Long-Term Pulmonary Responses of Three Laboratory Rodent Species to Subchronic Inhalation of Pigmentary Titanium Dioxide Particles

Edilberto Bermudez*,1, James B. Mangum*, Bahman Asgharian*, Brian A. Wong*, Edward E. Reverdy*,2, Derek B. Janszen*,3, Paul M. Hext{dagger}, David B. Warheit{ddagger} and Jeffrey I. Everitt*

* CIIT Centers for Health Research, P.O. Box 12137, 6 Davis Drive, Research Triangle Park, North Carolina 27709-2137; {dagger} Syngenta, Alderley Park, Macclesfield, Cheshire, SK104TJ, United Kingdom; {ddagger} DuPont Haskell Laboratories, Newark, Delaware 19714

Received May 14, 2002; accepted July 29, 2002


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Female mice, rats, and hamsters were exposed to 10, 50, or 250 mg/m3 pigmentary titanium dioxide (p-TiO2) particles for 6 h per day and 5 days per week for 13 weeks with recovery groups held for an additional 4, 13, 26, or 52 weeks postexposure (46 weeks for the p-TiO2-exposed hamsters). At each time point p-TiO2 burdens in the lung and lymph nodes and selected lung responses were examined. The responses studied were chosen to assess a variety of pulmonary parameters, including inflammation, cytotoxicity, lung cell proliferation, and histopathologic alterations. Burdens of p-TiO2 in the lungs and in the lung-associated lymph nodes increased in a concentration-dependent manner. Retained lung burdens following exposure were greatest in mice. Rats and hamsters had similar lung burdens immediately postexposure when assessed as milligrams of p-TiO2 per gram of dried lung. Particle retention data suggested that pulmonary overload was achieved in both rats and mice at the exposure levels of 50 and 250 mg/m3. Under the conditions of the present study, hamsters were better able to clear p-TiO2 particles than were similarly exposed mice and rats. Pulmonary histopathology revealed both species and concentration-dependent differences in p-TiO2 particle retention patterns. Inflammation was noted in all three species at 50 and 250 mg/m3, as evidenced by increases in macrophage and neutrophil numbers and in soluble indices of inflammation in bronchoalveolar lavage fluid (BALF; rats > mice, hamsters). In mice and rats, the BALF inflammatory responses remained elevated relative to controls throughout the entire postexposure recovery period in the most highly exposed animals. In comparison, inflammation in hamsters eventually disappeared, even at the highest exposure dose, due to the more rapid clearance of particles from the lung. Pulmonary lesions were most severe in rats, where progressive epithelial- and fibroproliferative changes were observed in the 250 mg/m3 group. These epithelial proliferative changes were also manifested in rats as an increase in alveolar epithelial cell labeling in cell proliferation studies. Associated with these foci of epithelial proliferation were interstitial particle accumulation and alveolar septal fibrosis. In summary, there were significant species differences in pulmonary responses to inhaled p-TiO2 particles. Under conditions in which the lung p-TiO2 burdens were similar and likely to induce pulmonary overload, rats developed a more severe and persistent pulmonary inflammatory response than either mice or hamsters. Rats also were unique in the development of progressive fibroproliferative lesions and alveolar epithelial metaplasia in response to 90 days of exposure to a high concentration of p-TiO2 particles.

Key Words: pigmentary titanium dioxide; lung response; rats; mice; hamsters; inhalation; bronchoalveolar lavage fluid; pulmonary lesions.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Titanium is an abundant element, 95% of which is used as titanium dioxide (TiO2), and has important commercial utility as white pigments in the manufacture of paints and other applications (Bingham et al., 2001Go). Two crystalline forms of TiO2 are produced, anatase and rutile, both with low solubility in water and primary particle diameters of less than 1 µm. TiO2 is virtually inert and has generally been classified as a nuisance dust and grouped with other particles of low toxicity and solubility as poorly soluble particulates (PSP).

The pulmonary toxicological effects of chronic exposure by inhalation to TiO2 particles have been examined in rodents. Continuous exposure for 2 years to 250 mg/m3 pigmentary TiO2 (p-TiO2) led to the development of lung tumors in male and female rats with a greater incidence found in females (Lee et al., 1985Go). Exposure to other PSP, such as diesel exhaust, carbon black, and talc particles, has resulted in pulmonary tumors in rats but not in mice or hamsters following chronic exposure (Heinrich et al., 1986Go; Hext, 1994Go; Mauderly et al., 1987Go, 1994Go; Muhle et al., 1998Go). Findings of fibrosis, bronchoalveolar hyperplasia, metaplasia, and pulmonary tumorigenesis following exposures to high doses of PSP for long periods of time appear to be unique to the rat relative to other rodent species. These PSP-induced effects occur under, and appear limited to, conditions of substantial particle lung burdens and corresponding impairment of alveolar macrophage-mediated lung clearance (pulmonary overload).

Subchronic inhalation toxicity studies of rats to PSP have demonstrated that at concentrations under which chronic exposure conditions would lead to overload, there is a persistent inflammatory response and an attendant impairment of particle clearance from the lung (Cullen et al., 2000Go; Warheit et al., 1997Go). Although particle overload conditions can be achieved in rats as well as mice, the specific pathway leading to tumor development only in rats remains unclear. The nonneoplastic lung responses induced in rodents by exposure to PSP have recently been reviewed, and the experimental evidence supports the hypothesis that there is a link between chronic inflammation and the epithelial changes leading to pulmonary cancer in rats (Donaldson, 2000Go). It is believed that the rat is, in some way, hyper responsive in terms of its tendency to mount an inflammatory response to particles.

