Background: Evidence suggested that abiotic airborne exposures may be associated with changes in body composition. However, more evidence is needed to identify key pollutants linked to adverse health effects and their underlying biomolecular mechanisms, particularly in sensitive older adults.
Objectives: Our research aimed to systematically assess the relationship between abiotic airborne exposures and changes in body composition among healthy older adults, as well as the potential mediating mechanisms through the serum lipidome.
Methods: From September 2018 to January 2019, we conducted a monthly survey among 76 healthy adults (60-69 years old) in the China Biomarkers of Air Pollutant Exposure (BAPE) study, measuring their personal exposures to 632 abiotic airborne pollutions using MicroPEM and the Fresh Air wristband, 18 body composition indicators from the InBody 770 device, and lipidomics from venous blood samples. We used an exposome-wide association study (ExWAS) and deletion/substitution/addition (DSA) model to unravel complex associations between exposure to contaminant mixtures and body composition, a Bayesian kernel machine regression (BKMR) model to assess the overall effect of key exposures on body composition, and mediation analysis to identify lipid intermediators.
Results: The ExWAS and DSA model identified that 2,4,5-T methyl ester (2,4,5-TME), 9,10-Anthracenedione (ATQ), 4b,8-dimethyl-2-isopropyl-phenanthrene, and 4b,5,6,7,8,8a,9,10-octahydro-(DMIP) were associated with increased body fat mass (BFM), fat mass indicators (FMI), percent body fat (PBF), and visceral fat area (VFA) in healthy older adults [Bonferroni-Hochberg false discovery rate ðFDRBHÞ < 0:05]. The BKMR model demonstrated a positive correlation between contaminants (anthracene, ATQ, copaene, di-epi-a-cedrene, and DMIP) with VFA. Mediation analysis revealed that phosphatidylcholine [PC, PC(16:1e/18:1), PC(16:2e/18:0)] and sphingolipid [SM, SM(d18:2/24:1)] mediated a significant portion, rang-ing from 12.27% to 26.03% (p-value <0:05), of the observed increase in VFA.
Discussion: Based on the evidence from multiple model results, ATQ and DMIP were statistically significantly associated with the increased VFA levels of healthy older adults, potentially regulated through lipid intermediators. These findings may have important implications for identifying potentially harmful environmental chemicals and developing targeted strategies for the control and prevention of chronic diseases in the future, partic-ularly as the global population is rapidly aging. https://doi.org/10.1289/EHP13865
Introduction
The global population is currently undergoing a notable aging trend, primarily driven by declining fertility rates and increased life expectancy.1 This demographic shift is evident as the number of individuals 65 years of age and above reached 761 million in 2021, surpassing the population growth rate of children.2,3 By 2050, the older adults aged 65 and above will account for 16% of the total population,4 attracting increasing attention to the issues
of healthy aging.5 Aging leads to gradual deterioration in physio-logical functions, such as growth in fat mass and losses in muscle mass, bone mineral density, and water content.6,7 These changes significantly affect the nutritional and endocrine status, quality of life, and the development of comorbidities in the older adults.8 Understanding dynamic changes in body composition in the older adults and their driving factors can greatly assist health care professionals in identifying health risks of the older population and implementing timely and appropriate treatment and preventive measures.9
Body composition can be affected by numerous factors such as genetics, dietary habits, lifestyle, and diseases.10,11 In recent years, growing evidence has recognized environmental factors as primary influencers of body composition change.12 Observational studies indicated that air pollutants [e.g., particulate matter with aerody-namic diameter <2.5 l m (PM2.5)] may have detrimental effects on adult body composition, leading to increased fat mass and decreased lean mass.13-15 Gaseous chemical substances, including airborne polycyclic aromatic hydrocarbons (PAHs), have the potential to serve as obesogens. Research suggests that when pregnant mothers and their unborn children are exposed to these substances, there is a heightened risk of obesity developing in children.16 Despite these links, the focus in most previous studies has primarily centered on hard outcomes (e.g., mortality, disease deterioration, and hospitalization) rather than exploring subclinical bioindicators.17 This gap in research emphasis may limit our com-prehension of the complex relationship between exposure to air pollution and health outcomes. Furthermore, most previous studies have solely examined the effects of a single exposure, resulting in fragmented research evidence and a high false positive rate.18 Consequently, it remains challenging to accurately and compre-hensively assess the complex array of airborne chemicals at perso-nal exposure levels and understand the major environmental drivers behind changes in body composition.19
To address these challenges, the notion of the "exposome" has emerged, which refers to the whole of environmental exposures during an individual's lifetime.20 This innovative perspective offers a promising approach to studying environmental health.20 Human health is influenced by both biotic and abiotic environmental variables, and while this study primarily focuses on abiotic factors, it acknowledges the role of biotic factors in shaping human health.21,22 Within the exposome framework, the abiotic airborne exposome refers to the composition of chemicals in the gas phase (e.g., alkanes, aromatic compounds, and heterocyclic compounds) and particle phase (e.g., PM2.5) in the surrounding air.21 The abiotic airborne exposome has been demonstrated to have a consider-able effect on human health, particularly with increasing emissions from modern industry and economy.23,24 Research on abiotic factors helps us understand the causes of air pollution, identify emis-sion sources, and assess the health effects of air pollution.21 This knowledge is crucial for formulating targeted environmental pro-tection policies.
Recently, advanced exposome tools have emerged, providing sophisticated capabilities for exposure assessments, data integra-tion, statistical analysis, and visualization to disentangle the intri-cate connections between exposure to the environment and health outcomes.25 At the population level, longitudinal environmental exposure monitoring approaches have encompassed remote sens-ing, fixed monitoring stations, and data modeling from various sources.26 At the individual level, portable devices (e.g., sensors and wearable passive samplers) have been deployed to enable precise and dynamic personal exposure measurements.21,26,27 Multi-omics analyses (e.g., lipidomics) can allow for a deep investigation into the biological mechanisms underlying precision health.28-30 Lipidomics is an emerging field that plays a crucial role in understanding various biological functions, including cel-lular and tissue architecture, membrane function, energy storage, signaling, and immune response.31 It involves profiling lipids and factors that interact with lipids at a systemic level.32 Numerous genetic research on human disorders have shown how important lipids are to the physiology of cells, tissues, and organs.33 The discovery of specific lipid molecules as promising biomarkers is an area of active and encouraging research.34 Several studies
have revealed the predictive value of lipid analysis in assessing the risk of clinical conditions such as diabetes and cardiovascular diseases.35,36 Due to the significant impact of obesity on human health, the correlation between lipidomics and body composition [such as body mass index (BMI) and body fat percentage] has received widespread attention from researchers in recent years.35-37 Various lipid molecules are considered to be sensitive indicators for monitoring obesity-related issues. To unravel the complex relationship between the exposome and health, numerous statistical toolsets [e.g., exposome-wide association study (ExWAS), deletion/substitution/addition (DSA) algorithm, and Bayesian kernel machine regression (BKMR)] have been applied to identify associations between independent and joint exposures and health phenotypes.38,39 Additionally, radiation-free and non-invasive techniques like bioelectrical impedance analysis (BIA) have been proven valuable in accurately assessing various body indicators (e.g., body fat mass and soft tissue muscle mass), particularly in the older adults.40,41 With the continuous advance-ment of exposomic tools, the integration of multiple technologies can allow for the exploration of exposome-internal-omes inter-actions and the study of potential mechanisms affecting individual health.24
As part of the China Biomarkers of Air Pollutant Exposure (BAPE) study,42,43 we conducted a longitudinal panel exposomics investigation, which aimed to a) systematically explore associations between components of the abiotic airborne exposure (e.g., bulk PM2.5 mass, PM2.5 inorganic elements, and organic chemicals) and changes in body composition (e.g., body components, body water contents, muscle indicators, and body fat indicators) among healthy adults (60-69 years old), b) employ a mixed exposure model to identify the most affected body composition indicators, and c) integrate lipidomics and mediation analyses to identify the underlying endogenous biomolecules and mechanisms driving the associations between exogenous exposures and body composition changes.
Methods
Study Design and Population
This study recruited 76 healthy adults 60-69 years of age (50% females) during September 2018-January 2019 in Jinan city, Shandong province, China. Monthly exposure assessment and biomonitoring were undertaken over a span of 5 months. The research period encompassed both the heating and nonheating seasons of fall and winter. This period was crucial for observing health effects due to heightened air pollution during colder months. Individual exposure monitoring was conducted prior to each follow-up visit, ensuring that participants had 3 d of monitoring data in their normal routines. Questionnaires and biological sample collection were performed during the visits. Prior to the physical examinations, participants were required to complete questionnaires regarding their personal information (such as age, gender, medical history, education level, and household income) and activity time in the past week. For lipidome analysis, fasting ve-nous blood samples were collected for each participant at 7:00 AM during the physical examinations and immediately stored at -80°C All participants received free daily meals for 5 d straight on each visit (2 d for acclimation and 3 d for monitoring) in order to reduce the impact of varying dietary patterns. A total of 293 visits (including participants with at least one follow-up) were included in the analysis (Figure 1).
