Formation of PAH–DNA adducts after in vivo and vitro exposure of rats and lung cells to different commercial carbon blacks

Toxicology and Applied Pharmacology 205 (2005) 157 – 167 www.elsevier.com/locate/ytaap

Formation of PAH–DNA adducts after in vivo and vitro exposure of rats and lung cells to different commercial carbon blacks Paul J.A. Borma,T, Gonca Cakmakb, Erich Jermannc, Christel Weishaupta, Pascal Kempersd, Frederik Jan van Schootend, Gu¨nter Oberdo¨rstere, Roel P.F. Schinsa a

Particle Research, Institut fu¨r Umweltmedizinische Forschung (IUF) at the Heinrich, Heine University, Dusseldorf, Germany b Department of Toxicology, Gazi University, Turkey c Department of Analytical Chemistry, Medizinisches Institut fu¨r Umwelthygiene at the Heinrich, Heine University Dusseldorf, Germany d Department of Health Risk Analysis and Toxicology, University of Maastricht, The Netherlands e Department of Environmental Medicine, University of Rochester, Rochester, NY 14627, USA Received 25 June 2004; accepted 13 October 2004 Available online 19 March 2005

Abstract Objective: The current study was designed to test the possible release and bioavailability of polycyclic aromatic hydrocarbons (PAHs) from a set of commercial carbon blacks (CBs) as well as the ability of these PAHs to form bulky DNA adducts. Methods: In four commercial CBs (Printex 90, Sterling V, N330, Lampblack 101), leaching of PAH was examined through (1) release of parent PAHs in saline with or without surfactant, and (2) PAH adducts in lung epithelial cells (A549) or in rat lungs after exposure to two CBs (Printex 90, Sterling V) for 13 weeks (50 mg/m3). In vitro experiments were done with original and extracted particles, as well as organic extracts of CB in DMSO. As positive controls, B[a]P (0.03 AM) and a mixture of 16 PAHs (0.1 AM) were used. Results: No leaching of PAHs was measured in saline or surfactant-containing saline. In vitro incubations with CB particles (30 – 300 Ag/cm2) revealed no adduct spots except for Sterling V. However, the spot was not concentration dependent and remains unidentified. Lung DNA from rats after inhalation of Printex 90 or Sterling V showed no spots related to PAH – DNA adduct formation compared to sham-exposed rats. Conclusion: The results suggest that PAHs are very tightly bound to these CBs. Only using organic extracts or particles of low-surface Sterling V, with high PAH content, PAHs may become available to form PAH – DNA adducts. However, the in vitro conditions showing this effect will not be encountered in vivo and renders this mechanism in particle-induced lung cancer at in vivo exposures highly unlikely. D 2004 Elsevier Inc. All rights reserved. Keywords: Carbon black; Bioavailability; PAHs; DNA adducts; Lung

Introduction The formation of tumors by poorly soluble particles (PSPs) in rat lungs is generally considered as the chronic endpoint of persistent inflammation and cell proliferation during exposures leading to overload (Borm et al., 2004; Greim et al., 2001; Schins, 2002). Among the responsible mediators, reactive oxygen and nitrogen species (ROS and

T Corresponding author. Zuyd University, Centre of Expertise in Life Sciences, Heerlen, The Netherlands. Fax: +31 45 4006800. E-mail address: [email protected] (P.J.A. Borm). 0041-008X/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2004.10.020

RNS) generated by inflammatory cells are considered to play a crucial role in genotoxicity, mutagenicity, and cell proliferation pathways (Knaapen et al., 2004). However, apart from this inflammation-associated DNA damage, an additional and potentially important mechanism of genotoxicity is contained in the fact that many particles, such as diesel exhaust particles (DEP) and carbon black (CB), can carry surface-adsorbed components into the peripheral lung. One important group of these compounds are polycyclic aromatic hydrocarbons (PAHs), among which some are established carcinogens, such as benzo[a]pyrene (B[a]P). Using organic extractions of ambient particulate matter (PM), it has been indicated that these compounds may play an important role in

158

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

the possible mutagenicity and carcinogenicity of automobile exhaust (Alink et al., 1983; Binkova et al., 2003; Buschini et al., 2001; Dukovich et al., 1981). The genotoxic and carcinogenic activity of PAH has usually been attributed to the formation of DNA adducts by reactive PAH metabolites (Hall and Grover, 1990). Although most DNA-reactive PAH metabolites are suggested to arise from biotransformation by cytochrome P450 enzymes, inflammatory cell-derived ROS (Borm et al., 1997) and quinones (Cavalieri et al., 2004) have been shown to play a role in the formation of DNA adducts by PAHs. In both cases, ROS either from inflammatory cells or generated by redox cycling are able to form adducts to PAHs leading to reactive metabolic intermediates that can react with DNA (Knaapen et al., 2004). However, it is still debated whether formation of DNA adducts from PAH metabolites plays a role at all in particle-induced lung tumor induction and whether adsorbed PAHs are bioavailable in the lung in vivo. Early studies showed that absorption of PAH onto organic particles clearly affects their retention in the lung and the elution of absorbed PAH also depends on particle size and surface area (Sun et al., 1984). Recent work by Gerde et al. (2001) in dogs showed for B[a]P-coated DEP particles that 36% of the B[a]P is rapidly released from DEP in the lung, but that its absorption rate is so high that all B[a]P enters the circulation without further metabolic transformation within the lung. The remaining fraction of the B[a]P (64%) was found not to be bioavailable and remained particle bound in the lung up to 5.6 months (Gerde et al., 2001). The role of PAH on carbon particles was further questioned by the findings of Dasenbrock et al. (1996), who found no increased tumor rate of DEP after coating particles with B[a]P (11 Ag/mg), compared to 0.9 Ag B[a]P/mg in control particles. In line with these observations, the tumorigenicity of native vs. organic-extracted DEP in rats appeared to be rather similar and with undetectable PAH – DNA adduct levels (Dasenbrock et al., 1996; Gallagher et al., 1994). The current study was designed to test the bioavailability and acellular release of PAHs from a set of commercial carbon blacks that differ widely in their PAH content, and test the ability of these PAH-containing particles to form bulky DNA adducts in target lung epithelial cells in vitro and in vivo. The outcomes will contribute to current issues of hazard classification and risk assessment of CBs (Borm et al., 2004; Rausch et al., 2004), since they provide quantitative data for whether these specific PSPs and PAHs are bioavailable and may play a significant role in the formation of DNA adducts.

