Effects of Subchronically Inhaled Carbon Black in Three Species. I. Re
http://www.100md.com
《毒物学科学杂志》
Department of Environmental Medicine, University of Rochester, Rochester, New York 14642
Department of Pediatrics, University of Rochester, Rochester, New York 14642
Proctor and Gamble Co., Cincinnati, Ohio 45202
Department of Pathobiology and Diagnostic Investigation, Michigan State University, East Lansing, Michigan 48824
ABSTRACT
Exposure to high concentrations of carbon black (Cb) produces lung tumors in rats, but not mice or hamsters, presumably due to secondary genotoxic mechanisms involving persistent lung inflammation and injury. We hypothesized that the lung inflammation and injury induced by subchronic inhalation of Cb are more pronounced in rats than in mice and hamsters. Particle retention kinetics, inflammation, and histopathology were examined in female rats, mice, and hamsters exposed for 13 weeks to high surface area Cb (HSCb) at doses chosen to span a no observable adverse effects level (NOAEL) to particle overload (0, 1, 7, 50 mg/m3, nominal concentrations). Rats were also exposed to low surface area Cb (50 mg/m3, nominal; LSCb). Retention and effects measurements were performed immediately after exposure and 3 and 11 months post-exposure; retention was also evaluated after 5 weeks of exposure. Significant decreases in body weight during exposure occurred only in hamsters exposed to high-dose HSCb. Lung weights were increased in high-dose Cb-exposed animals, but this persisted only in rats and mice up to the end of the study period. Equivalent or similar mass burdens were achieved in rats exposed to high-dose HSCb and LSCb, whereas surface area burdens were equivalent for mid-dose HSCb and LSCb. Prolonged retention was found in rats exposed to mid- and high-dose HSCb and to LSCb, but LSCb was cleared faster than HSCb. Retention was also prolonged in mice exposed to mid- and high-dose HSCb, and in hamsters exposed to high-dose HSCb. Lung inflammation and histopathology were more severe and prolonged in rats than in mice and hamsters, and both were similar in rats exposed to mid-dose HSCb and LSCb. The results show that hamsters have the most efficient clearance mechanisms and least severe responses of the three species. The results from rats also show that particle surface area is an important determinant of target tissue dose and, therefore, effects. From these results, a subchronic NOAEL of 1 mg/m3 respirable HSCb (Printex 90) can be assigned to female rats, mice, and hamsters.
Key Words: Carbon black; particle retention; species comparison; particle overload; inflammation; histopathology.
INTRODUCTION
Several studies have shown that inhaled carbon black (Cb) particles, when they are administered at doses that cause particle overload in the lungs as well as chronic inflammation and epithelial hyperplasia, induce lung tumors in rats but not in other rodent species (Driscoll et al., 1996; Heinrich et al., 1995; Mauderly et al., 1994; Rausch et al., 2004). Cb is not directly mutagenic (Kirwin et al., 1981); therefore, tumor formation in rats must occur via a secondary mechanism. Impaired clearance plays a role, but it is unlikely to be solely responsible for the observed tumors because rats, mice, and hamsters all demonstrate particle overload when the dose of Cb is high enough (Oberdrster, 1995a). Studies by Driscoll et al. (1997) demonstrated a connection between the severity of inflammation and ex vivo mutations by co-incubating lung lavage inflammatory cells from Cb-exposed rats with rat lung epithelial cells. When the percentage of neutrophils in the lavage fluid was 50%, the hprt mutation rate increased significantly, possibly related to the generation and release of reactive oxygen species and/or a depletion in antioxidants. Indeed, the mutagenic response was inhibited in the presence of an antioxidant (catalase). Thus, the persistence of particles in the lung, the overall lung burden, the severity of the inflammatory response, and the release of oxidants all likely combine to produce tumors in rats, but the reasons why mice and hamsters do not develop tumors under the same high Cb load conditions is not entirely clear.
We designed a species-comparative study to test the overarching hypothesis that rat-specific tumor responses are related to species differences in particle clearance kinetics, lung inflammation and injury, antioxidant cellular defense mechanisms, and secondary genotoxicity. In this study, rats, mice, and hamsters were exposed for 13 weeks to inhaled Cb such that equivalent lung burdens in terms of retained mass were achieved. We measured particle accumulation and clearance kinetics, cytotoxicity, inflammation, cell proliferation, lung morphology, oxidant and antioxidant production, ex vivo mutations, and genotoxicity as a function of time post-exposure. This article addresses the specific hypothesis that retention, inflammation, and lung injury as a consequence of inhaled Cb exposure are more severe and persistent in rats than in mice and hamsters, and it details the results of particle lung burden analyses and the accompanying changes in body and lung weights, lavage inflammatory cells, and cell proliferation and histopathology. The details of pro- and anti-inflammatory mediator production, oxidant stress responses, and hprt mutations will be presented in a forthcoming article.
METHODS
Animals.
All animals were obtained at 5 weeks of age and were specific pathogen–free. A total of 320 female F-344 rats were obtained from Harlan (Indianapolis, IN); 670 female B6C3F1 mice were purchased from Charles River (Wilmington, MA); and 276 F1B Syrian golden female hamsters were purchased from BioBreeders (Watertown, MA). Several extra animals from each species were exposed to allow replacement of any animals that died unexpectedly. Additional sentinels were euthanized throughout the study to monitor pathogen status. Females were chosen because they were previously shown to be more sensitive to the induction of lung tumors by poorly soluble, low toxicity particles than males (Oberdrster, 1995b). All animals were acclimatized for at least 2 weeks prior to the beginning of the studies and were uniquely identified with a subcutaneous transponder chip (BioMedic Data Systems, Seaford, DE). They were fed Purina rodent chow and water ad libitum and were housed in an AAALAC-accredited barrier facility with 12-h light/dark cycles. The hamster diet was supplemented with rolled oats (as suggested by a veterinarian at BioBreeders) during the post-exposure phase of the study to increase fiber content and avoid the development of diarrhea. Body weights were obtained every 2 weeks. Growth curves for each species are provided as Supplementary Data (see Figure S1a–c).
Study design.
Animals were randomized for placement in exposure groups (n = 5–6) using a computer-based random number generator. For the histopathological and particle dosimetry analyses, groups of six animals were used; for all other end points, groups of five animals were used. There were four nominal aerosol exposure levels (0, 1, 7, and 50 mg/m3 Printex-90; high surface area Cb [HSCb]) for all animals, as well as an additional Cb type (Sterling V, low surface area Cb [LSCb]; 50 mg/m3) to which only the rats were exposed. The HSCb exposure levels were chosen to span a range of pulmonary responses including a no-observable-adverse-effects-level (NOAEL) to lung particle overload. LSCb was used to test hypotheses about size-related or surface area–related variations in responses. The HSCb particles had a primary particle size of 14 nm with a particle surface area of 300 m2/g (Brunauer-Emmet-Teller nitrogen adsorption method); the LSCb particles had a primary particle size of 70 nm and a surface area of 37 m2/g. The aerosol concentration of LSCb was adjusted so that the predicted retained lung burden matched that of the mid-dose HSCb.
Exposures to carbon black.
Printex-90 was obtained from Degussa-Huels (Trostberg, Germany), and Sterling V was supplied by Cabot (Boston, MA). Polycyclic aromatic hydrocarbon (PAH) content was 0.039 mg/kg and 8.8 mg/kg for Printex 90 and Sterling V, respectively (Borm et al., 2005). Groups of five to six animals were exposed to filtered air or to three dose levels of HSCb: 1 (low), 7 (mid), and 50 (high) mg/m3 for 6 h/day, 5 days/week for 13 weeks. Additional groups of rats were also exposed to 50 mg/m3 LSCb. All exposures were carried out with the animals in compartmentalized, horizontal flow whole-body inhalation chambers (300 l). Each chamber can hold up to 32 rats or hamsters or up to 64 mice. For the hamster exposures, the bottom of each compartment was covered with solid plastic such that only the corners remained open. Total flow through the chambers was 100 l/min. After exposure, animals were transferred to plastic filter-top cages.
The particle-containing exposure atmospheres were generated using a Wright dust feeder (LSCb) or a Venturi jet generator (HSCb) with an AccuRate screw feeder. Aerosolized particle charges were brought to Boltzmann equilibrium by passage through a 85Kr source. Aerosol concentration was continuously monitored by a RAS-2 for the mid and high dose or by a RAM-1 (Monitoring Instruments for the Environment [MIE], Inc., Bedford, MA) for the low dose. Mass concentration and particle size were periodically measured by gravimetric filter and impactor sampling, respectively. The HSCb aggregate aerosols had aerodynamic diameters ranging from 1.2 to 2.4 μm (geometric standard deviations [GSD] = 2.0–3.1); the LSCb aggregate aerosols had aerodynamic diameters of 0.6–0.9 μm (GSD = 3.0–3.7).
Determination of carbon black lung burdens.
Groups of six animals were euthanized with an overdose of sodium pentobarbital (ip, 20 mg/100 g BW) and exsanguinated at the specified time points. Following euthanasia, the unlavaged lungs were removed and the wet weight was obtained (right lung lobes for rats, hamsters; entire lung for mice). The measurement of Cb in lung tissues was done according to the methods described by Rudd and Strom (1981). Briefly, the lung tissue was minced and digested in reagent-grade 25% KOH/methanol (w/v) at 60°C overnight. The mixture was then spun at 10,000 rpm for 30 min and the remaining pellet was mixed with reagent-grade 50% HNO3/methanol (v/v). The mixture was incubated again at 60°C overnight and then spun at 10,000 rpm for 30 min; any remaining supernate was removed by evaporation. The pellet was resuspended in surfactant water, 0.1% Brij-35 (v/v) (Sigma Chemical Co., St. Louis, MO) and sonicated. The turbidity (optical density) of the solutions was measured at 690 nm using a UV/visible spectrophotometer; samples were sonicated and re-read until a stable optical density was obtained. A standard curve was constructed and used to calculate the carbon black content of each sample. The detection limit of this assay is 0.1 μg/ml suspended solution. Retention half-times were calculated using a two-parameter monoexponential decay function (see exception for mice, Table 4).
Measurement of cellular and biochemical parameters in lavage fluid.
Groups of five animals were anaesthetized with sodium pentobarbital (ip, 20 mg/100 g BW) and exsanguinated by severing the carotid artery. A small incision was made in the trachea below the fifth cartilaginous ring, and the plastic sheath from an IV catheter was tied into the trachea. The lungs, trachea, and heart were removed en bloc and the heart and excess tissue was trimmed away. Rat lungs were lavaged 10 times with 5 ml each of sterile, pyrogen-free normal saline; mice were lavaged with 1 ml of saline; hamsters, with 4 ml of saline. The recovered volume from the first two lavages was kept separate from the others for biochemical analyses of the supernatant. The cells were pelleted at 300 x g for 12 min and then pooled from all fractions and resuspended in saline for viability determination, enumeration, and differential analysis. To obtain enough cells and lavage supernate for the cellular and biochemical analyses, lavages were combined from 3–5 mice (i.e., 15 mice lavaged per time point for low-, mid-, and high-dose HSCb; 25 per time point for controls).
A sample of the pooled lavage cells was used to assess cell number and viability (trypan blue exclusion) using a bright-line hemacytometer. A hematoxylin-eosin variant was used to stain cytocentrifuged cells for differential analysis. At least 500 cells per sample were identified to determine percentages and absolute numbers of alveolar macrophages, neutrophils, and lymphocytes. The total protein concentration and the activities of lactate dehydrogenase (LDH) and -glucuronidase in bronchoalveolar lavage (BAL) fluid were measured using commercially-available kits and reagents (Sigma Chemical Co., St. Louis, MO).
Pulmonary tissue processing for histopathology and morphometry.
Tissue from the left lung was obtained from the same rats and hamsters used for the Cb determinations (n = 6); however, whole lungs from separate groups of mice (n = 6) were used for the histopathological analyses. After death, the trachea, left extrapulmonary bronchus, and left lung lobe were removed intact from the carcass and fixed via the airway with a solution of 1% paraformaldehyde (Ted Pella, Inc., Redding, CA) and 0.1% glutaraldehyde (Electron Microscopy Sciences, Ft. Washington, PA) at a constant pressure (30 cm H2O) for 1 h. After intra-airway fixation, the lumen of the proximal bronchus was ligated, the trachea was removed, and the intact left bronchus and left lung lobe were immersed in a large volume of the same fixative for an additional 24 h.
A fractionator technique was used to ensure uniformly random sampling of the left lung lobe from all animals (Bolender et al., 1993; Geiser et al., 1990). Twenty-four hours after the start of fixation, the left lung lobe was sectioned perpendicular to its long axis at a section thickness of 2–3 mm to produce 15–20 lung sections per animal (fewer sections were obtained for mice). The first cut was made at a randomly selected position within the first 3 mm of the apical portion of the lung. Selection of the first lung section for further processing was determined using a random number (1–6). This section and every sixth section following were embedded in Immuno-Bed plastic embedding medium (Polysciences, Inc., Warrington, PA). The tissue blocks were cut at a thickness of 2–3 μm with a glass knife. For general histology, sections were stained with hematoxylin and eosin.
Cell proliferation assessment.
Cell proliferation was assessed by positive immunoreactivity for bromodeoxyuridine (BrdU; Sigma) in those animals for which there was also an increase in alveolar type II epithelial cell density. BrdU (50 mg/kg body weight) was delivered to the animals in vivo at a constant rate (10 μl/h for rats, hamsters; 1 μl/h for mice) for the 5 days preceding sacrifice by means of Alzet miniosmotic pumps (ALZA Corp., Palo Alto, CA), which were implanted subcutaneously along the dorsal mid-line. After tissue fixation and processing, lung sections were immunohistochemically stained to detect nuclear BrdU incorporation. All staining steps were carried out at 37°C unless otherwise noted. First, the sections were pre-blocked at room temperature for 10 min with 0.2 M glycine, rinsed in 1:5 automation buffer (AB; Bimeda Corp., Foster City, CA), and treated for 30 min with 0.01% proteinase K in 0.05M Tris-HCl. The slides were then rinsed in 1:5 AB, treated for 30 min with 0.2N HCl, followed by treatment with Borax buffer (13 mM Borax/48 mM boric acid, pH 8.5) for 10 min at room temperature, and rinsed again in 2x AB. The slides were then incubated for 30 min with blocking solution (1.5% normal horse serum in AB), followed by a 2-h incubation with monoclonal anti-BrdU (diluted 1:25 in blocking solution; Becton Dickinson Immunocytometry Systems, San Jose, CA). They were then incubated for 30 min with a biotinylated mouse IgG antibody (diluted 1:200 in blocking solution; Vector Laboratories, Inc., Burlingame, CA), followed by a 30-min incubation with alkaline phosphatase ABC-AP reagent (Vector) at the manufacturer's recommended concentration. To allow visualization of the BrdU-labeled nuclei, the slides were incubated for 23 min at room temperature with Vector Red Alkaline Phosphatase Substrate Solution (Vector), rinsed in deionized water, and counterstained with hematoxylin for 1 min. The slides were rinsed again in water, dipped briefly in 2x AB (to blue the hematoxylin), water, 100% ethanol, 100% xylene, and coverslipped with Permount mounting medium (Fisher Scientific, Hanover Park, IL). Cell nuclei that were immunoreactive for BrdU (i.e., were in the S phase of the cell cycle) appeared bright red, and non-S-phase nuclei were blue.
