Skip to main content
NCBI home page
Environment & Health logoLink to Environment & Health
. 2025 Sep 26;4(2):236–245. doi: 10.1021/envhealth.5c00143

Impact of Ozone Exposure on Oral Microbiome: A Controlled, Randomized, Crossover Trial

Hao Zhou 1, Shuyun Jiang 1, Liying Zhang 1, Jian Zhao 1, Tianzong Zhou 1, Luyun Zhang 1, Yuliang Ma 1, Shuhan Liu 1, Zhengkang You 1, Fengru Li 1, Shen Chen 1, Liping Chen 1, Qing Wang 1, Xiumei Xing 1, Wen Chen 1, Daochuan Li 1,*
PMCID: PMC12930307  PMID: 41743807

Abstract

Environmental ozone pollution may adversely impact human health, while dysbiosis is implicated in various health outcomes. However, the effects of ozone exposure on the human microbiome remain unclear. This study conducted a controlled, randomized, crossover ozone exposure trial with 29 participants exposed to filtered air and ozone. Oral microbiome samples were collected for 16S rRNA sequencing. A mixed-effects model was used to analyze the impact of ozone exposure on the diversity and composition of the oral microbiome, as well as their association with lung functions. The results showed that the ozone concentration remained stable at 5 ± 5.1 parts per billion (ppb) under filtered air and 282.2 ± 44 ppb during ozone exposure. Acute ozone exposure significantly impaired lung function and reduced oral microbiome diversity, with forced vital capacity (FVC) decreasing by 0.41 L, expiratory volume in one second (FEV1) by 0.36 L, and peak expiratory flow (PEF) by 0.81 L, while observed species richness (Sobs), abundance-based coverage estimator (Ace), and chao indices (Chao) decreased by 21.83, 22.24, and 21.87%, respectively. Males exhibited greater sensitivity to ozone exposure than females, reflected in both reduced lung function and alterations in the oral microbiome. The altered microbiome was mainly enriched in metabolic pathways. Among the microbiome, Treponema medium exhibited a significant correlation and mediating effect on ozone-induced lung function impairment, which may serve as a sensitive indicator of changes in lung function. These findings provide new insights into the health effects caused by ozone exposure.

Keywords: ozone, randomized controlled trial, oral microbiome, lung function, treponema medium

graphic file with name eh5c00143_0007.jpg

graphic file with name eh5c00143_0005.jpg

1. Introduction

Ozone has become a major air pollutant in the world, with increasing pollution concentration and expanding pollution range. Generated by photochemical reactions of nitrogen oxides (NOX) and volatile organic compounds (VOCs) emitted from motor vehicle exhaust and industrial waste gas, ozone exposure can adversely affect the respiratory, nervous, and cardiovascular systems, contributing to diseases such as asthma, chronic obstructive pulmonary disease (COPD), ischemic heart disease, and Alzheimer’s disease (AD). , The U.S. Environmental Protection Agency (EPA) has conducted a systematic assessment of the hazards of ozone exposure to human health and believes that there is a clear causal relationship between ozone exposure and lung damage. So far, the mortality rate caused by global ozone pollution is still on the rise, with a relative risk of death from respiratory causes increasing by 4.0% for every 10 parts per billion (ppb) increase in ozone concentration. Therefore, it is essential to understand the damage characteristics of ozone exposure to the human respiratory system.

Previous studies indicated that the microbiome could be affected by ozone exposure, leading to respiratory damage and metabolic changes. The proximity and continuity of the oral cavity and the lower respiratory tract make the oral microbiome the main source of the lung microbiome. Compared with other microbiomes, the oral microbiome can directly contact environmental ozone and is more susceptible to ozone exposure. Ozone is a highly active, oxidizing gas with low solubility in water. Since the removal rate of ozone inhaled by the mouth is lower than that of nasal inhalation, ozone in the mouth has a higher lung penetration rate. The human oral cavity harbors over 700 different bacterial species, with the predominant bacterial taxa that constitute the oral microbiome typically remaining conserved across individuals. The influence of the oral microbiome extends beyond the oral cavity, as oral-related microorganisms have been detected in distant organs, including the small intestine, lung, heart, placenta, and brain. Changes in the oral microbiome are also associated with the occurrence and development of respiratory diseases such as pneumonia, COPD, lung cancer, cystic fibrosis, and asthma. However, the impact of ozone exposure on oral microbiome is not fully understood.

Randomized Controlled Trials (RCTs) are recognized by the Cochrane Center in the United Kingdom as the highest level of evidence in evidence-based medicine and are widely regarded as the gold standard in clinical research. By effectively controlling for confounding factors and directly comparing the effects of different interventions, RCTs provide evidence with superior reliability and validity compared to other cross-sectional studies. In recent years, RCTs have been extensively applied in research on air pollutants such as ozone. Notably, several studies have employed exposure chambers with filtered air and ozone to investigate the effects of acute ozone exposure on the respiratory and cardiovascular systems in both elderly and young individuals. , These studies have also examined the impacts of ozone on blood metabolomics, transcriptomics, and proteomics, offering novel insights. − Nevertheless, evidence regarding the effects of acute ozone exposure on oral microbiome remains scarce.

In this study, we established an ozone exposure chamber and conducted a controlled, randomized crossover trial to investigate the effects of ozone exposure on respiratory function and oral microbiome in healthy adults, as well as their relationship between them, providing new evidence for exploring the health effects of ozone exposure on the human body.

2. Method

2.1. Participants

This study was approved by the Biomedical Research Ethics Review Committee of the School of Public Health, Sun Yat-sen University (SYSU Public Health Medical Ethics 2024, No. 109, Figure S4). Participants were recruited from students at Sun Yat-sen University in Yuexiu District, Guangzhou, Guangdong, China. All participants completed a standardized training protocol and provided written informed consent following full disclosure of the study objectives, potential risks, and participant rights. The consent process specifically guaranteed participants’ rights to immediate notification of any study-related adverse events and to unconditional withdrawal at their discretion without penalty.

