Ozone

Endpoint summary

Description of key information

As ozone is a gas, neither oral nor dermal exposure is considered as relevant exposure routes. The pulmonary NOAEC for ozone in the rat, based on chronic exposure (6h/d, 5d/week for 20 months) studies by the Health Effects Institute (HEI), is considered to be 0.12 ppm, bearing in mind the minor incidental structural changes reported at this exposure level. The structural changes present after 2 months of ozone exposure were reported to be similar to those observed after 20 months of exposure. These structural changes had minimal to no measurable effect on overall pulmonary function, so it was suggested that these changes might be a protective adaptive response, e.g. to build tolerance to the injurious effects of ozone.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion

Endpoint conclusion:

no study available

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed

Repeated dose toxicity: inhalation - local effects

Link to relevant study records

Reference

Reference 1

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study, effects on lung morphology were investigated in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male and female F344/N rats were purchased from Simonsen Laboratories (Gilroy, CA) at three weeks of age. After a quarantine period of 14 days, rats were randomly assigned to air or ozone exposure groups, and individually housed in stainless-steel wire-bottom cages. Rats were given NIH-07 open formula pellets (Zeigler Bros., Gardner, PA) and softened tap water ad libitum, except during exposure periods. Relative humidity (55°/rJ ± 15%}, temperature (24° ± 0.7°C), and lighting (12 hours light/12 hours dark) were maintained aulomatically.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

Ozone was generated from 100% oxygen corona discharge (OREC Model 03V5-0, Ozone Research and Equipment Corp., Phoenix, AZ). Ozone concentrations were measured with multiplexed ultraviolet spectrophotometric analyzers (Model 1003-AH, Dasibi Environmental Corp., Glendale, calibrated by a chemical method using neutral-buffered potassium iodide. Ozone in the control atmosphere was below the limit of detection (0.002 ppm). All the rats from the NTP/HEI collaborative study were exposed to 0, 0.12, 0.5, or 1.0 ppm ozone for 6 hours/day, 5 days/week for 20 months. . Animals in both studies were killed one week after the end of the exposure.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer.

Duration of treatment / exposure:

The ozone exposure was 6 hours/day, 5 days per week

Frequency of treatment:

Exposure took place for 20 months

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

A total of 38 rats, equally divided between males and females, were used for this study; the group size from each ozone exposure concentration was
10 from 0.0 ppm,
12 from 0.12 ppm,
8 from 0.5 ppm, and
8 from 1.0 ppm.

Control animals:

yes, concurrent vehicle

Details on study design:

Male and female rats were exposed to filtered air or to 1.0, 0.5, or 0.12 ppm ozone for 6 h/day, 5 days/week for 20 months. Animals were killed one week after the end of the exposure. Immediately after death, lungs were fixed, removed from the carcass and stained. Microdissection was done to prepare blocks for further processing for electron microscopic morphometric analysis.

Positive control:

Not applicable

Observations and examinations performed and frequency:

Not reported

Sacrifice and pathology:

Animals were killed by anaesthesia with sodium pentobarbital, and the tracheas were cannulated. The diaphragms were punctured to deflate the lungs, and the lungs were fixed by instillation at 30 cm of water pressure with 2% glutaraldehyde in 0.85 M sodium cacodylate buffer (350 mOsm; pH 7.4). After fixing the lungs in the chest for 15 minutes, they were removed and stored in fixative until processed. Lung volumes were measured by fluid displacement. Three 2-mm slices of the left lung were cubed into 4 mm x 4 mm pieces. To enhance the visibility of the interstitial matrix components, tissues were stained en bloc in 1% osmium tetroxide for four hours, 1 (%) tannic acid for 2.5 hours, and in 2% uranyl acetate for 2.5 hours (Mercer et al., 1991). The tissue was extensively washed in 8% sucrose after each step, dehydrated in an ethanol series, and embedded in epoxy resin with standard procedures. To assure adequate infiltration of the resin, the tissue blocks were allowed to incubate a long time (approximately 1 hour) in the propylene oxide-resin solution.

MICRODISSECTION
The microdissection technique used to isolate terminal bronchioles and proximal alveolar regions is described before(Chang et al. 1988). Tissue blocks selected arbitrarily were cut in a random orientation into slices 0.5 mm thick. They were examined sequentially with a dissecting microscope. Small airways, identified by their smooth circumference and thicker epithelium, were followed to the bronchiole-alveolarduct junction.
Terminal bronchioles were cut in cross sections to facilitate morphometric analysis of their tubular and oriented structure. The alveolar tissue surrounding the first alveolar duct bifurcation, designated as the proximal alveolar region, also was studied in cross section on sections that presented a distinct alveolar duct at each side of the first alveolar duct bifurcation. Three blocks of each anatomic location were selected randomly, and mounted on blank epoxy blocks. Since the orientation of the terminal bronchioles or proximal alveolar regions in the selected blocks were predetermined, thin sections of the blocks were cut by aligning the diamond knife to tho block faces. Sections were placed on 200-mesh copper grids, poststained in uranyl acetate and lead, and examined on a Zeiss lOC electron microscop (Carl Zeiss Inc., Germany).

Other examinations:

ELECTRON MICROSCOPIC MORPHOMETRY
Electron microscopic morphometric analysis was used to determine tissue volume, surface area, and cell characteristics regions, and terminal bronchioles. A total of 177 morphologic parameters were calculated for each animal (2 related to the whole animal, 94 in the proximal alveolar region, 47 in the random alveolar regions, and 34 in the terminal bronchioles).

Proximal Alveolar Regions
Three sites in the proximal alveolar region were randomly selected from each animal. Thin sections from each of these sites were stained and examined with an electron microscope. Two photomicrographs were taken of each grid square, one in the upper left corner and one in the lower right corner. All nuclear profiles in each grid square used for photography were counted on the electron microscope using Gundersen's rule of forbidden lines (Gundersen 1977). Micrographs were enlarged to x 8500 on 11- by 19-inch photographic paper that previously had been printed with a point-counting lattice of 448 lines, each 1.37 cm long. Points, intercepts, and nuclear profiles were counted to determine cell volume and cell surface densities. The formula used in the morphometric analysis has been described in detail {Barry and Crapo 1965; Chang and Crapo 1990).

Epithelium.
The epithelium is subdivided into alveolar (type I and type II) and bronchiolar-like (ciliated, Clara, and other) cells. The volume of total epithelium and of each of the components of the epithelium were determined. The cell area (numerical density), mean cell volume, and, for epithelial and endothelial cells only, mean cell surface area.

Interstitial Matrix.
Morphometric analysis of the interstitial matrix followed procedures designed by Vincent and associates (1992). The interstitial matrix was subdivided into four compartments: collagen fibers, elastin, basement membranes, and cellular space. Collagen fibers were defined as any arrangement of fibrils that can be delineated by an estimated perimeter and distinguished, according to the density of fibrils in the fibers, from other neighboring components.
Elastin was recognized as amorphous material uniformly stained by tannic acid and uranyl acetate. A basement membrane was juxta~posed to both epithelial and endothelial cells and was easily recognized. The volume of each interstitial matrix component was calculated and normalized to basement membrane surface area in the same manner described for tissues and cells in the proximal alveolar regions.

Endotbelium and Capillaries.
The volume of capillary endothelium normalized to the surface area of the basement cells were measured. The volume of the capillary bed was subdivided into red blood cells and plasma. The plasma component included all white blood cells. Because the results of numerous earlier studies of ozone exposure fail to show changes in white blood cells in the pulmonary vasculature, volume and cell characteristics of white blood cells were
not measured as a separate category.
Evidence of Inflammation.
Inflammatory cells (macro phages and neutrophils) in the alveolar spaces and in the interstitium were used as indicators of inflammation. The volumes and the cell characteristics of both were measured.

Random Alveolar Regions
Tissue and cell volumes were measured morphometrically as indices of the responses to ozone exposure in the total gas exchange region, referred to as the random alveolar region. Three blocks of tissue from each animal were randomly selected from the embedded tissue blocks without prior examination under the dissection microscope and without knowledge of the presence or absence of bronchiolealveolarduct junctions. Fifteen micrographs, taken from the upper left corner of 15 consecutive grid squares, were obtained. Photographs were printed and analyzed in a manner similar to those from the proximal alveolar regions. Only total cell volume, matrix volume, and surface area were measured. Cell characteristics and matrix components were not analyzed.

Terminal bronchioles
The terminal bronchioles were examined using morphometric techniques described by Barry and associates (1988) and Chang and associates (1 988). The complete epithelium of each terminal bronchiole examined was photographed by 25 to 30 overlapping micrographs taken at a magnification of x 2000. Pictures were enlarged to x 8500 and printed on 11- by a 14-inch photographic paper. A montage of each terminal bronchiole was constructed and the portion contributed by each micrograph was marked on the composite. The pictures were then placed under a Merz overlay sheet marked with 224 points. Points falling on each cell type and intercept between test lines and a luminal surface, a basal surface, or a cilium were counted. The number of nuclei of each type of cell in the montage was recorded. The area of the bronchiolar epithelium and the lengths of the luminal surface and the basement membrane were measured by a digitizer. The thickness of the epithelium was derived from these measurerments by assuming that the cross section of a bronchiole was a circle. The volume density of each cell type and the surface densities of the luminal and basement membrane surface areas for each type of cell in relation to the total volume of the terminal bronchiolar epithelium (the reference space) were calculated using point and intercept counts. For each terminal bronchiole, the number of cells is expressed in relation to the surface area of the bronchiolar epithelial basement membrane.

Statistics:

For the purposes of statistical analysis, we developed an analytical approach that would use all information most efficiently. Five categories of injury were established and one or two of the 177 measured parameters identified as the most sensitive indicators for each kind of injury. These we called primary variables. First, a multivariate analysis of variance (MANOVA) was done to test for statistical significance in this vector of primary variables. When the MANOVA revealed a significant relationship, univariate analysis of variance (ANOVA) was performed. If the MANOVA did not demonstrate significance, the variable was not tested further. In this study, six primary variables were established in the proximal and randomly selected alveolar regions, and seven primary variables in the terminal bronchioles were established.
If a primary variable showed statistical significance in the first MANOVA, then a second MANOV A was performed. The second vector for multivariate analysis consisted of 1 to 5 key variables that provided more information than the primary variable alone about the category of injury.
In this study, 15 key variables were identified in the proximal alveolar region and 11 in the random alveolar regions. No key variables were counted in the terminal bronchioles. As for the primary variables, multivariate significance was required for each of the five (corresponding to the five injury categories) key variable MANOVAs before examining univariate significance and comparisons among exposure concentrations. This analysis provides the statistical basis for the statements of significant effects given in this report.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Description (incidence and severity):

lung (see "Details on results"). Other information in this context is presented in the NTP study.

Histopathological findings: neoplastic:

not examined

Details on results:

EFFECTS OF OZONE CONCENTRATION AND GENDER ON THE PROXIMAL ALVEOLAR REGION
The MANOVA applied to the primary variables in the first stage of the statistical analysis revealed an effect due to ozone exposure concentration in the following injury classes: bronchiolarization, interstitial, vascular, and inflammation. No significant effects were attributable either to gender or to an interaction between gender and concentration. Therefore, effects from gender were not considered in subsequent stages of the statistical analysis. Significant effects due to ozone concentration were indicated by changes in the following primary variables: the percentage of bronchiolarization, the volume of interstitium, the volume of alveolar macrophages, and the surface area of capillaries. The volume of type I epithelium did not change following ozone exposure.
An uncontrolled ANOVA indicated that ozone concentration had an effect on a confirmatory variable, the total volume of tissue (epithelium, interstitium, and endothelium) due to increases in the volumes of both epithelium and interstitium.

Epithelium
The epithelium in the proximal alveolar regions was not significantly altered by exposure to 0.12 ppm ozone, but major changes were found in the epithelium of rats exposed to either 0.5 or 1.0 ppm ozone. The higher concentrations of ozone induced epithelial metaplasia (a change of the squamnous alveolar epithelium to cuboidal bronchiolar epithelium, also referred to as bronchiolarization) in the proximal alveolar regions. Normally, only a small number of bronchiolar epithelial cells, if any, can be found extending from terminal bronchioles into the proximal alveolar region.
In control rats, less than 2 % of the surface area in the proximal alveolar region was covered by bronchiolar cells. After exposure to either 0.5 or 1.0 ppm ozone, 7% or 12%, respectively, of basement membrane surface was covered by bronchiolar epithelium, and the volume of bronchiolar epithelium in the proximal alveolar region increased 2.5- fold or 4.5-fold, respectively. The extent of bronchiolarization induced by ozone, therefore, is dependent on dose, although only the effect at 1.0 ppm ozone was statistically significant. The bronchiolar cells lining the alveolar ducts and the alveoli consisted mainly of fully differentiated ciliated cells and Clara cells that were structurally identical to those found in terminal bronchioles. Both types of cells increased in number and volume by the prolonged ozone exposures. Unlike the terminal bronchiolar epithelium in normal proximal alveolar regions, in which the volume of ciliated cells is approximately twice as large as the volume of Clara cells, the metaplastic epithelium in the proximal alveolar regions contained approximately equal volumes of ciliated and Clara cells. There was also a large increase in the volume of unidentified cells (referred to as "other cuboidal cells"). These cells were observed only in the metaplastic proximal alveolar region, and typically contained differentiated features of both ciliated cells and Clara cells, including fiber bundles, secretory granules, basal bodies, and glycogen granules. Preciliated cells and brush cells that are normally found in the terminal bronchioles were not observed in the metaplastic cuboidal epithelium in the proximal alveolar regions. Morphometric analysis of the characteristics of the ciliated and Clara cells in the metaplastic epithelium indicated that there was no effect of ozone exposure on mean cell size or on mean cell surface area.
Although the total epithelial volume increased, duo to the metaplasia of the normal squamous epithelium to a cuboidal type of epithelium, in rats exposed to 0.5 or 1.0 ppm ozone, the total volumes of type I and type II epithelium were not changed by the exposures. Univariate ANOVA, however, showed an effect of ozone concentration on the characteristics of type I epithelial cells. The number of type I cells increased 64% to 74% after exposure to 0.5 or 1.0 ppm ozone, respectively. The size of the type I cells decreased approximately 40%, and the luminal and basal cell surface areas of alveolar type I cells decreased to approximately 50% of the control values. Few structural abnormalities wore noted, except for focal thickening of type I epithelium and small areas of cell necrosis. The characteristics of alveolar type II cells were not altered by the ozone exposures.

Interstitium
The volume of the interstitium increased as a function of ozone concentration. No difference in interstitial volume was noted between control rats and those exposed to 0.12 ppm ozone; however, the volume of interstitium was significantly increased after exposure to 0.5 or 1.0 ppm ozone. The vector of key variables for interstitial injury consisted of the cellular components and the compartments making up the interstitial matrix: acellular space, basement membrane, collagen, and elastin. Exposure to ozone had a significant effect on the volume of each of these variables. Elastin and acellular space were significantly elevated after exposure to 1.0 ppm ozone, and the volumes of cellular interstitial components, collagen, and the basement membrane were significantly increased after exposure to 0.5 or 1.0 ppm ozone.
The total interstitial volume increased 5:3 % after exposure to 0.5 ppm ozone, with a 40% increase in the cellular component and a 60% increase of the noncellular component. After 1.0 ppm exposure to ozone, total interstitial volume increased 71 % with a 44% increase of the cellular component and an 84% increase of the noncellular components. The increase in the volume of cellular interstitium was due mainly to an increase in interstitial fibroblasts that constituted approximately 80% of all interstitial cells. The increase of interstitial volume after exposure to 0.5 ppm ozone arises from small increases of both the number of fibroblasts and their mean cell size. Exposure to 1.0 ppm ozone, on the other hand, resulted in a significant increase in the number, but no increase in the mean cell volume, of interstitial fibroblasts. Furthermore, neither the number nor volume of interstitial cells was changed after ozone exposure.
Of the total noncellular interstitium, 40% to 50% is occupied by collagen. After 20 months of exposure to 0.5 ppm ozone, the volume of collagen increased 64%, and after exposure to 1.0 ppm ozone, the volume increased 78%. Basement membrane accounts for 22% to 26% of the noncellular interstitium. Exposure to 0.5 or 1.0 ppm ozone induced thickening of the basement membrane. The magnitude of changes were similar to those observed with collagen. The thickened basement membranes contained inclusion bodies the origin of which is not known. Elastin makes up only 2% of the total noncellular interstitium, and is mainly localized at septal tips. Exposure to 1.0 ppm ozone for 20 months caused an 80% increase in the volume of elastin. Acellular space made up 22% to 27% of the noncellular interstitium, and its volume increased 11% by exposure to 1.0 ppm ozone.

Endothelium and Capillaries
The vectors of primary variables for vascular injury were the volume of endothelium and the surface area of capillaries. No significant change in the volume of endothelium was observed in this study. The mean cell volume and the mean cell surface area of capillary endothelial cells remained unaltered after 20 months of exposure to either 0.12, 0.5, or 1.0 ppm ozone. The number of endothelial cells increased slightly after exposure to 1.0 ppm ozone. A concentration effect was observed and found to be due to a significant increase in the capillary surface area after exposure to 0.5 ppm ozone. However, exposure to 0.12 or 1.0 ppm ozone did not cause significant change in the parameter of capillary surfaces. The inconsistency of the trend of ozone effect on capillary surface suggests that the change found after 0.5 ppm ozone may not be biologically significant.

Evidence of Inflammation
The primary variable used for analyzing inflammation was the volume of the inflammatory cells (the sum of the alveolar and interstitial macrophages). An ozone concentration effect was found for inflammatory cells by MANOVA. Exposure to 0.12 or 0.5 ppm ozone for 20 months did not change either the number or the size of alveolar macrophages. However, rats exposed to 1.0 ppm ozone exhibited a 113% increase in alveolar macrophages in the proximal alveolar region. Interstitial macrophages were rare in all exposure groups, and no significant changes in the volume of interstitial macrophages were noted after ozone exposure. Other inflammatory cells such as neutrophils and monocytes were not included in the quantitative analysis because previous studies with subchronic and prolonged exposures to ozone had shown little or no involvement of these cells. Qualitative examination of the sections in this study confirmed the near absence of neutrophils and monocytes in the proximal alveolar regions of rats exposed to 0.12, 0.5 or 1.0 ppm ozone for 20 months.