Pulmonary morphometry.
Digital images of the BrdU-stained lung sections were captured with an Olympus 3-CCD color camera (Olympus America, Inc., Melville, NY) and analyzed with Scion Image software (v1.62; Scion Corp., Frederick, MD). The first image field for quantitative analysis was chosen randomly (1–5), and every fifth field following was quantified at a magnification of 790x. Only fields completely filled by lung parenchyma were analyzed (fields with conducting airways or large blood vessels were omitted). Three slides were analyzed for each animal, and 20–30 fields were analyzed per slide. Macrophages, neutrophils, and type II alveolar epithelial cells were identified morphologically and evaluated for positive BrdU staining. Cell numbers were expressed per mm2 of total alveolar parenchyma.
Statistical analyses.
Results were analyzed for statistically significant differences by two-way analysis of variance (ANOVA) with appropriate data transforms followed by Tukey t-tests using SigmaStat (SPSS Science, Chicago, IL). The two factors for the ANOVAs were exposure dose and time. Data were appropriately transformed (e.g., base 10 logarithm, natural logarithm) if an analysis of residuals suggested deviations from the assumptions of normality and equal variance. Comparisons were considered statistically significant when p 0.05.
RESULTS
Stability of Carbon Black Concentration in Exposure Chambers
There were two exposure chambers for each dose level of carbon black; the aerosol concentration did not differ significantly between the two chambers for any of the exposure levels or for any of the three species and are thus presented as single means in Table 1. For the exposures of the rats, the aerosol concentrations were quite stable around the targeted doses. Five weeks into the exposures of mice and hamsters, it was determined that the retained lung burden of HSCb (normalized to lung weight) was lower than was found in rats, which were exposed first. The mid- and high-dose levels for the mouse and hamster exposures were accordingly increased from 7 to 15 mg/m3 for the mid-dose and from 50 to 75 mg/m3 for the high dose for the remaining exposure period to produce equivalent predicted normalized lung burdens. For exposures in hamsters, the low dose also had to be increased from 1 to 1.1 mg/m3. Thus, the average chamber concentrations (for mice and hamsters) are higher than the original targets and the standard deviations are also higher. As can be seen from the mass median aerodynamic diameters (MMAD, 0.8–2.0 μm) and GSDs (2.3–3.2), the aerosols were respirable and polydispersed. The average aggregate aerosol size in the low- and high-dose mouse exposure chambers was slightly higher than in the others for reasons unknown, but that did not affect the aim of achieving low, mid, and high Cb lung burdens.
Effects of Exposure on Animal Body and Lung Weights
No rats died during exposure to carbon black. Nine control rats died in the post-exposure phase due to a blocked water line. One each in the low- and high-dose HSCb groups also died in the post-exposure phase. Thirteen mice died during exposure, seven of which were in the control group; two were in the low-dose group, three in the mid-dose group, and one in the high-dose group. In the post-exposure phase, six controls, one low-dose, two mid-dose, and six high-dose mice died. One hamster (high-dose group) died during exposure. Two hamsters each from the control, low-dose, and high-dose groups and four from the mid-dose group died in the post-exposure phase of the study. These deaths were most likely due to a change in housing conditions from wire-bottom cages (during exposure) to plastic cages (post-exposure) and reflects the heightened sensitivity of hamsters to environmental changes. Screening of sentinels (all three species) did not reveal parasitic, bacterial, or viral pathogens that would explain the unscheduled deaths. We conclude that unscheduled deaths were not related to exposure (i.e., not associated with Cb dose or time after exposure). Indeed, out of the total number of animals exposed, 4% of the animals per species died prematurely.
Body weights are summarized according to post-exposure time points in Table 2 (see also Figure S1). Rats and hamsters from the high-dose group lost weight during exposure, but this was only significant for the hamsters at the end of exposure. The differences probably reflect decreased weight gain instead of actual losses. Body weights returned to control values during the recovery period for both species (see Figure S1 a,c). There were no significant changes in body weight during exposure for other exposure doses or in the mice. In the post-exposure phase, we found that the hamsters began to lose weight after being placed in plastic cages, whereas during exposure, they gained weight faster than the sentinels (housed in plastic cages for the duration of the study). This again highlights the sensitivity to environmental change in this species.
For the rats, significant elevations in lung weight (up to more than twice the control lung weight) were found for the high-dose HSCb at the end of exposure and at all post-exposure time points (Table 3). The lung weights of rats exposed to LSCb were between those for the high- and mid-dose HSCb in magnitude and were significantly elevated 1 day and 3 months after exposure. In mice, lung weights were found to be elevated 1 day and 3 months post-exposure in the high- and mid-dose groups; at the end of the recovery period (i.e., 11 months post-exposure), significant elevations were found only in the high-dose group. In hamsters, significant elevations in lung weights were found only for the high-dose group, and these changes persisted up to 3 months post-exposure; after 11 months of recovery, no significant differences were found. Thus, the changes in lung weight were resolved most rapidly in hamsters.
Carbon Black Lung Burden
Figure 1a and b shows the carbon black lung burden and retention in rats. Lung burden was also expressed in terms of particle surface area dose (Fig. 2) because particle types of two different surface areas were used in the rat exposures, and the goal was to achieve the same lung burden in terms of surface area as with the mid-dose HSCb. The highest lung burdens in terms of mass were achieved in the high-dose HSCb and the LSCb groups. The mass burden due to LSCb (50 mg/m3) exposure was slightly but significantly higher than in the high-dose HSCb (50 mg/m3)–exposed rats at the end of exposure; the mass burdens were equivalent between the two groups thereafter. Both of these groups showed prolonged particle retention, as did the mid-dose HSCb rat group; LSCb was more efficiently cleared from the lungs than was high-dose HSCb. The mass burden values for high-dose HSCb-exposed rats were higher during recovery than at the end of exposure, which was most likely due to chance given that there was an n of 6 for each post-exposure time point. Clearly, this does not represent any further respiratory tract exposure, as rats were held in a separate holding room after the 13-week exposure; as can be seen from Figure 1b, there was no significant particle clearance in these rats. The low-dose group had the fastest clearance. However, the retention half times in rats for the mid and high doses of HSCb and for LSCb suggested that normal clearance mechanisms were overwhelmed (Table 4). In terms of surface area dose, the highest lung burdens by far were achieved in the high-dose HSCb-exposed rats (Fig. 2). The lung particle burden of LSCb-exposed rats, expressed as particle surface area, was similar to that of the mid-dose HSCb-exposed rats.
In mice, significant elevations in lung particle burdens were found during exposure and throughout recovery in all dose groups (Fig. 3a). The mice had the highest relative lung burdens of the three species when the exposures were finished, in part because the aerosol concentrations in the mid- and high-dose groups were over-adjusted after 5 weeks of exposure. Despite this, the mice cleared the particles faster at the highest dose (retention t1/2 = 322 days) than the rats (no clearance), so that by 11 months post-exposure the relative lung burdens were 48% versus 118% of end-exposure values. Although the retention as a percentage of end-exposure mass burden was lowest in the high-dose HSCb-exposed mice, clearance slowed over the 11-month recovery period such that retention in both the mid- and high-dose groups was higher than the low-dose group (Fig. 3b). The low- and mid-dose–exposed mice had very similar particle retention half times, whereas the high-dose group exhibited prolonged clearance (Table 4).
In hamsters, significant elevations in lung carbon black burden were found for all doses and at all time points (Fig. 4a). The particle retention as a percentage of mass burden at 1 day post-exposure in the high-dose group was higher than in the mid- and low-dose groups for all post-exposure times (Fig. 4b). The retention half times for the low- and mid-dose groups of the hamsters were very similar and were the lowest of the three species. The retention half-time for the high-dose group (309 days) was similar to that of the high-dose mice. As was also true for mice, hamsters exposed to high-dose HSCb exhibited impaired clearance (Table 4).
Cellular and Biochemical Parameters in Lavage Fluid
Summaries of all cellular and biochemical measurements in BAL fluid are presented for rats, mice, and hamsters, respectively, in Table S1 through Table S3. Figures 5 and 6 summarize the total cell number and percent PMN data, respectively. Total BAL cell numbers were elevated at 1 day post-exposure and all through the recovery period in the high-dose groups for all three species in comparison to all other exposure groups. Whereas total cell number gradually decreased over time in mice and hamsters, it stayed elevated in the rats exposed to high-dose HSCb and LSCb (Fig. 5). Total cell numbers in the controls did not change over time in mice and hamsters, whereas, in rats, there was a small decrease; the mean control values in chronological order were 1.85 x 107 ± 0.61, 1.15 x 107 ± 0.21, and 1.01 x 107 ± 0.14 (rats); 0.90 x 106 ± 0.09, 1.22 x 106 ± 0.07, and 1.02 x 106 ± 0.23 (mice); 1.30 x 107 ± 0.19, 1.56 x 107 ± 1.61, and 1.64 x 107 ± 0.29 (hamsters). The magnitude of the response to LSCb in rats was between those for high- and mid-dose HSCb. The mid-dose groups in rats and mice also exhibited significant elevations in total cell numbers after exposure, but recovery occurred before the study was terminated.
The presence of polymorphonuclear leukocytes (PMN) in lavage fluid was used as a sensitive indicator of lung inflammation. PMN were elevated in the mid- and high-dose HSCb and LSCb groups of rats at the end of exposure and throughout the recovery period. In rats, the PMN responses to high-dose HSCb and LSCb were similar (Fig. 6a; see also Table S1). The PMN remained elevated in rats from the high-dose HSCb and LSCb groups through the end of the study. At the end of exposure in mice and hamsters (Fig. 6b,c), the mid- and high-dose groups were also elevated compared to the controls and the low-dose group; the magnitudes of response were similar for the two species. During the recovery period, the percentage of PMN from the mid- and high-dose mice remained different from each other and the rest of the exposure groups. In hamsters, the PMN response decreased between 3 and 11 months post-exposure such that only the high-dose group retained a significant response at the termination of the study. The magnitude of the PMN response was the highest in rats and was the most prolonged of the three species. Thus, it appears that a maximal response was reached in mice and hamsters, but not in rats, with increasing mass dose of HSCb. The mean percentages of PMN in controls in chronological order were 0.47 ± 0.28%, 0.15 ± 0.16%, and 0.46 ± 0.23% (rats); 0.35 ± 0.26%, 0.23 ± 0.16%, and 0.67 ± 0.44% (mice); and 0.71 ± 0.15%, 0.81 ± 0.19%, and 2.51 ± 2.51% (hamsters). The concentration of total protein and activities of LDH and -glucuronidase in BAL fluid exhibited significant elevations in all three species (see Tables S1–S3) that were strongly correlated with changes in the PMN response. There were no significant elevations in cellular or biochemical parameters for any of the animals exposed to low-dose HSCb.
Rat Histopathology
No exposure-related alterations were observed microscopically at any of the post-exposure time points in the lungs of control rats (Fig. 7a). Likewise, no adverse histological effects of exposure were found at any post-exposure time point in the lungs of rats exposed to low-dose HSCb (not shown). The lungs of these rats were similar to those of the control rats, with the exception that some alveolar macrophages containing small amounts of Cb were widely scattered throughout the alveolar air spaces of the pulmonary parenchyma. However, the number and morphologic character of these particle-containing macrophages were not different from those in the controls.
Exposure-related histopathology was present in the lung lobes of rats exposed to mid- and high-dose HSCb, but the most severe alterations were present in the rats exposed to high-dose HSCb (Fig. 7b–d). At 1 day post-exposure, the principal exposure-related lung lesions were centered primarily in the centriacinar regions of the lungs with the most extensive epithelial and inflammatory responses in the alveolar ducts and the immediately surrounding alveolar parenchyma. These lesions were characterized by (1) the accumulation of large numbers of markedly hypertrophic and highly vacuolated alveolar macrophages laden with HSCb; (2) an associated mixed inflammatory cell influx composed of neutrophils and lesser numbers of mononuclear cells located in the alveolar walls, alveolar air spaces, and perivascular interstitium; and (3) hyperplasia and hypertrophy of alveolar type II cells (Fig. 7b). Many of the particle-laden macrophages were degenerative or necrotic indicating particle-mediated cytotoxicity. Cellular debris and free HSCb were also found in the affected alveolar air spaces. Though most of the HSCb was found within alveolar macrophages, lesser amounts were present in alveolar epithelium and interstitium of the affected parenchyma and in the bronchus-associated lymphoid tissue in the upper conducting airways (axial and preterminal airways). A mild to moderate alveolar proteinosis was also a consistent finding in the affected regions of the lung. Interestingly, these lesions were also evident in high-dose HSCb-exposed rats that were sacrificed 3 and 11 months post-exposure (Fig. 7c,d). There was only mild attenuation in the severity of the inflammatory and epithelial lesions at these later time points compared to those observed at 1 day post-exposure.
Interstitial fibrosis was not present in the lungs of rats sacrificed at 1 day post-exposure, but a mild, widely scattered interstitial fibrosis of the alveolar septa was present in the HSCb-exposed rats that were sacrificed 13 weeks after the end of the exposure. This particle-induced, multifocal fibrosis was markedly greater in both extent and severity in the lungs of rats exposed to the high concentration of HSCb and sacrificed 11 months post-exposure. Varying sized foci of alveolar, peribronchiolar, perivascular, and pleural fibrosis were scattered throughout the lungs of these rats, but they were always associated with HSCb-laden degenerative macrophages, neutrophilic inflammation, and alveolar epithelial proliferation. Conspicuous aggregates of free and phagocytized HSCb particles were often scattered throughout the areas of fibrosis. In addition, small foci of epithelial metaplasia composed of columnar ciliated and nonciliated cells lining alveolar ducts were commonly associated with the fibrotic areas of the affected centriacinar regions.