Prior to the study, personal information was collected from all participants, including age, gender, and body mass index (BMI). Eligibility criteria included Han ethnicity, age between 18 and 26 years, and a BMI ranging from 17.5 to 26 kg/m. Furthermore, participants were required to be nonsmokers (both active and passive), nonalcohol consumers, free from allergic disorders, underlying or chronic diseases, and had not taken any medications or dietary supplements in the past 3 months. Additionally, to ensure environmental stability for participants, they were instructed to refrain from leaving Guangzhou throughout the study period. Furthermore, participants were required to have no prior history of surgical intervention and to be free from active infections or febrile conditions. The ethical statement and sample size estimation are provided in the Supporting Information.

2.2. Study Design

In July 2024, we conducted this controlled, randomized, crossover ozone exposure trial. A randomization program was developed in R software, where the random seed was initialized using the “set.seed” function to ensure reproducibility, followed by random permutation of samples via the “sample_frac” function and equitable partitioning into two distinct experimental groups. Each participant was exposed to either filtered air or 280 ppb ozone, with each exposure lasting 2 h. This concentration lies between the levels used in previous studies, 200 and 300 ppb, and is also closer to the highest monthly average daily maximum 8 h ozone concentration of 286 ppb in Beijing, China. ,,− Under standard temperature and pressure conditions (25 °C, 1 atm), an ozone concentration of 100 ppb corresponds to 196 μg/m3. However, in this study, participants were exposed to this concentration for only 2 h, whereas in real-world scenarios, individuals would be exposed for longer durations. The two exposure groups were separated by a 2-week washout period, a duration adopted in multiple existing studies, which suggested that subjects could fully recover to their pre-exposure state after 2 weeks. ,, The study was conducted in a blinded design, with all participants unaware of their exposure conditions. To minimize dietary influences, a standardized breakfast and purified water were provided to each participant before the exposure sessions. Furthermore, the experiments were conducted on weekend mornings and completed within one month to control for potential temporal and seasonal effects.

During the 2-h exposure, participants alternated between 10 min of rest and 10 min of jumping jacks. An individualized exercise intensity was determined based on each participant’s capacity, and their average minute ventilation during exercise (VE) was measured using the Spirolab III Portable Spirometer (Mir Thoughts Medical Technology Co. Ltd., China). The target VE range was set at 7–11 L/min/m2 of body surface area for all participants to ensure comparable ozone inhalation levels.

2.3. Exposure Chamber Design

The exposure chamber used in this study measured 8.69 m × 8.13 m × 4.3 m and was thoroughly cleaned prior to the experiment (Figure S1). Participants were seated in dispersed positions across the chamber according to their assigned numbers. Throughout both the filtered air and ozone exposure sessions, environmental air was filtered using two air purifiers (Third Generation, Xiaomi, China), each equipped with specific filters capable of removing fine particulate matter (PM2.5) and total volatile organic compounds (TVOC) from the chamber. Two ozone generators (CH-ZTW, Guangzhou Chuanghuan Ozone Electric Equipment Co. Ltd., China) were placed at the center of the chamber. The generators were set with parameters of a flow rate of 3 L/min, ampere meter reading of 0.23 A, and ozone concentration of 40%. These parameters were pretested extensively to ensure a stable ozone concentration of 280 ppb during the exposures. Ozone concentrations in the chamber were monitored at 10-s intervals using three pump-suction electrochemical ozone detectors (GT-903-O3, Shenzhen Ke’ernuo Electronics Technology Co. Ltd., China). Additionally, four air quality detectors (Hong Kong Minhui Enterprise Co. Ltd., China) were placed in the chamber to measure TVOC, PM2.5, and carbon dioxide (CO2) levels, with measurements recorded every 20 min.

2.4. Health Measurements and Oral Sample Collection

After both air and ozone exposures, participants’ forced expiratory volume in one second (FEV1), forced vital capacity (FVC), and peak expiratory flow (PEF) were measured using a Spirometer Sp10 (Contec Medical Systems Co., Ltd., China). Subsequently, participants were instructed to rinse their mouths thoroughly with physiological saline for 10 s before sampling. The first swab was used to collect oral microbiome by moving it across the inner cheeks, upper and lower gums for 1 min. The second swab was swiped five times over the tonsillar crypts and lateral walls of the pharynx. The tips of both swabs were placed in the same 2 mL sterile tube, initially stored in liquid nitrogen, and later transferred to a −80 °C freezer for subsequent sequencing analysis.

2.5. Microbiome Assay and Enrichment Analysis

The analysis consisted of three main parts. First, genomic DNA of the oral microbiome was extracted using the MagPure Stool DNA KF Kit B (MAGEN, Guangzhou, China). Then, PCR amplification of the V3–V4 regions was performed using the primer pair: forward primer 5′-ACTCCTACGGGAGGCAGCAG-3′ and reverse primer 5′-GGACTACHVGGGTWTCTAAT-3′. Sequencing was carried out on the Illumina Miseq/Hiseq platform (BGI, Shenzhen, China). Finally, raw data underwent filtering and quality control to cluster operational taxonomic units (OTUs), and taxonomic annotation was conducted using the Silva database. Detailed protocols for DNA extraction, sequencing, and data processing are provided in the Supporting Information.

Following microbiome profiling, α-diversity indices were used to evaluate the species diversity of the participants’ oral microbiome, including shannon index (Shannon), simpson index (Simpson), goods coverage (Coverage), observed species richness (Sobs), chao index (Chao), and abundance-based coverage estimator (Ace). Specifically, both Shannon and Simpson indices measure community diversity, with the former emphasizing richness and the latter evenness. Coverage indicates the detection probability of strains in the sample. Sobs represents the actual species number, while ACE and Chao1 denote the estimated species number.

The relative abundance values were multiplied by 100 and then log-transformed using base 10, resulting in an approximate normal distribution. For species with zero relative abundance, half the smallest nonzero abundance across all samples was used to calculate the log-transformed value. Additionally, functional annotation of microbiome was performed using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg). Enriched biological processes were identified at three levels of annotation, and results were saved as absolute abundance values.

