Ozone

Endpoint summary

Key value for chemical safety assessment

Genetic toxicity in vitro

Description of key information

Several studies are available of which non was conducted in accordance with current OECD guidelines for genotoxicity testing. Ozone was tested negative in the bacterial reverse mutation assay (Dillon et al., 1992; Victorin, 1988), except for Salmonella typhimurium strain TA102 (Dillon et al., 1992). The results obtained from strain TA102 showed a huge variance between the six conducted experiments. Moreover, as no clear dose-response relationship was observed, the effect seen in tester strain TA102 is considered to be ambiguous. In further in vitro studies positive as well as negative results were obtained after exposure to ozone. Some of these studies showing methodological deficiencies and the results obtained must be interpreted with caution. In the study by Diaz-Llera, 2002 it is shown that the induction of DNA damage after ozone exposure is induced by the formation of hydrogen peroxide. Moreover, the study showed that post-treatment incubation of up to 90 min reduced the percentage of damaged cells to control levels, indicating that the cells recover rapidly from the induced DNA damage. These results show, that ozone as such is not considered genotoxic but may induce oxidative stress in some in vitro tests systems.

Link to relevant study records

Reference

Reference 1

Endpoint:

in vitro gene mutation study in bacteria

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 471 (Bacterial Reverse Mutation Assay)

Deviations:

yes

Remarks:

see box " Principles of method if other than guideline"

Principles of method if other than guideline:

A standard Ames test was modified for gasvapour exposure. The petri dishes were exposed to an atmosphere containing ozone for 35 minutes.

GLP compliance:

no

Type of assay:

bacterial reverse mutation assay

Specific details on test material used for the study:

Ozone was produced using an Ozone Generator, Type GLX, machine (Argentox, Hamburg, Germany) operating via an electrical discharge in dry oxygen. Different concentrations of ozone were achieved by regulating the flow rate of oxygen used as support gas and by varying the voltage. Ozone output was determined by titrating the iodine liberated from a solution of KI buffered with Na2HPO4 and KH2PO4 using standardised sodium thiosulphate.

Target gene:

Histidine locus

Species / strain / cell type:

S. typhimurium, other: TA 1535, TA 98, TA 100, TA102, TA104

Details on mammalian cell type (if applicable):

CELLS USED
- Source of cells: Prof. Bruce N. Ames (University of California, Berkeley)

MEDIA USED
- Nutrient medium: Oxoid No.2
- Top agar used to overlay Vogel-Bonner plates

Metabolic activation:

with and without

Metabolic activation system:

S9 mix

Test concentrations with justification for top dose:

6 concentrations per experiment ranging between 0.019 and 9 ppm.

Vehicle / solvent:

oxygen

Untreated negative controls:

yes

Remarks:

air

Negative solvent / vehicle controls:

yes

Remarks:

oxygen gas

True negative controls:

no

Positive controls:

yes

Positive control substance:

2-nitrofluorene

sodium azide

methylmethanesulfonate

mitomycin C

other: 2-Aminoanthracene (all strains, 1µg/plate, +S9) and formaldehyde (TA104, 50 µg/plate, -S9)

Details on test system and experimental conditions:

The standard plate incorporation protocol by Maron and Ames, 1983 were followed. Overnight bacterial culture (0.1 mL) and S9 mix or 0.05 M phosphate buffer, pH 7.4 (0.5 mL) were added to top agar (2.0 mL), mixed, and used to overlay Vogel-Bonner plates (3 plates per concentration). Cells were exposed on the plates stacked in glass jars of known volume with tapped, ground glass lids. Petri plate lids were slightly raised by the insertion of foil clips to facilitate circulation of the gas. All treatments comprised application of generator voltage for 5 min, after which time the jars were sealed to maintain ozone atmospheres for an additional 30 min. At the conclusion of this period, the residual ozone was purged with air. Plates were incubated at 37 °C for 2 days in the jars and then for an additional day outside the jars. Colonies were counted with a Biotran III colony counter. Qualitative indications of toxicity were obtained from the extent of background growth. Concurrent negative controls were performed by exposing cells, as for ozone, to the oxygen support gas with the generator switched off, or to air, with the gas flow set to zero. In total 6 experiments were conducted. Experiment 1-4 were conducted with an flow rate of 5 L/min and experiments 5 and 6 were conducted at flow rates of 7 L/min. Dosimetry was undertaken concurrently with each experiment.

Evaluation criteria:

Values significantly different (p< 0.01) from respective air controls (Dunnett's t-test) were regarded a positive response.

Statistics:

The parametric method of Dunnett (1955), involving calculation of the Student's t-statistic and a nonparametric, ranking procedure [Wahrendorf et al., 1985] were applied.

