Skip to main content
NCBI home page
Environment & Health logoLink to Environment & Health
. 2024 Feb 27;2(5):290–300. doi: 10.1021/envhealth.3c00194

Effects of PM2.5 and Its Components on Disease Severity in Patients with Schizophrenia and the Mediating Role of Thyroid Hormones

Jintao Liu †,‡,§,∥, Rong Song †,‡,§,∥, Xuanxuan Li †,‡,§,∥, Li Liu †,‡,§,∥, Ning Wei †,‡,§,∥, Jiajun Yuan †,‡,§,∥, Weizhuo Yi †,‡,§,∥, Rubing Pan †,‡,§,∥, Jian Cheng †,‡,§,∥, Xulai Zhang ∥,*, Hong Su †,‡,§,∥,*
PMCID: PMC11504621  PMID: 39474322

Abstract

graphic file with name eh3c00194_0005.jpg

Studies have indicated the different effects of PM2.5 components on human health. However, specific components that influence the severity of disease in schizophrenia patients and their underlying mechanisms remain unclear. Therefore, a repeated measures study for schizophrenia was constructed based on Anhui Mental Health Center. We collected information, including demographics and thyroid hormones (TH) levels, on repeat admissions of schizophrenia patients during 2017–2020, assessing their illness severity by positive and negative symptom scales (PANSS). Concentrations from the nearest component monitoring station in the 3 months before admission were assigned as the participant’s exposure level. We assessed the effects of PM2.5 components individually and in combination on schizophrenia and explored the mediating role of THs. Results indicated that benzo[a]pyrene, sulfate, nitrate, chloride, ammonia, cadmium, chromium, lead, selenium, and thallium exposure were associated with increased PANSS scores, with more significant results observed in males. Mixed exposure to PM2.5 components was found to be associated with increased PANSS scores and decreased free triiodothyronine (FT3). Mediation analysis suggested that the reduction in FT3 might mediate the association between the PM2.5 components and PANSS scores. The findings emphasize the impacts of PM2.5 components on schizophrenia and the potential value of focusing on changes in THs.

Keywords: PM2.5 components, Schizophrenia, Thyroid hormones, Mediation analysis

1. Introduction

Schizophrenia is a severe mental disorder characterized by positive symptoms (delusions, hallucinations, and disorganized thinking) and negative symptoms (aphasia, social withdrawal, and emotional slowness).1 According to the Global Burden of Disease (GBD) report, the global number, prevalence, and disability-adjusted life-years (DAYL) of schizophrenia have increased by 65%, 15%, and 65% from 1990 to 2019, respectively.2 In China, the prevalence of schizophrenia reaches 3.87‰ [95% confidence interval (CI): 3.44‰, 4.30‰], resulting in 3.57 million person-years of DAYL, which accounts for 23.63% of the total global burden.3 The growing burden of diseases has prompted researchers to deepen their exploration of risk factors for schizophrenia. In recent years, the impact of controllable environmental factors has received much attention.

Air pollution has become the fourth largest risk factor for attributable human mortality.4 Research shows that particulate matter with an aerodynamic diameter ≤2.5 μm (PM2.5) dominates the health effects caused by air pollution in China.5 By summarizing the available research evidence, a recent meta-analysis indicates a significant effect of PM2.5 on schizophrenia.6 Nevertheless, contradictions in the association between PM2.5 and schizophrenia as well as differences in impact effects were demonstrated in the reviewed studies. This heterogeneity or contradiction may arise from the characteristics of PM2.5 as a complex mixture with differences in the sources and toxicological mechanisms of its different components.7 For example, exposure to PM2.5 components (e.g., elemental carbon, organic carbon, copper, cadmium, nickel, and zinc) was found to increase the risk of depression and anxiety in a multicenter cross-sectional study, whereas no such detrimental effects were observed for other components (e.g., potassium, silicon, arsenic, and lead).8 As schizophrenia is a severe mental illness, schizophrenia may be similarly affected differently by PM2.5 components. However, there is still no evidence of the association between PM2.5 components and schizophrenia, and the potential mechanism is still unclear.

Thyroid hormones (THs) are human essential hormones produced by negative feedback regulation through the hypothalamic-pituitary-thyroid (HPT) axis.9 Significant associations between THs and mental health have been confirmed by numerous studies, with evidence of an essential role for THs in infant neurodevelopment and adult neurotransmitter regulation.10,11 A meta-analysis indicates that schizophrenia shows lower total triiodothyronine (TT3) and higher thyroid-stimulating hormone (TSH) levels compared to normal individuals.12 Abnormal thyroid function has been demonstrated significantly associated with anxiety, depression, and schizophrenia,13−15 while normal-range THs changes have also been reported to be related to cognitive functioning and disease manifestations in patients with schizophrenia.16,17 Mendelian randomization evidence similarly highlights a causal association between thyroid impairment and schizophrenia.18 In addition, recent review evidence suggests that exposure to PM2.5 is significantly associated with changes in human TH levels.19 Therefore, we hypothesized that changes in THs may play a key role in the effects of PM2.5 and its component exposure on schizophrenia.

Based on a comprehensive consideration of the lack of evidence for PM2.5 components with schizophrenia and the need to explore potential mechanisms, this repeated measurement study based on a schizophrenia cohort aimed to explore the association between PM2.5 components with schizophrenia and to investigate the key contribution of THs in this association.

2. Methods

2.1. Study Design

Based on the schizophrenia-specific cohort from the Anhui Mental Health Center (the only mental illness-specialized hospital located in Hefei, Anhui Province), we constructed a repeated measurement study. Details of patients with schizophrenia were collected at this center in 2017 and their repeat admissions were followed up until 2020. Schizophrenia patients who met the following criteria were included in the analysis: (1) met the diagnostic criteria of International Classification of Diseases 10 (ICD-10, F20.0–20.9); (2) residents of Hefei with at least two admissions during the study period; and (3) more than three months between last hospital discharge and next hospital admission. Finally, a total of 266 (769 visits) schizophrenia patients were included in our study. The study was approved by the Ethics Committee of the Anhui Medical University.

