International audience

PR | IFPRI3; ISI; CRP3.2; 4 Transforming Agricultural and Rural Economies | Development Strategies and Governance (DSG); Transformation Strategies | CGIAR Research Program on Maize (MAIZE)

Adipokines, cytokines secreted by adipose tissue, may contribute to obesity-related metabolic disease. The role of environmental phenols and parabens in racial difference in metabolic disease burden has been suggested, but there is limited evidence. We examined the cross-sectional associations of urinary phenols and parabens with adipokines and effect modification by race. Urinary concentrations of 6 phenols (bisphenol-A, bisphenol-F, 2,4-diclorophenol, 2,5-diclorophenol, triclosan, benzophenone-3) and 4 parabens (methyl-paraben, ethyl-paraben, propyl-paraben, butyl-paraben) were measured in 2002–2003 among 1200 women (mean age = 52.6) enrolled in the Study of Women's Health Across the Nation Multi-Pollutant Study. Serum adipokines included adiponectin, high molecular weight (HMW)-adiponectin, leptin, soluble leptin receptor (sOB-R). Linear regression models were used to estimate the adjusted percentage change in adipokines per inter-quantile range (IQR) increase in standardized and log-transformed levels of individual urinary phenols and parabens. Bayesian kernel machine regression (BKMR) was used to evaluate the joint effect of urinary phenols and parabens as mixtures. Participants included white (52.5%), black (19.3%), and Asian (28.1%) women. Urinary 2,4-dichlorophenol was associated with 6.02% (95% CI: 1.20%, 10.83%) higher HMW-adiponectin and urinary bisphenol-F was associated with 2.60% (0.48%, 4.71%) higher sOB-R. Urinary methyl-paraben was associated with lower leptin in all women but this association differed by race: 8.58% (−13.99%, −3.18%) lower leptin in white women but 11.68% (3.52%, 19.84%) higher leptin in black women (P interaction = 0.001). No significant associations were observed in Asian women. Additionally, we observed a significant positive overall effect of urinary phenols and parabens mixtures in relation to leptin levels in black, but not in white or Asian women. Urinary bisphenol-F, 2,4-dichlorophenol and methyl-paraben may be associated with favorable profiles of adipokines, but methyl-paraben, widely used in hair and personal care products, was associated with unfavorable leptin levels in black women. Future studies are needed to confirm this racial difference.

Exposure to metals has been linked with the risk of hypertensive disorders of pregnancy (HDP), but little is known about the potential effects of exposure to metal mixtures. Thus, our study aimed to investigated the impact of a complex mixture of metals on HDP, especially the interactions among metal mixtures. We did a population-based nested case-control study from October 2013 to October 2016 in Wuhan, China, including 146 HDP cases and 292 controls. Plasma concentrations of Aluminum (Al), Barium (Ba), Cobalt (Co), Copper (Cu), Lead (Pb), Mercury (Hg), Molybdenum (Mo), Nickel (Ni), Selenium (Se), Strontium (Sr), Thallium (Tl), and Vanadium (V) were measured and collected between 10 and 16 gestational weeks. We employed quantile g-computation, conditional logistic regression models, and Bayesian Kernel Machine Regression (BKMR) to assess the association of individual metals and metal mixtures with HDP risk. In the quantile g-computation, the OR for a joint tertile increase in plasma concentrations was 3.67 (95% CI: 1.70, 7.91). Hg contributed the largest positive weights and followed by Al, Ni, and V. In conditional logistic regression models, concentrations of Hg, Al, Ni, and V were significantly associated with the risk of HDP (p-FDR < 0.05). Compared to the lowest tertiles, the ORs (95% CI) for the highest tertiles of these four metals were 2.67 (1.44, 4.95), 3.09 (1.70, 5.64), 5.31 (2.68, 10.53), and 4.52 (2.26, 9.01), respectively. In the BKMR analysis, we observed a linear positive association between Hg, Al, V, and HDP, and a nonlinear relationship between Ni and HDP. A potential interaction between Al and V was also identified. We found that exposure to metal mixtures in early pregnancy, both individually and as a mixture, was associated with the risk of HDP. Potential interaction effects of Al and V on the risk of HDP may exist.

