(a) Poorly soluble particles: general introduction

Few studies are available in humans on the kinetics of clearance and retention of the specific inhaled particles that are discussed in this volume (i.e. carbon black, titanium dioxide and talc). However, for any particles, the probability of their deposition within a given region of the respiratory tract depends on their characteristics and the physical factors that influence their transport in the airstream (e.g. air velocity and airway structure; ICRP, 1994). Deposition by mechanisms of sedimentation and impaction depends on the aerodynamic diameter, while deposition by diffusion depends on the thermodynamic diameter of the particles (ICRP 1994; Environmental Protection Agency, 2004).

Several terms have been adopted in the measurement of aerosols and estimation of the probability of particle deposition in the human respiratory tract (ICRP, 1994; International Standards Organization, 1995; ACGIH®worldwide, 2005). The term ‘respirable’ refers to particles that can deposit in the alveolar (gas exchange) region of the lungs. Within this monograph, respirable size fractions are defined as ultrafine (< 0.1 µm diameter of primary particle), fine (0.1–2.5 µm) and coarse (> 2.5–10 µm) particles. ‘Thoracic’ refers to particles that can deposit in the lung airways, while ‘inhalable’ refers to particles that can deposit anywhere in the respiratory tract. It is recognized that primary ultrafine particles generally exist as aggregates that have a greater surface area than larger primary particles.

A detailed discussion of particle dosimetry in the human respiratory tract can be found elsewhere (Oberdörster, 1988; ICRP, 1994; NCRP, 1997; Environmental Protection Agency, 2004; Bennett & Brown, 2005; Brown et al., 2005; Martonen et al., 2005). In brief, inhaled particles may be either exhaled or deposited in the extrathoracic, tracheobronchial or pulmonary airways. The deposition of particles in the respiratory tract depends primarily on inhaled particle size, the route of breathing (i.e. through the nose and/or mouth) and the breathing pattern (e.g. volume and frequency). Particles close to 0.3 µm in diameter have minimal mobility, i.e. they are sufficiently large that their diffusive mobility is minimal, yet small enough that their sedimentation and impaction are also minimal. As a consequence, particles in this size range also have minimal deposition in the lung (Fig. 4.1). In general, the deposition fraction of most particle sizes (< 3–4 µm aerodynamic diameter) in humans is greater in the alveolar region than in the tracheobronchial airways. The deposition fraction for particles > 3–4 µm and < 0.01 µm in the alveolar region decreases due to their removal from the extrathoracic (particularly during nasal breathing) and tracheobronchial airways (Fig. 4.1). Of particular relevance to occupational exposures, particles that carry a charge due to the method of their generation (e.g. titanium dioxide) may have increased deposition efficiency in the lungs.

Figure 4.1

Probability of particle deposition in the human respiratory tract by region, according to the ICRP (1994) model. The average deposition has been modelled for an adult breathing through the nose at 25 L/min (light exercise), assuming nasal breathing of (more...)

Particles are frequently aggregates or agglomerates of smaller primary particles. The aerodynamic and thermodynamic properties of these aggregates (rather than the primary particles) affect their behaviour in the air and the probability of their deposition in the respiratory tract. Once deposited, properties such as the size and surface area of both aggregates and primary particles can potentially affect clearance kinetics.

Particle retention in the respiratory tract is determined by the balance between the rate of deposition and the rate of clearance. Particles that deposit in the tracheobronchial region are cleared by mucociliary clearance, which is relatively rapid (retention half-times of approximately 24–48 hours) (IARC, 1996), although some portion of the particles that deposit in the airways is cleared more slowly than expected (Stahlhofen et al., 1995). For particles that deposit in the alveolar region, the primary mechanism of clearance is by alveolar macrophage phagocytosis, migration to terminal bronchioles and the ‘mucociliary escalator’, through which particles are eventually swallowed or expectorated (Oberdörster, 1988). Particles that deposit in the alveolar region are associated with the slow clearance phase (retention half-times of months to years in humans) (Bailey et al., 1985; IARC, 1996). In a study of coal miners, little or no clearance of particles was observed (by magnetopneumography) one year after their retirement from the mine (Freedman & Robinson, 1988; Freedman et al., 1988). Translocation of particles to the interstitial region increases particle retention time in the lungs (ICRP, 1994). Some fraction of particles that deposit in the alveolar region may also be translocated to the lung-associated lymph nodes. This may occur by transepithelial migration of alveolar macrophages following phagocytosis of the particle or by translocation of free particles to the interstitium, where they may be phagocytosed by interstitial macrophages. Inflammation may alter mucociliary clearance, phagocytosis by alveolar macrophages and the uptake and transport of particles to and through the respiratory epithelium.

Particle deposition and clearance vary among individuals for several reasons, e.g. because of age, gender, smoking status and health status. Pre-existing lung diseases or conditions such as asthma or chronic obstructive pulmonary disease can influence the efficiency and pattern of deposition within the respiratory tract. Deposition also depends on the level of activity and breathing patterns. Deposition and retention determine the initial and retained particle dose to each region and may therefore influence the risk for developing diseases specific to those regions of the respiratory tract.

In summary, the pattern of deposition of particles depends on the particle diameter (aerodynamic or thermodynamic) and on the anatomical and physiological characteristics of the host. The deposition fraction for particles such as carbon black and titanium dioxide within the respiratory tract may vary depending on the size of the agglomerates and influences the dose to a given region of the respiratory tract. Pre-existing lung diseases or conditions can also influence deposition patterns.

(b) Deposition and retention of inhaled carbon black particles in the human respiratory tract

Several studies describe the deposition and retention of carbon black in the respiratory tract of exposed workers, as well as the health effects of these exposures, which are discussed in Section 4.2.

Although no quantitative data are available, studies of tissues from workers in carbon black factories have shown that widespread deposits of large amounts of carbon black are retained in the lungs (Rosmanith et al., 1969; Beck et al., 1985).

Lung diseases or conditions (either pre-existing or particle-related) may influence the deposition and retention of particles, e.g. by altering the size, structure and airflow patterns of the airways and by potentially affecting mechanisms of lung clearance. In a recent study of healthy humans who inhaled ultrafine carbon particles (count median diameter (CMD), 0.025 µm; geometric standard deviation (GSD), 1.6), bronchoconstriction may have caused the observed mild dysfunction of the small airways (increased airways resistance, seen as reduced maximal mid-expiratory flow rate [forced expiratory flow (FEF_25–75%)]) (Pietropaoli et al., 2004). The exposures were relatively low (single 2-hour exposures to 50 µg/m³, an ambient concentration that is found near major roads) and the individuals were healthy (no pre-existing lung disease). Bronchoconstriction was offered as the most probable mechanism, in part because pulmonary inflammation (as assessed by sputum), which would have been another possible explanation, was not observed. Reduced alveolar gas exchange (measured as reduced carbon monoxide diffusing capacity) was also observed, which was attributed to vasoconstriction. No adverse effects were observed in normal or asthmatic individuals who received single, 2-hour exposures to 10 µg/m³ ultrafine carbon; the effects observed in the group exposed to 50 µg/m³ were reversible. Particle deposition was not evaluated in this part of the study. In the same study, the deposition fraction of ultrafine carbon particles was measured in the respiratory tract in healthy and asthmatic subjects at rest and during exercise (Daigle et al., 2003; Chalupa et al., 2004; Frampton et al., 2004). The CMD of the ultrafine carbon aerosol was 0.025 µm (GSD, 1.6) (Pietropaoli et al., 2004), and 96% of the particles were elemental carbon (Frampton et al., 2004). The deposition fraction of the ultrafine particles in the respiratory tract was measured as the difference in the inspired and expired particle concentrations divided by the inspired concentration (using either mass or number concentration) (Frampton et al., 2004). [The Working Group noted that there may be methodological problems in relation to deposition measurements in the above series of studies as commented on by Kim and Jaques (2004).]

Compared with healthy individuals, asthmatics had an approximately 50% higher total deposition fraction of ultrafine carbon particles in the respiratory tract as either total number or mass deposited (Chalupa et al., 2004; Frampton et al., 2004). In a separate study, Brown et al. (2002) reported greater deposition of ultrafine particles in individuals who had obstructive lung disease.

Particle diameter influences deposition, even within the ultrafine particle size range. Daigle et al. (2003) and Frampton et al. (2004) reported the total ultrafine deposition fraction as particle mass or number, by the midpoint diameter of particle sizes from 7.5 to 75 nm. Within that particle size range, the deposition fraction increased with decreasing particle size, either at rest or with exercise, and among healthy or asthmatic individuals. For example, the deposition fraction for particles with median sizes of 65 nm and 8.7 nm increased from 0.63 to 0.74, respectively, in healthy subjects. During exercise, the deposition fraction increased from 0.84 to 0.94 for the same particle sizes and study group (Frampton et al., 2004).

Jaques and Kim (2000) and Kim and Jaques (2004) reported that the total deposition fraction of ultrafine aerosols in humans increased with decreasing particle size (from median diameter of 100 nm to 40 nm). The deposition fraction of particles increased to a similar extent with increases in either tidal volume or respiratory period. During exercise, tidal volume increased and residence time decreased relative to measurements taken at rest. These findings are in contrast to those of Daigle et al. (2003). Kim and Jaques (2004) noted that the methodology used by Daigle et al. (2003) may cause measurements of exhaled particle concentrations to be variable and erroneous. Furthermore, they asserted that the unusually high deposition values obtained by Daigle et al. (2003) were due to an improper sampling of exhaled aerosols. Although the deposition fraction of the particles decreases during exercice on a breath-by-breath basis, the total amount of particle deposition increases with exercise due to an increase in respiratory rate.

Observed and predicted deposition fractions were compared in a study of the total respiratory tract deposition of ultrafine carbon particles. In resting individuals, the observed fractions were found to be similar to those predicted by three deposition models (ICRP, 1994; NCRP, 1997; CIIT & RIVM, 2000 [the 1999 multiple path particle deposition model version from CIIT is cited, but not referenced]; Frampton et al., 2004). However, for exercising individuals, the total deposition fractions were higher than those predicted by the models (Frampton et al., 2004). This underprediction increased as the particle size increased from 10 to 100 nm; for 26-nm particles, the predicted deposition fraction was 22% lower than that measured in exercising individuals (Frampton et al., 2004). Among elderly subjects, the total deposition fraction observed was similar to that predicted from the ICRP (1994) model (although the model slightly overpredicted the deposition fraction of particles smaller than 0.04 or 0.05 µm, and slightly underpredicted the deposition fraction for particles larger than approximately 0.08 µm) (Kim & Jaques, 2005).

