Chronic Inhalation Exposure of Rats for up to 104 Weeks to a Non–Carbon-Based Magnetite Photocopying Toner

Abstract

Male and female Han Wistar rats were exposed for 6 h/day, 5 days/week for 13 or 104 weeks (whole body) to a magnetite photocopying toner. The toner contained 45% to 50% magnetite, with 45% to 50% styrene acrylic resin and less than 5% external additives, including hydrophobic amorphous silica and proprietary surface functional modifiers. Exposure levels were 1, 5, and 25 mg/m³ for the 13-week study and 1, 4, and 16 mg/m³ for the 104-week study. Lung toner burdens averaged 36, 288, and 604 μg per lung after 104 weeks’ exposure at 1, 4, and 16 mg/m³. The lung burdens were lower than have been reported in a similar study with a carbon-based toner. There were no significant effects on weight gain or food consumption in either study, or on clinical pathology parameters examined in the 13-week study. After 104 weeks’ exposure at 16 mg/m³, macroscopic examination revealed dark discoloration of the lungs and associated lymph nodes. Lung weights were significantly elevated by 21% and 14% for male and female rats, respectively. Microscopic findings indicative of a mild inflammatory response were similar in both studies, and included the presence of black-pigmented macrophages in the lungs and tracheobronchial and mediastinal lymph nodes; increased incidences of perivascular/peribronchiolar inflammatory cell infiltration; inflammation of the alveolar ducts (characterized by aggregations of black-pigmented alveolar macrophages and interstitial lymphocytic infiltration); increased cellularity of the bronchiole-associated lymphoid tissue; and a few instances of alveolar ciliated metaplasia. The 104-week study showed no increase in the incidence of pulmonary tumors.

Toners for electrophotographic printing are in widespread use throughout the world in office and consumer products, such as photocopiers, printers, and facsimile machines. Toners are composed of small particles containing resins, coloring agents, and other additives, such as surface functional modifiers. Coloring agents such as carbon black or magnetite are often used. Toner products are manufactured specifically to be composed predominantly of particles with a relatively large size range of 5.0 to 10 μm, and are used as printing media to form letters or images in the photocopying process. However, some of the particles are sufficiently small that they can become airborne and subsequently inhaled. This study was undertaken to investigate the possible effects of inhalation of a novel toner material.

Several rodent inhalation studies with a toner specially prepared by the Xerox Corporation, composed of 90% styrene/1-butyl-methacrylate copolymer and 10% carbon black, have been performed (Muhle et al. 1990, 1991). In those studies, mild to moderate inflammation and fibrosis were seen in the lungs of 92% of Fischer-344 rats exposed to the highest tested concentration of 16 mg toner/m³, and a minimal degree of fibrosis was seen in 22% of rats exposed to 4.0 mg toner/m³ for 6 h/day, 5 days/week for 2 years. There was no evidence of an excess of lung tumors in these animals, and the lung effects in the highest-dose group were attributed to lung overloading from saturation of clearance mechanisms. The toner tested in those inhalation studies contained carbon black as a coloring agent and did not contain surface functional modifiers.

The present studies were undertaken to investigate the long-term effects of inhalation of a toner containing magnetite (magnetic iron oxide; Fe₃O₄) as a coloring agent and proprietary external additives as surface function modifiers. The test toner contained 45% to 50% magnetite, with 45% to 50% styrene acrylic resin as the binding agent, and with less than 5% external additives, such as hydrophobic amorphous silica, as functional modifiers. The mean particle size was controlled at manufacture to be approximately 5.1 μm.

An initial 13-week subchronic inhalation study in rats was performed to obtain preliminary information for selection of appropriate exposure concentrations for a subsequent 104-week chronic inhalation study. In this 13-week study, rats were exposed to the particulate aerosol generated from the test toner for 6 h a day, 5 days a week for 13 weeks, at concentrations of 0 (air control), 1, 5, and 25 mg/m³. Estimation of lung burden was performed at intervals during the study using a satellite group of animals to investigate the possibility of any overload of lung clearance mechanisms. Lung burden measurements, together with histopathological effects on the respiratory tract, were primary considerations in dose selection for the subsequent long-term study because it is known that overloading of the lung clearance mechanism can lead to the formation of lung tumors in rats even with otherwise benign particles (Driscoll 1996; ILSI 2000). Although it is considered that the mechanism involved in the formation of such tumors is a rat-specific phenomenon of no relevance to human risk assessment, the present study was designed to avoid the possibility of lung overload. No significant clinical effects attributable to exposure were seen up to the highest dose of 25 mg/m³ in the 13-week study, but the extent of histopathological respiratory tract changes at this highest dose indicated that a high dose of 16 mg/m³ would be sufficient to produce clear nonneoplastic lung changes in the subsequent 104-week study. This concentration was also expected to induce increases in lung weights equal to or exceeding 10%, which would satisfy the requirement for a maximum tolerated dose (MTD) in a chronic toxicity/carcinogenicity study and provide information for direct comparison with the published data for the carbon black–based toner (Bellman et al. 1989, 1991; Muhle et al. 1990, 1991).