The pulmonary lesions in the rats exposed to the mid dose of HSCb and sacrificed 1 day post-exposure were generally similar in character but markedly less severe and extensive than those in the lungs of rats exposed to high-dose HSCb (Fig. 8). The major differences in these lung lesions compared to those in the high-dose rats, were (1) less involvement of the distal alveolar parenchyma (lesion mainly focused in the proximal alveolar ducts), (2) less degeneration of particle-laden macrophages, and (3) minimal associated inflammation and alveolar epithelial changes in the affected regions of the lung. There were focal accumulations of alveolar macrophages that were heavily laden with Cb particles in the centriacinar regions of the lung, but macrophages with microscopic features of degeneration and necrosis (e.g., cytoplasmic vacuolization, nuclear pyknosis, severe hypertrophy, cellular fragmentation) were infrequent. Widely scattered small aggregates of Cb particles were still present in the lungs of rats exposed to the mid dose at 3 and 11 months post-exposure. However, there was no associated influx of neutrophils or alveolar type II cell hyperplasia at these later time points post-exposure. Interestingly, there were more Cb particles in peribronchial and perivascular interstitial regions of the lung in the mid-dose HSCb-exposed rats sacrificed at 3 and 11 months post-exposure than in similarly exposed rats sacrificed at 1 day post-exposure. However, no particle-induced fibrosis was evident in any of these rats exposed to mid-dose HSCb.
Rats exposed to LSCb that were sacrificed 1 day post-exposure had similar pulmonary lesions as those in the rats exposed to mid-dose HSCb, but their lung lesions were greater in both severity and extent (Fig. 8i,j). These animals had a greater number of degenerative alveolar macrophages, inflammatory cells (neutrophils), and alveolar type II cells (hyperplasia) than the rats exposed to mid-dose HSCb, but the numbers of these cells were considerably fewer than those found in the lungs of rats exposed to high-dose HSCb. Rats exposed to LSCb and sacrificed 3 and 11 months after the end of exposure had lung lesions, including interstitial fibrosis, that were similar in character but that appeared to be slightly less in extent and severity than those observed in rats exposed to high-dose HSCb.
Quantitation of Pulmonary Morphometric Analyses in Rats
Numeric cell densities of alveolar type II cells in the alveolar parenchyma of Cb-exposed rats are presented in Table 5; those for neutrophils and alveolar macrophages/monocytes are shown in the Supplementary Data (see Tables S4 and S5, respectively). Compared to controls, rats exposed to HSCb had a dose-dependent increase in the numeric densities of all the measured cell types at 1 day post-exposure. There was a 129% increase in the numeric density of alveolar type II cells in rats exposed to high-dose HSCb (Table 5) as compared to controls. Approximately 8% of the type II cells in the animals exposed to high-dose HSCb had nuclear labeling for BrdU, indicating a significant increase in DNA synthesis compared to that in controls (1% labeling index). However, no HSCb-induced increase in alveolar type II cell number was found in rats exposed to the low or mid dose of HSCb compared to controls. At 3 and 11 months post-exposure, rats exposed to high-dose HSCb, but not to the low or mid doses, had significantly more alveolar type II cells than the control rats (156% and 98% more, respectively). In addition, the BrdU labeling index remained high in these epithelial cells at both 3 and 11 months post-exposure (7% and 12% respectively, compared to 1–3% in controls). Rats exposed to LSCb had similar increases in the numeric cell density of alveolar type II cells as compared to high-dose HSCb-exposed rats (106%, 135%, and 53% at 1 day and 3 and 11 months, respectively). The labeling index for BrdU in alveolar type II cells was also significantly elevated above that of controls in LSCb-exposed rats sacrificed at all post-exposure time points (7%, 3%, and 8% at 1 day and 3 and 11 months, respectively; Table 5).
Only those rats exposed to high-dose HSCb and LSCb had significant increases in the numeric density of alveolar neutrophils compared to controls at 1 day post-exposure (see Table S4). However, significantly fewer neutrophils were present in the rats exposed to LSCb (90 ± 20 cells/mm2) than in those exposed to high-dose HSCb (148 ± 13 cells/mm2 of parenchyma) at 1 day post-exposure. Interestingly, Cb-induced increases in alveolar neutrophils were maintained in both the high-dose HSCb- and LSCb-exposed rats at 3 and 11 months post-exposure. Compared to controls, there were 283% and 222% more neutrophils in rats exposed to high-dose HSCb and LSCb, respectively, at 11 months post-exposure. Carbon black–induced increases in the numeric cell density of alveolar macrophages were evident in rats exposed to the mid and high doses of HSCb and to LSCb at 1 day and 3 months post-exposure (see Table S5). At 11 months post-exposure, only the rats exposed to high-dose HSCb or LSCb had significantly more macrophages than the control rats. The densities of macrophages were not significantly different between rats exposed to high-dose HSCb and LSCb at any post-exposure time point.
Mouse Histopathology
No exposure-related alterations were observed microscopically at any post-exposure time point in the lungs of control mice (Fig. 9a); neither were any adverse histological effects of exposure found in the lungs of mice exposed to low-dose HSCb. The lungs of these mice were microscopically similar to those of the control mice, except that some alveolar macrophages containing small amounts of Cb were widely scattered throughout the pulmonary parenchyma. The number, size, and morphologic character of these particle-containing macrophages, however, were not different from those in the control mice.
In contrast, exposure-related lung lesions were present in mice exposed to mid- and high-dose HSCb and sacrificed at 1 day and 3 and 11 months post-exposure. Though the lesions were similar in these two exposure groups, the magnitude of the lesions (severity) was consistently greater in mice exposed to high-dose HSCb (Fig. 9b–d). The most conspicuous change in the lungs of these mice was the presence of numerous Cb particles that were widely scattered throughout the alveolar parenchyma, but more concentrated in centriacinar regions including alveolar ducts and adjacent alveoli. Aggregates of particles were either free in the alveolar air spaces or within enlarged alveolar macrophages. Though many of the alveolar macrophages were engorged with particles, few of these phagocytic cells had obvious features of cellular degeneration or necrosis like those commonly found in rats exposed to high-dose HSCb. Most of the particles were present in alveolar lumens, but much smaller numbers were present in macrophages within interstitial tissues surrounding terminal bronchioles or small blood vessels. At 3 and 11 months, but not 1 day, post-exposure, there was also an obvious increase in the numbers of alveolar macrophages in the lung parenchyma and a mild mononuclear cell infiltrate (mainly lymphocytes, monocytes and plasma cells) within the perivascular and peribronchiolar interstitium (Fig. 9c,d). In contrast to the lungs of rats exposed to HSCb, there were very few, if any, neutrophils and no conspicuous alveolar type II cell hyperplasia in the Cb-laden centriacinar and alveolar regions at 1 day post-exposure. Alveolar type II cell hyperplasia was present in mice exposed to high-dose HSCb and sacrificed 3 months post-exposure. However, this proliferation of type II cells did not persist in similarly exposed mice that were sacrificed 11 months after exposure. Interestingly, alveolar proteinosis was also a characteristic feature of all mice exposed to high-dose HSCb, as was seen in rats. This alveolar lesion was evident at all post-exposure time points. However, none of the mice developed alveolar septal fibrosis like that observed in the HSCb-exposed rats at 3 and 11 months post-exposure.
Quantitation of Pulmonary Morphometric Analyses in Mice
In contrast to rats, no Cb-exposed mice had numeric densities of alveolar type II cells that were significantly different from control mice at 1 day or 11 months post-exposure (Table 5). Only mice exposed to high-dose HSCb and sacrificed 3 months post-exposure had a significant increase (181%) in the numbers of alveolar type II epithelial cells compared to the control mice. There were no HSCb-related increases in neutrophils in the alveolar parenchyma of mice at 1 day or 3 months post-exposure. At 11 months post-exposure, there was a small but statistically significant increase in the number of neutrophils in the lungs of mice exposed to low- and mid-dose, but not high-dose, HSCb compared to that in the lungs of control mice sacrificed at the same time point (see Table S4). Modest exposure-related increases in alveolar macrophage numbers (2.7–3.5x those of controls) were only evident in mice exposed to high-dose HSCb and sacrificed 3 or 11 months post-exposure and to mid-dose HSCb and sacrificed 11 months post-exposure (see Table S5).
Hamster Histopathology
No exposure-related alterations were observed microscopically in the lungs of control (Fig. 10a) or low dose HSCb-exposed hamsters at any post-exposure time point. The lungs of low-dose HSCb-exposed hamsters (like those of similarly exposed rats and mice) were microscopically similar to those of the control animals, with the exception of some alveolar macrophages containing small amounts of Cb that were widely scattered throughout the pulmonary parenchyma. The number, size, and morphologic character of alveolar macrophages in these rodents were not different from those in the control hamsters. In contrast, hamsters exposed to the mid dose and the high dose of HSCb had widely scattered aggregates of enlarged alveolar macrophages and a few multinucleated giant cells engorged with phagocytized HSCb particles at all post-exposure time points (Figure 10 b-d). Most of these HSCb-laden macrophages were present in alveolar air spaces in or near centriacinar regions of the lungs. There was an increase in free and phagocytized particles in the interstitial spaces of alveolar ducts, terminal bronchioles, and blood vessels at the later time points. The number of widely scattered aggregates of HSCb-laden macrophages decreased with time, but it was consistently greater in the hamsters exposed to high-dose HSCb as compared to mid-dose HSCb. Associated with the particle-laden macrophages in the lungs of hamsters exposed to high-dose HSCb, there was a mild type II cell hyperplasia only at 1 day post-exposure. This alveolar epithelial lesion was not present in the mid- or high-dose–exposed hamsters. In addition, there was a conspicuous absence of neutrophilic inflammation, alveolar proteinosis, and septal fibrosis in the lungs of all of the hamsters.
Quantitation of Pulmonary Morphometric Analyses in Hamsters
At one day post-exposure, hamsters exposed to high-dose HSCb had significantly greater numbers of type II cells (Table 5) and alveolar macrophages (see Table S5) in alveolar parenchyma. These hamsters had 27% more alveolar type II cells and 62% more macrophages than controls. However, there was no increase in the numeric cell density of neutrophils above controls in the HSCb-exposed hamsters (see Table S4). Hamsters exposed to mid-dose HSCb and sacrificed 1 day post-exposure had a mild increase in the number of alveolar macrophages above controls (1.5x), but no increases in the number of neutrophils or alveolar type II cells were observed in the lung parenchyma. Hamsters exposed to low-dose HSCb had no significant differences in the numeric cell densities of alveolar type II cells, neutrophils, or alveolar macrophages at 1 day post-exposure as compared to controls. At 3 and 11 months post-exposure, high-dose HSCb-exposed hamsters had persistent increases in the numeric cell density of alveolar macrophages compared to control hamsters (53% and 38% more, respectively; see Table S5). There were no increases in alveolar type II cells or neutrophils at these later time points (Table 5; see also Table S4). Numeric cell densities for hamsters exposed to low- and mid-dose HSCb and sacrificed at 3 and 11 months post-exposure were not determined because there were no morphologic indications of increases in these cell types in the lungs of hamsters.
DISCUSSION AND CONCLUSIONS
The International Agency for Research on Cancer (IARC) has classified Cb as possibly carcinogenic to humans (Group 2B; IARC, 1996). This evaluation was based on inadequate evidence in humans, and sufficient evidence in experimental animals. In the case of the human epidemiological evidence, the U.S. studies on cancer in carbon black workers showed no excess of lung cancer (Robertson and Ingalls, 1980, 1989; Robertson and Inman, 1996). A UK study of Cb production workers showed a marginally significant increase in lung cancer (Hodgson and Jones, 1985). The IARC, in its 1996 evaluation, noted a variety of methodological shortcomings in the studies they reviewed, which precluded a definitive evaluation of the epidemiological findings. In a follow-up investigation of the Hodgson and Jones study, Sorahan et al. (2001) confirmed this increase in lung cancer, but they also demonstrated that it was not related to cumulative Cb exposure and suggested that it may be related to other factors.
The present study was designed to determine whether particle retention, lung inflammation, and pathology induced by subchronic inhalation of Cb are more pronounced in rats than in mice and hamsters, which could contribute to greater secondary genotoxic events in rats. A greater and more persistent inflammatory and related oxidative stress response in the rat lung is a likely mechanism by which high lung burdens of particles such as Cb can induce secondary genotoxicity resulting in epithelial cell mutations and subsequent lung tumor formation. Lung tumors induced by poorly soluble particles of low cytotoxicity (PSP) due to a condition of "particle overload" in the lung have only been observed in rats following chronic high-level exposures, but not in mice or hamsters. Thus, the overall study focused on a comparison between three rodent species of particle retention kinetics and pulmonary inflammatory, oxidative stress and genotoxic events; the present article reports on particle uptake and retention as well as Cb-induced lung inflammation and pathology.
Lung particle overload is a concept that applies only to PSP and was defined by Morrow (1988) in terms of retained particle volume as a state that arises when macrophage-mediated clearance mechanisms are overwhelmed. Evidence was summarized showing that retention kinetics are altered when mass burdens reach 1–3 mg particles/g lung tissue. According to this definition, particle overload was achieved in the present study in rats exposed to mid- and high-dose HSCb and LSCb, in mice exposed to the mid dose and the high doses, and in hamsters exposed to the high dose of HSCb. Further evidence in support of overload is demonstrated in the prolonged retention half times (Table 4) for rats (except the low dose) and for the high doses in mice and hamsters (assuming normal retention half times of 70 days for rats and 55 days for mice; Kreyling, 1990; Oberdrster, 1995a). The clearance kinetics and surface area burden of LSCb were similar to that of mid-dose HSCb, indicating the importance of particle surface area in particle retention. These data support a suggestion made by Oberdrster and Yu (1990) and, later, by Tran et al. (2000), regarding particle surface area–related effects of PSPs.