Methods Chemicals and test materials Dulbecco’s modified Eagle’s medium (DMEM), fetal calf serum (FCS), l-Glutamine, penicillin – streptomycin, Trypsin-EDTA, sodium pyruvate, h-NADH, 2,4-dinitro-

phenylhydrazine, and 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) and dipalmytoyl phosphatidylcholine (DPPC) were obtained from Sigma-Aldrich (Taufkirchen, Germany). Hank’s balanced salt solution (HBSS) was obtained from Life Technologies (Kahrlsruhe, Germany). 32P-ATP and Nuclease P1 were supplied by ICN (Indianapolis, USA). All other chemicals for postlabeling were from similar suppliers as in Van Schooten et al. (1997). Test materials. Carbon black pigments (CBs) were provided by Degussa (Printex 90, Flamruss 101) and Cabot (Sterling V) and Columbian Chemical Company (N330) in sealed 5-kg aliquots, along with data on surface area and particle size distribution of these materials (Table 1). Particle extracts were Table 1 Characteristics of study materials and concentrations used for in vitro studies with lung epithelial cells Material

PAH levelsa (mg/kg)

Type of treatment Concentrations in vitro

Printex 90, 300 m2/g

1: 0.039 2: 0.001

Original particle Extract (DMSO)

0.010 0.065 ND 0.008 8.8 0.4

Extracted particle

Sterling V, 30 – 40 m2/g

3: 4: 5: 6: 1: 2:

N330, 70 – 90 m2/g

3: 4: 5: 6: 1: 2:

21.5 202.5 6.8 89.7 2.4 0.2

Lampblack 101, 20 m2/g

3: 4: 5: 6: 1: 2:

7.2 191 1.4 44.8 0.057 0.002

Diesel (NIST)

TiO2 P25 (AF BaP), 50 m2/g TiO2 P25 (AF), 50 m2/g

3: 0.05 4: 0.813 5: 0.011 6: 0.172 See info RM 2975 5: 16.7

Original particle Extract (DMSO) Extracted particle

Original particle Extract (DMSO) Extracted particle

Original particle Extract (DMSO) Extracted particle

Original particle Extract (DMSO) Extracted particle Original particle Extract (DMSO) Extracted particle Original particle

100 Ag/cm2 0 AM BaP (1 mg/cm2)b 100 Ag/cm2

100 Ag/cm2 0.84 AM BaP (3.7 mg/cm2)b 100 Ag/cm2

100 Ag/cm2 0.16 AM BaP (3.2 mg/cm2)b 100 Ag/cm2

100 Ag/cm2 0.0006 AM BaP (1.6 mg/cm2) 100 Ag/cm2

100 Ag/cm2 AM BaP (ND)c 100 Ag/cm2 100 Ag/cm2 100 Ag/cm2 100 Ag/cm2

a Individual PAHs are denominated with (1) phenanthrene, (2) anthracene, (3) fluoranthene, (4) pyrene, (5) benzo(a)pyrene, and (6) benzo (ghi) perylene, but denomination is not extensive. Determinations done by HPLC – FLD after Soxhlet extraction as described in Methods. b Between brackets is shown the virtual dose of particles that would have been needed to achieve this level of BaP as obtained using extracts. c ND, not determined.

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

prepared by 8-h Soxhlet extraction of 7 –28 g of different carbon blacks with 220 ml toluene after precleaning the glass equipment for 3 h with 250 ml toluene (Haguenoer et al., 1996). Extracted particles were recovered by filtration and dried after two washings in acetone for 4 h at 110 -C. The toluene extracts were transferred into DMSO (5 ml) by selective distillation in order to avoid loss of volatile PAH and used for in vitro experiments along with the dried extracted particles (Table 1). The PAH content of the DMSO extracts was measured with high-pressure liquid chromatography (HPLC) and fluorescence detection. Standard diesel particles (SRM 2975) containing 0.0522 mg/kg B[a]P were obtained from the National Institute of Standards and Technology (Gaithersburg, MD). Ultrafine TiO2 (P25) was obtained from Degussa (Hanau, Germany) and coated with a known amount of B[a]P and used as positive particulate controls. As positive controls, either a mix of 16 standard PAHs (EPA – PAH mixture) and a B[a]P standard (gifts Dr. Ehrensdorfer, Augsburg, Germany) was used and prepared in DMSO to reach target concentrations of 0.1 and 0.03 AM, respectively, and frozen in aliquots at 80 -C. Extraction and determination of PAHs For the determination of solvent-extractable PAH content from carbon blacks, 75 mg of test materials was extracted for 8 h with 30 ml of toluene in a Soxhlet extractor. In order to avoid losses of PAHs, 1 ml DMSO was added to the toluene extract and the solvent was evaporated by means of a Rotation evaporator following a treatment under a flow of nitrogen down to 1 ml. Under the given conditions, the extraction of the PAHs was in a quantitative manner. The samples were stored for a maximum 3 days at room temperature, or otherwise at 18 -C. The determination of PAHs in extracts was performed by high-pressure liquid chromatography (HPLC) using fluorescence detection. Separation was done on a reversed-phase column (Grom-Sil PAH-1, tailor-made, 250  4.0 mm, by Grom, Herrenberg, Germany) using an acetonitrile/water gradient (starting with a acetonitrile/water ratio of 50:50 for 5 min, then increasing the acetonitrile content continuously over 15 min to 100%). Calibration was done by using standard solutions in acetonitrile/water (25/75/v/v) containing the PAHs at a concentration between 0 and 50 Ag/l. A certified reference material (PAH contaminated soil/sediment, by Promochem, Wesel, Germany) was used for quality control purposes. The detection limits for the PAHs ranged between 0.02 and 0.2 ng/ mg carbon black corresponding to 0.04 and 0.4 Ag/l of B[a]P. Acellular leaching of PAHs from carbon blacks To study the release of PAHs from the particle core, 3 –60 mg of the different carbon blacks was suspended in 3 ml of solutions containing from 100 to 10 000 Ag/ml dipalmitoyl phosphatidylcholine (DPPC) and incubated for 24 h at 37 -C in the dark in a shaking water bath. The carbon blacks were

