Effects of Ozone Exposure on House Dust Mites

IsaacSeow1✉

ZhenYunSiew1

BoonXuan1

Chee-MunFang1

MunSengKan1

ShewFungWong3

SiewTungWong3

MasitaArip4

HusnaFarhanahAhmad4

SuhailiZainalAbidin5Emailhyxis1@nottingham.edu.my

KennyVoon1✉Emailkenny.voon@nottingham.edu.my

Division of Biomedical Sciences, School of PharmacyUniversity of Nottingham Malaysia43500SemenyihMalaysia

2Faculty of Information Science and TechnologyMultimedia University75450Bukit Beruang, Melaka

3School of MedicineIMU University57000Kuala LumpurMalaysia

4Allergy and Immunology Research Centre, Institute for Medical ResearchNational Institutes of HealthPersiaran Setia Murni, Setia Alam40170Shah AlamSelangorMalaysia

5Acarology Unit, Infectious Diseases Research Centre, Institute for Medical ResearchNational Institutes of HealthPersiaran Setia Murni, Setia Alam40170Shah AlamSelangorMalaysia

Isaac Seow^1*, Zhen Yun Siew¹, Boon Xuan U², Chee-Mun Fang¹, Mun Seng Kan¹, Shew Fung Wong³, Siew Tung Wong³, Masita Arip⁴, Husna Farhanah Ahmad⁴, Suhaili Zainal Abidin⁵, Kenny Voon^1*

¹Division of Biomedical Sciences, School of Pharmacy, University of Nottingham Malaysia, 43500, Semenyih, Malaysia

²Faculty of Information Science and Technology, Multimedia University, 75450, Bukit Beruang, Melaka

³School of Medicine, IMU University, 57000, Kuala Lumpur, Malaysia

⁴Allergy and Immunology Research Centre, Institute for Medical Research, National Institutes of Health, Persiaran Setia Murni, Setia Alam, 40170 Shah Alam, Selangor, Malaysia

⁵Acarology Unit, Infectious Diseases Research Centre, Institute for Medical Research, National Institutes of Health, Persiaran Setia Murni, Setia Alam, 40170 Shah Alam, Selangor, Malaysia

*Corresponding author email: hyxis1@nottingham.edu.my & kenny.voon@nottingham.edu.my

Abstract

House dust mites (HDMs), particularly Dermatophagoides farinae, are commonly found in household dust and play a key role in allergic diseases such as asthma and allergic rhinitis. Beyond clinical management, allergen removal strategies are crucial for improving quality of life. Hence, this study investigated the effects of ozone exposure on D. farinae, focusing on changes in protein expression, surface bacterial composition, mortality, and mobility. Mites were exposed to ozone concentrations of 0.05, 0.5, and 1 ppm for 24, 48, and 72 hours in a controlled chamber, with non-exposed mites serving as controls. Western blotting using anti-Der f 1 and anti-Blo t 5 antibodies assessed changes in allergen profiles, while 16S rRNA sequencing characterised changes in surface bacterial communities. Mortality was evaluated using 100 mites per group under varying exposure durations. To assess behavioural responses, a three-chamber mobility assay was conducted, where mites were placed in a central compartment flanked by no-ozone and low-ozone chambers, and their distribution was recorded after 72 hours. Ozone exposure resulted in a concentration- and time-dependent reduction of Der f 1 protein intensity, suggesting allergen degradation. Surface bacterial profiling revealed distinct compositional shifts following ozone exposure. Mortality increased proportionally with ozone concentration and duration. In the mobility assay, mites predominantly remained in the no-ozone chamber, indicating avoidance of ozone. Collectively, these findings demonstrate that ozone exposure affects D. farinae at molecular, microbial, and behavioural levels, highlighting ozone’s potential role in modulating mite allergenicity and ecology.

Keywords:

Allergy

House Dust Mites

Ozone

Dermatophagoides farinae

Next generation sequencing

Microbial diversity

1.0 Introduction

House dust mites (HDMs) such as Dermatophagoides farinae (D. farinae), Dermatophagoides pteronyssinus (D. pteronyssinus), Euroglyphus maynei (E. maynei), and Blomia tropicalis (B. tropicalis) were widely acknowledged as culprits behind allergic diseases such as hay fever, rhinitis, asthma, and conjunctivitis [1]. They were present in almost every home and were mostly found in mattresses, carpets, sofas, and beddings because these places housed the primary food source for HDMs-shed human scales [2]. Previous surveys in Malaysia demonstrated that HDMs were highly prevalent in households, with all mattress dust samples showing evidence of infestation [3]. This showed that HDMs were heavily populated in Malaysian households, and since their existence was greatly linked to a variety of allergic diseases, finding effective ways to inactivate or kill HDMs was crucial for human health.