Several chronic and sub-chronic exposure studies on the effects of PSP have been conducted, one of the most recent ones by Bermudez et al. (2004), in which three rodent species were exposed to ultrafine titanium dioxide (uf-TiO2). There are similarities between uf-TiO2 and HSCb in terms of primary particle size (21 [uf-TiO2] versus 17 nm [HSCb]) and aerodynamic diameter (1.3–1.5 μm [uf-TiO2] versus 1.4–2.0 μm [HSCb]). Thus, it could be postulated that the two particle types would deposit and be retained similarly in the lungs and that they might produce similar effects. Considering lung burden in rats at the end of the exposure period (which was 13 weeks in the Bermudez study and in the present study) as an example, it is clear that the mass dose does not adequately predict outcome: 50 mg/m3 HSCb produced a mass burden (mg/g wet lung tissue) that was 2–3 times lower than 10 mg/m3 uf-TiO2. A suggestion from earlier studies (Oberdrster et al., 1994) is that particle surface area is a better predictor of lung responses to inhaled PSP than mass dose. Using the surface areas of each particle type (and correcting for wet versus dry tissue weights), the lung burdens for 10 mg/m3 uf-TiO2 and 7 mg/m3 HSCb are more alike (0.1 and 0.3 m2/g lung, respectively), whereas that for 50 mg/m3 HSCb is the highest (0.9 m2/g lung) at the end of exposure. The surface area burden for 50 mg/m3 LSCb is 0.2 m2/g lung. As reported by Bermudez et al. (2002, 2004), the surface area dose for 10 mg/m3 uf-TiO2 was similar to that for 50 mg/m3 pigmentary (p) TiO2. The following, then, is a set of four different PSPs that produce similar surface area tissue burdens: 50 mg/m3 LSCb, 7 mg/m3 HSCb, 10 mg/m3 uf-TiO2, and 50 mg/m3 p-TiO2. If surface area dose were the best predictor of outcome, then the retention kinetics and inflammatory and morphological changes should be similar for the dose levels given above for the four particle types. This is, however, not quite the case: the retention half times are somewhat shorter, the percentages of PMNs in BAL fluid are lower, and the histopathological changes are less severe for the two Cb particle types than for the TiO2.
the data presented here, as well as the preceding discussion, we propose that particle surface area dose may be the best metric to predict responses to PSPs as compared to other parameters, e.g., mass dose. Driscoll (1996) compiled data from several PSP chronic inhalation studies in rats and showed that the lung tumor response was better described by the retained particle surface area dose than by mass dose. For the present study, there are parameters other than surface area that could explain the differences in response that we observed in rats exposed to HSCb and LSCb. For one, the Cb particles were produced by different processes, possibly resulting in trace contaminant differences that could also contribute to the responses we observed. It is also possible that differences in the way inhaled aggregated aerosols disaggregate in the lungs could affect response, as suggested by Ferin et al. (1992), as well as differences in rates of interstitialization. Lastly, the PAH content of the LSCb was much higher (225x) than for the HSCb (8.8 versus 0.039 mg/kg). Given the fact that responses to LSCb were often between those for mid- and high-dose HSCb in magnitude, it is possible that the higher PAH content of LSCb is one explanation. Although recent work by Borm et al. (2005) showed that neither HSCb nor LSCb at the end of subchronic exposure induced DNA-PAH adduct formation in the lung (32P post-labeling analysis of lung samples from our study), it was also shown that LSCb did induce PAH adducts after acute in vitro exposure of A549 cells. This was not found after similar in vitro exposures with HSCb. These results indicate that DNA repair processes may have occurred in vivo in LSCB-exposed rats during and/or after subchronic inhalation exposure. Whether the greater PAH content of LSCb may have contributed to the observed effects in our study cannot be firmly determined from our results. We suggest, therefore, that PSP surface area dose plays a substantial role in determining outcome for PSPs of similar chemical composition and crystal structure, but that other factors may also contribute.
The level of inflammation found in the present study, as determined by the percentage of BAL PMNs, was comparable to that in another study conducted by Driscoll et al. (1996) to assess the inflammatory and mutagenic potential of Monarch 880 Cb particles (primary particle size, 16 nm; aerodynamic diameter, 0.88 μm; SA, 220 m2/g) in rats. Although the lung burdens in terms of total mass per lung in our study were not as high as reported by Driscoll, clearance impairment was observed in both studies. The present study extends these observations by performing the same panel of analyses in each of three rodent species as well as by exposing rats to a low surface area Cb.
The results of this comparative study in rats, mice, and hamsters confirm previous findings that the rat is the most sensitive of the three species with respect to PSP-induced adverse effects in the respiratory tract (Heinrich et al., 1995; Oberdrster et al., 1994). Retention of Cb in rats was longest, and pulmonary inflammation and histopathological changes were more severe and more persistent than in mice and hamsters. In addition, companion studies found that the generation of oxidants (superoxide anion, hydrogen peroxide, nitric oxide) by lavaged cells and mutational changes (hprt locus) were higher in rats than in mice and hamsters (Carter et al., 2003). Among other parameters, antioxidant levels (lung tissue and lavage fluid levels of superoxide dismutase and glutathione peroxidase) were also assessed. These indicators of oxidant/antioxidant balance, as a whole, suggested that rats experienced greater oxidant load in response to HSCb as compared to mice and hamsters. The results of these studies will be presented in a forthcoming article. Furthermore, progressive dose-related oxidative DNA damage in rat lung tissue was reported by Gallagher et al. (2003).
In summary, the results presented here demonstrate the importance of particle surface area in governing the retention kinetics of inhaled Cb particles as well as the subsequent inflammatory and morphological changes in rats. Specifically, a Cb with eightfold lower surface area (LSCb) exhibited similar mass burdens as high-dose HSCb, but it was cleared in a similar fashion and had similar toxicological effects as compared to mid-dose HSCb. Although we did not design our study to determine a surface area dose threshold, as that would have involved vastly more doses and data points we can estimate the surface area to be 0.03 m2/g lung for HSCb and LSCb in rats based on the PMN response. The responses to Cb particles inhaled for 13 weeks demonstrated clear relationships to dose and recovery time and, as other investigators have previously reported, we also found that inhaled Cb delivered at a high dose overloads the normal clearance mechanisms of rat, mouse, and hamster lungs. The rank order for Cb particle retention, irrespective of dose, is rat >>> mouse, hamster. Similarly, the order of sensitivity to Cb particles as found in this study is rats > mice > hamsters. No adverse inflammatory or morphological changes were observed at the lowest exposure concentration (1 mg/m3) of Cb in any of the three species. Interestingly, for the low- and mid-dose HSCb groups, the responses among the three species were similar in magnitude. Lastly, a subchronic NOAEL of 1 mg/m3 respirable HSCb (Printex 90) can be assigned to female rats, mice, and hamsters from these results.
SUPPLEMENTARY DATA
Supplementary data are available online at www.toxsci.oxfordjournals.org:
Table S1. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black–exposed rats. All exposure doses, particle types, and post-exposure evaluations are shown; Table S2. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black–exposed mice. All exposure doses and post-exposure evaluations are shown; Table S3. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black-exposed hamsters. All exposure doses and post-exposure evaluations are shown; Table S4. Numeric densities of neutrophils in the lungs of rats, mice, and hamsters following 13 weeks of exposure to carbon black; Table S5. Numeric densities of macrophages/monocytes in the lungs of rats, mice, and hamsters following 13 weeks of exposure to carbon black; Figure S1. Growth curves for rats (a), mice (b), and hamsters (c) exposed to carbon black for 13 weeks. The data shown cover the exposure and 11-month recovery periods. Green circles, controls; pink circles, low dose HSCb; blue circles, mid-dose HSCb; red circles, high-dose HSCb; and black circles, LSCb (rats only, panel a).
ACKNOWLEDGMENTS
The authors thank Nancy Corson, Kiem Nguyen, Pamela Wade-Mercer, Lori Bramble, and Drs. James Wagner and Jon Hotchkiss for their excellent technical assistance. This work was supported by a grant from the International Carbon Black Association and by a National Institute of Environmental Health Sciences (NIEHS) Environmental Health Sciences Center grant (P30 ESO1247).
REFERENCES
Bermudez, E., Mangum, J. B., Asgharian, B., Wong, B. A., Reverdy, E. E., Janszen, D. B., Hext, P. M., Warheit, D. B., and Everitt, J. I. (2002). Long-term pulmonary responses of three laboratory rodent species to subchronic inhalation of pigmentary titanium dioxide. Toxicol. Sci. 70, 86–97.
Bermudez, E., Mangum, J. B., Wong, B. A., Asgharian, B., Hext, P. M., Warheit, D. B., and Everitt, J. I. (2004). Pulmonary responses of rats, mice, and hamsters to subchronic inhalation of ultrafine titanium dioxide particles. Toxicol. Sci. 77, 347–357.
Bolender, R. P., Hyde, D. M., and Dehoff, R. T. (1993). Lung morphometry: a new generation of tools and experiments for organ, tissue, cell, and molecular biology. Am. J. Physiol. 265, L521–L548.
Borm, P. J. A., Cakmak, G., Jermann, E., Weishaupt, C., Kempers, P., van Schooten, F. J., Oberdrster, G., and Schins, R. P. F. (2005). Formation of PAH-DNA adducts after in vivo and in vitro exposure of rats and lung cells to different commercial carbon blacks. Toxicol. Appl. Pharmacol. 205, 157–167.
Carter, J. M., Oberdrster, G., Elder, A. P., Singh, K. A., and Driscoll, K. E. (2003). Species differences in response to the effects of inhaled carbon black particle overload in the lung. Am. J. Respir. Crit. Care Med. 167, A492.
Driscoll, K. E. (1996). Role of inflammation in the development of rat lung tumors in response to chronic particle exposure. Inhal. Toxicol. 8, 139–153.
Driscoll, K. E., Carter, J. M., Howard, B. W., Hassenbein, D. G., Pepelko, W., Baggs, R. B., and Oberdrster, G. (1996). Pulmonary inflammatory, chemokine, and mutagenic responses in rats after subchronic inhalation of carbon black. Toxicol. Appl. Pharmacol. 136, 372–380.
Driscoll, K. E., Deyo, L. C., Carter, J. M., Howard, B. W., Hassenbein, D. G., and Bertram, T. A. (1997). Effects of particle exposure and particle-elicited inflammatory cells on mutation in rat alveolar epithelial cells. Carcinogenesis 18, 423–430.
Ferin, J., Oberdrster, G., and Penney, D. P. (1992). Pulmonary retention of ultrafine and fine particles in rats. Am. J. Respir. Cell Mol. Biol. 6, 535–542.
Gallagher, J., Sams, R., Inmon, J., Gelein, R., Elder, A., Oberdrster, G., and Prahalad, A. K. (2003). Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine in rat lung DNA following subchronic inhalation of carbon black. Toxicol. Appl. Pharmacol. 190, 224–231.
Geiser, M., Cruz-Orive, L. M., Imhof, V., and Gehr, P. (1990). Assessment of particle retention and clearance in the intrapulmonary conducting airways of hamster lungs with the fractionator. J. Microsc. 160, 75–88.
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, 533–556.
Hodgson, J. T., and Jones, R. D. (1985). A mortality study of carbon black workers employed at five United Kingdom factories between 1947 and 1980. Arch. Environ. Health. 40, 261–268.
IARC (International Agency for Research on Cancer). (1996). Carbon Black, vol. 65, p. 149.
Kirwin, C. J., LeBlanc, J. V., Thomas, W. C., Haworth, S. R., Kirby, P. E., Thaligar, A., Bowman, J. T., and Brusick, D. J. (1981). Evaluation of the genetic activity of industrially produced carbon black. J. Toxicol. Environ. Health 7, 973–989.
Kreyling, W. G. (1990). Interspecies comparison of lung clearance of "insoluble" particles. J. Aerosol Med. 3F, S93–S110.
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., et al. (1994). Pulmonary Toxicity of Inhaled Diesel Exhaust and Carbon Black in Chronically Exposed Rats. Part I: Neoplastic and Nonneoplastic Lung Lesions. Health Effects Institute, Cambridge, MA. 68, 1–75.
Morrow, P. E. (1988). Possible mechanisms to explain dust overloading of the lungs. Fundam. Appl. Toxicol. 10, 369–384.
Oberdrster, G. (1995a). Lung particle overload: Implications for occupational exposures to particles. Reg. Toxicol. Pharmacol. 27, 123–135.
Oberdrster, G. (1995b). The NTP Talc Inhalation Study: A critical appraisal focused on lung particle overload. Reg. Toxicol. Pharmacol. 21, 233–241.
Oberdrster, G., Ferin, J., Soderholm, S., Gelein, R., Cox, C., Baggs, R., and Morrow, P. E. (1994). Increased pulmonary toxicity of inhaled ultrafine particles: Due to lung overload alone Ann. Occup. Hyg. 38, 295–302.
Oberdrster, G., and Yu, C. P. (1990). The carcinogenic potential of inhaled diesel exhaust: A particle effect J. Aerosol Sci. 21, S397–S401.
Rausch, L. J., Bisinger, E. C., Jr., and Sharma, A. (2004). Carbon black should not be classified as a human carcinogen based on rodent bioassay data. Reg. Toxicol. Pharmacol. 40, 28–41.
Robertson, J. M., and Ingalls, T. H. (1980). A mortality study of carbon black workers in the United States from 1935 to 1974. Arch. Environ. Health. 35, 181–186.
Robertson, J. M., and Ingalls, T. H. (1989). A case-control study of circulatory, malignant, and respiratory morbidity in carbon black workers in the United States. Am. Ind. Hyg. Assoc. J. 50, 510–515.
Robertson, J. M., and Inman, K. J. (1996). Mortality in carbon black workers in the United States. J. Occup. Environ. Med. 38, 569–570.
Rudd, C. J., and Strom, K. A. (1981). A spectrophotometric method for the quantitation of diesel exhaust particles in guinea pig lung. J. Appl. Toxicol. 1, 83–87.
Sorahan, T., Hamilton, L., van Tongeren, M., Gardiner, K., and Harrington, J. M. (2001). A cohort mortality study of U.K. carbon black workers, 1951–1996. Am. J. Ind. Med. 39, 158–170.