There have been several recent reviews concerning the topic of the species-specificity of nonneoplastic lung responses to PSP (Donaldson, 2000Go; Levy, 1994Go; Mauderly et al., 1994Go; Snipes, 1996Go). Despite the importance of the topic, there have been relatively few comparative studies of PSP that have employed multiple species in the same laboratory utilizing an identical exposure regimen; moreover, to date there have been no systematic species comparison studies with p-TiO2 particles. The objective of the present study was to test the hypothesis that the rat lung responds to high particulate loading in a manner that differs from that of mice and hamsters under identical exposure conditions to PSP. Selected pulmonary pathobiological responses of female rats, mice, and hamsters were compared following subchronic inhalation exposure to comparable concentrations of p-TiO2 particles.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals.
Six-week-old female B3C3F1/CrlBR mice, CDF (F344)/CrlBR rats, and Lak: LVG (SYR) BR hamsters (Charles River Breeding Laboratories, Wilmington, MA) free of parasites, mycoplasma, bacterial, and viral pathogens were used in these studies. Animals were housed in an AAALAC-accredited facility in 1-m3 H-1000 stainless-steel and glass inhalation chambers in suspended steel wire caging and were acclimated to this housing arrangement before beginning the exposures. Mice and rats were housed in similar stainless-steel suspended caging during the postexposure period. Hamsters were housed in filtered, microisolated, polycarbonate cages on direct-contact cellulose bedding during the postexposure period. Animals were supplied NIH07 cereal-based diet and water ad libitum. Animals were uniquely identified by implanted microchip transponders (Biomedic, Inc., NJ) in the subcutis and were randomly distributed to exposure groups using a computer-generated randomization algorithm. Room temperature was maintained at 64–79°F and humidity at 40–60% throughout the exposure and postexposure periods. All animals were acclimated for approximately 9 days prior to exposure. Body weights for each animal were recorded prior to exposure, weekly for the first 17 weeks, and biweekly thereafter.

Aerosol generation and monitoring.
P-TiO2 (rutile type) was obtained from the DuPont Company (Wilmington, DE). A single modified-rotating brush dust-feeding mechanism served as the generator. Inhalation exposures were conducted in 1-m3 H-1000 stainless-steel chambers (Lab Products, Maywood, NJ). During the exposures, particle concentrations were continuously monitored using light scatter (Model RAM-S, Monitoring Instruments for the Environment, Inc., Bedford, MA), and the time-averaged concentration was recorded at least six times over the 6-h exposure period. The particle size distribution of the aerosol was measured at least twice per exposure level (excluding the control chambers) during the course of the study. Measurements were made using a MOUDI impactor (micro-orifice uniform deposit impactor, model 100, MSP Corporation, Minneapolis, MN). Mean mass-median aerodynamic diameter was 1.40 µm (see Results).

Experimental design.
Animals were exposed to 10, 50, or 250 mg/m3 p-TiO2 for 6 h/day, 5 days/week, for 13 weeks. Controls were exposed to filtered air only. Hamsters were exposed separate from the mice and rats due to health consideration. Group size for rats was 65 animals and for mice and hamsters 73 animals. The difference in group size was to account for expected animal losses in the course of the study. Animals were sacrificed immediately following completion of the 13-week exposure and additional recovery groups were held for postexposure periods of 4, 13, 26, or 52 (46 for hamster) weeks in clean air. Following exposure and at each recovery time, the p-TiO2 burdens in the combined right lung lobes and the lung-associated lymph nodes were determined. The left lungs were used to assess lung cell proliferation and histopathology. A separate group of animals was used at each time point to assess the inflammatory response in bronchoalveolar lavage fluid (BALF) by assaying lactate dehydrogenase (LDH) and total protein levels and conducting cytologic characterization of BALF cells.

Titanium burden analysis.
Burden analysis was performed as described by Levine and colleagues (Levine et al., submitted manuscript). At necropsy, right lung and lymph node tissue samples were collected and frozen until ready for separate titanium burden analysis. Prior to burden analysis the tissues were thawed, weighed, dried overnight in a muffle furnace at 37°C, dessicated, and weighed again. Dry tissue samples were digested overnight in a mixture of nitric acid and hydrofluoric acid and then further digested in a microwave oven. The sample digests were diluted with deionized water to 25 ml, and samples were analyzed using an inductively coupled plasma optical emission spectrometer (ICP-OES; Thermo Jarrell Ash AtomScan 16,Thermo Elemental, Franklin, MA). Data for lung tissue are expressed as milligrams of p-TiO2 per gram dry lung. For lymph node tissues, the data are expressed as the total micrograms of p-TiO2 present because the harvest of lymph node tissues was subject to collection of extraneous tissue to assure collection of the nodes. The method quantitation limit (MQL) was 0.05 µg Ti/ml of sample digest. The minimum detectable concentrations (MDC) of p-TiO2 in pulmonary tissues were calculated using the MQL in combination with the average control lung weight, and they were calculated to be for lung 0.22, 0.03, and 0.03 mg p-TiO2/g dry weight for mouse, rat, and hamster, respectively, and for lymph nodes 2.09 µg p-TiO2/sample. Occasional values above the MDC were observed in control tissues (e.g., burden values for control hamsters at the end of the exposures), and the values reflect an increase in the baseline for that run of the assay rather than the presence of p-TiO2 in control tissues. Retention half-times for p-TiO2 in lung were calculated using the best-fit equations to the lung burden data at all time points.

Bronchoalveolar lavage.
Lungs were lavaged five times with equal volumes of phosphate-buffered saline. Fluid from the first two lavages was recovered, pooled, and placed on ice. The subsequent three lavages were pooled and also placed on ice. Recovered cells from all lavages were collected by centrifugation, resuspended in cell culture medium, and counted using an automated cell counter (Model ZM, Coulter, Marietta, GA). Cell differential counts were performed on Wright-Giemsa-stained cytocentrifuge slide preparations. LDH and total protein levels in cell-free fluid from the first two pooled lavages were quantitated spectrophotometrically using a COBAS FARA II automated analyzer (Roche Diagnostic Systems Inc., Montclair, NJ).