Numerous control methods have been explored, including chemical acaricides, natural essential oils, frequent vacuuming, hot laundering, and sunlight exposure. While these approaches can reduce mite levels to a certain extent, many of them are labour-intensive, provide only temporary effects, or pose concerns related to cost, safety, or practicality in daily household use [4]. In recent years, ozone has emerged as a potential alternative due to its strong oxidising properties and ability to disrupt biological structures, making it a promising candidate for HDM control [5–7]. However, most ozone-based HDM studies have been conducted under controlled laboratory conditions, and their real-world applicability remains limited, especially considering that indoor ozone levels must comply with the Malaysian Ambient Air Quality Standard (MAAQS) limit of 0.1 ppm [8]. Many studies on HDMs primarily focus on assessing mite mortality following ozone exposure, often overlooking the persistence of allergenic proteins, such as Der f 1 and Der p 1, which remain in the environment even after mite elimination, as their allergens were primarily found in the faeces [9]. Therefore, it is crucial to investigate how ozone exposure affects the expression of these allergenic proteins. Notably, several studies have shown that ozone exposure can regulate protein expression, as demonstrated using Western blot analyses in different biological contexts [10–14].

Besides, another limitation is that the impact of ozone on surface bacteria within the mite micro-environment has largely been overlooked, despite the potential for microbial communities to influence allergen load and mite survival. In addition, very few studies have explored behavioural responses such as ozone avoidance, which could provide a simple and non-chemical strategy to deter mites from human-occupied areas. To address these gaps, we evaluated the effects of ozone on D. farinae under controlled conditions. Western blotting was used to assess changes in major allergenic proteins, while mite mortality and behavioural preference assays evaluated ozone’s lethal and repellent effects. We also examined its impact on surface-associated bacteria to determine potential contributions to allergen reduction. These approaches provide a comprehensive evaluation of ozone as a multi-targeted strategy for HDMs control.

2.0 Materials and Methods

2.1 Mites Culture

D. farinae mites were obtained from colonies established since the 1960s in the Acarology Unit, Institute for Medical Research, Malaysia. The colonies were maintained in T75 culture flasks (Thermo Fisher Scientific, USA) and fed with ground fish flakes (TetraMin Crisps, Germany). All colonies were maintained at an average temperature of (26 ± 2) °C and 70–75% relative humidity in a dark room. Routine maintenance of the mites by random sampling from the culture flasks to detect cross-contamination was performed by microscopic examination of the mites mounted in Hoyer’s medium [15]. A microscopic image of the D. farinae mite is shown in Fig. 1.

Fig. 1

D. farinae female mite, at 40x/0.65 magnification, under a Nikon Eclipse phase contrast microscope.

2.2 Ozone Chambers Construction

Two customised airtight chambers were constructed based on a previous design with major modifications (Abidin & Ming, 2012): one for assessing mite mobility under ozone exposure and another for evaluating mortality and collecting mites for protein analysis. The design of the chambers was drawn using AutoCAD (Autodesk Inc., USA) and is shown in Figs. 2 and 3.

For the mortality and protein-exposure chamber, a customised Medklinn ozone generation system was constructed with patented Cerafusion™ technology, which produces ozone at 0.05, 0.5, and 1.0 ppm, using a 1 g/hr generator with a controlled gas flow of 0.05–0.25 L/min. The mobility assay chamber employed a Medklinn Autoplus model (Medklinn International, Malaysia) with a 6 mg/hr ozone generation capacity to maintain concentrations of up to 1.0 ppm at a flow rate of 1 L/min. In both setups, ozone levels were continuously monitored with a Standard Enclosure Model 106-L Ozone Monitor™ (2B Technologies, USA) operating under the Federal Equivalent Method (FEM), using closed-loop feedback to regulate ozone generation [16].

The mortality and protein-exposure chamber consisted of a single unit with two internal fans to ensure uniform ozone distribution. Relative humidity (75–80%) was maintained using a saturated sodium chloride solution, and temperature was controlled at 26 ± 2°C, both monitored with a digital hygrometer. The mobility assay chamber was constructed with three interconnected acrylic compartments, allowing mites to move freely between sections with different ozone concentrations, while maintaining the same humidity and temperature conditions.

Fig. 2

Design and airflow system of the mortality and protein-exposure chamber. (a) Isometric view of the custom-designed exposure chamber (AutoCAD schematic). (b) Labelled diagram showing internal layout and components of the chamber. (c) Airflow direction is regulated by dual fans to ensure uniform ozone distribution.

Fig. 3

Design of the mobility assessment chamber system. (a) Front view of the mobility assessment chambers illustrating the airflow pathway and directional flow across the compartments. (b) Isometric view and structural specifications of the chamber system used for ozone mobility assessment of D. farinae.

2.3 Mobility Assessments

The experiment was conducted to determine whether ozone exposure affects the mobility of D. farinae at varying concentrations (0.05 ppm, 0.5 ppm, and 1.0 ppm) over 72 hours. Three connected acrylic chambers were used to create a continuous environment that allowed the mites to move freely between sections, as shown in Fig. 4. Before the experiment, all chambers and lids were thoroughly cleaned using tissues and 75% ethanol. The ozone generator and ozone monitor were then switched on and allowed to stabilise for approximately 30 minutes before setting the desired ozone concentration.

A T75 culture flask containing D. farinae was examined under a stereomicroscope to ensure that the mites were active and suitable for testing. Two card papers measuring 11 cm × 11 cm were prepared with double-sided adhesive tape attached along their edges, while another card paper measuring 15 cm × 15 cm was left without adhesive. The taped cards were placed in the low ozone chamber (left) and the no ozone chamber (right), while the untaped card was placed in the middle chamber, which served as the ozone exposure zone. Figure 4 shows the setup of the ozone treatment chamber used for studying dust mites’ mobility.