Species / strain:

S. typhimurium, other: TA1535, TA98 and TA100 and TA 104

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Species / strain:

S. typhimurium, other: TA102

Metabolic activation:

with and without

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

cytotoxicity

Remarks:

at the two top doses (7.04 and 3.62 ppm)

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

Ozone was consistently non-mutagenic to strains TA1535, TA98, and TA100 and induced a slight but non reproducible and generally statistically non-significant increase in revertants of TA104 at a single concentration (not the high dose) and only in one experiment (maximum fold increase of 1.2 in comparison to control). Reproducible and statistically significant increases (p< 0.01) in revertant counts with TA102 were, induced at 0.022, 0.036, and 0.59 ppm ozone (Fig 2 below). In most experiments, mutagenicity was unaffected by the presence or absence of S9. Maximum fold increases of 3.1 over air controls were obtained with 0.19 ppm ozone, and small increases in revertants were recorded in separate experiments for the mean dose of 0.022 ppm, and over the range 0.019 to 0.22 ppm ozone. With higher concentrations, however, almost linear decreases in revertants occurred, probably due to toxicity, which was manifest as a reduction in the number of his+ revertants per plate and/or as a thinning of the background lawn at the highest dose levels. However, dose-related responses were not always obtained.

Remarks on result:

other: no dose response occurred

Conclusions:

In conclusion, the test item is considered overall as not genotoxic in the bacterial reverse mutation assay in the presence and absence of mammalian metabolic activation.

Executive summary:

In a modified reverse gene mutation assay in bacteria (similar to OECD 471) strains of S. typhimurium (TA1535, TA104, TA102, TA100 and TA98) were exposed to Ozone at six concentrations/experiment ranging from of 0.019 to 9 ppm in the presence and absence of mammalian metabolic activation for a total exposure time of 35 minutes. Positive controls induced the appropriate responses in the corresponding strains. There was no evidence of induced mutant colonies over background, except for strain 102. This strain is known for its sensitivity against oxygen radicals. Ozone, at two to three consecutive doses, induced weak, albeit statistically significant mutagenic responses with and without S9. But, these effects were not dose related and occurred only at the lower concentrations. Higher concentrations, which were not identified as toxic, showed revertant levels similar to the concurrent control. Moreover, a certain variance occured between the different experiments. Based on the results, ozone is not genotoxic in tester strains TA98, TA100, TA104 and TA1535 with and without metabolic activation. Ambigious results were obtained with tester strain TA102.

Reference 2

Endpoint:

in vitro DNA damage and/or repair study

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

test procedure in accordance with national standard methods with acceptable restrictions

Qualifier:

no guideline followed

Principles of method if other than guideline:

Primary peripheral blood leukocytes were exposed to ozone or hydrogen peroxide in vitro for 1 hour. Subsequently cell and DNA damage was evaluated by means of single cell gel electrophoresis. In addition, the effects of catalase treatment were studied.

GLP compliance:

no

Type of assay:

comet assay

Specific details on test material used for the study:

Ozone was produced in an ozone generator (Ozomed, Cuba).

Target gene:

n/a

Species / strain / cell type:

other: human peripheral blood leukocytes

Additional strain / cell type characteristics:

other: primary cells derived from non-smoking volunteers; 22 - 48 years of age

Metabolic activation:

not applicable

Test concentrations with justification for top dose:

range from 0.875 to 5.25 mM

Vehicle / solvent:

The blood was diluted in PBS

Untreated negative controls:

yes

Remarks:

whole blood

Negative solvent / vehicle controls:

yes

Remarks:

blood diluted in PBS

True negative controls:

no

Positive controls:

yes

Positive control substance:

other: hydrogen peroxide

Remarks:

Concentration PC: 4 mM and 40 mM

Details on test system and experimental conditions:

Cells and treatments:
Peripheral blood was obtained from six healthy, non-smoking volunteers, aged 22–48 years. The blood was diluted in PBS (60 µL blood in 1 mL PBS) and drawn into a 10 mL syringe. Ozone was produced in an ozone generator and the ozone-oxygen gas mixture was introduced into the syringe, which was incubated while slowly rocking for 1 h at 37 °C. In parallel experiments, hydrogen peroxide was added to the diluted blood at concentrations of 4 and 40 mM. Cell viability was determined with the Trypan blue exclusion method after treatment. To assess post-treatment recovery, cells from four donors treated with the highest dose of 03 (5.25 mM) were centrifuged, re-suspended in fresh PBS and incubated for 45 and 90 min after treatment. The effect of catalase was studied by pre-incubating the diluted blood from the two other donors with catalase (20 µg/mL) for 15 min before 03 and H2O2 treatment as above.