159

removed by a procedure slightly modified from Bevan and Worell (1985), including gradient centrifugation (22 min, 4000 rpm, 4 -C) after slow addition of 2.5 ml of a solution containing 50% saccharose. The supernatant containing the leachable fraction was then extracted for 60 s using 2 ml tertiary butylmethyl ether and centrifuged (10 min, 4000 rpm, 4 -C) to achieve a complete phase separation. One milliliter of the ether extract was evaporated down to dryness. The residue was then dissolved in 1 ml of a solution containing 250 Al acetonitrile and 750 Al ultrapure water. The determination of PAHs was performed by high-pressure liquid chromatography (HPLC) using fluorescence detection as described above. Calibration was done by using standard solutions in acetonitrile/water (25/75/v/v) containing the PAHs at a concentration between 0 and 50 Ag/l. A certified reference material (PAH contaminated soil/sediment, by Promochem) was used for quality control purposes. To avoid the contamination of the column by remains of the DPPC and improve chromatographic resolution, leachates were treated with phospholipase enzymatic digestion according to Alvarez et al. (1995). After addition of acetonitrile and centrifugation, the digested leachates were analyzed by HPLC with similar detection limits as noted above. Cell cultures A549 human lung epithelial cells (American Type Culture Collection) were grown in DMEM culture medium supplemented with 10% heat-inactivated fetal calf serum (SigmaAldrich), l-glutamine, and 30 IU/ml penicillin – streptomycin at 37 -C and 5% CO2. Experiments were performed in complete medium unless stated otherwise as described previously (Schins et al., 2002). Carbon black suspensions were made in HBSS, sonicated for 5 min in a water bath, and diluted into the culture dishes at the indicated final concentrations using two replicate wells per treatment or in 75-cm2 culture plates for extraction of DNA. Cytotoxicity was determined by two methods: lactate dehydrogenase (LDH) activity in the supernatant was measured at 540 nm using pyruvic acid as substrate on an automatic plate reader (Multiskan Labsystems, Helsinki, Finland), as described previously (Schins et al., 2002). LDH release into the medium was determined and compared to untreated cells. As a second method for cytotoxicity, the MTT assay was used, which is based on the activity of mitochondrial dehydrogenase in viable cells to convert 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) in a blue formazan dye. Details of the method are described elsewhere (Schins et al., 2002). Cell studies to investigate PAH –DNA adducts with A549 cells were done at equivalent mass (100 Ag/cm2) with all test particles and using equivalent volume (50 Al) of particle extracts reaching different PAH levels (Table 1). The particle concentration of 100 Ag/cm2 was used as the maximal value as determined by the absence of in vitro cytotoxicity of all carbon particles as determined by the MTT assay and the LDH assay (data not shown).

160

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

In vivo exposures

Results

All animal exposures took place at the University of Rochester, Department of Environmental Medicine. F-344 rats were purchased from Harlan (Indianapolis, IN), fed Purina rodent chow and water on an ad libitum basis. Three particle exposure levels (1, 7, and 50 mg/m3) were used for Printex-90 and one concentration (50 mg/m3) for Sterling V, as well a sham exposure group. All animals were exposed for 13 weeks and DNA was extracted from whole lung DNA from rats immediately after exposure. The lungs of rats for DNA analysis were not lavaged but the vascular system was perfused. DNA was extracted by Jane Gallagher (US-EPA) who used this DNA for determination of oxidative DNA damage (Gallagher et al., 2003). To study whether PAHs are available and subsequently transformed into DNA-binding metabolites, lungs of three animals from every exposure group immediately after exposure were analyzed for DNA adducts. For these samples, 32P postlabeling was performed using either nuclease P1 enrichment or butanol extraction to enhance DNA adduct detection. Both procedures are suitable to detect PAH-related DNA adducts with its own specificities.

Leaching of PAH from carbon blacks in DPPC-containing suspensions

PAH – DNA adduct formation Aromatic adducts to DNA were determined by 32P postlabeling according to the method previously described (Van Schooten et al., 1997). For isolation of DNA, A549 cells or tissues were lysed overnight with SET/SDS. DNA was extracted repeatedly with phenol, phenol/chloroform/ isoamyl alcohol (25:24:1) and chloroform/isoamyl alcohol (24:1), precipitated dissolved in 5 mM Tris/1 mM EDTA, pH 7.4. Quality and quantity of the DNA was determined spectrophotometrically and afterwards its concentration adjusted to 1 mg/ml. For postlabeling, DNA (5 Ag) was digested into deoxyribonucleoside 3V-monophosphates, treated with nuclease P1, and labeling was performed using AT-q-32P [50 ACi]. Subsequently, thin layer chromatography was done on polyethyleneimine (PEI)-cellulose sheets (Macherey Nagel, Du¨ren, Germany). For standardization, control samples of [3H]BPDE-modified DNA with known modification levels (1 adduct per 107 and 108 unmodified nucleotides) were run in all experiments. Quantification was done using a phosphoimager (Molecular Dynamicsi, Sunnyvale, CA) with a lower detection limit of 1 adduct per 109 nucleotides. In the BPDE – DNA adduct standards, a modification level of 1 adduct per 108 could easily be detected. Statistical analysis Data are expressed as mean T standard errors unless stated otherwise. Statistical analysis was performed using Student’s t test using SPSS v.9 for Windows. A difference was considered significant at P < 0.05.