The 104-week rat chronic inhalation study was performed to assess any potential tumorigenic potential primarily in the respiratory tract, and also for comparison with previously described toxic effects following long-term exposure to poorly soluble particles such as carbon black–based toners (Morrow et al. 1991).

MATERIALS AND METHODS

Test Article Generation and Exposure System

The test toner aerosol was generated using a single Wright Dust Feed Mechanism (WDFM) (Wright 1950) for each dose group. In this device, the powder is packed into a canister, scraped off at a fixed rate, and dispersed into the exposure chamber inlet air stream. Different operating gear ratios enabled different rates of packed test toner removal, resulting in the required chamber concentrations.

The purpose-built whole-body exposure chambers used were constructed from stainless steel and glass, had volumes of approximately 2.5 m³, and were operated at a flow rate of 650 L/min. The animals were held in individual wire mesh cages within the exposure chambers during exposure and the chamber level on which they were exposed was rotated on a weekly basis to minimize the potential effect of any inhomogeneity in chamber aerosol distribution. Chamber environmental conditions monitored included airflow, temperature, and relative humidity. Measurements were made at 30-min intervals during each 6-h exposure period using an automated data collection system. Determination of the test toner concentration by gravimetric analysis was performed 4 times daily during each exposure, using glass fiber open-face filters (Whatman GF/A) mounted in open-face filter holders to collect the chamber air samples. Relatively low concentrations were employed in both studies, requiring large sample volumes (200, 100, or 60 L for low, intermediate, and high dose, respectively) to obtain sufficient particulate for accurate weighing. Consequently, two control chamber samples were taken during each exposure in order to provide data to correct for any loss of filter fiber or collection of animal dander. Study mean variation in aerosol concentration between four sampling positions located in the animals’ breathing zone averaged 16%, 1%, and 7% over the 104-week study duration for the low-, intermediate-, and high-dose groups, respectively. The four samples taken each day were taken from different sampling ports to allow evaluation of spatial homogeneity.

Samples for the gravimetric determination of particle size distribution were taken using a Marple 298 cascade impactor (Andersen Instruments, Smyrna, GA, USA) at a flow rate of 3 L/min, at which rate the collection stage cut-points were 17.4, 12.1, 8.0, 4.9, 2.86, 1.27, 0.76, 0.42, and 0 μm.

During preliminary generation system trials, the toner aerosol was found to deposit rapidly on the walls of generation system pipework. This problem was solved by the inclusion of corona discharge air neutralizers (Model AFC2; Simco, Hatfield, PA, USA) into the chamber air supply immediately after the powder generator. These devices ionize air molecule, thus allowing them to neutralize airborne particles by carrying charge to the chamber walls and thence to earth. Following use of the air neutralizers, analysis of samples removed from four different positions in each exposure chamber demonstrated good correlation with target concentrations at all levels.

Test Substance

The test substance was a photocopying toner containing 45% to 50% magnetite (magnetic iron oxide; Fe₃O₄; CAS no. 1317-61-9) and 45% to 50% styrene acrylic resin (glass transition point 60°C) as the binding agent. The toner also contained less than 5% external additives as surface functional modifiers. The main constituent of the functional surface modifier particles was a silica, which was obtained by hydrophobizing a fumed amorphous silica (CAS no. 112945-52-5) the particle size of which is less than 0.1 μm. The tested toner had an average particle size of 5.1 μm.

The test substance toner was of the type used with photocopiers and laser printers produced by Canon, Japan, which employs an image formation process called the “jumping phenomenon.”