2.2. Assessment of Exposure

The PM2.5 components data were obtained from two supermonitoring stations in Hefei (Baohe and Yaohai districts), and the main components contained 16 polycyclic aromatic hydrocarbons (PAHs): naphthalene (Nap), acenaphthylene (Acy), fluorene (Fluo), acenaphthene (Ace), phenanthrene (Phe), anthracene (Ant), fluoranthene (Fl), pyrene (Pyr), chrysene (Chry), benz[a]anthracene (B[a]A), benzo[b]fluoranthene (B[b]F), benzo[k]fluoranthene (B[k]F), benzo[a]pyrene (B[a]P), dibenz[a,h]anthracene (D[ah]A), benzo[g,h,i]perylene (B[ghi]P), and indeno[1,2,3-cd]pyrene (InP); four water-soluble ions: sulfate (SO42–), nitrate (NO3–), chloride (Cl–), and ammonium (NH4+); and ten metals: stibium (Sb), aluminum (Al), arsenic (As), cadmium (Cd), chromium (Cr), lead (Pb), manganese (Mn), nickel (Ni), selenium (Se), and thallium (Tl). The detailed monitoring program and methods for the measurements of the components can be found in Appendix 1 of the Supporting Information. We similarly collected data on PM2.5 from 20 other air quality monitoring stations in Hefei and found that the PM2.5 concentrations measured at the two component monitoring stations were highly correlated with those measured at the other 20 monitoring stations (Figure S1), which suggests that PM2.5 concentrations are spatially homogeneous in this local area. This is consistent with the spatial distribution observed in previous studies exploring the health effects of PM2.5 components,20 and we, therefore, drew on that study to reflect individual exposure levels in terms of concentrations at the nearest monitoring station of the study participants. Current research evidence suggests that there are different time spans for analyzing the deleterious window of PM2.5 effects on schizophrenia, ranging from a few days to more than a decade.6,21 A recent review indicated that the association between PM2.5 and THs was mostly focused on an exposure window of three months.19 Therefore, the PM2.5 and its components concentration levels of schizophrenia patients in the three months before each admission were used uniformly as the exposure. In addition, meteorological information such as daily temperature and relative humidity were obtained from the real-time product data set of the China Land Data Assimilation System-V2.0 (CLDAS-V2.0) of the China Meteorological Administration (http://data.cma.cn/). The data set is developed into an isotropic grid fusion analysis product covering the Asian region (0–65°N, 60–160°E) with 0.0625° × 0.0625° and 1 h resolution using multiple grid variational assimilation, optimal interpolation, probability density function matching, physical inversion, and topographic correction techniques. Data on gaseous pollutants were collected from the Department of Ecology and Environment of Anhui Province (https://sthjt.ah.gov.cn/index.html).

2.3. Assessment of Outcome and Mediation Variables

All patients with schizophrenia were assessed for illness by a specialist on admission using positive and negative symptom scales (PANSS). PANSS is a specialized psychiatric rating tool developed by Kay, Fiszbein, and Opler22 for assessing positive and negative syndromes, and the overall severity of the illness in schizophrenia. It has been continually modified and refined during its development to become more standardized to meet the requirements of the principles of schizophrenia measurement.23 In clinical practice, the PANSS has been used to assess disease change in patients with schizophrenia following the application of many clinical antipsychotic medications and treatment regimens.24−26 In addition, the results of the norm and factor analyses of the Chinese version of PANSS that we used were generally consistent with those reported by the original authors,27 and showed promising reliability and validity in practical applications (intraclass correlation coefficients and Cronbach alphas for the overall scores were all over 0.8).28 Therefore, the PANSS score is utilized as a key outcome to reflect the severity of illness in patients with schizophrenia, with higher scores indicating more severe illness. In addition, serum THs were tested in schizophrenia patients at each admission, including free thyroxine (FT4, pmol/L), free triiodothyronine (FT3, pmol/L), and thyroid-stimulating hormone (TSH, mU/L).

2.4. Assessment of Covariates

Information about the participant was obtained from the medical electronic case system, including age, gender (male or female), education level (less than high school, high school, and more than high school), occupation (employed, retired, and unemployed), marital status (married, unmarried, and divorced/widowed), smoking (yes or no), alcohol use (yes or no), season of admission to the hospital (spring, summer, fall, and winter), history of schizophrenia, use of antipsychotic drugs, and chronic conditions (e.g., diabetes, hypertension, etc.). The chronic conditions of schizophrenic patients were diagnosed (implementing ICD-10 standards) and treated during hospitalization through various targeted tests. Based on records of diagnosis and treatment of chronic diseases including hypertension, diabetes, chronic liver disease, hyperlipidemia, chronic obstructive pulmonary disease, etc. in medical records, any one of these is considered suffering from a chronic disease. Antipsychotic drug intake was converted to chlorpromazine equivalents according to drug type and dosage.29 In addition, the average daily temperature, relative humidity, and gaseous pollutants in the three months before admission were likewise taken into account as important factors for further analysis.

2.5. Statistical Analysis

Mean ± standard deviation and frequency (percentage) were used to statistically characterize continuous and subtype variables, respectively. Spearman correlation analysis was used to assess the correlation between PM2.5 components, as well as other environmental variables. As our data were in the form of individual repeated measurements, a linear mixed effects model (LMEM) was applied to explore the association between PM2.5 components and schizophrenia. PM2.5 components and individual covariates were included in the model as fixed-effect terms, while variation between individual repeated measurements was included as a random effect. We considered the effects of age, gender, education level, occupation, marital status, disease duration, chlorpromazine equivalents, chronic disease, smoking, alcohol consumption, season of admission, mean temperature, and relative humidity in our main models. Model results were expressed as the change (β) in schizophrenia PANSS scores and 95% CI for each increase in the interquartile range (IQR) of the PM2.5 components. In addition, we divided the sample by gender to see if there were differences between males and females.

To assess the mixing effect of significant components in PM2.5, we further constructed Bayesian and machine-regression (BKMR) models. The BKMR model was based on a nonparametric Bayesian variable selection framework and used kernel machine regression to sort out high-dimensional exposure-response relationships in multiple exposures, identifying nonlinearities and interactions.30 Given the repeated measurement design in this study, a random-effects term for individual repeated measurement characteristics was added to the BKMR model. Referring to previous similar studies,31 we included in the BKMR the key components found in LMEM that exerted an influence on illness severity in schizophrenic patients (either positive or negative effects) to assess mixed effects. The overall effect of PM2.5 components on schizophrenia was assessed based on a Markov chain Monte Carlo (MCMC) algorithm with 10,000 iterations. After fixing the PM2.5 component exposure level at a specific quantile (50th), the mixed effect was reflected by quantifying the change in results caused by each five percentile increase in the mixture. To ensure the consistency of the results, all covariate control strategies were consistent with the LMEM.

Identifying the underlying mechanisms of human health damage by environmentally hazardous substances has been a major topic of focus. One common way to quantify biological mechanisms is to construct causal mediation analyses based on strong a priori evidence.32 However, as the dimensionality of exposure increases, complex interactions and nonlinear effects of mixed substances also need to be introduced into the modeling of mediators and outcomes, which makes unbiased estimation of mediating effects extremely difficult.33 Here, we utilized the BKMR-causal mediation analysis (BKMR-CMA) framework to estimate the mediating role of THs in the mixed effect of PM2.5 components on schizophrenia.33 The method decomposes the total effect (TE) of the mixture into natural direct effects (NDE) and natural indirect effect (NIE). Furthermore, BKMR-CMA can simultaneously address important modifiers that influence the relationship between exposure, mediation, and outcome by fixing the variable at some level of practical significance and observing how the associations of interest change at different levels of modifiers. We estimate the TE, NDE, and NIE at fixed levels of the modifiers (gender in this study) and for a fixed intervention in the exposures through simulation of counterfactuals from the posterior predictive distribution for each MCMC iteration and make inferences from these posterior samples of the mediation effects. In this study, the mixed exposure factor was the PM2.5 components, the outcome factor was the PANSS scores of schizophrenia patients (reflecting the severity of their illness), and the mediating variables were the THs (FT3, FT4, TSH). Since BKMR-CMA cannot consider multiple mediators at the same time, we performed independent mediation analyses for different hormones. In addition, based on the differences between men and women found by the LMEM, we considered gender as a key factor in the BKMR-CMA to assess its potential modifying effects. Detailed information about BKMR-CMA and the implementation code can be found at https://zc2326.github.io/causalbkmr/index.html.