The study design and inclusion/exclusion criteria were previ-ously described.42 The inclusion criteria included living in the community for more than 2 years, voluntary participation, independent ability to complete the questionnaire, and regular living habits. The exclusion criteria included history of smoking, alcohol abuse, previous diagnosis of chronic diseases, still working, fever, and infection using antibiotics and hormones in the past month. The exposures have been previously reported in existing studies, while the analysis of body composition and lipidomic data was conducted for the first time in this study.44-46 The study, approved by the Ethics Review Committee of the National Institute of Environmental Health (NIEH) of Chinese Center for Disease Control and Prevention (China CDC) (No. 201816), and for which written informed consent was obtained from each participant prior to enrollment, has been previously mentioned in the
literature.47
Abiotic Airborne Exposure Assessment
As previously reported,48 we measured personal exposures to a total of 632 abiotic airborne substances in our study subjects, con-sisting of PM2:5 mass concentration (n = 1), PM2:5 inorganic ele-ments (n= 18), and gaseous chemicals in 26 categories (n= 613). We monitored exposure to gaseous chemicals by means of a novel wearable passive sampler called the Fresh Air wristband.49 With the aim of enhancing analysis throughput, this sampler was specifi-cally designed to measure contaminants exposure solely through the inhalation route and employ a solvent-free extraction protocol. The analytical procedures for measuring these chemicals were reported elsewhere.44 Based on the exposure studies examining a range of chemical substances, it was observed that in indoor environments similar to the residences of participants in the BAPE study, compounds with log KOA (octanol-air partition ratio) values ranging from 4.15 to 11.72 experienced linear absorption over the course of a week.48 A 3-day exposure assessment period was
determinedto be the most effective in capturing a wide spectrum of airborne environmental contaminants, taking into account both the technical capabilities of Fresh Air wristbands and the practical aspects of conducting such a study.48,49 The Fresh Air wristband consisted of the wearable form using a wristband, and a custom-ized polytetrafluoroethylene (PTFE) chamber that could hold four polydimethylsiloxane (PDMS) sorbent bars. These PDMS sorbent bars were 1 cm long glass rods with a layer of PDMS sorbent mem-brane (effective sorbent thickness =0:33 mm), which can pas-sively collect gas-phase air pollutants via diffusion. After 3 days of wearing, we collected the PDMS samples from the participants locally, which were then transported through cold-chain shipping to Yale University for chemical analysis.44 Airborne chemical substan-ces collected by PDMS were thermally extracted, followed by chemical separation and identification using gas chromatography Orbitrap high-resolution mass spectrometry (GC-Orbitrap HRMS).44 The col-lected data was screened using multiple libraries, with ? 800,000 chemical substances from Wiley Registry 11th edition/National Institute of Standards and Technology (NIST) 2017 Mass Spectral Library (Wiley Science Solutions, John Wiley & Sons, Inc.), and vendor libraries from Thermo. Through the use of an internal script, differ-ences in the signal intensities of identified molecular characteristics throughout various collection periods and batches were normalized using the median normalization technique. The data filtering and workup process involved blank feature filtering (BFF) using labora-tory and transport blanks, removal of duplicate compound identifica-tions, and application of average score filtering. Compounds meeting the criterion of their 95th percentile surpassing the average of the blank (laboratory and transport) plus three times the standard deviation were retained to ensure inclusion. Features not meeting this criterion were uniformly removed. The identified features underwent an annotation workflow for accurate identifications with a low false positive rate. The annotation workflow described involved the use of Thermo's deconvolution plugin and stringent filtering parameters, which were based on exact mass, retention index, and electron ionization (EI) spec-tral match.44 Chemical substances with a detection rate exceeding 80% were subsequently chosen for further analysis.
In addition, individual PM2.5 concentrations and PM2.5 inorganic elements (e.g., Al, As, Ba, Br, Cr, Cs, Cu, Fe, Mn, Ni, Pb, Rb, S, Se, Si, Ti, V, and Zn) were monitored using MicroPEM sensors (version 3.2; RTI International).50 MicroPEM was used to monitor individu-al's indoor and outdoor PM2.5 concentrations continuously for 72 h. The MicroPEM was mandated to be worn at the waist level, with a plastic tube connecting it to the MicroPEM. The opposite end of the tube was to be maintained at the level of the respiratory belt to repli-cate real-life personal exposure scenarios. During sleep, the MicroPEM was to be placed on the bedside cabinet at a height corre-sponding to the participant's respiratory level. During the sampling period, the instrument operated at a flow rate of 500 mL/min, with a sampling duration of 60 s and an interval of 240 s between each sampling. To minimize losses of organic substances, all filters were promptly stored at - 20° C in plastic boxes after sampling. The filters were sent to RTI International (Research Triangle Park, NC) for X-ray fluorescence analysis to quantify the concentrations of all 33 elements. Eighteen elements were chosen for study because their detection rates were more than 80%, whereas the majority of the elements were below the detection limit. These specific testing methods were reported in previous literature.50,51
Body Composition Measurement
We evaluated body composition using the InBody 770 device (Biospace China Co., Ltd.). The principle of this technique is based on the fact that the time it takes for a low-voltage electric current to pass through the body is influenced by the unique characteristics of body composition.52 The trained professionals conducted on-site body composition assessments and completed health examination report forms during follow-up. The participants stood barefoot on the platform of the instrument, with the feet placed on the four corre-sponding electrodes, holding the handles with both hands, with the thumb and palm in contact with the corresponding electrodes, respectively, and with the arms spread out to ensure no contact between the arms and the torso. They stood still for ~ 1 min while keeping the elbow fully extended and the shoulder joint abducted at an angle of ~ 30 degrees. The participants remained in this position at all times during the assessment. The body composition indicators measured included four body components [protein, minerals, body cell mass (BCM), and bone mineral content (BMC)], four body water contents [total body water (TBW), intracellular water (ICW), extracellular water (ECW), and ECW/TBW], four muscle indicators [soft lean mass (SLM), skeletal muscle mass (SMM), fat free mass (FFM), and fat free mass indicator (FFMI)], and six body fat indicators [weight, BMI, body fat mass (BFM), percent body fat (PBF), fat mass index (FMI), and visceral fat area (VFA)].
Serum Lipidomics and Data Processing
Lipids were measured from the fasting venous blood samples (n = 353).53 Venous blood was centrifuged at 2,500 ×rpm for 10 min at 4°C, and the supernatant was taken to obtain serum. Samples were extracted by passing 100 lL of serum combined with 400 lL of CHCl3/MeOH 2:1 (vol/vol) (analytical grade; Merck) in order to extract lipids. The mixture was combined with an equivalent amount of organic solvents (CHCl3/MeOH 2:1) after it had been vortexed three times and centrifuged for 15 min at 4°C with 3,000×rpm. In a fresh glass tube, the organic phase (lower layer) was transferred, and the water phase was added in a
1:4 water to organic phase ratio. Vortexing the mixture again, it was centrifuged for 15 min at 4°C with 3,000 × rpm. The lower phase was extracted and then dried using nitrogen gas at 4°C. A Q-Exactive Orbitrap mass spectrometer (Thermo Scientific) equipped with an XSelect CSH C18 column at 45°C was used to perform lipidomics analysis on an ultra-high-performance liquid chromatography (UHPLC) UltiMate 3000 (UHPLC; Thermo Fisher). The mobile phase A: acetonitrile/10mM ammonium ace-tate solution (vol/vol) ratio was 3:2; and the mobile phase B: ace-tonitrile/isopropanol (vol/vol) ratio was 1:9, with a flow rate of 250 lL=min. Raw serum lipidomics data were processed using LipidSearch 4.1.30 (Thermo Fisher). For qualitative lipid identification, the LipidSearch database was utilized, and M scores were employed to compare fragment ions with database information.54 Lipid annotation was conducted using the following criteria: a) a mass tolerance of 5ppm for precursor masses in the database, b) a mass tolerance of 10 ppm for fragment masses, and c) a retention time shift of 0.15 min. False positives were manually checked and eliminated based on the M-score and chromatographic behavior. After data filtering, normalization to lipid internal standards repre-sentative of the analyte class was performed. To prevent contami-nation, fresh mobile phases were prepared for each analysis. To evaluate the data quality, equal aliquots of representative sub-groups of participants were combined to form quality control (QC) samples. One QC sample was injected per 10 samples after the first five QCsample injections were examined for column conditioning. To mitigate interference from system instability, the sequence order of all samples was randomly assigned.
Covariate Measurement
Every survey round involved face-to-face interviews with partici-pants utilizing computerized questionnaires to gather information about them.42,46 At the initial visit, all participants had their age, sex, education level, horizontal distance to the nearest traffic road, and financial income measured. Other time-varying factors (e.g., plasma cotinine concentration, drinking habits, activity time, and diet) were assessed using physical measurements and a daily time-activity questionnaire at each visit. The participants reported not smoking, but in order to determine whether they had been pas-sively exposed to tobacco, plasma cotinine levels were measured using a C18 column (Hypersil Gold; Thermo Fisher) interfaced with an LC-Q-Exactive Orbitrap mass spectrometer (Thermo Fisher). To address the correlation of repeated measurements on the same participant, we assigned an identity number (ID) to each participant. Within the longitudinal ExWAS analysis, the ID of each participant was included as a random intercept in the model. During the analyses using BKMR models, we adjusted the ID in the "id" parameter to account for the longitudinal data. Previous researches were used to adapt each primary model for predeter-mined criteria,55,56 including sex (male/female), age (continuous, years), income [continuous, 10k Chinese Yuan per year (CNY/ year)], plasma cotinine concentration (continuous, ng/mL), weekly activity time (continuous, h), educational level (below primary school, primary school, junior high school, senior high school, and university and above), drinking (alcohol intake at least once in the previous month, yes/no). Hourly monitoring data of temperature (° C) and relative humidity (%) were obtained from the environmen-tal monitoring station operated by the MicroPEM. Additionally, a 3-degree of freedom natural cubic spline was introduced for the 3-d average daily temperature and relative humidity.57,58
Statistical Analysis
Using the predictive mean matching approach and a chained equation from the mouse package in R, the missing data for
PM2:5 and inorganic elements was imputed. All exposed variables were then transformed to a base-10 logarithm and standar-dized using z-scores normalization for comparisons. The body composition indicators were also log10-transformed before the analysis. Descriptive analysis was conducted using different measures depending on the distribution of the data. Mean ± standard deviation (SD) was used for data that were normally dis-tributed, while median and interquartile range (IQR) were used for data that had a skewed distribution. Percentages (%) were used to display qualitative data. Spearman correlation analysis was performed to evaluate the internal correlation among all exposures and body fat indicators. All statistical analyses were performed using R (version 4.2.3; R Development Core Team), with the lme4, DSA, and BKMR packages.
ExWAS Analysis
The ExWAS analysis involved estimating the association between exposure and outcome by using independent linear regression models for each covariate.59 In comparison to traditional epidemiolog-ical methods, ExWAS analysis systematically assessed a vast array of potential environmental factors and their associations with health outcomes, with a capability of identifying previously unnoticed or unknown environmental exposures.59 ExWAS anal-ysis was employed to estimate the associations between the abi-otic airborne exposures (632 factors) and body composition (18 indicators). To translate the b estimate and standard error (SE) from the ExWAS models into the percentage change (with a 95% confidence interval), the following equation was used: percentage change ð%Þ = ð10b-1Þ×100.60 A p-value of <0:05 was regarded statistically significant, while a p-value of <0:05 after correction for Bonferroni-Hochberg false discovery rate (FDRBH) indicated good robustness.