The mean toner particle size was controlled at manufacture to be approximately 5.1 μm. The test toner aerosol generated by WDFMs appeared to be comprised of two distinct particle populations (Figures 1 and 2). One population (approximately 80.5%, 86.2%, or 91.7% by weight for low-, intermediate-, and high-dose groups, respectively) was composed of larger sized particles, with a mass median aerodynamic diameter (MMAD) between 5.1 and 5.8 μm and the remainder consisted of smaller particles (aerodynamic diameter <0.7 μm).

Scanning electron microscope (SEM) examination shows that the functional particles, such as amorphous silica, are attached to the surface of the toner particles and are uniformly dispersed. These functional particles attach to the surface of toner particles electrostatically and do not separate from the toner particles under normal handling. Before the inhalation studies were conducted, a sample of the test toner was passed through the test aerosol generator and the collected sample was examined for its particle size distribution and by SEM to determine whether aerosol generation affected the toner particles. The examinations showed that the functional particles remained uniformly dispersed on the surface of the toner particles without being detached and the particle size distribution did not change.

Furthermore, because the photocopying toner is made by grinding the resin mixture in which the iron oxide is uniformly dispersed, there is no apparent variation in the ratio of the binder resin and the iron oxide in differently sized particles. This has been confirmed by transmission electron microscopic (TEM) examination. Figure 7 depicts a TEM image of four particles of toner (three large and one small) showing very small (dark) particles of magnetite uniformly dispersed in the resin binder.

The stability of the test toner was determined using infrared (IR), differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) prior to study start and at the end of the 13-week study, and prior to the start and at 6-month intervals during the 2-year study. The results of each analysis confirmed the stability of the test material.

Animals and Husbandry

Male and female rats of the Han Wistar HsdBrl:WH strain were obtained from Harlan (UK), England. Rats used in the 13-week investigation were aged 8 to 10 weeks and in the weight range 241 to 292 g (males) and 170 to 207 g (females) at the start of exposure. This strain was selected as it has good survival characteristics and a well-documented background spontaneous tumor profile. Rats obtained for the 2-year study were 6 to 7 weeks of age and in the weight range 147 to 207 g (males) or 101 to 171 g (females) at the start of exposure. During the acclimatization period, selected animals of each sex in both investigations were sacrificed and examined macroscopically for health. The rats were housed five/cage in suspended stainless steel caging when not in the inhalation chambers. During exposure, each animal was housed separately within the exposure chamber. Animal husbandry complied with the requirements of the UK Home Office Animals (Scientific Procedures) Act 1986. Animals were fed pelleted SDS Rat and Mouse Diet No. 1 (Special Diet Services, Witham, Essex, UK) ad libitum. Tap water was supplied ad libitum, except during exposures, from water bottles, which were rinsed and refilled daily, and thoroughly cleaned periodically. Analysis of all batches of diet demonstrated them to be within nutritional requirements and to have no chemical or bacterial contamination that would interfere with the studies. Drinking water was analyzed every 6 months and found to be within safe limits for selected chemical contaminants. Animals were housed in temperature-controlled rooms with a 12-h light cycle.

Experimental Design: 13-Week Study

During the acclimation period, rats were assigned to four groups each of 84 males and 84 females by stratified randomization according to body weight. Three groups of rats were exposed to the particulate generated from test toner 6 h/day, 5 days/week for 13 consecutive weeks at target concentrations of 1, 5, and 25 mg/m³ in whole-body chambers with five levels of caging. The remaining group served as a control and was subjected to the same inhalation chamber conditions, but received air only. All animals were exposed simultaneously. Satellite and withdrawal groups of rats were included at each dose level to evaluate lung burdens and lung clearance, respectively. Urine samples from five male and five female satellite group rats were collected overnight prior to the last exposure of weeks 4, 8, and 13 for assay of mandelic acid using mass spectrometry. Mandelic acid is the major metabolite of the styrene component of the test toner. Following the last exposure in weeks 4, 8, and 13, these same satellite rats were sacrificed, the lungs dissected free, trimmed at the junction of each lobe’s main bronchus, weighed, and processed for test toner lung burden analysis.

To determine reversibility of any changes, five male and five female rats in each group were evaluated at 4, 8, and 13 weeks after the last exposure to toner.

The animals were individually observed for clinical signs at daily placement and removal from the exposure chamber. For all groups, observations were made during and following each exposure. Body weight and food consumption were determined weekly throughout the study. Assessment of water consumption was restricted to visual observation of water bottle contents. After dilation with a tropicamide ophthalmic solution, the eyes of all rats were examined using a Keeler indirect ophthalmoscope before exposures began, and from control and high-dose animals during the final week of exposure.