Gender was not found to affect the deposition fraction significantly in the Frampton et al. (2004) study in the two groups that had sufficient numbers to address this variable. In contrast, Jaques and Kim (2000) reported a greater total deposition fraction in women compared with men with the same breathing pattern, particularly for the smaller ultrafine particles (40 nm), although inter-subject variability was similar.

(c) Extrapulmonary translocation of carbon particles in humans

The translocation of coal dust particles of respirable size [specific size not noted] from the respiratory tract to other tissue sites has been observed in coal miners. Black pigment observed in the liver and spleen was associated with years in mining and severity of coal workers' pneumoconiosis (LeFevre et al., 1982). To reach the liver and spleen, the particles would have had to enter the blood circulation. It is not clear whether this was due to particles being cleared by the mucociliary clearance, being swallowed and entering the gastrointestinal tract and then being taken up in the blood, or whether the particles were able to pass through damaged epithelial and endothelial cells into the blood, as could occur under conditions of disease.

Several studies have been published on the clearance of agglomeration mode carbon particles (^99mTechnetium-labelled carbon particles < 100 nm in diameter [Technegas]; Nemmar et al., 2002). The primary particles that compose Technegas are in the range of 5–20 nm (Lemb et al., 1993; Lloyd et al., 1995). However, before inhalation, these primary particles coagulate into aggregates that have a median diameter in the range of 100 nm to 160 nm (Lemb et al., 1993; Lloyd et al., 1995; Roth et al., 1997). Pulmonary retention of Technegas 45 minute after inhalation was reported by Roth et al. (1997) to be 95% and by Isawa et al. (1991) to be 98%. According to Brown et al. (2002), pulmonary retention 45 minute after inhalation must on average have been less than 67% in the study of Nemmar et al. (2002), although these data were not reported. In view of the sharp contrast in the findings of Nemmar et al. (2002) and those of others, Brown et al. (2002) contended that the results of Nemmar et al. (2002) were consistent with the clearance of pertechnetate, but not with that of insoluble ultrafine particles. Mills et al. (2006) specifically investigated this supposition. Six hours after inhalation of Technegas, 95.6% of the particles remained in the lungs, and no accumulation of radioactivity was detected in the liver or spleen. In contrast to Nemmar et al. (2002), Mills et al. (2006) found that ultrafine carbon particles do not pass directly from the lungs into the systemic circulation.

(d) Excretion of particle-adsorbed substances

The retention of particles in the lungs may influence the bioavailability of adsorbed materials. As the retention of particles increases, the potential for adsorbed PAHs to be eluted and absorbed may also increase.

In a study of five nonsmoking warehouse packers in a carbon black (furnace black) manufacturing plant, daily average dust exposures were measured by air sampling, and urinary excretion of 1-hydroxypyrene (derived from pyrene) was measured in post-shift urine samples for five consecutive days during one work week. The mean ambient dust concentrations over the five days ranged from 1.5 to 13 mg/m³. Excretion of 1-hydroxypyrene ranged from 0.10 to 0.48 µmol/mol creatinine. A regression model showed a statistically significant relationship between weekly mean concentration of airborne dust and 1-hydroxypyrene excretion when the intercept was forced through zero—i.e. assuming zero 1-hydroxypyrene excretion with zero measured dust exposure—but not when the intercept was unconstrained—i.e. allowing for some level of 1-hydroxypyrene at zero measured dust concentration, such as from diet, as noted by the authors, or possibly from previous dust exposures. The urinary excretion was statistically significantly lower on Monday than on other days. The authors concluded that urinary excretion was affected by exposure to dust, and that the pyrene on the dust was bioavailable (Gardiner et al., 1992a). [The Working Group noted that the pyrene content of the carbon black was not measured. The airborne sampling method and particle size distribution were not described. Rather than performing the regression analyses based on the mean exposures of individuals for the week, it may be more informative to use the daily values in a mixed model that accounts for correlation within the values of each individual. Also, the use of a lag could be informative to account for the time between inhalation of dust, metabolism of pyrene and elimination of 1-hydroxypyrene.]

Thirty carbon black workers (eight of whom were involved in wet pelleting and 22 in packaging) were evaluated for their levels of exposure to PAHs. Urine samples were collected on day 1 pre-shift, day 1 post-shift and day 5 post-shift and tested for 1-hydroxypyrene. The inhalable particle-bound PAHs, gaseous PAHs and dermal exposure to PAHs were measured concomitantly. The sampling train contained a filter cassette to collect particles and determine particle-bound PAHs and a sorbent tube to measure gaseous PAHs. Exposure to pyrene was statistically significantly correlated with exposure to PAHs. The results of a multiple linear regression analysis showed no correlations on post-shift day 1, but the values for exposure of the packaging workers to gaseous PAHs and inhalable particle-bound PAHs and dermal exposure to particle-bound PAHs were significantly correlated on post-shift day 5 (Tsai et al., 2002).

(a) Rodent respiratory tract

As in humans and other species, the deposition of particles in the rodent respiratory tract depends on particle characteristics, airflow properties and airway structure. Rats are the most frequently used animals in experimental studies of inhaled particles, and some aspects of the rat respiratory tract that influence the kinetics of particle deposition therein include breathing pattern (nose or mouth), level of activity (resting or exercise) and lung structure (head airways and tracheobronchial branching pattern) (Miller, 2000). Rats are obligatory nose breathers, while humans breathe both through the nose and the mouth, the extent of which varies among individuals and also depends on level of activity (with exertion, the proportion of breathing through the mouth generally increases). Rats have more extensive airways in the nasal region; therefore, particle deposition in this region is greater in rats than in humans. The size of particles that are inhalable (capable of entering respiratory tract) differs between rats and humans (Ménache et al., 1995). The airway branching system is symmetric (bi- or tripodal) in humans and asymmetric (monopodal) in rats, which influences the site of deposition (airway impaction tends to be greater in the human tracheobronchial region), and, unlike humans, rats do not have respiratory bronchioles. These factors influence the kinetics of particle deposition in the respiratory tract (Ménache et al., 1996).

Once particles are deposited, their removal or retention are based on mechanisms of biological clearance. As for humans, particles in the tracheobronchial region of rats are cleared by the mucociliary pathway and by macrophages in the alveolar region. Particles that enter the interstitium may also enter the lymph and blood circulation. Differences in these physical and physiological factors can result in differences in the clearance rates among species. While tracheobronchial clearance is relatively rapid in both rats and humans (half-times of the order of hours to days), the normal alveolar clearance rate in rats is approximately 10 times faster than that in humans (Snipes, 1989).

Studies in rodents (primarily rats and mice) have shown that the long-term retention of particles is greater than would be predicted from rodent studies that used lower concentrations or durations of exposure. This increase in particle retention has been attributed to the excessive particle loading in alveolar macrophages and impairment of the clearance they mediate (Morrow, 1988; ILSI Risk Science Institute Workshop Participants, 2000). At sufficiently high doses, impaired clearance persists, especially in rat lungs (Bermudez et al., 2002, 2004; Elder et al., 2005). Muhle et al. (1990a) reported impaired alveolar clearance in rats that began at a retained particle mass dose of ∼0.5 mg/rat lung and had essentially ceased at ∼10 mg/rat lung (fine particles of unit density). In overloaded lungs, particles can translocate more readily to the lung interstitial and lymph nodes, and the fraction that migrates to the lymph nodes increases as the particle size decreases (Bellmann et al., 1989).

Lung responses to overloading in rats include increased lung weight, chronic inflammation, fibrosis and lung tumours (Muhle et al., 1991). Overloading was originally defined in terms of particle mass or volume dose (Morrow, 1988). However, Morrow (1992) noted that volumetric overloading did not explain the greater retention of ultrafine particles than that expected for a given mass or volume particle dose. Tran et al. (1999, 2000) developed a biomathematical exposure–dose–response model in which overloading in rats was based on particle surface area dose and provided a better fit to the experimental data evaluated. Ultrafine carbon black particles may be retained in the lungs to a greater extent than larger respirable particles because they escape alveolar macrophage phagocytosis (Renwick et al., 2001, 2004) and enter the lung interstitium (Ferin et al., 1992, 1994).

Overloading, as originally defined, refers only to poorly soluble, fine-sized particles of low toxicity. Other factors can also cause impaired clearance, increased particle retention and lung responses similar to those observed in overloading. These factors include cytotoxicity, such as generation of reactive oxygen species on the particle surface (e.g. crystalline silica) (Vallyathan et al., 1988), or escape from uptake by alveolar macrophages and entrance into the lung interstitium, as observed for ultrafine particles (Ferin et al., 1992, 1994; Renwick et al., 2001, 2004). Cytotoxic and ultrafine particles result in impaired clearance at mass doses that are much lower than those associated with classical overload (Muhle et al., 1990a; Bellmann et al., 1991; Morrow, 1992).

Several reviews, most of which focus on particle toxicity and carcinogenicity, have also described the retention kinetics of particles (including carbon black) after their deposition in the lungs of experimental animals (Morrow, 1988; Snipes, 1989; Kreyling, 1990; Morrow, 1992; Muhle et al., 1994; Oberdörster, 1995).

Several studies that are summarized in Table 4.1 evaluated the clearance and retention of different carbon black materials after deposition into the lung following intratracheal instillation into and inhalation by mice and rats. Bowden and Adamson (1984) instilled a very large dose (4 mg, i.e. 4% of the weight of a mouse lung) of colloidal carbon (primary particle size, 30 nm diameter) into the trachea of 60 Swiss Webster mice and followed its clearance in groups of three mice killed at intervals over a 6-month period. Most of the carbon black was cleared via the mucociliary escalator, but some transepithelial passage via type I cells also occurred. Heavily laden alveolar macrophages stayed in the lungs for the whole observation period, and there was some, although low, clearance via the lymphatic system. No quantitative results were reported.

Table 4.1

Kinetics of carbon black (CB) in experimental animals.