Comet assay protocol:
After treatment the cell suspensions were centrifuged. Re-suspended pellet (5 µl) was mixed with 75 µl of 0.5% low melting agarose (LMA, Fluka, USA) and added to conventional microscope slides pre-coated with 150 µl of 0.8% normal melting agarose. The slides were covered with coverslips and kept at 4 °C for 5 min to solidify the LMA. The coverslips were removed, a top layer of 75 µl LMA was added and the slides were kept again on ice for 2 min. After removal of the coverslips, the slides were immersed for 1 h at 4 °C in lysis solution; 2.5M NaCl (Fluka), 100mM EDTA (Sigma), 10mM Tris (Fluka), 1% Triton X-100 (BDH, UK), 10% DMSO (Sigma), pH 10.0. The slides were then placed in an electrophoresis chamber containing 0.3M NaOH (Sigma) and 1mM EDTA at 4 °C for 20 min before electrophoresis (250 mA, 25V, 30 min) in the same solution. The slides were washed 3× for 5 min in neutralising buffer (0.4M Tris–HCl, pH 7.5) and let dry at 37 °C for 20 min before staining with ethidium bromide (Sigma, 20 µg/ml). The slides were observed under a fluorescence microscope with a calibrated ocular micrometer. Images of 50 randomly selected cells (25 cells from each two replicate slides) were analysed from each treatment. From each cell, the length of the image (diameter of the nucleus plus migrated DNA) was measured in microns at a 400× magnification.

Statistics:

Differences between treatment means were tested for significance using Kruskall–Wallis and Dunn tests.

Species / strain:

other: primary human peripheral blood leukocytes

Metabolic activation:

without

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

The compiled data from the six donors showed a dose dependent effect for both H2O2 and O3. Pre-incubation of cells from two donors with catalase for 15 min significantly decreased the percentages of damaged cells and comet length, even in the controls, compared with the corressponding treatments without catalase. Moreover, cells from four donors treated with the highest dose of O3 (5.25 mM) were allowed to recover in the incubator for 45 or 90 min. In all cases, this post-treatment incubation reduced the percentages of damaged cell and the comet length to control levels, indicating that the cells recover rapidly from the genotoxic effect of O3 treatment.

Cell viability measurement:

Cell viability was determined with the Trypan blue exclusion method after treatment. The viability of the cells was 80-98% after treatment to ozone and hydrogen peroxide.

Conclusions:

The compiled data from the six donors showed a dose dependent effect for both H2O2 and O3 in human peripheral blood leukocytes, using the single cell gel electrophoresis assay. Pre-incubation of cells from two donors with catalase for 15 min significantly decreased the percentages of damaged cells and comet length, even in the controls, compared with the corresponding treatments without catalase. Moreover, cells from four donors treated with the highest dose of O3 (5.25 mM) were allowed to recover in the incubator for 45 or 90 min. In all cases, this post-treatment incubation reduced the percentages of damaged cell and the comet length to control levels, indicating that the cells recover rapidly from the genotoxic effect of O3 treatment.

Executive summary:

The genotoxic effect of ozone was studied in cultured primary human peripheral blood leukocytes in vitro, using the single cell gel electrophoresis (SCGE) assay. Cell treatment for 1 h at 37 °C with 0.9–5.3 mM ozone resulted in a dose-dependent increase of DNA damage, comparable to that induced by 4–40 mM of the positive control H2O2. This effect of ozone was reversed by post-treatment incubation of the cells for 45–90 min at 37 °C and prevented by pre-incubation of the cells with catalase (20 µg/mL). The inhibition of oxidative DNA damage by catalase implies a major role for hydrogen peroxide in generating genotoxic effects after ozone exposure. Moreover, the results showed, that in addition to the secondary genotoxic effect of ozone the observed DNA damage is rapidly repaired by cellular repair mechanisms.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Genetic toxicity in vivo

Description of key information

The available data was evaluated in a weight-of-evidence approach as contradictory results were observed in in vitro and in vivo studies.

After inhalation, 90% of O3 reacts in the epithelial lining fluid covering the respiratory airways and epithelial cellular membranes. The remaining 10% react with cell membranes. O3 reacts directly with polyunsaturated fatty acids, amino acids, electron donors such as vitamins or glutathione and proteins. Aldehydes, H2O2 and Criegee ozonides are reaction products and considered mediators of ozone toxicity. By this mechanism O3 is entirely consumed and systemically not available (Pryor et al., 1996). Especially, it does not reach the cell nucleus and/or nucleic acid. Several studies are available, of which none was conducted in accordance with current OECD guidelines for genotoxicity testing.

Bornholdt et al., 2002 showed that an in vivo short-term exposure to ozone of female mice for 90 minutes at 1 or 2 ppm induces lung inflammatory mediators and that DNA damage in cells of the lumen of the lung was observed within the first 200 minutes after exposure. After 200 minutes post-treatment no DNA damaging effects were observed. Moreover, this effect was not reflected by an induction of mutations in the lung of MutaTM Mice, indicating that oxidative DNA damage is not subsequently leading to persistent mutations.