Lung Burden Analysis

Determinations of lung burden were performed by digestion of weighed amounts of lung tissue in tetramethylammonium hydroxide for 64 h at 70°C. The digest containing the toner particles freed from lung tissue was centrifuged and the supernatant discarded. Methanol was added to suspend each sample followed by sonication for 10 min. The samples were centrifuged and the resulting pellet was allowed to air dry prior to resuspension in 2-butanone. The toner concentration was determined by quantifying spectrometric absorption at 540 nm. The mean values for lung burdens for the test groups for each sex and time point were adjusted, to compensate for absorption from undigested lung components, by the subtraction of the background values in digests of concurrent control animals. The methods to determine lung burden in this study are similar to procedures used for studies on carbon-based toners (Bellman et al. 1991) and diesel soot particles (Griffis et al. 1983). The lung burden measurement procedure included spiking lungs with toner to establish a standard reference curve. The test toner used for lung burden calibration purposes was a sample of toner particles with an aerodynamic diameter less than 8 μm collected from the impaction plates of a Marple cascade impactor sampler from a typical test atmosphere of the toner. This was considered more representative for method calibration purposes than using bulk toner because particles larger than 8 μm would not be expected to enter the upper or lower respiratory tract of the rat.

Hematology and Clinical Chemistry

Following an overnight fast, blood was obtained from the periorbital sinus of all rats during weeks 4 and 13 of exposure for hematological and clinical chemistry determinations. Hematological evaluation included hematocrit, hemoglobin, erythrocyte count (RBC), white cell count (WBC), reticulocyte count, platelet count, prothrombin time, and activated partial prothrombin time. WBC differential and RBC morphology were evaluated using Wright-stained blood smears. Clinical chemistry evaluation included creatine kinase (Roche Cobas centrifugal analyzer), glucose, total protein, albumin, globulin (by subtraction), urea, creatinine, alkaline phosphatase, alanine aminotrans-ferase, aspartate aminotransferase, γ-glutamyltransferase, total bilirubin, sodium, potassium, calcium, inorganic phosphorus, chloride, and cholesterol (Hitachi 737 analyzer).

Gross and Histopathology

Animals, including satellite and withdrawal rats, were euthanized by intraperitoneal overdose of pentobarbitone sodium followed by exsanguination from the brachial vessels. The main test group of rats was necropsied following week 13, and the following organs were dissected free and weighed: adrenals, brain, heart, kidneys, liver, lungs, ovaries, spleen, testes, epididymides, thymus, and uterus. Tissues were preserved in neutral buffered formalin; eyes were preserved in Davidson’s fixative (Latendresse et al. 2002). At necropsy, the lungs of main and withdrawal group rats were inflated with neutral-buffered formalin before immersion in fixative. At necropsy, the nasal cavity was flushed with fixative, the lower jaw removed, and the head immersed in neutral-buffered formalin.

Following fixation, the head was decalcified in Kristensen’s fluid (6 days), and samples of the nasal cavity taken from four levels (Young 1981). Tissues for histological evaluation were routinely processed, embedded in paraffin wax, and 4-μm sections prepared, stained with hematoxylin and eosin (H&E), and examined under the light microscope. The following tissues from the high-dose and control animals were examined: adrenals, esophagus, stomach, duodenum, femur /bone and marrow, jejunum, ileum, cecum, colon, rectum, aorta, brain, epididymides, eyes, heart, kidneys, lachrymal gland, liver, lungs, lymph nodes (cervical, mesenteric, mediastinal, and tracheobronchial), nasal passages, ovaries, pancreas, nasopharynx, pituitary, prostate, salivary gland, sciatic nerve, seminal vesicles, skeletal muscle, spinal cord, spleen, sternum, testes, thymus, thyroid glands, parathyroid glands, trachea, urinary bladder, and uterus, as well as any macroscopically abnormal tissue. In addition to histopathologic examination of the control and high-exposure animals, histopathologic examination of the nasal passages, larynx, trachea, lungs, tracheobronchial lymph nodes, and macroscopically abnormal tissues was performed on animals at the lower exposure levels. Also, histopathologic examination of the lungs, tracheobronchial lymph nodes, and macroscopically abnormal tissues was performed on withdrawal group animals.