DSA Method
To account for co-exposure, we screened the exposure variables with the DSA variable selection algorithm, with each exposure variable adjusted for the other exposures. The DSA is a widely used selection method known for its high sensitivity, low false detection rate, and selection of relevant exposure.39,61 The DSA algorithm started with an empty model and built the final model through the iterative process of exposure variables deletion, sub-stitution, and addition. Using 5-fold cross-validation to minimize the residual mean squared error (RMSE), the final model was chosen.62 In this study, DSA was run 100 times using the DSA package in R. If at least 5% (n > 3) of the results selected from the all exposures from the model, it was retained in the final model. The overlapping results of DSA and ExWAS were included in the subsequent analysis.
BKMR Model
Given the potential nonlinear relationship and multicollinearity among mixed exposures, the BKMR model was used to evaluate the overall impact of the abiotic airborne exposures and the dose-response relationships between each chemical on body composition.63 The exposure-response function was repeatedly estimated using Gaussian kernel functions in the BKMR model, which integrated Bayesian and statistical learning techniques. The posterior inclusion probability (PIP) was utilized to assess the relative significance of each mixture component on the out-come variable. As a nonparametric method, the BKMR model can be used to estimate the overall mixture effect.38,64 Using the results of the ExWAS and DSA models, we identified a set of key chemical exposures for each body composition indicator (Table S1). The overall impact of key chemical exposures on
body composition indicators was estimated by calculating the expected changes in quartiles at specific exposure levels in rela-tion to changes in body composition indicators. A 10,000-itera-tion stratified variable selection technique was carried out using a Markov chain Monte Carlo (MCMC) algorithm in light of the substantial correlation among contaminants. The relative rele-vance of each molecule with respect to the body composition in-dication was ascertained using a PIP threshold of 0.5.
Mediation Analysis
Mediation analysis was conducted using the lipidome as sus-pected lipid mediators in the association between the key abiotic airborne exposures and body composition. This method separated the overall impact of abiotic airborne exposures on body compo-sition into two parts: the direct exposure effect and the mediation effect via the lipidome.65 Based on the results of ExWAS, DSA, and BKMR models, we analyzed the relationship among key air-borne exposures-lipidome-VFA. We estimated the total and mediating effects by fitting two linear mixed effect models (LMEs). For estimating the mediation effect, we applied the following four criteria to determine whether: a) the key air-borne exposures and VFA were significantly correlated; b) the key airborne exposures were significantly associated with the expression of lipidome in LMEs; c) the mediation model for key airborne exposures-lipidome-VFA showed statistical sig-nificance (p-value <0:05), and the total effect direction aligned with the mediation effect; and d) reverse mediation analysis, namely the analysis of key airborne exposures-VFA-lipidome, confirmed the correctness of the mediation results. All of these analyses adjusted for covariates in a manner consistent with the ExWAS model.
Sensitivity Analysis
Sensitivity analyses were performed to evaluate the ExWAS, BKMR, and mediation analysis models by adjusting for the fol-lowing covariates in the main model: different horizontal distan-ces of the house from the nearest road (e.g., <50 m, 50?100 m, 101?200 m, 201?300 m, >300 m, and unknown), and other dietary factors (consuming extra food in addition to the provided standardized meals during the 3-d investigation, yes/no). In the sensitivity analysis, a two-sided p-value of <0:05 was considered statistically significant.
Literature Comparison
Initially, we established clear and concise objectives for our litera-ture search, including the following: a) to identify the accumulation of key abiotic airborne exposures in the environment and b) to iden-tify evidence of lipid intermediators affected bykey abiotic airborne exposure. We searched PubMed, Web of Science, and Google Scholar using the following search terms based on the results of ExWAS, DSA, and mediation analysis models: a) key abiotic air-borne exposures (e.g., 9,10-Anthracenedione OR 4b,8-Dimethyl-2-isopropylphenanthrene,4b,5,6,7,8,8a,9,10-octahydro- OR 2,4,5-T methyl ester) AND environmental occurrence and accumulation (e.g., water OR air OR soil OR blood OR urine); b) [glycerophos-pholipids OR sphingolipids OR SM(d18:2/24:1) OR Hex1Cer (d18:1/24:0) OR Hex3Cer(d18:1/22:0) OR Cer(d18:1/22:0) OR PC (16:2/18) OR PC(16:1/18:0) OR PC(34:1e) OR PC(20:1/18:2) OR PI(18:0/20:3)] AND (mechanism OR function OR obesity OR adi-pose tissue). Our approach commenced with a comprehensive ex-ploration of full text of the literature found using the above search terms, wherein the keywords pertaining to both the pollutants them-selves and their structural analogues were employed. Furthermore, the inclusion of literature was contingent upon its alignment with
the predefined temporal scope (January 2018 to June 2023) and ad-herence to the English language criterion. Additionally, to compare our findings with the existing literature, a thorough examination of the reference list of articles was undertaken to identify any pertinent literature that could support our study.
Results
Characteristics of the Study Population
This study included a total of 293 measurements taken from 76 partic-ipants over a 5-month period. The average age of participants was 65 years old, with 38 of them being female (50%), and the median annual family income was 84,000 CNY (about $12,000). Most participants had a junior (27.6%) or senior high school education (43.4%). The me-dian plasma cotinine concentration was 0.24 (0.23) ng/mL, indicating possible passive smoking. Table 1 presents an overview of body com-position measurements and participant characteristics. Table S2 shows the basic characteristics and relevant data of body composition of all survey subjects across five repeated measurements. Table 2 and Table S3 provide details on the factors of the abiotic airborne expo-sure, including PM2:5, 18 inorganic elements associated with PM2:5, and 613 organic gas phase chemicals. Excel Table S1 represents the CAS number, missing data status, median, mean, the 25th percentile (P25 ), the 75th percentile (P75), IQR, limit of detection, and units of the raw data for airborne exposures. Excel Table S2 illustrates the cor-relations between different levels of gaseous exposures.
Association between the Abiotic Airborne Exposure and Body Composition
The ExWAS model was used to assess the relationship between a total of 632 abiotic airborne exposure factors and 18 body composi-tion indicators in healthy older adults. The number and direction of exposures with significant differences in the ExWAS model are shown in Figure 2 and Table S4. A total of four abiotic airborne exposures were found to be associated with BFM, four exposures were related to FMI, five exposures were associated with PBF, and five exposures were related to VFA (FDRBH < 0:05). Among them, 2,4,5-T methyl ester (2,4,5-TME), 4b,8-dimethyl-2-isopropylphe-nanthrene,4b,5,6,7,8,8a,9,10-octahydro-(DMIP), 9,10-anthrace-nedione (ATQ), and 2,5-di-tert-butyl-1,4-benzoquinone (DTBQ) were positively associated with BFM, FMI, PBF, and VFA. For instance, BFM and VFA increased by 1.57% [95% confidence interval (CI): 0.95%, 2.18%] and 1.36% (95% CI: 0.68%, 2.05%), respectively, for each IQR increase in ATQ. Similarly, BFM, FMI, PBF, and VFA increased by 1.21% (95% CI: 0.80%, 1.61%), 1.10% (95% CI: 0.69%, 1.51%), 1.00% (95% CI: 0.66%, 1.34%), and 1.12% (95% CI: 0.67%, 1.57%), respectively, for every IQR increase in DMIP. Conversely, 1,2-diphenyl-2-butene was negatively corre-lated with PBF. Excel Table S3 shows the corresponding IQR changes, p-values, 95% CI, and FDRBH of the ExWAS model.
Combining the results of the ExWAS and DSA models, the final result identified eight key abiotic airborne exposures that were related to the changes in body composition indicators, including 1-Phenylnaphthalene, ATQ, 2,4,5-TME, DMIP, 5-Isopropenyl-3,8-di-methyl-1,2,4,5,6,7,8,8a-octahydroazulene (IDHOA), Anthracene, Copaene, and Di-epi-a-cedrene (Figure 3). The final results of the two models are shown in Excel Tables S4-S7. Specifically, after the dual screening process of ExWAS and DSA, we discovered that 2,4,5-TME was positively associated with BFM while ATQ and DMIP were also positively associated with both BFM and VFA. Additionally, 2,4,5-TME and DMIP were positively associated with FMI and PBF. The robustness of our overall results was demon-strated by the sensitivity analysis results, which agreed with the main model's conclusions (Figure S1; Excel Tables S8-S11).
The Mixture Impact of Abiotic Airborne Exposure on Body Composition
In light of the significant results from the cross-validation of the ExWAS and DSA models, four body fat indicators along with their corresponding abiotic airborne exposures were included in the BKMR analysis. Figure 4 illustrates the overall impact of a 3-d exposure to abiotic airborne contaminants on changes in body composition in healthy older adults. The Spearman correla-tion for the four body fat indicators can be found in Figure S2. In comparison to when they were at the 50th concentration percen-tile level, the impacts were studied across various exposure con-centration percentile levels. The findings show that there was a positive correlation between VFA and pollutants (anthracene, ATQ, copaene, di-epi-a-cedrene, and DMIP) when they were at or above the 55th percentile as opposed to the 50th percentile.
The trends of univariate exposure-response functions of the selected pollutants were shown in Figure S3. We found that body fat indicators showed an increasing trend with the increase of ATQ, DMIP, and 2,4,5-TME. The interaction between different levels of abiotic airborne exposures can be viewed in Figure S4. This analysis is conducted by fixing other abiotic airborne contaminants at the median to study the exposure-response function between one abiotic airborne exposure at the bottom 10th, 50th, and 90th percentiles and another exposure. The parallel relationship between different quartiles indicated no evi-dence of interaction. We found that there is interaction present between the abiotic airborne contaminants associated with VFA, such as di-epi-a-cedrene and anthracene. Table S5 dis-played the BKMR model-derived PIP for the four groups. Only the group PIP value for VFA exceeded 0.5, indicating a higher like-lihood of association. Among the pollutants, Di-epi-a-cedrene shows the highest group PIP value of 0.99. The sensitivity analy-sis's outcomes agreed with the primary model's conclusions (Table S6-S7; Figure S5).