EFFECTS OF OZONE EXPOSURE ON RANDOM ALVEOLAR REGIONS
Because previous studies have shown that the effects of 0.12 ppm ozone are strictly confined to the proximal alveolar regions, the morphometric study of random alveolar regions was carried out only with animals exposed to 0.5 or 1.0 ppm ozone. The set of primary variables was assessed with MANOVA. No statistically significant ozone concentration effect was found. Because the epithelium of the distal alveolar regions, which constitute the great majority of the gas exchange region from which the alveolar region blocks were randomly selected, is composed completely of alveolar type I and type II epithelial cells, the percentage of bronchiolarization measured in all exposure groups was not significantly different from the control group. Type I and type II epithelial volumes were not altered by the exposures. The volumes of cellular interstitium and interstitial matrix and the volumes of the components of the interstitial matrix also were not changed by exposure to any concentration of ozone. The ultrastructure of the alveolar epithelial cells, interstitial fibroblasts, and capillary endothelial cells was normal. No significant increase of inflammatory cells was noted in either the alveolar spaces or in the interstitium. Further studies of the group exposed to 0.12 ppm ozone were not performed because no effect was found for the higher ozone doses.

EFFECTS OF OZONE EXPOSURE ON TERMINAL BRONCHIOLES
Multivariate analysis was carried out using the primary variables listed. Statistically significant effects of exposure to 1.0 ppm ozone were observed in the number of ciliated cells, the number of Clara cells, and the mean cell volume of Clara cells. In addition, no effect was found for the thickness of the terminal bronchiole epithelium, the mean cell volume of ciliated cells, the mean luminal surface areas of both ciliated and Clara cells, and the average diameter of terminal bronchioles.
Five types of cells were studied from the terminal bronchioles, ciliated cells, Clara cells, brush cells, preciliated cells, and unidentified cells. Exposure to 1.0 ppm ozone for 20 months was found to cause a 13% reduction in the total number of cells per unit of basement membrane (mm2) in terminal bronchioles. This reflects the combined result of a cell population shift with loss of ciliated cells but an increase in Clara cells. The total number of ciliated cells reduced 37% from 11,508 cells/mm2 to 7,255 cells/mm2 of basement membrane surface area. The percentage of ciliated cells in terminal bronchioles decreased from 71% in control rats to 51% in rats exposed to 1.0 ppm ozone. Despite the loss of ciliated cells from terminal bronchioles, the characteristics of those cells remained normal except for a 20% increase of their basement membrane surface area. The surface area of cilia per cell was not changed. However, due to the reduced number of ciliated cells, the total density of ciliated surface (or cilia) in terminal bronchioles was reduced. In contrast to ciliated cells, the number of Clara cells in terminal bronchioles increased 54% after exposure to 1.0 ppm ozone, and the volume fraction of Clara cells increased from 21% to 36% of terminal bronchiolar cells. The size of the dome or luminal surface of Clara cells was not affected, but the average size of Clara cells decreased 13%. Aside from swollen cilia, the ultrastructure of ciliated cells and Clara cells was normal. Brush cells were identified by their brush Larders and filament bundles. They consisted of only 2% (by number) of the terminal bronchiolar cells in rats exposed to 0.0 ppm, 0.12 ppm, and 0.5 ppm ozone. Exposure to 1.0 ppm ozone increased the percentage of brush cells in terminal bronchioles to approximately 3%. Preciliated cells, a precursor of ciliated cells containing basal bodies and fibrinogen granules, were found to become smaller after exposure to 0.5 ppm ozone, and had a reduced basal surface area. These minor changes did not follow dose patterns and are probably random variability resulting from the small number of both cell types found in the terminal bronchioles.

Dose descriptor:

NOAEC

Effect level:

ca. 0.12 ppm

Based on:

test mat.

Sex:

male/female

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

yes

Lowest effective dose / conc.:

0.5 ppm

System:

respiratory system: lower respiratory tract

Organ:

alveolar duct

alveoli

bronchi

bronchioles

lungs

trachea

Treatment related:

yes

Conclusions:

The study provides valuable supportive information from morphometric studies of rats exposed to 0.0, 0.12, 0.5, or 1.0 ppm ozone for 20 months. The major finding was that exposure to 0.5 or 1.0 ppm ozone caused significant effects. Changes in the proximal alveolar region consisted of epithelial metaplasia, with replacement of squamous (type I) alveolar epithelial cell types by a cuboidal (type II) bronchiolar epithelium, and a thickening of the interstitium due to increases in both cells and extracellular matrix components. In the terminal bronchioles after exposure to 1.0 ppm ozone, decreases in cilia and in the volume and number of ciliated cells were noted, with a proportional increase in Clara cells. The number of alveolar macrophages examined also increased at this exposure level. No morphometric effects of exposure to 0.12 ppm ozone were observed in the proximal alveolar region or the terminal bronchioles, indicating no effects of a 20-month exposure to 0.12 ppm ozone in these lung locations in the animals. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed publication. The study provides valuable supportive information from morphometric studies of rats exposed to 0.0, 0.12, 0.5, or 1.0 ppm ozone for 20 months. Morphometric techniques were used to examine cellular and tissue changes occurring in male and female rat lungs exposed to ozone for a prolonged time.

F344/N rats were exposed to ozone concentrations of 0.0 (control) 0.12, 0.5, or 1.0 parts per million (ppm), six hours per day, five days per week for 20 months.

The major finding was that exposure to 0.5 or 1.0 ppm ozone caused significant effects. Changes in the proximal alveolar region consisted of epithelial metaplasia, with replacement of squamous (type I) alveolar epithelial cell types by a cuboidal (type II) bronchiolar epithelium, and a thickening of the interstitium due to increases in both cells and extracellular matrix components. In the terminal bronchioles after exposure to 1.0 ppm ozone, decreases in cilia and in the volume and number of ciliated cells were noted, with a proportional increase in Clara cells. The number of alveolar macrophages examined also increased at this exposure level. No morphometric effects of exposure to 0.12 ppm ozone were observed in the proximal alveolar region or the terminal bronchioles, indicating no effects of a 20-months exposure to 0.12 ppm ozone in these lung locations in the animals.

Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³

Reference 2

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study. Effects on lung function were investigated in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC model 03V5-ozonator (Ozone Research and Equipment Corp., Phoenix, AZ) with 100% oxygen. The O3 concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical­specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male and female Fischer-344 rats (four- to five-week-old) Fischer-344 rats were obtained from Simonsen Laboratories (Gilroy, CA). Animals were randomly assigned to exposure or control groups after a 10- to 14-day quarantine. Until the function tests the rats were housed in individual wire cages in 2.0-m3 inhalation exposure chambers (H2000, Hazleton Systems, Aberdeen, MD). The animal maintenance and observation procedures were standard for an NTP-sponsored inhalation bioassay. In brief, the chambers were maintained at approximately 24°C, 59% relative humidity, with a flow rate providing 10 air changes per hour. The exposure rooms were lighted on a 12-hour cycle (from 0600 to 1800). Untreated paper cage board beneath the cages was changed twice daily, and chambers were washed weekly. The rats were provided with a pelleted ration (NIH-07,
Zeigler Bros., Gardner, PA) ad libitum outside exposure hours and with water ad libitum at all times. Feed (NIH-07 open formula meal diet obtained from Zeigler Brothers, Gardners, PA) and water were available ad libitum except during exposure periods. Cage units were rotated vertically (for two-year studies) or horizontally (for lifetime studies) within each chamber on a weekly schedule. Light was provided on a 12-hour light/dark cycle. All animals were observed twice daily for morbidity and mortality. Body weights were recorded weekly for the first 13 weeks and thereafter.
Because of the scope and number of animals required for the NTP/HEI study, the animals that formed the basis of this study were received from several different exposure chambers. The average temperature range ( ± SD) within the exposure chambers over the course of the study was 23.9°C to 24.4°C ( ± 0.7 °C); the relative humidity range was 57.1% to 60.2% (± 7.3%).

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

All animals were housed individually in stainless-steel wire-bottom cages in modified Hazleton-2000 inhalation chambers (Hazleton Systems, Aberdeen, MD). Cage units were rotated vertically (for two-year studies) or horizontally (for lifetime studies) within each chamber on a weekly schedule. Ozone gas was generated from greater than 99.9% pure oxygen using a silent arc (corona) discharge ozonator (Model 03V5-0, OREC, Phoenix, AZ). The time to obtain a concentration 90% of the target concentration after initiation and 10% of the target concentration after termination of the test atmosphere generation and were 30 minutes and 5 to 9 minutes respectively.

Charcoal and high-efficiency particulate air filters were used to purify ambient air, and ambient ozone was removed using potassium permanganate filters.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

The ozone concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical - specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Duration of treatment / exposure:

20 months.

Frequency of treatment:

6h/day; 5d/wk;

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Lung function:
0 ppm: 9 males and 9 females;
0.12 ppm: 4 males and 4 females;
0.5 ppm: 8 males and 10 females;
1.0 ppm: 7 males and 10 females.

Control animals:

yes, sham-exposed

Details on study design:

Concentration levels were based on the National Ambient Air Quality standard (0.12 ppm) and the maximum concentration which the animals would tolerate (1.0 ppm). Animals were randomly assigned to the treatment groups. At one to six days after their last exposure lung function testing was performed.

Positive control:

no.

Other examinations:

Lung function
During six days after the last exposure, a comprehensive series of in vivo pulmonary function measurements were made by plethysmography, including assays of breathing pattern, lung mechanics, airflow limitation, lung volumes, gas distribution, and gas exchange. The following parameters were measured or calculated: respiratory frequency, tidal/minute volume, dynamic lung and quasistatic chord compliance, total pulmonary resistance, total lung capacity, vital capacity, functional residual capacity, residual volume, expiratory reserve volume, CO diffusing capacity, nitrogen washout, forced vital capacity (FVC), peak expiratory flow rate (PEFR), FVC/PEFR, Mean mid expiratory flow (MMEF), MMEF/FVC, expiratory flow (F) at 10, 25 and 50% FVC, F/FVC.

Statistics:

All data, unless indicated otherwise, are expressed as mean values ± 1 SD. The goal of the statistical analysis was to determine if pulmonary function values from ozone-exposed rats differed significantly from those from air-exposed control rats. The significance of exposure effects was analyzed by multiple pairwise comparisons using BMDP software (BMDP7D, "Description of Groups [Strata] with Histograms and Analysis of Variance," Edition 1983, Version 1987, BMDP Statistical Software, Los Angeles, CA). Three contrasts to control were tested using pooled variances and the Bonferroni adjustment for multiple comparisons. The criterion for statistical significance was set at a value of p < 0.05 for all comparisons. Values for the separate genders and for combined genders were analyzed for all parameters. To evaluate the possibility of a dose response to ozone for the respiratory function values, linear regressions were performed using the concentration of ozone and gender as explanatory variables. Gender was treated as a categorical variable and ozone concentration as a continuous variable, so that there was a single slope for ozone with separate intercepts for males and females. A retrospective (posterior) calculation of power was performed for ozone using the actual estimate of the SE of the regression model. The results of this calculation were expressed as the slope for ozone concentration necessary to have 95% power (Neter et al. 1985).

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Histopathological findings: non-neoplastic:

not examined

Histopathological findings: neoplastic:

not examined

Details on results:

Lung function:
Only the mean forced expiratory flow rate at 10% of the FVC for 1.0 ppm exposed males was significantly lower (30%) compared to control rats when normalized by FVC, but differed not when expressed in ml/s. In females a slight significant smaller TLC (3%) and larger VC (4%) resulted in a significantly higher VC/TLC (7%) ratio at 0.5 ppm. The RV and RV/TLC values in these animals were also significantly (40%) lower. Few significant changes on the lung function parameters due to ozone exposure were observed. A weak relationship was shown between a few parameters (vital capacity, RV/TLC, carbon monoxide diffusing capacity, FRC, MMEF/FVC, expiratory flow at 10% FVC and the ozone concentration.

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

male/female

Basis for effect level:

other: lung function

Critical effects observed:

yes

Lowest effective dose / conc.:

0.5 ppm

System:

respiratory system: lower respiratory tract

Organ:

lungs

Treatment related:

yes

Dose response relationship:

yes

Relevant for humans:

yes

Conclusions:

The overall conclusion of the investigations is that a 20 months exposure to 0.12, 0.5 or 1.0 ppm ozone produced minimal changes in the pulmonary function of F344/N rats. A statistically significant effect was observed only in female rats exposed to 0.5 ppm. These animals exhibited a decrease in lung residual volume. No differences from control rats were seen in the group exposed to 0.12 ppm ozone. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed journal. The impact of a 20-months exposure to ozone on the pulmonary function of rats was assessed. Four to ten male and female F344/N rats per group were exposed six hours per day, five days per week, for 20 months to ozone at 0.12, 0.5, or 1.0 parts per million (ppm), or to clean air as controls. One to three days after the last exposure, the rats were anesthetized using halothane, fitted with oral endotracheal and esophageal catheters, and measured using plethysmographic techniques.

The differences between mean values for control and treated rats were tested for significance by multiple comparisons. The values and intersubject variability for more than 30 measured and calculated parameters were similar to those reported previously for rats of similar age. The only consistent exposure-related effect was a small reduction of residual volume measured during slow lung deflation. This trend was observed in most exposure groups, but was most significant in females exposed to ozone at the 0.5 ppm level. Fibrosis and epithelial changes were observed in the terminal bronchiole-alveolar duct region in parallel studies of different rats from the same exposure groups. We hypothesized that these changes stiffened airspace walls and acted to maintain the patency of the air pathway at a lower than normal lung volume during deflation. Overall, the exposures had little impact on the integrated pulmonary function of the lung as measured in anesthetized rats.

No differences from control rats were seen in the group exposed to 0.12 ppm ozone. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Reference 3

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

No guideline followed, experimental research study

GLP compliance:

no

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male and female F344/N rats (Simonsen Laboratories, Gilroy, CA) were randomly assigned at four to five weeks of age to control or ozone exposure groups after a 10- to 14-day quarantine period. The rats were housed during the exposures in individual wire cages within 2.0-m3 inhalation exposure chambers (H2000, Hazelton Systems, Aberdeen, MD) The chambers were maintained at approximately 24°C, 59% relative humidity, and an air flow rate providing 10 air changes per hour. The exposure rooms were lighted on a 12-hour cycle (from 0600 to 1800). Untreated paper cage board beneath the cages was changed twice daily, and chambers were washed weekly. The rats were provided with a pelleted ration (NIH-07, Zeigler Bros., Gardner, PA) ad libitum outside exposure hours and with water ad libitum at all times.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

Ozone was generated from 100% oxygen corona discharge (OREC Model 03V5-0, Ozone Research and Equipment Corp., Phoenix, AZ). Ozone concentrations were measured with multiplexed ultraviolet spectrophotometric analyzers (Model 1003-AH, Dasibi Environmental Corp., Glendale, CA) calibrated by a chemical method using neutral-buffered potassium iodide. Ozone in the control atmosphere was below the limit of detection (0.002 ppm). All the rats from the NTP/HEI collaborative study were exposed to 0, 0.12, 0.5, or 1.0 ppm ozone for 6 hours/day, 5 days/week for 20 months. Animals in both studies were killed one week after the end of the exposure.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer.

Duration of treatment / exposure:

Exposure took place for 20 months.

Frequency of treatment:

The ozone exposure was 6 hour day, 5 days per week.

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Table 1 presented under "details of results " provides an overview of the number of animals examined

Control animals:

yes, concurrent vehicle

Details on study design:

Male and female rats were exposed to filtered air or to 1.0, 0.5, or 0.12 ppm ozone for 6 h/day, 5 days/week for 20 months. Animals were killed one week after the end of the exposure. Immediately after death, the nasal mucociliary apparatus was prepared for assessment of nasla mucociliary function. After functional analysis selectively chosen nasal tissues were prepared for light microscopy and transmission elelectron microscopy TEM examinations.

Positive control:

Not applicable

Observations and examinations performed and frequency:

Not reported

Sacrifice and pathology:

The animals were killed (no details reported) and the entire nasal cavities were taken for further examinations

Other examinations:

ASSESSMENT OF NASAL MUCOCILIARY FUNCTION
Mucociliary function was assessed as rapidly as possible after the time of death, as mucus continues to flow for only 20 to 30 minutes after death in rats (Morgan et al 1984a). The nasal passages were dissected to yield undamaged mucosal surfaces of selected regions of the nose. The mucosal surfaces examined were the lateral walls of the right lateral meatuses; the lateral aspect of the right nasoturbinate; the medial aspects of the right and left maxilloturbinates, the distal lateral wall and the ethmoid turbinates; the nasopharyngeal meatus; the nasal septum (most complete side); and the medial aspect of the left nasoturbinate. The tissue samples were immediately placed in an environmentally controlled chamber, designed for studies of mucociliary function by use of a microscope. The patterns and any abnormal features of mucous flow, such as altered mucous opacity, direction of flow, whirling flow, bubbling, or dissection artefacts were recorded. After collection of subjective data on mucociliary function, video recordings were made of 13 selected areas in the nose with a video camera. After recordings, three of the 13 areas in the nose selected for the functional analyses of mucous flow also were morphometrically analyzed for the amount of stored mucosubstances in the surface epithelium. These three regions were the proximal lateral wall (area 5), the proximal septum (area 9), and the medial aspect of the nasoturbinate (area 12).