Fig. 4

Set-up of ozone treatment chamber for mobility test.

A scoop (approximately 300 mg) of active mites was transferred onto a card in the middle ozone chamber, after which all chamber lids were securely closed to maintain airtight conditions. The ozone concentration was adjusted to the target level and maintained for 72 hours. During this period, mites were allowed to move freely across the connected chambers, and those that reached the low-ozone or no-ozone chambers became trapped on the double-sided adhesive tape. After 72 hours of exposure, the card papers from the low ozone and no ozone chambers were carefully removed and examined under a stereomicroscope. The number of mites trapped on each card was counted and recorded to evaluate the distribution of mites between the different ozone exposure chambers. Each ozone concentration test was performed in six replicates.

2.4 Mortality Assessments

The experiment was adapted from a previous study, with minor modifications [5]. The preparation steps were the same as with the mobility assessment.

A Petri dish (10 cm in diameter) was placed into the mortality and protein-exposure chamber (Fig. 2). The Petri dish was prepared by lining the base with filter paper and applying a thin layer of Vaseline along the inner wall and around the edge of the Whatman No. 1 filter paper to prevent mite escape. Approximately 100 active mites were selected and gently transferred from the T75 culture flask into each Petri dish. Each ozone concentration (0.05 ppm, 0.5 ppm, and 1.0 ppm) was tested with three replicates per exposure duration (24, 48, and 72 hours). Control groups were prepared in the same manner, just without turning on the ozone generator. After each experiment, the Petri dishes were removed from the chamber, and the number of live and dead mites was calculated under a stereomicroscope.

2.5 Protein Extraction and Western Blotting

The same ozone chamber setup used for the mortality assessment was employed for mite exposure, with minor modifications. Approximately 1g of active D. farinae from T75 cultures was placed into a Petri dish lined with Vaseline. Mites were exposed to ozone (0.05ppm, 0.5ppm, and 1ppm) for 24, 48, or 72 hours, while a separate set of dishes placed in the chamber without turning on the ozone generator served as unexposed controls. For protein analysis, six biological replicates were prepared per condition; three were pooled to generate one representative sample.

Following exposure, mites were homogenised in PBS containing protease inhibitor, and total protein was extracted for SDS-PAGE and Western blotting using a method adapted from our previous study [15]. BCA assay was performed using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, USA). Equalised protein concentrations were resolved on 12.5% SDS–PAGE gels, transferred to PVDF membranes (Merck Millipore, USA), and probed with anti-Der f 1 antibody (Sino Biological, China) to examine ozone-induced changes in the major D. farinae allergen. The anti-D. pteronyssinus (anti-DP) and anti-T. putrescentiae (anti-TP) antibodies were obtained from our previous study [17]. Additional probing was also performed using anti-D. pteronyssinus antibody, anti-T. putrescentiae antibody, and anti-Blo t 5 antibodyAdditional probing was also performed using anti-D. pteronyssinus antibody, anti-Tyrophagus putrescentiae antibody, and anti-Blo t 5 antibody (InBio, US) to observe broader protein responses across different mite-related targets. The immune complexes formed were detected using ECL chemiluminescence (Nacalai Tesque, Japan), and band intensities were quantified using Image Lab software v6.0 (Bio-Rad Laboratories, USA).

2.6 D. farinae Surface Microbiota and Bioinformatics Analysis

Live D. farinae were exposed for 72 hours to four conditions: control (0 ppm), 0.05 ppm, 0.5 ppm, and 1.0 ppm, using the same ozone chamber as shown in Fig. 2. Three Petri dishes (biological replicates) were used per condition. After exposure, the mites and associated material from each dish were sieved through a cotton-gauze funnel into a 15 mL tube containing 8 mL PBS. The suspension was briefly vortexed and centrifuged at 1200 rpm for 3 minutes to detach surface microbiota. The resulting supernatant was aliquoted into 1.5 mL tubes and centrifuged at 15,000 rpm to pellet the bacterial fraction, which was retained for DNA extraction.

Genomic DNA was extracted using the Monarch® Genomic DNA Purification Kit according to the manufacturer’s protocol. DNA purity and concentration were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific, US) (A260/A280 = 1.8–2.0; >20 ng/µL), and high-quality samples were used for 16S rRNA amplicon sequencing.

Samples that met the quality control standards were subjected to library construction. The sequencing library was prepared by random fragmentation of the DNA samples, followed by ligation of adapters to the 5′ and 3′ ends of the fragments. The adapter-ligated fragments were then amplified using PCR primers targeting the V3–V4 hypervariable regions, which annealed to the ends of each adapter. The resulting library templates were subsequently assessed for quality and quantified prior to sequencing.

The prepared library was loaded onto a flow cell, where the fragments were captured on a lawn of surface-bound oligonucleotides complementary to the library adapters. Each fragment was subsequently amplified into distinct clonal clusters through bridge amplification. Upon completion of cluster generation, the templates were made ready for sequencing. Raw images were generated by the Illumina sequencer using sequencing control software for system operation and base calling, through the integrated primary analysis software known as Real-Time Analysis (RTA). The resulting BCL/cBCL (base call) binary files were then converted into FASTQ files for downstream analysis.