Lung cell proliferation.
Five days prior to euthanasia, animals were subcutaneously implanted with osmotic pumps (Alza, Palo Alto, CA) containing bromodeoxyuridine (BrdU; Sigma Chemical Co., St. Louis, MO). Rats and hamsters were implanted with model 2ML1 (10 µl/h) pumps containing 5 mg/ml BrdU. Mice were implanted with model 2001 (1 µl/h) pumps containing 16 mg/ml BrdU. At necropsy, left lungs were pressure-infused intratracheally (20 cm H2O for mice; 30 cm H2O for rats and hamsters) with 10% neutral-buffered formalin. Lungs were fixed for approximately 48 h and then changed to 70% ethanol. Subsequently, the lungs were embedded in paraffin, sectioned at 5 µm, and stained for BrdU by established methods (Rutten et al., 1994Go). Terminal bronchiolar and alveolar cell labeling indices were determined for each animal, and the mean labeling index was calculated for each group of five animals.

Histopathology.
Paraffin-embedded left lung tissues were sectioned at 5 µm and stained with Masson’s trichrome. The trichrome-stained lung sections were evaluated for particle-induced histopathologic changes.

Statistical methods.
All data were tested for normality and homogeneity of variance. If the hypotheses for these assumptions were rejected (p < 0.01), common transformations (e.g., log, square root, arc sine) were applied and the data retested. Comparisons to controls were made using Dunnett’s test. The software package JMP (SAS Institute, Cary, NC) was used for the statistical analysis.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chamber Concentration
Target chamber concentrations of p-TiO2 aerosol were 10, 50, and 250 mg/m3 and the actual particle concentration in each chamber was monitored using a Real Aerosol Monitor (RAM). Mean, (± SD), particle concentrations over the exposure period were as follows: for mice 9.5 ± 1.2, 47.0 ± 4.6, and 240.3 ± 20.0; for rats 9.6 ± 1.1, 47.7 ± 5.1, and 239.1 ± 19.3; and for hamsters 9.9 ± 1.0, 49.7 ± 4.0, and 251.1 ± 17.3 mg/m3. The particle size characteristics for the 13-week exposure periods for each species are presented in Table 1Go.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Particle Size Analysis of Pigmentary TiO2 Aerosols HamsterMass median aerodynamic diameter0.00Geometric SD0.00
 
Animal Mortality and Clinical Observations
No significant losses of animals occurred during the exposure phase of the study. In the postexposure phase, there were unscheduled losses of rats (15) and mice (8) that generally occurred at least 10 months into the study and were distributed over the various treatment groups. The hamsters had greater morbidity and mortality than did mice and rats, presumably due to the occurrence of age-related spontaneous conditions such as chronic renal disease. A number of hamsters that died or were removed unscheduled from the study had bilateral, pale, small, firm kidneys with irregular pitted cortical surfaces. A few of these animals had evidence of ascitic fluid and subcutaneous edema suggestive of nephrotic syndrome. Many of the hamsters scheduled for sacrifice at the 46-week time point also had evidence of severe chronic renal disease at necropsy (bilateral, pale, firm, shrunken kidneys). These findings contributed to the decision to move up the terminal sacrifice so those hamsters under study would not be unduly compromised by potential renal dysfunction leading to pulmonary inflammatory changes from uremia.

Body Weights
Body weights were collected weekly for the exposure phase of the study and biweekly from week 17 through the end of the postexposure period. A depression in body weight was noted in all species and in all groups following the end of the exposure period (4–5% in mice, 2–3% in rats, and 5–11% in hamsters) with recovery occurring over the next three to four weeks in mice and rats (data not shown). This recovery was slower in hamsters, occurring within approximately six weeks. A likely reason for the depression in body weight in the three species is the need for a period of acclimation to changes in environmental conditions experienced by the animals during the postexposure recovery period when they were no longer housed in inhalation exposure chambers and were moved to new housing.

Mice exposed to the high concentration of p-TiO2 had a consistent depression, not exceeding 10% of controls, in body weight during the recovery period (data not shown). Although the cause for this depression in body weight is unclear, it is probably reflective of the high retained-p-TiO2 lung burdens found in these animals. In contrast, rats exposed to the high concentration of p-TiO2 demonstrated a consistent pattern of elevated weights (less than 10%), compared to controls, during the latter half of the recovery period (data not shown). No obvious cause for this weight elevation was evident.

Changes in hamster body weights were more frequent and abrupt than in mice or rats (data not shown). Weight loss during the exposure was attributed to housing conditions and was reversed following changes in housing from suspended-wire to solid-bottom shoebox caging during the postexposure periods. The observed drop in body weights near the end of the study probably reflected age-related changes as well as clinical symptoms consistent with kidney disease, as noted above.

P-TiO2 Burdens
There were dose-related changes in p-TiO2 lung burdens for all three species. Following 13 weeks of exposure, mice exhibited the greatest (170 mg/g dry lung) p-TiO2 lung burdens followed by rats (120 mg/g dry lung) then hamsters (114 mg/g dry lung; Fig. 1Go). Particle burdens decreased in the lung with time postexposure in mice, rats, and hamsters (Fig. 1Go). Burdens in the lymph nodes increased with time postexposure in rats of all dose groups, in mice of the low and mid-dose groups, and in hamsters of the high-dose group (Fig. 1Go). Notably, the lymph node p-TiO2 burdens of the animals exposed to 250 mg/m3 demonstrated the greatest increase between 4 and 13 weeks after the exposure ended. Decreases in lung burdens with time after exposure were sharply different between exposure concentration groups in both rats and mice; high- and mid-dose burdens decreased slowly to approximately 75% of the initial burden, whereas low-dose burdens decreased to approximately 15%. There were very significant differences between species in the pulmonary clearance kinetics of p-TiO2; rats and mice of the high-dose group retained more (approximately 75%) of the initial burden after 52 weeks of recovery than did hamsters after 46 weeks of recovery (approximately 10%).



View larger version (38K):
[in this window]
[in a new window]
 
FIG. 1. Retained lung (A) and lung-associated lymph node (B) pigmentary TiO2 burdens in rodents exposed for 13 weeks and held unexposed for up to 52 weeks. *Statistically significant differences from controls. (Dunnett’s, p < 0.05). Diamond, mouse; square, rat; triangle, hamster.