Experimental Design: 104-Week Study

During the acclimation period, rats were assigned to four groups each of 150 males and 150 females by stratified randomization according to body weight. Three groups of rats were exposed to the particulate generated from the test toner 6 h/day, 5 days/week for 104 consecutive weeks at target concentrations of 1, 4, and 16 mg/m³ in whole-body chambers. The remaining group served as a control and was subjected to the same inhalation chamber conditions, but received air only. The test and control animals were individually observed for clinical signs at daily placement and removal from the exposure chamber. Group observations were made during exposures. Body weight and food consumption were determined weekly up to week 16 and monthly there after throughout the remainder of the study.

Hematology

Peripheral blood was obtained from the tail of surviving main groups of rats at the end of weeks 52, 78, and 104, smears made, air dried, fixed in methanol, and stained. Smears from control and high-dose rats were examined for abnormalities. At study termination and at scheduled sacrifices, animals were given an intraperitoneal overdose of pentobarbitone sodium followed by exsanguination from the brachial vessels.

Lung Burden

A satellite group of 25 male and 25 female rats was included in each dose group to determine lung burden of magnetite toner following the last exposure in weeks 12, 26, 52, 78, and 104. At these time points, five male and five female satellite rats were sacrificed, the lungs dissected free, trimmed at the junction of each lobe’s main bronchus, weighed, and processed for toner lung burden analysis as described previously in the 13-week study.

Gross and Histopathology

Main group rats were necropsied over a 17-day period following week 104, with exposures of remaining animals continuing up to the last day of sacrifice. The procedures were as described for the 13-week study.

Regulatory Compliance

These studies were conducted in accordance with Good Laboratory Practice Standards (US EPA 1989a,b; UK HSE 1999; EC 1999; Japan MITI 1984; OECD 1998). Pathology observations were peer reviewed within the testing facility and by an independent consulting pathologist.

Statistics

Qualitative data were analyzed by Bartlett’s test (1937) for homogeneity; if the data were heterogeneous, a logarithmic transformation was performed to see if a more stable variance could be obtained. Data without significant heterogeneity were analyzed by one-way analysis of variance and Williams’ test (1971 (1972) for dose response. The Kruskal-Wallis (Kruskal and Wallis 1952, 1953) analysis was performed for nonparametric data. For organ weight data, the final bodyweight was used as covariate in the analysis of covariance. Pathologic data were analyzed using Fisher’s exact test (Sokal and Rohlf 1980). For the 2-year study, mortality was analyzed using the log ranks method (Mantel 1966). Tumor incidence was analyzed by methodology described by the International Agency for Research on Cancer (1980).

RESULTS

Exposure Conditions

The mean (mean of the mean values for each exposure) 13-week analyzed test toner concentrations, corrected for the study mean control filter weights, was 0.97, 5.1, and 23.2 mg/m³. These values were within 1%, 1%, and 8% of target for the 1, 5, and 25 mg/m³groups, respectively. The coefficient of variation for the samples was 19.3%, 10.6%, and 15.6% for groups exposed at 0.97, 5.1, and 23.2 mg/m³, respectively. As the samples were taken from four sample ports located in different regions of the chamber, the CV includes temporal and spatial variations in the distribution of the test atmospheres. The aerosols contained two populations of particles. The population of smaller particles had a mass median aerodynamic diameter (MMAD) <0.7 μm, whereas the major population of particles had an MMAD of 5.3, 5.2, and 6.5 μm for the low-, intermediate-, and high-dose groups, respectively. The proportion of particulates <0.7 μm was 16%, 10%, and 3% by weight for the low-, intermediate-, and high-dose groups, respectively. Chamber airflow during the 13 weeks averaged 648, 647, 647, and 646 L/min for the control, low-, intermediate-, and high-dose groups, respectively. Average chamber temperature and relative humidity were 22.4°C, 22.7°C, 22.5°C, and 21.9°C and 59%, 53%, 55%, and 61% for the control, low-, intermediate-, and high-dose groups.

The mean (CV) 104-week study analyzed test toner concentrations, corrected for the study mean control filter weights, was 0.96 (11.7%), 3.96 (9.2%), and 15.63 (10.3%) mg/m³. These values were within 4%, 1%, and 3% of target for low-, intermediate, and high-dose groups. The MMAD for the major population of particles was 5.2, 5.1, and 5.8 μm for the low-, intermediate-, and high-dose groups, respectively. The proportion of the test aerosol in the smaller population of particles with an MMAD less than 0.7 μm was 18%, 13%, and 8% for the low-, intermediate-, and high-dose groups, respectively. Chamber airflow averaged approximately 650 L/min for all groups. Average chamber temperature and relative humidity during the 104-week exposure period were 22.6°C, 21.8°C, 22.4°C, and 21.4°C, and 57%, 54%, 53%, and 59% for the control, low-, intermediate-, and high-dose groups. No significant effect upon study results was expected from the minimal intergroup differences in airflow, temperature, and relative humidity.