Sequence reads were imported into the CLC Genomics Workbench pipeline (Qiagen, Germany) for bioinformatics analysis. Following quality checking and trimming, operational taxonomic unit (OTU) clustering was performed for each region of all samples independently using the GreenGenes database, with clustering conducted at 97% sequence identity [18]. The flow chart below illustrates the template workflow used for 16S rRNA analysis in the CLC Genomics Workbench (Fig. 5).

Fig. 5

QIAGEN CLC Microbial Genomics Module.

2.7 Statistical Analyses

All statistical analyses were performed in Python (Google Colab) using the pandas, NumPy, SciPy, statsmodels, and matplotlib packages. A Preference Index (PI) was calculated to quantify mite distribution between the two chambers, following a two-choice behavioural metric adapted from Rharrabe et al. (2014) using a two-choice behavioural formula:

$\:\text{P}\text{I}=\frac{({\text{N}}_{{\text{No\:O}}_{3}}-{\text{N}}_{{\text{Low\:O}}_{3}})}{({\text{N}}_{{\text{No\:O}}_{3}}+{\text{N}}_{{\text{Low\:O}}_{3}})}$

where

$\:{\text{N}}_{{\text{No\:O}}_{3}}$

and

$\:{\text{N}}_{{\text{Low\:O}}_{3}}$

represent the number of mites in the no-ozone and low-ozone chambers, respectively [19]. The index ranges from − 1 to + 1, with positive values indicating preference for the no-ozone chamber and negative values indicating avoidance.

Differences in mite distribution between the paired chambers were analysed using paired t-tests. Mortality outcomes were compared using independent t-tests. Western blot band intensities were evaluated using one-way ANOVA with Tukey’s post-hoc test, while two-way ANOVA was applied to determine the effects of multiple factors. Statistical significance was set at p < 0.05.

3.0. Results

3.1. Mobility and Mortality Assessments

Ozone exposure affected both the mobility and survival of D. farinae. In the mobility assay, mites consistently preferred the ozone-free compartment across all tested concentrations (0.05, 0.5, and 1.0 ppm). The Preference Index remained relatively consistent across ozone concentrations, ranging from 0.14 ± 0.08 to 0.18 ± 0.15 (mean ± SD) and the proportion of mites in the ozone-free chamber was significantly higher than in the ozone-exposed chamber at every concentration tested (n = 6; 0.05 ppm: t = 8.24, p = 0.0004; 0.5 ppm: t = 3.283, p = 0.0219; 1.0 ppm: t = 4.366, p = 0.0073) as shown in Fig. 6a. Total movement declined in a concentration-dependent manner, with the greatest reduction observed at 1.0 ppm. The effect of ozone concentration on mite mobility is shown in Fig. 6b.

Ozone exposure also induced an obvious concentration- and time-dependent increase in mortality. After 24 h, mean mortality reached 11.9 ± 0.5%, 26.4 ± 1.4%, and 53.9 ± 1.8% for 0.05, 0.5, and 1.0 ppm, respectively, compared with 4.1 ± 1.1% of the control (n = 3; all p < 0.001). Mortality further increased to 26.0 ± 1.0%, 52.8 ± 0.8%, and 72.8 ± 1.2% at 48 h, and reached 41.7 ± 1.0%, 74.2 ± 1.0%, and 86.0 ± 0.7% after 72 h (n = 3; two-way ANOVA, concentration effect p < 0.001; time effect p < 0.001). The results were summarised in Fig. 6 (c-e). All ozone-treated groups showed significantly higher mortality than the control at every time point (p < 0.001). These results confirm that increasing ozone concentration and prolonged exposure strongly reduce D. farinae survival.

Fig. 6

Effects of ozone exposure on D. farinae mobility and survival. (a) Preference Index (PI) of D. farinae under varying ozone concentrations (0.05, 0.5, and 1.0 ppm). Bars represent mean PI values, and error bars indicate the standard deviation. Asterisks denote a significant avoidance response compared to zero preference (p < 0.05). The positive PI values observed across all ozone concentrations indicate that D. farinae consistently preferred the ozone-free chamber, demonstrating an avoidance behaviour toward ozone exposure. (b) Mobility response of D. farinae following 72-hour ozone exposure, showing mite distribution between ozone and ozone-free chambers. (c–e) Mortality rates of D. farinae after 24 h, 48 h, and 72 h of exposure to increasing ozone concentrations, demonstrating a time- and concentration-dependent rise in mortality.

3.2 Western Blot Analysis

Western blotting revealed a clear and detectable Der f 1 band in D. farinae extracts when probed with the anti-Der f 1 antibody, with band intensity decreasing progressively as ozone concentration increased. Probing with anti-D. pteronyssinus (anti-DP) antibody also demonstrated ozone-associated alterations in protein signal. In contrast, no bands were detected when membranes were probed with anti-T. putrescentiae (anti-TP) or anti-Blo t 5 antibodies, confirming the absence or undetectable levels of these proteins in D. farinae extracts and the species-specific nature of the observed bands. The densitometry analysis and representative blot images are shown in Fig. 7 (a-e).