Several sensitivity analyses were constructed to test the stability of the findings. First, the component-residual model was used to avoid collinearity and to assess the independent contribution of each component. Second, evidence from systematic reviews suggests that gaseous pollutants (NO2, SO2) are associated with schizophrenia risk.6 Therefore, we controlled both pollutants in our model to exclude potential confounding effects. Third, to exclude the influence of those with abnormal thyroid function in the sample on the results, we removed participants with abnormal THs according to the laboratory reference range (FT3: 2.77–6.31 pmol/L, FT4: 10.44–24.38 pmol/L, TSH: 0.38–4.34 mIU/L). Fourth, considering the impact of smoking as a key source of PM2.5 exposure, we further excluded smokers and repeated the analysis.

All statistical analyses and plots were performed based on the “lme4″, “bkmr”, and “ggplot2″ packages in R (version 4.3.0). The significance test threshold was set at a two-sided P-value of less than 0.05. The false discovery rate (FDR) correction method was used to correct for significance testing of multiple comparisons.34

3. Results

3.1. Statistical Description

A total of 266 (769 visits) patients with schizophrenia were included in this study. Table 1 shows the characteristics of the subjects in 769 visits. Overall, the mean age of the subjects was 39.4 years and the mean PANSS score was 74.81. More than half (402, 52.27%) of the subjects were educated below high school, and the majority of patients (663, 86.22%) were currently unemployed. There were 134 (17.43%), 31 (4.03%), and 294 (38.23%) participant visits that reported the presence of smoking, alcohol consumption, and chronic diseases, respectively. The mean duration of schizophrenia was 12.43 years and the drug administration was 244.82 mg chlorpromazine equivalent. The mean levels of THs in study participants were 4.13 pmol/L for FT3, 15.10 pmol/L for FT4, and 2.22 mU/L for TSH. The concentration of PM2.5 during the study period ranged from 29.25 to 139.00 μg/m3, which was higher than the World Health Organization air quality limit (25 μg/m3). Among the components of PM2.5, B[a]F (0.95 ng/m3), NO3– (11.83 μg/m3), and Al (76.31 ng/m3) were the highest concentrations in PAHs, water-soluble ions, and metals, respectively (Table S1). Correlation analysis indicated a significant positive correlation between most of the components (Figure S2).

Table 1. Demographic Characteristics of Patients with Schizophreniaa.

  N (%) or (mean ± SD)
  
characteristics total male female P
number (visits) 769 (100%) 407 (52.93%) 362 (47.07%)  
age (years) 39.4 ± 12.16 36.42 ± 12.07 42.75 ± 11.37 <0.001
educational level       0.154
less than high school 402 (52.27%) 206 (50.61%) 196 (54.14%)  
high school 150 (19.51%) 90 (22.11%) 60 (16.57%)  
more than high school 217 (28.22%) 111 (27.27%) 106 (29.28%)  
occupation       <0.001
currently employed 37 (4.81%) 9 (2.21%) 28 (7.73%)  
currently unemployed 663 (86.22%) 354 (86.98%) 309 (85.36%)  
retirement 69 (8.97%) 44 (10.81%) 25 (6.90%)  
marital status       <0.001
unmarried 321(41.74%) 239 (58.72%) 82 (22.65%)  
married 363 (47.21%) 124 (30.47%) 239 (66.02%)  
divorced or widowed 85 (11.05%) 44 (10.81%) 41 (11.33%)  
smoking       <0.001
yes 134 (17.43%) 132 (32.43%) 2 (0.60%)  
no 635 (82.57%) 275 (67.57%) 360 (99.40%)  
alcohol consumption       <0.001
yes 31 (4.03%) 29 (7.12%) 2 (0.55%)  
no 738 (95.97%) 378 (92.88%) 360 (99.45%)  
chronic disease       0.032
yes 294 (38.23%) 158 (38.82%) 136 (37.57%)  
no 475 (61.77%) 249 (61.18%) 226 (62.43%)  
course of schizophrenia (years) 12.43 ± 9.35 11.11 ± 8.87 13.92 ± 9.66 <0.001
chlorpromazine equivalent (mg) 244.82 ± 230.69 255.45 ± 230.56 232.87 ± 230.56 0.176
PANSS scores 74.81 ± 12.27 75.18 ± 11.87 74.40 ± 12.71 0.378
FT3 (pmol/L) 4.13 ± 1.43 4.16 ± 1.49 4.10 ± 1.36 0.572
FT4 (pmol/L) 15.10 ± 2.98 15.28 ± 3.05 14.90 ± 2.88 0.042
TSH (mU/L) 2.22 ± 2.68 1.82 ± 1.42 2.66 ± 3.56 <0.001
Open in a new tab
a

SD, standard deviation; PANSS, positive and negative symptom scales; FT3, free triiodothyronine; FT4, free thyroxine; TSH, thyroid-stimulating hormone.

3.2. Association between PM2.5 and Its Components with PANSS Scores and THs in Schizophrenia

Figure 1 shows the association between PM2.5 and its components and PANSS scores in schizophrenia patients. Specifically, each increment in total PM2.5 mass concentration by an IQR (29.5 μg/m3) was associated with a 4.82 (95% CI: 2.96, 6.67) increase in PANSS scores. We also found that exposure to B[a]P (β = 2.06, 95% CI: 0.83, 3.30), SO42– (β = 1.64, 95% CI: 0.69, 2.59), NO3– (β = 2.96, 95% CI: 1.43, 4.48), NH4+ (β = 4.69, 95% CI: 2.89, 6.49), Cl– (β = 3.79, 95% CI: 1.36, 6.22), Cd (β = 3.74, 95% CI: 2.11, 5.37), Cr (β = 1.81, 95% CI: 0.46, 3.17), Pb (β = 3.19, 95% CI: 1.39, 4.99), Se (β = 2.03, 95% CI: 1.04, 3.01), and Tl (β = 3.22, 95% CI: 1.68, 4.77) were associated with increased PANSS scores in patients with schizophrenia (all FDR-adjusted P-values <0.05). In addition, we observed that the association between PM2.5 components and schizophrenia showed some differences across genders (Figure 2). For example, of the above components found to be significant in the total population, we found that B[a]P (β = 1.96, 95% CI: 0.33, 3.58), NO3– (β = 5.12, 95% CI: 3.00, 7.24), Cl– (β = 5.10, 95% CI: 2.08, 8.11), Cr (β = 2.59, 95% CI: 0.78, 4.41), Pb (β = 4.61, 95% CI: 2.37, 6.85), and Tl (β = 4.45, 95% CI: 2.48, 6.41) exposures were significantly associated with PANSS scores only in males rather than females. Overall, the association strength of most PM2.5 components with PANSS scores in schizophrenia patients was slightly higher in males than in females. Figure S3 illustrates the effect of PM2.5 and its components on TH levels in patients with schizophrenia. We found that total PM2.5 mass was associated with a decrease in FT4 (β = −0.83, 95% CI: −1.36, −0.35) and FT3 (β = −1.06, 95% CI: −1.29, −0.83). Exposure to B[a]P, Sb, Pb, Mn, and Tl was associated with reduced levels of FT4. Most of the water-soluble ions (NO3–, NH4+, and SO42–) and metal fractions (Sb, Al, As, Cd, Pb, Mn, Se, and Tl) showed negative associations with FT3. However, we did not find any adjusted significant associations between PM2.5 and its components with TSH. In addition, we found a negative association between THs and PANSS scores, with PANSS scores changing by −0.83 (95% CI: −1.41, −0.24) and −0.37 (95% CI: −0.65, −0.09) per unit increase in FT3 and FT4, respectively (Table S2).