MORPHOLOGIC AND MORPHOMETRIC ASSESSMENTS OF THE NASAL MUCOSA
Immediately after video motion analysis, nasal tissues from each rat were fixed and decalcified. The septum, lateral and turbinates of half of the nasal cavity were transversely sectioned at four specific anatomic locations. These four tissue blocks were prepared for morphologic examination of surface epithelial cells and mucosubstances using image analysis. The area of stored stained mucosubstances within the surface epithelium lining the maxilloturbinate, the midseptum, the lateral wall, the lateral surface of the nasoturbinate, and the medial aspect of the nasoturbinate in the proximal nasal airway (tissue block 1) was calculated using imagine software program. Similar morphometry was conducted on the nasoturbinate, the later wall, and the maxilloturbinate in the middle nasal airway (tissue block 2).
Another 18 rats (4 to 5 per group) were killed seven or eight days after the end of the 20-month exposures. Nasal tissues from these animals were taken and prepared for ultrastructural investigations of the nasal surface epithelium by transmission electron microscopy (TEM) analysis to determine the intranasal distribution of the ozone-induced changes to the nasal mucosa. Epithelium sites of interest around one nasal passage were first detected with light microscopy, then complementary sites around the other nasal passage at the same level were sampled for analysis by TEM. Selected tissues were prepared for TEM examinations. Montages of tissue of interest were prepared at a final magnification of 3000x. Differential cell counts were based on counts of all nuclear profiles within the intact epithelium visible on these montages. The number of cells per millimeter of nasal were determined by counting the nuclear profiles per basal lamina length. Approximately 200 to 500 cells per montage, representing basal lamina lengths of 300 to 1000 micrometer, were counted. Type of nasal epithelium cells within the montage epithelium included non-ciliated cells, cuboidal cells, mucous cells, serous cells, mucoserous cells, ciliated cells, brush cells, and basal cells.

Statistics:

STATISTICS FOR MUCOUS FLOW DATA
The statistical significance of flow rate data was assessed using linear regression analysis of trends and Student's t test comparing each ozone-exposed group to controls. Dunnett's criterion for comparing several exposure groups to controls was used to account for multiple comparisons. Flow incidence data were analyzed using Fisher's exact test on all exposure groups and controls simultaneously and separately in pairwise comparisons of each exposure concentration versus controls. Bonferroni adjustments were used to allow for multiple comparisons in the pairwise incidence tests. An overall value of p < 0.05 was used to determine statistical significance in all tests.

STATISTICS FOR MORPHOMETRIC DATA
The natural logarithms of the morphometric data were used for statistical analysis. Data were tested for equality of group means by using an Student's t test with Bonferroni adjustment for multiple comparisons. The criterion for statistical significance was set at p < 0.05.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Description (incidence and severity):

In the study only the nose was examined (For details see section "Details on results".) See NTP-Study for further histopathological findings.

Histopathological findings: neoplastic:

not examined

Details on results:

FUNCTION OF THE NASAL MUCOCILIARY APPARATUS
In ozone-exposed rats there were clear exposure-associated changes in nasal mucociliary function, with some interanimal variation, especially in the group exposed to 0.5 ppm ozone. The maps indicating the extent and location of mucostasis and ciliastasis revealed distinct, concentration-related inhibition of mucociliary function in specific sites in the nose, with the most consistent changes occurring in the lateral meatuses (areas 1, 2, and 3) and the medial maxilloturbinates (areas 4, 5, and 6). These changes were most extensive at 1.0 ppm, more variable in extent at 0.5 ppm, and not observed in the animals exposed to 0.12 ppm ozone. Mucostasis was generally more extensive than ciliastasis, with a zone of active ciliary beating but no mucous flow usually occurring in regions directly distal to areas of ciliastasis. Animals exposed to 1.0 or 0.5 ppm ozone, in addition to ciliastasis and mucostasis in some areas, had distinct exposure-associated changes in areas where flow was maintained. These effects included altered character of the mucus, which was frequently opaque or "milky"; more copious mucus, in some cases with "strings" of viscid mucus adhering to fixed tissues and extending along the line of mucous flow streams; altered direction of mucous flow adjacent to areas of mucostasis; and whirling or "vortex-like" flow. These effects occurred in all areas of the nose in which mucus was flowing, except on the nasal septum. Flow of mucus over the septum appeared generally unaltered by ozone exposure, with little or no effect on the character of the mucus or the flow patterns.

Statistical of Mucous Flow Rates
Video motion analysis of mucous flow in ozone-treated animals generated a complex set of data, summarized in Table 1. A trend test was applied to the data using linear regression analysis. Pairwise t-tests were performed to compare each exposure group with the controls. Areas 1, 2, 4, 5, 6, and 8 (and to a lesser extent areas 3 and 7) showed statistically significant decreases in mucous flow rate in ozone-exposed rats. In the other, generally more distal areas of the nose, no statistically significant trends were even though in some cases flow changes were observed with ozone exposure. Changes at the 0.12 ppm level of exposure were less certain changes that tended to follow generally decreasing concentration-related flow rates were not statistically significant when compared with control values, probably as a consequence of the small sample sizes. However, in area 7, where there was a marginally significant trend for flow rate to decrease at the higher ozone concentration, the values at 0.12 ppm ozone tended to increase (p = 0.0086). A similar pattern of increasing mucous flow rate in the groups exposed to 0.12 ppm ozone was observed in areas 1, 3, 8, 10, and 13, but was not confirmed statistically. Taken together, these results indicate that although the higher concentrations of ozone inhibit mucous flow, 0.12 ppm ozone exposure may induce modest increases in mucous flow rates. Additional analyses were conducted taking into account the gender and terminal body weight of rats and the day they were killed. In areas 4 and 5, flow increased in males; otherwise, there were no significant differences between males and females. In areas 1, 10, 11, and 12, statistical analysis of the flow rate as a function of terminal body weight indicated a decreased flow rate with increasing weight after accounting for the effects of ozone and gender. The report shows the incidence of flow (number of rats in which flow is absent versus number in which flow is present) for areas where there was a significant association between the presence of flow and ozone exposure. Analysis of these data by Fisher's exact test indicated that areas where the number of zero flow values was large tended to occur more frequently at the higher ozone concentrations. The incidence of flow was also examined with respect to the gender of the rats and the day they were killed, but no significant associations were found.

HISTOPATHOLOGY OF THE NASAL MUCOSA
Both male and female rats exposed to 1.0 ppm ozone had significant morphologic alterations of the nasal mucosa that were detected by both light microscopy and TEM. Similar but less severe alterations were evident in rats exposed to 0.5 ppm ozone. No exposure-related lesions were evident in the nasal airways of rats exposed to 0.12 ppm ozone. There were no microscopically detectable differences in the nasal mucosa of rats exposed to 0.0 ppm ozone and those exposed to 0.12 ppm ozone. The most severe alterations induced 0.5 or 1.0 ppm ozone occurred in the nasal mucosa of the lateral wall, the nasoturbinate, and the maxilloturbinate lining the lateral meatus of the proximal and middle regions of the nasal passages. Conspicuous, but less severe, ozone-induced changes were also evident in the respiratory epithelium lining the mid-septum and lateral meatus in the middle nasal airways, the respiratory and olfactory epithelium the ethmoid turbinates in the distal nasal cavity and the respiratory epithelium lining the proximal nasopharynx. No significant histologic alterations were identified in the nasal respiratory epithelium lining the nasal septum in the proximal nasal airways (tissue block 1) of rats exposed to 0.5 or 1.0 ppm ozone. Ozone-induced lesions in the proximal regions (tissue block 1) of the nasal airways included marked mucous cell metaplasia and epithelial hyperplasia in the nasal transitional epithelium lining the entire surface of the lateral meatus (that is, surface epithelium covering the lateral wall, maxilloturbinates, and lateral surfaces of the nasoturbinates in the proximal nasal airways). Similar lesions were maxilloturbinate, and nasoturbinate in the middle region of the nose block (tissue 2) and in the respiratory epithelium lining the maxillary sinus. In rats exposed only to filtered air the nasal transitional epithelium was only one to two cells thick and was composed predominantly of nonciliated cuboidal or columnar cells with little histochemically detectable mucosubstances. In contrast, the nasal transitional epithelium exposed to 1.0 ppm ozone was four to six cells thick and contained numerous columnar mucous cells filled with copious histochemically detectable mucosubstances (both acidic AB-staining and neutral PAS-staining mucosubstances. Mucous cell metaplasia with increased intraepithelial mucosubstances was present to a lesser degree in rats exposed 0.5 ppm ozone. Epithelial cell hyperplasia was not a consistent feature in this latter group of rats. Mucous cell metaplasia was also present in some subepithelial ducts and glands within the lamina propria of the proximo-lateral aspects of the nasoturbinates. This glandular lesion was variable among ozone-exposed rats, but was usually more severe in rats exposed to 1.0 ppm ozone. Mucous cell metaplasia in the nasal transitional epithelium was accompanied by varying numbers of intraepithelial glands composed of several circumscribing mucous cells whose apical surfaces were directed to a common central lumen within the surface epithelium. Intraepithelial gland formation did not accompany the ozone induced mucous cell metaplasia in the respiratory epithelium lining the maxillary sinus.
A mild to moderate mixed inflammatory cell influx of lymphocytes, plasma cells, and neutrophils was present in the nasal mucosa of the lateral walls, maxilloturbinates, and lateral aspects of the nasoturbinates in the proximal half of the nasal airways. Surface or glandular epithelial cell necrosis did not accompany the chronic rhinitis in the ozone-exposed rats. However, a bony atrophy of the maxilloturbinates and the lateral ridges of the nasoturbinates was conspicuous in the proximal nasal airways of rats exposed to 0.5 or 1.0 ppm ozone. Areas of bone resorption (that is, Howship's lacunae) with numerous associated mononuclear leukocytes, osteoclasts, and osteoblasts were present in the atrophic bone of the affected turbinates. Though both male and female rats exposed to 0.5 or 1.0 ppm ozone had bony atrophy this alteration was more conspicuous in the turbinates of male rats.
In addition to the ozone-induced histopathology already described above, numerous eosinophilic globules were scattered throughout the respiratory and olfactory epithelium lining the ethmoid turbinates (that is, endoturbinates and ectoturbinates) in distal nasal passages (tissue blocks 3 and 4) and in the respiratory epithelium lining the nasoturbinates in the middle region of the nasal passages (tissue block 2) of rats exposed to 0.5 or 1.0 ppm ozone. This alteration was present in both males and females. These eosinophilic globules were identified, by TEM, as dilated cisternae of smooth endoplasmic reticulum in the cytoplasm of sustentacular cells in the olfactory epithelium and in the cytoplasm of nonciliated secretory, columnar cells in the respiratory epithelium. The enlarged cisternae were distended by copious proteinaceous material with a moderately electron-dense matrix. Only a few widely scattered eosinophilic globules were present in the similar nasal epithelia of rats exposed to 0.0 or 0.12 ppm ozone.

MORPHOMETRY OF INTRAEPITHELIAL MUCOSUBSTANCES
The effects of ozone exposure on the amount of intraepithelial mucosubstances in surface epithelium in various regions of the proximal and middle aspects (tissue blocks 1 and 2) of the rat nasal airways are presented in the report. Exposure to 1.0 ppm ozone induced increases in intraepithelial mucosubstances in the nasal transitional epithelium lining the lateral aspect of the nasoturbinate times the amount in controls), the maxilloturbinate times the amount in controls), and the lateral wall (27 times the amount in controls). Ozone-induced increases in intraepithelial mucosubstances were slightly smaller in the surface epithelium lining the nasoturbinate, maxilloturbinate, and lateral wall in the middle nasal airway (tissue block 2). No significant differences were found between the amounts of intraepithelial mucosubstances in the respiratory epithelium lining the nasal and the medial aspect of the nasoturbinate in the proximal nasal passage of these ozone-exposed rats and the amounts in controls. Mucous cell metaplasia with increases in intraepithelial mucosubstances was also present in both male and female rats exposed to 0.5 ppm ozone. There were no significant gender-related differences in these increases.
As in the rats exposed to 1.0 ppm ozone, in the rats exposed to 0.5 ppm ozone, there were dramatic increases in the stored intraepithelial mucosubstances in the nasal transitional epithelium lining the nasoturbinate (108 times the amount in controls), maxilloturbinate (78 times the amount in controls) and lateral wall times (14 times the amount in controls) in the proximal nasal passage. These increases, however, were approximately 50% to 70% less than in rats exposed to 1.0 ppm ozone, depending on the specific intranasal region examined. In addition, there were significant, but smaller increases in the amount of mucosubstances within the surface epithelium lining these structures in the middle nasal passage. As in the rats exposed to 1.0 ppm ozone, in the rats exposed to 0.5 ppm ozone, there were no detectable differences between the amounts of mucosubstances in the respiratory epithelium lining the septum and the medial aspect of the nasoturbinates in the proximal nasal passage and the amounts in controls. No significant differences in the amounts of intraepithelial mucosubstances were detected between rats exposed to 0.12 ppm ozone and controls exposed to 0.0 ppm ozone in any of the regions examined.

ULTRASTRUCTURAL MORPHOLOGY IN NASAL TRANSITIONAL EPITHELIUM
The principal ultrastructural difference in the nasal transitional epithelium of rats exposed to 1.0 or 0.5 ppm ozone and that of air-exposed controls was a marked increase in the number of luminal nonciliated cells with varying amounts of secretory granules. These secretory cells were classified as either mucous cells or nonciliated cuboidal cells with small numbers of secretory granules. The mucous cells were tall cuboidal to low columnar in shape with a microvillar luminal surface and abundant secretory granules filling most of the cytoplasm between the nucleus and the luminal surface. In contrast, nonciliated cuboidal cells with secretory granules were cuboidal to low columnar in shape with microvillar luminal surfaces and only a few secretory granules in the very apical of the cell. Both of these cells extended from the basal lamina to the luminal surface. The secretory granules in both cells were membrane-bound with a homogeneous electron-lucent matrix. The granules were either individual or coalescing. Neither mucous cells nor nonciliated cells with secretory granules were present in control rats exposed to 0.0 ppm ozone. Nonciliated cuboidal cells were the most common cell type in the nasal transitional epithelium of all rats. These cells were cuboidal to low columnar in shape and extended from the basal lamina to the luminal surface. These cells had microvillar luminal surfaces with abundant smooth endoplasmic reticulum and numerous mitochondria in their apical cytoplasms. These cells did not contain mucous secretory granules. Basal cells in the nasal transitional epithelium of all rats were elongated to an oval shape with basal surfaces attached to the basal lamina. The nucleus was central, and the cytoplasm contained few organelles. These cells in the nasal transitional epithelium of ozone-exposed rats were not morphologically different from those in controls. Brush cells had distinct ultrastructural characteristics. These cells were often pear-shaped with a wide base containing the nucleus and a narrow apical tip projecting into the airway lumen. The apical tip had long, dense, nonbranching microvilli and numerous microfilaments and microtubules. These cells were not altered by any of the ozone exposures.

MORPHOMETRY OF NASAL TRANSITIONAL EPITHELIUM
The differences in total and differential epithelial cells per millimeter of basal lamina in the nasal transitional epithelium lining the lateral wall in the proximal nasal passages of rats in the four experimental groups were determined. Total epithelial cells were significantly increased in rats exposed to 1.0 ppm ozone, compared with controls, but not in rats exposed to 0.5 or 0.12 ppm ozone. This increase in epithelial cells within the nasal transitional epithelium of rats exposed to 1.0 ppm ozone was due to significant increases in secretory cells (that is, in mucous cells with abundant secretory granules and in nonciliated cuboidal cells with few secretory granules). There were 94 ± 10 secretory cells per millimeter of basal lamina in rats exposed to 1.0 ppm ozone compared with 0 secretory cells per millimeter of basal lamina in controls. The numbers of ciliated cells, nonciliated cuboidal cells without secretory granules, brush cells, and basal cells were statistically similar in controls and rats exposed to 1.0 ppm ozone. Although there was no significant increase in total epithelial cells (that is, hyperplasia) in the rats exposed to 0.5 ppm ozone, the number increased to 71 ± 7 secretory cells per millimeter of basal lamina in ozone-exposed rats compared with 0 secretory cells per millimeter of basal lamina in controls. As in the rats exposed to 1.0 ppm ozone, the increase in secretory cells in the nasal transitional epithelium of rats exposed to 0.5 ppm ozone was due to increases in both mucous cells and nonciliated cuboidal cells with small numbers of secretory granules. There was a concurrent decrease in the number of nonciliated cells without secretory granules in these ozone-exposed rats, suggesting that some of the secretory cells in the ozone-exposed animals may have been derived from nonciliated cuboidal cells without secretory granules. There were no significant differences in the abundance of ciliated, brush, or basal cells between rats exposed to 0.5 ppm ozone and rats exposed to 0.0 ppm ozone (controls). Rats exposed to 0.12 ppm ozone had epithelial cell densities in the nasal transitional epithelium of the lateral wall that were similar to those in controls.

ULTRASTRUCTURAL MORPHOLOGY OF RESPIRATORY EPITHELIUM IN THE NASAL SEPTUM
The nasal respiratory epithelium in all the rats was composed of basal cells, secretory cells (that is, mucous cells, serous cells, and mucoserous cells), and ciliated cells. The respiratory epithelium from rats exposed to 1.0 ppm ozone had morphologic differences from that of controls. The principal ozone-induced alteration was a shift in the number of serous, mucoserous, and mucous cells. The animals exposed to 1.0 ppm ozone had more mucous cells and fewer mucoserous and serous cells than did controls. All of these secretory cells were columnar with microvillar luminal surfaces and basally located oval nuclei. Serous cells had numerous small, membrane-bound individual secretory granules with an electron-dense matrix. Granules were in the apical third of the cell. In contrast, mucous cells were characterized by numerous, large, electron-lucent secretory granules in the apical one-half to three-quarters of the cell. The mucoserous cell contained individual granules with some electron-lucent and some electron-dense granules. In addition, these cells usually contained secretory granules with a varying-sized electron-dense core surrounded an electron-lucent border. Ciliated cells, characterized luminal cilia with basal bodies and numerous mitochondria in the apical portions of the cells, were not morphologically different among the four experimental groups. No ultrastructural features of basal or brush cells were different from those of controls after any ozone exposure, except for mild to moderate basal cell hyperplasia in the respiratory epithelium of rats exposed to 1.0 ppm ozone.