Lee et al. (1987) exposed male Fischer 344 rats by inhalation to 6 mg/m³ carbon black with a MMAD of 0.22 µm in whole-body exposure chambers for 20 hours per day on 7 days per week for 1–11 weeks (for details, see Table 4.1). Immediately following exposure to carbon black, rats were exposed by nose-only inhalation to ¹⁴C-labelled diesel exhaust particulates for 45 minute and were followed for 1 year. Inhibition of lung clearance was inferred by the increased retention of radioactive diesel particles as a percentage of initial lung deposition. The percentage of retained particles increased with increasing exposures. The long-term retention half-times (estimated with a two-phase lung retention model with a sequestration term) were 57, 96 and 140 days for the 1-, 3- and 5-week exposure groups, respectively. The results at 11 weeks were not reported. The pulmonary retention of carbon black was similar to that reported for diesel exhaust by Strom et al. (1989).

Strom et al. (1989) measured the retention of carbon black (furnace black) in rat lungs and thoracic lymph nodes. Male Fischer 344 rats were exposed by inhalation (wholebody) to 7 mg/m³ carbon black for 20 hours per day on 7 days per week for 1, 3 or 6 weeks and were followed for 1 year. Particle size was 0.07 µm CMD with a MMAD of 0.24 µm. Lung burdens of 1.1, 3.5 and 5.9 mg carbon black were achieved after 1, 3 and 6 weeks of exposure, respectively; the 1-year retention fractions were 8, 46 and 61% of the lung burden at the end of the exposure periods, respectively. At the higher doses, clearance was reduced, and the main transport of particles from the lungs was to the lung-associated lymph nodes. The proportion of carbon black transported to the thoracic lymph nodes increased with increasing exposure—1, 21 and 27% of the initial lung burden at 1, 3 and 6 weeks of exposure, respectively. The authors concluded that a carbon black lung (macrophage compartment) burden in the rat of ∼0.8 mg results in a doubling of the normal retention half-time of about 50 days.

Impaired alveolar clearance and increased particle retention were also observed in an inhalation study with several particle types, including carbon black, in Wistar rats (Creutzenberg et al., 1990; Muhle et al., 1990b). The carbon black (furnace black) was Printex 90, with a primary particle size of approximately 0.014 µm and an MMAD of 0.64 µm. Female Wistar rats were exposed by inhalation (in whole-body chambers) to 7.4 ± 1.5 mg/m³ for 19 hours per day on 5 days per week for 4.5 months. The carbon black retained in the lungs at the end of 4.5 months of exposure was 13.7 ± 2.0 mg. The retention half-time of subsequently inhaled ⁸⁵Sr-labelled polystyrene test particles was 472 days in these rats compared with 61 days in air controls. After the 4.5-month exposure to 7.4 mg/m³, rats were subsequently exposed to 12 mg/m³ for 19 hours per day on 5 days per week for up to 24 months (Creutzenberg et al., 1990; Muhle et al., 1990a,b, 1994). Some groups were exposed for 18 months and then removed from exposure for 6 months. At 3, 6, 12, 18, 22 and 24 months of exposure, the lung and lung-associated lymph node burdens were measured. The highest carbon black lung burden was 50.2 ± 10.9 mg at 18 months, and the lymph node burden was 6.7 mg at 22 months. Interstitial fibrosis was observed in these rats at 12 and 18 months. The pulmonary retention half-times of radiolabelled tracer particles were determined at 3, 12 and 18 months of exposure, and at 18 months followed by 6 months of clean air. The retention half-time for carbon black was 550 days (95% CI, 322–1868 days) following termination of exposure. Test particle clearance of ⁵⁹Fe₂O₃ (0.35 µm in diameter) was significantly prolonged with increasing duration of exposure to carbon black, with a half-time that ranged from 244 to 591 days compared with 61–96 days in air controls. In contrast, clearance of ⁸⁵Sr-labelled polystyrene microsphere (3.5 µm diameter) showed only prolonged retention after 3 months of exposure to carbon black with a half-time of 472 days whereas, at the 12- and 18-month exposure time-points, test particle clearance returned to control values of about 50–60 days. The authors suggested that this was due to a change in the deposition site of the larger ⁸⁵Sr-labelled polystyrene microspheres as a result of altered lung architecture (in response to carbon black-induced inflammation and other changes) and breathing pattern, and concluded that the retardation of clearance was detectable in rats when the retained lung burden of various dusts exceeded 0.5 mg, and that a substantial decrease in the clearance rate was observed at lung burdens exceeding 10 mg (Creutzenberg et al., 1990; Muhle et al., 1990b).

Henderson et al. (1992) evaluated the pulmonary retention in Fischer 344 rats of furnace black (Elftex 12) inhaled at three different dose rates such that the product of concentration × time was very similar (392 mg × h/m³ per week). Lung burdens were 3–4 mg. The retention half-time determined over a 24-week period after exposure was not statistically significantly different among the different groups (∼520 days; 95% CI, 350–950 days).

Mauderly (1994) studied the retention of tracer doses of [⁷Be]furnace black (Elftex 12) in Fischer 344/N rats 3 and 18 months after chronic exposure to two concentrations of unlabelled carbon black (2.5 mg/m³ and 6.5 mg/m³). Clearance of the labelled carbon black followed a two-exponential model. The most striking difference was found in the slow-phase clearance component, which showed little or no clearance over a period of 126 days for the low- and high-dose groups compared with retention half-times of 113 and 135 days for control rats.

In a study of chronic inhalation in Wistar rats and NMRI mice exposed to furnace black (Printex 90; 11.6 mg/m³), pulmonary particulate accumulation was measured (Heinrich et al., 1995). The rats were exposed for 18 hours per day on 5 days per week for 24 months; the mice were similarly exposed for 13.5 months. Both rats and mice showed similar accumulation kinetics over the exposure time; at 1 year of exposure, the normalized lung burden (mg/g of control lung) was 32 mg in rats and 37 mg in mice. In addition, rats showed significantly prolonged retention of tracer particles compared with controls as early as 3 months after exposure, which persisted through 12 and 18 months of exposure and 3 months after the 18-month exposure (see Creutzenberg et al., 1990).

(b) Retention of intratracheally instilled ultrafine carbon black particles in healthy and injured lungs of mice

Using a rodent model of lung susceptibility, Adamson and Hedgecock (1995) and Adamson and Prieditis (1995) examined the particle distribution and retention of carbon black in healthy or injured (bleomycin-treated) lungs. Following treatment with bleomycin (0.15 units, by intratracheal instillation), male Swiss mice received 2 mg 40-nm carbon black in hydrolysed gel (also by intratracheal instillation) either three days or four weeks later. Groups of four mice were killed at various times up to 16 weeks after administration of the carbon black. Additional groups received carbon black only or bleomycin only. In the carbon black-only group, histological examination a few days after instillation showed that most of the carbon black was inside alveolar macrophages and polymorphonuclear leukocytes, although some particles were seen in the interstitium and interstitial macrophages (remaining for 16 weeks, when most of the alveoli were clear of inflammatory cells and particles); particles were also found in the hilar lymph nodes at 1 week, the amount of which increased by 16 weeks. In contrast, in the mice receiving carbon black 3 days after bleomycin, particles were seen to cross the denuded epithelial surface and, by 4 weeks, ‘many carbon black-laden cells’ were seen in the connective tissue; by 16 weeks, ‘a large amount of carbon black’ had been incorporated into the interstitium. In mice treated with carbon black four weeks after treatment with bleomycin, particles were again seen mostly in the air spaces (free or phagocytosed); although the alveolar surface was not denuded, cell composition was abnormal (cuboidal epithelium in fibrotic areas). The amount of carbon black retained in the lungs was assessed at 16 weeks (by digestion of the lungs in 40% potassium hydroxide). The weight of the insoluble residue at 16 weeks was statistically significantly greater (1.6 mg) in mice that received carbon black 3 days after bleomycin (when lung injury was greatest) than in mice that received either carbon black only or carbon black 4 weeks after bleomycin (∼1 mg). The unexposed mice and those treated with bleomycin only had approximately 0.2 mg of insoluble residue. This study shows that the retained lung dose of carbon black can increase significantly during a condition of pulmonary inflammation and epithelial cell injury.

(c) Comparison of clearance and retention of carbon black in lungs of three rodent species

The lung retention of and response to inhaled carbon black particles (Printex 90 and Sterling V) were investigated in three rodent species: Fischer 344 rats, B6C3F₁ mice and 276 F1B Syrian hamsters (all females) (Elder et al., 2005). The Printex 90 had a primary particle size of 14 or 17 nm (both were reported), a specific surface area of 300 m²/g and an MMAD of 1.4–2.0 µm (GSD, 2.3–2.8) for the various exposure chambers by rodent species and exposure group. The Sterling V had a primary particle size of 70 nm, a specific surface area of 37 m²/g and an airborne particle size of 0.8 µm MMAD (GSD, 3.2). Printex 90 was labelled as a high-surface area carbon black, while Sterling V was labelled as low-surface area carbon black. Rats were exposed to low, medium and high concentrations of Printex 90 of approximately 1, 7 and 50 mg/m³ for each species, respectively, for 6 hours per day on 5 days per week for 13 weeks. In addition, rats only were exposed to Sterling V at a concentration of approximately 50 mg/m³. Five or six animals were used per exposure group. The study was designed to provide the same dose, as either mass or surface area, for the two types of carbon black studied. Thus, although the mass doses were different, similar surface area doses were achieved in the rat lungs from exposure to 7 mg/m³ Printex 90 and to 50 mg/m³ Sterling V, i.e. approximately 0.3 m² (Elder et al., 2005). Although the surface area doses were different, similar mass doses were achieved at 50 mg/m³ Printex 90 or Sterling V, i.e. approximately 5.5 mg or 8 mg, respectively. Particle retention in the lungs was observed to be prolonged after exposure to the mid-(7 mg/m³) and high (50 mg/m³) concentrations of Printex 90 in rats and mice, and also for 50 mg/m³ Sterling V in rats. In hamsters, which had the most efficient clearance, pulmonary retention was prolonged only at the high dose.