 
Cytology
The number and types of cells recovered by lavage of the lungs was used to indicate the extent of pulmonary inflammation. The number of cells recovered by lavage from control animals was greatest in hamsters (2.8 x 106 ± 0.8) followed by rats (1.4 x 106 ± 0.4) then mice (1.1 x 105 ± 0.3) and was consistent with our previous studies with these species. Significant increases in the total number of cells recovered (number ± one SD) were observed at the end of exposures in rats (31.8 x 106 ± 10.0), mice (20.2 x 106 ± 11.6), and hamsters (19.0 x 106 ± 4.4) of the high-dose group and in rats of the mid-dose group (4.5 x 106 ± 1.2). These elevations in the number of cells recovered from p-TiO2-exposed animals declined with time after exposure in all species, but only in hamsters did the values return to control levels before the termination of the study (data not shown).

The number of recovered macrophages in rats, mice, and hamsters of the high-dose group was significantly elevated, relative to corresponding controls, following exposure. The absolute increase in this cell population was greatest in mice and approximately equivalent in rats and hamsters (Fig. 2Go). Postexposure recovery resulted in a decline in the number of macrophages recovered by lavage in all species; although these values remained significantly elevated over concurrent controls in mice of the mid- and high-dose groups and rats of the high-dose group at 52 weeks postexposure, hamsters had returned to control levels by 26 weeks postexposure (Fig. 2Go).



View larger version (42K):
[in this window]
[in a new window]
 
FIG. 2. Cytology of cells recovered by lung lavage. The data represent the number of macrophages (line graph) or the percentage of cells counted (200 per slide) identified as neutrophils (bar graph). Each data point represents the mean of five animals. Cross symbol indicates statistically significant differences (Dunnett’s, p < 0.05) from concurrent controls for macrophage numbers. *Statistically significant differences (Dunnett’s, p < 0.05) from concurrent controls for percentage neutrophils.

 
Neutrophils comprise a small proportion of the cells recovered by BAL from control mice and rats (approximately 0.3%) and in hamsters (approximately 3%). Exposure to p-TiO2 resulted in greatly increased proportions of neutrophils in all three species. The greatest response was measured in animals of the high-dose group (rats, 83%; hamsters, 58%; and mice, 32%) at the end of the exposure period (Fig. 2Go). With increasing time after exposure there was a diminution in the number of neutrophils in rats and mice of the mid- and high-dose group, although the number of neutrophils remained significantly elevated over controls at the 52-week postexposure time point (data not shown). In contrast, hamster neutrophil levels had returned to control values by 26 weeks postexposure.

Lymphocytes were minimal in control rats and mice (0.2 and 0.3% respectively), whereas these cells comprised from 0.2 to 2.0% of the recovered cell population in control hamsters (data not shown). Animals exposed to p-TiO2 (mid- and high-dose rats and mice and high-dose hamsters) demonstrated significantly elevated levels of lymphocytes, which remained so for rats (2.2%) and mice (1.4 and 1.0%) at the end of the 52-week postexposure period but had returned to control levels in hamsters by 46 weeks postexposure (data not shown).

Pulmonary Toxicity End Points
Bronchoalveolar lavage was performed on one group of five animals for each time point and dose group. Pooled fluids (BALF) from the first and second lavages were assayed for LDH and total protein as markers of lung injury.

LDH levels were significantly elevated in mouse BALF in the mid- and high-dose animals and remained so through 26 weeks postexposure for the mid-dose animals and 52 weeks postexposure for the high-dose animals. Similarly, rats in the high- and mid-dose groups had persistent elevation of LDH in BALF; however those in the mid-dose group had returned to control levels by 26 weeks postexposure. Although hamsters exposed to the high dose of p-TiO2 had elevated LDH levels, the elevations were significantly different from control only through four weeks postexposure (Fig. 3Go).



View larger version (31K):
[in this window]
[in a new window]
 
FIG. 3. Lactate dehydrogenase concentrations in bronchoalveolar lavage fluid. The data are expressed as the percentage of concurrent controls with a group size of five animals. *Statistically significant differences (Dunnett’s, p < 0.05) from concurrent controls.

 
Total protein concentrations in mouse BALF, collected from animals of the mid- and high-dose groups, were significantly greater than concurrent controls after exposure and remained elevated through 52 weeks postexposure. Similarly, rats had significantly elevated, persistent levels (to 52 weeks postexposure) of total protein in BALF but these were limited to the high-dose group. Elevations of total protein were also observed in hamsters but were limited to the high-dose group and had returned to control values by 26 weeks postexposure (Fig. 4Go).



View larger version (33K):
[in this window]
[in a new window]
 
FIG. 4. Total protein concentrations in bronchoalveolar lavage fluid. The data are expressed as the percentage of concurrent controls with a group size of five animals. *Statistically significant differences (Dunnett’s, p < 0.05) from concurrent controls.

 
Lung Cell Replication
Lung epithelial cell proliferation was quantified by immunocytochemical detection of BrdU incorporated by cells in the S-phase of the cell cycle.

Increased terminal bronchiolar cell replication was measured in mice and rats of the high-dose group following the 13-week exposure interval; however, the labeling index of these cells in both species had returned to control values by four weeks postexposure. No significant increases in the labeling index of bronchiolar cells of the hamster were observed (Fig. 5Go).



View larger version (34K):
[in this window]
[in a new window]
 
FIG. 5. Cell turnover at bronchiolar sites in the left lung of control and p-TiO2-exposed rodents. Portions of the lung were fixed, embedded, sectioned, and immunostained for the presence of BrdU. Data are expressed as the labeling index. *Statistically significant (Dunnett’s, p < 0.05) differences from concurrent controls.

 
Alveolar cells of the mouse and hamster lung had no significant increases in cell replication, whereas increased labeling was noted in this cell population in the rat in the high-dose group through 52 weeks postexposure (Fig. 6Go).