MORPHOMETRY OF RESPIRATORY EPITHELIUM OF THE NASAL SEPTUM
The effects of ozone on the abundance and differential densities of cells in the respiratory epithelium lining the proximal nasal septum are summarized in the report. The only exposure group that differed significantly from controls in the abundance of these epithelial cells was the group to 1.0 ppm ozone. These rats had a 21% increase, compared with controls, in total epithelial cells. This hyperplastic response was due primarily to an increase in basal cells. Though total secretory cell numbers did not increase after exposure to 1.0 ppm ozone in this epithelium, there was a noticeable decline in the numbers of serous and mucoserous cells and a concomitant increase in mucous cells (twice as many mucous cells in ozone-exposed respiratory epithelium as in that of controls).

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

male/female

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

yes

Lowest effective dose / conc.:

0.5 ppm

System:

respiratory system: upper respiratory tract

Organ:

other: nasal cavity

Treatment related:

yes

Dose response relationship:

yes

Relevant for humans:

yes

Conclusions:

The results of this study provide supportive information that rats chronically exposed to 0.5 or 1.0 ppm ozone have significant alterations in the function and structure of the nasal mucociliary apparatus. Minor functional changes observed in rats chronically exposed to 0.12 ppm ozone were interpreted as a physiologic, rather than a pathologic response to ozone exposure. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed journal. This study examined the effects of chronic ozone exposure on the structure and function of the nasal mucociliary apparatus of the rat. Male and female F344/N rats were exposed to ozone concentrations of 0.0 (control), 0.12, 0.5, or 1.0 parts per million, for six hours per day, five days per week for a total of 20 months. All rats were killed seven or eight days after the end of the exposure. Immediately after death mucous flow rates throughout the nasal passages were determined using in vitro video motion analysis. Following assessment of mucociliary function, the nasal tissues were processed for light microscopy and stained with Alcian blue/periodic acid-Schiff to detect intra-epithelium mucus. Image analysis was used to quantitate the amount of mucus within the nasal transitional epithelium.

In rats exposed to 0.5 or 1.0 ppm ozone, mucous flow rates were markedly slower over the lateral wall and turbinates of the proximal third of the nasal airways compared to 0.0 or 0.12 ppm ozone. These intranasal regions in the rats exposed to 0.5 or 1.0 ppm ozone contained marked mucous cell metaplasia and 25 to 300 times more mucous in nasal transitional epithelium (NTE) than controls. In addition rats exposed to 0.5 or 1.0 ppm had marked epithelial hyperplasia in NTE, increased eosinophilic globules in the surface epithelium lining the distal nasal airways and mild to moderate inflammatory cell influx in the nasal mucosa in the proximal and middle nasal passages. Male rats also had conspicuous bony atrophy in maxilloturbinates and nasoturbinates. There were no significant decreases between the mucous flow rates of rats exposed to 0.12 ppm ozone and those of control rats.

Reference 4

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study. Effects on nasal turbinates were investigated in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC model 03V5-ozonator (Ozone Research and Equipment Corp., Phoenix, AZ) with 100% oxygen. The 03 concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical­specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male and female Fischer-344 rats (four- to five-week-old) were obtained from Simonsen Laboratories (Gilroy, CA). Animals were randomly assigned to exposure or control groups after a 10- to 14-day quarantine. Animals were housed in modified Hazelton 2000 inhalation chambers.
Feed (NIH-07 open formula meal diet obtained from Zeigler Brothers, Gardners, PA) and water were available ad libitum except during exposure periods. The rats were housed during the exposures in individual wire cages within 2.0-m3 inhalation exposure chambers (H2000, Hazelton Systems, Aberdeen, MD). The chambers were maintained at approximately 24°C, 59% relative humidity, and an air flow rate providing 10 air changes per hour. The exposure rooms were lighted on a 12-hour cycle (from 0600 to 1800). Untreated paper cage board beneath the cages was changed twice daily, and chambers were washed weekly. The rats were provided with a pelleted ration (NIH-07, Zeigler Bros., Gardner, P A) ad libitum outside exposure hours and with water ad libitum at all times.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

All animals were housed individually in stainless-steel wire-bottom cages in modified Hazleton-2000 inhalation chambers (Hazleton Systems, Aberdeen, MD). Cage units were rotated vertically (for two-year studies) or horizontally (for lifetime studies) within each chamber on a weekly schedule. Ozone gas was generated from greater than 99.9% pure oxygen using a silent arc (corona) discharge ozonator (Model 03V5-0, OREC, Phoenix, AZ). The time to obtain a concentration 90% of the target concentration after initiation and 10% of the target concentration after termination of the test atmosphere generation and were 30 minutes and 5 to 9 minutes respectively.

Charcoal and high-efficiency particulate air filters were used to purify ambient air, and ambient ozone was removed using potassium permanganate filters.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

The ozone concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical - specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Duration of treatment / exposure:

20 months (NTP/HEI) study, and 24 months (NTP) study

Frequency of treatment:

6h/day; 5d/wk;

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Investigation nasal turbinates
NTP/HEI study: 37 animals; 4 to 5 males and 4 to 5 females from each of the four groups
NTP study 127 animals; 4 to 8 males and 23 to 28 females from each of the four groups

Control animals:

yes, sham-exposed

Details on study design:

Concentration levels were based on the National Ambient Air Quality standard (0.12 ppm) and the maximum concentration which the animals would tolerate (1.0 ppm). Animals were randomly assigned to the treatment groups. At one to six days after their last exposure lung function testing was performed.

Positive control:

no.

Sacrifice and pathology:

Animals in both studies were killed one week after the end of the exposure. After death, the entire nasal cavity of each animal in the NTP/HEI study was fixed in 2% glutaraldehyde and decalcified with 10% ethylenediaminetetraacetate in 0.1 M cacodylate buffer in preparation for light microscopy and TEM. Nasal tissues from rats in the NTP study were fixed in 10% neutral-buffered formalin, decalcified, embedded in paraffin, sectioned to a thickness of 5 to 6 µm, and stained with hematoxylin and eosin for microscopic examination.

Nasal tissues from rats in the NTP/HEI 20-month study were ultrastructurally examined using TEM.

MORPHOMETRIC ANALYSIS OF MAXILLOTURBINATES
Morphometrically analysis was done of the nasal tissues of 19 male and 18 female F344/N rats from the NTP/HEI 20-month study, and 25 male and 107 female F344/N rats from the NTP 24-month study. A semiautomatic image analysis system was used to morphometrically determine the total cross-sectional area of the maxilloturbinate in the proximal nasal airway. The individual areas of its three major tissue compartments (i.e., surface epithelium, lamina propria, and bone; also were determined using this imaging system and standard morphometric techniques

MORPHOMETRIC ANALYSIS OF NASAL AIRWAY LUMEN
The total lumenal area and perimeter of the nasal passages in the transverse nasal section from a subgroup of male and female rats exposed for 24 months (n 4 to 6 rats of each gender from each exposure group) were chosen for morphometric image analysis.

Statistics:

The effects of gender, chamber ozone concentration, and the interaction of these factors on the measured parameters were tested using a two-way analysis of variance (ANOVA) Dunnett's criterion for comparing several exposure groups to a control group was used to account for multiple comparisons. An overall value of p < 0.05 was used to determine statistical significance in all tests.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

not examined

Histopathological findings: neoplastic:

not examined

Details on results:

Morphological changes
Morphological changes were restricted to the 0.5 and 1.0 ppm exposure groups.
Both male and female rats exposed to 1.0 ppm and 0.5 ppm ozone had significant morphologic alterations that were detected by light microscopy. The effects were less severe at 0.5 ppm. The most severe alterations occurred in the nasal mucosa of the lateral wall, the nasoturbinate, and the maxilloturbinate lining the lateral meatus.
The principal ozone-induced lesions in this proximal region of the nasal airways included marked thickening of the lumenal surface epithelium, bony atrophy of the nasal turbinates (i.e., maxilla turbinates and lateral ridge of nasaturbinates), and a conspicuous thinning of the lamina propria (the tissue between the outer surface epithelium and the inner bone of the turbinates).

Inflammatory cells were present in the nasal mucosa of the lateral walls, the maxilloturbinates and the nasalturbinates of rats exposed to 1 ppm and 0.5 ppm, but less severe at this level.

Morphometry of maxilloturbinates.
Morpholometric changes were restricted to the 0.5 and 1.0 ppm exposure groups
Only in animals that were exposed to 0.5 and 1.0 ppm statistically significant alterations were found in the total cross-sectional area of the maxilloturbinates, and the area of the individual compartments comprising the maxilloturbinates (i.e. turbinate bone (atrophy, lamina propria (reduced area) and surface epithelium (thickening; hyperplasia and metaplasia). These observations were done in both the 20 month and 24 month exposure study.

Atrophy of the osseous tissue of the maxilloturbinates resulted in a conspicuous dorsoventral shortening of the maxilloturbinate in each nasal passage. The epithelial thickening was due to mucous cell metaplasia and epithelial hyperplasia in the nasal transitioneal epithelium lining the entire surface of the lateral meatus.
The nasal transitional epithelium rats exposed to 0.5 or 1.0 ppm ozone was four to six cells thick (2 cells in control) and contained numerous mucous cells. In addition, small focal areas of nonkeratinizing squamous cell metaplasia were occasionally found on the dorsal aspect of the maxilloturbinates, on the lateral scroll and lateral ridge of the nasoturbinates, and on the midaspect of the lateral wall.

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

male/female

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

yes

Lowest effective dose / conc.:

0.5 ppm

System:

respiratory system: upper respiratory tract

Organ:

other: nasal cavity

Treatment related:

yes

Dose response relationship:

yes

Conclusions:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed journal. The overall conclusion of the investigations is that chronic exposures to 0.5 or 1.0 ppm ozone result in morphological and morphometric changes in the nasal maxilloturbinates, including atrophy of the nasal bone. No differences from control rats were seen in the group exposed to 0.12 ppm ozone. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

The principal objective of the present study was to assess morphometrically the severity of the ozone-induced changes in the bone tissue of the maxilloturbinates in chronically exposed rats. Male and female F344/N rats were exposed to 0, 0.12, 0.5, or 1.0 ppm ozone, 6 hours/day, 5 days/week for 20 or 24 months. Rats were killed one week after the end of the exposure, and nasal tissues were processed for light and electron microscopy. Using image analysis and standard morphometric techniques, the amounts of bone, surface epithelium, and lamina propria comprising the maxilloturbinates were estimated by measuring the cross-sectional area of each tissue compartment at a defined location in the proximal nasal passage. Both male and female rats had significant morphologic and morphometric changes in the maxilloturbinates after prolonged exposures to 0.5 or 1.0 ppm ozone, but not to 0.12 ppm ozone. Ozone-exposed rats had significant reductions in the cross-sectional area of turbinate bone, reflecting the loss of bone in the maxilloturbinate after prolonged exposure. This ozone-induced bony atrophy was more severe in male than in female rats. Using electron microscopy, numerous bone-resorption sites were identified on the outer, periosteal, surface of the turbinate bone in ozone-exposed animals. Rats with bony atrophy also had a conspicuous influx of mixed inflammatory cells into the lamina propria surrounding the turbinate bone. In addition, ozone exposures caused reductions in the area of lamina propria, due to blood vessel constriction, and increases in the area of the surface epithelium, due to hyperplasia and metaplasia. The results of the present study demonstrated that prolonged exposure of rats to ozone can cause marked loss of turbinate bone. The severity of this ozone-induced bony atrophy in rats is dependent on both concentration and gender. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Reference 5

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study, effects on collagen in the lungs were investigated by means of histological and biochemical methods in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC model 03V5-ozonator (Ozone Research and Equipment Corp., Phoenix, AZ) with 100% oxygen. The O3 concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical­specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male and female Fischer-344 rats (four- to five-week-old) Fischer-344 rats were obtained from Simonsen Laboratories (Gilroy, CA). Animals were randomly assigned to exposure or control groups after a 10- to 14-day quarantine. Animals were housed in modified Hazelton 2000 inhalation chambers. The average temperature range ( ± SD) within the exposure chambers over the course of the study was 23.9°C to 24.4°C (± 0.7 °C); the relative humidity range was 57.1% to 60.2% (± 7.3%).

Feed (NIH-07 open formula meal diet obtained from Zeigler Brothers, Gardners, PA) and water were available ad libitum except during exposure periods. Cage units were rotated vertically (for two-year studies) or horizontally (for lifetime studies) within each chamber on a weekly schedule. Light was provided on a 12-hour light/dark cycle. All animals were observed twice daily for morbidity and mortality. Body weights were recorded weekly for the first 13 weeks and thereafter.

Because of the scope and number of animals required for the NTP/HEI study, the animals that formed the basis of this study were received from several different exposure chambers. The average temperature range ( ± SD) within the exposure chambers over the course of the study was 23.9°C to 24.4°C ( ± 0.7 °C); the relative humidity range was 57.1% to 60.2% (± 7.3%).

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

All animals were housed individually in stainless-steel wire-bottom cages in modified Hazleton-2000 inhalation chambers (Hazleton Systems, Aberdeen, MD). Animals were randomly assigned to O3 exposure or control groups after a 10- to 14-day quarantine. Cage units were rotated vertically (for two-year studies) or horizontally (for lifetime studies) within each chamber on a weekly schedule. Ozone gas was generated from greater than 99.9% pure oxygen using a silent arc (corona) discharge ozonator (Model 03V5-0, OREC, Phoenix, AZ). The time to obtain a concentration 90% of the target concentration after initiation and 10% of the target concentration after termination of the test atmosphere generation and were 30 minutes and 5 to 9 minutes respectively.

Charcoal and high-efficiency particulate air filters were used to purify ambient air, and ambient ozone was removed using potassium permanganate filters.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

The ozone concentration in each chamber was monitored by a multiplexed Dasibi model 1003-AH ultraviolet spectrophotometric analyzer (Dasibi Environmental Corp., Glendale, CA). The monitor was calibrated by comparing it with a chemical - specific, calibrated monitor (neutral-buffered potassium iodide method) simultaneously sampling the exposure chambers.

Duration of treatment / exposure:

20 months.

Frequency of treatment:

6h/day; 5d/wk;

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Biochemical and histopathological evaluations:
0, 0.5 and 1.0 ppm: 6;
0.12 ppm: 3.

Analysis of carbohydrates of connective tissue:
0 ppm: 7 males and 7 females;
0.12 ppm: 3 males and 3 females;
0.5 ppm: 6 males and 7 females;
1.0 ppm: 7 males and 7 females.

Lung function:
0 ppm: 9 males and 9 females;
0.12 ppm: 4 males and 4 females;
0.5 ppm: 8 males and 10 females;
1.0 ppm: 7 males and 10 females.

Control animals:

yes, sham-exposed

Details on study design:

Concentration levels were based on the National Ambient Air Quality standard (0.12 ppm) and the maximum concentration which the animals would tolerate (1.0 ppm). Animals were randomly assigned to the treatment groups. Rats were killed one week after pulmonary function testing. The right cranial and middle lobes were dissected free, trimmed of extraneous tissue, and frozen on dry ice for biochemical analysis (Last et al., 1994).

Positive control:

no.

Sacrifice and pathology:

Rats were killed one week after pulmonary function testing. The right cranial and middle lobes were dissected free, trimmed of extraneous tissue, and frozen on dry ice for biochemical analysis.
Right accessory lung lobes from 21 rats were inflated with 10% buffered formalin, embedded in paraffin and sections were cut. Sections were stained with hematoxylin and eosin for morphologic examination and with Masson's trichome for identification of semiquantitative analysis of interstitial collagen.
The animals were distributed among the groups as follows:
0 ppm: 2 males and 3 females;
0.12 ppm: 2 males and 2 females;
0.5 ppm: 3 males and 3 females;
1.0 ppm: 3 males and 3 females.

Other examinations:

Biochemical and histopathological evaluations:
1. collagen, assayed as 4-hydroxyproline. 4-hydroxyproline was measured as pyrrole by the method of Woessner
2 hydroxylysine-derived collagen cross-links.
3. the ratio of the collagen cross-links dihydroxylysinonorleucine (DHLNL) to hydroxylysinonorleucine (HLNL), a marker of active synthesis of fibrotic lung collagen. Analysis with a modified HPLC method of Reiser and Last, 1986
4. DNA content of lungs cranial lobe as index of the total cell count. DNA was isolated and quantified by a colorimetric reaction of deoxyribose with dephenylamine (Shatkin, 1969)
5. histopathology of the right accesory lobe. Masson's Trichrome staining for interstitial collagen.

Statistics:

All data, unless indicated otherwise, are expressed as mean values ± 1 SD. One-way analysis of variance (ANOVA). (Fisher protected least significant difference [PLSD]) was used to examine whether there were significant differences between groups. Linear regression analysis was used to test for concentration-related responses. The square of the correlation coefficient (r^2) was taken to represent the proportion variable that is related to concentration effects.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

no effects observed

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

effects observed, treatment-related

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Histopathological findings: neoplastic:

not examined

Details on results:

Organ weights: trend towards higher lung weights in females at 0.5 and 1 ppm.

DNA content of cranial lung lobes: For both male and female no statistically significant changes in the ozone exposed groups as compared to controls.

Collagen content (4-hydroxyproline): No statistically significant exposure-related changes in this value were observed in the male and female animals, when 4-hydroxyproline concentration was expressed per gram of lung weight.

Collagen cross-links: For the trifunctional cross-link hydroxypyridinium (OHP) no statistically significant exposure-related changes were observed in the males exposed to ozone. Control females had 12.7 ± 2.1 nmol of OHP per lobe. There was a concentration-related increase in lung OHP content in the females exposed to ozone, up to a value of 21.1 ± 7.7 nmol of OHP per lobe for 1 ppm group. The increase in OHP content was significant when the control group was compared directly with the females exposed to 1.0 ppm ozone. There was an apparent trend toward increased ratios of dihydroxulysinonorleucine ( DHLNL) to hydroxulysinonorleucine (HLNL) in both the male and female rats exposed to 0.5 or 1.0 ppm ozone. The whole lung lobe content of DHLNL, expressed as nanomoles per lobe, appeared to be greater in the rats of both genders exposed to either 0.5 or 1.0 ppm O3.
However, when DHLNL or HLNL levels were expressed as concentrations that were normalized to the lung lobe collagen content as mole per mole of collagen, there was a tendency toward higher values of DHLNL in the rats of both genders exposed to 0.5 or 1.0 ppm 0 3 . On the other hand, there were no exposure-related changes in the concentration of HLNL when the results were expressed on this basis.