Our investigations showed no measurable release of PAHs from standard diesel or carbon blacks (1 mg/ml) in aqueous DPPC solutions containing 100 –1000 Ag/ml DPPC. An increase in DPPC concentration up to 10 mg/ml also showed no measurable PAH release. The same was true when increasing the amount of diesel up to 20 mg/ml before the extraction experiments. Based on the detection limit for HPLC determination of PAH, the amount of a single PAH that possibly leaches from diesel particles under these circumstances is less than 1.2% for phenanthrene, <0.4% for pyrene, <1.0% for anthracene, <1.3% for chrysene, and <1.3% for fluoranthene. To avoid contamination of the column by remains of the DPPC and to improve chromatographic resolution, leachates were treated with phospholipase enzymatic digestion. Again, the PAH concentrations in the digestion solutions were without exception below the detection limit. These experiments thus confirmed the results that were obtained without digestion of the DPPC. Since no PAHs were found in the supernatant after addition of the EPA – PAH mixture to a diesel suspension of 20 mg/ml, we suspected that PAH leaching might occur but counteracted by a strong absorption to the particle surface. To test this hypothesis, we spiked several PAHs, including B[a]P, benzofluoranthene, and pyrene, into suspensions of diesel particles in incubation medium. At diesel particle concentrations between 1 and 20 mg/ml and PAH target concentration of 1 AM, no remaining PAH could be detected in supernatants when diesel concentrations exceeded 1 mg/ml. Similar investigations showed that the lowest surface CB (Sterling V) showed similar absorption of BaP than diesel (Fig. 1). The data suggest that particle surface is able to bind free PAHs in suspension, leading to low concentrations of free PAHs in the presence of particles. Similar tests with other CBs were not done, since their specific surface area is higher than diesel and Sterling V and therefore even larger adsorption effects are expected with other CBs. Studies in lung epithelial cells Incubations with A549 cells and CB particles were done using 0.03 AM B[a]P as a target concentration, assuming that all B[a]P from the CBs would be soluble. Experiments with standard B[a]P solutions and extracts of CB showed that at the 0.03 AM target concentration readily detectable levels of BPDE –DNA adducts were present (Fig. 2). The actual concentrations of B[a]P in the final incubations are given in Table 1, as well as the equivalent particle concentration that would have been necessary to obtain this concentration of B[a]P, assuming that all B[a]P is soluble. The four original CBs as well as their extracts and the extracted particles were incubated along with positive

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

161

Fig. 1. Absorption of PAHs on particle surface illustrated by the absorption of spiked B[a]P to diesel particles and sterling particles in water (incubation medium) suspension. Similar absorption was seen for pyrene and benzo[k]fluoranthene.

(Diesel-SRM-2975) and negative control (TiO2-uF) and EPA standard PAH mix (12 PAH including BaP at 0.1 AM). In addition, extracted Sterling particles were mixed with EPA mix during addition to the cells to see whether particles are able to (re) adsorb soluble PAHs as suggested from previous absorption experiments (Fig. 1). The results of cell incubations with soluble BaP or Sterling extracts are summarized in Fig. 2, showing a linear relationship between B[a]P concentration in the solution (standard) or extract (Sterling V) and the adduct formation in A549 cells after 24-h incubation. The results of a single experiment with all CBs, extracts, and extracted CBs are shown in Fig. 3A and show that extracts of both Sterling and N330 induce spots in the diagonal radioactive zone (DRZ) that colocate with PAH – DNA adducts. Among the original particles (at 100 Ag/cm2), only Sterling V showed a weak spot located in the same region where the BPDE – DNA spot is located (Fig. 3A). The PAH-free, extracted CB particles and the negative control (TiO2, AF, Ultrafine Degussa P25) particles did not show positive spots within the DRZ. However, also the positive control particles

(Diesel, SRM 2975) showed no detectable spots in the DRZ, whereas the soluble EPA standard PAH mix (12 PAH including B[a]P at 0.1 AM) showed clear spots (box 7, Fig. 3A). Interestingly, the EPA mix that gave a clear spot at 0.1 AM was almost absent when the mix was added along with extracted Sterling V particles. The mean and standard deviations of repeated experiments with Sterling V, its organic extract, and adding with EPA –PAH mix are shown in Fig. 3B. The fact that the adduct spots were higher at similar B[a]P concentration when using Sterling extracts (Fig. 2) led us to investigate the formation of benzo[g,h,i]perylene as a major constituent in Sterling V. However, incubations of RLE with benzo[g,h,i]perylene in concentrations up to 0.8 AM did not show detectable spots in the DRZ (data not shown). In vivo studies Three animals per treatment group of 13 weeks were investigated for lipophilic DNA adducts in lung tissue.

Fig. 2. Dose – response curves of PAH – DNA adducts in A549 cells after incubation with different standard concentrations of B[a]P (open symbols), or with organic extract of Sterling V, using the B[a]P concentration in the extract as the ordinate. Results are the mean and standard deviation of three different experiments, each in duplicate.

162

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

Fig. 3. (A) DNA from A549 epithelial cells incubated with carbon blacks, their organic extracts, and PAH standards. The different panels depict results from different treatments, including control (1, no treatment), blank (2, only vehicle DMSO), Sterling V extract (3), Lampblack 1001-extract (4), N-330 extract (5), Printex-90 extract (6), mix of EPA – PAH standards (7, all 0.1 AM), and EPA mix plus Sterling particles (8, 100 Ag/cm2). The panels in the bottom two rows results from different particle treatment including Sterling-extracted particles (9), and original particles at 100 Ag/cm2 of Sterling (10), Lampblack-101 (11), N-330 (12), Printex-90 (13), TiO2 (14), and Diesel-NIST (15). The last panels show B[a]P-diol-epoxide (BPDE)-modified DNA adduct standard with a level of 10 adducts per 108 unmodified nucleotides and control (untreated cells). (B) PAH – DNA adduct formation induced by Sterling V, its organic extract, and extracted particles in A549 cells. The data are the mean and standard deviation (bars) of three different experiments, each in duplicate. Concentrations used for particle experiments (100 Ag/cm2), Sterling extract (AM BaP), and EPA standards (0.1 AM).