(d) Translocation of carbon black particles from the site of deposition to other tissues

Female Swiss mice, aged 4 weeks and 18 months, were given with 7 mg ⁷Be-labelled furnace black particles (Elftex 8) by gavage. The distribution of the isotope was determined in the animals 4 hours and 1, 2, 5 and 14 days after exposure. The authors concluded that there was uptake and distribution from the gut and that transit was more rapid in young mice. Peyer's patches (a gut-associated lymphoid tissue) of older mice took up more radiolabel than those of younger mice (LeFevre & Joel, 1986). [It was not clear from the study whether the authors verified the stable binding of the radiolabel to the particles.]

In a study of ultrafine carbon black and other particles instilled in rat lungs, Oberdörster et al. (1992) determined that the translocation of particles from the alveolar lumen of the lungs was dependent on particle size. Following intratracheal instillation of 0.5 mg particles of different sizes, the smaller ultrafine particles (12 and 20 nm) penetrated the alveolar epithelial cell barrier and entered the lung interstitium to a greater extent than an equal mass of larger respirable particles (> 200 nm) within 24 hours. This proportion was shown to increase with increasing particle dose as either mass or surface area.

More recent studies have shown that ultrafine carbon and other particles can translocate beyond the lungs. Oberdörster et al. (2002) showed that inhaled spark-generated ultrafine ¹³C-carbon particles of approximately 25 nm in diameter were cleared rapidly from rat lungs and translocated to other organs (e.g. liver and spleen). Significant amounts of particles were found in the livers of rats in the high-exposure group (approximately fivefold higher amounts in the liver than in the lung at 24 hours). Clearance or translocation from the lungs may also depend on the composition of the particle. For example, ultrafine iridium particles inhaled by rats for 1 hour remained in the lungs to a much greater extent and only a small proportion was cleared (< 1% in 7 days) (Kreyling et al. 2002). However, of the iridium particles that did translocate from the lungs, 10 times more 15-nm particles translocated than 80-nm particles. In another study in rats, inhaled ultrafine elemental silver particles were found to enter the blood circulation (Takenaka et al., 2001).

Inhalation of ultrafine particles may also result in translocation of particles to the brain. Ultrafine insoluble ¹³C-carbon particles (CMD, 36 nm; GSD, 1.66) were found in the brains of Fischer 344 rats on days 1–7 following a 6-hour inhalation exposure to 160 µg/m³ (Oberdörster et al., 2004). Approximately 50% of the inhaled ultrafine particles was predicted to deposit in the olfactory mucosa (assuming equal distribution) of rats and approximately 20% of that amount was found in the olfactory bulb. On day 1 after exposure, 0.35 µg/g of added ¹³C was detected in the olfactory bulb; the amount increased on days 3 and 5 after exposure and reached 0.43 µg/g on day 7. The cerebrum and cerebellum contained significantly increased concentrations of ¹³C on day 1, but the levels tended to decrease subsequent to exposure. The study was not designed to distinguish between the possible paths through which ¹³C ultrafine particles could reach the brain, including crossing the blood–brain barrier (by particles that translocated into the blood following deposition anywhere in the respiratory tract) and transport of particles that deposited in the nasal olfactory mucosa along the olfactory nerve to the olfactory bulb. However, the authors concluded that the olfactory nerve pathway was the most probable explanation for the ¹³C found in the olfactory bulb because of the significant increase in amounts in that region and the consistency with previous studies that demonstrated an olfactory nerve pathway for ultrafine particles (Bodian & Howe, 1941; De Lorenzo, 1970). Studies in non-human primates have demonstrated the translocation of 30-nm viruses and 50-nm gold particles from the nasal region to the olfactory bulb of the brain. Hunter and Dey (1998) reported another pathway through which particles may enter the central nervous system, via the trigeminal nerve, which has synaptic innervation in the nasal epithelium.

The size of individual ultrafine particles may allow their entry into cells and cellular organelles more readily than larger particles or agglomerates. In a study of concentrated particles from air pollution (including carbon particles) in human bronchial epithelial cells and mouse alveolar macrophages, the ultrafine fraction (< 100 nm) was found to penetrate the cells, localize in mitochondria and cause oxidative damage to mitochondrial membranes (Li et al., 2003).

(e) Kinetics of carbon black-adsorbed material

Concern had been raised that material, including carcinogenic compounds, adsorbed onto carbon black particles are retained longer in the lung upon inhalation and will subsequently lead to a greater availability of carcinogens to target cells in the lung. In particular, this would be of importance for materials such as diesel exhaust particles, which are known to contain PAHs adsorbed onto the carbon core and which may contribute to the carcinogenic response of inhaled diesel exhaust. These studies are summarized in Table 4.2.

Table 4.2

Kinetics of carbon or carbon black (CB)-adsorbed compounds.

Pylev et al. (1970a,b) instilled [³H]benzo[a]pyrene adsorbed onto furnace black particles (26–160 nm) intratracheallly into Syrian hamsters and followed retention of radioactivity for 21 days. Compared with [³H]benzo[a]pyrene suspended in aminosol vitrum, retention of [³H]benzo[a]pyrene was longer when adsorbed onto carbon black (Pylev et al., 1970b).

In another study, male Fischer 344/Crl rats were exposed by inhalation for 30 days to Elftex 12 (furnace black; primary particle size, 37 nm; surface area, 43 m²/g) with adsorbed [7-¹⁴C]benzo[a]pyrene (Sun et al., 1989) or [4,5,9,10-¹⁴C]-1-nitropyrene (Wolff et al., 1989). A total concentration of 100 mg/m³ was used with the addition of either 0.2, 2 or 20% benzo[a]pyrene or 2 mg/m³ 1-nitropyrene. The long-term retention of radioactivity from both benzo[a]pyrene and 1-nitropyrene was increased when adsorbed onto carbon black. For both adsorbed compounds, a biphasic clearance was found, and most radioactivity was cleared from the lungs within 1–2 days. At all time-points, 16–60 times more radioactivity was retained after treatment with the adsorbed compounds compared with administration of the pure compound. Covalent interaction of these compounds with lung macromolecules was also greater when they were co-administered with carbon black particles.

These studies demonstrate that carbon black administered to rats and hamsters either by inhalation or intratracheal instillation can act as a carrier of adsorbed material, which is subsequently cleared from the lung much more slowly than the material given alone. In another study, Buddingh et al. (1981) reported that benzo[a]pyrene was poorly eluted from carbon black in vitro by human plasma or by swine serum, swine lung washing or lung homogenate, which is consistent with the findings of Borm et al. (2005) in surfactant-containing saline solution using four different carbon blacks.

(a) Observations in the general population

Chest radiographic features of small opacities that are consistent with pneumoconiosis have been observed in the general population. An analysis of nine study populations reported prevalences of small opacities (International Labour Organization (ILO) grade 1/0 or greater) ranging from 0.21 to 11.7%. A meta-analysis of these data yielded a population prevalence of 5.3% (95% CI, 2.9–7.7%). The prevalence was significantly greater in Europe (11.3%; 95% CI, 10.1–12.5%) than in North America (1.6%; 95% CI, 0.6–2.6%), which could not be explained on the basis of age, gender or smoking history. There was a greater prevalence of lung opacities in men (5.5%; 95% CI, 3.4–7.6%) than in women (3.5%; 95% CI, 1.3–5.8%). The age-specific pooled prevalence was higher in the study populations with a mean age of ≥ 50 years than in those with a mean age of < 50 years in both Europe (11.7% versus 9.6%) and North America (2.3% versus 0.6%). Environmental and unaccounted occupational exposures as well as reader variability may play a role in the determination of the prevalence of small opacities in these subjects and may explain the large differences between different regions (Meyer et al., 1997).

(b) Respiratory effects in carbon black workers

Gärtner and Brauss (1951) first described radiological changes analogous to pneumoconiosis in 31 workers in a carbon black factory. However, these individuals had no lung function abnormality. Since that time, a series of other reports have been published on pneumoconiosis in carbon black workers.

A health survey was conducted in two German factories that produced carbon black from acetylene or from oil that was burned with light gas, respectively. Among 56 workers, 16 had been employed for more than 10 years. Two of these workers had chest X-ray changes compared with none of the 52 controls who had had radiographs taken without suspicion of lung disease (von Mai, 1966). [The selection of workers was not clear, neither were the criteria for diagnosis.]

Most studies of respiratory morbidity have methodological shortcomings or provide insufficient detail for a reliable interpretation of the results (see review by Gardiner, 1995). Nevertheless, exposure–response relationships were evident for symptoms of chronic bronchitis, small opacities on chest radiographs and several respiratory parameters (forced expiratory volume in 1 second [FEV₁], FEF_25–75%). Studies in Germany (Küpper et al., 1996) and Poland (Szozda, 1994, 1996) provided evidence of a relationship between exposure to carbon black and lung function among smokers. The Polish studies also reported cases of hypertension and pneumoconiosis among carbon black workers.

Spirometry, body plethysmography and inhalation challenge tests were conducted among employees at a German carbon black production plant to assess the impact of fine carbon black dust on pulmonary function, to determine the prevalence of obstructive airway disease among the workers and to investigate whether exposure to fine dust is related to the prevalence of bronchial hyper-responsiveness. A total of 573 exposed workers (178 nonsmokers, 107 former smokers, 288 smokers) and 99 controls (46 nonsmokers, 13 former smokers, 40 smokers) participated in the study. Measurements of dust in air showed concentrations of 0.01–9.14 mg/m³ for fine dust (9–200 nm [includes fine and ultrafine sizes]) and 1.08–19.95 mg/m³ for total dust (mean dust concentrations, 0.58 mg/m³ for respirable dust; 1.08 mg/m³ for inspirable dust). Exposure to carbon black had a small but statistically significant impact on lung function in smokers (P < 0.01). Nevertheless, exposed smokers displayed signs of obstructive airway disease more frequently (7.3%) than exposed nonsmokers (3.9%). There was no effect of exposure to carbon black on lung function in former smokers or nonsmokers. Exposure to carbon black dust was not associated with an increased prevalence of bronchial hyper-reactivity (Küpper et al., 1996).