2.6. Statistical Analysis

Linear mixed-effects (LME) models were employed to analyze the impact of ozone exposure on lung function, diversity and composition of the oral microbiome and KEGG pathways analysis. Ozone exposure was treated as the independent variable, categorized as a binary variable. Participant ID was included as a random intercept to account for individual differences. Covariates such as BMI, gender, and age were included in the models, and the α-diversity indices were log-transformed to achieve an approximate normal distribution.

In sensitivity analyses, three sets of models were fitted. First, to assess the robustness of the single-pollutant models, pollutants such as PM2.5, CO2, and TVOC would be incorporated into the model. Prior to this, a linear regression model was employed to evaluate collinearity among all air pollutants. In the presence of collinearity, regression residuals were used as substitutes. Second, exposure order was added to the model to evaluate potential effects of exposure sequence. Third, VE per unit body surface area, which varied among participants during exposure, was included as a covariate to account for its potential impact.

Spearman correlation tests were used to assess associations between the oral microbiome and lung function indicators. Subsequently, mediation analysis was performed to examine the mediation effects in this association. Specifically, the effect was partitioned into two components: average direct effects (ADE) and average causal mediation effects (ACME). A statistically significant ACME indicated that ozone impacted lung function via the mediating effect of microbiome. Additionally, sex-stratified analysis was performed to explore whether ozone-induced changes in microbiome differed by gender.

All statistical analyses were conducted in R Studio (Version 4.3) using the ″lme4″ and ″mediation″ packages for LME modeling and mediation analysis. Data were presented as means ± standard deviations. All statistical tests were two-sided, and p values < 0.05 were considered statistically significant.

3. Results

3.1. Participant Characteristics and Exposure Levels

A total of 88 participants were recruited for this study. Of these, 36 were excluded due to cross-city travel, and 22 declined participation due to scheduling conflicts, resulting in 30 participants ultimately enrolled (Figure ). One participant withdrew due to menstrual-related physical discomfort, leaving 29 participants who completed both exposure experiments and sample collections (19 males, 10 females). The mean age was 21.97 ± 2.29 years, with a mean BMI of 21.51 ± 1.83 kg/m2 (Table S1). No significant differences were observed between the baseline characteristics of the two exposure groups.

1.

1

Open in a new tab

Flowchart of participant recruitment. Diagram illustrating the process of participant recruitment, screening, and group assignment.

The mean ozone concentration was 282.2 ± 44 ppb during ozone exposure, while it was 5 ± 5.1 ppb during filtered air exposure. Air quality in the exposure chamber was monitored throughout the study (Table S2). The temperature and humidity were maintained at comparable levels of 23.2 ± 1.2 °C and 57.8 ± 2.7%, respectively. In addition, other air pollutants including PM2.5, TVOC and CO2 were also examined, and their exposure levels were very low in both groups. During exposure to ozone or filtered air, the actual VE of all participants was controlled between 7–11L/min/m2 of body surface area to ensure the uniformity of exposure.

3.2. Effects of Ozone on Lung Function

The lung functions of all participants were examined and the results showed that compared with filtered air, ozone exposure led to an immediate and significant decline in lung function of the participants. After ozone exposure, the mean FVC, FEV1 and PEF decreased by 11.98% (−0.41 L, 95% CI: −0.63 L to −0.20 L), 12.22% (−0.36 L, 95% CI: −0.59 L to −0.13L), and 14.21% (−0.81 L/min, 95% CI: −1.30 L/min to −0.31 L/min), respectively (Figure A; Table S2). It was noteworthy that the FEV1, FVC and PEF in males significantly decreased by 14.23% (−0.50 L, 95% CI: −0.80 L to −0.21 L), 15.41% (−0.47 L, 95% CI: −0.75 L to −0.20 L), and 17.15% (−1.03 L/min, 95% CI: −1.68 L/min to −0.37 L/min) after ozone exposure, respectively. However, no changes were observed in females (Figure A-B; Table S4). The above results indicate that there are gender differences in the decline of lung function caused by ozone.

2.

2

Open in a new tab

Results on lung function and oral microbiota characteristics. This figure shows the results of lung function (A) α-diversity of oral microbiota (B), the composition of oral microbiota (C), and the KEGG functional enrichment results of oral microbiota at different taxonomic levels, level 1 (D) and level 2 (E). In all panels, the x-axis represents the effect estimates ± 95% confidence intervals. Red indicates a statistically significant difference, with * indicating p < 0.05, ** indicating p < 0.01, and *** indicating p < 0.001. Blue indicates no statistically significant difference. Note: FEV1/FVC, forced expiratory volume in 1 s/forced vital capacity; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; PEF, peak expiratory flow; Simpson, Simpson index; Coverage, goods coverage; Shannon, Shannon index; Sobs, observed species; Ace, abundance-based coverage estimator; Chao, Chao index; Haemophilus sp., Haemophilus sputorum.

3.

3

Open in a new tab

Stratified analysis results by sex. This figure includes the stratified analysis of sex-specific results for various outcomes: male pulmonary function (A), female pulmonary function (B); male composition of oral microbiota (C), female composition of oral microbiota (D); male KEGG functional enrichment at level 1 (E), female KEGG functional enrichment at level 1 (F). In all panels, red indicates a statistically significant difference, with * indicating p < 0.05, ** indicating p < 0.01, and *** indicating p < 0.001. Blue indicates no statistically significant difference. Note: FEV1/FVC, forced expiratory volume in 1 s/forced vital capacity; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; PEF, peak expiratory flow; Simpson, Simpson index; Coverage, goods coverage; Shannon, Shannon index; Sobs, observed species; Ace, abundance-based coverage estimator; Chao, Chao index.