Further supportive data was published by Ferng et al., 1997 who showed, that an in vivo short-term exposure to 1 ppm ozone in guinea pigs for 72 hours could result in an increase of DNA strand breaks in tracheobronchial epithelial cells. This increase occurred together with an increase of total protein indicating an ozone-mediated inflammatory reaction. 96 hours after exposure the DNA damage decreased.

Lee et al., 1997 presented data in which guinea pigs were exposed to ozone for 2 hours to 0.4 and 1.0 ppm ozone. Within one hour after exposure bronchoalveolar cells were obtained from animals and analysed for DNA damage in the Comet assay. An increase of DNA damage was observed. But, in parallel an increase of cytotoxicity was observed. Similar results were obtained from the study published by Rithidech et al., 1990.

In the study by Haney & Connor, 1999 the results of a comet assay in mice indicated, that acute ozone exposure can lead to DNA damage, which is in line with the reactivity of ozone with a variety of biological target structures such as olefin structures or electron donors. The chemical reaction between O3 and the olefins structures follows the mechanism of ozonolysis and leads to the production of hydrogen peroxide and aldehydes. Thus, it can be assumed, that the effect of ozone are secondary via oxidized reaction products.

Several publications are available, which showed that in vivo exposure to ozone in hamster or mouse did not induce chromosomal damage in peripheral blood or bone marrow cells (Tice et al., 1978; Gooch et al., 1976).

In studies on the genotoxic effects of ozone on human test person in vivo (McKenzie et al., 1977; Guerrero et al., 1979; Finkenwirth et al., 2013) no genotoxic effects of ozone were observed.

Thus, it can be concluded that adverse effects of ozone in certain studies were not driven by ozone itself. Instead, effects can be attributed to products of the reactions of ozone with cell components or by induced inflammatory processes. In addition, acute exposure to ozone and related DNA damage is not persistent but handled through activation of repair mechanisms or increase of antioxidant levels. Due to the high reactivity of ozone and consequential lack of systemic bioavailability, no classification for mutagenicity is warranted.

Link to relevant study records

Reference

Reference 1

Endpoint:

in vivo mammalian cell study: DNA damage and/or repair

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Acceptable, well-documented publication which meets basic scientific principles

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 489 (In vivo Mammalian Alkaline Comet Assay)

Principles of method if other than guideline:

The genotoxic and inflammatory effects of ozone were investigated in female mice exposed to 1 and 2 ppm ozone once for 90 min. The tail moment in bronchoalveolar lavage (BAL) cells from mice was determined by the comet assay as a measure of DNA strand breaks at different intervals after termination of the exposure. To determine whether the exposures were mutagenic, MutaTMMice were exposed to 2 ppm ozone, 90 min per day for 5 days and examined for induction of mutations.

GLP compliance:

not specified

Type of assay:

mammalian comet assay

Specific details on test material used for the study:

Ozone was generated photochemically by a thermostated mercury lamp in a flow of 0.15 L 99.999% oxygen.

Species:

mouse

Strain:

other: female BALB/c mice and Muta TM mice

Sex:

female

Details on test animals or test system and environmental conditions:

The mice were allowed to acclimatise for minimally 7 days, were housed in polypropylene cages with sawdust bedding, were given a standard diet and water ad libitum. They were kept at controlled temperature (20 ± 2 °C), humidity (50 ± 10%) and a 12 h light/dark cycle. The mean weight at the time of exposure for the BALB/c mice was 20.6 g with S.D. of 1.6 g. The mean weight for the Muta TM mice was 26.0 g with S.D. of 3.4 g.

Route of administration:

inhalation: gas

Vehicle:

air

Details on exposure:

Both the BALB/c mice and the Muta TM mice were exposed to ozone for 90 min in an 18 L glass/stainless steel chamber. Ozone was generated photochemically by a thermostated mercury lamp in a flow of 0.15 L 99.999% oxygen. The ozone was led through 2 m of Teflon tube and mixed into the flow of 24.5 L/min ambient air just before the chamber. The ozone concentration in the breathing zone was measured throughout the entire exposure by an API photometric O3 analyser model 400 (API, San Diego, CA). Control animals (both BALB/c and Muta TM Mice) were exposed to ambient air or ambient air plus 0.15 L/min 99.999% oxygen.

Duration of treatment / exposure:

90 min

Frequency of treatment:

BALB/c mice received a single exposure. Muta TM Mice were exposed on five consecutive days.

Post exposure period:

After exposure the BALB/c mice were allowed to recover between 20 and 1400 min before they were sacrificed to study the time course of the effects after the exposure. Muta Tm mice were allowed to recover for 14 days before sacrifice.