Representative examples of chromatograms are shown in Figs. 4 and 5, and show no clear diagonal radioactive zones (DRZs). Generally, after exposure to complex mixtures containing PAHs, DRZs are observed that represent a large number of aromatic DNA adducts (Van Schooten et al., 1997). As can be seen in Figs. 4 and 5, predominantly adduct spots were observed that are also present in the nonexposed control animals. Using the NP1 enrichment procedure, more intense spots were observed at the right uppersite at one exposure level (7 mg/m3) of Printex 90 only (Fig. 4; panels E and F). No intense spot was seen at the other exposure used and the spot is outside of the typical DRZ zone. Additional spots were only observed in

Printex 90-exposed animal lung DNA using butanol enrichment (1 and 7 mg/m3) (Fig. 5; panels C, D, and F). However, all these spots are also out of the DRZ and are suggested to be endogenous adducts related to oxidative stress (Randerath et al., 1999).

Discussion Our results using acellular, in vitro, and in vivo approaches demonstrate that under the conditions and sensitivity of the assays used, (i) PAHs when present on different carbon blacks do not leach into aqueous media

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

Fig. 4. 32P postlabeling detection of aromatic DNA adducts in lung DNA of rats after in vivo exposure to carbon black. Nuclease P1 enrichment was used with multidirectional ion-exchange chromatography. The different panels depict results from single rats, including control samples (A and B) and rats exposed for 13 weeks by inhalation to 1 mg/m3 (C and D), 7 mg/m3 (E and F), or 50 mg/m3 (G and H) Printex 90. Panels I and J show postlabeling results of rats exposed 13 weeks to 50 mg/m3 Sterling V. The last panel (K) shows B[a]P-diol-epoxide (BPDE)-modified DNA adduct standard with a level of 10 adducts per 108 unmodified nucleotides. The small b indicated background spots.

with or without suppletion of surfactant and (ii) that PAHs are not bioavailable from three out of four tested commercial products that contain little PAHs. Some evidence suggests that in the carbon black sample with high PAH content (Sterling V) some PAHs are bioavailable for biotransformation and induction PAH – DNA adducts. The main candidate PAH for this (benzo[g,h,i] perylene) appeared negative in our test system and therefore the identity of the PAH remains obscure. Furthermore, no dose response was observed with regard to the spot intensity of this carbon black type. Bioavailability of PAHs from organic particles such as diesel and carbon black has been the subject of a large series of in vitro and in vivo studies. In vitro leaching of B[a]P loaded onto several rubber-grade carbon blacks using phospholipids vesicles as used in our study showed a biphasic elution pattern, noting that not all B[a]P was eluted (Bevan and Worell, 1985). The concentrations of B[a]P were, however, 100-fold higher (160 – 330 ppm) than the endogenous B[a]P associated with these CB (1.5 – 5 ppm). Interestingly, a similar biphasic elution of B[a]P was observed after in vivo administration of diesel particles that were organic denuded and then loaded with a high (14.5 ng/ Ag) amount of B[a]P (Gerde et al., 2001). These outcomes suggest that below a certain percentage of surface monolayer (15%) PAHs are strongly bound and that bioavailability is negligible. Binding of PAHs beyond 15% of surface monolayer results in rapid leaching from the particle. This is confirmed by studies using four rubbergrade CBs evaluating the elution of endogenous and artificially added PAH from the CBs (Bevan and Yonda, 1985). Using dimyristoyl phophatidylcholine (PMPC) vesicles as a medium reflecting surfactant in the lung,

163

elution of PAH was observed in two out of four CBs (Bevan and Yonda, 1985). Leaching of B[a]P and benzo[ghi]perylene was between around 0.2% of the total amount and observed in the CB with the lowest surface area and highest B[a]P content. A release of 0.2% BaP in our test system using 3 mg/ml of CB particles under our experimental conditions would have led to a maximal concentration of 40 pg/ml in the case of Sterling V, which is well below the analytical detection limit of 40 ng/ml. The strong absorption of PAH to CB and diesel was also shown by our experiments in aqueous medium as well as previous work by Bevan (Bevan and Worell, 1985; Bevan and Yonda, 1985) in toluene and cyclohexane. We showed that all B[a]P in aqueous solution is absorbed to particles at a concentration up to 300 Ag B[a]P/ml when 1 mg/ml particle is present. This is also the explanation why no adduct spots were seen in DNA adducts of epithelial cells coincubated with CBs and PAHs that were added concomitantly to cell incubations in the form of EPA mixture. Although surfactant was used to mimic surface tension in the lung, which is suggested to facilitate leaching of PAHs, particle uptake in vivo occurs within a few hours after deposition in macrophages. After phagocytosis, they are contained in an acidic milieu in the phagolyzome (pH 4.4), which is well know to degrade mineral and organic fibers, but unstudied with regard to leaching of components absorbed on high-surface materials. Therefore, we cannot exclude that leaching or metabolization of PAHs occurs in inflammatory cells by reactive oxygen species (Borm et al., 1997), lysosomal enzymes, or the acidic environment. However, the in vivo findings in rat lungs after different doses of Printex 90 do not support such a mechanism. At three different lung burdens of Printex 90 (and one of

Fig. 5. 32P postlabeling detection of aromatic DNA adducts in vivo using butanol enrichment with multidirectional ion-exchange chromatography. The different panels depict results from single rats as shown in Fig. 2, including control samples (A and B) and rats exposed for 13 weeks by inhalation to 1 mg/m3 (C and D), 7 mg/m3 (E and F), or 50 mg/m3 (G and H) Printex 90. Panels I and J show postlabeling results of rats exposed 13 weeks to 50 mg/m3 Sterling V. The last panel (K) shows B[a]P-diol-epoxide (BPDE)-modified DNA adduct standard with a level of 10 adducts per 108 unmodified nucleotides. The small b indicated background spots.