3.3. Effects of Ozone on Oral Microbiome Diversity and Composition

For the oral microbiome of the participants, the α-diversity was significantly changed by ozone exposure. Specifically, Sobs decreased by 21.83% (95% CI: −41.74% to −1.93%), Ace decreased by 22.24% (95% CI: −42.19% to −2.28%), and Chao decreased by 21.87% (95% CI: −41.78% to −1.95%) in ozone exposure group compared with filtered air. No significant differences were observed in other indices (Figure B; Table S5). These findings suggest ozone-driven selective suppression of microbial diversity, manifested through pronounced declines in species numbers, independent of richness/evenness alterations. For the oral composition, the identified OTUs were classified into 9 phyla, 13 classes, 25 orders, 38 families, 49 genera, and 76 species. At the phylum level, Firmicutes, Proteobacteria, and Bacteroidota accounted for more than 80% of the total abundance. At the genus level, the most abundant genera, with average abundances exceeding 10%, were Streptococcus (20.4%), Prevotella (13.7%), Veillonella (12.5%), and Haemophilus (11.4%). Using the LME model, one order and four species of microbiome were found to be associated with ozone exposure. Specifically, Treponema medium, Actinomyces naeslundii, and Haemophilus sputorum were positively associated with ozone exposure, showing increases of 107.49% (95% CI: 26.77% to 239.63%), 54.88% (95% CI: 2.80% to 133.35%), and 79.89% (95% CI: 8.39% to 198.54%), respectively. In contrast, Capnocytophaga sputigena and Flavobacteriales were negatively associated with ozone exposure, with decreases of 30.66% (95% CI: −50.80% to −2.50%) and 36.03% (95% CI: −56.65% to −5.81%), respectively (Figure C; Table ). For microbiome of different genders, significant changes were observed in 14 species in males and 7 species in females, with Treponema medium being the only overlapping species which showed a consistent increase of 96.06% (95% CI: 4.09% to 269.32%) in males and 135.14% (95% CI: 13.43% to 387.45%) in females (Figure C–D; Table S6).

1. Changes in Microbiome Composition.

outcome estimate std error t value p value lower CI upper CI
Treponema medium 0.317 0.109 2.898 0.007 0.103 0.531
Actinomyces naeslundii 0.190 0.091 2.097 0.045 0.012 0.368
Haemophilus sp. 0.255 0.112 2.267 0.031 0.035 0.475
Capnocytophaga sputigena –0.159 0.076 –2.103 0.044 –0.308 –0.011
Flavobacteriales –0.194 0.086 –2.259 0.032 –0.363 –0.026
Open in a new tab
a

Change in the log-transformed relative abundance values of the oral microbiome.

3.4. Enrichment Analysis of Oral Microbiome-Related Pathways

To identify the potential function of the differential microbiome, KEGG pathway and the LME model analysis were used to find out the significant pathways they got involved. All pathways were classified into three levels, at the first level, the most significantly affected pathway was metabolism, followed by genetic information processing, cellular processes, and organismal systems (Figure D; Table S7). The altered pathways enriched in males were consistent with the overall results, while no significant pathway changes were found in females (Figure E-F; Table S8). At the second level, detailed pathway alterations were identified, including changes in the metabolism of vitamins, amino acids, carbohydrates, glycans, nucleotides, and terpenoids and polyketides. Additionally, various biological processes were affected, including biosynthesis, DNA replication, repair, translation, and RNA transcription (Figure E; Table S7). In the stratified analysis, similar pathway changes were observed in males. In contrast, only a single immunity-related pathway was identified in females (Figure S2; Table S8). We also analyzed the changes in specific substances within each pathway, categorized as the third level. Among these, metabolic changes still accounted for the largest number of alterations, including substances such as d-glutamine, d-alanine, and thiamine (Figure S3).

3.5. Correlation and Mediation Analysis of Oral Microbiome and Lung Function

Spearman correlation analysis was conducted to examine associations between the oral microbiome and lung function indicators. The results showed significant positive correlations between Treponema medium and Flavobacteriales with both FVC (r = 0.407, p = 0.031; r = 0.382, p = 0.003) and FEV1 (r = 0.478, p = 0.01; r = 0.269, p = 0.041), respectively (Figure ; Table S9). Based on these results, mediation analysis was performed Among all microbiome, and the results showed that Treponema medium exhibited both a statistically significant ADE (p < 0.001) and ACME (p = 0.036) for FVC, indicating a mediating role of Treponema medium in the decline of FVC (Table ).

4.

4

Open in a new tab

Correlation and mediation analysis of oral microbiome and lung function. This figure presents the correlation analysis results between differential oral microbiome and lung function indicators. Red indicates a positive correlation, while blue indicates a negative correlation. * Denotes p < 0.05, ** denotes p < 0.01, and *** denotes p < 0.001. Note: FEV1/FVC, forced expiratory volume in 1 s/forced vital capacity; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; PEF, peak expiratory flow.

2. Mediation Effect Analysis in Microbiome and Lung Function .

outcome alpha diversity ACME ADE total effect p value ACME p value ADE
FVC Flavobacteriales –0.047 –0.368 –0.415 0.178 0.004
FEV1 Flavobacteriales –0.007 –0.358 –0.365 0.898 0.028
PEF Flavobacteriales 0.016 –0.842 –0.826 0.854 0.066
FEV1/FVC Flavobacteriales 0.010 –0.016 –0.007 0.352 0.690
FVC Treponema medium 0.111 –0.526 –0.415 0.036 0.000
FEV1 Treponema medium 0.092 –0.457 –0.365 0.126 0.016
PEF Treponema medium 0.163 –0.989 –0.826 0.352 0.068
FEV1/FVC Treponema medium 0.000 –0.007 –0.007 0.950 0.844
FVC Actinomyces naeslundii –0.034 –0.382 –0.415 0.362 0.004
FEV1 Actinomyces naeslundii –0.021 –0.343 –0.365 0.766 0.054
PEF Actinomyces naeslundii –0.010 –0.816 –0.826 0.924 0.144
FEV1/FVC Actinomyces naeslundii –0.002 –0.005 –0.007 0.890 0.920
FVC Haemophilus sputorum 0.007 –0.422 –0.415 0.818 0.002
FEV1 Haemophilus sputorum 0.007 –0.372 –0.365 0.842 0.012
PEF Haemophilus sputorum 0.004 –0.830 –0.826 0.908 0.098
FEV1/FVC Haemophilus sputorum 0.000 –0.007 –0.007 0.870 0.824
FVC Capnocytophaga sputigena 0.001 –0.416 –0.415 0.994 0.006
FEV1 Capnocytophaga sputigena –0.011 –0.353 –0.365 0.856 0.032
PEF Capnocytophaga sputigena 0.007 –0.833 –0.826 0.932 0.064
FEV1/FVC Capnocytophaga sputigena –0.005 –0.002 –0.007 0.558 0.996
Open in a new tab
a

