Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2026 Jan 16;16:5592. doi: 10.1038/s41598-026-36243-5

Environmental pollutants associated with blood glucose levels in healthy individuals

Xuekui Liu 1,#, Gangshan Peng 1,#, Yanhong Lin 2,#, Wenruo Chen 2, Houfa Geng 1,✉, Jun Liang 2,✉
PMCID: PMC12891712  PMID: 41545583

Abstract

The occurrence of diabetes is closely linked to environmental pollution. In the pre-diabetic stage, the human body is actually already exposed to the environment. Exploring the correlation between environmental exposure substances and blood glucose levels in individuals with normal blood glucose is of great and positive practical significance for the effective prevention and precise control of diabetes. This study meticulously collected detailed information from 307 individuals in northern China, and sampled the serum of the participants for analysis using liquid chromatography-tandem mass spectrometry (LC–MS/MS) technology. A total of 203 environmental exposure substances were detected. Results: Through in-depth exposome research, it was found that environmental exposure substances such as α-hexabromocyclododecane (α-HBCD), 4-methylbenzylidene camphor (4-MBC), and (1,2-diphenylethane-1,2-dione) benzil showed a significant positive correlation with the increase in blood glucose levels. In contrast, substances such as isopropylphenyl diphenyl disulfide (IPPD), tris(1,3-dichloro-2-propyl) phosphate (TDCIPP), and dibenzothiophene sulfone (PES) were negatively correlated with blood glucose elevation. Both the Weighted Quantile Sum (WQS) analysis and the Bayesian Kernel Machine Regression (BKMR) analysis showed that there was a positive correlation between the overall effect of environmental exposure substances and blood glucose elevation (95% confidence interval = 1.141–1.721; P < 0.001). Environmental pollutants are important risk factors for the increase in blood glucose levels in healthy individuals. Moreover, these pollutants may interact with and affect each other, leading to a further rise in blood glucose levels.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-36243-5.

Keywords: Environmental pollutants, Fasting plasma glucose, Weighted quantile Sum, Bayesian kernel machine regression

Subject terms: Environmental social sciences, Endocrinology

Introduction

Diabetes has emerged as a significant global public health challenge1. According to the latest epidemiological data, the number of individuals with diabetes reached 529 million in 2021, with projections estimating a rise to 1.31 billion by 20502. China, the country most affected by diabetes, reported an adult prevalence rate of 12.4% (based on 2018 data), accounting for over 118 million patients—approximately 22% of the global diabetic population3. However, traditional risk factors, such as obesity, genetics, and lifestyle choices, explain only about 50% of diabetes cases4. This suggests that environmental exposure factors, especially chemical pollutants, may play a significant role in the disease’s development.

Recent epidemiological studies have confirmed that environmental chemicals, including persistent organic pollutants (POPs) such as polychlorinated biphenyls (PCBs) and organochlorine pesticides (OCPs), as well as endocrine disruptors like phthalates and bisphenol A, and per- and polyfluoroalkyl substances (PFAS), significantly increase the risk of diabetes5–7. These chemicals can contribute to diabetes through mechanisms involving insulin resistance, β-cell dysfunction, and oxidative stress8. While existing research has primarily focused on populations with diagnosed diabetes, the critical and potentially reversible pre-diabetic stage has largely been neglected.

The progression to diabetes is a gradual process, and interventions targeting impaired fasting glucose (IFG) can reduce diabetes risk by as much as 58% through lifestyle modifications or environmental risk management9. Despite this potential for early intervention, current research lacks a systematic analysis of how environmental pollutants influence blood glucose fluctuations in the pre-diabetic stage10. Specifically, the relationship between environmental exposures and subclinical hyperglycemia in non-diabetic populations remains poorly understood.

Although several studies have established a clear link between air pollution and the onset of type 2 diabetes, the specific mechanisms underlying this association remain inadequately explored11. For example, airborne particulate matter can affect insulin sensitivity by triggering inflammatory responses and inducing oxidative stress, ultimately leading to elevated blood glucose levels12,13. Similarly, chemicals such as pesticides and heavy metals may disrupt endocrine or immune systems, indirectly influencing glucose metabolism14.

To address these critical knowledge gaps, there is a pressing need for systematic research aimed at elucidating the various biological mechanisms through which environmental factors impact blood glucose levels, particularly in the pre-diabetic population. Such research could yield essential insights for developing early intervention strategies to prevent the progression to diabetes.

Methods

Research subjects

Participants were recruited from individuals undergoing health check-ups at the Health Examination Center of Xuzhou Central Hospital between August 2024 and December 2024. Inclusion criteria: Age ≥ 18 years; Voluntary participation in the study; No prior hospitalization history for hypertension, type 2 diabetes, or malignant tumors. Exclusion criteria: Incomplete physical examination data or lack of blood glucose testing; Non-compliance with blood sample collection procedures; Diagnosed mental illnesses that could impair the ability to complete the study.

