Association Between Serum Pyrethroid Levels And Incident Type 2 Diabetes And The Mediating Role Of Plasma Lipid Metabolites

Type 2 diabetes is a major non-communicable disease,which is caused by a combination of multiple environmental and genetic factors.Type 2 diabetes could lead to severe complications,which shorten life expectancy and lower the standard of life,bringing severe challenges to the socio-economic and medical system worldwide.However,traditional risk factors only partially explain the present diabetes epidemic,while studies on the potential risks of environmental chemicals such as endocrine disruptors on type 2 diabetes are still limited,and the molecular mechanism remains largely unknown.Identifying emerging environmental chemicals that confer potential type 2 diabetes risk and elucidating the possible mechanisms would greatly expand our understanding on the environmental risk factors of the diabetes pandemic and provide critical evidence for preventing type 2 diabetes and discriminating populations at high risk.Pyrethroid insecticides(hereinafter referred to as "pyrethroids"),serving as "nontoxic" alternatives to classical pesticides,are widely used in agricultural and household settings and ranked second in the global insecticide market.In vivo and in vitro studies suggested that long-term low-dose exposure to pyrethroids might lead to glucose or lipid metabolism disorders probably through oxidative stress.Several epidemiological studies with a cross-sectional design have also suggested that pyrethroid exposure was associated with higher odds of altered glucose homeostasis or prevalent diabetes.However,prospective studies linking pyrethroid exposure to incident type 2 diabetes are still lacking.Besides,genetic polymorphisms not only affect the biotransformation and metabolism of environment contaminants in the human body,but some are also the key cause of type 2 diabetes.Therefore,it is warranted to explore the modifying effect of genetic susceptibility on the association between pyrethroid exposure and incident type 2 diabetes.Several in vivo studies demonstrated that pyrethroid exposure could lead to abnormal lipid metabolism,which could be involved in insulin resistance and the development of diabetes.On the one hand,plasma lipid metabolites can act as critical indicators of pyrethroid-induced dyslipidemia;on the other hand,they could participate in the occurrence and development of type 2 diabetes caused by pyrethroid.Therefore,identifying the lipid metabolic profile related to pyrethroid exposure and its mediating role in the association between pyrethroid exposure and incident type 2 diabetes might aid in elucidating the potential mechanism underlying the type 2 diabetes caused by pyrethroid exposure.Therefore,the main objectives of this study were 1)to examine the association between pyrethroid exposure and the risk of incident type 2 diabetes and the modifying effect of genetic predisposition on the association;2)to explore the lipid metabolic profile of altered lipid metabolite levels caused by pyrethroid exposure and its mediating role in the association between serum pyrethroid levels and the risk of incident type 2 diabetes.The study consists of the following two parts:Part Ⅰ.Association of serum pyrethroid levels with incident type 2 diabetes and the modifying role of genetic predispositionObjective: To assess the association between pyrethroid exposure and the risk of incident type 2 diabetes and the modifying effect of diabetes-related genetic predisposition.Methods: The study was a nested case-control study based on the 2008 baseline population of the prospective Dongfeng-Tongji cohort,which comprised 1 006 incident type 2 diabetes cases identified during the follow-up period from the baseline to 2013,and another 1 006 randomly selected controls(1:1 matched on age [± 5 years] and sex).Baseline serum levels of eight pyrethroids(including bifenthrin,fenpropathrin,cyhalothrin,cyfluthrin,permethrin,cypermethrin,fenvalerate,and deltamethrin)were determined using gas chromatographytandem triple quadrupole mass spectrometry.Type 2 diabetes was diagnosed by at least one of the following criteria: fasting blood glucose levels ≥ 7.0 mmol/L,glycosylated hemoglobin ≥ 6.5%,self-reported physician-diagnosed diabetes or receiving insulin or oral glucose medication.Odds ratio(OR)with corresponding 95% confidence interval(CI)for the associations of serum levels of eight pyrethroids with incident type 2 diabetes was examined using conditional logistic regression models.Furthermore,weighted quantile sum regression(WQS)and Bayesian kernel machine regression(BKMR)models were performed to examine the association of exposure to pyrethroid mixture with the risk of incident type 2 diabetes.Based on the large-scaled meta-analysis of genome-wide association studies on type 2 diabetes among East Asian populations,161 single nucleotide polymorphisms were selected to construct genetic risk score(GRS)and weighted genetic risk score(w GRS)to assess genetic predisposition to type 2 diabetes.To assess the modifying role of genetic susceptibility in the association between serum pyrethroid levels and incident type 2 diabetes,conditional logistic regression models or "epi R" R package was used to examine the multiplicative and additive interaction effect and to calculate the relative excess risk due to interaction(RERI)and the attributable proportion to interaction(API).Results: The detection rates were 17.59% for bifenthrin,8.40% for fenpropathrin,13.27% for cyhalothrin,39.56% for cyfluthrin,92.45% for permethrin,99.80% for cypermethrin,90.61% for fenvalerate and 99.95% for deltamethrin.Compared to the participants with serum fenvalerate levels in the lowest quartile,ORs(95% confidence intervals [CIs])of the risk of incident type 2 diabetes were 1.20(0.86–1.67),1.41(0.97–2.05),and 2.29(1.27–4.11)for the second,third,and fourth quartile(Ptrend = 0.01)with adjustment for education level,cigarette-smoking status,alcohol-drinking status,physical activity,nighttime sleep duration,frequency