One-way ANOVA was performed to compare Der f 1 expression across ozone concentrations at each time point. At 24 hours, ozone exposure produced a significant overall effect on Der f 1 levels (F = 7.506, p = 0.0405), and Tukey’s post-hoc test identified a significant reduction only at 1.0 ppm compared with the control (p = 0.0472). At 48 hours, no significant differences were detected among treatment groups (F = 5.123, p = 0.0742), and all pairwise comparisons were non-significant. By 72 hours, ANOVA again showed a significant effect of ozone concentration (F = 7.473, p = 0.0408), with post-hoc analysis confirming that only the 1.0 ppm group differed significantly from the control (p = 0.0309). Lower ozone concentrations (0.05 and 0.5 ppm) showed no significant changes at any time point.

Two-way ANOVA revealed that ozone concentration had a significant effect on the measured outcomes (p < 0.001), whereas exposure time did not produce a significant main effect (p > 0.05). No interaction between concentration and time was detected (p > 0.05), indicating that the impact of ozone was consistent across all time points. Post-hoc Tukey comparisons confirmed that only the highest concentration (1.0 ppm) differed significantly from the control, while 0.05 and 0.5 ppm groups showed no significant differences. These findings indicate that ozone concentration, rather than exposure duration, is the primary determinant of the observed protein response.

Since two-way ANOVA showed no significant effect of exposure time (24–72 h) or interaction between time and ozone, all time points were combined to calculate mean ± SD for each concentration. Figure 7f shows a Hill curve fit (red line) applied to quantify the dose-response relationship. The IC50 (green dashed line) indicates the ozone concentration causing ~ 50% reduction in protein expression. The maximum inhibition point is highlighted in purple. Figure 7g shows the percentage reduction in protein expression across ozone concentrations (0.05, 0.5, and 1 ppm) compared to the control. Error bars represent standard deviation. A horizontal dashed line indicates the 20% reduction cutoff. Red asterisks mark concentrations that are significantly different from control (Tukey HSD, p < 0.05).

Fig. 7

Ozone-induced changes in Der f 1 expression in D. farinae. (a–c) Fold-change in Der f 1 protein abundance following 24 h, 48 h, and 72 h of ozone exposure (0.05–1.0 ppm), relative to unexposed controls. Densitometric analysis (Bio-Rad Image Lab) revealed a concentration-dependent reduction in Der f 1 levels. The decrease was statistically significant at 24 h and 72 h (24 h: F(3,4) = 7.506, p = 0.0405; 72 h: F(3,4) = 7.473, p = 0.0408), while 48 h showed a downward trend that did not reach significance (F(3,4) = 5.123, p = 0.0742). Tukey’s HSD post-hoc test confirmed that the 1.0 ppm group differed significantly from the control at 24 h and 72 h (p < 0.05). Bars represent mean ± SD, and p < 0.05 was considered statistically significant. (d) Representative Western blot of Der f 1 (~ 25 kDa) at 24, 48, and 72 h under increasing ozone concentrations. (e) Immunoblot of ozone-exposed mites (24–72 h) probed with D. pteronyssinus hyperimmune sera. (f) Hill-curve model fitting Der f 1 expression to ozone concentration, illustrating the concentration–response relationship. (g) Percent Der f 1 expression across ozone concentrations, demonstrating the progressive reduction in allergen abundance.

3.3 Ozone Effects on D. farinae Surface Microbiota

16S rRNA gene sequencing revealed clear compositional shifts in the bacterial community associated with D. farinae following 72h of ozone exposure. In control samples, the most abundant genera were Brevibacillus, Paenibacillus, Sphingomonas, Achromobacter, and an unclassified Alcaligenaceae genus (Fig. 8a). Their relative abundance declined progressively at 0.05, 0.5, and 1 ppm ozone, with the strongest reduction observed at 1 ppm. Low-abundance genera such as Staphylococcus, Oceanobacillus, and Candidatus Cardinium were detected mainly in ozone-treated samples. At the class level, the control group’s microbiota consisted predominantly of Bacilli and Alphaproteobacteria, with minor contributions from Actinobacteria and Clostridia (Fig. 8b). After ozone exposure, the relative proportion of Bacilli decreased with increasing ozone concentration, whereas Alphaproteobacteria and Betaproteobacteria became more represented. These community-level changes were consistent across all ozone-treated conditions and distinct from the control profile, indicating a clear ozone-associated alteration in the surface microbiota composition of D. farinae mites.

Fig. 8

Ozone-induced shifts in the surface microbiota of D. farinae. (a) Hierarchical clustering heatmap of bacterial genera associated with D. farinae after 72 h of ozone exposure, illustrating distinct clustering patterns and compositional shifts in response to increasing ozone concentrations. (b) Relative class-level composition of the surface microbiota of D. farinae after 72 h of ozone exposure, showing changes in bacterial community structure following ozone treatment.