Figure 1.

Figure 1

Open in a new tab

Estimates (βs) and 95% CIs of PANSS scores for schizophrenia patients according to each interquartile range (IQR) increase in PM2.5 and its components. Red asterisks indicate P < 0.05 for false discovery rate (FDR) correction.

Figure 2.

Figure 2

Open in a new tab

Estimates (βs) and 95% CIs of PANSS scores for gender-specific schizophrenia patients according to each interquartile range (IQR) increase in PM2.5 and its components. Red asterisks indicate P < 0.05 for false discovery rate (FDR) correction.

3.3. Mixing Effects of PM2.5 Components Found by BKMR Model

In order to exclude the nonlinearity and additivity between individual components, BKMR models were constructed to explore the mixing effect of PM2.5 components. As shown in Figure 3, an increase in the mixed concentration of PM2.5 components was positively associated with an increase in PANSS scores in schizophrenia patients. In the association of PM2.5 component mixtures with THs, we found that FT3 tended to decrease gradually with increasing mixture concentration. In addition, we did not find a significant association between PM2.5 component mixtures and FT4 and TSH levels in schizophrenia patients.

Figure 3.

Figure 3

Open in a new tab

Combined effects of significant PM2.5 components mixture on PANSS scores, FT3, FT4, and TSH in schizophrenia patients estimated by Bayesian kernel machine regression.

3.4. Mediation Effect of THs on the Associations of PM2.5 and Its Components with Schizophrenia

Table S3 presents the results of the mediation analysis of THs between individual PM2.5 components and schizophrenia. We found that FT3 and FT4 significantly mediated the effects of B[a]P, NO3–, NH4+, SO42–, Mn, Ni, Se, Tl, and Pb on the severity of schizophrenia, with the mediation proportion ranging from 3.64% to 20.01%. Figure 4 shows the mediating role of THs in mixed exposures obtained from BKMR-CMA. When gender was fixed at male, we found that FT3 partially mediated the effect of PM2.5 significant constituents mixture exposure on disease severity in schizophrenia patients, with a TE of 4.59 (95% CI: 3.01, 6.18), a NDE of 3.38 (95% CI: 1.75, 5.02), a NIE of 1.21 (95% CI: 0.03, 2.39), and the mediating proportion amounted to 26.36% (P < 0.05). In addition, no significant mediating associations were found in the analyses with FT4 and TSH as mediating variables in either males or females.

Figure 4.

Figure 4

Open in a new tab

Mediating effects of THs in the association between PM2.5 component mixtures and PANSS scores in schizophrenia. Red asterisks indicate P < 0.05.

3.5. Sensitivity Analysis

To assess the stability of the findings, we constructed several sensitivity strategies were constructed. First, we used component-residual modeling to evaluate the independent effects of each component after controlling for PM2.5. The results showed that the effect of PM2.5 components on the severity of schizophrenia did not change significantly compared to the original model (Table S4). Second, the effects of PM2.5 and its components on schizophrenia remained after the interference of gaseous pollutants (NO2 and SO2) was excluded (Table S5). Third, the findings were stable after excluding schizophrenia patients with abnormal THs (658 visits remaining) (Table S6). Fourth, analyses conducted in groups that excluded smokers (635 visits remaining) indicated no significant changes in the effects of most components (Table S7).

4. Discussion

In the current repeated measures study, we investigated the effect of PM2.5 components on schizophrenia and explored the mediating role of THs in this association. Overall, we found that exposure to B[a]P, SO42–, NO3–, Cl–, NH4+, Cd, Cr, Pb, Se, and Tl was associated with schizophrenia. The BKMR model showed a significant effect of PM2.5 component mixtures on schizophrenia disease severity. Furthermore, by constructing a BKMR-CMA, we identified the mediating role of FT3 in the association between the PM2.5 component mixture and schizophrenia. Considering the increasing disease burden of schizophrenia, our evidence may be instructive in future targeting to prevent or reduce the risk of schizophrenia caused by PM2.5.

Among the many PAHs components, only B[a]P was observed to have a significant hazardous effect on schizophrenia. B[a]P is produced during the incomplete combustion of organic materials in industrial and domestic processes and has been recognized by the International Agency for Research on Cancer (IARC) as a carcinogen and mutagen.35 Emerging evidence suggests that low-dose B[a]P exposure induces neurotoxicity earlier than carcinogenicity and teratogenicity.36 Animal studies have found that zebrafish larvae exposed to B[a]P develop neurobehavioral changes, such as lethargy and anxiety-like behavior in adulthood.37 Early postnatal B[a]P exposure in rats causes persistent neurobehavioral impairments that emerge postnatally and continue into adolescence and adulthood.38 Population epidemiologic evidence indicated high prenatal B[a]P exposure, whether characterized by personal air monitoring or maternal and cord adducts, was positively associated with symptoms of anxiety/depression and attention problems.39 Occupational B[a]P exposure has also been reported to be significantly associated with impaired learning and memory in coke oven workers.40 Our findings further add to the population evidence for the hazardous effects of B[a]P on mental health, which emphasizes the need for effective protection against this pollutant in the future.