This study was approved by the Ethics Committee of Xuzhou Central Hospital, and all participants provided written informed consent prior to their enrollment. All procedures involving human subjects in this study were conducted in accordance with the ethical standards of Xuzhou Central Hospital, as well as the 1964 Helsinki Declaration and its subsequent amendments.

Data collection from participants

Venous blood samples were collected from all study participants following an 8-h fasting period. Complete blood counts were conducted using EDTA-coated tubes, while serum separation tubes containing a clot activator were employed for biochemical analyses. For lipoprotein-related tests, specialized vacuum tubes with inert gel barriers were utilized.

To maintain sample integrity, blood samples were stored upright at 4 °C in sealed containers shielded from light for a maximum of 6 h prior to centrifugation, which was conducted at 2,500 × g for 10 min at 4 °C to prevent hemolysis and degradation of metabolites. Serum aliquots intended for long-term storage were immediately frozen at −80 °C in cryovials with minimal headspace. The processed specimens were subsequently analyzed using biochemical and blood glucose testing instruments to measure fasting plasma glucose (FPG) and blood lipid profiles, which included total cholesterol, triglycerides, high-density lipoprotein (HDL), and low-density lipoprotein (LDL). These lipid profiles were assessed using a Japanese fully automated biochemical analyzer, specifically the Hitachi LABOSPECT 008 α model.

In addition to blood analyses, anthropometric measurements, including height and weight, as well as blood pressure readings, were recorded. Prior to measuring blood pressure, participants were instructed to remain seated and rest for 5 min to ensure accurate readings. This comprehensive data collection protocol facilitated the acquisition of robust physiological datasets, which were essential for subsequent analyses exploring the associations between exposure biomarkers and glucose metabolism.

High-throughput screening and targeted quantification of exposure by LC–MS/MS

This study utilized a high-throughput screening method for serum endocrine disrupting chemicals (EDCs) through high-performance liquid chromatography-triple quadrupole tandem mass spectrometry, selecting 203 compounds from 11 major categories for targeted quantitative analysis. The evaluated exposures included traditional environmental contaminants such as Polycyclic Aromatic Hydrocarbons (PAHs), Phthalates (PAEs), and Perfluorooctanoic Acid (PFOA), as well as emerging pollutants like Persistent Organic Pollutants (POPs), Per- and Polyfluoroalkyl Substances (PFAS), Organophosphate Esters (OPEs), Environmental Endocrine Disruptors (EDCs), and Pharmaceuticals and Personal Care Products (PPCPs). Specifically, the evaluated pollutants were classified into the following 11 categories: Organophosphate Esters (OPEs), Polycyclic Aromatic Hydrocarbons (PAHs), Ultraviolet Stabilizers (UV-S), Synthetic Antioxidants (SA), Phthalates (PAEs), Novel Brominated Flame Retardants (NBFRs), Pesticides (Pstc), Pharmaceuticals and Personal Care Products (PPCPs), Per- and Polyfluoroalkyl Substances (PFAS), Bisphenols (BAs), and Brominated Flame Retardants (BFRs) (Supplement Table S2). The experiment employed a Shimadzu LC-40D XS ultra-high performance liquid chromatography system coupled with an LCMS-8060 NX triple quadrupole mass spectrometer equipped with an electrospray ionization source (ESI). Detection was performed in multiple reaction monitoring (MRM) mode (Supplement Table S1). Key mass spectrometry parameters were as follows: 1) Ionization mode: ESI +/ESI-; 2) Spray voltages: + 4.5 kV (for positive ions) and −3.5 kV (for negative ions); 3) Capillary voltage: 4000 V;4) Desolvation temperature: 300 °C; 4) Nebulizing gas flow rate: 3 L/min (nitrogen, ≥ 99.99% purity); 5) Collision gas: argon (≥ 99.999% purity, at 270 kPa); 6) Drying gas and sheath gas: nitrogen (flow rate 15 L/min, temperature 325 °C).

Analytical quality control

Before instrumental analysis, serum samples underwent a solid-phase extraction (SPE) procedure to eliminate matrix interferences and enrich the target analytes. Specifically, 200 μL of each serum sample was spiked with isotope-labeled internal standards (e.g., diisobutyl phthalate-D4, atrazine-D5, methylparaben-D4, triphenyl phosphate-D15) to monitor extraction efficiency and compensate for potential matrix effects. The samples were then subjected to liquid–liquid extraction using a mixture of ethyl acetate and hexane (1:1, v/v) containing 0.6% (v/v) formic acid. After vortexing and centrifugation, the organic layer was collected, dried under a gentle stream of nitrogen at 40 °C, and reconstituted in 100 μL of methanol–water (1:1, v/v) prior to LC–MS/MS analysis.