Tran, C. L., Buchanan, D., Cullen, R. T., Searl, A., Jones, A. D., and Donaldson, K. (2000). Inhalation of poorly soluble particles. II. Influence of particle surface area on inflammation and clearance. Inhal. Toxicol. 12, 1113–1126.(Alison Elder, Robert Gelein, Jacob N. Fi)
Department of Pediatrics, University of Rochester, Rochester, New York 14642
Proctor and Gamble Co., Cincinnati, Ohio 45202
Department of Pathobiology and Diagnostic Investigation, Michigan State University, East Lansing, Michigan 48824
ABSTRACT
Exposure to high concentrations of carbon black (Cb) produces lung tumors in rats, but not mice or hamsters, presumably due to secondary genotoxic mechanisms involving persistent lung inflammation and injury. We hypothesized that the lung inflammation and injury induced by subchronic inhalation of Cb are more pronounced in rats than in mice and hamsters. Particle retention kinetics, inflammation, and histopathology were examined in female rats, mice, and hamsters exposed for 13 weeks to high surface area Cb (HSCb) at doses chosen to span a no observable adverse effects level (NOAEL) to particle overload (0, 1, 7, 50 mg/m3, nominal concentrations). Rats were also exposed to low surface area Cb (50 mg/m3, nominal; LSCb). Retention and effects measurements were performed immediately after exposure and 3 and 11 months post-exposure; retention was also evaluated after 5 weeks of exposure. Significant decreases in body weight during exposure occurred only in hamsters exposed to high-dose HSCb. Lung weights were increased in high-dose Cb-exposed animals, but this persisted only in rats and mice up to the end of the study period. Equivalent or similar mass burdens were achieved in rats exposed to high-dose HSCb and LSCb, whereas surface area burdens were equivalent for mid-dose HSCb and LSCb. Prolonged retention was found in rats exposed to mid- and high-dose HSCb and to LSCb, but LSCb was cleared faster than HSCb. Retention was also prolonged in mice exposed to mid- and high-dose HSCb, and in hamsters exposed to high-dose HSCb. Lung inflammation and histopathology were more severe and prolonged in rats than in mice and hamsters, and both were similar in rats exposed to mid-dose HSCb and LSCb. The results show that hamsters have the most efficient clearance mechanisms and least severe responses of the three species. The results from rats also show that particle surface area is an important determinant of target tissue dose and, therefore, effects. From these results, a subchronic NOAEL of 1 mg/m3 respirable HSCb (Printex 90) can be assigned to female rats, mice, and hamsters.
Key Words: Carbon black; particle retention; species comparison; particle overload; inflammation; histopathology.
INTRODUCTION
Several studies have shown that inhaled carbon black (Cb) particles, when they are administered at doses that cause particle overload in the lungs as well as chronic inflammation and epithelial hyperplasia, induce lung tumors in rats but not in other rodent species (Driscoll et al., 1996; Heinrich et al., 1995; Mauderly et al., 1994; Rausch et al., 2004). Cb is not directly mutagenic (Kirwin et al., 1981); therefore, tumor formation in rats must occur via a secondary mechanism. Impaired clearance plays a role, but it is unlikely to be solely responsible for the observed tumors because rats, mice, and hamsters all demonstrate particle overload when the dose of Cb is high enough (Oberdrster, 1995a). Studies by Driscoll et al. (1997) demonstrated a connection between the severity of inflammation and ex vivo mutations by co-incubating lung lavage inflammatory cells from Cb-exposed rats with rat lung epithelial cells. When the percentage of neutrophils in the lavage fluid was 50%, the hprt mutation rate increased significantly, possibly related to the generation and release of reactive oxygen species and/or a depletion in antioxidants. Indeed, the mutagenic response was inhibited in the presence of an antioxidant (catalase). Thus, the persistence of particles in the lung, the overall lung burden, the severity of the inflammatory response, and the release of oxidants all likely combine to produce tumors in rats, but the reasons why mice and hamsters do not develop tumors under the same high Cb load conditions is not entirely clear.
We designed a species-comparative study to test the overarching hypothesis that rat-specific tumor responses are related to species differences in particle clearance kinetics, lung inflammation and injury, antioxidant cellular defense mechanisms, and secondary genotoxicity. In this study, rats, mice, and hamsters were exposed for 13 weeks to inhaled Cb such that equivalent lung burdens in terms of retained mass were achieved. We measured particle accumulation and clearance kinetics, cytotoxicity, inflammation, cell proliferation, lung morphology, oxidant and antioxidant production, ex vivo mutations, and genotoxicity as a function of time post-exposure. This article addresses the specific hypothesis that retention, inflammation, and lung injury as a consequence of inhaled Cb exposure are more severe and persistent in rats than in mice and hamsters, and it details the results of particle lung burden analyses and the accompanying changes in body and lung weights, lavage inflammatory cells, and cell proliferation and histopathology. The details of pro- and anti-inflammatory mediator production, oxidant stress responses, and hprt mutations will be presented in a forthcoming article.
METHODS
Animals.
All animals were obtained at 5 weeks of age and were specific pathogen–free. A total of 320 female F-344 rats were obtained from Harlan (Indianapolis, IN); 670 female B6C3F1 mice were purchased from Charles River (Wilmington, MA); and 276 F1B Syrian golden female hamsters were purchased from BioBreeders (Watertown, MA). Several extra animals from each species were exposed to allow replacement of any animals that died unexpectedly. Additional sentinels were euthanized throughout the study to monitor pathogen status. Females were chosen because they were previously shown to be more sensitive to the induction of lung tumors by poorly soluble, low toxicity particles than males (Oberdrster, 1995b). All animals were acclimatized for at least 2 weeks prior to the beginning of the studies and were uniquely identified with a subcutaneous transponder chip (BioMedic Data Systems, Seaford, DE). They were fed Purina rodent chow and water ad libitum and were housed in an AAALAC-accredited barrier facility with 12-h light/dark cycles. The hamster diet was supplemented with rolled oats (as suggested by a veterinarian at BioBreeders) during the post-exposure phase of the study to increase fiber content and avoid the development of diarrhea. Body weights were obtained every 2 weeks. Growth curves for each species are provided as Supplementary Data (see Figure S1a–c).
Study design.
Animals were randomized for placement in exposure groups (n = 5–6) using a computer-based random number generator. For the histopathological and particle dosimetry analyses, groups of six animals were used; for all other end points, groups of five animals were used. There were four nominal aerosol exposure levels (0, 1, 7, and 50 mg/m3 Printex-90; high surface area Cb [HSCb]) for all animals, as well as an additional Cb type (Sterling V, low surface area Cb [LSCb]; 50 mg/m3) to which only the rats were exposed. The HSCb exposure levels were chosen to span a range of pulmonary responses including a no-observable-adverse-effects-level (NOAEL) to lung particle overload. LSCb was used to test hypotheses about size-related or surface area–related variations in responses. The HSCb particles had a primary particle size of 14 nm with a particle surface area of 300 m2/g (Brunauer-Emmet-Teller nitrogen adsorption method); the LSCb particles had a primary particle size of 70 nm and a surface area of 37 m2/g. The aerosol concentration of LSCb was adjusted so that the predicted retained lung burden matched that of the mid-dose HSCb.
Exposures to carbon black.
Printex-90 was obtained from Degussa-Huels (Trostberg, Germany), and Sterling V was supplied by Cabot (Boston, MA). Polycyclic aromatic hydrocarbon (PAH) content was 0.039 mg/kg and 8.8 mg/kg for Printex 90 and Sterling V, respectively (Borm et al., 2005). Groups of five to six animals were exposed to filtered air or to three dose levels of HSCb: 1 (low), 7 (mid), and 50 (high) mg/m3 for 6 h/day, 5 days/week for 13 weeks. Additional groups of rats were also exposed to 50 mg/m3 LSCb. All exposures were carried out with the animals in compartmentalized, horizontal flow whole-body inhalation chambers (300 l). Each chamber can hold up to 32 rats or hamsters or up to 64 mice. For the hamster exposures, the bottom of each compartment was covered with solid plastic such that only the corners remained open. Total flow through the chambers was 100 l/min. After exposure, animals were transferred to plastic filter-top cages.
The particle-containing exposure atmospheres were generated using a Wright dust feeder (LSCb) or a Venturi jet generator (HSCb) with an AccuRate screw feeder. Aerosolized particle charges were brought to Boltzmann equilibrium by passage through a 85Kr source. Aerosol concentration was continuously monitored by a RAS-2 for the mid and high dose or by a RAM-1 (Monitoring Instruments for the Environment [MIE], Inc., Bedford, MA) for the low dose. Mass concentration and particle size were periodically measured by gravimetric filter and impactor sampling, respectively. The HSCb aggregate aerosols had aerodynamic diameters ranging from 1.2 to 2.4 μm (geometric standard deviations [GSD] = 2.0–3.1); the LSCb aggregate aerosols had aerodynamic diameters of 0.6–0.9 μm (GSD = 3.0–3.7).
Determination of carbon black lung burdens.
Groups of six animals were euthanized with an overdose of sodium pentobarbital (ip, 20 mg/100 g BW) and exsanguinated at the specified time points. Following euthanasia, the unlavaged lungs were removed and the wet weight was obtained (right lung lobes for rats, hamsters; entire lung for mice). The measurement of Cb in lung tissues was done according to the methods described by Rudd and Strom (1981). Briefly, the lung tissue was minced and digested in reagent-grade 25% KOH/methanol (w/v) at 60°C overnight. The mixture was then spun at 10,000 rpm for 30 min and the remaining pellet was mixed with reagent-grade 50% HNO3/methanol (v/v). The mixture was incubated again at 60°C overnight and then spun at 10,000 rpm for 30 min; any remaining supernate was removed by evaporation. The pellet was resuspended in surfactant water, 0.1% Brij-35 (v/v) (Sigma Chemical Co., St. Louis, MO) and sonicated. The turbidity (optical density) of the solutions was measured at 690 nm using a UV/visible spectrophotometer; samples were sonicated and re-read until a stable optical density was obtained. A standard curve was constructed and used to calculate the carbon black content of each sample. The detection limit of this assay is 0.1 μg/ml suspended solution. Retention half-times were calculated using a two-parameter monoexponential decay function (see exception for mice, Table 4).
Measurement of cellular and biochemical parameters in lavage fluid.
Groups of five animals were anaesthetized with sodium pentobarbital (ip, 20 mg/100 g BW) and exsanguinated by severing the carotid artery. A small incision was made in the trachea below the fifth cartilaginous ring, and the plastic sheath from an IV catheter was tied into the trachea. The lungs, trachea, and heart were removed en bloc and the heart and excess tissue was trimmed away. Rat lungs were lavaged 10 times with 5 ml each of sterile, pyrogen-free normal saline; mice were lavaged with 1 ml of saline; hamsters, with 4 ml of saline. The recovered volume from the first two lavages was kept separate from the others for biochemical analyses of the supernatant. The cells were pelleted at 300 x g for 12 min and then pooled from all fractions and resuspended in saline for viability determination, enumeration, and differential analysis. To obtain enough cells and lavage supernate for the cellular and biochemical analyses, lavages were combined from 3–5 mice (i.e., 15 mice lavaged per time point for low-, mid-, and high-dose HSCb; 25 per time point for controls).
A sample of the pooled lavage cells was used to assess cell number and viability (trypan blue exclusion) using a bright-line hemacytometer. A hematoxylin-eosin variant was used to stain cytocentrifuged cells for differential analysis. At least 500 cells per sample were identified to determine percentages and absolute numbers of alveolar macrophages, neutrophils, and lymphocytes. The total protein concentration and the activities of lactate dehydrogenase (LDH) and -glucuronidase in bronchoalveolar lavage (BAL) fluid were measured using commercially-available kits and reagents (Sigma Chemical Co., St. Louis, MO).
Pulmonary tissue processing for histopathology and morphometry.
Tissue from the left lung was obtained from the same rats and hamsters used for the Cb determinations (n = 6); however, whole lungs from separate groups of mice (n = 6) were used for the histopathological analyses. After death, the trachea, left extrapulmonary bronchus, and left lung lobe were removed intact from the carcass and fixed via the airway with a solution of 1% paraformaldehyde (Ted Pella, Inc., Redding, CA) and 0.1% glutaraldehyde (Electron Microscopy Sciences, Ft. Washington, PA) at a constant pressure (30 cm H2O) for 1 h. After intra-airway fixation, the lumen of the proximal bronchus was ligated, the trachea was removed, and the intact left bronchus and left lung lobe were immersed in a large volume of the same fixative for an additional 24 h.
A fractionator technique was used to ensure uniformly random sampling of the left lung lobe from all animals (Bolender et al., 1993; Geiser et al., 1990). Twenty-four hours after the start of fixation, the left lung lobe was sectioned perpendicular to its long axis at a section thickness of 2–3 mm to produce 15–20 lung sections per animal (fewer sections were obtained for mice). The first cut was made at a randomly selected position within the first 3 mm of the apical portion of the lung. Selection of the first lung section for further processing was determined using a random number (1–6). This section and every sixth section following were embedded in Immuno-Bed plastic embedding medium (Polysciences, Inc., Warrington, PA). The tissue blocks were cut at a thickness of 2–3 μm with a glass knife. For general histology, sections were stained with hematoxylin and eosin.
Cell proliferation assessment.
Cell proliferation was assessed by positive immunoreactivity for bromodeoxyuridine (BrdU; Sigma) in those animals for which there was also an increase in alveolar type II epithelial cell density. BrdU (50 mg/kg body weight) was delivered to the animals in vivo at a constant rate (10 μl/h for rats, hamsters; 1 μl/h for mice) for the 5 days preceding sacrifice by means of Alzet miniosmotic pumps (ALZA Corp., Palo Alto, CA), which were implanted subcutaneously along the dorsal mid-line. After tissue fixation and processing, lung sections were immunohistochemically stained to detect nuclear BrdU incorporation. All staining steps were carried out at 37°C unless otherwise noted. First, the sections were pre-blocked at room temperature for 10 min with 0.2 M glycine, rinsed in 1:5 automation buffer (AB; Bimeda Corp., Foster City, CA), and treated for 30 min with 0.01% proteinase K in 0.05M Tris-HCl. The slides were then rinsed in 1:5 AB, treated for 30 min with 0.2N HCl, followed by treatment with Borax buffer (13 mM Borax/48 mM boric acid, pH 8.5) for 10 min at room temperature, and rinsed again in 2x AB. The slides were then incubated for 30 min with blocking solution (1.5% normal horse serum in AB), followed by a 2-h incubation with monoclonal anti-BrdU (diluted 1:25 in blocking solution; Becton Dickinson Immunocytometry Systems, San Jose, CA). They were then incubated for 30 min with a biotinylated mouse IgG antibody (diluted 1:200 in blocking solution; Vector Laboratories, Inc., Burlingame, CA), followed by a 30-min incubation with alkaline phosphatase ABC-AP reagent (Vector) at the manufacturer's recommended concentration. To allow visualization of the BrdU-labeled nuclei, the slides were incubated for 23 min at room temperature with Vector Red Alkaline Phosphatase Substrate Solution (Vector), rinsed in deionized water, and counterstained with hematoxylin for 1 min. The slides were rinsed again in water, dipped briefly in 2x AB (to blue the hematoxylin), water, 100% ethanol, 100% xylene, and coverslipped with Permount mounting medium (Fisher Scientific, Hanover Park, IL). Cell nuclei that were immunoreactive for BrdU (i.e., were in the S phase of the cell cycle) appeared bright red, and non-S-phase nuclei were blue.