Water-soluble ions are important components of PM2.5 and have been identified in recent years by a large body of evidence as a hazardous factor in the progression of respiratory and cardiovascular disease.41−43 However, only a handful of studies have reported on the association between these components and neurological and mental health. A large prospective cohort study from China found that long-term exposure to NH4+ and SO42– in PM2.5 was associated with depression risk.44 NO3– and SO42– exposure has also been shown to increase the risk of dementia in a cohort of the elderly population in the United States.45 In addition to long-term exposures, short-term exposure to SO42–, and NO3– has been associated with an increased risk of psychiatric hospitalization.46 In the present study, all four water-soluble ions detected were found to be significantly associated with schizophrenia. Although the mechanisms underlying these associations have not yet been elucidated, it has been suggested that exposure to harmful ions (e.g., NO3–) triggering increased oxidative damage in vivo may be a potentially explainable cause of mental health harm.47 There is also evidence from animal studies that NH4+ induces astrocyte swelling and free radical production in rats, which may be related to neurotoxicity.48

We found significant positive effects of Pb, Cd, Cr, Se, and Tl on the severity of schizophrenia, which is consistent with previous evidence that metallic exposures affect mental health. Cohort studies have indicated that increased blood Pb levels in children lead to elevated general psychopathological symptoms, including internalizing symptoms and thought disorders.49 Evidence from the National Health and Nutrition Examination Survey has previously reported positive associations between serum Cd and Cr exposure and depression.50,51 Exposure to Se has also been found to be associated with an increased risk of cognitive impairment in older adults.52 Animal experiments revealed that heavy metal Tl can induce mitochondrial dysfunction and neuronal shortening resulting in neurotoxicity.53 A recent systematic review emphasized the differences in the concentrations of these elements in schizophrenia compared to normal subjects and hypothesized that it is crucial regarding the onset and exaggeration of psychotic dysfunctions.54 Our study identified several key hazard components (Pb, Cd, Cr, Se, and Tl) that would have an impact on schizophrenia, which further affirmed this hypothesis. In the future, targeted control of the production of these harmful substances through precise traceability will be of significance for the protection of the public’s mental health.

The specific biological pathways underlying the effects of harmful factors on clinical outcomes have been the focus of research in the field of environmental epidemiology. In the present study, we found a mediating role for changes in FT3 and FT4 in the association between PM2.5 components exposure and schizophrenia. Of the total THs in the body, the majority are bound to plasma proteins, while the unbound free fraction (FT4 and FT3) can be taken up by different types of cells to perform regulatory actions.55 There is evidence that FT4 and FT3 levels in patients with schizophrenia showed positive associations with their cognitive function, attention, and vigilance.16,17 A large survey of 81,761 participants (40,843 hypothyroid patients, 40,918 healthy controls) from Israel found an independent positive association between hypothyroidism and schizophrenia.56 FT4 is converted in vivo by deiodinase to FT3, the main active form of THs that can interact with glial cells that regulate the immune response, modulate neurotransmitter release, and control neuronal metabolism.57 The effects of FT3 on adult neurogenesis occur primarily in two regions of the brain, the subventricular and subgranular regions, which are commonly associated with cognitive deficits, psychiatric disorders, and depression.58,59 Decreased levels of THs have been demonstrated to reduce microglial processes in postnatal rats, whereas daily injections of T3 promote the survival of microglial cells and the growth of their processes.60 Our findings regarding the reduction of FT3 levels by PM2.5 components mixed exposure further suggest the possibility of altered TH levels as a key factor in the impact of PM2.5 components on mental health. Although clear causal inferences could not be provided, the significant mediating effect may also provide some implications for future prevention of mental health impairment by PM2.5 components.

Another important finding is that males with schizophrenia appear to be more susceptible to the PM2.5 component than females. However, the clear mechanism for this gender difference remains unclear. Some scholars have argued that males may be more susceptible to biological risk factors for schizophrenia due to slower prenatal maturation of the cerebral cortex and excessive synaptic pruning during adolescence.61 At the same time, owing to differences in job characteristics, more exposure to outdoor or other high air pollution environments makes males suffer more from the persistent effects of PM2.5.62 Some animal models indicate that estrogen has significant psychoprotective properties, showing that women have milder symptoms and better antipsychotic drug responses.61,63 Not only that, but current evidence suggests that FT3 may likewise have sex-dependent effects, as it has been demonstrated that exposure of male and female mice to THs results in different states of glial cells (activation in males and inactivation in females).57 THs modulate different brain signaling and regulate key genes in a male- and female-specific manner may account for the differences in THs-mediated effects across sexes we have found.64

A primary strength of this study is the provision of the first research evidence of an association between PM2.5 components and schizophrenia. Besides, BKMR and BKMR-CMA were utilized as emerging methods to explore the effects of PM2.5 component mixture exposure on schizophrenia, and the mediating role of THs can provide some implications for future precision interventions. However, several research limitations still cannot be ignored. First, due to differences in environmental emissions, building types, wind speeds, as well as individual indoor and outdoor activities, there may be some differences in the concentrations of specific components between regions, and it is, therefore, unavoidable to introduce an exposure bias by using only the two monitoring stations in Hefei City to represent individual exposure levels. It would be worthwhile to further increase the number and coverage of environmental monitoring stations in the future. Combined with land use information to provide a refined assessment of exposure levels at specific locations. More precise individual monitoring also deserves to be applied in further studies. Second, although repeated measures cover a certain degree of temporal attributes, we have to recognize the limitations of statistically derived mediation effects in making firm causal inferences. Not only that, but we were similarly unable to fully specify the temporal order of THs testing and PANSS assessment for each schizophrenic patient, which somewhat limited our inferences about potential mediating effects. Nevertheless, based on our findings, further cohort or pilot studies to explore the key role of THs in the development of schizophrenia after exposure to PM2.5 and its components are still full of practical implications. Third, although we controlled for confounding factors as much as possible, potentially unknown factors affecting the study results still cannot be ignored.

5. Conclusions

Our study shows that PM2.5 and its components (B[a]P, SO42–, NO3–, NH4+, Cl–, Cd, Cr, Pb, Se, and Tl) are associated with the severity of schizophrenia disease, and this association is more pronounced in males. Changes in FT3 and FT4 levels may play a mediating role in the effects of PM2.5 component exposure on schizophrenia. Overall, our findings emphasize the potentially harmful effects of PM2.5 and its components on mental health. Reducing the concentration of harmful PM2.5 components and paying timely attention to changes in TH levels after exposure may be of value in protecting the health of patients with schizophrenia.

Acknowledgments

This study was funded by the National Natural Science Foundation of China (Grant No. 42375184), the Research Funds of Center for Big Data and Population Health of IHM (Grant No. JKS2022011), and the Natural Science Research Key Project of University of Anhui Province (Grant No. 2023AH050652).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/envhealth.3c00194.

  • Sampling and detailed detection methods for PM2.5 components; statistical description of the concentrations of PM2.5 and its constituents and meteorological parameters; association between THs and PANSS scores in patients with schizophrenia; mediating role of thyroid hormones in the effects of single component exposure on schizophrenia; results of sensitivity analysis; location of ambient air pollution monitoring stations in Hefei and distribution of study subjects; correlation analysis between environmental exposure variables; estimates (βs) and 95% CIs of thyroid hormones for schizophrenia patients according to each IQR increase in PM2.5 and its components (PDF)

Author Contributions

The contributions of the two corresponding authors are equal. The study was jointly completed by School of Public Health at Anhui Medical University and Anhui Mental Health Center. Prof. Xulai Zhang’s group conducted the field work and collected the data, and Prof. Hong Su’s group conducted the data analysis and manuscript writing. Jintao Liu, Song Rong: Conceptualization, Methodology, Data curation, Formal analysis, Writing - original draft, Writing - review and editing; Xuanxuan Li, Li Liu: Data curation, Formal analysis, Visualization, Writing - review and editing; Ning Wei, Jiajun Yuan: Formal analysis, Methodology, Data curation, Writing - review and editing; Weizhuo Yi, Rubing Pan, Jian Cheng: Writing - review and editing; Xulai Zhang, Hong Su: Supervision, Conceptualization, Validation, Writing - original draft, Writing review and editing.