To ensure data accuracy and reproducibility, rigorous quality assurance/quality control (QA/QC) procedures were implemented throughout the analytical process. Blank samples (solvent only) were analyzed intermittently between real samples to monitor potential carry-over or contamination. Additionally, calibration curves were constructed using matrix-matched standards across a concentration range of 0.5–10,000 ng/L, with a weighting factor of 1/x applied during linear regression to enhance fitting accuracy at low concentrations. The correlation coefficients (R2) for all calibration curves exceeded 0.99.

Method validation was performed in accordance with international bioanalytical guidelines. For the 97 target analytes with available reference standards (Supplement Table 2), the limit of detection (LOD), defined as the concentration yielding a signal-to-noise ratio (S/N) of 3, ranged from 0.1 to 3 ng/L. The lower limit of quantification (LLOQ), corresponding to S/N = 10, was established at 0.5 ng/L for all compounds and fell within the validated range of 0.5–10 ng/L, ensuring sensitive detection of trace-level exposure biomarkers in human serum.

Table 2.

Multivariate logistic regression to explore the association between exposures and the risk of higher glucose.

Exposoures Beta S.E Wald P OR 95%C.I
IPPD −0.306 0.079 15.046  < 0.001 0.737 0.631 ~ 0.860
Benzil 0.097 0.042 5.232 0.022 1.101 1.014 ~ 1.196
4-MBC 0.132 0.05 6.918 0.009 1.141 1.034 ~ 1.260
PES −0.331 0.081 16.924  < 0.001 0.718 0.613 ~ 0.841
TDCIPP −0.354 0.08 19.638  < 0.001 0.692 0.600 ~ 0.821
α-HBCD 0.295 0.071 17.411  < 0.001 1.344 1.170 ~ 1.544
Combined 1.381 0.195 50.213  < 0.001 4.182 2.884 ~ 6.071
Open in a new tab

All models adjusted gender, age, BMI, SBP, DBP, TC, TG, HDL and LDL.

Spiked recovery experiments were conducted by fortifying drug-free human serum with known amounts of each target analyte at three concentration levels (low: 10 ng/L; medium: 100 ng/L; high: 1000 ng/L), followed by the full sample preparation workflow. Mean recovery rates for the 97 quantified compounds ranged from 70 to 120%, with relative standard deviations (RSDs) below 15% (n = 6), demonstrating acceptable accuracy and robustness in the complex serum matrix. Matrix effects, evaluated by comparing the peak areas of post-extraction spiked samples with those of neat standards, were within ± 15% for all analytes.

Intra-day and inter-day precision and accuracy were assessed by analyzing quality control (QC) samples over three consecutive days. All results met the acceptance criteria, with intra- and inter-day RSDs maintained below 10%. These comprehensive QA/QC measures ensured the reliability, consistency, and reproducibility of the LC–MS/MS data reported in this study.

Statistical analysis

In this study, statistical analysis was conducted using R version 4.4.2 software.All quantitative data were tested for normality prior to analysis. The exposure data underwent a normality test after log2 transformation, and data that did not conform to a normal distribution were subjected to necessary transformations for normalization. In this study, all clinical data met the criteria for normal distribution, and the exposure data also conformed to a normal distribution after log2 transformation. During the clinical data collection phase, various parameters were analyzed, including gender, age, height, weight, BMI, systolic blood pressure (SBP), diastolic blood pressure (DBP), and four lipid profile metrics (total cholesterol [TC], triglycerides [TG], high-density lipoprotein [HDL], and low-density lipoprotein [LDL]). The detailed analysis steps are outlined below: Grouping: All samples were divided into two groups based on their FPG levels. There were 77 participants in the higher glucose group(FPG ≥ 5.6 mmol/L) and 230 participants in the normal glucose group(FPG < 5.6 mmol/L). Initial Data Analysis: To determine differences in clinical data between the two groups, the chi-square test was used for categorical variables, while the two-independent-samples t-test was applied for continuous variables. Exposure Analysis: After quantifying exposure substances using LC–MS/MS, exposome analysis methods were employed to identify differential exposure substances between the two groups. A multivariate logistic regression model was then utilized to analyze the correlation between these differential exposure substances and elevated blood glucose levels. Combined Effects Analysis: The Weighted Quantile Sum (WQS) and Bayesian Kernel Machine Regression (BKMR) methods were adopted to evaluate the combined effects of the differential exposure substances on elevated blood glucose (Detailed information regarding WQS and BKMR is provided in Supplementary Information 2.). These methods allowed for a comprehensive assessment of how multiple exposures interact to influence blood glucose levels. All statistical tests were performed with a significance level set at p < 0.05.

