| The clinical detection of serum lipids mainly includes total cholesterol（TC）,triglycerides（TG）,high-density lipoprotein cholesterol（HDL-C）,and low-density lipoprotein cholesterol（LDL-C）.The detection of those serum lipids is of great significance for assessing the risk of cardiovascular disease（CVD）.Similarly,the atherosclerosis index（AI）based on serum lipid levels can also be well used to predict CVD risk.Bisphenol A（BPA）,bisphenol F（BPF）,and triclosan（TCS）are the most common phenols,which can interfere with the normal endocrine function.Studies have shown that BPA exposure interferes with lipid metabolism and exacerbates atherosclerosis.BPF is structurally like BPA and has a biologically similar endocrine-disrupting effect to BPA.TCS also has been reported to affect lipid metabolism and cause atherosclerosis.There are currently limited studies on the associations of BPA,BPF,and TCS exposure with serum lipid levels and AI in the population study.Further understanding of the associations of BPA,BPF,and TCS exposure with serum lipid levels and AI in different populations is needed.Serum lipid levels and AI are not only affected by environmental factors,but also by the interaction between genes and the environment.When phenols enter the body,they are metabolized by metabolic enzymes to interfere with normal biological functions.The interaction effects between BPA,BPF,and TCS and metabolism-related pathway genes may influence serum lipid and AI levels.However,there are few studies of the interactions between phenols and metabolism-related pathway genes on serum lipid level and AI.Phenols metabolized by metabolic enzymes in vivo may exert biological functions through various pathways.Available evidence suggests that phenols exposure can increase the level of oxidative stress and induce inflammatory responses.Meanwhile,oxidative stress and inflammatory responses play important roles in lipid metabolism and atherosclerosis development.Therefore,oxidative stress and inflammatory responses may be involved in the associations of environmental phenols exposure with serum lipid levels and AI.Telomeres are nuclear protein structures at the ends of eukaryotic chromosomes.Studies have shown significant associations between phenols exposure and shortened telomere length.And shorter telomere length was significantly associated with serum lipid levels and atherosclerosis risk.Therefore,whether telomere length plays a role between phenols exposure and serum lipid levels and AI deserves further investigation.Based on the above evidence,this study evaluated the association of BPA,BPF,and TCS exposure with serum lipid levels and AI in healthy people who underwent a physical examination at Wuhan Tongji Hospital.At the same time,the interactions between phenols exposure and metabolism-related pathway genes on serum lipid level and AI were explored in this population,and the significant interactions were verified in an independent population from Wuhan Union Hospital.Finally,the mediating effects of oxidative stress,inflammatory response,and relative telomere length（RTL）in the associations of phenols exposure with serum lipid levels and AI were investigated.This research is mainly divided into the following three parts:Part one:Association of phenols exposure with serum lipid levels and atherosclerosis indexObjective:To explore the associations of environmental phenols BPA,BPF and TCS exposure with serum lipid and AI levels.Methods:From November 2017 to October 2019,a total of 1,017 subjects in Wuhan Tongji Hospital were recruited.Ultra-high-performance liquid chromatography high-resolution mass spectrometry（UHPLC-HRMS）system was used to determine the concentrations of BPA,BPF,and TCS;an automatic biochemical analyzer was used to determine the concentrations of serum TC,TG,HDL-C,and LDL-C,and the level of AI[AI=（TC-HDL-C）/HDL-C]was calculated.Multiple linear regressions were used to analyze the associations of urinary BPA,BPF,and TCS with serum lipid levels and AI,and stratified analysis was used to evaluate the association between environmental phenols exposure and serum lipid and AI levels in different subgroups.Results:After adjusting for sex,age,smoking status,drinking