The authors declare no competing financial interest.

Supplementary Material

eh3c00194_si_001.pdf (378.1KB, pdf)

References

  1. Jauhar S.; Johnstone M.; McKenna P. J. Schizophrenia. Lancet 2022, 399 (10323), 473–486. 10.1016/S0140-6736(21)01730-X. [DOI] [PubMed] [Google Scholar]
  2. Solmi M.; Seitidis G.; Mavridis D.; Correll C. U.; Dragioti E.; Guimond S.; et al. Incidence, prevalence, and global burden of schizophrenia - data, with critical appraisal, from the Global Burden of Disease (GBD) 2019. Mol. Psychiatry 2023, 10.1038/s41380-023-02138-4. [DOI] [PubMed] [Google Scholar]
  3. Institute for Health Metrics and Evaluation. GBD . Schizophrenia. 2019. https://vizhub.healthdata.org/gbd-results/.
  4. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2020, 396, 1223–1249. 10.1016/s0140-6736(20)30752-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Yin P.; Brauer M.; Cohen A. J.; Wang H.; Li J.; Burnett R. T.; et al. The effect of air pollution on deaths, disease burden, and life expectancy across China and its provinces, 1990–2017: an analysis for the Global Burden of Disease Study 2017. Lancet Planet Health 2020, 4 (9), e386 10.1016/S2542-5196(20)30161-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Song R.; Liu L.; Wei N.; Li X.; Liu J.; Yuan J.; et al. Short-term exposure to air pollution is an emerging but neglected risk factor for schizophrenia: A systematic review and meta-analysis. Sci. Total Environ. 2023, 854, 158823. 10.1016/j.scitotenv.2022.158823. [DOI] [PubMed] [Google Scholar]
  7. Liu F.; Liu C.; Liu Y.; Wang J.; Wang Y.; Yan B. Neurotoxicity of the air-borne particles: From molecular events to human diseases. J. Hazard Mater. 2023, 457, 131827. 10.1016/j.jhazmat.2023.131827. [DOI] [PubMed] [Google Scholar]
  8. Shi W.; Li T.; Zhang Y.; Sun Q.; Chen C.; Wang J.; et al. Depression and Anxiety Associated with Exposure to Fine Particulate Matter Constituents: A Cross-Sectional Study in North China. Environ. Sci. Technol. 2020, 54 (24), 16006–16016. 10.1021/acs.est.0c05331. [DOI] [PubMed] [Google Scholar]
  9. Mendoza A.; Hollenberg A. N. New insights into thyroid hormone action. Pharmacol Ther 2017, 173, 135–145. 10.1016/j.pharmthera.2017.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Patel J.; Landers K.; Li H.; Mortimer R. H.; Richard K. Thyroid hormones and fetal neurological development. J. Endocrinol 2011, 209 (1), 1–8. 10.1530/JOE-10-0444. [DOI] [PubMed] [Google Scholar]
  11. Liu Y. Y.; Brent G. A. Thyroid hormone and the brain: Mechanisms of action in development and role in protection and promotion of recovery after brain injury. Pharmacol Ther 2018, 186, 176–185. 10.1016/j.pharmthera.2018.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Misiak B.; Stańczykiewicz B.; Wiśniewski M.; Bartoli F.; Carra G.; Cavaleri D.; et al. Thyroid hormones in persons with schizophrenia: A systematic review and meta-analysis. Prog. Neuropsychopharmacol Biol. Psychiatry 2021, 111, 110402. 10.1016/j.pnpbp.2021.110402. [DOI] [PubMed] [Google Scholar]
  13. Gyllenberg D.; Sourander A.; Surcel H. M.; Hinkka-Yli-Salomäki S.; McKeague I. W.; Brown A. S. Hypothyroxinemia During Gestation and Offspring Schizophrenia in a National Birth Cohort. Biol. Psychiatry 2016, 79 (12), 962–970. 10.1016/j.biopsych.2015.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bode H.; Ivens B.; Bschor T.; Schwarzer G.; Henssler J.; Baethge C. Association of Hypothyroidism and Clinical Depression: A Systematic Review and Meta-analysis. JAMA Psychiatry 2021, 78 (12), 1375–1383. 10.1001/jamapsychiatry.2021.2506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Siegmann E. M.; Müller H. H. O.; Luecke C.; Philipsen A.; Kornhuber J.; Grömer T. W. Association of Depression and Anxiety Disorders With Autoimmune Thyroiditis: A Systematic Review and Meta-analysis. JAMA Psychiatry 2018, 75 (6), 577–584. 10.1001/jamapsychiatry.2018.0190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ichioka S.; Terao T.; Hoaki N.; Matsushita T.; Hoaki T. Triiodothyronine may be possibly associated with better cognitive function and less extrapyramidal symptoms in chronic schizophrenia. Prog. Neuropsychopharmacol Biol. Psychiatry 2012, 39 (1), 170–174. 10.1016/j.pnpbp.2012.06.008. [DOI] [PubMed] [Google Scholar]
  17. Barbero J. D.; Gutiérrez-Zotes A.; Montalvo I.; Creus M.; Cabezas A.; Solé M.; et al. Free thyroxine levels are associated with cognitive abilities in subjects with early psychosis. Schizophr Res. 2015, 166 (1–3), 37–42. 10.1016/j.schres.2015.04.030. [DOI] [PubMed] [Google Scholar]
  18. Freuer D.; Meisinger C. Causal link between thyroid function and schizophrenia: a two-sample Mendelian randomization study. Eur. J. Epidemiol 2023, 38, 1081. 10.1007/s10654-023-01034-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Liu J.; Zhao K.; Qian T.; Li X.; Yi W.; Pan R.; et al. Association between ambient air pollution and thyroid hormones levels: A systematic review and meta-analysis. Sci. Total Environ. 2023, 904, 166780. 10.1016/j.scitotenv.2023.166780. [DOI] [PubMed] [Google Scholar]
  20. Michikawa T.; Sasaki J.; Yamazaki S.; Takami A.; Asakura K.; Imamura H.; et al. A Case-Crossover Analysis of the Association between Exposure to Total PM(2.5) and Its Chemical Components and Emergency Ambulance Dispatches in Tokyo. Environ. Sci. Technol. 2022, 56 (11), 7319–7327. 10.1021/acs.est.1c08219. [DOI] [PubMed] [Google Scholar]
  21. Antonsen S.; Mok P. L. H.; Webb R. T.; Mortensen P. B.; McGrath J. J.; Agerbo E.; et al. Exposure to air pollution during childhood and risk of developing schizophrenia: a national cohort study. Lancet Planet Health 2020, 4 (2), e64–e73. 10.1016/S2542-5196(20)30004-8. [DOI] [PubMed] [Google Scholar]
  22. Kay S. R.; Fiszbein A.; Opler L. A. The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophrenia Bulletin 1987, 13 (2), 261–276. 10.1093/schbul/13.2.261. [DOI] [PubMed] [Google Scholar]