Histopathology: Ozone-induced lesions were restricted to the centriacinar regions of the lungs in rats exposed to 0.5 or 1.0 ppm ozone. There was a conspicuous increase in the number of respiratory bronchioles in the rats exposed to 1.0 ppm. The walls of terminal bronchioles, respiratory bronchioles, and proximal alveolar ducts were markedly thickened due to epithelial hyperplasia and interstitial fibrosis. These alterations were less severe in rats exposed to 0.5 ppm. Rats exposed to 1.0 ppm had moderate to marked centriacinar fibrosis, with an average score of 3.8 for interstitial collagen. Rats exposed to 0.5 ppm had less centriacinar fibrosis, with an average score of 2.5 for intramural and interstitial collagen in the centriacinar regions. Lungs from rats exposed to 0.12 ppm could not be distinguished from lungs from control rats. Only minimal amounts of intramural collagen were evident within the centriacinar regions in control animals and in rats exposed to 0.12 ppm. The average scores for intramural and interstitial collagen in the centriacini of rats exposed to 0 and 0.12 ppm were 1 and 1.5, respectively.

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

female

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

yes

Lowest effective dose / conc.:

0.5 ppm

System:

respiratory system: lower respiratory tract

Organ:

lungs

Treatment related:

yes

Dose response relationship:

yes

Conclusions:

This research study is part of the NTP/HEI Collaborative Ozone Project and published in a peer-review journal. The investigations of this study show that lungs from rats chronically exposed to 0.12 ppm ozone could not be distinguished from control rats by histopathological and biochemical examinations. In rats exposed to 0.5 or 1.0 ppm lesions were restricted to the centriacinar regions of the lung. The walls of terminal bronchioles, respiratory bronchioles, and proximal alveolar ducts were markedly thickened due to epithelial hyperplasia and interstitial fibrosis. It is concluded that chronic exposure of rats for 20 months to ozone at concentrations of 0.5 ppm or above for six hours per day, five days per week, causes mild to moderate lung fibrosis, as defined histologically and, in female rats, biochemically.
Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³

Executive summary:

This research study is part of the NTP/HEI Collaborative Ozone Project and published in a peer-reviewed journal. Male and female Fischer-344 rats were exposed either to filtered air (controls) or to 0.12, 0.5, or 1.0 ppm ozone for 6 hours per day, 5 days per week for 20 months. Collagen deposition in lung tissue from these animals was examined to determine whether or not chronic exposure of rats to ozone causes pulmonary fibrosis, as defined biochemically. Several techniques were used to study collagen deposition in the lungs of the animals. These methods included biochemical quantification by analysis of 4-hydroxyproline in lung tissue hydrolysates. The hydroxylysine-derived cross-links in mature collagen were quantified to estimate biochemically the excess of fibrotic collagen in the lung tissue. Biochemical analysis indicated excess collagen in the female rats exposed to 0.5 or 1.0 ppm ozone. Collagen in the lungs of the females also contained relatively more hydroxylysine-derived cross-links than did the lung collagen from age-matched control animals that had breathed only filtered air. Exposure of Fischer-344 rats for 20 months to 0.5 or 1.0 ppm ozone was associated with excess fibrotic lung collagen deposition as defined histologically. In female rats, exposure was also associated with excess deposition as determined biochemically. There was no indication of any significant changes in the lungs of any of the rats exposed to 0.12 ppm ozone, but the number of animals in this group was far too small to conclude whether this was a true no observable-effect level. It is concluded that chronic exposure of rats for 20 months to ozone at concentrations of 0.5 ppm or above for six hours per day, five days per week, causes mild to moderate lung fibrosis, as defined histologically and, in female rats, biochemically.

Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³

Reference 6

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

No guideline followed, experimental research study

GLP compliance:

no

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen.

Species:

rat

Strain:

Fischer 344

Sex:

male

Details on test animals or test system and environmental conditions:

Male Fischer 344 rats were obtained from Simonsen Laboratories (Gilroy, CA) at 4 to 5 weeks of age. Animals were randomly assigned to ozone exposure or control groups after a 10- to 14-day quarantine period. Animals were housed in modified Hazelton 2000 inhalation chambers and exposures were for 6 hours per day (between 7:30 a.m. and 5:30 p.m.), 5 days per week for 20 months. Due to the scope and number of animals required for the NTP/HEI study, the animals that formed the basis of this study were received from a total of 6 different exposure chambers (3 for ozone and 3 for filtered air). The average temperature range (+S.D.) within the exposure chambers over the course of the study was 23.9 to 24.4 (±0.7) C; the relative humidity range was 57.1 to 60.2 (±7.3) percent.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Details on inhalation exposure:

Equal numbers of male and female Fischer 344 rats were exposed to 0, 0.12, 0.50, or 1.0 ppm (nominal concentrations) of ozone, generated from medical-grade oxygen, for 6 h/day, 5 days/week for 20 months in the same rooms in which they were housed for the remainder of the time. Exposure to ozone was performed at Battelle, Pacific Northwest Laboratories (Richland, WA) as part of a collaborative, multilevel study with the National Toxicology Program (NTP) and Health Effects Institute (HEI) to examine the long-term effects of ozone. The average temperature range (± S.D.) within the exposure chambers over the course of the study was 23.9-24.4 (±0.7)°C; the relative humidity range was 57.1-60.2 (± 7.3) %. Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen. Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer. Calibration of the monitor was accomplished by comparison with a chemical-specific calibrated (Neutral buffered potassium iodide method) monitor simultaneously sampling the exposure chambers.
In this study, only male rats exposed to 1.00 ppm were taken for investigations.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer.

Duration of treatment / exposure:

Exposure took place for 20 months

Frequency of treatment:

The ozone exposure was 6 hour day, 5 days per week

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

For this study, a total of four animals from the control group and four animals from the 1.0 ppm ozone exposure group were evaluated.

Control animals:

yes, concurrent vehicle

Details on study design:

Male and female rats were exposed to filtered air or to 1.0, 0.5, or 0.12 ppm ozone for 6 h/day, 5 days/week for 20 months. Four male rats exposed to clean air and four rats exposed to 1.00 ppm were used in this study. Light microscopic, morphometric, and immunohistological approaches were used to determine the distribution and degree of differentiation of ciliated and nonciliated bronchiolar epithelial (Clara) cells lining alveolar ducts of the central acinus, a primary target of ozone.

Positive control:

Not applicable

Observations and examinations performed and frequency:

Not reported

Sacrifice and pathology:

Animals were sacrificed one week after the end of the exposure. Methods of scarifice is not reported in this paper.

Other examinations:

Lung Fixation and Processing for Microscopy

For this study, a total of four animals from the control group and four animals from the 1.0 ppm ozone exposure group were evaluated. Animals for analysis were selected by random number assignment. The lungs were fixed by intratracheal instillation of 2% glutaraldehyde in cacodylate buffer (pH of 7.4, 350 mOsm). In situ fixation of the lungs in the thorax was for 15 minutes at 30 cm fixative pressure. The lungs were collapsed by diaphragmatic puncture before fixative infusion. The fixed lungs were removed by thoracotomy and stored in the same fixative until processed for immunohistochemistry, histochemistry, light microscopy, and scanning electron microscopy. The fixed lungs were trimmed free of all mediastinal contents and the lung volumes measured by fluid displacement. Six to eight regions were selected from the left lung lobe for complementary immunohistochemistry, histochemistry, and embedding for high resolution light microscopy.

Immunohistochemistry and Histochemistry

Portions of the left lung were embedded in glycol methacrylate. One p sections were cut with glass knives on a JB4 microtome. Serial and serial-step sections were stained either with periodic acid Schiff and Alcian blue (pH 2.5) or for Clara cell secretory protein (CCSP) with immunoperoxidase. For immunohistochemistry, the primary antibody, raised in rabbits against rat Clara cell secretory protein, was followed by localization with avidin biotin peroxidase. The avidin biotin peroxidase reagents were obtained from Vector Laboratories, (Burlingame, CA) and used as recommended by the supplier with minor modifications. The distribution of the reaction product was identified on 1 p sections using a scanning laser confocal microscope (Biorad, Watford, England) in the reflectance mode. Methods controls included substitution of the primary antibody with normal rabbit serum.

Scanning Electron Microscopy

For scanning electron microscopy, the distal third of the right middle lobe was critical-point dried using ethanol and carbon dioxide. The distal conducting airways and the BADJs were identified by microdissection of critical-point-dried specimens. The dissected lungs were mounted on stubs sputtercoated with gold and examined with a Phillips 501 microscope.

Bronchiole-Alveolar Duct Isolations for Light Microscopy

Tissue slices cranial and caudal to the hilar level of the left lobe were embedded as large blocks for isolation of centriacinar regions by the methods of Pinkerton et al.. Briefly, tissue slices (approximately 2 x 4 x 6 mm in size) were post-fixed in 1% osmium tetroxide in Zetterquist's buffer, followed sequentially by 1% tannic acid and 1% uranyl acetate in maleate buffer, dehydrated in ethanol and propylene oxide and embedded in either Epon 812 or Araldite 502. Centriacinar regions were isolated by cutting the entire tissue block into slices approximately 0.3 to 0.4 mm thick. Each slice was examined under a dissecting microscope to identify BADJs in longitudinal profile. Criteria for selection was a symmetrical pair of alveolar ducts arising from a single terminal bronchiole. Isolations meeting this selection criteria consistently contained alveolar duct paths in longitudinal profile that extended two to four generations beyond the BADJ. Selected isolations were remounted on BEEM capsules and sectioned at a thickness of 0.5 p with glass knives. Sections were stained with toluidine blue (0.5% in 1% borate buffer).

Quantitative Microscopy

To measure the extent of differentiated bronchiolar epithelium into alveolar ducts, a rigid sampling scheme was employed. Because rearrangement of the BADJ is a feature of chronic exposure to ozone, the distal end of the terminal bronchiole was identified based on locating the most proximal alveolar outpocketing in the airway wall. At the level of the proximal border of the first alveolar outpocketing, a reference point was placed in the geometric center of the airway lumen. From this reference point, a pattern of concentric circles at 100 p intervals was drawn. Overlaying the concentric circle pattern onto each BADJ isolation was facilitated through the use of a Macintosh Ilci computer interfaced to an Olympus BH-2 microscope via a Dage MTI video camera (Michigan City, IN). The bull's eye pattern of concentric circles (set at 100 micro intervals relative to the resolution used) was stored as a computer fle and could be directly overlaid and oriented over the captured computerized image of each isolation. Using a high resolution objective lens (x60), the most distal extent of the most distal ciliated and nonciliated bronchiolar cells down each alveolar duct path was identified. The extent of cell extension down alveolar duct paths was determined using the concentric arc pattern as a measure of distance from a single reference point. If the distance that bronchiolar epithelial cells extended into the alveolar duct was less than 200 p, the actual distance was measured from a distance perpendicular to the most proximal edge of the first alveolar outpocketing. A total of eight alveolar duct isolations per animal were examined from four animals in each group.

Statistics:

Comparisons between control and exposure groups were made using a one-way analysis of variance. To examine the heterogeneity of bronchiolar epithelial cell distance into alveolar ducts, Chi square analysis was used to evaluate and contrast the distribution of values for each group. The level of statistical significance in each instance was set at a P-value less than 0.057.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Description (incidence and severity):

.

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

effects observed, treatment-related

Description (incidence and severity):

see below

Immunological findings:

effects observed, treatment-related

Description (incidence and severity):

see below

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Description (incidence and severity):

see below

Histopathological findings: neoplastic:

not examined

Details on results:

Scanning Electron Microscopy

To obtain a three-dimensional perspective on the centriacinar reorganization occurring with long-term ozone exposure, scanning electron microscopy was performed on complementary halves of BADJs. Using the tilt function on the scanning electron microscope, it was possible to view all sides of the walls in this region. In animals exposed to air and to ozone, the most proximal alveolar outpocketings were within 100 p of the BADJ. In both control and ozone-exposed animals, the epithelial surface of the airway proximal to the first alveolar outpocketing was a mixture of ciliated cells and nonciliated cells, most of which had apical projections into the airway lumen. In control animals, the surface pattern of epithelial cells characteristic of those found in the terminal bronchiole was observed to extend to a variable distance beyond the most proximal alveolar outpocketing. This variation was observed from animal to animal, and from BADJ to BADJ within the same animal. Bronchiolar epithelial cells identified by their surface features, were occasionally found as far distally as the first bifurcation point of the alveolar duct. In animals exposed to ozone, the three-dimensional perspective on all portions of the airway wall indicated that the bronchiolar epithelium extended a number of generations into the alveolar ducts well beyond the first alveolar duct bifurcation ridge. The surface characteristics of these cells within alveolar ducts appeared to be similar to those cells of the terminal bronchiole, although occasionally small regions within some alveolar duct generations contained squamous epithelial cells on the surfaces of alveolar mouth openings as well as in the alveolar outpockets. The extent of bronchiolarization varied from animal to animal and from BADJ to BADJ. However, the surface pattern of bronchiolar epithelial cell types was relatively equal in extent around the entire circumference of affected alveolar ducts. Polarization of the bronchiolar epithelium in relation to the position of the pulmonary arteriole or the number of generations of branching of alveolar ducts in which the epithelium was observed was not apparent. In both control and exposed animals, where cuboidal cells with apical projections were observed in groups, ciliated cells were also present.

Immunohistochemistry of CCSP

As a marker of the differentiation status of the cuboidal cells which we observed by scanning electron microscopy, we used the presence of antigen for CCSR In control animals, the reaction product specific for CCSP was detectable in almost all of the nonciliated cells lining terminal bronchioles. The distribution of the product was clearly discernible by reflected light on 1 p sections. The distribution of the ciliated and nonciliated cells on the same section could be characterized using DIC Normarsky optics. In BADJs where bronchiolar epithelium was present on the bifurcation ridge, CCSP could be detected in the nonciliated cells in this cuboidal epithelium. At high magnification, the distribution of ciliated cells and their relationship to nonciliated cells was detectable. The reaction product for CCSP was distributed most heavily in secretory granules but could also be observed in a reticulated pattern throughout the cell cytoplasm, with the exception of the nucleus. In exposed animals, the reaction product for CCSP was detectable in numerous generations of alveolar ducts. All of the cuboidal epithelium we identified in alveolar ducts as being bronchiolar epithelium by morphological criteria contained nonciliated cells with a positive reaction for CCSP.

Distribution and Extent of Ciliated and Nonciliated Cells

In toluidine blue-stained, 1 p sections, the bronchiolar epithelium was characterized by its dense staining pattern. The heterogeneity in extent of bronchiolar epithelium into alveolar ducts was also discernible. Bronchiolar epithelium occupied a variable extent of the alveolar ducts down different pathways in the same pulmonary acinus. At higher magnification, the densely staining epithelium in these distal regions contained cuboidal cells with identifiable cilia and nonciliated cells with apical projections. Using these criteria, we observed a significant difference in the average. Distance ciliated and nonciliated (Clara) cells extended into alveolar ducts. In control animals, ciliated cells were found an average of 106 p (±23) into the alveolar duct. In contrast, ciliated cells in exposed animals extended an average of 4 times deeper (476 p ± 34) into the alveolar ducts compared to age-matched controls. The heterogeneity of distribution for ciliated cells within alveolar ducts of control animals and animals exposed to ozone. For control animals, approximately half of the isolated alveolar duct paths had no ciliated cells distal to the most proximal alveolus. In 30% of alveolar ducts, ciliated cells were found between 20 and 200 p distal to the most proximal alveolus. In less than 30% of alveolar duct paths, ciliated cells were observed more than 200 p distal. In contrast, for animals exposed to ozone, every alveolar duct examined contained ciliated cells that extended beyond the most proximal alveolus. In over 35% of the alveolar duct paths, ciliated cells extended between 200 and 400 p into the duct. In almost 50% of the alveolar duct paths in exposed animals, ciliated cells were found more than 400 p from the most proximal alveolus. In almost 10% of the isolations, ciliated cells were found 800 p or more into the acinus. x analysis confirmed a significant alteration in the relative distribution of ciliated cell distribution down alveolar duct paths in exposed animals compared with control animals. Clara cells in control animals extended slightly further down alveolar duct paths (122 p ± 23) than did ciliated cells. In animals exposed to ozone, Clara cells were observed 4 times deeper down alveolar duct paths (481 p ± 34) compared to that observed in control animals. Almost 50% of the alveolar duct paths in control animals had Clara cells that extended less than 20 p into the pathway. The variation in extent of Clara cells in different alveolar ducts was similar for both Clara and ciliated cells in control animals . In none of the alveolar ducts in exposed animals were Clara cells found less than 20 p from the most proximal alveolar outpocket. The heterogeneity in the extent of Clara cell distribution down alveolar duct pathways was similar to that observed for ciliated cells . As with ciliated cells, the extent of Clara cell distribution into alveolar ducts was significantly different between control and exposed animals by Chi-square-analysis.

Dose descriptor:

other:

Remarks:

not applicable, only one dose group evaluated

Remarks on result:

not determinable due to adverse toxic effects at highest dose / concentration tested

Critical effects observed:

not specified

Conclusions:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed journal. The study provides supportive information that exposure to 1.0 ppm ozone for 20 months cause reorganization of the epithelium lining in the bronchiole-alveolar duct junction. Well-differentiated ciliated and non-ciliated bronchiolar epithelial cells occur on the epithelial surfaces lining the alveolar duct lumen extending down into alveolar duct pathways. No evidence of inflammation was present in alveolar ducts, suggesting that epithelial cell transformations in alveolar ducts is a natural consequence of lifetime exposures to oxidant gases.