Pulmonary morphometry.
Digital images of the BrdU-stained lung sections were captured with an Olympus 3-CCD color camera (Olympus America, Inc., Melville, NY) and analyzed with Scion Image software (v1.62; Scion Corp., Frederick, MD). The first image field for quantitative analysis was chosen randomly (1–5), and every fifth field following was quantified at a magnification of 790x. Only fields completely filled by lung parenchyma were analyzed (fields with conducting airways or large blood vessels were omitted). Three slides were analyzed for each animal, and 20–30 fields were analyzed per slide. Macrophages, neutrophils, and type II alveolar epithelial cells were identified morphologically and evaluated for positive BrdU staining. Cell numbers were expressed per mm2 of total alveolar parenchyma.
Statistical analyses.
Results were analyzed for statistically significant differences by two-way analysis of variance (ANOVA) with appropriate data transforms followed by Tukey t-tests using SigmaStat (SPSS Science, Chicago, IL). The two factors for the ANOVAs were exposure dose and time. Data were appropriately transformed (e.g., base 10 logarithm, natural logarithm) if an analysis of residuals suggested deviations from the assumptions of normality and equal variance. Comparisons were considered statistically significant when p 0.05.
RESULTS
Stability of Carbon Black Concentration in Exposure Chambers
There were two exposure chambers for each dose level of carbon black; the aerosol concentration did not differ significantly between the two chambers for any of the exposure levels or for any of the three species and are thus presented as single means in Table 1. For the exposures of the rats, the aerosol concentrations were quite stable around the targeted doses. Five weeks into the exposures of mice and hamsters, it was determined that the retained lung burden of HSCb (normalized to lung weight) was lower than was found in rats, which were exposed first. The mid- and high-dose levels for the mouse and hamster exposures were accordingly increased from 7 to 15 mg/m3 for the mid-dose and from 50 to 75 mg/m3 for the high dose for the remaining exposure period to produce equivalent predicted normalized lung burdens. For exposures in hamsters, the low dose also had to be increased from 1 to 1.1 mg/m3. Thus, the average chamber concentrations (for mice and hamsters) are higher than the original targets and the standard deviations are also higher. As can be seen from the mass median aerodynamic diameters (MMAD, 0.8–2.0 μm) and GSDs (2.3–3.2), the aerosols were respirable and polydispersed. The average aggregate aerosol size in the low- and high-dose mouse exposure chambers was slightly higher than in the others for reasons unknown, but that did not affect the aim of achieving low, mid, and high Cb lung burdens.
Effects of Exposure on Animal Body and Lung Weights
No rats died during exposure to carbon black. Nine control rats died in the post-exposure phase due to a blocked water line. One each in the low- and high-dose HSCb groups also died in the post-exposure phase. Thirteen mice died during exposure, seven of which were in the control group; two were in the low-dose group, three in the mid-dose group, and one in the high-dose group. In the post-exposure phase, six controls, one low-dose, two mid-dose, and six high-dose mice died. One hamster (high-dose group) died during exposure. Two hamsters each from the control, low-dose, and high-dose groups and four from the mid-dose group died in the post-exposure phase of the study. These deaths were most likely due to a change in housing conditions from wire-bottom cages (during exposure) to plastic cages (post-exposure) and reflects the heightened sensitivity of hamsters to environmental changes. Screening of sentinels (all three species) did not reveal parasitic, bacterial, or viral pathogens that would explain the unscheduled deaths. We conclude that unscheduled deaths were not related to exposure (i.e., not associated with Cb dose or time after exposure). Indeed, out of the total number of animals exposed, 4% of the animals per species died prematurely.
Body weights are summarized according to post-exposure time points in Table 2 (see also Figure S1). Rats and hamsters from the high-dose group lost weight during exposure, but this was only significant for the hamsters at the end of exposure. The differences probably reflect decreased weight gain instead of actual losses. Body weights returned to control values during the recovery period for both species (see Figure S1 a,c). There were no significant changes in body weight during exposure for other exposure doses or in the mice. In the post-exposure phase, we found that the hamsters began to lose weight after being placed in plastic cages, whereas during exposure, they gained weight faster than the sentinels (housed in plastic cages for the duration of the study). This again highlights the sensitivity to environmental change in this species.
For the rats, significant elevations in lung weight (up to more than twice the control lung weight) were found for the high-dose HSCb at the end of exposure and at all post-exposure time points (Table 3). The lung weights of rats exposed to LSCb were between those for the high- and mid-dose HSCb in magnitude and were significantly elevated 1 day and 3 months after exposure. In mice, lung weights were found to be elevated 1 day and 3 months post-exposure in the high- and mid-dose groups; at the end of the recovery period (i.e., 11 months post-exposure), significant elevations were found only in the high-dose group. In hamsters, significant elevations in lung weights were found only for the high-dose group, and these changes persisted up to 3 months post-exposure; after 11 months of recovery, no significant differences were found. Thus, the changes in lung weight were resolved most rapidly in hamsters.
Carbon Black Lung Burden
Figure 1a and b shows the carbon black lung burden and retention in rats. Lung burden was also expressed in terms of particle surface area dose (Fig. 2) because particle types of two different surface areas were used in the rat exposures, and the goal was to achieve the same lung burden in terms of surface area as with the mid-dose HSCb. The highest lung burdens in terms of mass were achieved in the high-dose HSCb and the LSCb groups. The mass burden due to LSCb (50 mg/m3) exposure was slightly but significantly higher than in the high-dose HSCb (50 mg/m3)–exposed rats at the end of exposure; the mass burdens were equivalent between the two groups thereafter. Both of these groups showed prolonged particle retention, as did the mid-dose HSCb rat group; LSCb was more efficiently cleared from the lungs than was high-dose HSCb. The mass burden values for high-dose HSCb-exposed rats were higher during recovery than at the end of exposure, which was most likely due to chance given that there was an n of 6 for each post-exposure time point. Clearly, this does not represent any further respiratory tract exposure, as rats were held in a separate holding room after the 13-week exposure; as can be seen from Figure 1b, there was no significant particle clearance in these rats. The low-dose group had the fastest clearance. However, the retention half times in rats for the mid and high doses of HSCb and for LSCb suggested that normal clearance mechanisms were overwhelmed (Table 4). In terms of surface area dose, the highest lung burdens by far were achieved in the high-dose HSCb-exposed rats (Fig. 2). The lung particle burden of LSCb-exposed rats, expressed as particle surface area, was similar to that of the mid-dose HSCb-exposed rats.
In mice, significant elevations in lung particle burdens were found during exposure and throughout recovery in all dose groups (Fig. 3a). The mice had the highest relative lung burdens of the three species when the exposures were finished, in part because the aerosol concentrations in the mid- and high-dose groups were over-adjusted after 5 weeks of exposure. Despite this, the mice cleared the particles faster at the highest dose (retention t1/2 = 322 days) than the rats (no clearance), so that by 11 months post-exposure the relative lung burdens were 48% versus 118% of end-exposure values. Although the retention as a percentage of end-exposure mass burden was lowest in the high-dose HSCb-exposed mice, clearance slowed over the 11-month recovery period such that retention in both the mid- and high-dose groups was higher than the low-dose group (Fig. 3b). The low- and mid-dose–exposed mice had very similar particle retention half times, whereas the high-dose group exhibited prolonged clearance (Table 4).
In hamsters, significant elevations in lung carbon black burden were found for all doses and at all time points (Fig. 4a). The particle retention as a percentage of mass burden at 1 day post-exposure in the high-dose group was higher than in the mid- and low-dose groups for all post-exposure times (Fig. 4b). The retention half times for the low- and mid-dose groups of the hamsters were very similar and were the lowest of the three species. The retention half-time for the high-dose group (309 days) was similar to that of the high-dose mice. As was also true for mice, hamsters exposed to high-dose HSCb exhibited impaired clearance (Table 4).
Cellular and Biochemical Parameters in Lavage Fluid
Summaries of all cellular and biochemical measurements in BAL fluid are presented for rats, mice, and hamsters, respectively, in Table S1 through Table S3. Figures 5 and 6 summarize the total cell number and percent PMN data, respectively. Total BAL cell numbers were elevated at 1 day post-exposure and all through the recovery period in the high-dose groups for all three species in comparison to all other exposure groups. Whereas total cell number gradually decreased over time in mice and hamsters, it stayed elevated in the rats exposed to high-dose HSCb and LSCb (Fig. 5). Total cell numbers in the controls did not change over time in mice and hamsters, whereas, in rats, there was a small decrease; the mean control values in chronological order were 1.85 x 107 ± 0.61, 1.15 x 107 ± 0.21, and 1.01 x 107 ± 0.14 (rats); 0.90 x 106 ± 0.09, 1.22 x 106 ± 0.07, and 1.02 x 106 ± 0.23 (mice); 1.30 x 107 ± 0.19, 1.56 x 107 ± 1.61, and 1.64 x 107 ± 0.29 (hamsters). The magnitude of the response to LSCb in rats was between those for high- and mid-dose HSCb. The mid-dose groups in rats and mice also exhibited significant elevations in total cell numbers after exposure, but recovery occurred before the study was terminated.
The presence of polymorphonuclear leukocytes (PMN) in lavage fluid was used as a sensitive indicator of lung inflammation. PMN were elevated in the mid- and high-dose HSCb and LSCb groups of rats at the end of exposure and throughout the recovery period. In rats, the PMN responses to high-dose HSCb and LSCb were similar (Fig. 6a; see also Table S1). The PMN remained elevated in rats from the high-dose HSCb and LSCb groups through the end of the study. At the end of exposure in mice and hamsters (Fig. 6b,c), the mid- and high-dose groups were also elevated compared to the controls and the low-dose group; the magnitudes of response were similar for the two species. During the recovery period, the percentage of PMN from the mid- and high-dose mice remained different from each other and the rest of the exposure groups. In hamsters, the PMN response decreased between 3 and 11 months post-exposure such that only the high-dose group retained a significant response at the termination of the study. The magnitude of the PMN response was the highest in rats and was the most prolonged of the three species. Thus, it appears that a maximal response was reached in mice and hamsters, but not in rats, with increasing mass dose of HSCb. The mean percentages of PMN in controls in chronological order were 0.47 ± 0.28%, 0.15 ± 0.16%, and 0.46 ± 0.23% (rats); 0.35 ± 0.26%, 0.23 ± 0.16%, and 0.67 ± 0.44% (mice); and 0.71 ± 0.15%, 0.81 ± 0.19%, and 2.51 ± 2.51% (hamsters). The concentration of total protein and activities of LDH and -glucuronidase in BAL fluid exhibited significant elevations in all three species (see Tables S1–S3) that were strongly correlated with changes in the PMN response. There were no significant elevations in cellular or biochemical parameters for any of the animals exposed to low-dose HSCb.
Rat Histopathology
No exposure-related alterations were observed microscopically at any of the post-exposure time points in the lungs of control rats (Fig. 7a). Likewise, no adverse histological effects of exposure were found at any post-exposure time point in the lungs of rats exposed to low-dose HSCb (not shown). The lungs of these rats were similar to those of the control rats, with the exception that some alveolar macrophages containing small amounts of Cb were widely scattered throughout the alveolar air spaces of the pulmonary parenchyma. However, the number and morphologic character of these particle-containing macrophages were not different from those in the controls.
Exposure-related histopathology was present in the lung lobes of rats exposed to mid- and high-dose HSCb, but the most severe alterations were present in the rats exposed to high-dose HSCb (Fig. 7b–d). At 1 day post-exposure, the principal exposure-related lung lesions were centered primarily in the centriacinar regions of the lungs with the most extensive epithelial and inflammatory responses in the alveolar ducts and the immediately surrounding alveolar parenchyma. These lesions were characterized by (1) the accumulation of large numbers of markedly hypertrophic and highly vacuolated alveolar macrophages laden with HSCb; (2) an associated mixed inflammatory cell influx composed of neutrophils and lesser numbers of mononuclear cells located in the alveolar walls, alveolar air spaces, and perivascular interstitium; and (3) hyperplasia and hypertrophy of alveolar type II cells (Fig. 7b). Many of the particle-laden macrophages were degenerative or necrotic indicating particle-mediated cytotoxicity. Cellular debris and free HSCb were also found in the affected alveolar air spaces. Though most of the HSCb was found within alveolar macrophages, lesser amounts were present in alveolar epithelium and interstitium of the affected parenchyma and in the bronchus-associated lymphoid tissue in the upper conducting airways (axial and preterminal airways). A mild to moderate alveolar proteinosis was also a consistent finding in the affected regions of the lung. Interestingly, these lesions were also evident in high-dose HSCb-exposed rats that were sacrificed 3 and 11 months post-exposure (Fig. 7c,d). There was only mild attenuation in the severity of the inflammatory and epithelial lesions at these later time points compared to those observed at 1 day post-exposure.
Interstitial fibrosis was not present in the lungs of rats sacrificed at 1 day post-exposure, but a mild, widely scattered interstitial fibrosis of the alveolar septa was present in the HSCb-exposed rats that were sacrificed 13 weeks after the end of the exposure. This particle-induced, multifocal fibrosis was markedly greater in both extent and severity in the lungs of rats exposed to the high concentration of HSCb and sacrificed 11 months post-exposure. Varying sized foci of alveolar, peribronchiolar, perivascular, and pleural fibrosis were scattered throughout the lungs of these rats, but they were always associated with HSCb-laden degenerative macrophages, neutrophilic inflammation, and alveolar epithelial proliferation. Conspicuous aggregates of free and phagocytized HSCb particles were often scattered throughout the areas of fibrosis. In addition, small foci of epithelial metaplasia composed of columnar ciliated and nonciliated cells lining alveolar ducts were commonly associated with the fibrotic areas of the affected centriacinar regions.