  23. Mass R.; Schoemig T.; Hitschfeld K.; Wall E.; Haasen C. Psychopathological syndromes of schizophrenia: evaluation of the dimensional structure of the positive and negative syndrome scale. Schizophr Bull. 2000, 26 (1), 167–177. 10.1093/oxfordjournals.schbul.a033437. [DOI] [PubMed] [Google Scholar]
  24. Fleischhacker W.; Galderisi S.; Laszlovszky I.; Szatmári B.; Barabássy A.; Acsai K.; et al. The efficacy of cariprazine in negative symptoms of schizophrenia: Post hoc analyses of PANSS individual items and PANSS-derived factors. Eur. Psychiatry 2019, 58, 1–9. 10.1016/j.eurpsy.2019.01.015. [DOI] [PubMed] [Google Scholar]
  25. Lisoni J.; Baldacci G.; Nibbio G.; Zucchetti A.; Butti Lemmi Gigli E.; Savorelli A.; et al. Effects of bilateral, bipolar-nonbalanced, frontal transcranial Direct Current Stimulation (tDCS) on negative symptoms and neurocognition in a sample of patients living with schizophrenia: Results of a randomized double-blind sham-controlled trial. J. Psychiatr Res. 2022, 155, 430–442. 10.1016/j.jpsychires.2022.09.011. [DOI] [PubMed] [Google Scholar]
  26. Lin C. H.; Lin H. S.; Lin S. C.; Kuo C. C.; Wang F. C.; Huang Y. H. Early improvement in PANSS-30, PANSS-8, and PANSS-6 scores predicts ultimate response and remission during acute treatment of schizophrenia. Acta Psychiatr Scand 2018, 137 (2), 98–108. 10.1111/acps.12849. [DOI] [PubMed] [Google Scholar]
  27. Jiang J.; Sim K.; Lee J. Validated five-factor model of positive and negative syndrome scale for schizophrenia in Chinese population. Schizophr Res. 2013, 143 (1), 38–43. 10.1016/j.schres.2012.10.019. [DOI] [PubMed] [Google Scholar]
  28. Phillips M. R.; Xiong W.; Wang R. W.; Gao Y. H.; Wang X. Q.; Zhang N. P. Reliability and validity of the Chinese versions of the Scales for Assessment of Positive and Negative Symptoms. Acta Psychiatr Scand 1991, 84 (4), 364–370. 10.1111/j.1600-0447.1991.tb03161.x. [DOI] [PubMed] [Google Scholar]
  29. Andreasen N. C.; Pressler M.; Nopoulos P.; Miller D.; Ho B. C. Antipsychotic dose equivalents and dose-years: a standardized method for comparing exposure to different drugs. Biol. Psychiatry 2010, 67 (3), 255–262. 10.1016/j.biopsych.2009.08.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Bobb J. F.; Valeri L.; Claus Henn B.; Christiani D. C.; Wright R. O.; Mazumdar M.; et al. Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures. Biostatistics 2015, 16 (3), 493–508. 10.1093/biostatistics/kxu058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Yi W.; Cheng J.; Song J.; Pan R.; Liang Y.; Sun X.; et al. Associations of polycyclic aromatic hydrocarbons, water-soluble ions and metals in PM(2.5) with liver function: Evidence from schizophrenia cohort. Sci. Total Environ. 2023, 868, 161624. 10.1016/j.scitotenv.2023.161624. [DOI] [PubMed] [Google Scholar]
  32. VanderWeele T. J.; Vansteelandt S. Conceptual issues concerning mediation, interventions and composition. Stat. Interface 2009, 2 (4), 457–468. 10.4310/SII.2009.v2.n4.a7. [DOI] [Google Scholar]
  33. Devick K. L.; Bobb J. F.; Mazumdar M.; Claus Henn B.; Bellinger D. C.; Christiani D. C.; et al. Bayesian kernel machine regression-causal mediation analysis. Stat Med. 2022, 41 (5), 860–876. 10.1002/sim.9255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Benjamini Y.; Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological) 1995, 57 (1), 289–300. 10.1111/j.2517-6161.1995.tb02031.x. [DOI] [Google Scholar]
  35. Straif K.; Baan R.; Grosse Y.; Secretan B.; El Ghissassi F.; Cogliano V. Carcinogenicity of polycyclic aromatic hydrocarbons. Lancet Oncol 2005, 6 (12), 931–932. 10.1016/S1470-2045(05)70458-7. [DOI] [PubMed] [Google Scholar]
  36. Chepelev N. L.; Moffat I. D.; Bowers W. J.; Yauk C. L. Neurotoxicity may be an overlooked consequence of benzo[a]pyrene exposure that is relevant to human health risk assessment. Mutat Res. Rev. Mutat Res. 2015, 764, 64–89. 10.1016/j.mrrev.2015.03.001. [DOI] [PubMed] [Google Scholar]
  37. Vignet C.; Devier M. H.; Le Menach K.; Lyphout L.; Potier J.; Cachot J.; et al. Long-term disruption of growth, reproduction, and behavior after embryonic exposure of zebrafish to PAH-spiked sediment. Environ. Sci. Pollut Res. Int. 2014, 21 (24), 13877–13887. 10.1007/s11356-014-2585-5. [DOI] [PubMed] [Google Scholar]
  38. Chen C.; Tang Y.; Jiang X.; Qi Y.; Cheng S.; Qiu C.; et al. Early postnatal benzo(a)pyrene exposure in Sprague-Dawley rats causes persistent neurobehavioral impairments that emerge postnatally and continue into adolescence and adulthood. Toxicol. Sci. 2012, 125 (1), 248–261. 10.1093/toxsci/kfr265. [DOI] [PubMed] [Google Scholar]
  39. Perera F. P.; Tang D.; Wang S.; Vishnevetsky J.; Zhang B.; Diaz D.; et al. Prenatal polycyclic aromatic hydrocarbon (PAH) exposure and child behavior at age 6–7 years. Environ. Health Perspect 2012, 120 (6), 921–926. 10.1289/ehp.1104315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Niu Q.; Zhang H.; Li X.; Li M. Benzo[a]pyrene-induced neurobehavioral function and neurotransmitter alterations in coke oven workers. Occup Environ. Med. 2010, 67 (7), 444–448. 10.1136/oem.2009.047969. [DOI] [PubMed] [Google Scholar]
  41. Achilleos S.; Kioumourtzoglou M. A.; Wu C. D.; Schwartz J. D.; Koutrakis P.; Papatheodorou S. I. Acute effects of fine particulate matter constituents on mortality: A systematic review and meta-regression analysis. Environ. Int. 2017, 109, 89–100. 10.1016/j.envint.2017.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Chung Y.; Dominici F.; Wang Y.; Coull B. A.; Bell M. L. Associations between long-term exposure to chemical constituents of fine particulate matter (PM2.5) and mortality in Medicare enrollees in the eastern United States. Environ. Health Perspect 2015, 123 (5), 467–474. 10.1289/ehp.1307549. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Park S.; Ku J.; Lee S. M.; Hwang H.; Lee N.; Kim H.; et al. Potential toxicity of inorganic ions in particulate matter: Ion permeation in lung and disruption of cell metabolism. Sci. Total Environ. 2022, 824, 153818. 10.1016/j.scitotenv.2022.153818. [DOI] [PubMed] [Google Scholar]