Results

The clinical characteristics of participants

A total of 307 participants were included in this study, with 230 exhibiting FPG levels below 5.6 mmol/L and 77 having FPG levels equal to or greater than 5.6 mmol/L. Consequently, the participants were categorized into two groups based on their FPG levels. As shown in Table 1, there were no statistically significant differences in weight, body mass index (BMI), SBP, DBP, TC,TG, HDL, and LDL between the two groups, except for the variable of age.

Table 1.

The clinical characteristics of all participants.

FPG
 < 5.6 mmol/L  ≥ 5.6 mmol/L p
n 230 77
Gender(M) 122 47 0.276
Age (years) 51.17(13.49) 59.04(13.44)  < 0.001
Weight (kg) 63.47(8.20) 64.39(8.44) 0.4
BMI(kg/m2) 23.02(2.18) 23.41(2.33) 0.182
SBP(mmHg) 117.88(11.76) 120.73(12.20) 0.07
DBP (mmHg) 73.10(7.20) 73.39(7.43) 0.762
TC(mmol/L) 4.31(0.59) 4.20(0.72) 0.167
TG(mmol/L) 0.99(0.38) 1.02(0.32) 0.586
HDL(mmol/L) 1.32(0.20) 1.34(0.26) 0.442
LDL(mmol/L) 2.34(0.45) 2.23(0.53) 0.07
Open in a new tab

Results of exposure detection

In this study, we conducted a comprehensive analysis of 203 exposure substances across 11 categories, utilizing Principal Component Analysis (PCA) to investigate the underlying relationships among these substances. The analysis of the scree plot indicated significant correlations among the detected exposure substances, suggesting that they do not operate in isolation (Fig. 1A).

Fig. 1.

Fig. 1

Open in a new tab

The differences in all exposures between the two groups were described using PCA analysis.

While examining the PCA plot, we observed clustering patterns of the exposure substances in relation to glucose levels. Samples from the higher glucose group exhibited a degree of similarity in their exposure profiles; however, there was no clear visual separation between the higher glucose group and the other groups. It is important to note that the first two principal components accounted for approximately 31% of the total variance, indicating that while trends may be present, the data does not show strong structural clarity in distinguishing between the different glucose groups (Fig. 1B). Supplement Table S1 showed the expression value of exposures between two groups.

Analysis results of 203 exposed substances

The exposomics analysis identified notable differences in the expression levels of certain exposure substances between the normal glucose group and the higher glucose group among participants. Figure 2A (Volcano Plot): This volcano plot demonstrates the differential expression of exposure substances between the higher glucose group and the normal glucose group. The x-axis shows the log₂(Fold Change), with “Up” substances (e.g., 4-MBC, Benzil, α-HBCD) exhibiting higher expression in the higher glucose group, while “Down” substances (e.g., IPPD, TDCIPP, PES) display lower expression. All marked substances show high -log10(Adjusted P-value), indicating statistically significant differential expression. Figure 2B (Heatmap): The heatmap visualizes expression patterns of key substances across samples, with rows representing substances and columns corresponding to individual samples grouped by glucose status. Distinct clustering reveals consistent expression patterns, with 4-MBC, Benzil, and α-HBCD showing higher expression and IPPD, PES, and TDCIPP exhibiting lower expression in the higher glucose group, consistent with the volcano plot. Figure 2C (Violin Plots): These violin plots illustrate expression level distributions. 4-MBC, Benzil, and α-HBCD present upward shifts in the higher glucose group, indicating higher expression and significant p-values (e.g., p < 0.0001). In contrast, IPPD, PES, and TDCIPP show downward shifts, confirming lower expression with strong statistical significance (e.g., p < 0.0001).

Fig. 2.

Fig. 2

Open in a new tab

The analysis of differentially exposed substances between two groups was conducted using volcano plots, heat maps, and intergroup difference violin plots.