The pulmonary lesions in the rats exposed to the mid dose of HSCb and sacrificed 1 day post-exposure were generally similar in character but markedly less severe and extensive than those in the lungs of rats exposed to high-dose HSCb (Fig. 8). The major differences in these lung lesions compared to those in the high-dose rats, were (1) less involvement of the distal alveolar parenchyma (lesion mainly focused in the proximal alveolar ducts), (2) less degeneration of particle-laden macrophages, and (3) minimal associated inflammation and alveolar epithelial changes in the affected regions of the lung. There were focal accumulations of alveolar macrophages that were heavily laden with Cb particles in the centriacinar regions of the lung, but macrophages with microscopic features of degeneration and necrosis (e.g., cytoplasmic vacuolization, nuclear pyknosis, severe hypertrophy, cellular fragmentation) were infrequent. Widely scattered small aggregates of Cb particles were still present in the lungs of rats exposed to the mid dose at 3 and 11 months post-exposure. However, there was no associated influx of neutrophils or alveolar type II cell hyperplasia at these later time points post-exposure. Interestingly, there were more Cb particles in peribronchial and perivascular interstitial regions of the lung in the mid-dose HSCb-exposed rats sacrificed at 3 and 11 months post-exposure than in similarly exposed rats sacrificed at 1 day post-exposure. However, no particle-induced fibrosis was evident in any of these rats exposed to mid-dose HSCb.
Rats exposed to LSCb that were sacrificed 1 day post-exposure had similar pulmonary lesions as those in the rats exposed to mid-dose HSCb, but their lung lesions were greater in both severity and extent (Fig. 8i,j). These animals had a greater number of degenerative alveolar macrophages, inflammatory cells (neutrophils), and alveolar type II cells (hyperplasia) than the rats exposed to mid-dose HSCb, but the numbers of these cells were considerably fewer than those found in the lungs of rats exposed to high-dose HSCb. Rats exposed to LSCb and sacrificed 3 and 11 months after the end of exposure had lung lesions, including interstitial fibrosis, that were similar in character but that appeared to be slightly less in extent and severity than those observed in rats exposed to high-dose HSCb.
Quantitation of Pulmonary Morphometric Analyses in Rats
Numeric cell densities of alveolar type II cells in the alveolar parenchyma of Cb-exposed rats are presented in Table 5; those for neutrophils and alveolar macrophages/monocytes are shown in the Supplementary Data (see Tables S4 and S5, respectively). Compared to controls, rats exposed to HSCb had a dose-dependent increase in the numeric densities of all the measured cell types at 1 day post-exposure. There was a 129% increase in the numeric density of alveolar type II cells in rats exposed to high-dose HSCb (Table 5) as compared to controls. Approximately 8% of the type II cells in the animals exposed to high-dose HSCb had nuclear labeling for BrdU, indicating a significant increase in DNA synthesis compared to that in controls (1% labeling index). However, no HSCb-induced increase in alveolar type II cell number was found in rats exposed to the low or mid dose of HSCb compared to controls. At 3 and 11 months post-exposure, rats exposed to high-dose HSCb, but not to the low or mid doses, had significantly more alveolar type II cells than the control rats (156% and 98% more, respectively). In addition, the BrdU labeling index remained high in these epithelial cells at both 3 and 11 months post-exposure (7% and 12% respectively, compared to 1–3% in controls). Rats exposed to LSCb had similar increases in the numeric cell density of alveolar type II cells as compared to high-dose HSCb-exposed rats (106%, 135%, and 53% at 1 day and 3 and 11 months, respectively). The labeling index for BrdU in alveolar type II cells was also significantly elevated above that of controls in LSCb-exposed rats sacrificed at all post-exposure time points (7%, 3%, and 8% at 1 day and 3 and 11 months, respectively; Table 5).
Only those rats exposed to high-dose HSCb and LSCb had significant increases in the numeric density of alveolar neutrophils compared to controls at 1 day post-exposure (see Table S4). However, significantly fewer neutrophils were present in the rats exposed to LSCb (90 ± 20 cells/mm2) than in those exposed to high-dose HSCb (148 ± 13 cells/mm2 of parenchyma) at 1 day post-exposure. Interestingly, Cb-induced increases in alveolar neutrophils were maintained in both the high-dose HSCb- and LSCb-exposed rats at 3 and 11 months post-exposure. Compared to controls, there were 283% and 222% more neutrophils in rats exposed to high-dose HSCb and LSCb, respectively, at 11 months post-exposure. Carbon black–induced increases in the numeric cell density of alveolar macrophages were evident in rats exposed to the mid and high doses of HSCb and to LSCb at 1 day and 3 months post-exposure (see Table S5). At 11 months post-exposure, only the rats exposed to high-dose HSCb or LSCb had significantly more macrophages than the control rats. The densities of macrophages were not significantly different between rats exposed to high-dose HSCb and LSCb at any post-exposure time point.
Mouse Histopathology
No exposure-related alterations were observed microscopically at any post-exposure time point in the lungs of control mice (Fig. 9a); neither were any adverse histological effects of exposure found in the lungs of mice exposed to low-dose HSCb. The lungs of these mice were microscopically similar to those of the control mice, except that some alveolar macrophages containing small amounts of Cb were widely scattered throughout the pulmonary parenchyma. The number, size, and morphologic character of these particle-containing macrophages, however, were not different from those in the control mice.
In contrast, exposure-related lung lesions were present in mice exposed to mid- and high-dose HSCb and sacrificed at 1 day and 3 and 11 months post-exposure. Though the lesions were similar in these two exposure groups, the magnitude of the lesions (severity) was consistently greater in mice exposed to high-dose HSCb (Fig. 9b–d). The most conspicuous change in the lungs of these mice was the presence of numerous Cb particles that were widely scattered throughout the alveolar parenchyma, but more concentrated in centriacinar regions including alveolar ducts and adjacent alveoli. Aggregates of particles were either free in the alveolar air spaces or within enlarged alveolar macrophages. Though many of the alveolar macrophages were engorged with particles, few of these phagocytic cells had obvious features of cellular degeneration or necrosis like those commonly found in rats exposed to high-dose HSCb. Most of the particles were present in alveolar lumens, but much smaller numbers were present in macrophages within interstitial tissues surrounding terminal bronchioles or small blood vessels. At 3 and 11 months, but not 1 day, post-exposure, there was also an obvious increase in the numbers of alveolar macrophages in the lung parenchyma and a mild mononuclear cell infiltrate (mainly lymphocytes, monocytes and plasma cells) within the perivascular and peribronchiolar interstitium (Fig. 9c,d). In contrast to the lungs of rats exposed to HSCb, there were very few, if any, neutrophils and no conspicuous alveolar type II cell hyperplasia in the Cb-laden centriacinar and alveolar regions at 1 day post-exposure. Alveolar type II cell hyperplasia was present in mice exposed to high-dose HSCb and sacrificed 3 months post-exposure. However, this proliferation of type II cells did not persist in similarly exposed mice that were sacrificed 11 months after exposure. Interestingly, alveolar proteinosis was also a characteristic feature of all mice exposed to high-dose HSCb, as was seen in rats. This alveolar lesion was evident at all post-exposure time points. However, none of the mice developed alveolar septal fibrosis like that observed in the HSCb-exposed rats at 3 and 11 months post-exposure.
Quantitation of Pulmonary Morphometric Analyses in Mice
In contrast to rats, no Cb-exposed mice had numeric densities of alveolar type II cells that were significantly different from control mice at 1 day or 11 months post-exposure (Table 5). Only mice exposed to high-dose HSCb and sacrificed 3 months post-exposure had a significant increase (181%) in the numbers of alveolar type II epithelial cells compared to the control mice. There were no HSCb-related increases in neutrophils in the alveolar parenchyma of mice at 1 day or 3 months post-exposure. At 11 months post-exposure, there was a small but statistically significant increase in the number of neutrophils in the lungs of mice exposed to low- and mid-dose, but not high-dose, HSCb compared to that in the lungs of control mice sacrificed at the same time point (see Table S4). Modest exposure-related increases in alveolar macrophage numbers (2.7–3.5x those of controls) were only evident in mice exposed to high-dose HSCb and sacrificed 3 or 11 months post-exposure and to mid-dose HSCb and sacrificed 11 months post-exposure (see Table S5).
Hamster Histopathology
No exposure-related alterations were observed microscopically in the lungs of control (Fig. 10a) or low dose HSCb-exposed hamsters at any post-exposure time point. The lungs of low-dose HSCb-exposed hamsters (like those of similarly exposed rats and mice) were microscopically similar to those of the control animals, with the exception of some alveolar macrophages containing small amounts of Cb that were widely scattered throughout the pulmonary parenchyma. The number, size, and morphologic character of alveolar macrophages in these rodents were not different from those in the control hamsters. In contrast, hamsters exposed to the mid dose and the high dose of HSCb had widely scattered aggregates of enlarged alveolar macrophages and a few multinucleated giant cells engorged with phagocytized HSCb particles at all post-exposure time points (Figure 10 b-d). Most of these HSCb-laden macrophages were present in alveolar air spaces in or near centriacinar regions of the lungs. There was an increase in free and phagocytized particles in the interstitial spaces of alveolar ducts, terminal bronchioles, and blood vessels at the later time points. The number of widely scattered aggregates of HSCb-laden macrophages decreased with time, but it was consistently greater in the hamsters exposed to high-dose HSCb as compared to mid-dose HSCb. Associated with the particle-laden macrophages in the lungs of hamsters exposed to high-dose HSCb, there was a mild type II cell hyperplasia only at 1 day post-exposure. This alveolar epithelial lesion was not present in the mid- or high-dose–exposed hamsters. In addition, there was a conspicuous absence of neutrophilic inflammation, alveolar proteinosis, and septal fibrosis in the lungs of all of the hamsters.
Quantitation of Pulmonary Morphometric Analyses in Hamsters
At one day post-exposure, hamsters exposed to high-dose HSCb had significantly greater numbers of type II cells (Table 5) and alveolar macrophages (see Table S5) in alveolar parenchyma. These hamsters had 27% more alveolar type II cells and 62% more macrophages than controls. However, there was no increase in the numeric cell density of neutrophils above controls in the HSCb-exposed hamsters (see Table S4). Hamsters exposed to mid-dose HSCb and sacrificed 1 day post-exposure had a mild increase in the number of alveolar macrophages above controls (1.5x), but no increases in the number of neutrophils or alveolar type II cells were observed in the lung parenchyma. Hamsters exposed to low-dose HSCb had no significant differences in the numeric cell densities of alveolar type II cells, neutrophils, or alveolar macrophages at 1 day post-exposure as compared to controls. At 3 and 11 months post-exposure, high-dose HSCb-exposed hamsters had persistent increases in the numeric cell density of alveolar macrophages compared to control hamsters (53% and 38% more, respectively; see Table S5). There were no increases in alveolar type II cells or neutrophils at these later time points (Table 5; see also Table S4). Numeric cell densities for hamsters exposed to low- and mid-dose HSCb and sacrificed at 3 and 11 months post-exposure were not determined because there were no morphologic indications of increases in these cell types in the lungs of hamsters.
DISCUSSION AND CONCLUSIONS
The International Agency for Research on Cancer (IARC) has classified Cb as possibly carcinogenic to humans (Group 2B; IARC, 1996). This evaluation was based on inadequate evidence in humans, and sufficient evidence in experimental animals. In the case of the human epidemiological evidence, the U.S. studies on cancer in carbon black workers showed no excess of lung cancer (Robertson and Ingalls, 1980, 1989; Robertson and Inman, 1996). A UK study of Cb production workers showed a marginally significant increase in lung cancer (Hodgson and Jones, 1985). The IARC, in its 1996 evaluation, noted a variety of methodological shortcomings in the studies they reviewed, which precluded a definitive evaluation of the epidemiological findings. In a follow-up investigation of the Hodgson and Jones study, Sorahan et al. (2001) confirmed this increase in lung cancer, but they also demonstrated that it was not related to cumulative Cb exposure and suggested that it may be related to other factors.
The present study was designed to determine whether particle retention, lung inflammation, and pathology induced by subchronic inhalation of Cb are more pronounced in rats than in mice and hamsters, which could contribute to greater secondary genotoxic events in rats. A greater and more persistent inflammatory and related oxidative stress response in the rat lung is a likely mechanism by which high lung burdens of particles such as Cb can induce secondary genotoxicity resulting in epithelial cell mutations and subsequent lung tumor formation. Lung tumors induced by poorly soluble particles of low cytotoxicity (PSP) due to a condition of "particle overload" in the lung have only been observed in rats following chronic high-level exposures, but not in mice or hamsters. Thus, the overall study focused on a comparison between three rodent species of particle retention kinetics and pulmonary inflammatory, oxidative stress and genotoxic events; the present article reports on particle uptake and retention as well as Cb-induced lung inflammation and pathology.
Lung particle overload is a concept that applies only to PSP and was defined by Morrow (1988) in terms of retained particle volume as a state that arises when macrophage-mediated clearance mechanisms are overwhelmed. Evidence was summarized showing that retention kinetics are altered when mass burdens reach 1–3 mg particles/g lung tissue. According to this definition, particle overload was achieved in the present study in rats exposed to mid- and high-dose HSCb and LSCb, in mice exposed to the mid dose and the high doses, and in hamsters exposed to the high dose of HSCb. Further evidence in support of overload is demonstrated in the prolonged retention half times (Table 4) for rats (except the low dose) and for the high doses in mice and hamsters (assuming normal retention half times of 70 days for rats and 55 days for mice; Kreyling, 1990; Oberdrster, 1995a). The clearance kinetics and surface area burden of LSCb were similar to that of mid-dose HSCb, indicating the importance of particle surface area in particle retention. These data support a suggestion made by Oberdrster and Yu (1990) and, later, by Tran et al. (2000), regarding particle surface area–related effects of PSPs.