Lung Burden Assay

Results of analysis of lung tissue for burden of test toner content at week 12 of the 104-week study were very similar to those found at week 13 of the 13-week study (Table 1). Mean lung burdens for males and females combined, adjusted by subtraction of background values in concurrent control animals, were 4, 110, and 221 μg per lung for week 12 of exposure at 1, 4, and 16 mg/m³, respectively, in the 104-week study, compared with 29, 165, and 253 μg per lung for exposure at 1, 5, and 25 mg/m³, respectively, in the 13-week study. Subsequent to week 12 in the 104-week study, the results of the lung burden analysis were similar to week 12 values, but showed increased variability with time, probably due to a variable background contribution from undissolved cartilage and tissue debris. This undigested material gradually increased as the rats grew older (due in part to increasing amounts of cartilage in the lungs), and this is considered a factor influencing increased variability of lung burden results with time. Variability was most apparent in the low-dose group but was also seen to a lesser degree at the higher dose levels. At the high dose level in the 104-week study (Table 1), the lung burden, averaged for males and females, was 221 μg per lung at week 12 and varied between 321 and 652 μg per lung for the remainder of the study. At week 104 of the study, lung burdens, averaged for males and females, were 36, 288, and 604 μg per lung. The lung burden of toner was dose related at all time points and was roughly constant from week 26 onwards, indicating that lung overload had not occurred.

Survival and Clinical Observations

A dose-related darkening of the fur and tails of rats due to deposition of the toner on the fur of exposed rats was evident in both studies. However, no adverse effects of exposure to test toner was evident in animal behavior or clinical signs at concentrations up to 25 mg/m³ for 13-weeks, or 16 mg/m³ for up to 104 weeks. Ophthalmoscopic examination did not reveal any changes in the 13-week study attributable to exposure to test toner. Statistical analysis of survival data showed that there was no evidence that exposure to the test toner had any effect on survival (Figure 3). No signs indicative of systemic toxicity were evident.

Body Weight and Food Consumption

In the 13- and 104-week studies, no treatment-related differences were seen in body weight gain and food consumption between control and test groups. A slight, statistically significant increase in body weight gain for male high-dose rats in the 104-week study was not seen in the female rats of the same group, and in the absence of any indicated systemic toxicity, the difference was not considered to be treatment related.

Hematology, Clinical Chemistry, and Mandelic Acid Assay

In measurements performed in the 13-week study, no toxicologically significant differences were seen in a range of hematological and blood chemical parameters as a result of exposure to test toner. Urinary mandelic acid results, indicative of metabolism of the styrene-acrylic resin component in the toner, did not indicate any treatment-related differences between the groups. Values were very variable (50 to 1000 ng/ml of urine) and there was no indication of a group difference. Peripheral blood smears taken at intervals during the 104-week study showed no treatment-related changes.

Organ Weights

No toxicologically significant or treatment-related effects were seen in the 13-week study. However, at the end of the 104-week study, statistical analysis of organ weights, using body weight as covariate, revealed significantly higher lung weights for both sexes in the high-dose animals. Mean absolute lung weights were 21% and 14% higher for male and female high-dose rats, respectively, compared with control values. This increase in lung weights was considered treatment related and likely attributable to retention of the toner particles and a consequent inflammatory response as evident in both gross and microscopic pathology findings.

Gross and Microscopic Pathology

Gross examination revealed dark discoloration of the lungs in 30% of high-dose male rats at termination of the 13 week study. Lung congestion was observed in a proportion of withdrawal male rats in all toner-exposed recovery groups 4 weeks after the end of exposures. However, by week 13 following the last exposure, no treatment-related macroscopic signs were evident. Microscopic examination of respiratory tract tissues from the 13-week study revealed the presence of pigmented macrophages in the lungs and tracheobronchial lymph nodes at all dose levels and the bronchus associated lymphoid tissue (BALT) in the intermediate- and high-dose groups. Granulomatous inflammation was evident in the lungs of rats from the intermediate- and high-dose groups. Examination of tissues from rats killed 13 weeks after the cessation of exposure showed that all these findings were, at least partially, reversible (Table 2).