To investigate the occurrence of medical conditions related to exposure to carbon black, a large study was conducted in 18 carbon black production plants (including the German plant; Küpper et al., 1996) in seven European countries between mid-1987 and mid-1989. A total of 1298 respirable [SIMPEDS cyclone method] and 1317 total inhalable [Institute of Occupational Medicine head method] samples were taken and included in the study. The distributions of the TWA values were best described by a log-normal distribution and exposure was characterized by GMs and standard deviations (Gardiner et al., 1992b). In a subsequent study, exposure-related health effects were assessed in 3086 employees in these plants through respiratory health questionnaires, spirometry and chest radiographs. Personal monitoring was used to measure current exposure to inhalable and respirable carbon black, sulfur dioxide and carbon monoxide. The final analysis comprised 1742 employees in 15 plants (81% response rate) who provided data on respiratory symptoms and spirometry, and 1096 chest radiographs were available from 10 plants (74% response rate). In addition to the respirable (1298) and inspirable (total inhalable; 1317) dust samples mentioned above, 1301 sulfur dioxide and 1322 carbon monoxide samples were also collected. This study thus included a comprehensive assessment of current occupational exposure to carbon black dust and its associated gaseous contaminants. In respirable dust samples, the geometric mean level was 0.21 (GSD, 2.7) mg/m³ and in total inhalable dust, the GM level was 0.57 (GSD, 4.0) mg/m³. Associations were found between cough, sputum production, the symptoms of chronic bronchitis (mean prevalence, 10%) and indices of increasing current exposure (from 0.14 to > 0.45 mg/m³). There was a small reduction in lung function with increasing dust exposure in both smokers and nonsmokers. Nearly 25% of the chest radiographs showed small opacities (ILO category 0/1 or greater), which were strongly associated with indices of cumulative dust exposure, after accounting for production plant and current smoking habits. The findings were consistent with a non-irritant effect of carbon black dust on the airways combined with dust retention in the lungs (Gardiner et al., 1993).

Chronic inflammation has also been associated with non-neoplastic lung diseases in workers with dusty jobs. Rom (1991) found a statistically significant increase in the percentage of polymorphonuclear neutrophils in the bronchoalveolar lavage (BAL) fluid of workers with respiratory impairment who had been exposed to asbestos, coal or silica (4.5% in cases versus 1.5% in controls). Elevated levels of such cells have been observed in the BAL fluid of miners with simple coal workers' pneumoconiosis (31% of total BAL cells versus 3.4% in controls; Vallyathan et al., 2000) and in patients with acute silicosis (a 10-fold increase over controls; Goodman et al., 1992; Lapp & Castranova, 1993).

The results of two additional studies (phases 2 and 3) of respiratory health of European carbon black workers showed exposure-related adverse effects of carbon black on the respiratory system, which were evident from an increase in the prevalence of cough and sputum production, and reductions in lung function, based on measurements of FEV₁, FEF_25–75% and the FEV₁/forced vital capacity (FVC) ratio. An increase in exposure to inhalable dust of 1 mg/m³ was associated with an increase of 80% in the prevalence of respiratory symptoms of chronic bronchitis (odds ratio, 1.8; 95% CI, 1.3–2.6) in phase 2, but not in phase 3. The prevalence of respiratory symptoms such as cough and cough and sputum production, however, was significantly affected by an increase of 1 mg/m³ in exposure. Working for 40 years with a mean exposure of 1 mg/m³ (480 mg.month/m³) was expected to increase the prevalence of cough by almost 70% (odds ratio, 1.7; 95% CI, 1.2–2.1) and that of cough and sputum production by 60% (odds ratio, 1.6; 95% CI, 1.2–2.1). Similarly, a 1-mg/m³ increase in exposure to carbon black was associated with significant decrements in FEV₁, FEF_25–75% and FEV₁/FVC ratio (Gardiner et al., 2001).

Van Tongeren et al. (2002) carried out a longitudinal analysis of workers in the European carbon black manufacturing study who had provided a full-size chest radiograph in each of the three cross-sectional surveys between 1987 and 1995. All chest radiographs were read independently according to the ILO classification by three experiences readers who were blind to all factors, including the sequence in which the chest radiographs were taken. After exclusion of all workers from a factory that had a low participation rate (< 60%) in the first survey and all workers who had reported various lung injuries, operations or respiratory disease (asthma, pleurisy or pulmonary tuberculosis), data from 675 employees were available for analysis. The prevalence of small opacities with ILO category ≥ 1/0 was 13.9, 19.9 and 19.7% in the first, second and third survey, respectively. An association between cumulative exposure during the study and progression of small opacities was observed, although only four cases of existing small opacities (≥ 1/0) in the first survey progressed to higher ILO categories. The authors concluded that exposure to carbon black was associated with the incidence of small opacities, although this effect may be reversible after cessation of exposure.

Harber et al. (2003) investigated whether exposure to carbon black was associated with decrements in lung function and increased prevalence of respiratory symptoms among 1755 employees from 22 North American carbon black manufacturing plants. Multiple linear regression analyses showed that cumulative exposures to ‘total’ and inhalable dust were both was associated with a statistically significant decrement in FEV₁ and with FVC. The slopes were −2 mL and −0.7 FEV₁/mg–year/m³ for cumulative exposure to ‘total’ and inhalable dust, respectively. Cumulative exposure was also associated with an increased prevalence of chronic bronchitis in nonsmokers.

(a) Inhalation exposure

The effects of subchronic inhalation of carbon black on pulmonary inflammation, expression of inflammatory cytokines and growth factors, and on lung histopathology were studied in male Fischer 344 rats exposed for 6 hours per day on 5 days per week for up to 13 weeks to 1.1, 7.1 and 52.8 mg/m³ carbon black (Monarch 880, Cabot; diameter, 16 nm; surface area, 220 m²/g). Effects on the lung were assessed after 6.5 and 13 weeks of exposure and after 3 and 8 months of recovery. After 13 weeks, lung burdens were 354, 1826 and 7861 µg carbon black at the three exposure concentrations, respectively. Inhalation of 1.1 mg/m³ carbon black did not cause any of the adverse effects on the lung that were measured, but lung clearance appeared to be impaired after exposure to 7.1 and more severely so after exposure to 52.8 mg/m³. Analysis of BAL fluid showed no effect of the lowest dose and a relative increase in the number of neutrophils after exposure to the intermediate dose that persisted. At the highest dose of carbon black, an increase in total cell number and neutrophils and a decrease in macrophages were observed. The BAL fluid concentrations of lactate dehydrogenase, β-glucuronidase and total protein were also increased at this dose. All these effects persisted until 8 months after exposure. mRNA expression of macrophage inflammatory protein 2 (MIP-2) and monocyte chemotactic protein 1 (MCP-1)—two chemotactic cytokines—was minimal in the lungs of rats in the low-dose group, but MIP-2 mRNA was clearly present at all time-points after exposure to 7.1 and 52.8 mg/m³. MCP-1 mRNA was also increased at these doses, but this effect was persistent for up to 8 months after exposure to the high dose only. In lung tissue sections, particle-containing macrophages were seen in alveolar and alveolar duct regions after the 1.1-mg/m³ dose. The intermediate dose produced acute inflammation (characterized by accumulation of neutrophils and macrophages within alveolar spaces), mild epithelial hyperplasia and mild interstitial fibrosis. The 52.8-mg/m³ dose caused mainly lesions in the alveolar ducts, with pronounced epithelial hyperplasia and fibrosis. Alveolar type II cell hypertrophy and hyperplasia seen after exposure to intermediate and high doses persisted throughout the 8-month recovery period (Driscoll et al., 1996).

To investigate whether the inflammatory response induced by inhaled ultrafine particles involves an increased release of systemic clotting factor, adult male Wistar rats were exposed by inhalation for 7 hours to fine or ultrafine carbon black particles. The attained total suspended particle concentrations were 1.66 mg/m³ for ultrafine (Printex 90; diameter, 14 nm) and 1.40 mg/m³ for fine carbon black (Huber 990; diameter, 260 nm). Particle concentration (number of particles/m³) of the ultrafine carbon black was more than 10 times greater than that of the fine particles; the average CMDs were 114 nm for ultrafine and 268 nm for fine carbon black. Exposure to ultrafine particles caused an increase in total cell number and in the number of neutrophils in BAL fluid immediately after exposure. Both fine and ultrafine carbon black caused twofold and fourfold increases, respectively, in the number of polymorphonuclear leukocytes in BAL 16 hours after exposure. Exposure to ultrafine but not to fine carbon black particles was associated with a significant increase in the total number of blood leukocytes. Blood coagulation-related plasma, fibrinogen, factor VII and von Willebrand factor were all unaffected by exposure to particles. MIP-2 mRNA was significantly increased in BAL cells 48 hours after the end of exposure to ultrafine carbon black. The data showed a small but consistent pro-inflammatory effect of ultrafine particles that was greater than that of the same exposure (on a weight/volume basis) to fine carbon black (Gilmour et al., 2004).

The retention kinetics, inflammation and histopathology following exposure to carbon black were examined in female Fischer 344 rats, B6C3F₁ mice and F1B Syrian golden hamsters exposed to 0, 1, 7 and 50 mg/m³ (nominal concentrations) carbon black particles (Printex 90; diameter, 14 nm; surface area, 300 m²/g) for 6 hours per day on 5 days per week for 13 weeks. Rats were also exposed to 50 mg/m³ (nominal) low-surface area carbon black (Sterling V; diameter, 70 nm; surface area, 37 m²/g). Retention and effects were measured immediately after exposure and 3 and 11 months later; retention was also evaluated after 5 weeks of exposure. Significant decreases in body weight were observed only in hamsters exposed to the high dose of carbon black. Lung weights were increased in all groups exposed to this dose, but this persisted only in rats and mice up to 11 months after exposure. Lung inflammation and histopathology (lung lesions located primarily in the centriacinar regions, with the most extensive epithelial and inflammatory responses in the alveolar ducts and surrounding parenchyma) were more severe and prolonged in rats than in mice and hamsters, and were similar in rats exposed to ‘surface-area equivalent’ concentrations of 7 mg/m³ Printex 90 and 50 mg/m³ Sterling V. Hamsters had the most efficient clearance and least severe responses of the three species. The results obtained in rats suggest that the surface area of the particles is an important determinant of dose to the target tissue and subsequent effects (Elder et al., 2005).