ADE, average direct effects; ACME, average causal mediation effects; FVC, forced vital capacity; FEV1, forced expiratory volume in 1 s; PEF, peak expiratory flow and FEV1/FVC, FEV1 to FVC Ratio.

3.6. Results of Sensitivity Analysis

Pollutants (PM2.5, CO2, and TVOC), exposure order, and VE were included in LME model for sensitivity analysis. We observed that once air pollutants were adjusted for in Model 2, some statistically significant associations in the third level of KEGG pathways disappeared. Aside from this, most results across the unadjusted model and the three adjusted models remained consistent, indicating that our model and results are relatively robust (Table S10).

4. Discussion

In this study, we reinforced the detrimental effects of ozone exposure on respiratory system of human populations by using a controlled, randomized, crossover trial in ozone exposure chamber. Interestingly, we found that respiratory function of males was more sensitive to ozone exposure than that of females. Besides, our findings suggested that ozone exposure could alter oral microbiome, which also had gender differences. Ozone could reduce the oral microbiome diversity which might be associated with the impaired lung function. Treponema medium was found to be mediating the decline of lung function. These findings suggest that disturbances in the oral microbiome may contribute to ozone induced decline in respiratory function.

Current research suggests that both short-term and long-term ozone exposure can lead to declines in pulmonary function. Specifically, multiple crossover and cohort studies have demonstrated that ozone exposure leads to declines in FVC, PEF, and FEV1, which are consistent with our results. − Regarding the FEV1/FVC ratio, some studies have reported that it remains unaffected following ozone exposure. This is likely because ozone exerts similar effects on both FVC and FEV1, resulting in parallel changes that are not reflected in the ratio. Furthermore, our study performed stratified analyses by sex and observed similar results. Multiple lung function indicators were significantly reduced in males, while no changes were observed in females, which was supported by existing animal studies. For instance, studies on C57BL/6J mice have shown that males experience greater effects than females in terms of airway hyperresponsiveness, lung and gut microbiome. , Some mechanistic investigations suggest that this difference may be mediated by the microbiome through the interleukin-33/ST2 pathway, as the knockout of ST2 has abolished the sex differences in the response to ozone.

Studies investigating ozone-induced changes in microbiome are limited. Environmental studies indicate that ozone exposure could not only reduce fungal and microbial communities in soil and water, but also exhibit bactericidal effects on antibiotic-resistant microbiome, highlighting its potential for various disinfection applications. − In humans, a few studies have explored the relationship between ozone exposure and respiratory microbiome. A similar crossover experiment has found that acute ozone exposure reduces the α-diversity in the nasal microbiome. Moreover, a 7-day follow-up study has indicated that ozone exposure can also affect the α-diversity of the oral microbiome. Consistently, our findings suggest that ozone exposure could significantly reduce the Sobs, Ace, and Chao of the oral microbiome, confirming that acute ozone exposure may have a negative impact on the oral microbiome. Regarding the differences in the microbiome, the number of changes in the microbiome in females was lower than that in males, similar to the decline in lung function, suggesting that the microbiome in males is more sensitive than that in females.

For the microbiome in our study, Actinomyces naeslundii, Haemophilus sputorum, Treponema medium, Capnocytophaga sputigena, and Flavobacteriales were found to be significantly different after ozone exposure. Most of them were reported to be associated with oral diseases such as dental caries and plaques, periodontitis, and oral squamous cell carcinoma. − Haemophilus sputorum was also found to play a dominant role in pulmonary inflammation and could be detected in approximately 60% of patients of purulent nasopharyngitis. − Besides, it is also believed that the oral microbiome may affect the human body’s response to pollutants, as the microbiome can metabolize pollutants and regulate inflammatory responses. For example, smoking could reduce the abundance of oral microbiome while increasing pathogen abundance, further exacerbating pulmonary inflammation. , Our study revealed that pathways related to metabolism, genetic information, cellular processes, were predominantly affected by the disturbance of oral microbiome. The change of this pathway suggests that oral microbiome may get involved in mediating the biological effects of ozone. Our results also indicate that Treponema medium is a mediating effect on FVC changes. Therefore, these findings provide the possibility that oral microbiome may play a role in the effects of ozone exposure, such as mediating respiratory system damage. However, the effects and specific mechanisms of oral microbiome in ozone-related biological effects still need to be clarified in future research.

This study employed a randomized, crossover and blind cohort design, where all participants were screened and randomly assigned to groups. Individual differences in diet, lung ventilation, seasonal variations, temperature, humidity, and environmental pollutants were controlled, and mixed-effects models were used to adjust for confounding factors. These elements ensured a high level of scientific rigor in the study design and methodology. However, several limitations remain. We attempted to apply a false discovery rate (FDR) correction for multiple comparisons in the microbiome analysis; however, none of the results reached statistical significance. Therefore, unadjusted p values were retained for the exploratory analysis. Additionally, this study did not conduct in-depth mechanistic investigations into oxidative stress or inflammatory responses, which require further exploration in subsequent research.