Several chronic and sub-chronic exposure studies on the effects of PSP have been conducted, one of the most recent ones by Bermudez et al. (2004), in which three rodent species were exposed to ultrafine titanium dioxide (uf-TiO2). There are similarities between uf-TiO2 and HSCb in terms of primary particle size (21 [uf-TiO2] versus 17 nm [HSCb]) and aerodynamic diameter (1.3–1.5 μm [uf-TiO2] versus 1.4–2.0 μm [HSCb]). Thus, it could be postulated that the two particle types would deposit and be retained similarly in the lungs and that they might produce similar effects. Considering lung burden in rats at the end of the exposure period (which was 13 weeks in the Bermudez study and in the present study) as an example, it is clear that the mass dose does not adequately predict outcome: 50 mg/m3 HSCb produced a mass burden (mg/g wet lung tissue) that was 2–3 times lower than 10 mg/m3 uf-TiO2. A suggestion from earlier studies (Oberdrster et al., 1994) is that particle surface area is a better predictor of lung responses to inhaled PSP than mass dose. Using the surface areas of each particle type (and correcting for wet versus dry tissue weights), the lung burdens for 10 mg/m3 uf-TiO2 and 7 mg/m3 HSCb are more alike (0.1 and 0.3 m2/g lung, respectively), whereas that for 50 mg/m3 HSCb is the highest (0.9 m2/g lung) at the end of exposure. The surface area burden for 50 mg/m3 LSCb is 0.2 m2/g lung. As reported by Bermudez et al. (2002, 2004), the surface area dose for 10 mg/m3 uf-TiO2 was similar to that for 50 mg/m3 pigmentary (p) TiO2. The following, then, is a set of four different PSPs that produce similar surface area tissue burdens: 50 mg/m3 LSCb, 7 mg/m3 HSCb, 10 mg/m3 uf-TiO2, and 50 mg/m3 p-TiO2. If surface area dose were the best predictor of outcome, then the retention kinetics and inflammatory and morphological changes should be similar for the dose levels given above for the four particle types. This is, however, not quite the case: the retention half times are somewhat shorter, the percentages of PMNs in BAL fluid are lower, and the histopathological changes are less severe for the two Cb particle types than for the TiO2.
the data presented here, as well as the preceding discussion, we propose that particle surface area dose may be the best metric to predict responses to PSPs as compared to other parameters, e.g., mass dose. Driscoll (1996) compiled data from several PSP chronic inhalation studies in rats and showed that the lung tumor response was better described by the retained particle surface area dose than by mass dose. For the present study, there are parameters other than surface area that could explain the differences in response that we observed in rats exposed to HSCb and LSCb. For one, the Cb particles were produced by different processes, possibly resulting in trace contaminant differences that could also contribute to the responses we observed. It is also possible that differences in the way inhaled aggregated aerosols disaggregate in the lungs could affect response, as suggested by Ferin et al. (1992), as well as differences in rates of interstitialization. Lastly, the PAH content of the LSCb was much higher (225x) than for the HSCb (8.8 versus 0.039 mg/kg). Given the fact that responses to LSCb were often between those for mid- and high-dose HSCb in magnitude, it is possible that the higher PAH content of LSCb is one explanation. Although recent work by Borm et al. (2005) showed that neither HSCb nor LSCb at the end of subchronic exposure induced DNA-PAH adduct formation in the lung (32P post-labeling analysis of lung samples from our study), it was also shown that LSCb did induce PAH adducts after acute in vitro exposure of A549 cells. This was not found after similar in vitro exposures with HSCb. These results indicate that DNA repair processes may have occurred in vivo in LSCB-exposed rats during and/or after subchronic inhalation exposure. Whether the greater PAH content of LSCb may have contributed to the observed effects in our study cannot be firmly determined from our results. We suggest, therefore, that PSP surface area dose plays a substantial role in determining outcome for PSPs of similar chemical composition and crystal structure, but that other factors may also contribute.
The level of inflammation found in the present study, as determined by the percentage of BAL PMNs, was comparable to that in another study conducted by Driscoll et al. (1996) to assess the inflammatory and mutagenic potential of Monarch 880 Cb particles (primary particle size, 16 nm; aerodynamic diameter, 0.88 μm; SA, 220 m2/g) in rats. Although the lung burdens in terms of total mass per lung in our study were not as high as reported by Driscoll, clearance impairment was observed in both studies. The present study extends these observations by performing the same panel of analyses in each of three rodent species as well as by exposing rats to a low surface area Cb.
The results of this comparative study in rats, mice, and hamsters confirm previous findings that the rat is the most sensitive of the three species with respect to PSP-induced adverse effects in the respiratory tract (Heinrich et al., 1995; Oberdrster et al., 1994). Retention of Cb in rats was longest, and pulmonary inflammation and histopathological changes were more severe and more persistent than in mice and hamsters. In addition, companion studies found that the generation of oxidants (superoxide anion, hydrogen peroxide, nitric oxide) by lavaged cells and mutational changes (hprt locus) were higher in rats than in mice and hamsters (Carter et al., 2003). Among other parameters, antioxidant levels (lung tissue and lavage fluid levels of superoxide dismutase and glutathione peroxidase) were also assessed. These indicators of oxidant/antioxidant balance, as a whole, suggested that rats experienced greater oxidant load in response to HSCb as compared to mice and hamsters. The results of these studies will be presented in a forthcoming article. Furthermore, progressive dose-related oxidative DNA damage in rat lung tissue was reported by Gallagher et al. (2003).
In summary, the results presented here demonstrate the importance of particle surface area in governing the retention kinetics of inhaled Cb particles as well as the subsequent inflammatory and morphological changes in rats. Specifically, a Cb with eightfold lower surface area (LSCb) exhibited similar mass burdens as high-dose HSCb, but it was cleared in a similar fashion and had similar toxicological effects as compared to mid-dose HSCb. Although we did not design our study to determine a surface area dose threshold, as that would have involved vastly more doses and data points we can estimate the surface area to be 0.03 m2/g lung for HSCb and LSCb in rats based on the PMN response. The responses to Cb particles inhaled for 13 weeks demonstrated clear relationships to dose and recovery time and, as other investigators have previously reported, we also found that inhaled Cb delivered at a high dose overloads the normal clearance mechanisms of rat, mouse, and hamster lungs. The rank order for Cb particle retention, irrespective of dose, is rat >>> mouse, hamster. Similarly, the order of sensitivity to Cb particles as found in this study is rats > mice > hamsters. No adverse inflammatory or morphological changes were observed at the lowest exposure concentration (1 mg/m3) of Cb in any of the three species. Interestingly, for the low- and mid-dose HSCb groups, the responses among the three species were similar in magnitude. Lastly, a subchronic NOAEL of 1 mg/m3 respirable HSCb (Printex 90) can be assigned to female rats, mice, and hamsters from these results.
SUPPLEMENTARY DATA
Supplementary data are available online at www.toxsci.oxfordjournals.org:
Table S1. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black–exposed rats. All exposure doses, particle types, and post-exposure evaluations are shown; Table S2. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black–exposed mice. All exposure doses and post-exposure evaluations are shown; Table S3. Cellular and biochemical parameters in bronchoalveolar lavage fluid from carbon black-exposed hamsters. All exposure doses and post-exposure evaluations are shown; Table S4. Numeric densities of neutrophils in the lungs of rats, mice, and hamsters following 13 weeks of exposure to carbon black; Table S5. Numeric densities of macrophages/monocytes in the lungs of rats, mice, and hamsters following 13 weeks of exposure to carbon black; Figure S1. Growth curves for rats (a), mice (b), and hamsters (c) exposed to carbon black for 13 weeks. The data shown cover the exposure and 11-month recovery periods. Green circles, controls; pink circles, low dose HSCb; blue circles, mid-dose HSCb; red circles, high-dose HSCb; and black circles, LSCb (rats only, panel a).
ACKNOWLEDGMENTS
The authors thank Nancy Corson, Kiem Nguyen, Pamela Wade-Mercer, Lori Bramble, and Drs. James Wagner and Jon Hotchkiss for their excellent technical assistance. This work was supported by a grant from the International Carbon Black Association and by a National Institute of Environmental Health Sciences (NIEHS) Environmental Health Sciences Center grant (P30 ESO1247).
REFERENCES
Bermudez, E., Mangum, J. B., Asgharian, B., Wong, B. A., Reverdy, E. E., Janszen, D. B., Hext, P. M., Warheit, D. B., and Everitt, J. I. (2002). Long-term pulmonary responses of three laboratory rodent species to subchronic inhalation of pigmentary titanium dioxide. Toxicol. Sci. 70, 86–97.
Bermudez, E., Mangum, J. B., Wong, B. A., Asgharian, B., Hext, P. M., Warheit, D. B., and Everitt, J. I. (2004). Pulmonary responses of rats, mice, and hamsters to subchronic inhalation of ultrafine titanium dioxide particles. Toxicol. Sci. 77, 347–357.
Bolender, R. P., Hyde, D. M., and Dehoff, R. T. (1993). Lung morphometry: a new generation of tools and experiments for organ, tissue, cell, and molecular biology. Am. J. Physiol. 265, L521–L548.
Borm, P. J. A., Cakmak, G., Jermann, E., Weishaupt, C., Kempers, P., van Schooten, F. J., Oberdrster, G., and Schins, R. P. F. (2005). Formation of PAH-DNA adducts after in vivo and in vitro exposure of rats and lung cells to different commercial carbon blacks. Toxicol. Appl. Pharmacol. 205, 157–167.
Carter, J. M., Oberdrster, G., Elder, A. P., Singh, K. A., and Driscoll, K. E. (2003). Species differences in response to the effects of inhaled carbon black particle overload in the lung. Am. J. Respir. Crit. Care Med. 167, A492.
Driscoll, K. E. (1996). Role of inflammation in the development of rat lung tumors in response to chronic particle exposure. Inhal. Toxicol. 8, 139–153.
Driscoll, K. E., Carter, J. M., Howard, B. W., Hassenbein, D. G., Pepelko, W., Baggs, R. B., and Oberdrster, G. (1996). Pulmonary inflammatory, chemokine, and mutagenic responses in rats after subchronic inhalation of carbon black. Toxicol. Appl. Pharmacol. 136, 372–380.
Driscoll, K. E., Deyo, L. C., Carter, J. M., Howard, B. W., Hassenbein, D. G., and Bertram, T. A. (1997). Effects of particle exposure and particle-elicited inflammatory cells on mutation in rat alveolar epithelial cells. Carcinogenesis 18, 423–430.
Ferin, J., Oberdrster, G., and Penney, D. P. (1992). Pulmonary retention of ultrafine and fine particles in rats. Am. J. Respir. Cell Mol. Biol. 6, 535–542.
Gallagher, J., Sams, R., Inmon, J., Gelein, R., Elder, A., Oberdrster, G., and Prahalad, A. K. (2003). Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine in rat lung DNA following subchronic inhalation of carbon black. Toxicol. Appl. Pharmacol. 190, 224–231.
Geiser, M., Cruz-Orive, L. M., Imhof, V., and Gehr, P. (1990). Assessment of particle retention and clearance in the intrapulmonary conducting airways of hamster lungs with the fractionator. J. Microsc. 160, 75–88.
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, 533–556.
Hodgson, J. T., and Jones, R. D. (1985). A mortality study of carbon black workers employed at five United Kingdom factories between 1947 and 1980. Arch. Environ. Health. 40, 261–268.
IARC (International Agency for Research on Cancer). (1996). Carbon Black, vol. 65, p. 149.
Kirwin, C. J., LeBlanc, J. V., Thomas, W. C., Haworth, S. R., Kirby, P. E., Thaligar, A., Bowman, J. T., and Brusick, D. J. (1981). Evaluation of the genetic activity of industrially produced carbon black. J. Toxicol. Environ. Health 7, 973–989.
Kreyling, W. G. (1990). Interspecies comparison of lung clearance of "insoluble" particles. J. Aerosol Med. 3F, S93–S110.
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., et al. (1994). Pulmonary Toxicity of Inhaled Diesel Exhaust and Carbon Black in Chronically Exposed Rats. Part I: Neoplastic and Nonneoplastic Lung Lesions. Health Effects Institute, Cambridge, MA. 68, 1–75.
Morrow, P. E. (1988). Possible mechanisms to explain dust overloading of the lungs. Fundam. Appl. Toxicol. 10, 369–384.
Oberdrster, G. (1995a). Lung particle overload: Implications for occupational exposures to particles. Reg. Toxicol. Pharmacol. 27, 123–135.
Oberdrster, G. (1995b). The NTP Talc Inhalation Study: A critical appraisal focused on lung particle overload. Reg. Toxicol. Pharmacol. 21, 233–241.
Oberdrster, G., Ferin, J., Soderholm, S., Gelein, R., Cox, C., Baggs, R., and Morrow, P. E. (1994). Increased pulmonary toxicity of inhaled ultrafine particles: Due to lung overload alone Ann. Occup. Hyg. 38, 295–302.
Oberdrster, G., and Yu, C. P. (1990). The carcinogenic potential of inhaled diesel exhaust: A particle effect J. Aerosol Sci. 21, S397–S401.
Rausch, L. J., Bisinger, E. C., Jr., and Sharma, A. (2004). Carbon black should not be classified as a human carcinogen based on rodent bioassay data. Reg. Toxicol. Pharmacol. 40, 28–41.
Robertson, J. M., and Ingalls, T. H. (1980). A mortality study of carbon black workers in the United States from 1935 to 1974. Arch. Environ. Health. 35, 181–186.
Robertson, J. M., and Ingalls, T. H. (1989). A case-control study of circulatory, malignant, and respiratory morbidity in carbon black workers in the United States. Am. Ind. Hyg. Assoc. J. 50, 510–515.
Robertson, J. M., and Inman, K. J. (1996). Mortality in carbon black workers in the United States. J. Occup. Environ. Med. 38, 569–570.
Rudd, C. J., and Strom, K. A. (1981). A spectrophotometric method for the quantitation of diesel exhaust particles in guinea pig lung. J. Appl. Toxicol. 1, 83–87.
Sorahan, T., Hamilton, L., van Tongeren, M., Gardiner, K., and Harrington, J. M. (2001). A cohort mortality study of U.K. carbon black workers, 1951–1996. Am. J. Ind. Med. 39, 158–170.
Tran, C. L., Buchanan, D., Cullen, R. T., Searl, A., Jones, A. D., and Donaldson, K. (2000). Inhalation of poorly soluble particles. II. Influence of particle surface area on inflammation and clearance. Inhal. Toxicol. 12, 1113–1126.(Alison Elder, Robert Gelein, Jacob N. Fi)