The Identification of the Lipidome Mediators
Based on the previously described results, DMIP and ATQ were selected as key abiotic airborne exposures for further lipidome mediation analysis. DMIP was found to induce increases in VFA that were associated with six lipids: PC(16:2e/18:0), PC(16:1e/ 18:0), PC(34:1e), Hex3Cer(d18:1/22:0), PI(18:0/20:3), and Cer (d18:1/22:0). Conversely, ATQ-induced increases in VFA were associated with three lipids: SM(d18:2/24:1), Hex1Cer(d18:1/ 24:0), and PC(20:1/18:2). The results of lipidome mediation of the relationship between individual exposure and VFA are shown in Figure 5. Increase in sphingolipid [SM, SM(d18:2/24:1)] mediated 26.03% (95% CI: 2.64%, 64.99%) of the total effect of ATQ on VFA. Decrease in Glycerophospholipids [PC, PC(16:2e/ 18:0), and PC(16:1e/18:1)] mediated 12.43% (95% CI: 1.72%, 28.08%) and 12.00% (95% CI: 1.29%, 27.79%) of the total effect of DMIP on VFA, respectively (Excel Table S12). No statisti-cal differences were found in the reverse mediation results (all p-values >0:05) (Excel Table S13).
Discussion
As far as we know, this study provides a novel perspective on systematically examining the relationship between the personal abiotic airborne exposures and changes in body composition, employing exposome-wide analysis, within healthy adults (60-69 years old). In addition, we explored potential lipid intermediators and the underlying biological mechanisms. Our findings offer novel insights by revealing that ATQ and DMIP were signifi-cantly associated with increased VFA among healthy older indi-viduals. The mixed exposure model demonstrated a positive correlation between the combined effect of these key abiotic air-borne exposures and VFA. Mediation analysis further suggested that alterations in glycerophospholipids and sphingolipids may play a crucial role as lipid intermediators in the association between these exposures and increased VFA. These findings may extend our current understanding of body composition changes in older adults.
Identification of the Key Abiotic Airborne Exposure
Prior epidemiological research has mostly looked at the relation-ship between children's body composition and air pollution, with an emphasis on the bulk measurements of PM2:5 mass concentra-tion. Longitudinal research conducted in Spain, the United States,
Europe, and China has consistently found a correlation between increasing outdoor PM2:5 concentrations and higher BMI in chil-dren.56,66,67 Investigations conducted in New York revealed a link between prenatal exposure to airborne PAHs and early childhood PBF, obesity risk, and BMI increases.68,69 According to some research, long-term exposure to ambient air pollution (e.g., PM2:5 mass concentration) was positively correlated with older persons' BMI.70-72 However, to our knowledge, there are no studies on the association between personal exposure to PAHs and changes in body composition in the older adults. In this study, we found that three key abiotic airborne exposures (ATQ, DMIP, and 2,4,5-TME) were significantly positively associated with body fat indicators (BFM, FMI, PBF, and VFA) in the older adults, but no significant correlation was found between bulk PM2:5 mass or PM2:5 elemental components with body composition indicators.
The presence of these key exposures in the environment has been previously documented in the literature (Table S8). For example, ATQ is an oxygen-containing polycyclic aromatic hydrocarbon (OPAH) that is primarily emitted into the atmos-phere through sources similar to PAHs (e.g., incomplete combus-tion of coal).73 OPAHs have been found in both urban and rural environments, indicating their widespread presence in the air.74-77 Furthermore, due to increased use of biofuels for residential heating, OPAH concentrations are higher in the atmosphere during the winter season, exhibiting a seasonal pattern.78,79 Additionally, stable detec-tions of OPAHs have been observed in water and soil contaminated with PAHs, indicating the persistence of OPAHs and their potential for environmental accumulation.80,81 Although the study indicates that OPAHs make a significant contribution to air pollution and can even exhibit carcinogenic effects without enzymatic activa-tion, they have not received sufficient attention in population health risk assessments.73,82,83 This lack of attention may be attributed to limited awareness and understanding of OPAHs, as well as the challenges associated with their detection and analysis. Although the exact toxicity mechanism of ATQ is yet unknown, prior research has shown that OPAHs can signifi-cantly increase lipid peroxidation and induce reactive oxygen species generation.73,84,85 DMIP, a phenanthrene compound produced through burning fuels or petrochemical manufactur-ing, has limited reports on its presence in the environment.86 While it has the potential to cause harm to adipose tissue, there are currently insufficient data to substantiate this claim.87 However, phenanthrene is considered to be very persistent based on comprehensive assessments. It was discovered in high amounts within contaminated air and water and could be absorbed and accumulate in tissues.88,89 Several epidemiologi-cal studies provide evidence of a positive association between phenanthrene and adult obesity, as well as fat mass percent, which is consistent with our findings.90,91 2,4,5-TME, the main component of Agent Orange, is a synthetic organic substance mainly used as an herbicide and banned in many countries. However, no studies that we found have investigated its associa-tion with adipose tissue. 2,4,5-TME is the methyl ester form of 2,4,5-Trichlorophenoxyacetic acid (2,4,5-T). 2,4,5-T is an herbicide that has been banned in many countries due to its use as a plant growth regulator and the presence of toxic dioxins as one of its by-products.86 As a derivative of 2,4,5-T, 2,4,5-TME may possess similar biological activities and risks. As a lipophilic com-pound, 2,4,5-T has been found to accumulate in the adipose tissues of organisms.92 By using an exposomic approach, we identified significant associations between these three environmental exposures (e.g., ATQ, DMIP, and 2,4,5-TME) and body fat indicators. And, it is crucial to investigate the environmental abundance of such sub-stances in the atmosphere, determine exposure levels, and conduct quantitative analyses to improve risk assessment efforts. Additional research, including cohort studies and both in vivo and in vitro experiments, is essential to thoroughly evaluate the health risks linked to these exposures, as well as to understand the intricate inter-play among various pollutants.
Mixed Effects Analysis of Changes in Body Composition
Assessing the effects of the exposure on health effects is challeng-ing because the possibility of adverse antagonistic consequences resulting from a range of environmental exposures acting through similar pathways.93-95 The BKMR model is a powerful statistical tool for analyzing complex relationships between mixed exposures and health outcomes.63 Our BKMR results show that abiotic air-borne exposure was only significantly associated with VFA but was not associated with other body fat indicators, such as BFM, FMI, and PBF. Additionally, the trends of the univariate exposure- response of ATQ, DMIP, and 2,4,5-TME in BKMR were consistent with the results of the ExWAS. Furthermore, there are varying degrees of interaction observed between abiotic airborne contami-nants associated with VFA. This suggests that specific airborne contaminants may have a more pronounced effect on VFA com-pared to other body fat measures.
VFA refers to the amount of fat stored in the abdominal lumen, especially around the visceral organs such as the liver, pancreas, and intestine.96 It is a measure of central obesity and is strongly associated with multiple health problems such as cardiovascular disease, type 2 diabetes, and metabolic syndrome.96-98 Existing evidence has demonstrated that the distribution of regional fat, rather than overall fat amount, is closely linked to morbidity and mortality rates among obese individuals.99-102 Although the associ-ation between mixed exposure to air pollutants and VFA was not currently supported by epidemiological studies, animal experimen-tal studies have demonstrated that exposure to PAHs can lead to increased visceral adipose tissue and impaired systemic glucose tol-erance.103,104 Consequently, our results offered more proof that air-borne environmental variables contribute to the build-up of visceral adipose tissue.
Identification of the Key Lipid Intermediators
Lipidomics provides new insights into the exploration of the rela-tionship between exposure to airborne chemicals and changes in body composition. Mediation analysis was further performed to examine the effects of key abiotic airborne exposures (ATQ and DMIP) on the serum global lipid profile and their potential role in the increase of VFA. The results show that glycerophospholipids and sphingolipids may be involved in the increase of VFA. Among them, the mediating effects of SM(d18:2/24:1), PC(16:2e/18:0), and PC(16:1e/18:1) were more prominent. Our study indicates that certain lipid intermediators are linked to the increase in visceral fat accumulation induced by key pollutants.
Elevated plasma levels of sphingolipids can serve as bio-markers for chronic diseases (e.g., type 2 diabetes and coronary heart disease),105,106 highlighting their potential as valuable bio-markers. Glycerophospholipids, conversely, play a crucial role in cell membranes, signaling molecules, and systemic lipid homeo-stasis.107,108 Previous toxicological studies have provided insights into the biomolecular functions of these lipid intermediators (Table S9). For example, SM(d18:2/24:1) has emerged as a promising predictive marker for metabolic risk, playing a significant role in the development of metabolic disorder.109 Recent studies have highlighted the association between exposure to air pollution and increased plasma SM levels,110 and elevated plasma SM was significantly associated with obesity111,112; these results were consistent with those of our study. In addition, the impact of DMIP on PC(16:2e/18:0) and PC(16:1e/18:1) remains largely unexplored, as no studies have yet demonstrated a direct effect. Nevertheless, animal experiments have suggested that ex-posure to PAHs can lead to reduced serum PC levels.113 PC has been inversely associated with obesity-related complications and liver fat content.114 Evidently, lipid signaling pathways are complex, and understanding the integrated lipidomic networks and coordinated pathways remains a major research goal.107 Over the past decade, researchers have conducted extensive studies, but the mechanism by which lipid analysis mediates the occurrence of obesity is still unclear.108,115 Providing such lipid annotation results helps researchers understand the specific properties of particular lipid molecules and how they are associated with bio-logical processes, diseases, or metabolic pathways. Oxidative stress and inflammasome activation are possible mechanisms. Existing research results show that increased ROS activity can stimulate the production of sphingolipids, glycerophospholipids, and their metabolites.116,117 At the same time, the formation and activation of sphingolipids and inflammasomes are closely related (such as NLR3 inflammasome molecules), and body weight is directly correlated with the presence of NLRP3 inflam-masome units in adipose tissue.118
In this study, the application of multimodel cross-validation provided us with a more comprehensive perspective. The progressive nature of the results from each model ensured a higher level of credibility. This is crucial for a deeper understanding of the impact of airborne contaminants, particularly on lipid indicators, on human body composition. ExWAS and DSA models pro-vided comprehensive validation of the associations between ATQ, DMIP, 2,4,5-TME, and adiposity indicators in healthy older adults. The BKMR model suggested a positive combined effect of key abiotic airborne exposures on VFA in this population. Additionally, the trends observed in the univariate exposure- response analysis of ATQ and DMIP in the BKMR model were consistent with the results from the ExWAS analysis. The BKMR model also indicated the potential presence of interactions among abiotic airborne exposures related to adipose tissue, which may explain the decreasing trends observed in FMI and PBF in the mixed effects analysis. Therefore, future research should pay atten-tion to the interactions between gaseous chemicals in order to better understand their combined effects. Our research provides a basis for future exploration of significant airborne pollutant exposures, but the underlying mechanisms still require further validation through more rigorous toxicological experiments and integrated multi-omics analysis.