Dose / conc.:

1 ppm (nominal)

Remarks:

BALB C mice

Dose / conc.:

2 ppm (nominal)

Remarks:

BALB/c mice and Muta M Mice

No. of animals per sex per dose:

single exposure: 3 to 8 females/ exposure group
repeated exposure: 5 females / exposure group

Control animals:

yes, concurrent vehicle

other: air + 0.6% oxygen

Positive control(s):

not applicable

Tissues and cell types examined:

Bronchoalveolar lavage (BAL) and lung tissue cells, blood

Details of tissue and slide preparation:

- BALB/C mice (single exposure): The animals were anaesthetised and a 0.8 mL blood sample was withdrawn from the heart, for analysis of oxidised protein in the serum. The blood was stabilised in 72 microliter 0.17 M K2EDTA and kept on ice until plasma was isolated by centrifugation at 4 °C and 12000 × g for 10 min. Immediately after withdrawing the heart blood, a bronchoalveolar lavage (BAL) was performed three times with 1 mL of 0.9% sterile saline through the trachea. The BAL was immediately put on ice until recovering the cells at 1250 rpm at 4 °C for 10 min after maximum 90 min. The cells were re-suspended in Merchant’s medium (4 °C) and used directly in the comet assay and for cell identification. The lungs were frozen immediately in cryotubes (NUNC) in liquid nitrogen and stored at −80 °C.

- Muta Mice (repeated exposure): Mice were allowed to recover for 14 days before sacrifice. The mice were anaesthetised and 0.8 mL blood was withdrawn from the heart. The blood was stabilised in 72 microliter 0.17 M K2EDTA and stored at −80 °C. The lungs were isolated, quickly frozen in liquid nitrogen and stored at −80 °C. DNA was isolated from the right lung of the mice using. The vectors were packaged into phages. Phages with mutations in the cII gene were screened by use of the positive selection model Select-cII mutation detection system for Big Blue rodents.

- Cell viability: The total number and viability of the cells were determined by staining with 0.25% trypan blue exclusion.

- Cell identification: Cells were collected on microscope slides by centrifugation at 1000 rpm for 4 min. The slides were fixed with 96% ethanol and stained with May–Grünwald–Giemsa stain. Two hundred cells were evaluated for each preparation.

- Cytokine (interleukin-6, TNF alpha, interleukin-1) and ERCC1 mRNAs: Lung tissue was homogenised with an ultrathorax and RNA and subsequently cDNA was prepared and stored at -80 °C. (RT-)PCR was used to evaluated the levels on cytokines and ERCC1.

- 8-oxo-deoxyguanosine (8-oxo-dG) content in lung tissue from BALB/c mice: 8-oxo-dG relative to dG was measured in lung tissue by HPLC with electrochemical detection.

- Comet assay: To investigate the extent of DNA strand breaks caused by ozone, the tail moment in the comet assay was determined in the lung tissue and in the freshly prepared BAL cells. A frozen lung was immediately placed in a stainless steel cylindrical sieve in 2mL Merchant’s medium. After disruption of the tissue the extract was filtered through a 53 µm nylon mesh. The cell concentration was adjusted to 1 × 10^6 cells/mL with Merchant’s medium. The BAL cells were used without further preparation. One percent (w/v) normal melting point and 1% (w/v) low melting point agarose solutions were freshly prepared with PBS (0.14M NaCl, 1.8mM KH2PO4, 8.1mM Na2HPO4, 2.7mM KCl, pH 7.4). First a layer of agarose was prepared by transfer of 95 µl normal melting point agarose at 65 °C onto a fully frosted slide. A second layer of agarose was prepared by mixing 95 µl low melting point agarose with 10 µl cell suspension (1×10^6 cells/mL) at 37 °C. The mixture was applied onto the first layer of agarose and covered with a cover slip. The embedded cells were lysed in ice-cold lysis buffer (2.5M NaCl, 0.1M Na2EDTA, 10mM Tris, 1% Triton X-100, pH 10) overnight. After lysis the slides were washed briefly in freshly made ice-cold alkaline electrophoresis solution (0.3M NaOH, 1mM Na2EDTA, pH >13). The slides were then placed in a Maxicell EC360M horizontal electrophoresis tank containing ice-cooled alkaline electrophoresis solution for 40 min. Electrophoresis was carried out at 25V and 292–296mA for 20 min. The slides were washed two times for 5 min with cold (4 °C) neutralising buffer (0.4M Tris, pH 7.5). The DNA was stained by applying 25 µl of 0.6 µM TOTO TM-1 iodide to each gel. The samples were analysed on a Leica DM BL fluorescence microscope with 400× magnification, a 450–490 nm excitation filter and a LP520 suppression filter. Using a Kinetics® image analysing system (version 3.0) the DNA damage was measured as tail moment. For each sample, three gels were analysed and 25 cells on each gel were measured

Evaluation criteria:

DNA strand breaks in BAL cells and lung tissue were evaluated by a comet assay. The tail was defined as the area where the intensity was lower than 10% of intensity in the head. Due to day-to-day variation in the comet assay, each tail moment was normalised by dividing the tail moment by the mean of tail moments of the untreated control mice for that day.