Executive summary:

This research study is part of the NTP/HEI Collaborative Ozone Project published in a peer-reviewed journal. In this publication only the effects of exposure to 1.0 ppm of ozone for twenty months were studied in male Fischer 344 rats. Light microscopic, morphometric, and immunohistological approaches were used to determine the distribution and degree of differentiation of ciliated and non-ciliated bronchiolar epithelial (Clara) cells lining alveolar ducts of the central acinus, a primary target of ozone induced lung injury. Alveolar duct pathways extending beyond the level of the most proximal alveolar out pocketing of terminal bronchioles were isolated in longitudinal profile. The distance that ciliated and non-ciliated bronchiolar epithelial (Clara) cells projected down each alveolar duct pathway was determined by placing concentric arcs radiating outward from a single reference point at the level of the first alveolar out pocketing. A high degree of heterogeneity in the magnitude of bronchiolar epithelial cell extension into alveolar ducts was noted for each isolation and animal age-matched control animals also demonstrated variation in the degree of bronchiolar epithelial cell extension down alveolar ducts. In animals exposed to ozone, a striking similarity was noted by scanning electron microscopy in the surface characteristics of cells lining both terminal bronchioles and alveolar ducts. The presence of Clara cell secretory protein in cells of bronchioles and alveolar ducts was also detected immunohistochemically and visualized using confocal laser scanning microscopy in the reflectance mode. Well-differentiated ciliated and non-ciliated bronchiolar epithelial cells were found lining alveolar septal tips and alveoli up to a depth of 1,000 µ into the pulmonary acinus after 20 months of exposure to ozone. No evidence of inflammation was present in alveolar ducts, suggesting that epithelial cell transformations in alveolar ducts is a natural consequence of lifetime exposures to oxidant gases.

Reference 7

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study as part of the NTP/HEI study, effects on lung morphology were investigated in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen.

Species:

rat

Strain:

Fischer 344

Sex:

male/female

Details on test animals or test system and environmental conditions:

Male F344/N rats were obtained from Simonsen Laboratories (Gilroy, at 4 to 5 weeks of age. Each animal was randomly assigned to an ozone exposure or control group after a 10- to 14-day quarantine period. Animals were maintained at the California Regional Primate Research Center at UC Davis, where they were housed in stainless-steel and glass inhalation chambers of 4.2m3 capacity. The average temperature range within the exposure chambers over the course ofthe was 23.9 to 24.4 °C; the relative humidity range was 57.1% to 60.2%.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

Ozone was generated from 100% oxygen corona discharge (OREC Model 03V5-0, Ozone Research and Equipment Corp., Phoenix, AZ). Ozone concentrations were monitored with multiplexed ultraviolet spectrophotometric analyzers (Model 1003-AH, Dasibi Environmental Corp., Glendale, calibrated by a chemical method using neutral-buffered potassium iodide. The target concentrations for ozone were 0.0 ppm for the control chambers and 0.12, 0.5, and 1.0 ppm for the ozone chambers. The actual ozone exposure concentration over the course of the study in the control chambers was less than 0.002 ppm (below the limit of detection), and the mean concentration (± SD) was 0.12 (± 0.01), 0.51 (± 0.02), and 1.01 (± 0.05) ppm in the ozone chambers.

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer.

Duration of treatment / exposure:

The ozone exposure was 6 hour day, 5 days per week

Frequency of treatment:

Exposure took place for 20 months

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

0.5 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Groupsize used in this study was 8 animals (4 males and 4 females for all groups)

Control animals:

yes, concurrent vehicle

Details on study design:

Male and female rats were exposed to filtered air or to 1.0, 0.5 or 0.12 ppm ozone for 6 h/day, 5 days/week for 20 months. Animals were killed one week after the end of the exposure. Immediately after death, the lung tissue was prepared for assessment of effect on the tracheobronchial epithelium and pulmonary acinus using morphometric analysis and antioxidant enzyme activity.

Positive control:

Not applicable

Observations and examinations performed and frequency:

Not reported

Sacrifice and pathology:

Animals were killed with an overdose of sodium pentobarbital. The lungs were collapsed by diaphragmatic puncture and fixed in situ by intratracheal instillation of 2% glutaraldehyde in cacodylate buffer (pH 7.4, 350 mOsm) for 15 minutes at 30 cm of fixative pressure (Plopper 1990). The fixed lungs were removed by thoracotomy and stored in the same fixative until processing for analysis by histochemistry, light microscopy, and scanning electron microscopy. The fixed lungs were trimmed of all mediastinal contents, and the lung volumes were measured by fluid displacement. Regions of the left lung lobe were selected for complementary studies by histochemistry and high-resolution light microscopy.

Beginning with the trachea, airways were dissected along their long axes to approximately the level of the terminal bronchiole (Plopper 1990}. The dissections were done with the aid of a dissecting microscope (M8, Wild Heerbrngg Instruments, Farmingdale, NY) and fibre-optic illumination. The areas selected for study are identified as the cranial, central, and caudal regions. The cranial region included a medium-sized conducting airway with a short path length, a large cumulative branch angle (change in the path direction), and a relatively small diameter. The cranial centriacinar regions were selected from the acini supplied by this bronchial pathway. The conducting airway selected from the central region had the same number of generations of branching as that from the cranial region, but was much larger in diameter and was part of the axial pathway for conducting airways in the left lobe. The airway selected from the caudal region had a much greater path length and approximately the same diameter as that of the cranial bronchus; however, the angle of deviation was much smaller in the caudal airway. The caudal centriacinar regions were selected from the acini supplied by this airway.

The areas selected to study pulmonary acini were chosen from cranial and caudal regions of the lung at the distal ends of two conducting airway paths, one in which air travels a short path with a large cumulative angle of change in the path direction, and the other selected from a region served by a much longer path with a low cumulative angle of deviation.

Identification of the Bronchiole-Alveolar Duct Junction (BADJ) To measure systematically the distribution of tissue changes in ventilatory units of animals exposed to ozone, a strategy is required that ensues the proper identification of the BADJ. If these junctions are not easily identified, then comparisons between control and treated animals may involve inappropriately matched alveolar duct generations. This problem arises with exposures to high levels of ozone, which cause the bronchiolar epithelium to extend from the level of the terminal bronchiole into the proximal alveolar regions of the ventilatory unit (Boorman et aL 1980; Barret al. 1988). When such a change has occurred, the first alveolar duct generations in control animals may be inadvertently compared with the second and third alveolar duct generations in the lungs of these ozone exposed animals. However, if no alveoli are obliterated during the process of epithelial reorganization, the first alveolar outpocketing along the airway path may serve as a landmark for identifying the original BAD). From each embedded tissue block taken from the cranial and caudal sites in the left lung lobe, Bronchiole-Alveolar Duct Junctions were isolated by the methods of Pinkerton and colleagues {1993).

Other examinations:

Morphometry tracheobronchial epithelium
The thickness and relative abundance (volume fraction) of conducting airway epithelial cells were evaluated by procedures that are discussed in detail elsewhere (Hyde et al. 1990, 1992b; Plopper et al. 1992). All measurements were made using high-resolution light microscopy (4.0 x objective and 0.5 to 1.0-) um sections). Measurements were made from video images captured with a video camera (DAGE MTJ, Michigan City, IN) mounted on an Olympus BH-2 microscope, which was interfaced with a Macintosh Ilci computer running National Institutes of Health IMAGE software. The analysis was performed using a cycloid grid overlay and software for counting points and intercepts (Stereology Toolbox, Davis, CA).

For each bronchus and the trachea, four fields were evaluated. Fields were selected at random by dividing the cross section of the airway into four quadrants, choosing a random angle (between 0° and 90°) from a random number table, and centering the field for evaluation on that angle. In the centriacinar region, at least five centriacinar areas were used for each analysis. Epithelium evaluated for the terminal bronchiole was defined as the epithelium just proximal to the first alveolar outpocketing. Epithelium evaluated for the proximal bronchiole was obtained from a site contiguous with the terminal bronchiole, but 0.5 to 1 mm more proximal.

Morphometric Analysis of Alveolar Septal Tips
Each ventilatory unit isolation was captured as a digital video image on the computer. From a single reference point at the level of the first alveolar outpocketing, a pattern of concentric circles at 100 um intervals was placed over the isolation. These digital images for each isolation served as guides to identify arc intercepts of each circle with tissues along the ducts of each ventilatory unit. Only those arc intercepts with tissues along open duct paths within a 30° angle incident to either side of a line bisecting the ventilatory unit profile were measured.

Morphometric Analysis of the Alveolar Duct Wall
The same images that were captured at 400x magnification for alveolar septal tip thickness measurements based on arc intercept lengths were reanalyzed using a test lattice overlay (21lines) to count points and intercepts in order to derive volume densities. All volume density measurements were normalized to the area of the alveolar surface. In this study, the surface area was defined as the alveolar tissue-air interface. In total, four to eight isolations per animal were used for this analysis. These measurements were made for each 100 um interval down the alveolar duct. The measurements per 100 um interval down the alveolar duct were averaged and expressed as epithelial, interstitial, and capillary density.

Extent of Bronchiolar Epithelium Down Alveolar Duct Paths
To measure the extension of bronchiolar epithelium into alveolar ducts, the same rigid sampling scheme for each ventilatory unit isolation was employed. Using the same concentric circle grid at 100 um intervals, we generated a map of each ventilatory unit. At high magnification (40x), the most distal ciliated and nonciliated bronchiolar cells within the ventilatory units were identified. The extension within the ventilatory unit was determined using the concentric circle pattern as a measure of distance from the central reference point.

Preparation of Tissue Specimens for Antioxidant Activity Determinations
The procedure for obtaining defined specimens of the lung by blunt dissection has been described in detail (Plopper et al. 1991a; Duan et al. 1993).
The following subcompartments were obtained: distal trachea, lobar bronchi, the axial pathway of the largest (major daughter) branch, the pathways of the first and second of the largest daughter branches from the major axial pathway (minor daughters), the most distal three to four generations (distal bronchioles-central acinus) of conducting airway and the proximal acinus, and lung parenchyma that was free of all of the other components. Before dissection, the caudal one-quarter of the left lobe was removed and homogenized for estimating activity in the whole lung. All dissections were completed within 90 minutes of the animal's death. As pieces were removed, they were homogenized in 300 to 1,000 ul of phosphate buffer (50 mM, pH 7.4, 4°C) with microglass homogonizers; the resulting homogenate was centrifuged at 9,000 x g for 10 minutes. Either activities were measured immediately after centrifugation, or supernatants were stored at -80°C for up to one week.

Enzyme Activity Assays
Total SOD activity was measured by the xanthine oxidase-NET assay first described by Beauchamp and Fridovich (1971), and modified for small samples by Oberley and Spitz (1984).
Selenium-dependent GPx was measured by a modification of the method of Paglia and Valentine (1967).
Glutathione S-transferase activity was assayed using 1-chloro-2,4-dinitrobenzene (CDNB) as a substrate. Incubations contained 6 to 15 µg of protein in a final volume of 650 µl (Habig et al. 1974).
DNA content was determined using 4'-6-diamidino-2-phenylindole (DAPI) as a derivatizing agent (Meyer and Grundmann 1984).

Statistics:

Statistical comparisons were performed using a similarly sophisticated, multiple-stage, "step-down" analysis. The investigators identified primary variables that were analyzed at the multivariate level followed by analysis at the univariate level if significance was found at the multivariate level. However, although this strategy was rigorously followed for the data presented in the first and third sections of the Investigators' Report, for the data reported in the second section of the Report on the architectural modeling of the pulmonary acini, it was used only to analyse the mathematical model.
The statistical analyses of the airway data, the ozone dosimetry model, and the antioxidant enzyme activities all were appropriately conducted. However, the statistical analysis of the data on the architectural remodeling of pulmonary acini does not match the complexity and sophistication of the study design. In performing this analysis, the investigators departed from their multiple-stage analysis.
The investigators' departure from their statistical plan accounts, in large part, for the difficulties in interpreting the remodeling data. Bronchiolarization in the centriacinar region was measured using several approaches, and a large amount of biological data was collected. However, most of it was not statistically analyzed. Only the small portion of the data used to generate the mathematical model was statistically analyzed. Unfortunately, examining of all of the results in this second section leads to the conclusion that the changes seen are not consistent in magnitude or site, particularly at the exposure level of 0.12 ppm ozone.
Because no statistical testing was done on the large amount of biological data presented, one is left with only a qualitative impression of the changes seen. As a consequence, the results of these remodeling experiments are not as powerful as they could be, (according the HEI riewers, 1995).

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

not examined

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Description (incidence and severity):

lung

Histopathological findings: neoplastic:

not examined

Details on results:

TRACHEA
Carbohydrate histochemical staining identified AD/PAS positive material in the epithelium of the trachea. Most of the staining was confined to circular inclusions within the apices of nonciliated cells, which ranged in color from a purplish-pink to deep purple. There was little difference in the color range among different exposure groups. There was approximately one half as much material in the tracheas of animals exposed to 1.0 ppm ozone as in control animals exposed to filtered air. Although there was a consistent decrease in stored secretory product as ozone exposure concentration increased, this difference was statistically significant only at 1.0 ppm. The histologic appearance of epithelial cells lining the distal trachea was not appreciably different in control rats and rats exposed to 0.12 or 0.5 ppm ozone.

Morphometric assessment of the average thickness of the tracheal epithelium indicated some reduction in epithelial thickness in the group exposed to 1.0 ppm ozone, but due to the variability at different sites in all animals, these differences were not statistically significant. There was no significant change in the volume density of nonciliated cells.

Cranial Bronchus
There was little difference in the distribution of AB/PAS positive material in this airway generation in animals exposed to 0.12 or 0.5 ppm ozone compared with that in controls. However, in animals exposed to 1.0 ppm ozone, there was a marked increase in the amount of purplish-red to dark purple material observed in spherical and circular inclusions in nonciliated cells and an apparent increase in the number of cells containing AB/PAS-positive material. In this airway, the amount of AD/PAS-positive material stored in epithelial cells, per unit area of basal lamina, was more than six times greater in animals exposed air or to lower ozone concentrations. The histologic appearance of the epithelial cells lining this airway in all exposure groups did not differ remarkably. There were two categories of cells present: ciliated and nonciliated. There was variability from animal to animal and from region to region within the same airway in an individual animal. Differences in the thickness of the epithelium in exposed groups, in the volume fraction of the epithelium composed ofnonciliated cells, and in the volume density of nonciliatod cells lining this airway were not statistically significant.

Central Bronchus
There was approximately one-third as much AD/PAS-positive material stored in the nonciliated cells of this airway as in the trachea of control animals (Table 4), but no significant change in the amount of stored material in ozone-exposed animals. The epithelium lining this airway was composed almost exclusively of nonciliated and ciliated cells varying in shape from cuboidal to low columnar. The thickness of the epithelium, the volume fraction of the epithelium composed of nonciliated cells, and the volume density of nonciliated cells (per unit of surface area) were not significantly altered by ozone exposure.

Caudal Bronchus
In ozone-exposed animals, AB/PAS-positive material increased, in terms of both the amount per cell and the number of cells with positive material. In comparison with control animals, the amount of stored intraepithelial AB/PAS-positive material increased with increasing ozone concentration and was significantly (p < 0.05) greater in the animals exposed to 1.0 ppm ozone. The cellular composition of the epithelium lining this airway was similar to that in the cranial and central airways. The thickness of the epithelium lining the caudal bronchus decreased in relation to rising ozone concentration. The volume fraction of the epithelium composed of nonciliated cells and the volume density of nonciliated cells (per unit surface area) were unchanged by ozone exposure.

Table 4. Stored Secretory Product in Tracheobronchial Airways of Rats Exposed to Ozone for 20 Months (a)

Ozone Concentration (ppm)

Airway Region 0.0 0.12 0.5 1.0


Distal trachea 0.64 ± 0.07 0.53 ± 0.08 0.42 ± 0.06 0.:34 ± 0.06b

Bronchus Cranial 0.04 ± 0.03 0.03 ± 0.03 0.02 ± 0.03 0.25 ± 0.03b

Bronchus Central 0.14 ,, 0.03 0.23 ± 0.03 0.18 ± 0.03 0.16 ± 0.03

Bronchus Cauda 0.06 ± 0.04 0.1.4 ± 0.05 0.10 ± 0.04 0.26 ± 0.04b


a) Values are least-squares means ± SE expressed in µm^3/ µm^2; n = 8 rats (4 male and 4 female) for each concentration group.
b) p < 0.05 compared with the value for animals exposed to 0.0 ppm ozone.


CENTRACINAR AIRWAYS
Terminal bronchioles
There were no statistically significant differences in the thickness of this epithelium in any exposure groups. The volume density of nonciliated cells in the caudal terminal bronchiole after any ozone exposure was greater than that in control animals. This effect was observed only in male rats. There was no detectable AB/PAS-positive material in the nonciliated cells lining this region.

Proximal bronchioles
The epithelium lining the proximal bronchioles had a cellular composition very similar to that of the terminal bronchioles. The epithelium was thicker than that of the terminal bronchioles. The volume fractions and volume densities of ciliated and nonciliated cells were approximately the same in exposed and control animals. There was some variability in the histologic composition of this epithelium both within and between animals.

PULMONARY ACINUS
Scanning Electron Microscpopy
In animals exposed to ozone, the three-dimensional perspective on all portions of the airway wall indicated that the bronchiolar epithelium extended several generations into the alveolar ducts, well beyond the first alveolar duct bifurcation ridge. These cells within alveolar ducts appeared to have surface characteristics similar to those of the terminal bronchiole cells, although occasionally small regions within some alveolar duct generations contained squamous epithelial cells on the surfaces of alveolar mouth openings as well as in the alveolar out pockets. The extent of bronchiolarization varied from animal to animal and from BADJ to BAD]. However, the surface pattern of bronchiolar epithelial cell types was relatively equal in extent around the entire circumference of affected alveolar ducts. Preferential location of the bronchiolar epithelium in relation to the position of the pulmonary arteriole or to the number of generations of branching of alveolar ducts in which the epithelium was observed was not apparent.
Exposure to 0.1.2 or 0.5 ppm ozone also was associated with significant changes in the pattern of epithelial interdigitation of the terminal bronchiole with the alveolar duct. In contrast to the abrupt transition from bronchiolar to alveolar epithelium at BADJs in the lungs of control animals, bronchiolar epithelial cells extended for two to three alveoli into the alveolar duct in animals exposed to 0.12 ppm ozone and to greater depths in the lungs of animals exposed to 0.5 ppm ozone. Unlike the epithelium in animals exposed to 1.0 ppm ozone, the bronchiolar epithelium within alveolar ducts of animals exposed to 0.12 or 0.5 ppm ozone appeared to be limited to surfaces forming the mouth openings of each alveolus in the central acinus.