Strengths and Limitations
Our study has several notable strengths. First, to the best of our knowledge, this is the first study to systematically examine the relationship between the personal abiotic airborne exposure and changes in body composition within a healthy older population. Second, we used advanced exposome technologies and devices (e.g., Fresh Air wristband, MicroPEM, nontargeted analysis, and InBody 770) to accurately assessed the external exposures and measure body composition indicators in healthy older adults. Third, we implemented rigorous inclusion and exclusion criteria to minimize potential confounders such as unhealthy lifestyle habits (e.g., smoking and drinking) or the use of medication. All participants followed a standardized diet, reducing the impact of diet on body composition indicators. Fourth, we employed a state-of-art statistical strategy combining ExWAS analysis, DSA model, and the BKMR model to effectively reduce false discov-ery rate and explore the impact of abiotic airborne contaminants mixtures on body composition in healthy older adults. Last, we conducted mediation analysis to investigate the mediating effect of global lipid factors on key abiotic airborne exposures and VFA, providing valuable insights into the pathways through which these chemicals may affect fat distribution.
This study also has some limitations. First, despite the longitu-dinal panel design, the limited sample size and unmeasured con-founding factors may hinder our ability to establish causal relationships. Second, due to the relatively short period of observa-tion (i.e., 5 months),we were unable to assess long-term impacts of abiotic airborne exposures on body composition. Third, our study focused solely on the effects of nonbiological airborne exposures but not biotic factors (e.g., viruses and bacteria). Fourth, DSA results need to be interpreted with caution, as they cannot fully account for the causal relationship between exposure and outcome over time. Fifth, there was a failure to resolve the interactions among complex chemicals, and limitations exist in the mediation model analysis when addressing the intricate interactions among lipid molecules. Finally, the specific selection of healthy older adults in one city restricts the generalizability of our findings to other age groups or general populations.
Conclusions
The present study explored the independent and combined effects of individual abiotic airborne exposureonbody composition indicators in healthy older individuals. Our findings revealed that the specific PAH substances, DMIP and ATQ, primarily increased VFA by modifying glycerophospholipids and sphingolipids. These findings contribute to the growing body of evidence linking air pollution ex-posure with adverse effects on body composition in older individu-als. Furthermore, our study uncovered specific lipid intermediators [e.g., SM(d18:2/24:1), PC(16:2e/18:0), and PC(16:1e/18:1)] that are responsive to airborne chemicals and implicated in the altera-tions of body composition indicators. The results from this study may have important implications for identifying potentially harmful and widespread environmental chemicals and developing targeted strategies for control and prevention of chronic disease among the older adults in the future, particularly as the global population is rap-idly aging.
Acknowledgments
We thank all the participants of the China BAPE Study, the Dianliu Community, the Ankang Community Hospital, the Shandong CDC, the Jinan CDC, the China BAPE Study team, and the Metabolomics Facilities in Tsinghua University Protein Research Center.
This study was financially supported by the National Key Research and Development Program of China (No. 2022YFC3702700), the National Natural Science Foundation of China (No. 82025030), and the National Research Program for Key Issues in Air Pollution Control of China (No. DQGG0401) to X.S.
The data are not publicly available due to them containing information that might compromise research participant privacy/ consent.
References
1. Luo Y, Guo C, Wang Y, Zheng X. 2020. Trends and challenges for population health and migration-China, 2015-2050. China CDC Wkly 2(28):520-524, PMID: 34594693, https://doi.org/10.46234/ccdcw2020.141.
2. United Nations Department of Economic and Social Affairs Population Dynamics. 2019. World Population Prospects 2019 Revision. https://population. un.org/wpp2019/ [accessed 21 December 2023].
3. Jarzebski MP, Elmqvist T, Gasparatos A, Fukushi K, Eckersten S, Haase D, et al. 2021. Ageing and population shrinking: implications for sustainability in the urban century. NPJ Urban Sustain 1(1):17, https://doi.org/10.1038/s42949-021-00023-z.
4. United Nations Department of Economic and Social Affairs Population Dynamics. 2022. World Population Prospects 2022 Revision. https://population. un.org/wpp/ [accessed 24 December 2023].
5. Chakravarti D, LaBella KA, DePinho RA. 2021. Telomeres: history, health, and hallmarks of aging. Cell 184(2):306-322, PMID: 33450206, https://doi.org/10. 1016/j.cell.2020.12.028.
6. Forbes GB. 2012. Human Body Composition: Growth, Aging, Nutrition, and Activity. Berlin, Germany: Springer Science & Business Media.
7. Dittmar M, Reber H. 2002. Evaluation of different methods for assessing intra-cellular fluid in healthy older people: a cross-validation study. J Am Geriatr Soc 50(1):104-110, PMID: 12028254, https://doi.org/10.1046/j.1532-5415.2002.50015.x.
8. Marra M, Sammarco R, De Lorenzo A, Iellamo F, Siervo M, Pietrobelli A, et al. 2019. Assessment of body composition in health and disease using bioelectri-cal impedance analysis (BIA) and dual energy X-ray absorptiometry (DXA): a critical overview. Contrast Media Mol Imaging 2019:3548284, PMID: 31275083, https://doi.org/10.1155/2019/3548284.
9. IAEA (International Atomic Energy Agency). 2023. Nutrition and Ageing. https:// www.iaea.org/topics/nutrition-and-ageing [accessed24December 2023].
10. Bray GA, Krauss RM, Sacks FM, Qi L. 2019. Lessons learned from the POUNDS lost study: genetic, metabolic, and behavioral factors affecting changes in body weight, body composition, and cardiometabolic risk. Curr Obes Rep 8(3):262- 283, PMID: 31214942, https://doi.org/10.1007/s13679-019-00353-1.
11. Costello E, Goodrich J, Patterson WB, Rock S, Li Y, Baumert B, et al. 2022. Diet quality is associated with glucose regulation in a cohort of young adults. Nutrients 14(18):3734, PMID: 36145110, https://doi.org/10.3390/nu14183734.
12. Duque-Cartagena T, Mundstock E, Dala Bernardina Dalla M, Vontobel Padoin A, Cañon-Montañez W, Mattiello R. 2023. The role of environmental pollutants in body composition: systematic review and meta-analysis. Environ Res 228:115840, PMID: 37024033, https://doi.org/10.1016/j.envres.2023.115840.
13. Tung NT, Lee YL, Lin SY, Wu CD, Dung HB, ThuyTPC, et al. 2021. Associations of ambient air pollution with overnight changes in body composition and sleep-related parameters. Sci Total Environ 791:148265, PMID: 34119796, https://doi.org/10.1016/j.scitotenv.2021.148265.
14. Wang X, Karvonen-Gutierrez CA, Gold EB, Derby C, Greendale G, Wu X, et al. 2022. Longitudinal associations of air pollution with body size and composition in midlife women: the study of women's health across the nation. Diabetes Care 45(11):2577 2584, PMID: 36084038, https://doi.org/10.2337/dc22-0963.
15. Yang S, Hong F, Li S, Han X, Li J, Wang X, et al. 2023. The association between chemical constituents of ambient fine particulate matter and obesity in adults: a large population-based cohort study. Environ Res 231(pt2):116228, PMID: 37230219, https://doi.org/10.1016/j.envres.2023.116228.
16. Poursafa P, Moosazadeh M, Abedini E, Hajizadeh Y, Mansourian M, Pourzamani H, et al. 2017. A systematic review on the effects of polycyclic aromatic hydrocarbons on cardiometabolic impairment. Int J Prev Med 8:19, PMID: 28479961, https://doi.org/10.4103/ijpvm.IJPVM_144_17.
17. Vineis P, Robinson O, Chadeau-Hyam M, Dehghan A, Mudway I, Dagnino S. 2020. What is new in the exposome? Environ Int 143:105887, PMID: 32619912, https://doi.org/10.1016/j.envint.2020.105887.
18. Vrijheid M. 2014. The exposome: a new paradigm to study the impact of environment on health. Thorax 69(9):876-878, PMID: 24906490, https://doi.org/10. 1136/thoraxjnl-2013-204949.
19. Karlsson O, Rocklöv J, Lehoux AP, Bergquist J, Rutgersson A, Blunt MJ, et al.
2021. The human exposome and health in the anthropocene. Int J Epidemiol 50(2):378-389, PMID: 33349868, https://doi.org/10.1093/ije/dyaa231.
20. Wild CP. 2005. Complementing the genome with an "exposome": the outstand-ing challenge of environmental exposure measurement in molecular epidemiol-ogy. Cancer Epidemiol Biomarkers Prev 14(8):1847 1850, PMID: 16103423, https://doi.org/10.1158/1055-9965.EPI-05-0456.
21. Jiang C, Zhang X, Gao P, Chen Q, Snyder M. 2021. Decoding personal biotic and abiotic airborne exposome. Nat Protoc 16(2):1129-1151, PMID: 33437065, https://doi.org/10.1038/s41596-020-00451-8.
22. Thakur IS, Roy D. 2020. Environmental DNA and RNA as records of human exposome, including biotic/abiotic exposures and its implications in the assessment of the role of environment in chronic diseases. Int J Mol Sci 21(14):4879, PMID: 32664313, https://doi.org/10.3390/ijms21144879.
23. Kampa M, Castanas E. 2008. Human health effects of air pollution. Environ Pollut 151(2):362-367, PMID: 17646040, https://doi.org/10.1016/j.envpol.2007.06.012.
24. Vermeulen R, Schymanski EL, Barabási AL, Miller GW. 2020. The exposome and health: where chemistry meets biology. Science 367(6476):392-396, PMID: 31974245, https://doi.org/10.1126/science.aay3164.
25. Gao P, Shen X, Zhang X, Jiang C, Zhang S, Zhou X, et al. 2022. Precision environmental health monitoring by longitudinal exposome and multi-omics profil-ing. Genome Res 32(6):1199-1214, PMID: 35667843, https://doi.org/10.1101/gr. 276521.121.
26. Niedzwiecki MM, Walker DI, Vermeulen R, Chadeau-Hyam M, Jones DP, Miller GW. 2019. The exposome: molecules to populations. Annu Rev Pharmacol Toxicol 59:107-127, PMID: 30095351, https://doi.org/10.1146/annurev-pharmtox-010818-021315.