In the 104-week study (Table 3), black areas on the lungs were seen macroscopically in 96%, 92%, and 60% of male rats, and in 86%, 10%, and 18% of female rats in the high-, intermediate-, and low-dose groups, respectively. In addition, a general grey discoloration of the lungs was seen in 20% male and 26% female high-dose rats, but not in the lower-dose groups. These findings were not seen in control rat lungs of either study. Microscopically the nonneoplastic treatment-related changes seen in the lungs were those associated with the normal clearance of particles from the lungs, together with a mild inflam-matory response to the presence of particulate. The principal nonneoplastic findings included black-pigmented macrophages with associated changes in the lungs, tracheobronchial and mediastinal lymph nodes, together with lesser changes in the trachea, tracheal bifurcation, and nasal passages (Figure 4A to D). Black-pigmented lung alveolar and perivascular/peribronchiolar macrophages were seen in all intermediate- and high-dose animals and in a high proportion of low-dose animals (70% to 80% and 50% to 66% for males and females, respectively). A similar dose-related increase in the incidence of proliferation of type 2 pneumocytes associated with alveolar macrophages was also noted in males and females of all three treated groups. This is a common response to epithelial injury (Schwartz et al. 1994). An increased incidence of brown-pigmented alveolar macrophages was seen in high-dose animals. Alveolar eosinophilic debris and fibrillar material associated with foamy macrophages was seen at increased incidence and degree in high-dose animals. There was also a tendency for the incidence of foamy alveolar macrophages to be greater at the high dose level. Increased incidences of perivascular/peribronchiolar inflammatory cell infiltration (mainly lymphocytes) were seen in male and female high-dose animals. Inflammation of the alveolar ducts (characterized by aggregations of black-pigmented alveolar macrophages with inclusions of black test material and interstitial lymphocytic infiltration) was seen in all toner-exposed groups, mainly in the intermediate- and high-dose groups (Figures 5 and 6B ).

Increased cellularity of the bronchiole-associated lymphoid tissue was noted in high-dose animals (Figure 6A ). Increased incidences of focal or multifocal alveolar epithelial hyperplasia (not statistically significant) were seen in high-dose animals (Figure 4A ), and increased incidences of alveolar ciliated metaplasia (Figures 4B and 5) were seen in male and female high-dose animals (p < .001) and in female intermediate-dose animals (p < .01).

Black-pigmented macrophages in the tracheobronchial and mediastinal lymph nodes, often containing brown inclusions (Figure 4C ), were seen in animals of all treated groups, the incidence and degree being dosage related. Females tended to show greater incidences of these findings than males, but there was little difference in incidence or degree between terminal and decedent animals. An increased incidence of generalized increased cellularity was noted in high-dose animals (Table 4). At the tracheal bifurcation, black-pigmented macrophages, sometimes containing brown inclusions, and often associated with lymphoid tissue, were seen in the submucosa of the bronchi in animals of all three treated groups, the incidence and degree being generally dose related. In the nasal passages, there was an increase in the incidence of eosinophilic inclusions in the respiratory epithelium in male and female high-dose animals when compared with controls. This effect was greater in males that survived to termination than in decedents, although this was not the case in females. Increases in the incidence of eosinophilic inclusions in the olfactory epithelium were seen in male and female animals in the intermediate- and high-dose groups, and in females of the low-dose group, compared with controls. There was some evidence of dosage relationship, although there was little difference between females in the intermediate- and high-dose groups. The effect in females was greater in terminal animals than in decedents, although this was not the case in males. These findings were considered to reflect mild local irritation.

Statistical analysis of the tumor incidence in the 104-week study (WHO 1980) revealed no treatment-related differences in tumor profiles.

The only neoplastic findings in the lungs were a bronchioalveolar adenoma in one male control, one female control, and one male of the low-dose group, whereas bronchioalveolar adenocarcinoma was seen in one male of the intermediate-dose group. No pulmonary tumors were seen in any animal in the high-dose group.