Carbonaceous aerosols pose significant climatic impact, however, their sources and respective contribution to light absorption vary and remain poorly understood. In this work, filter-based PM₂.₅ samples were collected in winter of 2021 at three urban sites in Yibin, a fast-growing city in the south of Sichuan Basin, China. The composition characteristics of PM₂.₅, light absorption and source of carbonaceous aerosol were analyzed. The city-wide average concentration of PM₂.₅ was 87.4 ± 31.0 μg/m³ in winter. Carbonaceous aerosol was the most abundant species, accounting for 42.5% of the total PM₂.₅. Source apportionment results showed that vehicular emission was the main source of PM₂.₅ during winter, contributing 34.6% to PM₂.₅. The light absorption of black carbon (BC) and brown carbon (BrC) were derived from a simplified two-component model. We apportioned the light absorption of carbonaceous aerosols to BC and BrC using the Least Squares Linear Regression with optimal angstrom absorption exponent of BC (AAEBC). The average absorption of BC and BrC at 405 nm were 51.6 ± 21.5 Mm⁻¹ and 17.7 ± 8.0 Mm⁻¹, respectively, with mean AAEBC = 0.82 ± 0.02. The contribution of BrC to the absorption of carbonaceous reached 26.1% at 405 nm. Based on the PM₂.₅ source apportionment and the mass absorption cross-section (MAC) value of BrC at 405 nm, vehicle emission was found to be the dominant source of BrC in winter, contributing up to 56.4%. Therefore, vehicle emissions mitigation should be the primary and an effective way to improve atmospheric visibility in this fast-developing city.

Environmental heavy metal exposure has been considered to be the risk factor for neurodevelopmental disorders in children. However, the available data on the associations between multiple metals exposure and the risk of dyslexia in China are limited. The purpose of our study was to examine the associations between urinary metal concentrations and Chinese dyslexia risk. A total of 56 Chinese dyslexics and 60 typically developing children were recruited. The urinary concentration of 13 metals were measured by inductively coupled plasma-mass spectrometer (ICP-MS). Binary logistic regression and the Probit extension of Bayesian kernel machine regression (BKMR-P) were used to explore the associations between multiple metal exposure and the risk of Chinese dyslexia. Our results indicated that Co, Zn and Pb were significantly associated with Chinese dyslexia in the multiple-metal exposure model. After adjusting the covariates, a positive association was observed between Pb and the risk of Chinese dyslexia, with the odds ratio (OR) in the highest quartiles of 6.81 (95%CI: 1.07–43.19; p–trend = 0.024). Co and Zn were negatively associated with the risk of Chinese dyslexia. Compared to the lowest quartile, the ORs of Co and Zn in the highest quartile are 0.13 (95%CI: 0.02–0.72; p–trend = 0.026) and 0.18 (95%CI: 0.04–0.88; p–trend = 0.038), respectively. In addition, BKMR-P analysis indicated that with the cumulative level across Co, Zn and Pb increased, the risk of Chinese dyslexia gradually declined and then rebounded, albeit non-significantly, and Pb was the major contributor in this association. In general, the urinary concentrations of Co, Zn and Pb were significantly associated with Chinese dyslexia. More prospective studies are needed to confirm the health effects of multiple metals exposure in children with Chinese dyslexia.

The association between polycyclic aromatic hydrocarbons (PAHs) exposure and hearing loss is rarely assessed. We aimed to evaluate the relationship of polycyclic aromatic hydrocarbons (PAHs) exposure and hearing loss among US adults and adolescents, and to explore the mediating role of systemic inflammation in the associations. Participants from the National health and Nutrition Examination Surveys (NHANES, 2001–2016) were included. Multiple logistic regression models were used to explore the associations between PAH metabolites and hearing loss. A total of 4200 adults and 1337 adolescents were included in the present analysis. For adults, we found positive association between urinary PAH metabolites and hearing loss, including total, speech-frequency and high-frequency hearing loss. The odds ratios (ORs) and 95% confidence intervals (CIs) for each one-unit increase in the log-transformed level of 3-Hydroxyfluorene (3-OHFlu), 2-Hydroxyfluorene (2-OHFlu) and 2 & 3-Hydroxyphenanthrene (2&3-OHPh) with total hearing loss were 1.17 (1.04–1.31), 1.24 (1.07–1.43), and 1.18 (1.03–1.37), respectively. For adolescents, urinary PAH metabolites were positively associated with total and speech-frequency hearing loss, not with high-frequency. The ORs and 95% CIs for each one-unit increase in the log-transformed level of 3-OHFlu, 2-OHFlu and total urinary PAH metabolites with total hearing loss were 1.34 (1.06–1.68), 1.48 (1.13–1.93), and 1.33 (1.04–1.72), respectively. Each one-unit increase in the log-transformed level 2-OHFlu (β = 0.112, 95%CI = 0.018–0.206) and 2&3-OHPh (β = 0.145, 95%CI = 0.037–0.253) were positively associated with C-reactive protein (CRP) among adolescents, but not among adults. No mediating effect for CRP on the association of urinary PAH metabolites with hearing loss was found (all P > 0.05). 3-OHFlu and 2-OHFlu are associated with increased prevalence of hearing loss among adults and adolescents. Systemic inflammation does not mediate the associations. Further studies should be conducted to verify the results.