164

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

Sterling), no different adduct levels were found, despite significant differences in lung inflammation (Gallagher et al., 2003). Bulky DNA adducts in rodent lungs have been analyzed previously using 32P postlabeling after exposure to carbon black and diesel emissions. These studies show rather contrasting findings (Bond et al., 1990; Iwai et al., 2000; Sato et al., 2000; Wong et al., 1986). Initial studies showed increased PAH –DNA adduct formation in rat lungs after inhalation of DEP (Wong et al., 1986), but it was pointed out that at the resulting exposure level (0.35 mg/m3) no increased tumors were seen (Mauderly et al., 1987). Importantly also, DNA adduct formation as observed with a nontumorigenic dose of DEP appeared to show similar adduct levels as seen at much higher dose (Bond et al., 1990). Other studies showed increased BPDE – DNA adducts in the lungs of F344 rats following exposure to B[a]P (2 mg/m3) absorbed onto CB (97 mg/m3), but the adduct levels were not different from those seen upon treatment of B[a]P alone, despite a longer retention (Wolff et al., 1989). As such, the latter study indicates high availability of B[a]P from the particles, likely as a result of the high B[a]P load (>20 mg/g). Similar studies on PAH – DNA adducts in animals chronically exposed to lower concentrations DEP or carbon black showed no increased DNA adducts in comparison to the adduct levels as observed in the lungs of sham-exposed rats (Gallagher et al., 1994). In the current study, we used both Nuclease P1 and butanol enrichment to determine bulky adducts from PAHs as well as endogenous adducts that could arise from lipid peroxidation and inflammation that are out of the DRZ (Randerath et al., 1999). No adducts were found in DNA from lung homogenates isolated immediately after 13-week inhalation of 50 mg/m3 Sterling V or Printex 90, which resulted in particle lung burdens of 7.6 mg (Sterling V) and 4.9 mg, respectively (Printex 90). Although this burden is significantly lower compared to the lung burden of Printex 90 after a 6- and 24-month inhalation (Gallagher et al., 1994; 20 and 44 mg, respectively), Sterling V contains at least 1000-fold the amount of PAHs as compared to Printex 90. If all of the PAHs from Sterling would be bioavailable, the total dose of B[a]P for Sterling over 13 weeks would be 52 ng and for Benzo[ghi]perylene even 684 ng per lung. If 10% of this amount would be metabolized and subsequently form PAH – DNA adducts, an adduct level of 5.2 ng BPDE/ mg DNA would have been observed, which conforms to 130 adducts per 107 nucleotides. In comparison to a detection limit for BPDE – DNA of 1 adduct per 108 nucleotides, a release of 0.1% combined to a metabolic conversion of 10% from CB would have been detectable. Several reasons may explain why we did not detect an increase in total DNA adduct levels (DRZ) or specific spots at the 13-week time point. No data are available on in vivo DNA repair in the rat lung, and extrapolating in vitro repair from immortalized cells to in vivo data is subject to considerable doubt. DNA

repair in the lung was suggested to explain the rapid decrease in PAH – DNA adducts after an initial increase at day 1 after intratracheal instillation of non-particle bound B[a]P (Godschalk et al., 2000). In the current protocol of sub-chronic inhalation with particle-associated PAH, we might expect a steady-state of DNA repair activity. On the other hand the data from a previous study (Godschalk et al., 2000) suggest that the rate of repair is driven by the initial adduct level, since at low levels (0.5  108 nucleotides) in the lung caused by dermal application, repair over 20 days was not detectable. On the other hand, intratracheal and oral application of soluble BaP caused higher adduct levels (20 and 7  108 nucleotides, respectively) and repair was according to these levels. The in vitro adduct levels as observed in our current study (1– 5 per 108 nucleotides with CB particles) are indicative of a low level of DNA adducts and do not support a role of DNA repair in not detecting PAH –DNA adducts in our in vivo experiments. Secondly, epithelial cells along the bronchial and alveolar lining contain the highest amount of CYP450 enzyme systems that are able to metabolize PAH. Adducts that predominantly occur in these target cells are in the present study diluted with DNA from cells in the lung that are no targets or have little to no CYP450 isozymes that can convert PAHs. The relevance of this ‘‘dilution effect’’ is however not clear. Although the epithelium is the relevant, proliferating cell for tumor initiation in the lung (Knaapen et al., 2004), the target cells in the rat for lung cancer may not be the same as those for humans (Borm et al., 2004) and adducts will also occur in other cells following exposure (Borm et al., 1997). With regard to this aspect it is worthwhile to mention that the increased cells in the lung due to the inflammatory cells are included in the total lung DNA. Subchronic exposure to the CBs led to a 10-fold (Sterling V) and 20-fold (Printex 90) increase of total cells obtained by lavage (Gallagher et al., 2003). On the one hand this increased the possibility to find PAH –DNA adducts since these cells contain most of the CB particles (Ferin et al., 1992). On the other hand if no adducts are there, the DNA of this cell-pool further dilutes the PAH –DNA level in the total lung homogenate. Also of importance will be that the subchronic exposure and associated inflammatory response affects cell turnover, e.g., epithelial cell proliferation. This may lead to a further dilution of the adducts recovered. A third reason for our negative in vivo data may be related to the findings of Gerde et al. (2001), who instilled B[a]p-loaded diesel particles in dogs and showed that only 20% of the desorbed B[a]P was absorbed into the metabolically competent epithelium. This leads to a different metabolic profile between soluble and particle-associated B[a]P. Finally, the inflammatory cell influx and the enhanced lung cell turnover as reported in Gallagher et al. (2003) provide a further dilution of the possibly adducted DNA by cell proliferation and hyperplasia (Borm et al., 2004). Indirectly, this also suggests that the formation of extra PAH –DNA adducts in epithelial cells by activated