4.0 Discussion

Ozone exposure exerts multiple effects on D.farinae, influencing their behaviour, survival, allergen production, and the composition of surface-associated microbiota. Our findings provide compelling evidence that ozone can induce both sublethal and lethal biological consequences in HDMs and may, under controlled conditions, serve as a potential strategy to reduce mite populations and their associated allergenic burden. The ozone concentrations selected in this study, 0.05, 0.5, and 1.0 ppm, were chosen to represent a realistic-to-elevated exposure range. The lowest concentration (0.05 ppm) corresponds to the maximum level permitted in indoor environments by regulatory bodies such as the US Environmental Protection Agency, and reflects the typical output of consumer ionizers and ozone-generating purifiers [20]. Higher ozone concentrations (0.5 and 1.0 ppm) were included because values within this range are commonly used in controlled laboratory investigations examining ozone’s biological effects on mites and other small arthropods, providing a benchmark for experimental comparisons across previous oxidative-stress and fumigation studies[21–23]. The exposure durations of 24, 48, and 72 hours were selected to capture acute and early sub-acute biological responses, aligning with established definitions of short-term ozone exposure and reflecting time frames previously used to demonstrate cumulative oxidative effects on arthropod physiology [5, 20, 24].

Behavioural assays revealed a clear and consistent avoidance response across all tested ozone concentrations. Mites preferentially migrated toward the ozone-free compartment, reflected by a stable Preference Index, indicating that D. farinae can detect and actively avoid oxidative environments. However, mobility decreased at higher ozone levels, suggesting that early sublethal effects, such as oxidative stress, neuromuscular impairment, or reduced energy metabolism, may limit sustained movement, as similarly reported in other oxidant-exposed arthropods [25–27]. Mortality patterns further supported a concentration-dependent effect, with extended exposure to 1.0 ppm leading to over 85% mite death by 72 hours. Ozone-induced oxidative damage to lipids, proteins, and cuticular structures likely contributed to the progressive lethality [23, 24]. Compared with previous ioniser-based studies, the present experiment showed a faster and more pronounced lethal effect, most likely due to the controlled chamber design, real-time ozone monitoring, and homogenous and direct exposure, which minimised fluctuation and ensured stable ozone delivery [5].

In addition to behavioural and survival impacts, ozone exposure significantly suppressed the expression of the major allergen Der f 1 in a concentration-dependent manner. Der f 1 was chosen as the primary target due to its high prevalence in Southeast Asia, being more commonly detected than Der p 1 in Malaysia, Thailand, and Vietnam[28]. As a cysteine protease that is abundant in mite faeces and highly stable in indoor environments, Der f 1 can disrupt epithelial barriers and strongly drives allergic sensitisation, making it the most clinically relevant marker for assessing ozone-mediated allergen reduction [29].

Although one-way ANOVA at individual time points initially suggested that only 1.0 ppm had a significant effect, two-way ANOVA confirmed that ozone concentration, rather than exposure duration, was the primary driver of allergen suppression. Notably, even low-level ozone (0.05 ppm), corresponding to indoor safety recommendations, showed reduced Der f 1 expression when data were pooled across time points. This outcome is biologically relevant because Der f 1 is a cysteine protease with a redox-sensitive active site [30, 31]. Oxidation of its thiol groups may disrupt tertiary structure, inhibit proteolytic function, and interfere with its ability to cleave epithelial junction proteins such as ZO-1 and occludin, which are the key steps in allergen sensitisation [32]. Thus, ozone-mediated suppression of Der f 1 may theoretically reduce allergenic potential at the source, aligning with its concentration-dependent acaricidal effects. Additional protein band changes observed with anti–D. pteronyssinus sera suggest broader ozone-induced proteomic alterations, likely due to the close phylogenetic relationship between Dermatophagoides species [33].

Furthermore, 16S rRNA sequencing indicated that ozone exposure reshaped the surface microbiota of D. farinae, even at 0.05 ppm. Mite-associated bacteria can contribute lipopolysaccharides and other microbe-associated molecular patterns (MAMPs) that potentiate Th2-type immune responses [34–36]. As such, ozone-driven microbiome shifts may modify allergen–microbe interactions and influence downstream host inflammation [37, 38]. This finding is significant in the context of household air-purification devices, which can generate low-level ozone. Depending on which taxa are suppressed or enriched, ozone exposure may reduce, maintain, or potentiate allergenicity.

Several limitations must be acknowledged. Biological replicates for Western blotting and microbiome analysis were pooled, preventing statistical assessment of inter-sample variation. Laboratory-maintained mite colonies may not fully represent the heterogeneous microbiota or stress responses of environmental mite populations. In addition, the study quantified protein expression but not protease activity; thus, the inhibitory effects observed likely underestimate the full functional impact of ozone, as inactivation can occur without reduced protein abundance.

Despite these limitations, the present study provides integrated evidence that ozone can impair mite mobility, induce mortality, suppress a major allergen, and alter surface microbial communities. These combined effects highlight ozone’s potential as a non-chemical environmental intervention in reducing mite-related allergen burden, particularly in enclosed indoor settings. However, because ozone is itself a respiratory irritant, any practical application must be carefully balanced against human safety guidelines [39]. Future research should employ mite activity assays, proteomics, metagenomics, and immunological models to determine whether ozone-driven biochemical and microbiome changes translate into reduced allergenicity in vivo [4].