  44. Ju K.; Lu L.; Liao W.; Yang C.; Xu Z.; Wang W.; et al. Long-term exposure of PM(2.5) components on the adults’ depressive symptoms in China - Evidence from a representative longitudinal nationwide cohort. Sci. Total Environ. 2023, 857 (Pt 1), 159434. 10.1016/j.scitotenv.2022.159434. [DOI] [PubMed] [Google Scholar]
  45. Li J.; Wang Y.; Steenland K.; Liu P.; van Donkelaar A.; Martin R. V.; et al. Long-term effects of PM(2.5) components on incident dementia in the northeastern United States. Innovation (Camb) 2022, 3 (2), 100208. 10.1016/j.xinn.2022.100208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Qiu X.; Wei Y.; Amini H.; Wang C.; Weisskopf M.; Koutrakis P.; et al. Fine particle components and risk of psychiatric hospitalization in the U.S. Sci. Total Environ. 2022, 849, 157934. 10.1016/j.scitotenv.2022.157934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Huang W.; Wang G.; Lu S. E.; Kipen H.; Wang Y.; Hu M.; et al. Inflammatory and oxidative stress responses of healthy young adults to changes in air quality during the Beijing Olympics. Am. J. Respir Crit Care Med. 2012, 186 (11), 1150–1159. 10.1164/rccm.201205-0850OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Murthy C. R.; Rama Rao K. V.; Bai G.; Norenberg M. D. Ammonia-induced production of free radicals in primary cultures of rat astrocytes. J. Neurosci Res. 2001, 66 (2), 282–288. 10.1002/jnr.1222. [DOI] [PubMed] [Google Scholar]
  49. Reuben A.; Schaefer J. D.; Moffitt T. E.; Broadbent J.; Harrington H.; Houts R. M.; et al. Association of Childhood Lead Exposure With Adult Personality Traits and Lifelong Mental Health. JAMA Psychiatry 2019, 76 (4), 418–425. 10.1001/jamapsychiatry.2018.4192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Chen J.; Kan M.; Ratnasekera P.; Deol L. K.; Thakkar V.; Davison K. M. Blood Chromium Levels and Their Association with Cardiovascular Diseases, Diabetes, and Depression: National Health and Nutrition Examination Survey (NHANES) 2015–2016. Nutrients 2022, 14 (13), 2687. 10.3390/nu14132687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wu Y.; Wu Q.; Pan R.; Yi W.; Li Y.; Jin X.; et al. Phenotypic aging mediates the association between blood cadmium and depression: a population-based study. Environ. Sci. Pollut Res. Int. 2023, 30 (15), 44304–44315. 10.1007/s11356-023-25418-6. [DOI] [PubMed] [Google Scholar]
  52. Gu L.; Yu J.; Fan Y.; Wang S.; Yang L.; Liu K.; et al. The Association Between Trace Elements Exposure and the Cognition in the Elderly in China. Biol. Trace Elem Res. 2021, 199 (2), 403–412. 10.1007/s12011-020-02154-3. [DOI] [PubMed] [Google Scholar]
  53. Bramanti E.; Onor M.; Colombaioni L. Neurotoxicity Induced by Low Thallium Doses in Living Hippocampal Neurons: Evidence of Early Onset Mitochondrial Dysfunction and Correlation with Ethanol Production. ACS Chem. Neurosci. 2019, 10 (1), 451–459. 10.1021/acschemneuro.8b00343. [DOI] [PubMed] [Google Scholar]
  54. Baj J.; Forma A.; Sitarz E.; Karakuła K.; Flieger W.; Sitarz M. Beyond the Mind-Serum Trace Element Levels in Schizophrenic Patients: A Systematic Review. Int. J. Mol. Sci. 2020, 21 (24), 9566. 10.3390/ijms21249566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Hennemann G.; Docter R.; Friesema E. C.; de Jong M.; Krenning E. P.; Visser T. J. Plasma membrane transport of thyroid hormones and its role in thyroid hormone metabolism and bioavailability. Endocr Rev. 2001, 22 (4), 451–476. 10.1210/edrv.22.4.0435. [DOI] [PubMed] [Google Scholar]
  56. Sharif K.; Tiosano S.; Watad A.; Comaneshter D.; Cohen A. D.; Shoenfeld Y.; et al. The link between schizophrenia and hypothyroidism: a population-based study. Immunol Res. 2018, 66 (6), 663–667. 10.1007/s12026-018-9030-7. [DOI] [PubMed] [Google Scholar]
  57. Noda M. Thyroid Hormone in the CNS: Contribution of Neuron-Glia Interaction. Vitam Horm 2018, 106, 313–331. 10.1016/bs.vh.2017.05.005. [DOI] [PubMed] [Google Scholar]
  58. Remaud S.; Gothié J. D.; Morvan-Dubois G.; Demeneix B. A. Thyroid hormone signaling and adult neurogenesis in mammals. Front Endocrinol (Lausanne) 2014, 5, 62. 10.3389/fendo.2014.00062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Bernal J.; Nunez J. Thyroid hormones and brain development. Eur. J. Endocrinol. 1995, 133 (4), 390–398. 10.1530/eje.0.1330390. [DOI] [PubMed] [Google Scholar]
  60. Lima F. R.; Gervais A.; Colin C.; Izembart M.; Neto V. M.; Mallat M. Regulation of microglial development: a novel role for thyroid hormone. J. Neurosci. 2001, 21 (6), 2028–2038. 10.1523/JNEUROSCI.21-06-02028.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Falkenburg J.; Tracy D. K. Sex and schizophrenia: a review of gender differences. Psychosis 2014, 6 (1), 61–69. 10.1080/17522439.2012.733405. [DOI] [Google Scholar]
  62. Ministry of Environmental Protection of the People’s Republic of China . Exposure Factors Handbook of Chinese Population; China Environmental Science Press: Beijing, China, 2013; p 270. [Google Scholar]
  63. Kulkarni J.; Gavrilidis E.; Hayes E.; Heaton V.; Worsley R. Special biological issues in the management of women with schizophrenia. Expert Rev. Neurother 2012, 12 (7), 823–833. 10.1586/ern.12.62. [DOI] [PubMed] [Google Scholar]
  64. Baksi S.; Pradhan A. Thyroid hormone: sex-dependent role in nervous system regulation and disease. Biol. Sex Differ 2021, 12 (1), 25. 10.1186/s13293-021-00367-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

eh3c00194_si_001.pdf (378.1KB, pdf)

Articles from Environment & Health are provided here courtesy of American Chemical Society

RESOURCES