status,and body mass index,per 10 units increase in the log-transformed BPA was associated with a 0.14 unit decrease in log-transformed HDL-C（β=-0.14,95%CI=-0.26～-0.02,P<0.05）and 0.21 unit increase in log-transformed AI（β=0.21,95%CI=0.01～0.41,P<0.05）;per 10 units increase in the log-transformed BPF values was associated with 0.12 unit decrease in log-transformed HDL-C（β=-0.12,95%CI=-0.21～-0.02,P<0.05）.Stratified analysis showed that elevated urinary TCS was associated with elevated TC levels in women（β=0.45,95%CI=0.05～0.85,P<0.05）.In people aged>60 years,urinary TCS was associated with increased TC（β=0.77,95%CI=0.18～1.35,P<0.05）,LDL-C（β=0.70,95%CI=0.16～1.23,P<0.05）and AI（β=0.28,95%CI=0.04～0.52,P<0.05）.Conclusions:BPA exposure was significantly associated with decreased HDL-C levels and increased AI levels;BPF exposure was also significantly associated with decreased HDL-C levels.Elevated urinary TCS was found to be associated with elevated TC levels in women;and in people aged>60 years,elevated urinary TCS was found to be associated with elevated levels of TC,LDL-C,and AI.Part two:The interaction effects of phenols exposure and metabolism-related pathway genes on serum lipid levels and atherosclerosis indexObjective:In the first part of the study,we found that BPA exposure was significantly associated with decreased HDL-C levels and increased AI levels;BPF exposure was significantly associated with decreased HDL-C levels.In this part,we further explored the interactions between environmental BPA or BPF exposure and metabolism-related genetic variants on the levels of HDL-C and AI.Methods:A two-stage study strategy was used in this part of the study.A total of 468subjects were included in the discovery phase based on the first part of the study population.A total of 487 subjects were included in the validation phase,and they were recruited at Wuhan Union Hospital from May 2016 to May 2017.Genomic DNA was extracted and genotyped using a genome-wide chip（Infinium?Global Screening Array v1.0）for a two-stage population.Metabolism-related pathway genes were screened through the gene set enrichment analysis database,and single nucleotide polymorphisms（SNPs）in the upstream and downstream 2 kb regions of the genes were extracted.After the quality control of SNPs,237 SNPs were finally included for gene-environment interaction analysis.Multiple linear regression models were used to evaluate the interactions between BPA or BPF and SNPs on HDL-C and AI.The models with interaction two-sided P<5×10-3 in the discovery stage were validated in the second stage,and a two-sided P<0.05 in the validation stage was considered statistically significant.Finally,the associations between urinary BPA or BPF and HDL-C and AI levels were analyzed in different genotype carriers.Results:In the discovery stage,significant interactions of IGFBP7 rs3806744(Pinteraction=1.59×10-4),rs9992658(Pinteraction=1.59×10-3)and rs4865180(Pinteraction=4.91×10-3),APOE rs405509(Pinteraction=5.05×10-4)and rs439401(Pinteraction=4.58×10-3),ABCG1rs4148106(Pinteraction=3.46×10-3)and ELOVL6 rs11098065(Pinteraction=4.77×10-3)and urinary BPA levels on HDL-C levels were found;significant interaction of STAT5B rs185478092(Pinteraction=2.83×10-3)and urinary BPF levels on HDL-C levels was found;and significant interactions of EPAS1 rs59901247(Pinteraction=7.69×10-4),ABCG1 rs492338(Pinteraction=1.15×10-3),IGFBP7 rs11573104(Pinteraction=1.34×10-3)and BMP5 rs16887173(Pinteraction=3.76×10-3)and urinary BPA levels on AI levels were found.The above models of gene-environment interaction were included in the second stage for validation.In the validation stage,it was found that IGFBP7 rs9992658(Pinteraction=9.57×10-3)and rs4865180(Pinteraction=3.03×10-3)still had significant interactions with urinary BPA levels on HDL-C.In combined stage,IGFBP7 rs9992658(β=-0.025,95%CI=-0.041～-0.010,Pinteraction=1.22×10-3)and rs4865180(β=-0.026,95%CI=-0.039～-0.012,Pinteraction=3.17×10-4)also showed significant interactions with urinary BPA on HDL-C levels.In addition,we found that in individuals with the rs9992658 AA genotype,urinary BPA levels were negatively associated with HDL-C(β=-0.020,95%CI=-0.031～-0.009,P=3.76×10-4);in individuals