In summary, this study was a controlled, randomized, crossover ozone exposure trial, which found that acute ozone exposure led to a decrease in lung function and oral microbiome diversity in participants which were more pronounced in males than in females. For the microbial species, Treponema medium was identified as a sensitive indicator and might play a role in mediating the effects of ozone exposure. These findings provide new insights into the biological effects induced by ozone exposure.

Supplementary Material

eh5c00143_si_001.pdf (936.4KB, pdf)

Acknowledgments

We sincerely thank all the participants for their dedication and contribution to this study.

Glossary

Abbreviations

NOX

nitrogen oxides

VOCs

volatile organic compounds

AD

Alzheimer’s disease

RCTs

randomized controlled trials

TVOC

total volatile organic compounds

COPD

chronic obstructive pulmonary disease

PM2.5

fine particulate matter

CO2

carbon dioxide

FEV1

forced expiratory volume in 1 s

FVC

forced vital capacity

PEF

peak expiratory flow

FEV1/FVC

forced expiratory volume in 1 s/forced vital capacity

Coverage

goods coverage

Chao

Chao index

Simpson

Simpson index

Shannon

Shannon index

Sobs

observed species richness

Ace

abundance-based coverage estimator

OTUs

operational taxonomic units

PICRUSt

Phylogenetic Investigation of Communities by Reconstruction of Unobserved States

KEGG

Kyoto Encyclopedia of Genes and Genomes

LME

linear mixed-effects

VE

average minute ventilation during exercise

BMI

body mass index

ADE

average direct effects

ACME

average causal mediation effects

ROS

reactive oxygen species

FDR

false discovery rate

ppb

parts per billion

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/envhealth.5c00143.

  • Additional experimental details, supplementary figures, supplementary tables, and ethical statement for studies involving human participants (PDF)

1.

H.Z. and S.J. contributed equally to this work.

This work was supported by the State Key Program of National Natural Science of China (82130094).

The authors declare no competing financial interest.