Morphometric analysis
Interstitium, the data for the cranial and the caudal sites were averaged. Significant gender and concentration effects were found (Table 10). The males had significantly greater interstitial volume densities than the females among rats exposed to 0.12 or 0.5 ppm ozone. Although the values for males were still elevated relative to those for females following 1.0 ppm ozone exposure, the difference was no longer statistically significant. Within each gender, there was a trend of increasing interstitial volume density with increasing ozone concentrations; however, volume density was significantly elevated relative to the control value only following exposure to 1.0 ppm ozone for both the male and female rats.

Table 10. Interstitial Volume Density of the Pulmonary Acinus for Cranial and Caudal Regions Combined (a)

Ozone Concentration
(ppm)
Gender 0.0 0.12 0.5 1.0

Females 1.32 ± 0.11 1.41 ±0.09 1.50 ± 0.06 1.92 ± 0.03b
(5) (6) (5) (4)
Males 1.66 ± .014 1.80 ± 0.15c 1.92 ± 0.06c 2.25 ± 0.22b
(4) (4) (4) (4)

a) Values are means ± SE. Sample size is given in parentheses on the line below the data
b) Significantly different from the control group at a level of p < 0.05.
c) Airways from male rats were significantly different from airways from female rats (p < 0.05).

Table 11. Capillary Volume Density of the Pulmonary Acinus (a)

Ozone Concentration
(ppm)
Gender 0.0 0.12 0.50 1.00
----------------------------------------------------------------------------------------------------------------------------------------------------
Cranial Region
Females 1.17 ± 0.19 1.53± 0.12 1.78 ± 0.19b 1.19 ± 0.10
(4) (6) (5) (3)
Males 1.90±0.17c 1.89 ± 0,19 1.95±0.19 1.37 ± 0.23
(3) (4) (4) (3)

Caudal Region
Females 1.41±0.05 1.65 ± 0.18 1.67 ± 0.16 1.24 ± 0.02
(5) (6) (4) (3)
Males 2.06 ± 0.20c 2.16 ± 0.14c 2.05 ± 0.14 1.51 ± 0.09b
(4) (4) (4) (4)
------------------------------------------------------------------------------------------------------------------------------------------------------
a Values are means± SE. Sample size is given in parentheses on the line below the data.
b Signiflcantly different from the control group at a level of p < 0.05.
c Airways from male rats were significantly different from airways from female rats at a particular site ( p < 0.05).


ANTIOXIDANT ENZYME ACTIVITY

Superoxide dismutase
Exposure of rats to ozone for 20 months produced significant elevations in SOD activity in two subcompartments, the distal trachea and the distal bronchiole-central acinus (Table 12). The activity in the distal bronchiole-central acinus in animals exposed to 1.0 ppm ozone was 1.7 times that in control animals; exposure to 0.5 ppm ozone elevated SOD levels by 1.5 times. The distal trachea showed a concentration-dependent increase in SOD activity; however, there were no significant differences in SOD activity in samples derived from whole-lung homogenates.

Table 12. Superoxide Dismutase Activity in Lung Subcompartments of rats Exposed to Ozone for 90 Days or 20 Months(a)
90-Day Exposure (b) 20-Month Exposure

Compartment 0.0 ppm 0.12 ppm 1.0 ppm 0.0 ppm 0.50 ppm 1.0 ppm
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Distal trachea 295.2 ± 64.3 395.0 ± 88.7 394.4 ± 136.0 136.8 ± 40.7c,d 238.5 ± 80.7d 311.3 ± 140.8d
Lobar bronchus 239.5 ± 49.5 514.1 ± 161.8 397.8 ± 101.2 449.5 ± 108.1c 309.2 ±194.5 370.3 ± 138.8
Major daughter
bronchus 379.5 ± 59.9d 409.1 ± 70.8d 255.9 ± 76.9d 292.6 ± 68.7 308.0 ± 953 215.5 ± 44.1
Minor daughter
bronchus 619.1 ± 179.2d 954.8 ± 84.1d 1,235.6± 276.2d 398.1 ± 115.4 4 89.2 ± 173.0 517.6 ± l83.6
Distal
bronchiolecentral
central acinus 550.3 ± 55.9d 641.4 ± 43.2d,e 1,113.0 ± 238.9d,e 353.8 ± 53.3c,d 543.3 ± 131..4d,e 590.9 ± 200.1d,e
Parenchyma 452.9 ± 37.2 514.0 ± 130.1 604.3 ± 148.9 452.2 ± 149.1 417.9 ± 217.9 409.2 ± 173.8
Whole lung
homogenate 573.2 ± 23.8d 575.5 ± 58.7d 741.l ± 103.1d 397.7 ± 144.5 406.8 ± 172.3 367.5 ± 169.5
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
a) Values are expressed in units/mg of DNA.
b) The 90-day exposure was conducted at the University of California at Davis under conditions identical to those for the 20-month exposure (Plopper et. al. 1994b),
c) Significant (p < 0.05) compared with the group exposed to 0.0 ppm ozone for 90 days.
d) Significance was found to be dependent on concentration; p < 0.05 by regression analysis.
e) Significant (p < 0.05) compared with control group exposed to 0.0 ppm ozone from the same study.

Exposure of rats to ozone for 20 months produced concentration-dependent elevations in GPx activity in two subcompartments: minor daughter bronchus and distal bronchiol-central acinus (Table 13). Glutathione peroxidase activity of animals exposed to 1.0 ppm ozone was approximately 50% greater in the distal bronchiole and minor daughter bronchus than in the corresponding sites in control animals. After exposure to 0.5 ppm ozone, activity was also approximately 1.5 times the control values in the distal bronchiole. There was a significant concentration-dependent decrease in GPx activity in the major daughter bronchus. The value after exposure to 1.0 ppm ozone was approximately 60% of that for filtered air controls. There were no differences in other subcompartments or in the whole-lung homogenates.

Table 13. Glutathione Peroxidase Activity in Lung Subcompartments of Rats Exposed to Ozone for 90 Days or 20 Months (a)

90-Day Exposure(b) 20-Month Exposure

Compartment 0.0 ppm 0.12 ppm 1.0 ppm 0.0 ppm 0.5 ppm 1.0 ppm
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Distal trachea 0.330 ± 0.059 0.413 ± 0.110 0.425 ± 0.165 0.379 ± 0.197 0.466 ± 0.139 0.512 ± 0.043
Lobar bronchus 0.315 ± 0.107 0.52B ± 0.166 0.455 ± 0.151 0.609 ± 0.214 0.597 ± 0.295 0.643± 0.117
Major daughter
bronchus 0.544 ± 0.065c 0.525 ± 0.141c 0.367 ± 0.074c 0.775 ± 0.238c 0.597 ± 0.245c 0.436 ± 0.128c
Minor daughter
bronchus 0.692 ± 0.107 0.794 ± 0.122 0.850 ± 0.100 0.691± 0.184c 0.684 ± 0.097c 1.055 ± 0.275c
Distal bronchiole~
central acinus 0.583 ± 0.073c 0.545 ± 0.101c 0.712 ± 0.099c 0.612 ± 0.145c 0.947 ± 0.327c 1.000 ± 0.172c
Parenchyma 0.829 ± 0.149 0.966 ± 0.221 0.819 ± 0.208 0.787 ±0.249 0.787 ± 0.301 0.691 ± 0.073
Whole lung
homogenate 0.981 ± 0.079 0.961 ± 0.212 0.958 ± 0.083 0.865 ± 0.257 0.915 ± 0.241 1.016 ± 0.383
-------------------------------------------------------------------------------------------------------------------------------------------
a Values are expressed in units/mg of DNA.
b The 90-day exposure was conducted at the University of California at Davis under conditions identical to those for the 20-month exposure: (Plopper el al. 1994b).
c Significance was found to be dependent on concentration; p < 0.05 by regression analysis.

Twenty-month exposure to ozone produced no concentration-dependent changes in GST activity in any lung subcompartments.

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

male/female

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

yes

Lowest effective dose / conc.:

1 ppm

System:

respiratory system: lower respiratory tract

Organ:

trachea

Treatment related:

yes

Conclusions:

This study, part of the NTP/HEI Collaborative Ozone Project, is published in a peer-review report. The study provide information that exposure to 0.5 and 1.0 ppm ozone for 20 months causes cellular reorganization, epithelium thickening and bronchiolarization of the alveolar duct in the centracinar region. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

The effect of ozone on the respiratory system is not confined to a single region or a specific cell type. Ozone-induced injury can occur at all levels of the respiratory system. However, the effects of this oxidant gas throughout the tracheobronchial tree and the lung parenchyma can be highly variable. The doses of ozone delivered to the various regions may also be different, and these differences may have a significant effect on the extent of injury. This study is part of the NTP/HEI Collaborative Ozone Project. To examine the effects of chronic exposure to ozone on the lungs, a systematic sampling approach was used to perform morphometric, histochemical, and enzymatic analyses of selected airway generations and pulmonary acini arising from short and long airway paths of the tracheobronchial tree. The objectives of this study were to define compositional, cytochemical, and architectural changes that occur in epithelial cells of the airways and major tissue components of the pulmonary acini after 20 months of exposure to 0.0, 0.12, 0.5, or 1.0 parts per million (ppm) ozone in male and female F344/N rats. In the trachea and bronchi significant alterations in stored secretory product following exposure to ozone were found, but no changes in epithelial thickness or the volume density of non-ciliated cells were identified. The volume density of non-ciliated cells was significantly increased in terminal bronchioles arising from a long airway path (caudal region) of the left lung. The predominant change within the pulmonary acini was the extension of bronchiolar epithelium beyond the bronchiole-alveolar duct junction into alveoli. This change was concentration-dependent and site-specific, with ventilatory units arising from a short path (cranial region) of the left lung in male rats being most affected. The antioxidant enzymes superoxide dismutase, glutathione peroxidase, and glutathione S-transferase were significantly elevated in the distal bronchiole to central acinus following 20 months of exposure to 0.5 or 1.0 ppm ozone. Changes in antioxidant enzyme levels were more variable in other airway generations. It is concluded in the study that the effects of long-term (20-month) exposure to ozone are dose-dependent and site-specific along the tracheobronchial tree and pulmonary acini of the lungs. With the tissue sampling strategies used in this study, for the first time microdosimetric relations between ozone concentrations and biological changes in precisely delineated regions of the lungs can be defined along the entire lower respiratory tract. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Reference 8

Endpoint:

chronic toxicity: inhalation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Research study is part of the NTP/HEI study. The NTP study is similar to OECD 453. Effects on lung morphology and ultrastructure were investigated in subgroups of rats from a long-term inhalation study.

GLP compliance:

yes

Limit test:

no

Specific details on test material used for the study:

Ozone was generated by corona discharge using an OREC Model 03V5-0 ozonator (Ozone Research and Equipment Corporation, Phoenix, AZ) with 100% oxygen

Species:

rat

Strain:

Fischer 344

Sex:

male

Details on test animals or test system and environmental conditions:

Male F344/N rats were obtained from Simonsen Laboratories (Gilroy, at 4 to 5 weeks of age. Each animal was randomly assigned to an ozone exposure or control group after a 10- to 14-day quarantine period. Animals were maintained at the California Regional Primate Research Center at UC Davis, where they were housed in stainless-steel and glass inhalation chambers of 4.2m3 capacity. The average temperature range within the exposure chambers over the course of the was 23.9 to 24.4 °C; the relative humidity range was 57.1% to 60.2%.

Route of administration:

inhalation: gas

Type of inhalation exposure:

whole body

Vehicle:

clean air

Remarks on MMAD:

MMAD / GSD: n.a.

Details on inhalation exposure:

Animals were exposed to either ozone or filtered air for a six-hour period five days per week for 3 months.
Ozone was generated from 100% oxygen corona discharge (OREC Model 03V5-0, Ozone Research and Equipment Corp., Phoenix, AZ). Ozone concentrations were monitored with multiplexed ultraviolet spectrophotometric analyzers (Model 1003-AH, Dasibi Environmental Corp., Glendale, calibrated by a chemical method using neutral-buffered potassium iodide. The target concentrations for ozone were 0.00 ppm for the control chamber and 0.12 ppm and 1.00 ppm for the ozone chambers Ozone in the control atmosphere was below the limit of detection (0.002 ppm).

All the rats from the NTP/HEI collaborative study were exposed to 0, 0.12, 0.5, or 1.0 ppm ozone for 6 hours/day, 5 days/week for 20 months, (see Harkema 1994)

Analytical verification of doses or concentrations:

yes

Details on analytical verification of doses or concentrations:

Ozone concentration in each chamber was monitored by a multiplexed Dasibi Model 1003-AH (Dasibi Environmental Corporation, Glendale, CA) ultraviolet spectrophotometric analyzer.

Duration of treatment / exposure:

The ozone exposure was 6 hour day, 5 days per week

Frequency of treatment:

Exposure took place for 3 months

Dose / conc.:

0 ppm (nominal)

Dose / conc.:

0.12 ppm (nominal)

Dose / conc.:

1 ppm (nominal)

No. of animals per sex per dose:

Total number of rats per groups is not reported. For some, but not all, investigations the number of animals per group used is reported.

Control animals:

yes, concurrent vehicle

Details on study design:

Male and female rats were exposed to filtered air or to 1.0 or 0.12 ppm ozone for 6 h/day, 5 days/week for 3 months. Animals were killed one week after the end of the exposure. Immediately after death, the lung tissue was prepared for assessment of effect on the tracheobronchial epithelium and pulmonary acinus using morphometry parameters (total epithelium volume density and nonciliated cell volume density), volume density measurements (epithelium, interstitium, macrophages and capillary lumen) and antioxidant enzyme localization.
The study was designed to be able to compare the generated data with those from almost identical studies with animals exposed to the same concentrations of ozone for 20 months as part of the NTP/HEI Collaborative Ozone Project.

Positive control:

Not applicable

Observations and examinations performed and frequency:

Not reported

Sacrifice and pathology:

The animals were killed one day after the conclusion of the exposure. Rats were killed with an overdose of sodium pentobarbital. The lung were fixed, taken and prepared as indicated for the various investigations.

Other examinations:

The technical evaluation of the Health Review Committee was as follows:
The study was performed with rigorous attention to the details of ozone dosimetry and morphometric and immunohistologic analyses. A particular strength of the experimental design was the sampling technique, which ensured that comparable regions of the lung were studied in each animal. The investigators used microdissection techniques to examine sites within the tracheobronchial tree and pulmonary acini by morphometric and immunohistochemical analyses. Morphometric methods were used to determine ozone induced changes in epithelial thickness and the volume fraction of various cell types (epithelial cells, interstitial cells and macrophages) and anatomic compartments such as capilary lumina. An important aspect of these analyses was the proper location of the bronchio-alveolar duct junction (BADJ) which was designated as the site of the first alveolar out-pocketing Pinkerton and colleagues studied changes at BADJ sites that arose from short (cranial) or long (caudal) tracheobronchial paths to the acinar region. They used immunohistologic approaches with both light and electron microscopy to locate and quantify the levels of Cu-Zn SOD and Mn SOD.

Statistics:

For data analyses, the investigators used a series of statistical procedures, established in their earlier study (Pinkerton et al. 1995), that deal effectively with multivariate data. One method consisted of classifying dependent or outcome variables into sets related to various hypotheses. As a first step, these underwent appropriate multivariate analysis of variance (MANOVA) to establish if significance existed. These analyses examined the dependent variables jointly such as ozone dose, length of exposure (3 or 20 months) and their interactions. If statistical (p<0.05) by Hotelling-Lawlet trace test was established by the multivariate analyses, each dependent variable was subject to a second analyses by a two way analysis of variance (ANOVA). A third step consistent of separate t tests performed on variables identified as significant in the second step by the ANOVA F test. The investigators stopped their analyses at any step that did no show statistical significance. They also developed variations of these procedures that incorporated repeated measures analysis. In some cases, they first determined summary statistics, such as slopes, for later use in their analyses. Because they used this step-down procedure, the investigators did not need to adjust for multiple comparisons.