Statistics:

To define the time period when an effect on DNA strand breaks and IL-6 mRNA expression could be detected a 95% reference interval was determined for the control group. The interval was given by x¯ ± (sk), where s is the standard derivation for (n − 1) and k is defined as, k = (√(1/n) + 1)t (f )1−(α/2). The effect time window border lines were defined by the first two ubsequent data that were outside the reference interval. To test for dose effect in the time interval defined where the effect occurred, we used a Student’s t-test and ANOVA. The mutation frequencies, 8-oxo-dG/dG results, protein oxidation and cell identification data were all compared by the Student’s t-test or the modified Student’s t-test. Homogeneity was tested with a variance ratio test.

Sex:

female

Genotoxicity:

negative

Toxicity:

no effects

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

not examined

Remarks on result:

other: lung tissue were used for the comet assay

Sex:

female

Genotoxicity:

negative

Remarks:

after 200 min recovery

Toxicity:

no effects

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

not examined

Remarks on result:

other: BAL cells used for the comet assay

Sex:

female

Genotoxicity:

positive

Remarks:

within 0-200 min recovery

Toxicity:

no effects

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

not examined

Remarks on result:

other: BAL cells used for the comet assay

Sex:

female

Genotoxicity:

negative

Toxicity:

no effects

Vehicle controls validity:

not examined

Negative controls validity:

valid

Positive controls validity:

not examined

Remarks on result:

other:

Remarks:

No mutations observed in the cII gene in lung homogenates from Muta TM mice

Additional information on results:

BALB/c mice (single exposure): see Table1 and Figure 1.
The recovery period when the tail moments were increased above the reference interval was in the time interval 0–200 min after the exposure. During this time, the tail moments were dose-dependently increased (P< 0.001, ANOVA) with a 1.6- and 2.6-fold increase over the untreated controls at 1 and 2 ppm, respectively. No changes were observed after 200 min. Cells prepared from the lung tissue of mice exposed to 2 ppm ozone had no increase in tail moment compared to the air-exposed mice. Staining and microscope analysis of the cell composition in the BAL fluid revealed that most cells were macrophages, and a small but significant infiltration of polymorphic neutrophils or lymphocytes occurred at 2 ppm ozone. The viability of BAL cells was 90–97% and showed no difference between cells from exposed and unexposed mice. No difference in the 8-oxo-dG/dG ratio between mice exposed to 1 or 2 ppm ozone and the unexposed mice could be detected at any time after exposure. mRNA level of ERCC1 was not changed due. IL-6 was induced up to 150-fold in the ozone-exposed animals (2 ppm) sacrificed after a recovery period of 150–200 min but had returned to control values at 16 hours. Other cytokines, IL-1 and TNF alpha were not affected.

Muta TM mice (multiple treatment):
The Muta TM mice did not show a difference in mutation frequency between exposed and control animals. In contrast, the ozone-treated mice tended to have fewer mutations that controls.

Table 1: DNA single strand breaks in BALB/c mice after 90 min exposure to air, 1 or 2 ppm ozone measured in a comet assay

	BAL cells	Lung cells
Air Ozone(1ppm) Ozone(2ppm)	1.01 ± 0.35   (n= 21) 1.56 ± 0.68∗   (n= 11) 2.57 ± 0.66∗∗ (n= 12)	1.0 ± 0.6 (n= 6) n.d. 1.1 ± 0.9 (n= 7)

The values are the mean ± S.D. of the relative tail moments for all mice in the recovery period between 0 and 200 min.

∗ P <0.05 for difference to air (Student’s t-test).

∗∗ P <0.01 for difference to air (Student’s t-test).

Conclusions:

The data show supportive information that an acute 90 min exposure to 1-2 ppm ozone concentrations caused DNA strand-breaks in BAL cells in the lungs of mice shortly after exposure and reversal there-off. No mutagenic effect were detected at the transgenic cII gene in the lungs of MutaTM mouse model.