Relationship between differentially expressed exposures and hyperglycemia

Multivariate logistic regression analysis was performed to assess the associations between six significant exposure substances and the risk of hyperglycemia, with adjustments for potential confounding factors such as gender, age, BMI, SBP, DBP, TC, TG, HDL, and LDL. The results, as shown in Table 2, indicated that even after these adjustments, significant correlations remained. Benzil (odd ratio, OR = 1.101), α-HBCD (OR = 1.344), and 4-MBC (OR = 1.141) were associated with an increased risk of hyperglycemia. In contrast, IPPD (OR = 0.737), PES (OR = 0.718), and TDCIPP (OR = 0.692) were associated with a reduced risk of hyperglycemia, suggesting a potential protective effect. Factor analysis was carried out to extract a common factor named “Combined”. The results demonstrated that the “Combined” factor was a risk factor for hyperglycemia, with an OR of 4.182. Furthermore, subgroup analysis was conducted to explore the correlations between exposure substances and hyperglycemia under different genders and age groups. Figure 3G showed that the combined factor of exposure substances was consistently a risk factor for hyperglycemia across different genders and age groups. Figure 3A revealed that the relationship between α-HBCD and hyperglycemia varied with age and gender. Specifically, α-HBCD had no significant statistical correlation with hyperglycemia in individuals < 45 years old. However, in both male and female groups, as well as in individuals over 45 years old, α-HBCD was positively correlated with hyperglycemia. Figure 3B shows that the association between TDCIPP and hyperglycemia varied with age and gender. In populations of different genders, TDCIPP shows a negative correlation with hyperglycemia. This negative correlation is also observed in individuals < 45 years old. However, in individuals under 45 years old, there is no significant correlation. Figure 3C, Fig. 3D, Fig. 3E, and Fig. 3F illustrate the correlations between PES, Benzil, IPPD, and 4-MBC with hyperglycemia, respectively. The relationships between these exposures and hyperglycemia vary across different genders and age groups.

Fig. 3.

Fig. 3

Open in a new tab

The correlations between six differentially exposed substances and hyperglycemia were analyzed by stratification according to different genders and age levels.

Analysis of the joint effects of exposures on blood glucose elevation using WQS and BKMR

Using the WQS method, we analyzed the comprehensive impact of six exposure substances on blood glucose elevation. After adjusting for factors such as gender, age, BMI, SBP, DBP, TC, TG, HDL, and LDL, the results indicated that for every 1 interquartile range increase in the WQS score, the risk of blood glucose elevation increased by 1.401 (95% Confidence Interval, CI = 1.141–1.721; P < 0.001). This demonstrates a significant positive correlation between overall exposure increase and blood glucose elevation. As shown in Fig. 4A, α-HBCD had the highest weight (weight = 0.53), followed by 4-MBC (weight = 0.31) and Benzil (weight = 0.17), suggesting that these three pollutants might be key contributors to blood glucose elevation.

Fig. 4.

Fig. 4

Open in a new tab

The correlations between the joint effects of six differentially exposed substances and hyperglycemia were analyzed using WQS and BKMR.

Subsequently, we used BKMR to evaluate the joint impact of the six exposure substances on blood glucose elevation. After adjusting for the same confounding factors as above, the data demonstrated that the overall effect of the exposure mixture was statistically significant, showing a positive correlation between exposure level and blood glucose elevation. Specifically, when the exposure level increased from the 50th percentile to the 75th percentile, the risk of blood glucose elevation increased fivefold, further confirming the significant positive correlation(Fig. 4C). The single-effect analysis revealed that Benzil and α-HBCD had a particularly strong positive correlation with blood glucose elevation, indicating their significant role in influencing this health outcome (Fig. 4B).

Discussion

This study explored the relationship between environmental pollutants and blood glucose levels in a cohort of 307 individuals from northern China. The results indicated that, among non-diabetic participants, several exposure substances are closely linked to increases in FPG. Notably, α-HBCD, 4-MBC, and Benzil were identified as significant risk factors contributing to elevated blood glucose.

Utilizing the WQS and BKMR methods, this present study demonstrated that an overall increase in exposure levels was significantly positively correlated with higher blood glucose levels. It is well known that the increase in environmental pollutants significantly raises the risk of elevated blood glucose15,16. Numerous studies have revealed that long-term exposure to ultrafine particles (UFP) and per- and polyfluoroalkyl substances (PFAS) is closely associated with human blood glucose levels and the risk of developing type 2 diabetes17–19. Exposure to high concentrations of UFP can remarkably increase the prevalence of diabetes and the risk at various stages of the disease, having an adverse impact on the levels of FPG and glycated hemoglobin (HbA1c)20. Among PFAS, certain compounds such as perfluorodecanoic acid (PFDA) exhibit a negative correlation with the risk of type 2 diabetes, yet the complex underlying mechanisms still require in-depth exploration21. These research findings strongly highlight the crucial significance of effectively controlling environmental pollution for the prevention of diabetes.