The potential effect of gestational exposure to phthalates on the lung function levels during childhood is unclear. Therefore, we examined this association at different ages (from 4 to 11 years) and over the whole childhood. Specifically, we measured 9 phthalate metabolites (MEP, MiBP, MnBP, MCMHP, MBzP, MEHHP, MEOHP, MECPP, MEHP) in the urine of 641 gestating women from the INMA study (Spain) and the forced vital capacity (FVC), forced expiratory volume in 1 s (FEV₁) and FEV₁/FVC in their offspring at ages 4, 7, 9 and 11. We used linear regression and mixed linear regression with a random intercept for subject to assess the association between phthalates and lung function at each study visit and for the overall childhood, respectively. We also assessed the phthalate metabolites mixture effect on lung function using a Weighted Quantile Sum (WQS) regression. We observed that the phthalate metabolites gestational levels were consistently associated with lower FVC and FEV₁ at all ages, both when assessed individually and jointly as a mixture, although most associations were not statistically significant. Of note, a 10% increase in MiBP was related to lower FVC (−0.02 (−0.04, 0)) and FEV₁ z-scores (−0.02 (−0.04, −0.01) at age 4. Similar significant reductions in FVC were observed at ages 4 and 7 associated with an increase in MEP and MnBP, respectively, and for FEV₁ at age 4 associated with an increase in MBzP. WQS regression consistently identified MBzP as an important contributor to the phthalate mixture effect. We can conclude that the gestational exposure to phthalates was associated with children's lower FVC and FEV₁, especially in early childhood, and in a statistically significant manner for MEP, MiBP, MBzP and MnBP. Given the ubiquity of phthalate exposure and its established endocrine disrupting effects in children, our findings support current regulations that limit phthalate exposure.

Many cities across the world face the challenge of severe fine particulate matter (PM₂.₅) pollution. Among the many factors that affect PM₂.₅ pollution, there is an increasing interest in the impacts of urban structure. However, quantifying these impacts in China has been difficult due to differences of study area and scale in existing research, as well as limited sample sizes. Here, we conducted a continental study focusing on 301 prefectural cities in mainland China to investigate the effects of urban structure, including urban size and urban compactness, on PM₂.₅ concentrations. Based on PM₂.₅ raster and land cover data, we used quantile regression and a general multilinear model to estimate the effects and relative contributions of urban size and urban compactness on urban PM₂.₅ pollution, with explicit consideration for pollution level, urban size and geographical location. We found: (1) nationwide, the larger and more compact that cities were, the heavier the PM₂.₅ pollution tended to be. Additionally, this relationship became stronger with increasing levels of pollution. (2) In general, urban size played a more important role than urban form, and there were no significant interactive effects between the two metrics on urban PM₂.₅ concentrations at the national scale. (3) The impacts of urban size and form varied by city size and geographical location. The impacts of urban size were only significant for small or medium-large cities but not for large cities. Among large cities, only urban form had a significantly positive effect on urban PM₂.₅ concentrations. The further north and west that cities were, the more dependent PM₂.₅ pollution was on urban form, whereas the further south and east that cities were, the greater the impact of urban size. These results provide insights into how urban design and planning can be used to alleviate air pollution.

The inherent oxidation potential (OP) of atmospheric particulate matter has been shown to be an important metric in assessing the biological activity of inhaled particulate matter and is associated with the composition of PM₂.₅. The current study examined the chemical composition of 388 personal PM₂.₅ samples collected from students and guards living in urban and suburban areas of Beijing, and assessed the ability to predict OP from the calculated metrics of carcinogenic risk, represented by ELCR (excess lifetime cancer risk), non-carcinogenic risk represented by HI (hazard index), and the composition and sources of the particulate matter using multiple linear regression methods. The correlations between calculated ELCR and HI and the measured OP were 0.37 and 0.7, respectively. HI was a better predictor of OP than ELCR. The prediction models based on pollutants (Model_1) and pollution sources (Model_2) were constructed by multiple linear regression method, and Pearson correlation coefficients between the predicted results of Model_1 and Model_2 with the measured volume normalized OP are 0.81 and 0.80, showing good prediction ability. Previous investigations in Europe and North America have developed location-specific relationships between the chemical composition of particulate matter and OP using regression methods. We also examined the ability of relationships between OP and composition, sources, developed in Europe and North America, to predict the OP of particulate matter in Beijing from the composition and sources determined in Beijing. The relationships developed in Europe and North America provided good predictive ability in Beijing and it suggests that these relationships can be used to predict OP from the chemical composition measured in other regions of the world.