Executive summary:

The genotoxic and inflammatory effects of ozone were investigated in female mice exposed to ozone for 90 min. The tail moment in bronchoalveolar lavage (BAL) cells from BALB/c mice was determined by the comet assay as a measure of DNA strand breaks. Lung tissue and BAL was isolated from animals with a different recovery period after exposure. Within the first 200 min after exposure, the BAL cells from the mice exposed to 1 or 2 ppm ozone had 1.6- and 2.6-fold greater tail moments than unexposed mice. 200 min after exposure there was no effect observed anymore. It could be ruled out that the effect during the first 200 min was due to major infiltration of lymphocytes or neutrophils. No increase of tail moments in comparison to the negative control was observed in lung tissue cells from BALB/c mice. Unexpectedly, ozone had no effect on the content of 8-oxo-deoxyguanosine (8-oxo-dG) in nuclear DNA or on oxidized amino acids in the lung tissue. The mRNA level of the repair enzyme ERCC1 was not increased in the lung tissue. Inflammation was measured by the cytokine mRNA level in lung homogenates. An up to 150-fold induction of interleukin-6 (IL-6) mRNA was detected in the animals exposed to 2 ppm ozone compared to the air-exposed control mice. Also at 1 ppm ozone, the IL-6 mRNA was induced. The large induction of IL-6 mRNA in the lung took place after DNA strand breaks were induced in BAL. This does not support the notion that inflammatory reactions are the cause of DNA damage. To determine whether these exposures were mutagenic, Muta Mice were exposed to 2 ppm ozone, 90 min per day for 5 days. No treatment-related mutations could be detected in the cII transgene. These results indicate that a short episode of ozone exposure at five times the threshold limit value (TLV) in US induces lung inflammatory mediators and DNA damage in the cells in the lumen of the lung. This was not reflected by an induction of mutations in the lung of Muta TM Mice. This data support the idea that acute exposure to ozone and related damage can be handled by the cell through activation of repair mechanisms or increase of antioxidant levels and subsequently to avoid the introduction of cellular DNA mutations.

Reference 2

Endpoint:

in vivo mammalian cell study: DNA damage and/or repair

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Acceptable, well-documented publication which meets basic scientific principles

Qualifier:

no guideline followed

Principles of method if other than guideline:

40 human volunteers were exposed to 0.21 ppm ozone or filtered air for two hours. 30min and 4.5 h after exposure, DNA damage was determined in isolated lymphocytes using the Fast Micromethod, which detects DNA single-strand breaks.

GLP compliance:

no

Remarks:

Academic research study under non-GLP conditions

Type of assay:

other: Fast Micromethod

Species:

other: human volunteers

Sex:

male

Details on test animals or test system and environmental conditions:

A randomized parallel study with 40 healthy, nonsmoking male subjects exposed to 0.21 ppm O3 or filtered ambient air for 2 h in an exposure chamber at the University Medical Center, Johannes Gutenberg University, Mainz, Germany. O3 concentrations are the highest during summer time. To minimize previous ozone contact before the experiment and to reduce ozone exposure for the control group, the study was conducted during the winter season. To reduce the possible influence of other irritants, the measurements were performed on days without smog, fog, or atmospheric inversion. Subjects were advised to avoid excessive outdoor activities prior to their exposure. They did not take any medication. Three subjects were subsequently excluded because urine samples indicated inhalative smoking. 19 subjects (median age: 25 years) were included in the placebo and 18 in the ozone group (median age: 23.5 years).

Route of administration:

inhalation: gas

Vehicle:

none

Details on exposure:

The volunteers were exposed to O3 in an exposure chamber. During exposure, subjects had to cycle on an ergometer with a power of 100 W two times for 15 min to provide sufficient inhalation.

Duration of treatment / exposure:

2 h

Frequency of treatment:

once

Post exposure period:

0.5 and 4.5 h

Dose / conc.:

0.21 ppm

No. of animals per sex per dose:

19 human volunteers in the control group and 18 human volunteers in the ozone group

Control animals:

yes, concurrent no treatment

Positive control(s):

none

Tissues and cell types examined:

Peripheral blood lymphocytes

Details of tissue and slide preparation:

Lymphocytes were separated from other blood fractions by Ficoll gradient centrifugation. The cells were cryopreserved for 24 h at - 80 °C and finally stored in liquid nitrogen. For the analysis of single-strand breaks, the cell concentration was adjusted to 1 x 10^5 cells/mL.

Tthe Fast Micromethod was used, which is a very sensitive method to detect DNA single-strand breaks. A fluorescent dye, which preferentially binds to double-stranded DNA but not to singlestranded DNA is used. After adding an alkaline solution the cellular double-stranded DNA starts to unwind and the fluorescence decreases. The rate of DNA unwinding and hence the decrease in fluorescence intensity correlates with the number of single-strand breaks.

Statistics:

SSF was calculated for the samples collected 30 min and 4.5 h after exposure. Values of both the groups were compared using Mann–Whitney test for nonparametrical, independent samples. Significance level was set at p< 0.05.

Sex:

male

Genotoxicity:

negative

Toxicity:

not examined

Vehicle controls validity:

valid

Negative controls validity:

not valid

Positive controls validity:

not applicable

Additional information on results:

The mean outdoor atmospheric O3 concentration during the experiments was 0.03 ppm (SD = 0.01). There were no statistical differences between the SSF values of the O3 group and the control group, determined 30 min and 4.5 hours after exposure (See Figure 1 and 2).