View larger version (39K):
[in this window]
[in a new window]
 
FIG. 6. Cell turnover at alveolar sites in the left lung of control and p-TiO2-exposed rodents. Portions of the lung were fixed, embedded, sectioned, and immunostained for the presence of BrdU. Data are expressed as the labeling index. *Statistically significant (Dunnett’s, p < 0.05) differences from concurrent controls.

 
Baseline values for alveolar cells of the hamsters increased at the latter time points and may be indicative of the age or health status of these animals. Similar increases in bronchiolar labeling indices were observed in some hamsters of the control and high-dose groups at 46 weeks postexposure resulting in elevated mean values and statistically (but not biologically) significant decreases in the low and mid-dose groups (Fig. 6Go).

With the counting scheme used, no attempt was made to quantify cell replication in focal lesions associated with particle and particle-laden macrophage accumulation; however, it was apparent that labeled cells were often associated with these sites.

Lung Histopathology
The microscopic pattern of particle retention and lesions varied by species, by exposure concentration level, and with time postexposure. A characteristic finding in all three species was the intraluminal aggregation of particle-laden macrophages filled with opaque particles of p-TiO2. In each of the three concentration groups, the majority of the particles were found in macrophages within alveolar and alveolar duct lumens, although some of the intraluminal particulate material was extracellular. Extracellular particles were most prominent immediately following the exposure and were by far most numerous in the high-concentration group. Over time postexposure, aggregations of particle-laden macrophages coalesced and extracellular particulate material diminished. In all three species, the low concentration exposure groups had only a minimal diffuse increase in alveolar macrophages, some of which contained p-TiO2 particles. No associated pathology was present in these animals.

Rats.
The mid- and high-dose rats had a variety of lung lesions associated with retained p-TiO2 particles. Particles and particle-laden alveolar macrophages were most numerous in centriacinar lung regions, although in the high-dose animals there was a more diffuse panacinar distribution at the end of the 13-week exposure period. While the nature of the epithelial lesions was similar in mid- and high-dose animals, there were significant differences in lesion severity. The number of particle-laden macrophages observed individually and in aggregates was much greater in the high-dose animals. Although most particles were retained intraluminally, a minimal to mild interstitial accumulation of particle-laden macrophages was present in mid- and high-dose rats and remained over the course of the 52-week recovery. Immediately postexposure, both mid- and high-dose rats had alveolar hypertrophy and hyperplasia of type II epithelial cells surrounding aggregations of particle-laden macrophages (Fig. 7BGo). These lesions were generally minimal to mild in mid-dose and mild to moderate in high-dose animals. In both of these exposure concentration groups at this time point, histological evidence of chronic active inflammation was noted by the infiltration of neutrophils. Histological recognition of neutrophil infiltrates diminished in mid-dose rats by four weeks recovery although they persisted throughout the 52-week recovery period in the high-dose rats.



View larger version (126K):
[in this window]
[in a new window]
 
FIG. 7. Histopathology of the lungs of mice, rats, and hamsters immediately following exposure to p-TiO2 for 13 weeks. (A) Lung tissue from control rat following 13-week exposure to air in control chamber. (B) High-dose rat lung at the completion of the 13-week p-TiO2 exposure period. Metaplastic alveolar epithelium surrounds aggregates of pigment-laden alveolar macrophages. Inset: High magnification depicts the cuboidal character of metaplastic alveolar epithelial cells. (C) Control mouse lung following 13-week exposure to air in control chamber. (D) High-dose mouse lung following 13-week p-TiO2 exposure period. Note the numerous pigment-laden macrophages and the lack of pulmonary parenchymal interstitial reaction. Inset: Free pigment particles and macrophages within the alveolar space surrounded by relatively normal flattened alveolar epithelium. (E) Lung tissue from control hamster after 13-week exposure to air in control chamber. (F) High-dose hamster lung following the completion of the 13-week p-TiO2 exposure period. Note the numerous pigment-laden macrophages and the lack of pulmonary parenchymal interstitial reaction. Inset: Free pigment particles and macrophages within the alveolar space. All photomicrographs Masson’s trichrome stain, original magnification x75 (insets, x300).

 
Over increasing time postexposure, there was significant coalescence of the aggregates of particle-laden macrophages. By 52 weeks postexposure, mid-dose rats developed small intraluminal lesions of tightly aggregated, heavily particle-laden macrophages surrounded by minimal alveolar epithelial type II cell hypertrophy and alveolar metaplasia (bronchiolization). These lesions differed from those of the high-dose group in that they lacked significant alveolar septal fibrosis. In many instances at the later time points, lesions in mid-dose animals had intraluminal aggregations of particle-laden macrophages with relatively little epithelial reaction present. High-dose animals developed more severe alveolar type II cell hypertrophy and hyperplasia and alveolar metaplasia that were associated with septal fibrosis and interstitialization of particles, often within macrophages (Fig. 8BGo). These lesions were noted by four weeks postexposure, were prominent at the 26-week time point, and progressed through the 52-week recovery time point. In some instances, the alveolar lumens in lesion areas were characterized by lipoproteinosis and cholesterol cleft development (Fig. 8BGo). Lesions in both mid- and high-dose animals were scattered throughout the lung lobes and were commonly observed in subpleural regions. Airway lesions in rats consisted of minimal bronchiolar hypertrophy in mid-dose animals that did not progress over time of recovery and mild to moderate bronchiolar hypertrophy and hyperplasia in high-dose animals with progressive bronchiolization in areas with lesions.