Brominated flame retardants (BFRs), such as hexabromocyclododecane (HBCD), are extensively used in building materials and electronic products22. A wealth of research has confirmed that they pose potential health risks. Studies have shown that HBCD and its stereoisomers, such as α-HBCD and γ-HBCD, can interfere with the human endocrine system and disrupt metabolic processes, leading to insulin resistance and disorders of glucose metabolism23,24. Our study also suggests that α-HBCD is an independent risk factor for the elevation of FPG. Long-term exposure to HBCD significantly increases the risk of obesity, type 2 diabetes, and other metabolic diseases. Notably, HBCD exposure can also damage liver function, induce hepatotoxicity, and further exacerbate metabolic abnormalities25,26. 4-MBC, as a widely used ingredient in cosmetics, has been proven to possess remarkable endocrine-disrupting properties27. Studies have shown that 4-MBC can activate estrogen receptors, interfere with the synthesis and metabolism processes of sex hormones, and thus trigger functional abnormalities in the reproductive system28. Moreover, 4-MBC can also disrupt the normal physiological functions of the thyroid gland, affecting the synthesis and regulatory mechanisms of thyroid hormones. Long-term exposure to 4-MBC is highly likely to have adverse effects on the nervous system and kidney functions, and may even increase the risk of developing cancer29,30. Benzil is an extremely crucial organic synthesis intermediate, occupying a significant position in the field of organic synthesis31. In the field of medical research, 1,2-diphenylethane-1,2-dione (benzil) and its derivative compounds exhibit abundant potential for biological activity32. Scientific research has revealed that some benzil derivatives possess multiple biological activities such as antibacterial, anti-inflammatory, and antioxidant properties33. In our study, we found that benzil is a risk factor for elevated blood glucose. This conclusion is drawn based on epidemiological characteristics and still needs further verification.

In this study, we found that there is a negative correlation of IPPD, TDCIPP, PES with the elevation of blood glucose, which is inconsistent with the conclusions of other related studies34–36. This discrepancy may be closely related to the design characteristics of this study. Since this study adopted a cross-sectional research method, it has certain limitations and is unable to accurately verify the causal relationship between IPPD, TDCIPP, PES and the elevation of blood glucose. In addition, the relationship between the exposure substances and the body is significantly influenced by various factors such as the exposure duration and the exposure dose. Under the effect of low-dose exposure substances, the body may activate its own compensatory functions to cope with the situation37, and this speculation still needs to be verified through further experimental studies. However, in the in-depth analysis of the impact of the combined effects of multiple exposure substances on the elevation of blood glucose, we have clearly found that the overall effect of these exposure substances is a risk factor contributing to the increase in blood glucose levels. This finding strongly indicates that when assessing and analyzing the potential harm caused by exposure substances to human health, we should not merely consider the effects of individual exposure substances in isolation. Instead, a comprehensive analysis from an overall perspective is necessary. Such an analytical approach is evidently more practical and instructive, providing a more reliable basis for us to have a more comprehensive understanding of the relationship between exposure substances and health.

This study has several limitations that should be acknowledged. First, while cross-sectional studies can provide valuable insights for clinical practice and help identify potential associations and trends, their ability to establish causal relationships is relatively weak due to the nature of their study design. This makes it challenging to conclusively prove causal connections between variables. Second, although the study successfully identified certain exposure substances associated with elevated blood glucose levels in individuals with normal glucose levels—a significant finding—the relatively small sample size may not adequately represent the broader population’s characteristics and conditions. Therefore, large-scale population studies are essential to further validate the reliability and generalizability of these findings. Lastly, the study utilized two different blood matrices: plasma to measure clinical outcomes (e.g., fasting plasma glucose, lipid profiles) and serum to analyze environmental chemical exposures. While method validation was performed to ensure the accuracy of target chemical quantification in serum, and previous research has demonstrated a high correlation of most environmental pollutants between these two matrices, potential limitations still exist when using different blood matrices.

In conclusion, this study, which focused on 307 individuals from northern China with normal blood glucose levels, identified several environmental exposure substances associated with blood glucose elevation. It is worth noting that there is a high probability of a synergistic combined effect among these environmental exposure substances. They interact with and act on each other, thus exerting a comprehensive influence on human blood glucose levels.

Supplementary Information

Supplementary Information 1. (19.6KB, docx)
Supplementary Information 2. (69.9KB, docx)

Acknowledgements

We acknowledge and thank all participants for their cooperation and sample contributions.

Author contributions

XK Liu was drafted this manuscript, GS Peng, WR Chen and YH Lin collected the information of sample and analyzed the data. J Liang and HF Geng are responsible for the integrity of the work as a whole. All authors read and approved the final version of the manuscript.

Funding

This work was supported by the Science and Technology Project of Xuzhou Municipal Health Commission (No. 2025DF07) and the General Program of Jiangsu Provincial Preventive Medicine Research Project (No. YM2023070).