References

  1. Li K., Jacob D. J., Liao H., Shen L., Zhang Q., Bates K. H.. Anthropogenic drivers of 2013–2017 trends in summer surface ozone in China. Proc. Natl. Acad. Sci. U. S. A. 2019;116(2):422–427. doi: 10.1073/pnas.1812168116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Dang R., Liao H., Fu Y.. Quantifying the anthropogenic and meteorological influences on summertime surface ozone in China over 2012–2017. Sci. Total Environ. 2021;754:142394. doi: 10.1016/j.scitotenv.2020.142394. [DOI] [PubMed] [Google Scholar]
  3. Zhang J. J., Wei Y., Fang Z.. Ozone Pollution: A Major Health Hazard Worldwide. Front Immunol. 2019;10:2518. doi: 10.3389/fimmu.2019.02518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Smith A. E., Glasgow G.. Integrated Uncertainty Analysis for Ambient Pollutant Health Risk Assessment: A Case Study of Ozone Mortality Risk. Risk Anal. 2018;38(1):163–176. doi: 10.1111/risa.12828. [DOI] [PubMed] [Google Scholar]
  5. Turnock S. T., Reddington C. L., West J. J., O’Connor F. M.. The Air Pollution Human Health Burden in Different Future Scenarios That Involve the Mitigation of Near-Term Climate Forcers, Climate and Land-Use. Geohealth. 2023;7(8):e2023GH000812. doi: 10.1029/2023GH000812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Nuvolone D., Petri D., Voller F.. The effects of ozone on human health. Environ. Sci. Pollut Res. Int. 2018;25(9):8074–8088. doi: 10.1007/s11356-017-9239-3. [DOI] [PubMed] [Google Scholar]
  7. Baker J. L., Mark Welch J. L., Kauffman K. M., McLean J. S., He X.. The oral microbiome: diversity, biogeography and human health. Nat. Rev. Microbiol. 2024;22(2):89–104. doi: 10.1038/s41579-023-00963-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Baker J. L., Bor B., Agnello M., Shi W., He X.. Ecology of the Oral Microbiome: Beyond Bacteria. Trends Microbiol. 2017;25(5):362–374. doi: 10.1016/j.tim.2016.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Garmendia J., Cebollero-Rivas P.. Environmental exposures, the oral-lung axis and respiratory health: The airway microbiome goes on stage for the personalized management of human lung function. Microb Biotechnol. 2024;17(6):e14506. doi: 10.1111/1751-7915.14506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Higgins, J. ; Thomas, J. ; Chandler, J. ; Cumpston, M. ; Li, T. ; Page, M. ; Welch, V. . Cochrane Handbook for Systematic Reviews of Interventions version 6.5; Cochrane, 2024. [Google Scholar]
  11. Dijkgraaf M. G. W., Haukoos J., Itani K. M. F.. Practical Guide to Design Choice of Randomized Clinical Trials in Surgery. JAMA Surg. 2022;157(12):1154–1155. doi: 10.1001/jamasurg.2022.4889. [DOI] [PubMed] [Google Scholar]
  12. Devlin R. B., Duncan K. E., Jardim M., Schmitt M. T., Rappold A. G., Diaz-Sanchez D.. Controlled exposure of healthy young volunteers to ozone causes cardiovascular effects. Circulation. 2012;126(1):104–111. doi: 10.1161/CIRCULATIONAHA.112.094359. [DOI] [PubMed] [Google Scholar]
  13. Arjomandi M., Balmes J. R., Frampton M. W., Bromberg P., Rich D. Q., Stark P., Alexis N. E., Costantini M., Hollenbeck-Pringle D., Dagincourt N.. et al. Respiratory Responses to Ozone Exposure. MOSES (The Multicenter Ozone Study in Older Subjects) Am. J. Respir Crit Care Med. 2018;197(10):1319–1327. doi: 10.1164/rccm.201708-1613OC. [DOI] [PubMed] [Google Scholar]
  14. Niu Y., Li H., Wang W., Wang C., Liu C., Du X., Zhang Q., Li J., Shi S., Meng X.. et al. Ozone exposure and prothrombosis: Mechanistic insights from a randomized controlled exposure trial. J. Hazard Mater. 2022;429:128322. doi: 10.1016/j.jhazmat.2022.128322. [DOI] [PubMed] [Google Scholar]
  15. Wang C., Lin J., Niu Y., Wang W., Wen J., Lv L., Liu C., Du X., Zhang Q., Chen B.. et al. Impact of ozone exposure on heart rate variability and stress hormones: A randomized-crossover study. J. Hazard Mater. 2022;421:126750. doi: 10.1016/j.jhazmat.2021.126750. [DOI] [PubMed] [Google Scholar]
  16. Du X., Niu Y., Wang C., Wang W., Liu C., Meng X., Chu C., Chen R., Kan H.. Ozone exposure and blood transcriptome: A randomized, controlled, crossover trial among healthy adults. Environ. Int. 2022;163:107242. doi: 10.1016/j.envint.2022.107242. [DOI] [PubMed] [Google Scholar]
  17. Fang Y., Du X., Ji X., Wang W., Wang C., Chen R., Niu Y., Kan H.. Genome-wide profiling of long non-coding RNA following ozone exposure: A randomized, controlled exposure trial. Environ. Res. 2024;263(Pt 2):120101. doi: 10.1016/j.envres.2024.120101. [DOI] [PubMed] [Google Scholar]
  18. National Urban Air Quality Report for June 2023. China National Environmental Monitoring Centre, https://www.cnemc.cn/jcbg/kqzlzkbg/202311/t20231103_1054516.shtml (accessed 2023 Nov 3). [Google Scholar]
  19. Niu Y., Chen R., Wang C., Wang W., Jiang J., Wu W., Cai J., Zhao Z., Xu X., Kan H.. Ozone exposure leads to changes in airway permeability, microbiota and metabolome: a randomised, double-blind, crossover trial. Eur. Respir. J. 2020;56(3):2000165. doi: 10.1183/13993003.00165-2020. [DOI] [PubMed] [Google Scholar]
  20. Tao; Wang. Strong ozone production in urban plumes from Beijing, China. Geophys. Res. Lett. 2006. [Google Scholar]
  21. Hu D., Cui L., Qi Y., Jia X., Chen J., Niu W., Miller M. R., Loh M., Zhou H., Deng F.. et al. Identification of potential markers for internal exposure to ambient ozone in oral cavity of healthy adults. Environ. Res. 2020;190:109907. doi: 10.1016/j.envres.2020.109907. [DOI] [PubMed] [Google Scholar]
  22. Tsiavos A., Antza C., Trakatelli C., Kotsis V.. The Microbial Perspective: A Systematic Literature Review on Hypertension and Gut Microbiota. Nutrients. 2024;16(21):3698. doi: 10.3390/nu16213698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Chen H., Masood S., Rappold A. G., Diaz-Sanchez D., Samet J. M., Tong H.. Effects of Controlled Ozone Exposure on Circulating microRNAs and Vascular and Coagulation Biomarkers: A Mediation Analysis. Noncoding RNA. 2023;9(4):43. doi: 10.3390/ncrna9040043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Xu F., Wu Q., Yang L., Sun H., Li J., An Z., Li H., Wu H., Song J., Chen W.. et al. Modification of gut and airway microbiota on ozone-induced airway inflammation. Sci. Total Environ. 2024;954:176604. doi: 10.1016/j.scitotenv.2024.176604. [DOI] [PubMed] [Google Scholar]
  25. Gao X., Xu Y., Cai Y., Shi J., Chen F., Lin Z., Chen T., Xia Y., Shi W., Zhao Z.. Effects of filtered fresh air ventilation on classroom indoor air and biomarkers in saliva and nasal samples: A randomized crossover intervention study in preschool children. Environ. Res. 2019;179(Pt A):108749. doi: 10.1016/j.envres.2019.108749. [DOI] [PubMed] [Google Scholar]
  26. Guo Q., Zhao Y., Zhao J., Bian M., Qian L., Xue T., Zhang J. J., Duan X.. Acute change of lung function to short-term exposure to ambient air pollutants with and without physical activity: A real-world crossover study. Environ. Pollut. 2023;316(Pt 1):120481. doi: 10.1016/j.envpol.2022.120481. [DOI] [PubMed] [Google Scholar]