of dietary intake(including fish,meat,dairy products,vegetables,and fruits),body mass index,family history of diabetes,estimated glomerular filtration rate,hypertension,and hyperlipidemia.Spline analysis demonstrates a significant positive association between serum fenvalerate levels and incident type 2 diabetes(P for overall association = 0.006).In models not adjusted for estimated glomerular filtration rate,hypertension,and hyperlipidemia,serum deltamethrin levels show a significant linear association with the risk of incident type 2 diabetes.Compared with the first quartile,OR of incident type 2 diabetes for the highest quartile was 1.93(95% CI: 1.16–3.21;Ptrend = 0.002).This association attenuated after adjusting for those factors(highest vs lowest quartile: OR [95% CI] = 1.61 [0.95–2.73];Ptrend = 0.02).The study did not observe a significant association between other pyrethroids and the risk of incident type 2 diabetes.Furthermore,two mixture models(WQS and BKMR)revealed positive associations between mixture exposure to four pyrethroids with detection rates > 70% with the risk of incident type 2 diabetes,with fenvalerate ranked as the top contributor(contribution proportion: > 70%).Gene-environment interaction analyses based on 916 pairs with qualified genotyping data show that although there was nonsignificant association between serum permethrin levels and the risk of incident type 2 diabetes among the whole participants,a significant and positive association was observed among participants in the highest tertile of w GRS.In the highest and lowest tertile of the w GRS,each doubling increase in serum permethrin levels,the ORs(95% CIs)were 1.12(1.00–1.26)and 0.92(0.81–1.04),respectively.Serum permethrin levels exhibited significant multiplicative interaction(Pinteraction = 0.003)and additive interaction(RERI = 1.20 [95% CI: 0.03–2.36];API = 0.39 [95% CI: 0.12–0.66]).Conclusions: Serum levels of fenvalerate and deltamethrin were positively associated with the risk of incident type 2 diabetes.Mixture exposure to serum pyrethroid insecticides was positively associated with the risk of incident type 2 diabetes,which was largely attributable to fenvalerate.Furthermore,diabetes genetic risk might amplify the association between permethrin and incident type 2 diabetes.Further investigations are urgently needed to confirm these findings and elucidate the underlying pathogenic mechanisms,given the widespread use of pyrethroids and the global pandemic of diabetes.Part Ⅱ.Pyrethroid-related lipid metabolic profile and its mediating role in the association between serum pyrethroids and the risk of incident type 2 diabetesObjectives: This part aimed to explore the pyrethroid-associated lipid metabolic profile for two pyrethroids(deltamethrin and fenvalerate)that showed significant association with type 2 diabetes in part one and then evaluate whether lipid metabolites play a mediating role in the association between serum pyrethroid levels and incident type 2 diabetes.Methods: Study participants and the quantification of serum pyrethroids in this section were consistent with that in the first section.The plasma lipid profile of 166 lipid metabolites was determined using high-performance liquid chromatography-tandem triple quadrupole mass spectrometry.Lipidome-wide association study identified lipid metabolites significantly associated with serum pyrethroids among 2 012 participants.Based on the identified lipid metabolites significantly associated with pyrethroids(FDR < 0.05),pathway enrichment analysis was further performed to explore their possible involved metabolism pathway.Furthermore,multiple mediation analyses were performed via PROCESS macro to assess whether plasma lipid metabolites might mediate the association between serum pyrethroid levels and the risk of incident type 2 diabetes.In addition,to explore whether highly correlated lipid metabolites act through similar metabolic pathways,using Clust Var LV R package,these pyrethroid-associated metabolites were clustered into different latent lipid variables and were enriched for important sets of metabolite pathways.Then multiple mediation analyses were performed via PROCESS macro to assess whether these latent variables might mediate the association between serum pyrethroid levels and the risk of incident type 2 diabetes.Results: A total of 59 lipid metabolites were identified to be significantly associated with serum deltamethrin levels,among which eight were also significantly associated with serum fenvalerate(FDR < 0.05).Even after Bonferroni correction,three sphingomyelins(SMs),including SM 38:1,SM 32:1,and SM 40:2,exhibit significantly negative associations with both pyrethroids.Pathway enrichment analysis shows that deltamethrin-associated lipid metabolites,which mainly belong to phosphatidylcholines(PCs)and lysophosphatidylcholines(LPCs),were significantly enriched in the glycerophospholipid metabolism pathway(FDR < 0.05).Furthermore,several deltamethrin-associated lipid metabolites,such as PC 32:0,PC 34:4,cholesterol ester(CE)20:0,and triacylglycerol(TAG)52:5 [18:2],were found to significantly mediate the association between serum deltamethrin and the risk of incident type 2 diabetes,with mediation proportions being 44.81%,15.92%,16.85%,and 16.66%,respectively.Besides,lipid metabolites associated with deltamethrin and fenvalerate were clustered into five and two latent variables,respectively.The latent variable enriched for glycerophosphorylethanolamine significantly mediates the association between serum deltamethrin and the risk of incident type 2 diabetes(mediation proportion: 22.86%).Conclusions: Serum pyrethroids,particularly deltamethrin,were associated with circulating levels of multiple lipid metabolites and mainly involved in the glycerophospholipid metabolic pathway represented by PCs and LPCs.Some lipid metabolites,such as PC 32:0,PC 34:4,CE 20:0 and TAG 52:5[18:2] and glycerophosphorylethanolamine-enriched metabolic pathways might mediate the association between serum deltamethrin levels and incident type 2 diabetes.