(b) Intratracheal or intranasal instillation

Respiratory syncytial virus causes bronchiolitis and pneumonia in infants and may lead to the development of asthma in childhood. To determine whether exposure to particles modulates the immune response to this virus, 8-week-old female BALB/c mice received an intratracheal instillation of 40 µg ultrafine carbon black particles (150 m²/g) in 100 µL saline. The following day, mice were instilled with either 10⁶ plaque-forming units of respiratory syncytial virus or uninfected medium. Compared with animals that received the virus alone, tumour necrosis factor-α (TNFα) protein was reduced in the BAL fluid on days 1 and 2 of infection in mice exposed to both carbon black and the virus. There was a reduction in the number of lymphocytes in the BAL fluid on day 4, and decreased levels of interferon (IFN)-γ-inducible protein lymphotactin and IFN-γ mRNAs in the lungs of mice exposed to carbon black plus virus. On days 2–4 of infection, viral titres in these mice were lower than those in animals that had received respiratory syncytial virus alone. By day 7, however, the numbers of neutrophils, expression of pro-inflammatory cytokine mRNA, TNF-α and Th2 cytokine interleukin (IL)-13 protein levels were increased in the lungs of mice exposed to carbon black plus virus, which indicated an exacerbation of infection. The data showed that pre-exposure to ultrafine particles induces an inflammatory condition that promotes Th2-type immune responses rather than the production of IFN-γ Th1, which is necessary for microbial defence (Lambert et al., 2003).

The ability of ultrafine and fine particles to induce inflammation, cause epithelial injury and affect alveolar macrophage clearance (phagocytosis, chemotaxis) was studied in Wistar rats instilled with 125 or 500 µg fine titanium dioxide (mean diameter, 250 nm; 6.6 m²/g), ultrafine titanium dioxide (mean diameter, 29 nm; 49.8 m²/g), fine carbon black (Huber 990; mean diameter, 260.2 nm; 7.9 m²/g) or ultrafine carbon black (Printex 90; mean diameter, 14.3 nm; 253.9 m²/g) in 0.5 mL saline. Inflammation was quantified by counting the number of neutrophils in BAL fluid. The ultrafine particles recruited more polymorphonuclear neutrophils, caused more epithelial damage and were more cytotoxic than fine particles at equal mass concentrations. Both ultrafine and fine particles significantly impaired the ability of alveolar macrophages to phagocytose fluorescent indicator beads, but only treatment with ultrafine particles enhanced the C5a-stimulated chemotactic potential of the macrophages. This study showed that ultrafine particles [of two very different materials] induced inflammation and epithelial damage to a greater extent than their larger-sized mass counterparts. In general, the effect of ultrafine carbon black was greater than that of ultrafine titanium dioxide, which suggests that there are differences in the potential hazard of different types of ultrafine particle. Epithelial injury and toxicity were associated with the inflammatory response that followed exposure to ultrafine particles. Increased sensitivity to a C5a chemotactic stimulus as a result of exposure to ultrafine particles could retain the macrophages in the lung at the site of particle deposition, and thus allow the dose to accumulate (Renwick et al., 2004).

To explore the role of vascular endothelial growth factor (VEGF) in the induction of alveolar capillary permeability by ultrafine particles, male ICR mice received an intratracheal instillation of 200 µg carbon black (Printex 90; diameter, 14 nm before grinding; surface area, 253.9 m²/g). A significant and sustained increase in total proteins was observed in BAL fluid, which was maximal at 21 hours after instillation. The level of TNFα was significantly elevated only at 4 hours, but significant increases in VEGF were seen throughout the 42-hour study period, with a peak increase at 16 hours. The results showed that ultrafine carbon black particles induce the production of VEGF, which is associated with an increase in alveolar capillary permeability. The involvement of oxidative stress in this process was supported by the observation in an in-vitro study that N-acetylcysteine (a scavenger of reactive oxygen species) prevents the induction of VEGF by ultrafine carbon black particles (Chang et al., 2005).

The kinetics of airway toxicity or inflammation and allergic sensitization to ovalbumin in response to ultrafine carbon black particles (diameter, 30–50 nm) was studied in BALB/cANNCrl mice exposed intranasally to ovalbumin (10 µg in 20 µL) alone or in combination with 2, 20 or 200 µg carbon black particles. Airway toxicity and inflammation were assessed on days 4 and 8, immune adjuvant effects were measured in the lung-draining peribronchial lymph nodes on day 8, antigen-specific immunoglobulin E (IgE) was measured on days 21 and 28 and allergic airway inflammation was studied after ovalbumin challenges on day 28. The dose of 200 µg carbon black particles, but not 20 µg or 2 µg, induced immediate airway inflammation and had immune adjuvant activity that involved enlargement of the peribronchial lymph nodes and an increased ovalbuminspecific production of Th2 cytokines IL-4, IL-5 and IL-10. Serum levels of ovalbuminspecific IgE were increased on day 21, which was indicative of systemic sensitization. This was supported by allergic airway inflammation after challenges with ovalbumin. The authors concluded that there is a correlation between early airway toxicity and adjuvant effects of carbon black particles and that local cytokine production early after exposure to these particles is predictive of airway inflammation (de Haar et al., 2005).

The size-specific effects of particles on pulmonary immune response, translocation to lymph nodes and expression of chemokine mRNA were studied in the lung and lymph nodes of 8-week-old male BALB/c mice exposed to ultrafine or fine carbon black particles by intratracheal instillation. In a first experiment, 25, 125 or 625 µg ultrafine carbon black particles (Printex 90; diameter, 14 nm; 300 m²/g) were administered once a week for 4 weeks. In a second experiment with the same dose regimen, larger-sized carbon black (Flammruss 101; diameter, 95 nm; 20 m²/g) was instilled. Total and differential cell counts and release of cytokines and chemokines were measured in BAL fluid 24 hours after the last instillation. In a third experiment, a dose of 125 µg ultrafine carbon black or larger-sized carbon black was administered according to the same schedule, and lungs and mediastinal lymph nodes were isolated 4 hours after the last instillation to measure expression of chemokine mRNA. The total cell count and differential cell counts (macrophages, lymphocytes, neutrophils) in BAL fluid increased significantly in mice exposed to the ultrafine carbon black particles in a dose-dependent manner, as did the release of IL-1β, IL-6 and TNFα. MIP-1 α/CCL-3 protein and mRNA expression were also increased in the lungs and lymph nodes of these mice. The effects seen with the 95-nm carbon black particles were weaker than those obtained with the smaller-sized particles. Particle translocation to the mediastinal lymph nodes was greater in mice given the ultrafine particles than in those that received the larger-sized carbon black. The study showed that repeated intratracheal instillation of ultrafine carbon black particles in mice leads to pulmonary inflammation, translocation of particles to mediastinal lymph nodes and enhanced expression of chemokine mRNA in the lung and lymph nodes. These effects were stronger with ultrafine than with fine particles (Shwe et al., 2005).

The effects of ultrafine and fine particles on immune function in the mouse brain were investigated by the instillation of 125 µg carbon black (Printex 90; diameter, 14 nm; 300 m²/g; or Flammruss 101; diameter, 95 nm; 20 m²/g) into the nostrils of 8-week-old male BALB/c mice once a week for 4 weeks. Four hours after the last instillation, the olfactory bulb and hippocampus were isolated. The mRNA expression of pro-inflammatory cytokines (IL-1β and TNFα) and chemokines (MCP-1/CCL2, MIP-1α/CCL3) and monokine-induced INF-γ /CXC chemokine ligand was enhanced in the brain olfactory bulb but not in the hippocampus of mice instilled with 14-nm carbon black particles. The 95-nm particles did not show effects in either organ at the doses used (Shwe et al., 2006).

Yang et al. (1999) tested the combined effect of particulates and organic compounds on the alveolar macrophage response to bacteria or bacterial products (such as lipopolysaccharide) and the secretion of pro-inflammatory cytokines (IL-1 and TNFα). A comparative study of the pulmonary responses to exposure to diesel exhaust particles, carbon black and silica was conducted in male Sprague-Dawley rats that were exposed to a single intratracheal dose (5 or 35 mg/kg bw) of diesel exhaust particles (NIST; MMAD, 0.5 µm), carbon black (Elftex 12 furnace black; MMAD, 0.1–0.6 µm), silica (Min-U-Sil; MMAD, < 5 µm) or saline. The alveolar macrophages isolated from the particle-exposed rats were challenged ex vivo with lipopolysaccharide (0.1 µg/10⁶ alveolar macrophages) and cytokines were monitored. In addition, rats were exposed to a single dose of diesel exhaust particles (5 mg/kg bw) followed 3 days later by exposure to lipopolysaccharide (1 mg/kg bw) for 3 hours in vivo. Exposures to diesel exhaust particles, carbon black and silica resulted in polymorphonuclear neutrophil infiltration and elevated levels of albumin and lactate dehydrogenase in the BAL fluid. The alveolar macrophages from the carbon black- and silica-exposed rats showed an increased production of TNFα but not of IL-1 and did not show a decreased response to a subsequent challenge with lipopolysaccharide. Upon ex-vivo challenge with lipopolysaccharide, the alveolar macrophages from diesel exhaust particle-exposed rats showed a significant decrease in TNFα. The authors concluded that diesel exhaust particles, carbon black and silica all induced a pulmonary response due to particle stimulation, but only diesel exhaust particles suppressed cytokine release in alveolar macrophages in response to stimulation with lipopolysaccharide.

Nilsen et al. (1997) studied the adjuvant activity of diesel exhaust particles (NIST 1650; MMAD, 0.03 µm; 64 m²/g) and carbon black (Regal 250R; MMAD, 0.035 µm; 65 m²/g) on systemic IgE production in ovalbumin-treated mice after intranasal administration. Female Balb/cA mice were immunized four times with ovalbumin (20 µg) alone or in combination with diesel exhaust particles (25 µg) or carbon black (25 µg). One and 2 weeks later, increased responses in both the number of responding animals and serum IgE antibody were seen in the animals treated with ovalbumin and either of the particles; the activity of diesel exhaust particles was more pronounced than that of carbon black, which indicated that the organic matter adsorbed to the diesel exhaust particles and the non-extractable carbon cores were both responsible for the observed adjuvant effect.