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

neutrophils in vitro (Borm et al., 1997) may not be that important in vivo despite the large inflammatory response in both Sterling V- and Printex 90-treated rats (Gallagher et al., 2003). On the other hand, the levels of 8-OHdG, also shown to be mediated in epithelial cell DNA by coincubation with PMN in vitro (Knaapen et al., 1999), were found increased in Printex 90-treated rats. In these rats, total cell count was increased 15-fold reaching PMN levels of 50% of total cells. To overcome some of the above experimental limits and uncertainties of the in vivo experiments, in vitro studies using metabolically competent target epithelial cells were done. First of all, cell incubations were done with organic extracts of CBs at target concentrations B[a]P that were shown to lead to detectable PAH – DNA adducts, estimated at 0.02 AM B[a]P. Indeed, the carbon blacks without spots in DRZ (Printex 90 and Lampblack 101) did not reach this final concentration of B[a]P in the medium, despite extraction and concentration of large amounts of CB (6 – 20 g). The cells were incubated with original and extracted carbon black particles in concentrations between 30 and 300 Ag/cm2 that were shown to be noncytotoxic using two independent assays (LDH, MTT). The results show no detectable PAH –DNA adducts in A549 cells, except for Sterling V that showed an adduct that co-located with BPDE – DNA at 100 Ag/cm2 but not at 30 and 300 Ag/cm2. Several potential explanations were tested. First of all, a contamination was excluded as a possible cause since a second independent batch of Sterling V gave the same results. Secondly, Sterling V also showed PAH – DNA adducts in another epithelial cell line (data not shown) used previously for genotoxicity work with quartz (Schins et al., 2002). Thirdly, as the major PAH in Sterling V, benzo [g,h,i] perylene was investigated for its ability to cause DNA adducts in A549 cells and showed no adducts up to 8.5 AM final concentration. This in itself is disconcordant with findings from Hughes and Phillips (1993) who found evidence that benzo[g,h,i] perylene has DNA binding capacities by showing B[ghi]P – DNA adducts in topically treated mouse skin and in vitro in the presence of 3-methylcholanthrene-induced rat liver microsomal preparations. It needs to be emphasized that all conclusions and extrapolations are related to the details and conditions of the in vitro models and the sensitivity of the 32P-postlabeling assay. Although very sensitive, a concentration of 0.01 AM of soluble B[a]P is needed to produce a detectable amount of DNA adducts (Fig. 2). Particle concentrations needed to reach such a virtual concentration of B[a]P would be 1000 Ag/cm2 for Sterling V assuming that all B[a]P is soluble. At this level, particle concentrations are toxic to cells and can only be achieved using repeated in vitro administrations followed by washing steps. In addition, such particle mass/ cell surface concentrations are totally irrelevant with regard to surface dose, which is achieved after in vivo inhalation or instillation. For instance, an instillation of 60 mg of Printex 90, known to cause lung tumors in 60.4% of rats (Borm et al.,

165

2004), would only lead to a dose of 20 Ag/cm2 CB in vivo considering a lung surface for the rat of 3000 cm2. The concentration at which our in vitro experiments were done (100 Ag/cm2) would be equivalent to a 300-mg intratracheal instillation. Therefore, the in vitro data with both the organic extracts and original particles should be regarded as a highly artificial situation that will occur neither under normal nor excessive exposure conditions. Apart from the difference in particle surface dose, the PAH dose rate between in vivo and in vitro experiments is even more different since the desorption of PAHs in the 7.6-mg Sterling (maximum 0.05 Ag BaP) is spread over 12 weeks (= 2016 h) and leads to an in vivo dose rate of 0.007 pg/h cm2 BaP in the lung. A similar calculation in the in vitro system shows that a concentration of 100 Ag/cm2 over 24 h equals an in vitro dose rate of 28 pg/ h cm2 of B[a]P. In our in vitro study we did not use particles pretreated in surfactant which partly limits the conclusions of our in vitro studies with epithelial cells. Previous studies have demonstrated that although primary components of that surfactant do not extract materials with strong in vitro genotoxic activity from DEP (Wallace et al., 1987), mixing DEP into a DPPC in saline dispersion without filtering out the particles results in a surfactant dispersion of DEP that expresses in vitro genotoxic activity. This genotoxic activity was not associated with a surfactant extract of the DEP, but with the surfactant-coated particles made biologically available. That is, the particles are solubilized but not dissolved. In particular, DEP did not express gene mutation activity in the Salmonella histidine reversion assay when mixed into DPPC-saline (Wallace et al., 1987). DEP mixed into a dispersion of DPPC in saline also has been shown to express mammalian cell genotoxic activities including sister chromatid exchange (Keane et al., 1991), micronucleus induction (Gu et al., 1992), and unscheduled DNA synthesis (Gu et al., 1994). These observations suggest that surfactant does play a role in the genotoxic endpoints of DEP, and therefore we cannot exclude that CB particles by some of these actions may cause DNA damage. However, the genotoxic effects of DEP and CB (Don Porto Carrero et al., 2001) may also be due to other mechanisms than transcytosis of CB particles including intracellular release of ROS, mitochondrial damage, membrane lipid peroxidation, and formation of 4-HNE (Knaapen et al., 2004). In addition, even though genotoxicity was studied and observed, the central question would be how these findings have to be interpreted to in vivo events in rat and human (Borm et al., 2004). Regardless of the exact mechanism of genotoxicity by DEP and CB, the main purpose of the study was to evaluate the release and bioavailability of PAHs in vitro and in vivo in order to facilitate the process of risk assessment which would be clearly different if free genotoxic PAHs in cells are found. Using the methods and approaches as described, we were unable to show that PAHs are bioavailable from the four tested CBs. Previous work with leaching of parent PAHs from CB suggested a release of 0.2% of endogenous PAHs on CB

166

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167

(Bevan and Yonda, 1985) and that below a certain monolayer surface load (15%) with exogenous PAHs little to no leaching takes place. Our work more or less confirms that in one CB with the highest PAH content and a moderate surface area, release of PAH is just enough to cause unidentified DNA adduct spots in the DRZ. The data also add up to the evidence that PAHs play no significant role in the induction of lung tumors upon inhalation of ultrafine and fine CB in rats (Borm et al., 2004; Dasenbrock et al., 1996). The use of high doses of organic particles loaded with PAHs is therefore limited to mechanistic interpretations. The carcinogenic action of CBs in the rat is generally considered as the chronic endpoint of persistent inflammation and cell proliferation during exposures leading to overload (Borm et al., 2004; Gallagher et al., 2003; Greim et al., 2001). Among the responsible mediators, reactive oxygen and nitrogen species (ROS, RNS) play a crucial role in genotoxicity, mutagenicity, and cell proliferation pathways (Knaapen et al., 2004). Although PAHs can also be activated by ROS generated by neutrophils and form adducts in epithelial cells (Borm et al., 1997), the contribution of this mechanism to PAH – DNA adduct formation was not detectable in the current as well as in previous (Gallagher et al., 1994) in vivo studies reaching overload conditions.