  27. Li L., Zhang W., Liu S., Wang W., Ji X., Zhao Y., Shima M., Yoda Y., Yang D., Huang J.. et al. Cardiorespiratory effects of indoor ozone exposure during sleep and the influencing factors: A prospective study among adults in China. Sci. Total Environ. 2024;924:171561. doi: 10.1016/j.scitotenv.2024.171561. [DOI] [PubMed] [Google Scholar]
  28. Chen H., Tong H., Shen W., Montilla T. S., Case M. W., Almond M. A., Wells H. B., Alexis N. E., Peden D. B., Rappold A. G.. et al. Fish oil blunts lung function decrements induced by acute exposure to ozone in young healthy adults: A randomized trial. Environ. Int. 2022;167:107407. doi: 10.1016/j.envint.2022.107407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Bergamaschi E., De Palma G., Mozzoni P., Vanni S., Vettori M. V., Broeckaert F., Bernard A., Mutti A.. Polymorphism of quinone-metabolizing enzymes and susceptibility to ozone-induced acute effects. Am. J. Respir Crit Care Med. 2001;163(6):1426–1431. doi: 10.1164/ajrccm.163.6.2006056. [DOI] [PubMed] [Google Scholar]
  30. Birukova A., Cyphert-Daly J., Cumming R. I., Yu Y. R., Gowdy K. M., Que L. G., Tighe R. M.. Sex Modifies Acute Ozone-Mediated Airway Physiologic Responses. Toxicol. Sci. 2019;169(2):499–510. doi: 10.1093/toxsci/kfz056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Cho Y., Abu-Ali G., Tashiro H., Brown T. A., Osgood R. S., Kasahara D. I., Huttenhower C., Shore S. A.. Sex Differences in Pulmonary Responses to Ozone in Mice. Role of the Microbiome. Am. J. Respir. Cell Mol. Biol. 2019;60(2):198–208. doi: 10.1165/rcmb.2018-0099OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kasahara D. I., Wilkinson J. E., Cho Y., Cardoso A. P., Huttenhower C., Shore S. A.. The interleukin-33 receptor contributes to pulmonary responses to ozone in male mice: role of the microbiome. Respir Res. 2019;20(1):197. doi: 10.1186/s12931-019-1168-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Zhang K., Zentella R., Burkey K. O., Liao H. L., Tisdale R. H.. Long-term tropospheric ozone pollution disrupts plant-microbe-soil interactions in the agroecosystem. Glob Chang Biol. 2024;30(3):e17215. doi: 10.1111/gcb.17215. [DOI] [PubMed] [Google Scholar]
  34. Sun W., Lu Z., Zhang Z., Zhang Y., Shi B., Wang H.. Ozone and Fenton oxidation affected the bacterial community and opportunistic pathogens in biofilms and effluents from GAC. Water Res. 2022;218:118495. doi: 10.1016/j.watres.2022.118495. [DOI] [PubMed] [Google Scholar]
  35. Liu Y., Zhou H., Wang J., Liu S., Xie G. J., Liu B. F., Xing D.. Regulatory Mechanisms of Quorum Sensing System of Bacteria in Response to Chlorine and Ozone Disinfection. Environ. Sci. Technol. 2024 doi: 10.1021/acs.est.3c08305. [DOI] [PubMed] [Google Scholar]
  36. Hazucha M. J., Folinsbee L. J., Bromberg P. A.. Distribution and reproducibility of spirometric response to ozone by gender and age. J. Appl. Physiol (1985) 2003;95(5):1917–1925. doi: 10.1152/japplphysiol.00490.2003. [DOI] [PubMed] [Google Scholar]
  37. Schüpbach P., Osterwalder V., Guggenheim B.. Human root caries: microbiota in plaque covering sound, carious and arrested carious root surfaces. Caries Res. 1995;29(5):382–395. doi: 10.1159/000262097. [DOI] [PubMed] [Google Scholar]
  38. Umemoto T., Nakazawa F., Hoshino E., Okada K., Fukunaga M., Namikawa I.. Treponema medium sp. nov., isolated from human subgingival dental plaque. Int. J. Syst. Bacteriol. 1997;47(1):67–72. doi: 10.1099/00207713-47-1-67. [DOI] [PubMed] [Google Scholar]
  39. Dong L., Yin J., Zhao J., Ma S. R., Wang H. R., Wang M., Chen W., Wei W. Q.. Microbial Similarity and Preference for Specific Sites in Healthy Oral Cavity and Esophagus. Front Microbiol. 2018;9:1603. doi: 10.3389/fmicb.2018.01603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Silva P. H. F., Oliveira L. F. F., Cardoso R. S., Ricoldi M. S. T., Figueiredo L. C., Salvador S. L., Palioto D. B., Furlaneto F. A. C., Messora M. R.. The impact of predatory bacteria on experimental periodontitis. J. Periodontol. 2019;90(9):1053–1063. doi: 10.1002/JPER.18-0485. [DOI] [PubMed] [Google Scholar]
  41. Wootton D. G., Cox M. J., Gloor G. B., Litt D., Hoschler K., German E., Court J., Eneje O., Keogan L., Macfarlane L.. et al. A Haemophilus sp. dominates the microbiota of sputum from UK adults with non-severe community acquired pneumonia and chronic lung disease. Sci. Rep. 2019;9(1):2388. doi: 10.1038/s41598-018-38090-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Yang K., Wang Y., Zhang S., Zhang D., Hu L., Zhao T., Zheng H.. Oral Microbiota Analysis of Tissue Pairs and Saliva Samples From Patients With Oral Squamous Cell Carcinoma - A Pilot Study. Front Microbiol. 2021;12:719601. doi: 10.3389/fmicb.2021.719601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Eriksson K., Lundmark A., Delgado L. F., Hu Y. O. O., Fei G., Lee L., Fei C., Catrina A. I., Jansson L., Andersson A. F.. et al. Salivary Microbiota and Host-Inflammatory Responses in Periodontitis Affected Individuals With and Without Rheumatoid Arthritis. Front Cell Infect Microbiol. 2022;12:841139. doi: 10.3389/fcimb.2022.841139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Schwartz R. H., Rodriguez W. J., Bryan C.. Ampicillin-resistant Haemophilus in the oropharynx: prevalence in three groups of young, middle-class children. Pediatrics. 1983;72(4):464–468. [PubMed] [Google Scholar]
  45. Adar S. D., Huffnagle G. B., Curtis J. L.. The respiratory microbiome: an underappreciated player in the human response to inhaled pollutants? Ann. Epidemiol. 2016;26(5):355–359. doi: 10.1016/j.annepidem.2016.03.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Kumar P. S., Matthews C. R., Joshi V., de Jager M., Aspiras M.. Tobacco smoking affects bacterial acquisition and colonization in oral biofilms. Infect. Immun. 2011;79(11):4730–4738. doi: 10.1128/IAI.05371-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Biedermann L., Zeitz J., Mwinyi J., Sutter-Minder E., Rehman A., Ott S. J., Steurer-Stey C., Frei A., Frei P., Scharl M.. et al. Smoking cessation induces profound changes in the composition of the intestinal microbiota in humans. PLoS One. 2013;8(3):e59260. doi: 10.1371/journal.pone.0059260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Jiang M., Wang X., Gao X., Cardenas A., Baccarelli A. A., Guo X., Huang J., Wu S.. Association of DNA methylation in circulating CD4­(+)­T cells with short-term PM(2.5) pollution waves: A quasi-experimental study of healthy young adults. Ecotoxicol Environ. Saf. 2022;239:113634. doi: 10.1016/j.ecoenv.2022.113634. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

eh5c00143_si_001.pdf (936.4KB, pdf)

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

RESOURCES