Al-Humadi et al. (2002) exposed Brown Norway rats intratracheally to saline, carbon black or diesel exhaust particles at 5 mg/kg bw followed by exposure for 30 minute to ovalbumin (90 mg/m³) or saline 1, 8, 18 and 29 days later. Exposure to diesel exhaust particles, carbon black or ovalbumin alone did not result in abnormal levels of inflammatory cells, lactate dehydrogenase or total protein in the BAL fluid. However the combinations of ovalbumin with diesel exhaust particles or carbon black increased these markers, and also the level of IL-4 mRNA in lung tissue and serum levels of ovalbuminspecific IgG and IgE.

The effect of acute exposure to diesel exhaust particles on phase I and phase II enzymes was investigated in rat lung. Intratracheal administration of these particles enhanced cytochrome P450 (CYP) 1A1 protein levels and enzyme activity one day after exposure; enzyme levels returned to control values after five days. Carbon black particles (35 mg/kg bw) did not induce CYP1A1 protein or enzyme activity. However, both particle types (at 5 and 35 mg/kg bw) caused a significant decrease in CYP2B1 protein and enzyme activity at day 1, which was persistent up to day 7 with 35 mg/kg bw. Similarly, both treatments significantly attenuated glutathione S-transferase (GST)-Pi protein on day 1 after exposure and decreased the activities of GST and catalase on days 1 and 7. The diesel exhaust particles, but not carbon black, significantly induced quinone reductase activity on day 7. The authors suggested that diesel exhaust particles may induce CYP1A1 and quinone reductase enzymes by a chemical effect, while the carbonaceous core may be involved in the attenuation of CYP2B1, GST and catalase protein levels and enzyme activities (Rengasamy et al., 2003).

Zhao et al. (2004) evaluated the change in lung metabolic enzymes in response to concentrations of 35 mg/kg bw saline, diesel exhaust particles and carbon black intratracheally instilled into rats that were then killed 1, 3 or 7 days later. Metabolically activated fractions (S9) were extracted from control and exposed rat lungs. The mutagenic activity of 2-aminoanthracene, 2-aminofluorene, 1-nitropyrene and an organic extract of diesel exhaust particles was then determined in Salmonella typhimurium YG1024. The S9 from the control and exposed rats showed a dose-dependent increase in mutagenic activity of all four compounds. Compared with the saline control, the S9 from the carbon black-exposed rats was a less potent inducer of mutagenicity of 2-aminoanthracene. When inhibitors of CYP1A1 (α-naphthoflavone, 1 µM/plate) or CYP2B1 (metyrapone, 10 µM/plate) were added to the reaction mixture to monitor the involvement of CYP1A1 in S9 metabolic activity in the lung, α-naphthoflavone inhibited the metabolic activation of 2-aminoanthracene induced by S9 from carbon black-exposed rats to a lesser extent than the metabolic activity induced by S9 from rats exposed to saline and diesel exhaust particles. The rats exposed to both particle types revealed a significant change in phase I and II enzymes in the lungs, including CYP1A1, CYP2B1, GST and nicotinamide adenine dinucleotide phosphate quinone-oxidoreductase (Rengasamy et al., 2003). The authors suggested that, after exposure to carbon black, the reduction in the constitutive enzyme CYP2B1 in the lung may play a role in the pulmonary handling of mutagenic agents (Zhao et al., 2004).

(c) Other routes

The effect of ultrafine particles on the microcirculation in extrapulmonary organs was investigated in C57BL/6 mice that received intra-arterial infusions of 1 × 10⁷ or 5 × 10⁷ ultrafine carbon black particles (Printex 90; diameter, 14 nm; surface area, 300 m²/g) suspended in 200 µL buffer containing 15% human albumin. Two hours after infusion, platelet–and leukocyte–endothelial cell interactions, sinusoidal perfusion, endothelial fibrin deposition and the phagocytic activity of Kupffer cells were analysed by intravital video fluorescence microscopy in the liver microvasculature. The particles induced accumulation of platelets in the hepatic microvessels, which was associated with pro-thrombotic changes on their endothelial surface. Accumulation of particles in the liver had a strong procoagulatory effect, but did not trigger an inflammatory reaction or induce microvascular or hepatocellular tissue injury (Khandoga et al., 2004).

The possible adjuvant effect of diesel exhaust particles (NIST; diameter, 30 nm; 64 m²/g) and carbon black (Regal 250R; diameter, 35 nm; 60 m²/g), which was used as a surrogate for a non-extractable core of diesel exhaust particles with a similar size and surface area, on the response to the allergen ovalbumin was studied in BALB/c mice. A footpad inoculation was followed by a popliteal lymph node assay and other immunotoxic evaluations, including the weight change of popliteal lymph nodes, cell numbers and proliferation and specific serum IgE anti-ovalbumin antibody levels. Carbon black, although less potent than diesel exhaust particles, exhibited a similar capacity to increase the local lymph node response and specific serum IgE response to ovalbumin. Both particles had a significant adjuvant effect on the local immune-mediated inflammatory response and systemic specific IgE response to allergen, which suggested that the non-extractable particle core contributed substantially to the adjuvant activity of diesel exhaust particles (Løvik et al., 1997).

(d) Ex-vivo and in-vitro studies

In a study on the cytotoxicity of diesel exhaust particles, their phagocytosis and the resulting immune response, carbon black particles (FR103; diameter, 95 nm) that were included as a surrogate of the carbonaceous core of the diesel exhaust particles were reported to contain 1.5% of the PAH content of the diesel exhaust particles. Human bronchial epithelial cells (16HBE14o-) and human nasal epithelial cells in primary culture were exposed to the two particle types. Treatment with carbon black particles (10 µg/cm² for 48 hours; concentrations were expressed per square centimeter since the particles sediment rapidly onto the culture) stimulated the release of granulocyte macrophage colony-stimulating factor (GM–CSF) and IL-8, but to a lesser extent than diesel exhaust particles (Boland et al., 1999).

The expression of human leukocyte antigen-DR (HLA-DR) on the cell membrane of antigen-presenting cells is of major importance for the induction of an allergic response in the airways. Because environmental particulates may induce or enhance allergic sensitization, a study was conducted to investigate the potential of carbon black (Vulcan M; CMD, 90 nm), diesel exhaust particles and urban air particulates (0.1–1000 ng/cm²) to induce the expression of HLA-DR in cultures of differentiated THP-1 human monocytes, which are used as a model for alveolar macrophages. The ‘adjuvant’ potential of the particles on IFN-γ, a known enhancer of HLA-DR, was also studied. The particles alone did not induce HLA-DR on the THP-1 cells after 48 hours of incubation. However, even at very low concentrations, carbon black (1 ng/cm² and above) and diesel exhaust particles (0.1 ng/cm² and above) interacted with IFN-γ (100 U/mL) to enhance HLA-DR expression up to 2.5-fold. This in-vitro finding suggests the existence of a mechanism by which particles exert an adjuvant activity and which may partially explain how exposure to particles can enhance allergic sensitization (Don Porto Carero et al., 2002).

The effects of 10, 20 or 30 µg/cm² ambient particulate matter, diesel exhaust particles and carbon black particles (FR103; diameter, 95 nm) on cultured human bronchial epithelial (16HBE14o-) cells were compared. No significant effects on cell viability were observed after incubation with either particle type. In contrast to ambient particulate matter and diesel exhaust particles, carbon black particles did not disturb cell growth or induce the production of peroxides or the release of GM–CSF. Carbon black particles were more actively phagocytosed than the two other particle types (Baulig et al., 2003a). In a subsequent study, the same carbon black particles (10, 20 or 30 µg/cm², equivalent to 50, 100 or 150 µg/mL) did not cause an increase in reactive oxygen species or induce the expression of CYP1A1 mRNA in 16HBE14o- cells, whereas diesel exhaust particles did (Baulig et al., 2003b).

In another study that compared the effects of various forms of diesel exhaust and carbon black particles (10 µg/cm²) on 16HBE cells, the latter weakly induced the release of GM–CSF and activated nuclear factor-κB (Bonvallot et al., 2001).

To investigate whether reduced clearance from the lung after exposure to ultrafine particles may be due to impaired phagocytosis by alveolar macrophages, an in-vitro study was conducted with the macrophage cell line J774.2 MΦ. The cells were exposed for 8 hours to fine titanium dioxide (mean diameter, 250 nm; 6.6 m²/g), ultrafine titanium dioxide (mean diameter, 29 nm; 49.8 m²/g), carbon black (Huber 990; mean diameter, 260.3 nm; 7.9 m²/g) or ultrafine carbon black (Printex 90; mean diameter, 14.3 nm; 253.9 m²/g). The particles had no cytotoxic effects. The ability of the macrophages to phagocytose 2-µm latex beads was significantly reduced (P < 0.001) after exposure to 0.39 µg/mm² ultrafine carbon black and 0.78 µg/mm² of all particle types compared with the control. Furthermore, ultrafine carbon black induced a significant (P < 0.001) reduction in macrophage phagocytosis at a lower dose than fine carbon black (0.39 and 0.78 µg/mm², respectively). At all doses, exposure to ultrafine carbon black resulted in a larger number (P < 0.001) of non-phagocytic macrophages compared with the other particle types. The culture medium collected after exposure of macrophages to particles had no significant effect on the phagocytic ability of naive macrophages, which suggests that cell-to-cell contact rather than a soluble factor was responsible for the defective phagocytosis. The authors concluded that slowed clearance of particles, especially ultrafine particles, can in part be attributed to a particle-mediated impairment of macrophage phagocytosis (Renwick et al., 2001; see also the in-vivo study by Renwick et al., 2004, discussed above).

Because ultrafine particles and transition metals have been postulated to be important determinants of the toxicity and potential adverse health effects of particulate air pollution, the interactions between transition metal salts and fine and ultrafine carbon black particles were studied. In all experimental systems used, the ultrafine particles were more reactive than the larger-sized particles. Incubation of ultrafine carbon black (Printex 90; diameter, 14 nm; 253.9 m²/g) with the reactive oxygen species-sensitive probe dichlorofluorescin diacetate in a cell-free system generated significantly more reactive oxygen species than the larger-sized carbon black particles (Huber 990; 260 nm; 7.9 m²/g). Addition of cupric sulfate, ferrous sulfate or ferric chloride further increased the generation of reactive oxygen species induced by ultrafine carbon black. In Mono Mac 6 macrophages (a human monocytic cell line), the 14-nm carbon black again produced more reactive oxygen species than the 260-nm particles, but iron salts had no additive effect. Ultrafine carbon black decreased the cellular content of glutathione (GSH) and adenosine triphosphate (ATP) in the murine macrophage cell line J774. Further reductions in GSH and ATP were seen after the addition of iron salts but only at the highest concentration tested (500 µM). A concentration-dependent increase in the production of TNFα was also observed in J774 cells after exposure to ultrafine carbon black, but this effect was not further enhanced by the addition of iron salts even at the highest concentration tested (500 µm). In the rat lung, ultrafine carbon black (125 µg) induced a significant influx of neutrophils in the BAL fluid. This inflammatory effect was enhanced by the addition of ferric chloride (100 µM), which was inactive alone. The authors concluded that ultrafine particles and metal salts interact by chemical potentiation in a cell-free system to generate reactive oxygen species. This potentiation was not observed in the presence of macrophages, probably because the iron is sequestered or chelated by the cells. In the lung, ultrafine particles and iron salts interacted synergistically in generating inflammation (Wilson et al., 2002).