27. Okeme JO, Koelmel JP, Johnson E, Lin EZ, Gao D, Pollitt KJG. 2023. Wearable passive samplers for assessing environmental exposure to organic chemicals: current approaches and future directions. Curr Environ Health Rep 10(2):84-98, PMID: 36821032, https://doi.org/10.1007/s40572-023-00392-w.
28. Gao P. 2021. The exposome in the era of one health. Environ Sci Technol 55(5):2790-2799, PMID: 33417434, https://doi.org/10.1021/acs.est.0c07033.
29. Jiang C, Wang X, Li X, Inlora J, Wang T, Liu Q, et al. 2018. Dynamic human environmental exposome revealed by longitudinal personal monitoring. Cell 175(1):277-291, PMID: 30241608, https://doi.org/10.1016/j.cell.2018.08.060.
30. Schüssler-Fiorenza Rose SM, Contrepois K, Moneghetti KJ, Zhou W, Mishra T, Mataraso S, et al. 2019. A longitudinal big data approach for precision health. Nat Med 25(5)792-804, PMID: 31068711, https://doi.org/10.1038/s41591-019-0414-6.
31. Shevchenko A, Simons K. 2010. Lipidomics: coming to grips with lipid diver-sity. Nat Rev Mol Cell Biol 11(8):593-598, PMID: 20606693, https://doi.org/10. 1038/nrm2934.
32. Lagarde M, Géloën A, Record M, Vance D, Spener F. 2003. Lipidomics is emerging. Biochim Biophys Acta 1634(3):61, PMID: 14643793, https://doi.org/ 10.1016/j.bbalip.2003.11.002.
33. Wenk MR. 2005. The emerging field of lipidomics. Nat Rev Drug Discov 4(7):594-610, PMID: 16052242, https://doi.org/10.1038/nrd1776.
34. Quehenberger O, Armando AM, Brown AH, Milne SB, Myers DS, Merrill AH, et al. 2010. Lipidomics reveals a remarkable diversity of lipids in human plasma. J Lipid Res 51(11):3299-3305, PMID: 20671299, https://doi.org/10.1194/ jlr.M009449.
35. Rhee EP, Cheng S, Larson MG, Walford GA, Lewis GD, McCabe E, et al. 2011. Lipid profiling identifies a triacylglycerol signature of insulin resistance and improves diabetes prediction in humans. J Clin Invest 121(4):1402-1411, PMID: 21403394, https://doi.org/10.1172/JCI44442.
36. Stegemann C, Pechlaner R, Willeit P, Langley SR, Mangino M, Mayr U, et al. 2014. Lipidomics profiling and risk of cardiovascular disease in the prospec-tive population-based Bruneck study. Circulation 129(18):1821-1831, PMID: 24622385, https://doi.org/10.1161/CIRCULATIONAHA.113.002500.
37. Duchez AC, Fauteux-Daniel S, Sut C, Ebermeyer T, Heestermans M, Arthaud CA, et al. 2023. Bioactive lipids as biomarkers of adverse reactions associated with apheresis platelet concentrate transfusion. Front Immunol 14:1031968, PMID: 37138863, https://doi.org/10.3389/fimmu.2023.1031968.
38. Yu L, Liu W, Wang X, Ye Z, Tan Q, Qiu W, et al. 2022. A review of practical statis-tical methods used in epidemiological studies to estimate the health effects of multi-pollutant mixture. Environ Pollut 306:119356, PMID: 35487468, https://doi.org/ 10.1016/j.envpol.2022.119356.
39. Barrera-Gómez J, Agier L, Portengen L, Chadeau-Hyam M, Giorgis-Allemand L, Siroux V, et al. 2017. A systematic comparison of statistical methods to detect interactions in exposome-health associations. Environ Health 16(1):74, PMID: 28709428, https://doi.org/10.1186/s12940-017-0277-6.
40. Di Vincenzo O, Marra M, Di Gregorio A, Pasanisi F, Scalfi L. 2021. Bioelectrical impedance analysis (BIA) -derived phase angle in sarcopenia: a systematic review. Clin Nutr 40(5):3052-3061, PMID: 33183880, https://doi.org/ 10.1016/j.clnu.2020.10.048.
41. Aleixo GFP, Shachar SS, Nyrop KA, Muss HB, Battaglini CL, Williams GR. 2020. Bioelectrical impedance analysis for the assessment of sarcopenia in patients with cancer: a systematic review. Oncologist 25(2):170-182, PMID: 32043785, https://doi.org/10.1634/theoncologist.2019-0600.
42. Tang S, Li T, Fang J, Chen R, Cha Y, Wang Y, et al. 2021. The exposome in practice: an exploratory panel study of biomarkers of air pollutant exposure in Chinese people aged 60-69 years (China BAPE study). Environ Int 157:106866, PMID: 34525388, https://doi.org/10.1016/j.envint.2021.106866.
43. Ding E, Deng F, Fang J, Li T, Hou M, Liu J, et al. 2023. Association between or-ganophosphate ester exposure and insulin resistance with glycometabolic disorders among older Chinese adults 60-69 years of age: evidence from the China BAPE study. Environ Health Perspect 131(4):047009, PMID: 37042841, https://doi.org/10.1289/EHP11896.
44. Koelmel JP, Lin EZ, Guo P, Zhou J, He J, Chen A, et al. 2021. Exploring the external exposome using wearable passive samplers - The China BAPE study. Environ Pollut 270:116228, PMID: 33360595, https://doi.org/10.1016/j.envpol. 2020.116228.
45. Wang J, Li T, Fang J, Tang S, Zhang Y, Deng F, et al. 2022. Associations between individual exposure to fine particulate matter elemental constituent mixtures and blood lipid profiles: a panel study in Chinese people aged 60-69 years. Environ Sci Technol 56(18):13160-13168, PMID: 36043295, https://doi.org/ 10.1021/acs.est.2c01568.
46. Zhao L, Fang J, Tang S, Deng F, Liu X, Shen Y, et al. 2022. PM2.5 and serum metabolome and insulin resistance, potential mediation by the gut micro-biome: a population-based panel study of older adults in China. Environ Health Perspect 130(2):027007, PMID: 35157499, https://doi.org/10.1289/ EHP9688.
47. Shi W, Tang S, Fang J, Cao Y, Chen C, Li T, et al. 2022. Epigenetic age strati-fies the risk of blood pressure elevation related to short-term PM2.5 exposure in older adults. Environ Res 212(pt D):113507, PMID: 35636465, https://doi.org/ 10.1016/j.envres.2022.113507.
48. Guo P, Lin EZ, Koelmel JP, Ding E, Gao Y, Deng F, et al. 2021. Exploring perso-nal chemical exposures in China with wearable air pollutant monitors: a repeated-measure study in healthy older adults in Jinan, China. Environ Int 156:106709, PMID: 34153889, https://doi.org/10.1016/j.envint.2021.106709.
49. Lin EZ, Esenther S, Mascelloni M, Irfan F, Godri Pollitt KJ. 2020. The fresh air wristband: a wearable air pollutant sampler. Environ Sci Technol Lett 7(5):308- 314, https://doi.org/10.1021/acs.estlett.9b00800.
50. Du Y, Wang Q, Sun Q, Zhang T, Li T, Yan B. 2019. Assessment of PM2.5 moni-toring using MicroPEM: a validation study in a city with elevated PM2.5 levels. Ecotoxicol Environ Saf 171:518-522, PMID: 30641312, https://doi.org/10.1016/j. ecoenv.2019.01.002.
51. Fang J, Tang S, Zhou J, Zhou J, Cui L, Kong F, et al. 2020. Associations between personal PM2.5 elemental constituents and decline of kidney function in older individuals: the China BAPE study. Environ Sci Technol 54(20):13167-13174, PMID: 32929958, https://doi.org/10.1021/acs.est.0c04051.
52. Kyle UG, Bosaeus I, De Lorenzo AD, Deurenberg P, Elia M, Gómez JM, et al. 2004. Bioelectrical impedance analysis-part I: review of principles and meth-ods. Clin Nutr 23(5):1226-1243, PMID: 15380917, https://doi.org/10.1016/j.clnu. 2004.06.004.
53. Wilmanski T, Rappaport N, Earls JC, Magis AT, Manor O, Lovejoy J, et al. 2019. Blood metabolome predicts gut microbiome a-diversity in humans. Nat Biotechnol 37(10):1217-1228, PMID: 31477923, https://doi.org/10.1038/s41587-019-0233-9.
54. Huynh K, Barlow CK, Jayawardana KS, Weir JM, Mellett NA, Cinel M, et al.
2019. High-throughput plasma lipidomics: detailed mapping of the associa-tions with cardiometabolic risk factors. Cell Chem Biol 26(1):71-84, PMID: 30415965, https://doi.org/10.1016/j.chembiol.2018.10.008.
55. Shi W, Gao X, Cao Y, Chen Y, Cui Q, Deng F, et al. 2022. Personal airborne chemical exposure and epigenetic ageing biomarkers in healthy Chinese el-derly individuals: evidence from mixture approaches. Environ Int 170:107614, PMID: 36375280, https://doi.org/10.1016/j.envint.2022.107614.
56. Vrijheid M, Fossati S, Maitre L, Márquez S, Roumeliotaki T, Agier L, et al. 2020. Early-life environmental exposures and childhood obesity: an exposome-wide approach. Environ Health Perspect 128(6):067009, PMID: 32579081, https://doi.org/ 10.1289/EHP5975.
57. Cai M, Li H, Wu Y, Zhang S, Wang X, Zhang Z, et al. 2022. Ambient air pollution associated with body fat percentages at different body compartments: a cohort study of UK biobank participants. Environ Health Perspect 130(6):067702, PMID: 35767013, https://doi.org/10.1289/EHP10920.
58. Chen Z, Newgard CB, Kim JS, Ii O, Alderete TL, Thomas DC, et al. 2019. Near-roadway air pollution exposure and altered fatty acid oxidation among ado-lescents and young adults - the interplay with obesity. Environ Int 130:104935, PMID: 31238265, https://doi.org/10.1016/j.envint.2019.104935.
59. Agier L, Portengen L, Chadeau-Hyam M, Basagaña X, Giorgis-Allemand L, Siroux V, et al. 2016. A systematic comparison of linear regression-based sta-tistical methods to assess exposome-health associations. Environ Health Perspect 124(12):1848-1856, PMID: 27219331, https://doi.org/10.1289/EHP172.
60. Li MX, Yeung JM, Cherny SS, Sham PC. 2012. Evaluating the effective num-bers of independent tests and significant p-value thresholds in commercial genotyping arrays and public imputation reference datasets. Hum Genet 131(5):747-756, PMID: 22143225, https://doi.org/10.1007/s00439-011-1118-2.