View larger version (121K):
[in this window]
[in a new window]
 
FIG. 8. Histopathology in the lungs of mice, rats, and hamsters exposed to p-TiO2 for 13 weeks and allowed to recover for up to 52 weeks. (A). Lung tissue from air exposed control rat after 52-week recovery period. (B). High-dose rat after 52-week postexposure recovery period. Note the collagen (blue) within lesions of metaplastic proliferative epithelium. Inset: A fibroproliferative lesion showing cholesterol cleft (arrow) and interstitial collagen (blue). (C). Lung tissue from air exposed control mouse following 52-week recovery period. (D). High-dose mouse at the end of the 52-week postexposure recovery period. Note the lack of fibroproliferative changes versus the rat lesion in (B). Inset: High magnification of lesion showing normal thickness alveolar septae. (E) Lung tissue from air exposed control hamster after 46-week recovery period. (F) High-dose hamster at the completion of the 46-week postexposure recovery period. Note the coalescence of the aggregates of heavily particle-laden macrophages versus that noted in Figure 7FGo. Inset: High magnification of lesion showing normal thickness alveolar septae. All photomicrographs Masson’s trichrome stain, original magnification x75 (insets, x300).

 
Mice.
Mice in the mid- and high-dose groups had particle retention patterns that differed from those of rats and had a greater number of particles and particle-laden macrophages in the lungs at all postexposure time points. In certain respects, the particle retention patterns in mice were similar to those in rats in that particles and particle-laden macrophages were primarily intraluminal, but the pattern of accumulation differed in that it was more diffuse corresponding to a panacinar distribution (Fig. 9Go). Numerous particles and particle-laden macrophages were present in pleural and subpleural regions of mid-dose and especially high-dose animals immediately postexposure, but over time there was a marked preferential loss of particulate material from the periphery of the lung lobes during the postexposure recovery period (Fig. 9Go). In general, mice had less aggregation of particle-laden macrophage than did either rats or hamsters.



View larger version (178K):
[in this window]
[in a new window]
 
FIG. 9. Retained p-TiO2 particles in mouse and rat lungs. (A) High-dose mouse at the completion of the 13 week p-TiO2 exposure period. Note the panlobar distribution of p-TiO2 particles. Inset: Heavy aggregation of opaque particulate in the pleural region is shown. (B) High-dose mouse after 52- week postexposure recovery period. As compared with the photomicrograph depicted in (A), there is substantial diminution of p-TiO2 particles. Inset: High magnification of particulate in bronchus-associated lymphoid tissue is shown. (C) Mid-dose mouse at the completion of the 13 week p-TiO2 exposure period. Note the prominent central lobar distribution of particles. (D) Mid-dose mouse at the end of the 52-week postexposure recovery period, for comparison of particle distribution with (C). (E) Mid-dose rat at the completion of the 13-week p-TiO2 exposure period. Note the focal, widely scattered areas of particle aggregation, primarily in centriacinar regions. (F) High-dose rat at the completion of the 13-week p-TiO2 exposure period. Note the diffuse panlobar distribution of p-TiO2, as compared with (E). Paraffin sections, original magnification x20 (insets, x150).

 
The most marked difference between rats and mice in the pathologic reaction of the lung was the lack of alveolar epithelial hyperplasia, metaplasia, and septal fibrosis in high-dose mice at the recovery time points (Fig. 8CGo). Particles and associated lesions, characterized by alveolar type II cell hypertrophy, were most prevalent in central lobar regions and virtually absent in the periphery of lung lobes. Microscopically there was increased peribronchiolar aggregation of lymphoid cells in the lungs of mice (Fig. 9BGo). This is an age-related finding in B6C3F1 mice as noted in the finding of minimal to mild infiltrates in three of five animals in the control exposure group at 52 weeks postexposure. P-TiO2 exposure increased the incidence and severity of this finding and appeared to decrease the latency of the lesion in the high-dose mice as it was noted at each time point postexposure.

Hamsters.
The particle retention pattern in hamsters reflected the rapid clearance of p-TiO2. Similar to findings in mice, particles and particle-laden macrophages in exposed hamsters appeared to be concentrated in central portions of the lung lobes, although there were occasional subpleural focal accumulations of heavily particle-laden macrophages, similar to those noted in rats. Minimal type II epithelial cell hypertrophy and hyperplasia were observed associated with free particles and particle-laden macrophages, in both mid- and high-dose animals. These effects were reduced in incidence and severity with increasing time postexposure, correlating with the rapid removal of p-TiO2 particles from the hamster lung. Alveolar metaplasia and fibroproliferative changes were not present in hamster lungs (Fig. 8FGo). Polymorphonuclear leukocytes, both neutrophils and eosinophils, were found in all high-dose hamster lungs at each postexposure time point with the exception of the 46-week final time point. The severity of inflammation subsided with increasing time postexposure and with the associated diminution of particle burden.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The aerosol concentrations of p-TiO2 particles used in the present study were chosen to achieve retained lung burdens that would span from minimal accumulation in alveolar macrophages to those that have been associated with impairment of macrophage-mediated lung clearance (so-called pulmonary overload) in long-term rat studies (Lee et al., 1985Go). Overload is generally characterized by a deposition of material into the lung that exceeds the ability of the alveolar macrophages to remove it (Mauderly and McCunney, 1996Go). In the studies presented here, as expected, the retained lung and lung-associated lymph node (LALN) burdens in all three species increased in a concentration dependent manner. Particle burdens in LALN significantly increased in rats at all exposure concentrations, in mice in the mid and high exposure groups, and in hamsters at the highest exposure concentration. Translocation of more particles to the LALN in rats, regardless of the particle lung burdens at the end of the exposures, suggests that rats may be more susceptible to the establishment of pulmonary overload than mice or hamsters.

The retained lung p-TiO2 particle burdens in this study indicated that mid- and high-dose rats and mice developed significant impairment of alveolar macrophage-mediated clearance and thus overloaded lungs. Animals in the mid- and high exposure concentration groups had calculated particle retention half-times of 417 and 621 days (mice) or 324 and 838 days (rats) respectively. This contrasts with retention half-times of approximately 50 days for mice and 100 days for rats in the low concentration group and less than 110 days for all hamsters. The prolongation of p-TiO2 particle retention in exposed rats in the present study is consistent with the retention half-times of 68, 110, and 330 days previously reported for rats exposed for four weeks by inhalation to aerosol concentrations of 5, 50, and 250 mg/m3 p-TiO2 particles, respectively (Warheit et al., 1997Go).