Data availability

All data generated or analyzed during this study are included in this manuscript.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The study was reviewed and approved by the ethics committee of the Xuzhou central hospital. The NO. of ethics committee approval is XZXY-LJ-20201110–060.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Xuekui Liu, Gangshan Peng and Yanhong Lin are contributed equally to this work.

Contributor Information

Houfa Geng, Email: genghoufa@xzhmu.edu.cn.

Jun Liang, Email: mwlj521@njmu.edu.cn.

References

  • 1.Collaborators GFaF. Global fertility in 204 countries and territories, 1950–2021, with forecasts to 2100: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. Lancet403(10440), 2057–2099 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.(NCD-RisC) NRFC. Worldwide trends in diabetes prevalence and treatment from 1990 to 2022: a pooled analysis of 1108 population-representative studies with 141 million participants. Lancet404(10467), 2077–2093 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Xu, Y. et al. Diabetes in China part 1: epidemiology and risk factors. Lancet Public Health9(12), e1089–e1097 (2024). [DOI] [PubMed] [Google Scholar]
  • 4.Xu, H. et al. Environmental pollution, a hidden culprit for health issues. Eco Environ. Health1(1), 31–45 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Cohen, M. Environmental toxins and health–the health impact of pesticides. Aust. Fam. Physician36(12), 1002–1004 (2007). [PubMed] [Google Scholar]
  • 6.Dendup, T., Feng, X., Clingan, S. & Astell-Burt, T. Environmental risk factors for developing type 2 diabetes Mellitus: a systematic review. Int. J. Environ. Res. Public Health15(1), 78 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Laakso, M. Biomarkers for type 2 diabetes. Mol Metab27(27s(Suppl)), S139–S146 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Gong, T. et al. Study on the mechanism of plant metabolites to intervene oxidative stress in diabetic retinopathy. Front. Pharmacol.16, 1517964 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Schauer, P. R. et al. Bariatric surgery versus intensive medical therapy for diabetes - 5-year outcomes. N. Engl. J. Med.376(7), 641–651 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Duan, Y. & Sun, H. Studies on the association between environmental organic pollutants and diabetes. Environ. Chem.36(4), 753–766 (2017). [Google Scholar]
  • 11.Rajagopalan, S. & Brook, R. D. Air pollution and type 2 diabetes: mechanistic insights. Diabetes61(12), 3037–3045 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Lim, C. C. & Thurston, G. D. Air Pollution, Oxidative Stress, and Diabetes: a Life Course Epidemiologic Perspective. Curr. Diab. Rep.19(8), 58 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Xu, H. et al. The mediating role of vascular inflammation in traffic-related air pollution associated changes in insulin resistance in healthy adults. Int. J. Hyg. Environ. Health239, 113878 (2022). [DOI] [PubMed] [Google Scholar]
  • 14.Wang, X., Mukherjee, B. & Park, S. K. Associations of cumulative exposure to heavy metal mixtures with obesity and its comorbidities among U.S. adults in NHANES 2003–2014. Environ. Int.121(Pt 1), 683–694 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Akash, M. S. H., Sabir, S. & Rehman, K. Bisphenol A-induced metabolic disorders: From exposure to mechanism of action. Environ. Toxicol. Pharmacol.77, 103373 (2020). [DOI] [PubMed] [Google Scholar]
  • 16.Mudumbi, J. B. N., Ntwampe, S. K. O., Mekuto, L., Matsha, T. & Itoba-Tombo, E. F. The role of pollutants in type 2 diabetes mellitus (T2DM) and their prospective impact on phytomedicinal treatment strategies. Environ. Monit. Assess190(5), 262 (2018). [DOI] [PubMed] [Google Scholar]
  • 17.Li, Z. H. et al. Association of physical activity and air pollution exposure with the risk of type 2 diabetes: a large population-based prospective cohort study. Environ. Health21(1), 106 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Luo, H. et al. Long-term exposure to ambient air pollution is a risk factor for trajectory of cardiometabolic multimorbidity: A prospective study in the UK Biobank. EBioMedicine84, 104282 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Yang, T. et al. Associations between long-term exposure to ultrafine particles and type 2 diabetes: A large-scale, multicenter study in China. J. Hazard. Mater.488, 137364 (2025). [DOI] [PubMed] [Google Scholar]