Conclusions:

This experimental study provides evidence that an acute two hour exposure of male human subjects to 0.21 ppm ozone did not lead to increased DNA single strand breaks in the peripheral blood lymphocytes 30 min or 4.5 hours after exposure in comparison to the negative control.

Executive summary:

In this randomized parallel study, it was examined whether an acute ozone exposure leads to increased DNA strand breaks in human lymphocytes. 18 human male volunteers were exposed to 0.21 ppm ozone and 19 human male volunteers were exposed to filtered air for two hours in an whole body exposure chamber. During exposure, subjects had to cycle on an ergometer with a power of 100 W two times for 15 min to provide sufficient inhalation. Before exposure, 30 min and 4.5 h after exposure, blood was taken from the human volunteers. DNA single-strand breaks were determined in isolated peripheral blood lymphocytes using the Fast Micromethod. There was no detectable effect after ozone exposure. It was concluded that an acute ozone exposure at the tested concentration does not lead to persistent DNA damage.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Justification for classification or non-classification

Evaluation of the genotoxic properties of a new substance is based on a tiered approach. First in vitro tests in pro- and eukaryotic cells are performed. Based on the outcome of these tests it is decided to further investigate the effects of the substance on these end points in vivo. However ozone is not a new substance and its toxicity, including the effects on genetic material, has already been studied for several decades in in vitro and in vivo tests.

From the collected studies, the most relevant were used in a weight-of evidence approach (summarized in IUCLID section 7.6). Some of the results of the in vitro studies have to be interpreted with caution, because they are generated in non-standardized in vitro tests missing the in vivo anti-oxidant protective means and often without using proper reference controls. The same is true for collected in vivo studies. They are either limited with respect to number of exposure groups, group size and/or exposure duration, and individually do not allow to draw firm conclusions on the genotoxic potential of ozone.

Based on the weight of evidence, Ozone is not considered to warrant classification as genotoxic, mainly based on its reactivity. After inhalation, 90% of ozone is consumed in the epithelial lining fluid covering the respiratory airways and epithelial cellular membranes. The remaining O3 reacts with polyunsaturated fatty acids, amino acids, electron donors such as vitamins or glutathione and proteins. Aldehydes, H2O2 and Criegee ozonides are reaction products and can be seen as mediators of ozone toxicity. By this mechanism O3 is entirely consumed and systemically not available (Pryor et al., 1996), and does not reach the cell nucleus and/or nucleic acid.

Diaz-Llera et al., 2002 showed that ozone induced DNA damage in vitro is mediated by H2O2 which is a reaction product of ozone (and other oxidizing substances) in the organism. Moreover, post-treatment incubation of the treated cells reduced the DNA damage levels back to control level, indicating that cellular defence mechanism as e.g. DNA repair is able to prevent oxidative stress related DNA damage induced by reaction products of ozone.

Similar results were reported by Bornholdt et al., 2002, who showed, that an in vivo short-term exposure to ozone of female mice for 90 minutes at 1 or 2 ppm induces lung inflammatory mediators and that DNA damage in cells of the lumen of the lung was observed within the first 200 minutes after exposure. After 200 minutes post-treatment no DNA damaging effects were observed anymore. Moreover, this effect was not reflected by an induction of mutations in the lung of MutaTMMice, indicating that oxidative DNA damage is not subsequently leading to persistent mutations.

Further supportive data was published by Ferng et al., 1997 who showed, that an in vivo short-term exposure to 1 ppm ozone in guinea pigs for 72 hours could result in an increase of DNA strand breaks in tracheobronchial epithelial cells. This increase occurred together with an increase of total protein indicating an ozone-mediated inflammatory reaction. 96 hours after exposure the DNA damage decreased.

Lee et al, 1997 presented data in which guinea pigs were exposed to ozone for 2 hours to 0.4 and 1.0 ppm ozone. Within one hour after exposure bronchoalveolar cells were obtained from animals and analysed for DNA damage in the Comet assay. An increase of DNA damage was observed, but only at cytotoxic concentrations.

In studies on the genotoxic effects of ozone on human volunteers no genotoxic effects were observed (McKenzie et al., 1977, Guerrero et al., 1979; Finkenwirth et al., 2013).

It can be concluded, that effects in certain studies were not driven by ozone itself, but by reaction products of ozone with cell components or by induced inflammatory processes (like for other strong oxidizing chemicals). These oxidative stress related effects on DNA is transient through activation of repair mechanisms or increase of antioxidant levels and subsequently to avoid the introduction of cellular genotoxicity. The lack of substance specific genotoxic effects is supported by the negative results from a carcinogenicity study (NTP 1994).

The high reactivity of ozone prevents the substance from being systemic bioavailability. Together with the assessment of the available data on genotoxicity and carcinogenicity in a weight-of-evidence approach no classification for mutagenicity is warranted.