In addition to significant differences in total retained lung and LALN p-TiO2 burdens, all three rodent species had distinctive particle microdosimetry retention patterns. Despite these differences, the three species had primarily airspace aggregation of particles and particle-laden macrophages, as opposed to the interstitial patterns of particle retention that have been reported in people and in nonhuman primate models (Nikula et al., 2001Go). Although particle distribution has been proposed to account for marked differences in rat and human responses to inhaled particles (Nikula et al., 2001Go), the marked interspecies differences in pulmonary toxicity and pathology findings in the present rodent study cannot be explained by dose and anatomic site of particle distribution alone.

Although lung overload can be achieved in hamsters it requires prolonged exposure and higher lung burdens than are necessary to achieve this condition in rats (Muhle et al., 1990Go). The retained p-TiO2 lung particle burdens indicated that under the conditions of the present study, hamsters were better able to clear particles than were similarly exposed rats and mice. In part, the efficient lung clearance and corresponding decreased retention of p-TiO2 particles may explain the relative lack of long-term pulmonary toxicity and pathology noted in the heavily exposed hamsters. The lung burden findings in hamsters following the subchronic p-TiO2 exposure were somewhat surprising since we did not anticipate such dramatic differences between rats and hamsters. There have been relatively few clearance studies of PSP in Syrian golden hamsters, although a previous chronic study had demonstrated significant differences between the response of rats and Syrian golden hamsters to the inhalation of particles including p-TiO2 (Muhle et al., 1990Go). Creutzenberg and colleagues noted the stimulation of clearance of radiolabeled polystyrene tracer particles in female Syrian golden hamsters following PSP particle exposure and speculated that hamsters as burrowing animals might have a well-developed adaptive mechanism due to living in a dry, dusty environment (Creutzenberg et al., 1998Go).

The most striking findings of the present study are the interspecies differences noted in inflammation and lung lesions found in highly exposed animals. These experiments demonstrate that rats, mice, and Syrian golden hamsters vary significantly in their lung responses to inhaled p-TiO2 particles, confirming the generally held view that rats respond differently to PSP inhalation at overload exposure concentrations (Donaldson, 2000Go). The inflammatory changes, epithelial proliferative changes, and fibroproliferative lung lesions noted in rats in this study are identical to findings previously noted in subchronic (Warheit et al., 1997Go) and chronic (Lee et al., 1985Go) p-TiO2 studies that have used similar particles and exposure concentrations. Our findings differ from those of Baggs and colleagues (Baggs et al., 1997Go) who observed the regression of pulmonary lesions with time postexposure, whereas in the present study, animals with equivalent initial lung burdens did not exhibit fibroproliferative lung lesions. A major difference between these two studies is the gender of the rats used. Since the incidence of pulmonary tumors in female rats has been shown to be greater than in male rats (Lee et al., 1985Go), and gender differences in defense mechanisms (e.g., phagocytosis and neutrophil function) are known to exist (Spitzer, 1999Go) it is possible that this would result in the observed differences.

The lowest exposure concentration of p-TiO2 particles (10 mg/m3) employed in this study did not result in inflammation or lesions, confirming the low toxicity of this particle-type and giving credibility to the notion that these results can be extrapolated to PSPs in general. Clearly, particle size, physicochemical attributes, and airborne concentration are known to play a role in PSP-induced effects. Therefore other interspecies studies using additional particle compositions are warranted (Cullen et al., 2000Go).

Our results demonstrate that p-TiO2-induced epithelial and fibroproliferative lesions associated with pulmonary overload and high lung burdens in the rat are progressive even following cessation of particle exposure and the diminution of pulmonary inflammation. The time course of lesion development in the present study shows that alveolar cell metaplasia (bronchiolization) occurs prior to the development of severe interstitial fibrosis in the lesion regions in those animals with high lung burdens. Further, the present studies confirm that pulmonary overload occurs in mice but with less neutrophilic inflammation and without the histopathological lesions observed in rats. The epithelial metaplasia (bronchiolization), persistent alveolar cell hyperplasia, and septal fibrosis noted in p-TiO2 particle-overloaded rats was a concentration-dependent effect and unique to that species when compared to the mice and hamsters. These lesions appeared to be highly correlated with the severe neutrophilic inflammation that characterized the rat response to inhaled p-TiO2 particles at exposure concentrations that resulted in pulmonary overload. Pulmonary neutrophilia in the rat was much more severe and persistent than in mice or hamsters, and may in large part explain the genesis of the species-specific pulmonary parenchymal lesions that occur. These lesions are believed to be strongly associated with PSP-induced tumorigenesis occurring in chronic inhalation bioassays using rats (Bellmann et al., 1992Go; Heinrich et al., 1995Go; Mauderly et al., 1987Go, 1994Go; Muhle et al., 1990Go).

Utilizing multiple species exposed in the same experiment and incorporating a long-term postexposure component, we demonstrated significant species differences in pulmonary responses to inhaled p-TiO2 particles. Under conditions where the retained lung p-TiO2 burdens were similar and likely to induce pulmonary overload, rats demonstrated a more severe and persistent pulmonary inflammatory response than either mice or Syrian golden hamsters. Rats also were unique in the development of a progressive fibroproliferative lesion and alveolar epithelial metaplasia in response to a subchronic exposure to a high concentration of p-TiO2 particles.