61. Sinisi SE, van der Laan MJ. 2004. Deletion/substitution/addition algorithm in learning with applications in genomics. Stat Appl Genet Mol Biol 3:18, PMID: 16646796, https://doi.org/10.2202/1544-6115.1069.
62. Steen B. 1988. Body composition and aging. Nutr Rev 46(2):45-51, PMID: 3279343, https://doi.org/10.1111/j.1753-4887.1988.tb05386.x.
63. Bobb JF, Valeri L, Claus Henn B, Christiani DC, Wright RO, Mazumdar M, et al. 2015. Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures. Biostatistics 16(3):493-508, PMID: 25532525, https://doi.org/ 10.1093/biostatistics/kxu058.
64. Yu G, Jin M, Huang Y, Aimuzi R, Zheng T, Nian M, et al. 2021. Environmental exposure to perfluoroalkyl substances in early pregnancy, maternal glucose homeostasis and the risk of gestational diabetes: a prospective cohort study. Environ Int 156:106621, PMID: 33984575, https://doi.org/10.1016/j.envint.2021. 106621.
65. Albert JM. 2008. Mediation analysis via potential outcomes models. Stat Med 27(8):1282-1304, PMID: 17691077, https://doi.org/10.1002/sim.3016.
66. de Bont J, Hughes R, Tilling K, Díaz Y, de Castro M, Cirach M, et al. 2020. Early life exposure to air pollution, green spaces and built environment, and body mass index growth trajectories during the first 5 years of life: a large longitudi-nal study. Environ Pollut 266(pt 3):115266, PMID: 32745901, https://doi.org/10. 1016/j.envpol.2020.115266.
67. Starling AP, Moore BF, Thomas DSK, Peel JL, Zhang W, Adgate JL, et al.
2020. Prenatal exposure to traffic and ambient air pollution and infant weight and adiposity: the healthy start study. Environ Res 182:109130, PMID: 32069764, https://doi.org/10.1016/j.envres.2020.109130.
68. Rundle AG, Gallagher D, Herbstman JB, Goldsmith J, Holmes D, Hassoun A, et al. 2019. Prenatal exposure to airborne polycyclic aromatic hydrocarbons and childhood growth trajectories from age 5-14 years. Environ Res 177:108595, PMID: 31352299, https://doi.org/10.1016/j.envres.2019.108595.
69. Rundle A, Hoepner L, Hassoun A, Oberfield S, Freyer G, Holmes D, et al. 2012. Association of childhood obesity with maternal exposure to ambient air polycy-clic aromatic hydrocarbons during pregnancy. Am J Epidemiol 175(11):1163- 1172, PMID: 22505764, https://doi.org/10.1093/aje/kwr455.
70. Li M, Qian Z, Vaughn M, Boutwell B, Ward P, Lu T, et al. 2015. Sex-specific dif-ference of the association between ambient air pollution and the prevalence of obesity in Chinese adults from a high pollution range area: 33 communities Chinese health study. Atmospheric Environment 117:227-233, https://doi.org/10. 1016/j.atmosenv.2015.07.029.
71. Liu X, Tu R, Qiao D, Niu M, Li R, Mao Z, et al. 2020. Association between longterm exposure to ambient air pollution and obesity in a Chinese rural popula-tion: the Henan rural cohort study. Environ Pollut 260:114077, PMID: 32041030, https://doi.org/10.1016/j.envpol.2020.114077.
72. Alver K, Meyer HE, Falch JA, Søgaard AJ. 2010. Outdoor air pollution, bone density and self-reported forearm fracture: the Oslo health study.
Osteoporos Int 21(10): 1751 1760, PMID: 20077108, https://doi.org/10.1007/ s00198-009-1130-8.
73. Sadiktsis I, de Oliveira Galvão MF, Mustafa M, Toublanc M, Ünlü Endirlik B, Silvergren S, et al. 2023. A yearlong monitoring campaign of polycyclic aromatic compounds and other air pollutants at three sites in Sweden: source identification, in vitro toxicity and human health risk assessment. Chemosphere 332:138862, PMID: 37150457, https://doi.org/10.1016/j.chemosphere. 2023.138862.
74. Andersson JT, Achten C. 2015. Time to say goodbye to the 16 EPA PAHs? Toward an up-to-date use of PACs for environmental purposes. Polycycl Aromat Compd 35(2-4):330-354, PMID: 26823645, https://doi.org/10.1080/ 10406638.2014.991042.
75. Chen L, Liu W, Tao S, Liu W. 2021. Spatiotemporal variations and source identification of atmospheric nitrated and oxygenated polycyclic aromatic hydrocarbons in the coastal cities of the Bohai and Yellow seas in Northern China. Chemosphere 279:130565, PMID: 33866095, https://doi.org/10.1016/j.chemosphere. 2021.130565.
76. Clergé A, Le Goff J, Lopez C, Ledauphin J, Delépée R. 2019. Oxy-PAHs: occur-rence in the environment and potential genotoxic/mutagenic risk assessment for human health. Crit Rev Toxicol 49(4):302-328, PMID: 31512557, https://doi.org/ 10.1080/10408444.2019.1605333.
77. Zhu J, Sheng M, Shang J, Kuang Y, Shi X, Qiu X. 2022. Photocatalytic role of atmospheric soot particles under visible-light irradiation: reactive oxygen species generation, self-oxidation process, and induced higher oxidative potential and cytotoxicity. Environ Sci Technol 56(12)7668-7678, PMID: 35537182, https://doi.org/10.1021/acs.est.2c00420.
78. Degrendele C, Kanduc T, Kocman D, Lammel G, Cambelová A, Dos Santos SG, et al. 2021. NPAHs and OPAHs in the atmosphere of two Central European cities: seasonality, urban-to-background gradients, cancer risks and gas-to-particle par-titioning. Sci Total Environ 793:148528, PMID: 34328964, https://doi.org/10.1016/j. scitotenv.2021.148528.
79. Lim H, Silvergren S, Spinicci S, Mashayekhy Rad F, Nilsson U, Westerholm R, et al. 2022. Contribution of wood burning to exposures of PAHs and oxy-PAHs in Eastern Sweden. Atmos Chem Phys 22(17):11359 11379, https://doi.org/10. 5194/acp-22-11359-2022.
80. Li X, Zhao H, Qu B, Tian Y. 2022. Photoformation of environmentally persistent free radicals on particulate organic matter in aqueous solution: role of anthracene and formation mechanism. Chemosphere 291 (pt 1):132815, PMID: 34752830, https://doi.org/10.1016/j.chemosphere.2021.132815.
81. Titaley IA, Simonich SLM, Larsson M. 2020. Recent advances in the study of the remediation of polycyclic aromatic compound (PAC)-contaminated soils: transformation products, toxicity, and bioavailability analyses. Environ Sci Technol Lett 7(12):873-882, PMID: 35634165, https://doi.org/10.1021/acs.estlett. 0c00677.
82. Idowu O, Semple KT, Ramadass K, O'Connor W, Hansbro P, Thavamani P.
2019. Beyond the obvious: environmental health implications of polar polycyclic aromatic hydrocarbons. Environ Int 123:543-557, PMID: 30622079, https://doi.org/10.1016/j.envint.2018.12.051.
83. Lammel G, Kitanovski Z, Kukucka P, Novák J, Arangio AM, Codling GP, et al.
2020. Oxygenated and nitrated polycyclic aromatic hydrocarbons in ambient air-levels, phase partitioning, mass size distributions, and inhalation bioaccessibil-ity. Environ Sci Technol 54(5):2615-2625, PMID: 31950831, https://doi.org/10.1021/ acs.est9b06820.
84. Funk MI, Conde MA, Piwien-PilipukG, Uranga RM. 2021. Novel antiadipogenic effect of menadione in 3T3-L1 cells. Chem Biol Interact 343:109491, PMID: 33945810, https://doi.org/10.1016/j.cbi.2021.109491.
85. Shang Y, Fan L, Feng J, Lv S, Wu M, Li B, et al. 2013. Genotoxic and inflamma-tory effects of organic extracts from traffic-related particulate matter in human lung epithelial A549 cells: the role of quinones. Toxicol In Vitro 27(2):922-931, PMID: 23333790, https://doi.org/10.1016/j.tiv.2013.01.008.
86. Ravit B, Cooper K, Buckley B, Yang I, Deshpande A. 2019. Organic compounds associated with microplastic pollutants in New Jersey, USA surface waters. AIMS Environmental Science 6(6):445-459, https://doi.org/10.3934/environsci. 2019.6.445.
87. Mallah MA, Changxing L, Mallah MA, Noreen S, Liu Y, Saeed M, et al. 2022. Polycyclic aromatic hydrocarbon and its effects on human health: an overe-view. Chemosphere 296:133948, PMID: 35151703, https://doi.org/10.1016/j. chemosphere.2022.133948.
88. Yaar S, Filatova TS, England E, Kompella SN, Hancox JC, Bechtold DA, et al. 2023. Global air pollutant phenanthrene and arrhythmic outcomes in a mouse model. Environ Health Perspect 131 (11 ):117002, PMID: 37909723, https://doi.org/ 10.1289/EHP12775.
89. Al-Moubarak E, Shiels HA, Zhang Y, Du C, Hanington O, Harmer SC, et al.
2021. Inhibition of the hERG potassium channel by phenanthrene: a polycyclic aromatic hydrocarbon pollutant. Cell Mol Life Sci 78(23)7899-7914, PMID: 34727194, https://doi.org/10.1007/s00018-021-03967-8.
90. Wang Y, Zhu L, James-Todd T, Sun Q. 2022. Urinary polycyclic aromatic hydrocarbon excretion and regional body fat distribution: evidence from the U.S. National health and nutrition examination survey 2001-2016. Environ Health 21(1):75, PMID: 35945606, https://doi.org/10.1186/s12940-022-00890-8.
91. Liu C, Liu Q, Song S, Li W, Feng Y, Cong X, et al. 2023. The association between internal polycyclic aromatic hydrocarbons exposure and risk of obe-sity-a systematic review with meta-analysis. Chemosphere 329:138669, PMID: 37059208, https://doi.org/10.1016/j.chemosphere.2023.138669.