The capacity of fine carbon black (Huber 990; diameter, 260 nm; 7.9 m²/g) to activate serum factors that stimulate the migration of murine alveolar macrophages was compared with that of ultrafine carbon black (Printex 90; diameter, 14 nm; 254 m²/g). Incubation of fetal bovine serum with 5 and 10 mg/mL ultrafine carbon black caused a 1.4- and 1.8-fold increase, respectively, in migration of macrophages compared with untreated serum. These effects were partially inhibited by further incubation with antioxidants (Trolox or Nacystelin). An equivalent mass of fine carbon black (10 mg/mL) did not show chemotactic activity. On an equal mass basis, ultrafine carbon black particles activated serum factors, possibly C5a-like proteins, to a greater extent than fine carbon black particles (Barlow et al., 2005).

The mouse monocyte/macrophage cell line RAW264.7 was used to determine the adverse effects of exposure in vitro to 30 and 120 µg/mL size-fractionated urban air particles (particulate matter (PM) 2.5–10; PM_2.5) collected in the city of Rome and carbon black (Huber NG90; diameter, 200–250 nm). Urban air particles induced a significant release of arachidonic acid after a 5-hour exposure at both concentrations, while carbon black was effective only at 120 µg/mL. After 5 hours, the 120-µg/mL concentration of the two PM fractions stimulated the production of TNFα about 10-fold more strongly than carbon black particles, but the stimulation diminished after 24 hours. In contrast, carbon black-stimulated TNFα production did not show such a decrease. Production of IL-6 was enhanced by incubation with urban air particles but not with carbon black. Carbon black was consistently less effective than the urban particles (Pozzi et al., 2003).

Chin et al. (1998) evaluated the role of adsorbed mutagens, such as benzo[a]pyrene, on carbon black particles in cellular response and signal transduction. A cultured macrophage cell line (RAW264.7) was exposed to carbon black (N339; diameter, 0.1 µm) and benzo[a]pyrene-adsorbed carbon black (2 µg/mL benzo[a]pyrene) for up to 24 hours. The benzo[a]pyrene-adsorbed carbon induced time-dependent expression and release of TNFα and apoptosis in RAW cells, which were inhibited by a TNFα-neutralizing antibody. Neither carbon black nor benzo[a]pyrene alone induced these effects. TNFα activates mitrogen-activated protein kinase (MAPK) activity and the extracellular signal-regulated kinases p42/44 in a time-dependent manner, and treatment of RAW264.7 cells with the MAPK inhibitor PD-098059 inhibited the apoptosis and TNFα secretion induced by benzo[a]pyrene-adsorbed carbon black. The results indicated that adsorbed mutagens on carbonaceous particles may play a role in the induction of apoptosis and inflammatory responses, such as the release of cytokines like TNFα.

Ultrafine particles including carbon black (Printex 90, diameter, 12 nm; 300 m²/g), elemental carbon (diameter, 90 nm; 600 m²/g) and diesel exhaust particles (diameter, 120 nm; 108 m²/g; 10–320 µg/mL/10⁶ cells) caused a variety of cytoskeletal dysfunctions including impaired phagocytosis (approximately 50% of controls), inhibited cell proliferation and decreased cell viability in primary alveolar macrophages from dogs and a mouse alveolar macrophage cell line (J774A.1) within 24 hours of treatment with a dose of 320 µg/mL/10⁶ cells (Möller et al., 2002).

(a) Mechanistic considerations (overall mode of action)

A cascade of events that was proposed to describe the biological process that starts with some particle deposition at critical target cells or tissues within the rat lung and results in rat lung tumours includes: sustained inflammation, production of reactive oxygen species, depletion of antioxidants and/or impairment of other defence mechanisms, cell proliferation and gene mutations. These individual steps comprise an overall mode of action that can be used to compare responses of rats with those of other species including humans (see Fig. 4.2). The dose metric that best describes the dose–response relationship for poorly soluble particles in the rat lung has been examined in various studies (Driscoll et al., 1996; Pott & Roller, 2005; Morfeld et al., 2006). The surface area, volume and size of particles have all been shown to be related to the tumour response in rats.

At particle lung burdens that are associated with impaired clearance in rats (e.g. beginning at a mass dose of ∼0.5 mg/lung and completely overloaded at ∼10 mg/lung for fine particles of unit density; Muhle et al., 1990a), a sustained and widespread cellular inflammatory response occurs. The cell population is dominated by activated and probably (under these conditions) persistent polymorphonuclear neutrophils and secretes a collection of mediators (reactive oxygen species, pro-/anti-inflammatory cytokines, proteases, cytotoxins, fibrogenic and other growth factors) that act through the pulmonary milieu on surrounding cells or tissues and surrounding structures (Castranova, 1998, 2000; Knaapen et al., 2004). The degree of sustained inflammation experienced by rodents (most notably rats) at high lung burdens has not been observed in humans. However, humans may experience sustained inflammation in certain disease states. One such human condition (which may be particle-stimulated—e.g. by silica—or may be cytogenic) is late-stage, interstitial pulmonary fibrosis (Daniels & Jett, 2005). Patients with interstitial pulmonary fibrosis and chronic inflammation have been reported to develop a higher incidence of lung tumours, frequently in the most inflamed areas. In addition, chronic inflammation has been associated with non-neoplastic lung diseases in workers with dusty jobs. Rom (1991) found a statistically significant increase in the percentage of polymorphonuclear neutrophils in the BAL fluid of workers with respiratory impairment who had been exposed to asbestos, coal, or silica (4.5% in cases versus 1.5% in controls). Elevated levels of polymorphonuclear neutrophils have been observed in the BAL fluid of miners with simple coal workers' pneumoconiosis (31% of total BAL cells versus 3% in controls) (Vallyathan et al., 2000) and in patients with acute silicosis (a 10-fold increase over controls) (Goodman et al., 1992; Lapp & Castranova 1993). An elevated incidence of lung cancer has been observed in some workers exposed to poorly soluble particles including crystalline silica (Rice et al., 2001; Attfield & Costello, 2004) and diesel exhaust particles (Stayner et al., 1998), although these materials may have greater inherent toxicity than other poorly soluble particles such as titanium dioxide, carbon black and talc.

The precise role of chronic inflammation in the development of cancer is uncertain, but there is considerable evidence that chronic inflammation may have a multifaceted role in this process. Activated cells in the lung are known to release various reactive intermediates, most notably those derived from oxygen. Sustained excess of oxidant activity is known to deplete antioxidant defences gradually. These defence mechanisms in the lungs of humans and rats clearly differ, in that humans are overall relatively deficient in some of them (Slade et al., 1985). Reactive oxygen species within cells may damage DNA directly and potentially induce mutations. Moreover, cell damage and promitotic stimuli initiated by reactive oxygen species promote cell turnover and proliferation, events that may enhance the risk for DNA replication error and/or expand a mutated or transformed cell to initiate a tumour.

The mechanism that involves inflammation and oxidative stress which lead to tumour formation is considered to be a secondary genotoxic mechanism, in contrast to a primary genotoxic mechanism in which the agent interacts directly with DNA (Knaapen et al., 2004).

(b) Interspecies extrapolation

There remains uncertainty with regard to identifying in detail the cascade of events that lead to rat lung cancer following inhalation of poorly soluble particles that include talc, carbon black and titanium dioxide. However, as shown in Fig. 4.2, several important steps can be identified that are supported by a substantial rodent database. An important question that needs to be addressed is the extent to which the steps outlined in Fig. 4.2 for rat lung cancer are also operative in other animal species including humans. The majority of animal studies that have evaluated the effects of poorly soluble particles on the respiratory tract have been conducted in rats; species differences such as particle inhalability, breathing conditions, respiratory tract structure and pulmonary defences need to be considered when toxicological findings from rodents are extrapolated to humans (Brown et al., 2005). Exposure to airborne particles would generally need to be higher in rats to result in equivalent particle doses in the human lung (Brown et al., 2005).

All animals species used routinely in particle toxicology and humans are susceptible to impairment of clearance of poorly soluble particles from the lungs. Impaired clearance is probably one of the first steps necessary to initiate a sequence of events that may lead to lung cancer in rats exposed to poorly soluble particles (see Fig. 4.2). However, different animal species exhibit differences in particle-induced impairment of clearance, which can result in different lung burdens (expressed as mass or surface area) following exposures to the same particle concentrations (Elder et al., 2005).

Similarly, pulmonary inflammation has been reported to be a consequence of exposures to poorly soluble particles in both experimental animals and humans (Driscoll et al., 1996; Elder et al., 2005; Shwe et al., 2005; Rom, 1991; Vallyathan et al., 2000). The pathophysiology of particle-induced fibrosis in humans and fibrosis and lung cancer in rats from lung overload—with lung burden being expressed as mass, volume and/or surface area for fine and ultrafine particles—involves chronic inflammation, hyperplasia and cell proliferation, altered collagen deposition and architecture.

Rats and mice, in contrast to hamsters, exhibit sustained inflammation associated with particle lung burden, but lung tumours induced by poorly soluble particles have only been observed in rats. It has been shown that rats are uniquely susceptible to poorly soluble particle-induced lung cancer relative to mice and hamsters. While some of the steps indicated in Fig. 4.2 have been demonstrated in humans exposed to poorly soluble particles, it is not known to what extent humans are susceptible to particle-induced lung cancers associated with titanium dioxide, carbon black or talc.