5.0 Conclusion

In conclusion, our study demonstrated that ozone exposure exerts many effects on D. farinae, particularly influencing their behaviour, survival, allergen production, and surface-associated microbiota. Mite mobility was consistently reduced in ozone-treated conditions, with mites strongly avoiding ozone-containing chambers, while mortality increased in a concentration- and time-dependent manner, indicating that ozone impairs both activity and survival. At the molecular level, ozone exposure caused a clear, concentration-dependent suppression of the major allergen Der f 1, with even low-level exposure (0.05 ppm) producing measurable reductions, suggesting potential mitigation of allergenic potential. Furthermore, 16S rRNA gene analysis revealed that ozone modulated the surface microbiota of the mites, altering bacterial community composition even at indoor-safe ozone levels. Taken together, these findings provide comprehensive evidence that ozone impacts D. farinae across behavioural, molecular, and microbiological dimensions, highlighting its potential as a controlled, non-chemical strategy in reducing HDM allergen burden and associated health risks in indoor environments.

Data Availability

The raw sequencing data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1354801(https://www.ncbi.nlm.nih.gov/bioproject/1354801). The corresponding BioSample accession numbers are SAMN52955989, SAMN52955990, SAMN52955991, and SAMN52955992. All other data supporting the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgement

This work is supported by an industrial grant under code NVLJ0002 from Medklinn International Sdn. Bhd. The authors would like to thank Ms. Ngai Zi Ni for her guidance in protein work, and also Ms. Hoo Yong Qi and Mr. Daniel Lu for their technical advice in designing the ozone chambers.

Competing interests

All authors declare no competing interests.

Author Contribution

I.S. designed and performed the experiments, analysed the data, and wrote the manuscript. B.X.U. performed statistical analyses. S.Z.A. provided expertise in dust mite taxonomy and propagation. Z.Y.S. contributed to data interpretation and manuscript editing. K.V. conceptualized and supervised the study. C.M.F., M.S.K., S.F.W., M.A., H. F. A., and S.T.W. contributed to project supervision and provided critical review of the manuscript. All authors discussed the results, contributed to the manuscript, and approved the final version.

Correspondence and requests for materials should be addressed to I.S. and K.V.

References

Tang, J. C., Wong, S. F., Mak, J. W. & Ho, T. M. Antigenic profile of Blomia tropicalis, Aleuroglyphus ovatus and Glycycometus malaysiensis. Trop. Biomed. 28, 223–236 (2011).

Colloff, M. J. Springer Netherlands, Dordrecht,. Methods in house dust mite ecology and biology. in Dust Mites (ed. Colloff, M. J.) 255–271 (2009). 10.1007/978-90-481-2224-0_6

Mariana, A., Ho, T. M., Sofian-Azirun, M. & Wong, A. L. House dust mite fauna in the Klang Valley, Malaysia. Southeast. Asian J. Trop. Med. Public. Health. 31, 712–721 (2000).

Seow, I. et al. House dust mites eradication treatments: Current updates emphasizing on tropical countries. Trop. Biomed. 41, 450–460 (2024).

Abidin, S. Z. & Ming, H. T. Effect of a commercial air ionizer on dust mites Dermatophagoides pteronyssinus and Dermatophagoides farinae (Acari: Pyroglyphidae) in the laboratory. Asian Pac. J. Trop. Biomed. 2, 156–158 (2012).

Han, J. H. et al. Killing Effect of Ozone on House Dust Mites, the Major Indoor Allergen of Allergic Disease. Ozone: Sci. Eng. 28, 191–196 (2006).

Pumnuan*, J., Insung, A. & Wangapai, T. The Use of Ozone for Controlling European House Dust Mite, Dermatophagoides pteronyssinus (Trouessart). Curr Appl. Sci. Technol 420–428 (2020).

Baidrulhisham, S. E. et al. Effects of Weather and Anthropogenic Precursors on Ground-Level Ozone Concentrations in Malaysian Cities. Atmosphere 13, 1780 (2022).

Tovey, E. R. & Chapman, M. D. Platts-Mills, T. A. Mite faeces are a major source of house dust allergens. Nature 289, 592–593 (1981).

10.

Abd-Elrahim Batran, S. Toxicity of environmental ozone exposure on mice olfactory bulbs, using Western blot technique. Toxicol. Rep. 7, 453–459 (2020).

11.

Hussain, S. et al. Intermittent ozone inhalation during house dust mite-induced sensitization primes for adverse asthma phenotype. Redox Biol. 76, 103330 (2024).

12.

Michaudel, C. et al. Ozone exposure induces respiratory barrier biphasic injury and inflammation controlled by IL-33. J. Allergy Clin. Immunol. 142, 942–958 (2018).

13.

Rodríguez-Martínez, E., Nava-Ruiz, C., Escamilla-Chimal, E., Borgonio-Perez, G. & Rivas-Arancibia, S. The Effect of Chronic Ozone Exposure on the Activation of Endoplasmic Reticulum Stress and Apoptosis in Rat Hippocampus. Front Aging Neurosci 8, (2016).

14.

Vo, T., Paudel, K., Choudhary, I., Patial, S. & Saini, Y. Ozone exposure upregulates the expression of host susceptibility protein TMPRSS2 to SARS-CoV-2. Sci. Rep. 12, 1357 (2022).

15.

Wong, S. F. et al. Molecular identification of house dust mites and storage mites. Exp. Appl. Acarol. 55, 123–133 (2011).

16.