    ACKNOWLEDGMENTS
 
This work was supported in part by the American Chemistry Council and the Conseil Europeen des Federations de L’Industrie Chimique. Our thanks to Ms. V. Wong and Ms. F. Trasti for their expert aid with the molecular end points and to Dr. B. Kuyper for her editorial review of the manuscript.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (919) 558-1300. E-mail: bermudez{at}ciit.org. Back

2 Present address: Advanced Inhalation Research, Cambridge, MA 02139-3789. Back

3 Present address: Wyeth Research, Collegeville, PA 19087-4517. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Baggs, R. B., Ferin, J., and Oberdorster, G. (1997). Regression of pulmonary lesions produced by inhaled titanium dioxide in rats. Vet. Pathol. 34, 592–597.[Abstract]

Bellmann, B., Muhle, H., Creutzenberg, O., and Mermelstein, R. (1992). Irreversible pulmonary changes induced in rat lung by dust overload. Environ. Health Perspect. 97, 189–191.[ISI][Medline]

Bingham, E., Corhssen, B., and Powell, C. H., Eds. (2001). Pattys Toxicology. Wiley, New York.

Creutzenberg, O., Bellmann, B., Muhle, H., Dasenbrock, C., Morrow, P., and Mermelstein, R. (1998). Lung clearance and retention of toner, TiO2, and crystalline silica, utilizing a tracer technique during chronic inhalation exposure in Syrian golden hamsters. Inhal. Toxicol. 10, 731–751.[ISI]

Cullen, R. T., Tran, C. L., Buchanan, D., Davis, J. M., Searl, A., Jones, A. D., and Donaldson, K. (2000). Inhalation of poorly soluble particles. I. Differences in inflammatory response and clearance during exposure. Inhal. Toxicol. 12, 1089–1111.[ISI][Medline]

Donaldson, K. (2000). Nonneoplastic lung responses induced in experimental animals by exposure to poorly soluble nonfibrous particles. Inhal. Toxicol. 12, 121–139.[Medline]

Heinrich, U., Fuhst, R., Rittinghausen, S., Creutzenberg, O., Bellmann, B., Koch, W., and Levsen, K. (1995). Chronic inhalation exposure of Wistar rats and two different strains of mice to diesel engine exhaust, carbon black, and titanium dioxide. Inhal. Toxicol. 7, 533–356.[ISI]

Heinrich, U., Muhle, H., Takenaka, S., Ernst, H., Fuhst, R., Mohr, U., Pott, F., and Stober, W. (1986). Chronic effects on the respiratory-tract of hamsters, mice and rats after long-term inhalation of high-concentrations of filtered and unfiltered diesel-engine emissions. J. Appl. Toxicol. 6, 383–395.[ISI][Medline]

Hext, P. M. (1994). Current perspectives on particulate induced pulmonary tumours. Hum. Exp. Toxicol. 13, 700–715.[ISI][Medline]

Lee, K. P., Trochimowicz, H. J., and Reinhardt, C. F. (1985). Pulmonary response of rats exposed to titanium dioxide (TiO2) by inhalation for two years. Toxicol. Appl. Pharmacol. 79, 179–192.[ISI][Medline]

Levy, L. S. (1994). Squamous lung lesions associated with chronic exposure by inhalation of p-aramid fibres (fine fibre dust) and to titanium dioxide: Findings of a pathology workshop. In Toxic and Carcinogenic Effects of Solid Particles in the Respiratory Tract (D. L. D. U. Mohr, J. L. Mauderly, and G. Oberdorster, Eds.), pp. 473–478. ILSI Press, Washington, DC.

Mauderly, J. L., Jones, R. K., Griffith, W. C., Henderson, R. F., and McClellan, R. O. (1987). Diesel exhaust is a pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. Appl. Toxicol. 9, 208–221.[ISI][Medline]

Mauderly, J. L., and McCunney, R. J., Eds. (1996). Particle Overload in the Rat Lung and Lung Cancer: Implications for Human Risk Assessment. Taylor and Francis, Washington, D.C.

Mauderly, J. L., Snipes, M. B., Barr, E. B., Belinsky, S. A., Bond, J. A., Brooks, A. L., Chang, I. Y., Cheng, Y. S., Gillett, N. A., Griffith, W. C., Henderson, R. F., Mitchell, C. E., Nikula, K. J., and Thomassen, D. G. (1994). Pulmonary toxicity of inhaled diesel exhaust and carbon black in chronically exposed rats. Part I: Neoplastic and nonneoplastic lung lesions. Res. Rep. Health Eff. Inst. 68–75.

Muhle, H., Bellmann, B., Creutzenberg, O., Koch, W., Dasenbrock, C., Ernst, H., Mohr, U., Morrow, P., and Mermelstein, R. (1998). Pulmonary response to toner, TiO2 and crystalline silica upon chronic inhalation exposure in Syrian golden hamsters. Inhal. Toxicol. 10, 699–729.[ISI]

Muhle, H., Creutzenberg, O., Bellmann, B., Heinrich, U., and Mermelstein, R. (1990). Dust overloading of lungs—investigations of various materials, species-differences, and irreversibility of effects. J. Aerosol Med.-Depos. Clear. Eff. Lung 3, S111–S128.

Nikula, K. J., Vallyathan, V., Green, F. H., and Hahn, F. F. (2001). Influence of exposure concentration or dose on the distribution of particulate material in rat and human lungs. Environ. Health Perspect. 109, 311–318.[ISI][Medline]

Rutten, A. A., Bermudez, E., Mangum, J. B., Wong, B. A., Moss, O. R., and Everitt, J. I. (1994). Mesothelial cell proliferation induced by intrapleural instillation of man-made fibers in rats and hamsters. Fundam. Appl. Toxicol. 23, 107–116.[ISI][Medline]

Snipes, M. B. (1996). Current information on lung overload by nonrodent mammals: Contrast with rats. Inhal. Toxicol. 8(Suppl.), 91–109.

Spitzer, J. A. (1999). Gender differences in some host defense mechanisms. Lupus 8, 380–383.[ISI][Medline]

Warheit, D. B., Hansen, J. F., Yuen, I. S., Kelly, D. P., Snajdr, S. I., and Hartsky, M. A. (1997). Inhalation of high concentrations of low toxicity dusts in rats results in impaired pulmonary clearance mechanisms and persistent inflammation. Toxicol. Appl. Pharmacol. 145, 10–22.[ISI][Medline]