Clinical signs:

not examined

Mortality:

not examined

Body weight and weight changes:

not examined

Food consumption and compound intake (if feeding study):

not examined

Food efficiency:

not examined

Water consumption and compound intake (if drinking water study):

not examined

Ophthalmological findings:

not examined

Haematological findings:

not examined

Clinical biochemistry findings:

not examined

Urinalysis findings:

not examined

Behaviour (functional findings):

effects observed, treatment-related

Description (incidence and severity):

see below

Immunological findings:

not examined

Organ weight findings including organ / body weight ratios:

not examined

Gross pathological findings:

not examined

Neuropathological findings:

not examined

Histopathological findings: non-neoplastic:

effects observed, treatment-related

Description (incidence and severity):

see below

Histopathological findings: neoplastic:

not examined

Details on results:

The technical evaluation of the Health Review Committee summarized the results as follows:

The lungs of male F344/N rats exposed to 0.12 ppm ozone for 3 months did not show any structural changes compared with control rats that breathed clean air. This is consistent with the findings of the 20-month exposure study Pinkerton et al. 1995). Altough the results of both studies indicate that exposure to 0.12 ppm ozone does not produce structural changes in healthy animal lungs, they do not exclude the possibility that prolonged exposure to this or even lower ozone levels may alter lung responses to other agents such as viruses, nor do they exclude the possibility that prolonged exposure may cause deleterious effects to humans with lung disease.
Rats exposed to 1.0 ppm ozone for 2 or 3 months Showed morphometric changes that were similar to those seen after 20 months. For example bronchiolarized metaplasia in alveolar ducts had occurred after 2 months of exposure. This change was confined to the alveoli within 200 µm of a terminal bronchiole. Also in agreement with the findings of the 20-month study was the observation that the largest effects of 1.0 ppm ozone were at BADJs in cranial sites. As the authors point out, because cranial BADJs are associated with a shorter respiratory tract path than BADJs at caudal sites, they would be expected to receive a greater dose of ozone. Thus ozone's effects decreased as the path length leading to the BADJ increased.
The effect of exposure to 1.0 ppm ozone was greater in the epithelium than in the interstitium, which is consistent with ozone being most reactive with the first tissue it encounters.
Exposure to 1.0 ppm ozone for 2 months had no effect of Cu-Zn SOD in any cell type in the terminal bronchioles or the alveolar ducts. In contrast, exposure to this ozone concentration increased the expression of Mn SOD in the mitochondria of II alveolar epithelium cells located within 400 um of the BADJ. Mn SOD expression in Clara cells in the BADJ was unaffected by exposure to 1.0 ppm ozone. These findings suggest that upregulation of Mn SOD by alveolar Type II cells confers resistance to ozone. In their earlier study (Pinkerton et al 1995), the investigators reported an increase in total SOD activity in the BADJ after exposure to 1.0 ppm ozone. The results of the present study suggest that the increased expression of mitochondrial Mn SOD in type II cells was responsible for this change
The investigators propose that Mn SOD in Clara cells plays a lesser role in protection against ozone because its expression was not enhanced. However, the basal level of Mn SOD was higher in Clara cells than in type II cells and the investigators consider Clara cells to be relatively resistant to ozone. Thus, it is possible that the basal level of Mn SOD in Clara cells is sufficient to protect them. In addition to induction of the enzyme in type II cells, the increase in total SOD activity in the BADJ region seen in their NTP/HEI study could have been caused, in part, by bronchiolar metaplasia in alveolar ducts that replaced epithelial cells containing low levels of Mn SOD with Clara cells containing high levels of Mn SOD

Dose descriptor:

NOAEC

Effect level:

0.12 ppm

Based on:

test mat.

Sex:

male

Basis for effect level:

histopathology: non-neoplastic

Critical effects observed:

not specified

Conclusions:

This study, part of the NTP/HEI Collaborative Ozone Project, is published in a peer-reviewed journal. The study provided supportive information that exposure to 1.0 ppm ozone for 3 months caused epithelium thickening and bronchiolarization of the alveolar duct in the centracinar region close to the bronchiole-alveolar duct junction (BADJ). The degree of these changes was similar to that seen after a 20 months exposure to 1.0 ppm ozone. Increase MnSOD activity occur in the type II epithelial cells close to the BADJ. Lungs of male rats exposed to 0.12 ppm for 3 months did not show any structural changes compared to control rats. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³.

Executive summary:

In this study, which is part of the NTP/HEI Collaborative Ozone Project, male F344/N rats were exposed to 0, 0.12 or 1.0 ppm ozone for either 2 or 3 months and compared their observations with those found after 20 months of exposure. The results of this study confirm and extend the original findings. The lungs of male rats exposed to 0.12 ppm ozone for 3 months did not show any structural changes compared with control rats that breathed clean air. The greatest effects of exposure to 1.0 ppm ozone were seen in the centriacinar region close to the BADJ, a site that, as mathematical models predict, is a target area for inhaled ozone. Morphometric changes in the centriacinar region (epithelial thickening, bronchiolarization of the alveolar duct) were seen after 2-3 months of exposure to 1.0 ppm ozone, and the degree of these changes was similar to that seen after 20 months of exposure to 1.0 ppm ozone. Thus, the ozone-induced changes do not appear to have been affected by aging. These changes alter the anatomy of the centriacinar region and it is not known if this is detrimental to lung health over the long term. The increased centriacinar antioxidant enzyme activity reported in the investigators NTP /HEI study after exposure to 1.0 ppm ozone for 20 months could be accounted for by an increase in MnSOD in type II alveolar epithelial cells close to the BADJ after 2 or 3 months of exposure to 1.0 ppm ozone. Taken together, the results indicate that the cellular and antioxidant enzyme changes in response to long-term exposure to ozone occur early and are stable. Accordingly, the early responses to ozone may represent changes that protect the rat lungs during continued exposure to ozone. Suggested NOAEC chronic = 0.12 ppm = 0.24 mg/m³

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

NOAEC

0.235 mg/m³

Study duration:

chronic

Species:

rat

Repeated dose toxicity: dermal - systemic effects

Endpoint conclusion

Endpoint conclusion:

no study available

Repeated dose toxicity: dermal - local effects

Endpoint conclusion

Endpoint conclusion:

no study available

Additional information

No studies, in which internationally accepted guidelines (OECD Test Guideline 412, 413 and 452; EC or EPA, etc.) were followed for evaluation of the toxicity of pure ozone following repeated inhalation exposure, were found in the public domains or the archives of the applicants. However, the scientific literature published in public domains during the last 50 years were screened for data regarding toxic effects of ozone following repeated exposure of animals. This resulted in approximately 100 studies, from which the most relevant ones were retained and summarised in IUCLID. The focus was on studies in rats (Chang et al., 1991, 1992, 1995; Harkema and Mauderly, 1994; Harkema et al., 1994, 1997, 1999; Last et al., 1994; Pinkerton et al. 1993, 1995, 1997; Szarek et al., 1994; Martrette et al., 2011; Gordon et al., 2013, 2014; Radhakrishnamurthy et al., 1994; Pereyra-Munoz et al., 2006; Schelegle et al., 2003a; van Bree et al., 2001; Rivas-Arancibia et al., 2010; Guevara-Guzman et al., 2009; Tesfaigzi et al., 1998) and monkeys (Harkema et al., 1987, 1993; Murphy et al., 2014; Maniar-Hew et al., 2011; Fannuchi et al., 2006; Schelegle et al., 2003b; Murphy et al., 2013; Eustis et al., 1981; Tyler et al., 1988; Carey et al., 2011; Kajekar et al., 2007). Despite the fact that the most of the selected studies are not considered as key studies (e.g. none of the studies was designed in compliance with OECD Test Guidelines), after careful evaluation, this set of studies is considered to have an acceptable study reliability to derive values such as NOAEC for risk assessment (shown in Table 1).

Table 1: Overview of NOAECs

Species	Type of response	Estimated NOAEC	Duration of exposure
Rat	pulmonary	0.12 ppm	6h/d, 5 d/week, for 20 months
Nonhuman primate	pulmonary	below 0.15 ppm	8h/day for 90 days
Rat	brain	below 0.25 ppm	4h/day for 90 days

Pulmonary response

Rat

Late in the 1980s, the US Health Effects Institute (HEI), amongst others, nominated ozone for carcinogenicity and toxicity testing by the National Toxicology Program (NTP) (Boorman et al., 1994, 1995; refer to Section 7.7 Carcinogenicity). The NTP and HEI had a collaboration that allowed HEI-funded investigators access to rats that were exposed to ozone using the same rigorous exposure protocol and quality assurance procedures as the NTP studies. Healthy male and female F344/N rats were exposed for 6h/day, 5 days/week, for 20 months to filtered air (control) or one of three tested concentrations of ozone : the US EPA standard at the time of the study (0.12 ppm), the maximum concentration considered to be compatible with long term survival (1.0 ppm) and an intermediate level (0.5 ppm). The endpoints examined by the HEI included investigations of lung biochemical constituents, structural and cellular changes, lung function, and nasal structure and function. The publications have been appointed as key studies in IUCLID and are described in detail in the next section.

The authors (e.g. from Pinkerton et al., Harkema et al., Tesfaigzi et al.) concluded that repeated exposures to ozone for 20 month in the range of 0.12 to 1 ppm resulted in an initial acute response, evidenced by increased protein and albumin content and increased neutrophil influx in bronchioalveolar fluid, which subsided when exposure continued. Continued exposure resulted in the induction of structural changes in the nasal cavity and the centriacinar region (collagen formation and bronchiolisation). The structural changes present after 2 months of ozone exposure were reported to be similar to those observed after 20 months of exposure. These structural changes had minimal to no measurable effect on overall pulmonary function, so it was suggested that these changes might be a protective adaptive response, e.g. to build tolerance to the injurious effects of ozone. In parallel to the HEI studies, Chang et al. (1991) performed studies on rats exposed to low levels of ozone (0.06-0.25 ppm) for up to 78 weeks. Relative persistent ozone-induced changes (such as epithelium hyperplasia, fibroblast proliferation and bronchiolisation) were observed after 17 weeks of recovery after long-term exposure to 0.25 ppm ozone.

Overall, the pulmonary NOAEC for ozone in the rat, based on chronic exposure (6h/d, 5d/week for 20 months) studies by the HEI, is considered to be 0.12 ppm, bearing in mind the minor incidental structural changes reported at this exposure level. These structural changes had minimal to no measurable effect on overall pulmonary function, so it was suggested that these changes might be an adaptive response.

Monkey

Exposures of aged and young nonhuman primates to low levels of ozone have identified two regions of the respiratory tract that are particularly sensitive to ozone: the transitional epithelial region of the nasal cavity and the centriacinar region. Exposure early in life to low levels of ozone compromises postnatal morphogenesis of tracheobronchial airways resulting in persistent structural changes and persistent effects on pulmonary and systemic innate immune responses later in life. For example, Carey et al. (2011) exposed infant monkeys at age 6 months to ozone (0.5 ppm), to determine effects on the developing nasal airways. Eleven cyclic ozone exposures induced persistent rhinitis, squamous metaplasia, and epithelial hyperplasia in the anterior nasal airways of infant monkeys, resulting in a 39% increase in the numeric density of epithelial cells. Ozone induced a 65% increase in glutathione concentrations at this site. Structural changes in the nasal cavity and the lung centriacinar region were also reported for exposures as low as 0.15 ppm ozone for 90 days (8h/d) (Harkema et al. 1987, 1993). Therefore, the NOAEC for sub-chronic ozone exposure (8h/d for 90 days) of nonhuman primates is below 0.15 ppm.

Non-pulmonary response

The NTP carcinogenicity study (Boorman et al., 1994, 1995), as described briefly above and in detail in section 7.7 Carcinogenicity, was performed in accordance to the OECD 452 test guideline with minor exceptions (e.g. organ weights, ophthalmology, haematology, urinalysis). Animals were observed for clinical signs twice daily and recorded monthly until week 92 of the study and every two weeks until the end of the study. A complete histopathology was performed on all rats and body weight development was recorded throughout the study. A significantly lower body weight was reported for the high dose group (1.0 ppm O₃) when compared to control. The histopathological examinations revealed only local effects in lung, larynx and nose as described above. Based on these observations a systemic NOAEC of 0.5 ppm can be established.

Justification for classification or non-classification

The main toxicity target of repeated exposure of ozone is the respiratory tract.

Late in the 1980s, the US Health Effects Institute (HEI), amongst others, nominated ozone for carcinogenicity and toxicity testing by the National Toxicology Program (NTP). The NTP and HEI had a collaboration that allowed HEI-funded investigators access to rats that were exposed to ozone using the same rigorous exposure protocol and quality assurance procedures as the NTP studies. Healthy male and female F344/N rats were exposed for 6h/day, 5 days/week, for 20 months to filtered air (control) or one of three tested concentrations of ozone : the US EPA standard at the time of the study (0.12 ppm), the maximum concentration considered to be compatible with long term survival (1.0 ppm) and an intermediate level (0.5 ppm). The NTP carcinogenicity study does correspond to the OECD 452 guideline with minor exceptions (e.g. organ weights, ophthalmology, haematology, and urinalysis). Animals were observed for clinical signs twice daily and recorded monthly until week 92 and every two weeks until the end of the study. A complete histopathology was performed on all rats. Body weight development was recorded throughout the study. A significantly lower body weight was reported for the high dose group (1.0 ppm ozone) when compared to control. Based on these observations a systemic NOAEC of 0.5 ppm can be established (based on body weight changes). At 0.5 and 1 ppm, histopathological examinations revealed only local effects in lung, larynx and nose. Consequently, 0.12 ppm is considered the local NOAEC.

Several pulmonary endpoints have been examined in detail in separate publications as part of the HEI study, including investigations of lung biochemical constituents, structural and cellular changes, lung function, and nasal structure and function. The authors reported that repeated exposures to ozone for 20 month in the range of 0.12 to 1.0 ppm resulted in an initial acute response, evidenced by increased protein and albumin content and increased neutrophil influx in bronchioalveolar fluid, which subsided when exposure continued (e.g. as shown in studies from Pinkerton et al.; Harkema et al. 1987, 1993, 1994, 1997, 1999; Tesfaigzi et al. 1998). Continued exposure resulted in the induction of structural changes in the nasal cavity and the centriacinar region (collagen formation and bronchiolisation). The structural changes present after 2 months of ozone exposure were reported to be similar to those observed after 20 months of exposure. These structural changes had minimal to no measurable effect on overall pulmonary function, so it was suggested that these changes might be a protective adaptive response, e.g. to build tolerance to the injurious effects of ozone. The results of these studies support the conclusion of a chronic (local) NOAEC of 0.12 ppm, and a LOAEC of 0.5 ppm for effects on the respiratory tract.

Under CLP, classification for specific target organ toxicity Category 1 (STOT RE 1) is applicable if significant toxic effects from a 90-day repeated-dose study conducted in experimental animals were shown to occur at or below the guidance values (here: C ≤ 50 ppm, 6h/day) as indicated in Table 3.9.2 of the CLP regulation. For classification of ozone as STOT RE 1, the adverse effects of ozone need to be assessed to determine whether they are (1) significant and relevant for human health or (2) “adaptive responses” that are not considered toxicologically relevant.

Adaptive response (e.g. tolerance) of ozone has been noted in laboratory animals. In an early study, as citied in Patty (2012 in section 7.12), it was noted that pretreating animals to ozone levels below 1,000 ppb protected them from subsequent exposures to lethal concentrations (above 3,000 ppb). Pinkerton et al. (1989) reported that structural changes (epithelium thickening and bronchiolarization of the alveolar duct in the centriacinar region) present after 2 months of ozone exposure were similar to those observed after 20 months of exposure. Accordingly, the author suggested that the early responses to ozone may represent changes that protect the rat lungs during continued exposure to ozone. However, several studies suggested that tolerance to the irritant effects and the subsequent diminution of lung function may not be beneficial.

Also in monkeys, chronic exposures to ozone manifested bronchiolitis, altered epithelial cell proliferation, nasal secretory hyperplasia and other effects, including focal lung lesions, which persisted after the cessation of exposure (Eustis et al., 1981; Harkema et al., 1987; Tyler et al., 1988). Other lines of evidence support the concept that repeated ozone exposure may result in chronic lung disease (Patty, 2012). Ozone inactivates lysozyme, an antimicrobial protein secreted by airway cells, and human a1-antitrypsin, a protease inhibitor that protects the lung from emphysema. It also increases the synthesis, deposition, and degradation of collagen in rat lung (van Bree et al., 2001; Last et al., 1994). With respect to the significance of the toxicological effect, it has to be emphasized that available animal studies are mostly conducted at low concentrations. Consequently, there is only limited data for exposure above 1 ppm. Based on the available data it is questionable that exposure to ozone at concentrations of 1 ppm and above leads to an adaptive (beneficial) response.

In human studies (discussed in more details in Section 7.10 Exposure related observations in humans), acute exposure to as little as 80 ppb ozone can induce neutrophilic inflammation, peaking in bronchoalveolar lavage fluid or biopsies of the bronchial mucosa 12–18 h after a single exposure (Adam et al., 2002). These data indicate that the amount of acute inflammatory damage from ozone repeatedly demonstrated in animals is likely to be duplicated in the human lung at concentrations that are lower than those used in animal studies. In addition, chronic pathological effects in lung specimens from accident victims in southern California were discussed by Patty (2012). More than 25% of the tissues examined had severe and extensive injury to the centriacinar region (with monocytic infiltrates).

The possible mode of action of ozone is the interaction with biological macromolecules. It is unlikely that ozone as such can penetrate beyond the layer of fluid covering the cells of the lung. Instead, ozone lesions are propagated by a cascade of secondary reaction products, such as aldehydes and hydroxyperoxides produced by ozonolysis of fatty acids and other substrates in the lung lining fluid, and by reactive oxygen species arising from free radical reactions. At low concentrations, this interaction may be reversible/adaptive. At higher concentrations it might be possible that ozone is depleting scavenger molecules or repair mechanism and subsequently react with structure on the cell surfaces. This correlation is also discussed by Patty (2012):

“At concentrations below 200 ppb, most, if not all, ozone is likely to react with the biological macromolecules in the respiratory lining fluid. Ozone is a powerful oxidant and will react with amino acids (particularly cysteine, tryptophan, methionine, phenylalanine, and tyrosine) and with lipids (particularly the unsaturated fatty acids contained in membrane phospholipids). The former can yield disulfides and methionine sulfoxide; the latter can yield hydrogen peroxide, aldehydes, and hydroxyhydroperoxides. Antioxidants in mucus and other fluids lining the respiratory tract, as well as those in the tissues themselves, may be protective. In the past, ozone has been purported to act as a free radical. However, although clearly a strong oxidant, ozone is not a free radical. In addition, supportive evidence for this mechanism at best is only indirect and comes from studies showing that vitamin E, a free radical scavenger, retards or prevents ozone’s effects on polyunsaturated fatty acids in vitro. In addition, vitamin E deficiency in experimental animals enhances ozone’s toxicity. It is not known, however, whether supplemental vitamin E in the diet can protect humans against ozone’s effects.”

While effects might be “adaptive” changes al low concentrations (< 1 ppm), slightly higher concentrations of ozone (3.2–12 ppm) can already be fatal and can produce pulmonary oedema and hemorrhage in experimental animals (Mittler et al., 1956, 1957; Svirbely and Saltzman, 1957; Diggle and Gage, 1995). Even though the margin of effects between adaptive changes and acute toxicity is small, it seems appropriate to classify ozone as STOT RE 1 to acknowledge chronic effects to the respiratory system at sub-lethal concentrations.