Snyder, E. G. et al. The Changing Paradigm of Air Pollution Monitoring. Environ. Sci. Technol. 47, 11369–11377 (2013).

17.

Tung, W. S. Production of recombinant group 10 allergens of Dermatophagoides pteronyssinus and Tyrophagus putrescentiae mites and monoclonal antibodies against them for the study of mite allergy. PhD thesis, International Medical University (2013).

18.

DeSantis, T. Z. et al. Greengenes, a Chimera-Checked 16S rRNA Gene Database and Workbench Compatible with ARB. Appl. Environ. Microbiol. 72, 5069–5072 (2006).

19.

Rharrabe, K., Jacquin-Joly, E. & Marion-Poll, F. Electrophysiological and behavioral responses of Spodoptera littoralis caterpillars to attractive and repellent plant volatiles. Front Ecol. Evol 2, (2014).

20.

US EPA. O. Ozone Generators that are Sold as Air Cleaners. (2014). https://www.epa.gov/indoor-air-quality-iaq/ozone-generators-are-sold-air-cleaners Accessed 9 Nov 2025.

21.

Botondi, R. et al. The Use of Ozone Technology: An Eco–Friendly Method for the Sanitization of the Dairy Supply Chain. Foods 12, (2023).

22.

Hasan, M. M., Aikins, M. J., Schilling, W. & Phillips, T. W. Efficacy of Controlled Atmosphere Treatments to Manage Arthropod Pests of Dry-Cured Hams. Insects 7, 44 (2016).

23.

Bi, J. et al. Ozone reduces lifespan and alters gene expression profiles in Rhyzopertha dominica (Fabricius). 3 Biotech 12, 345 (2022).

24.

Holmstrup, M. et al. Effects of ozone on gene expression and lipid peroxidation in adults and larvae of the red flour beetle (Tribolium castaneum). J. Stored Prod. Res. 47, 378–384 (2011).

25.

Messi, M., Giorgione, R. & Astolfi, M. L. Insects as Sentinels of Oxidative Stress Induced by Environmental Contaminants: Biomarkers and Analytical Approaches. Toxics 13, 698 (2025).

26.

Telesnicki, M., Martinez, M., Arneodo, J. & Ghersa, C. Direct effect of ozone pollution on aphids: Revisiting the evidence at individual and population scales. Entomologia Exp. et Applicata 155, (2015).

27.

Vanderplanck, M., Lapeyre, B., Lucas, S. & Proffit, M. Ozone Induces Distress Behaviors in Fig Wasps with a Reduced Chance of Recovery. Insects 12, 995 (2021).

28.

Kim, D. et al. Characteristics of house dust mite allergens in Southeast and East Asia with the effect of hygienic practices. Indoor Environ. 1, 100010 (2024).

29.

Ouyang, X., Reihill, J. A., Douglas, L. E. J. & Martin, S. L. Airborne indoor allergen serine proteases and their contribution to sensitisation and activation of innate immunity in allergic airway disease. Eur. Respir Rev. 33, 230126 (2024).

30.

Chruszcz, M. et al. Crystal structures of mite allergens Der f 1 and Der p 1 reveal differences in surface exposed residues that may influence antibody binding. J. Mol. Biol. 386, 520–530 (2009).

31.

Gough, L., Schulz, O., Sewell, H. F. & Shakib, F. The Cysteine Protease Activity of the Major Dust Mite Allergen Der P 1 Selectively Enhances the Immunoglobulin E Antibody Response. J. Exp. Med. 190, 1897–1902 (1999).

32.

Cataldo, F. On the action of ozone on proteins. Polym. Degrad. Stab. 82, 105–114 (2003).

33.

Miller, J. D. The Role of Dust Mites in Allergy. Clin. Rev. Allergy Immunol. 57, 312–329 (2019).

34.

de Daan, J. et al. Lipopolysaccharide inhibits Th2 lung inflammation induced by house dust mite allergens in mice. Am. J. Respir Cell. Mol. Biol. 48, 382–389 (2013).

35.

Huang, Y. J., Porsche, C., Kozik, A. J. & Lynch, S. V. Microbiome-Immune Interactions in Allergy and Asthma. J. Allergy Clin. Immunol. Pract. 10, 2244–2251 (2022).

36.

Lin, T. H., Su, H. H., Kang, H. Y. & Chang, T. H. The Interactive Roles of Lipopolysaccharides and dsRNA/Viruses on Respiratory Epithelial Cells and Dendritic Cells in Allergic Respiratory Disorders: The Hygiene Hypothesis. Int. J. Mol. Sci. 18, 2219 (2017).

37.

Cho, Y. et al. The Microbiome Regulates Pulmonary Responses to Ozone in Mice. Am. J. Respir Cell. Mol. Biol. 59, 346–354 (2018).

38.

Tashiro, H. & Shore, S. A. The Gut Microbiome and Ozone-induced Airway Hyperresponsiveness. Mechanisms and Therapeutic Prospects. Am. J. Respir Cell. Mol. Biol. 64, 283–291 (2021).

39.

Lee, Y. Z., Kow, A. S. F., Jacquet, A., Lee, M. T. & Tham, C. L. House dust mite allergy in Malaysia: review of research gaps in the current scenario and the way forward. Exp. Appl. Acarol. 91, 509–539 (2023).

Yes