with the rs4865180 TT genotype,urinary BPA levels were also negatively associated with HDL-C(β=-0.022,95%CI=-0.033～-0.010,P=1.85×10-4).Conclusions:The interaction between BPA exposure and metabolism-related pathway gene IGFBP7（rs9992658 and rs4865180）was associated with HDL-C levels.BPA exposure was significantly associated with lower HDL-C in individuals with the rs9992658AA genotype or rs4865180 TT genotype.Part three:Mediating effects of oxidative stress,inflammatory response,or relative telomere length in the associations of phenols exposure with serum lipid levels and atherosclerosis indexObjective:In the first part of the study,we found that BPA exposure was significantly associated with decreased HDL-C levels and increased AI levels;BPF exposure was significantly associated with decreased HDL-C levels.We further explored the mediating effects of oxidative stress,inflammatory response,and RTL in the associations of BPA or BPF exposure with HDL-C and AI levels.Methods:Based on the population in the first part,liquid chromatography-tandem mass spectrometry（LC-MS/MS）was used to determine the urinary oxidative stress markers8-isoprostaglandin F2α(8-iso PGF2α)and 4-hydroxy-2-nonene-mercaptoacetic acid（HNE-MA）concentrations;platelets,total white blood cells（WBC）and its five subtypes（neutrophils,lymphocytes,monocytes,eosinophils,and basophils）counts were determined by an automatic biochemical analyzer.And the system immune-inflammation index（SII）was calculated.Total WBC and its subtype count and SII were used as inflammatory response indicators.A real-time quantitative polymerase chain reaction（q PCR）instrument was used to determine the level of RTL in peripheral blood mononuclear cells;Mediating effects of oxidative stress,inflammatory response,and RTL in the associations of BPA or BPF exposure with HDL-C and AI levels were assessed by mediation analysis.Results:Urinary BPA was associated with increased urinary 8-iso PGF2α（β=0.07,95%CI=0.02～0.13,P<0.05）and HNE-MA（β=0.19,95%CI=0.13～0.25,P<0.05）.Urinary BPF was also associated with increased urinary 8-iso PGF2α（β=0.07,95%CI=0.03～0.11,P<0.05）and HNE-MA（β=0.08,95%CI=0.04～0.12,P<0.05）.We did not find that 8-iso PGF2αand HNE-MA had mediating effects in the associations between BPA and HDL-C and AI levels,and the association between BPF and HDL-C（P>0.05）.Urinary BPF values was significantly associated increased total WBC count（β=0.19,95%CI=0.08～0.31,P<0.05）,lymphocyte count（β=0.16,95%CI=0.03～0.29,P<0.05）,monocyte count（β=0.25,95%CI=0.10～0.39,P<0.05）,and eosinophil count（β=0.41,95%CI=0.05～0.77,P<0.05）.Total WBC count,neutrophil count,lymphocyte count,monocyte count,and eosinophil count were negatively associated with HDL-C levels;and positively associated with AI levels（all P<0.05）.Mediation analysis showed that in the association between urinary BPF concentration and HDL-C level,total WBC count（mediating effect=-0.004,95%CI=-0.007～-0.001,P<0.05）,lymphocyte count（mediating effect=-0.002,95%CI=-0.003～-0.0003,P<0.05）,monocyte count（mediating effect=-0.003,95%CI=-0.006～-0.001,P<0.05）,and eosinophil count（mediating effect=-0.002,95%CI=-0.003～-0.0003,P<0.05）all had significant mediating effects,and mediation ratios were 27.40%,11.20%,20.80%and 10.50%,respectively.No association was found between urinary BPA and BPF with RTL（P>0.05）.In addition,no association was found between RTL and HDL-C levels and AI levels（P>0.05）.RTL did not mediate the associations between BPA and HDL-C and AI levels,and the association between BPF and HDL-C.Conclusion:BPA and BPF exposure were associated with increased urinary oxidative stress markers.Urinary BPF concentration was positively associated with peripheral blood total WBC count,lymphocyte count,monocyte count,and eosinophil count.Total WBC count,neutrophil count,lymphocyte count,monocyte count,and eosinophil count were negatively associated with HDL-C levels but positively associated with AI levels.Total WBC count,lymphocyte count,monocyte count,and eosinophil count significantly mediated the association of BPF exposure with HDL-C levels.The inflammatory response may be one of the underlying biological processes of HDL-C reduction caused by BPF exposure. |
PDF Full Text Request