  • 20.Yang, T. et al. Associations of residential green space with incident type 2 diabetes and the role of air pollution: A prospective analysis in UK Biobank. Sci. Total Environ.866, 161396 (2023). [DOI] [PubMed] [Google Scholar]
  • 21.Wang, M. et al. Per- and polyfluoroalkyl substances and type 2 diabetes among older adults: Synthesizing cross-sectional population study and meta-analysis. Int. J. Hyg. Environ. Health266, 114560 (2025). [DOI] [PubMed] [Google Scholar]
  • 22.Xiong, P. et al. A review of environmental occurrence, fate, and toxicity of novel brominated flame retardants. Environ. Sci. Technol.53(23), 13551–13569 (2019). [DOI] [PubMed] [Google Scholar]
  • 23.Miller, I. et al. Hexabromocyclododecane (HBCD) induced changes in the liver proteome of eu- and hypothyroid female rats. Toxicol. Lett.245, 40–51 (2016). [DOI] [PubMed] [Google Scholar]
  • 24.Szabo, D. T., Pathmasiri, W., Sumner, S. & Birnbaum, L. S. Serum Metabolomic Profiles in Neonatal Mice following Oral Brominated Flame Retardant Exposures to Hexabromocyclododecane (HBCD) Alpha, Gamma, and Commercial Mixture. Environ. Health Perspect.125(4), 651–659 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Marques, M. L. & Cairrao, E. Occurrence and health effects of Hexabromocyclododecane: an updated review. Toxics11(5), 409 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Mohammed, N. A., Lewis, K., Hodges, N. & Michelangeli, F. Mechanisms of cell death induced by hexabromocyclododecane (HBCD) involves apoptosis, autophagy, and ER stress. J. Biochem. Mol. Toxicol.37(8), e23397 (2023). [DOI] [PubMed] [Google Scholar]
  • 27.Cuccaro, A., Freitas, R., De Marchi, L., Oliva, M. & Pretti, C. UV-filters in marine environments: a review of research trends, meta-analysis, and ecotoxicological impacts of 4-methylbenzylidene-camphor and benzophenone-3 on marine invertebrate communities. Environ. Sci. Pollut. Res. Int.29(43), 64370–64391 (2022). [DOI] [PubMed] [Google Scholar]
  • 28.Lax, C. et al. The effect of sunscreen 4-methylbenzylidene camphor in different and reproductive models, its bioaccumulation and molecular effects on ligand-receptor interaction, and protein expression. Basic Clin. Pharmacol. Toxicol.133(2), 130–141 (2023). [DOI] [PubMed] [Google Scholar]
  • 29.Nohynek, G. J., Borgert, C. J., Dietrich, D. & Rozman, K. K. Endocrine disruption: fact or urban legend?. Toxicol. Lett.223(3), 295–305 (2013). [DOI] [PubMed] [Google Scholar]
  • 30.Komrakova, M. et al. Impact of 4-methylbenzylidene camphor, daidzein, and estrogen on intact and osteotomized bone in osteopenic rats. J. Endocrinol.211(2), 157–168 (2011). [DOI] [PubMed] [Google Scholar]
  • 31.Nguyen, H. T., Truong, V. A. & Tran, P. H. Preparation of polysubstituted imidazoles using AC-SO(3)H/[Urea](7)[ZnCl(2)](2) as an efficient catalyst system: a novel method, and α-glucosidase inhibitor activity. RSC Adv.13(18), 12455–12463 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Mary, Y. S. et al. Detailed quantum mechanical, molecular docking, QSAR prediction, photovoltaic light harvesting efficiency analysis of benzil and its halogenated analogues. Heliyon5(11), e02825 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Yaylı, N. et al. Synthesis, biological evaluation (antioxidant, antimicrobial, enzyme inhibition, and cytotoxic) and molecular docking study of hydroxy methoxy benzoin/benzil analogous. Bioorg. Chem.115, 105183 (2021). [DOI] [PubMed] [Google Scholar]
  • 34.Zhang, L. et al. Exposure to pyrazosulfuron-ethyl induces immunotoxicity and behavioral abnormalities in zebrafish embryos. Fish Shellfish Immunol.131, 119–126 (2022). [DOI] [PubMed] [Google Scholar]
  • 35.Wang, C. et al. Review of emerging contaminant tris(1,3-dichloro-2-propyl)phosphate: Environmental occurrence, exposure, and risks to organisms and human health. Environ. Int.143, 105946 (2020). [DOI] [PubMed] [Google Scholar]
  • 36.Zhong, L. et al. IPPD-induced growth inhibition and its mechanism in zebrafish. Ecotoxicol. Environ. Saf.239, 113614 (2022). [DOI] [PubMed] [Google Scholar]
  • 37.Wolff, S. Aspects of the adaptive response to very low doses of radiation and other agents. Mutat. Res.358(2), 135–142 (1996). [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Information 1. (19.6KB, docx)
Supplementary Information 2. (69.9KB, docx)

Data Availability Statement

All data generated or analyzed during this study are included in this manuscript.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES