adjusted for the same covariates as the single-metal model.The restricted cubic spline(RCS)model was used to explore the dose-response relationship between plasma metals and hsCRP levels.Finally,we further explored the potential interactions between metals on hsCRP levels.Results:After adjusting for potential confounders,plasma Cu(?:1.046;95%CI:0.909,1.184;FDR<0.001)was positively associated with hsCRP levels,whereas Se was negatively associated with hsCRP levels(?:-0.224;95%CI:-0.377,-0.070;FDR=0.04)in the single-metal model.Plasma Cu and Se were also identified to be significantly associated with hsCRP levels in the LASSO penalized regression model(lambda.1se=0.051).The results of RCS model showed that the positive association of plasma Cu with hsCRP levels and the inverse association of plasma Se with hsCRP levels were both in a linear manner(P linearity<0.001 and P linearity=0.02,respectively).We found significant interactions between Cu and Se,Cu and Zn,with respect to serum hsCRP levels(all P interactions<0.001),and higher plasma Se and Zn concentrations may attenuate the positive association between plasma Cu and hsCRP levels.Conclusions:Our results suggested that hsCRP levels were positively associated with plasma Cu concentrations,and inversely associated with plasma Se concentrations.Moreover,higher Se and Zn concentrations may attenuate the positive association between Cu and hsCRP.Further investigations are warranted to explore the underlying mechanisms for the associations of plasma Cu and Se with hsCRP levels.Part II Associations of plasma metal concentration-associated DNA methylation profiles with hypersensitive C-reactive protein levelsObjectives:In the first part of the study,we found that plasma Cu was positively associated with hsCRP levels,whereas plasma Se was inversely associated with hsCRP levels.In order to better understand the underlying mechanisms for the associations of plasma Cu and Se with hsCRP levels,we further explored the role of DNA methylation involved in their relationships.Methods:We conducted a cross-sectional study in a total of 1055 participants,including682 participants from the baseline of the DFTJ cohort(Dongfeng-Tongji group),230participants from the baseline of the WHZH cohort(Wuhan-Zhuhai group),and 143healthy individuals recruited from Shiyan city(Shiyan-resident group).The Illumina Infinium Human Methylation450K beadchip was used to measure DNA methylation profiles including around 485000 Cp G sites of blood leukocytes in all of the 1055participants.The ICP-MS was used to detect the plasma Cu and Se concentrations in all of the 1055 participants.The latex immunoturbidimetric method was used to measure the serum hsCRP levels in 682 participants from the Dongfeng-Tongji group.The Human HT-12 version 4 Expression Bead Chip was used to perform m RNA profiles in 143 participants from the Shiyan-resident group.In each of the three groups,the general linear regression model was applied to explore the relationship of plasma Cu and Se concentrations with the methylation levels of each Cp G,with adjustment for age,sex,smoking status,drinking status,BMI,regions(only for Wuhan-Zhuhai group),the proportions of major leukocytes,and the surrogate variables(SVs).The inverse variance weighted fixed-effect meta-analysis was used to combine the results.For plasma Cu or Se-related Cp Gs,we analyzed the associations of their methylation levels with hsCRP levels and their mediating effects on the associations between plasma Cu,Se and hsCRP levels in the Dongfeng-Tongji group.Furthermore,we evaluated the associations of plasma Cu or Se-related Cp Gs with the m RNA levels of their targeted or nearby genes in the Shiyan-resident group.Finally,we used gene set enrichment analysis to explore the GO pathways associated with genes annotated to the top 300 Cu or Se-related Cp Gs.Results:We identified eight Cp Gs that were associated with plasma Cu concentrations,including cg25112191 on RORC gene,cg20995564 on ZEB2 gene,cg21852842 on MAD1L1 gene,cg07798295 on CHMP1A gene,cg18608055 on SBNO2 gene,cg26470501 on BCL3 gene,cg05825244 on EBF4 gene,and cg09349128 near ZBED4gene(P<1.0×10–5).The associations of plasma Cu with methylation levels at cg07798295(?=-0.14;FDR=0.05),cg26470501(?=-0.15;FDR=0.05),cg05825244(?=0.15;FDR=0.05),and cg20995564(?=-0.14;FDR=0.06)were nearly reached the genome-wide significance.Among the Cp Gs associated with plasma Cu concentrations,the methylation levels of cg25112191(?=-0.17;P=1.95×10–5),cg20995564(?=-0.09;P=2.05×10–2),cg21852842(?=-0.10;P=1.61×10–2),cg26470501(?=-0.11;P=7.96×10–3),and cg09349128(?=-0.19;P=5.19×10–6)were also significantly associated with hsCRP levels.The mediation analyses suggested that the DNA methylation levels of cg25112191 and cg09349128 may mediate the association between Cu and hsCRP,with a mediation proportion of 4.01%(P for mediations=0.004)and 3.94%(P for mediations<0.001),respectively.In addition,the methylation levels of cg25112191 were inversely associated with m RNA levels of KIAA1026 genes;the methylation levels of cg18608055 were positively associated with m RNA levels of SBNO2 and HMHA1 genes,while inversely associated with m RNA levels of GPX4 genes;the methylation levels of cg26470501 were inversely associated m RNA levels of BCL3 genes;the methylation levels of cg05825244were positively associated with m RNA levels of CPXM1 and EBF4 genes;and the methylation levels of cg09349128 were positively associated with m RNA levels of ZBED4 genes,while inversely associated with m RNA levels of PIM3 genes(P<0.05).The genes annotated to the top 300 Cu-related Cp G sites were enriched in several immune and inflammatory pathways including alpha-beta T cell differentiation,white fat cell proliferation,type I interferon signaling pathway,and tumor necrosis factor binding pathway.The study also determined five Cp Gs that were associated with plasma Se concentrations,including cg23799313 on the PAQR7 gene,cg05445326 on TM4SF1 gene,cg02396116 in the intergenic region,cg13068379 on GALNT9 gene,and cg00097639 on RIPK4 gene(P<1.0×10–5).However,we did not observe the significant associations of Se-related Cp Gs with hsCRP levels or m RNA levels of targeted genes(P>0.05).The genes annotated to the top 300 Se-related Cp G sites were enriched in biological pathways including glutamate metabolic process,the apoptotic process involved in blood vessel morphogenesis,cellular defense response,and positive regulation of fat cell differentiation.Conclusions:We identified eight DNA methylation sites that were associated with plasma Cu concentrations and five DNA methylation sites that were associated with plasma Se concentrations.The methylation levels of five Cu-related Cp Gs were also related to hsCRP levels,two of which significantly mediated the relationship between plasma Cu and hsCRP levels.In addition,the methylation levels of Cu-related Cp Gs were significantly associated with m RNA levels of targeted genes including SBNO2,BCL3,and EBF4 genes,and genes annotated to Cu-related Cp Gs were enriched in immune and inflammation-related pathways.Our study indicated that that Cu exposure may induce changes in m RNA levels by altering the methylation levels of specific DNA methylation sites,leading to the increase of hsCRP levels. |
PDF Full Text Request