92. Schecter A, Dai LC, Thuy LT, Quynh HT, Minh DQ, Cau HD, et al. 1995. Agent Orange and the Vietnamese: the persistence of elevated dioxin levels in human tissues. Am J Public Health 85(4):516-522, PMID: 7702115, https://doi.org/10. 2105/ajph.85.4.516.
93. Mustieles V, Arrebola JP. 2020. How polluted is your fat? What the study of adipose tissue can contribute to environmental epidemiology. J Epidemiol Community Health 74(5):401-407, PMID: 32019765, https://doi.org/10.1136/jech-2019-213181.
94. Yilmaz B, Terekeci H, Sandal S, Kelestimur F. 2020. Endocrine disrupting chemicals: exposure, effects on human health, mechanism of action, models for testing and strategies for prevention. Rev Endocr Metab Disord 21(1):127- 147, PMID: 31792807, https://doi.org/10.1007/s11154-019-09521-z.
95. Braun JM, Gennings C, Hauser R, Webster TF. 2016. What can epidemiologi-cal studies tell us about the impact of chemical mixtures on human health? Environ Health Perspect 124(1):A6-A9, PMID: 26720830, https://doi.org/10. 1289/ehp.1510569.
96. Ryo M, Kishida K, Nakamura T, Yoshizumi T, Funahashi T, Shimomura I. 2014. Clinical significance of visceral adiposity assessed by computed tomography: a Japanese perspective. World J Radiol 6(7):409-416., PMID: 25071881, https://doi.org/10.4329/wjr.v6.i7.409.
97. Piché ME, Poirier P, Lemieux I, Després JP. 2018. Overview of epidemiology and contribution of obesity and body fat distribution to cardiovascular dis-ease: an update. Prog Cardiovasc Dis 61(2):103-113, PMID: 29964067, https://doi.org/10.1016/j.pcad.2018.06.004.
98. Piché ME, Tchernof A, Després JP. 2020. Obesity phenotypes, diabetes, and cardiovascular diseases. Circ Res 126(11):1477-1500, PMID: 32437302, https://doi.org/10.1161/CIRCRESAHA.120.316101.
99. Kissebah AH, Vydelingum N, Murray R, Evans DJ, Hartz AJ, Kalkhoff RK, et al. 1982. Relation of body fat distribution to metabolic complications of obesity. J Clin Endocrinol Metab 54(2):254-260, PMID: 7033275, https://doi.org/10.1210/ jcem-54-2-254.
100. Björntorp P. 1985. Obesity and the risk of cardiovascular disease. Ann Clin Res 17(1):3-9, PMID: 3893310.
101. Kaplan NM. 1989. The deadly quartet. Upper-body obesity, glucose intoler-ance, hypertriglyceridemia, and hypertension. Arch Intern Med 149(7):1514- 1520, PMID: 2662932, https://doi.org/10.1001/archinte.149.7.1514.
102. Després JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C. 1990. Regional distribution of body fat, plasma lipoproteins, and cardiovascular dis-ease. Arteriosclerosis 10(4):497-511, PMID: 2196040, https://doi.org/10.1161/01. atv.10.4.497.
103. Lee I, Park H, Kim MJ, Kim S, Choi S, Park J, et al. 2022. Exposure to polycy-clic aromatic hydrocarbons and volatile organic compounds is associated with a risk of obesity and diabetes mellitus among Korean adults: Korean national environmental health survey (KoNEHS) 2015-2017. Int J Hyg Environ Health 240:113886, PMID: 34864598, https://doi.org/10.1016/j.ijheh.2021.113886.
104. Jackson E, Shoemaker R, Larian N, Cassis L. 2017. Adipose tissue as a site of toxin accumulation. Compr Physiol 7(4):1085-1135, PMID: 28915320, https://doi.org/10.1002/cphy.c160038.
105. Yeboah J, McNamara C, Jiang XC, Tabas I, Herrington DM, Burke GL, et al.
2010. Association of plasma sphingomyelin levels and incident coronary heart disease events in an adult population: multi-ethnic study of atherosclerosis. Arterioscler Thromb Vasc Biol 30(3):628-633, PMID: 20032291, https://doi.org/ 10.1161/ATVBAHA.109.199281.
106. Haus JM, Kashyap SR, Kasumov T, Zhang R, Kelly KR, Defronzo RA, et al. 2009. Plasma ceramides are elevated in obese subjects with type 2 diabetes and correlate with the severity of insulin resistance. Diabetes 58(2):337-343, PMID: 19008343, https://doi.org/10.2337/db08-1228.
107. Quehenberger O, Dennis EA. 2011. The human plasma lipidome. N Engl J Med 365(19):1812-1823, PMID: 22070478, https://doi.org/10.1056/NEJMra1104901.
108. Wolfgang MJ. 2021. Remodeling glycerophospholipids affects obesity-related insulin signaling in skeletal muscle. J Clin Invest 131(8):e148176, PMID: 33855969, https://doi.org/10.1172/JCI148176.
109. Willinger CM, Yin X, Juhasz P, Courchesne P, Muntendam P, Gordon N, et al. 2014. Discovery lipidomic and metabolomic profiling predicts longitudi-nal changes in metabolic risk factors. Circulation 130(suppl 2):A18728, https://doi.org/10.1161/circ.130.suppl_2.18728.
110. Li R, Wang Y, Hou B, Lam SM, Zhang W, Chen R, et al. 2020. Lipidomics insight into chronic exposure to ambient air pollution in mice. Environ Pollut 262:114668, PMID: 32443199, https://doi.org/10.1016/j.envpol.2020.114668.
111. Hanamatsu H, Ohnishi S, Sakai S, Yuyama K, Mitsutake S, Takeda H, et al. 2014. Altered levels of serum sphingomyelin and ceramide containing distinct acyl chains in young obese adults. Nutr Diabetes 4(10):e141, PMID: 25329603, https://doi.org/10.1038/nutd.2014.38.
112. Yin X, Willinger CM, Keefe J, Liu J, Fernández-Ortiz A, Ibáñez B, et al. 2020. Lipidomic profiling identifies signatures of metabolic risk. EBioMedicine 51:102520, PMID: 31877415, https://doi.org/10.1016/j.ebiom.2019.10.046.
113. Li F, Xiang B, Jin Y, Li C, Li J, Ren S, et al. 2019. Dysregulation of lipid metabo-lism induced by airway exposure to polycyclic aromatic hydrocarbons in C57BL/6 mice. Environ Pollut 245:986-993, PMID: 30682755, https://doi.org/10. 1016/j.envpol.2018.11.049.
114. Wu G, Cheng H, Guo H, Li Z, Li D, Xie Z. 2023. Tea polyphenol EGCG amelio-rates obesity-related complications by regulating lipidomic pathway in leptin receptor knockout rats. J Nutr Biochem 118:109349, PMID: 37085056, https://doi.org/10.1016/j.jnutbio.2023.109349.
115. Boini KM, Xia M, Koka S, Gehr TW, Li PL. 2017. Sphingolipids in obesity and related complications. Front Biosci (Landmark Ed) 22(1):96-116, PMID: 27814604, https://doi.org/10.2741/4474.
116. Summers SA. 2006. Ceramides in insulin resistance and lipotoxicity. Prog Lipid Res 45(1):42-72, PMID: 16445986, https://doi.org/10.1016/j.plipres.2005.11. 002.
117. Boini KM, Xia M, Abais JM, Li G, Pitzer AL, Gehr TW, et al. 2014. Activation of inflammasomes in podocyte injury of mice on the high fat diet: effects of ASC gene deletion and silencing. Biochim Biophys Acta 1843(5):836-845, PMID: 24508291, https://doi.org/10.1016/j.bbamcr.2014.01.033.
118. Vandanmagsar B, Youm YH, Ravussin A, Galgani JE, Stadler K, Mynatt RL, et al.
2011. The NLRP3 inflammasome instigates obesity-induced inflammation and in-sulin resistance. Nat Med 17(2):179-188, PMID: 21217695, https://doi.org/10.1038/ nm.2279.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2024. This work is published under Reproduced from Environmental Health Perspectives (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
Background: Evidence suggested that abiotic airborne exposures may be associated with changes in body composition. However, more evidence is needed to identify key pollutants linked to adverse health effects and their underlying biomolecular mechanisms, particularly in sensitive older adults.Objectives: Our research aimed to systematically assess the relationship between abiotic airborne exposures and changes in body composition among healthy older adults, as well as the potential mediating mechanisms through the serum lipidome.Methods: From September 2018 to January 2019, we conducted a monthly survey among 76 healthy adults (60-69 years old) in the China Biomarkers of Air Pollutant Exposure (BAPE) study, measuring their personal exposures to 632 abiotic airborne pollutions using MicroPEM and the Fresh Air wristband, 18 body composition indicators from the InBody 770 device, and lipidomics from venous blood samples. We used an exposome-wide association study (ExWAS) and deletion/substitution/addition (DSA) model to unravel complex associations between exposure to contaminant mixtures and body composition, a Bayesian kernel machine regression (BKMR) model to assess the overall effect of key exposures on body composition, and mediation analysis to identify lipid intermediators.Results: The ExWAS and DSA model identified that 2,4,5-T methyl ester (2,4,5-TME), 9,10-Anthracenedione (ATQ), 4b,8-dimethyl-2-isopropyl-phenanthrene, and 4b,5,6,7,8,8a,9,10-octahydro-(DMIP) were associated with increased body fat mass (BFM), fat mass indicators (FMI), percent body fat (PBF), and visceral fat area (VFA) in healthy older adults [Bonferroni-Hochberg false discovery rate ðFDRBHÞ < 0:05]. The BKMR model demonstrated a positive correlation between contaminants (anthracene, ATQ, copaene, di-epi-a-cedrene, and DMIP) with VFA. Mediation analysis revealed that phosphatidylcholine [PC, PC(16:1e/18:1), PC(16:2e/18:0)] and sphingolipid [SM, SM(d18:2/24:1)] mediated a significant portion, rang-ing from 12.27% to 26.03% (p-value <0:05), of the observed increase in VFA.Discussion: Based on the evidence from multiple model results, ATQ and DMIP were statistically significantly associated with the increased VFA levels of healthy older adults, potentially regulated through lipid intermediators. These findings may have important implications for identifying potentially harmful environmental chemicals and developing targeted strategies for the control and prevention of chronic diseases in the future, partic-ularly as the global population is rapidly aging.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
1 China CDC Key Laboratory of Environment and Population Health, National Institute of Environmental Health, Chinese Center for Disease Control and Prevention, Beijing, China