DISCUSSION

In these studies, rats were exposed to the magnetite test toner by inhalation at concentrations up to 26 mg/m³ (13-week study) or 16 mg/m³ (104-week study). Microscopic examination of the lungs and the lung-associated lymph nodes clearly showed the presence of large quantities of black toner particles. In addition, brown and green particles were seen, particularly in the tracheobronchial lymph nodes that were often green, grey, or black at macroscopic examination. Quantitative measurement of the lung burden of toner particles showed toner burdens of around 0.2 to 0.3 mg per lung in rats exposed to 25 mg/m³ for 13-weeks, and 0.6 mg per lung in rats exposed to 16 mg/m³ for 2 years. The lung burdens in the present study were considerably less than those seen in the published data for the carbon black–based toner, where burdens of over 2 mg/lung were seen after 13-weeks and 16 mg after 2 years exposure to 16 mg/m³ (Muhle et al. 1990, 1991; Bellman et al. 1989, 1991). In the long-term study with the carbon black–based toner, a moderate degree of fibrosis was seen in the lungs, a finding not seen with the currently tested toner. This may have impaired lung clearance. Lung toner burden remained relatively constant from week 26 of the present 104-week study and the ratio of toner burden between the groups remained similar to the ratio of the exposure levels, indicating that lung overload had not occurred. The results also show that a maximum of approximately 650 μg/lung was attained during the periods analyzed in Table 1. With typical lung weights in rats throughout the study in a range from 1.0 to 2.5 g, lung burdens were in a maximum range of 260 to 650 μg/g lung. If approximately 1 mg/g lung tissue is considered to be the lung overload threshold (Oberdorster 1997), these results also indicate that lung overload did not occur in the current studies with magnetite toner.

Particle size distribution analyses performed at intervals during both the 13- and 104-week studies indicated the presence of two populations of particles in the test atmospheres. One population was composed of larger-sized particles, most with an aerodynamic diameter between 3 and 12 μm, and the second was composed of particles with an aerodynamic diameter less than 0.7 μm. The presence of two populations of particles was evident visually in photomicrographs of the powder of the test toner (Figure 1). TEM examination of the smaller particles and the larger particles (Figure 7) shows that they appear similar in composition. Study average MMAD values of 5.3, 5.2, and 6.5 μm for the 13-week study and 5.2, 5.1, and 5.8 μm for the 104-week study were achieved, which were close to the average size of the test toner before aerosolization (5.1 μm). These values were slightly greater than published data from the studies with the toner prepared by the Xerox Corporation (Muhle et al. 1990, 1991), which showed an MMAD of 4.2 μm at the high-dose level. Although this value is slightly lower than the 5.1 to 6.5 μm obtained in the current investigations, the submicron population present in atmospheres of the magnetite toner would contribute to an overall increased inhaled fraction, probably resulting in similar total respirable fraction for both studies. The test toner was used as supplied to simulate potential exposures to users, with no attempts to improve generation by reduction of particle size by milling, and it is recognized that the particle size was greater than ideal for rat inhalation studies. However the evidence from increased lung weights, together with significant lung histopathology findings, indicates that a considerable quantity of test toner was inhaled. The USEPA recommended MMAD range of 1 to 3 μm for rat inhalation studies assumes a normal distribution of aerostable particulate. However, in this study two particle populations were present. Assuming for an ‘ideal’ USEPA MMAD range of 1 to 3 μm, the proportion of particles <7 μm and considered respirable, would be 78%. This is in good agreement with the actual achieved proportion of 69% to 78% <7 μm in the present study when the mass of the respirable particles of the larger populations is added to the mass of the particles in the smaller sized population.

In the 13-week study, histopathological findings in lung and tracheobronchial lymph nodes were typical of a nonspecific response to inhalation of a poorly soluble particle and no evidence of specific changes related to the test toner exposure was seen. However, there were indications in the histopathology from recovery rats that the toner had not been cleared significantly by the end of 13-weeks after withdrawal from exposure. The nonneoplastic histopathological findings in this study were similar to those described in rats exposed to a carbon black–based toner for 13-weeks at 16 and 64 mg/m³ (Muhle et al. 1990).

In the 104-week chronic study, treatment-related histopathological changes in the respiratory tract were similar to but more marked than those seen in the 13-week preliminary study. These changes included inflammatory changes, black-pigmented macrophages in the lungs, tracheobronchial and mediastinal lymph nodes, together with lesser changes in the trachea, tracheal bifurcation, and nasal passages. There were no treatment-related differences in tumor profiles and no pulmonary tumors at all in the high-dose group.

The absence of treatment-related neoplastic findings in this study indicates that respirable dust from the magnetite toner used in this study is not tumorigenic in the rat.