Acknowledgments The authors gratefully acknowledge the contribution of Dr. Jane Gallagher who provided the extracted DNA from animals exposed to different carbon blacks, and Dr. Nils Kruger (Degussa) and Dr Len Levy (University of Leicester) for acting as study monitors. This research was partially supported by the International Carbon Black Association (ICBA) and the DFG (SFB 503).

References Alink, G.M., Smith, H.S., van Houft, J.J., Kolkman, J.R., Boleij, J.S., 1983. Mutagenic activity of airborne particulates at non-industrial locations. Mutat. Res. 116, 21 – 34. Alvarez, J.G., Slomovic, B., Ludmir, J., 1995. Analysis of dipalmitoyl phosphatidylcholine in amniotic fluid by enzymatic hydrolysis and high-performance thin-layer chromatography reflectance spectrodensitometry. J. Chromatogr. 665, 79 – 87. Bevan, D.R., Worell, W.J., 1985. Elution of benzo[a]pyrene from carbon blacks into biomembranes in vitro. J. Toxicol. Environ. Health 15, 697 – 710. Bevan, D.R., Yonda, N.T., 1985. Elution of polyarmotica hydrocarbons from carbon blacks into biomembranes in vitro. Toxicol. Ind. Health 1, 57 – 67. Binkova, B., Cerna, M., Pastorkova, A., Jelinek, R., Benes, I., Novak, J., Sram, R.J., 2003. Biological activities of organic compounds adsorbed onto ambient air particles: comparison between the cities of Teplice and Prague during the summer and winter seasons 2000 – 2001. Mutat. Res. 525, 43 – 59. Bond, J.A., Mauderly, J.L., Wolff, R.K., 1990. Concentration- and timedependent formation of DNA adducts in lungs of rats exposed to diesel exhaust. Toxicology 60, 127 – 135. Borm, P.J.A., Knaapen, A.M., Schins, R.P.F., Godschalk, R.W.L., van

Schooten, F.J., 1997. Neutrophils amplify the formation of DNAadducts by benzo(a)pyrene in lung target cells. Environ. Health Perspect. 105 (Suppl. 5), 1089 – 1093. Borm, P.J.A., Schins, R.P.F., Albrecht, C., 2004. Inhaled particles and lung cancer. Part B: paradigms and risk assessment. Int. J. Cancer 110, 3 – 14. Buschini, A., Cassoni, F., Anceschi, E., Pasini, L., Poli, P., Rossi, C., 2001. Urban airborne particulate: genotoxicity evaluation of different size fractions by mutagenesis tests on micro-organisms and comet assay. Chemosphere 44, 1723 – 1736. Cavalieri, E., Rogan, E., Chakravarti, D., 2004. The role of endogenous catechol quinines in the initiation of cancer and neurodegenerative diseases. Methods Enzymol. 382, 293 – 319. Dasenbrock, C., Peters, L., Creutzenberg, O., Heinrich, U., 1996. The carcinogenic potency of carbon particles with and without PAH after repeated intratracheal administration in the rat. Toxicol. Lett. 88, 15 – 21. Don Porto Carrero, A., Hoet, P.H.M., Verschaeve, L., Schoeters, G., Nemery, B., 2001. Genotoxic effects of carbon black particles, diesel exhaust particles, and urban air particulates and their extracts on a human alveolar epithelial cell line (A549) and a human monocytic cell line (THP-1). Environ. Mol. Mutagen. 37, 155 – 163. Dukovich, M., Yasbin, R.E., Lestz, S.S., Risby, T.H., Zweidinger, R.B., 1981. The mutagenic and SOS-inducing potential of the soluble organic fraction collected from diesel particulate emissions. Environ. Mutagen. 3 (3), 253 – 264. Ferin, J., Oberdorster, G., Penney, D.P., 1992. Pulmonary retention of ultrafine and fine particles in rats. Am. J. Respir. Cell Mol. Biol. 6, 535 – 542. Gallagher, J., Heinrich, U., George, M., Hendee, L., Phillips, D.H., Lewtas, J., 1994. Formation of DNA adducts in rat following chronic inhalation of diesel emissions, carbon black and titanium dioxide. Carcinogenesis 15, 1291 – 1299. Gallagher, J., Sams II, R., Inmon, J., Gelein, R., Elder, A., Oberdorster, G., Prahalad, A.K., 2003. Formation of 8-oxo-7,8dihydro-2V-deoxyguanosine in rat lung DNA following subchronic inhalation of carbon black. Toxicol. Appl. Pharmacol. 190, 224 – 231. Gerde, P., Muggenburg, B.A., Lundborg, M., Dahl, A.R., 2001. The rapid alveolar absorption of diesel soot-adsorbed benzo[a]pyrene: bioavailability, metabolism and dosimetry of an inhaled particle-borne carcinogen. Carcinogenesis 22, 741 – 749. Godschalk, R.W., Moonen, E.J., Schilderman, P.A., Broekmans, W.M., Kleinjans, J.C., van Schooten, F.J., 2000. Exposure route dependent DNA adduct formation by polycyclic aromatic hydrocarbons. Carcinogenesis 21, 87 – 92. Greim, H., Borm, P.J.A., Schins, R.P.F., Donaldson, K., Driscoll, K.E., Hartwig, A., Kuempel, E., Oberdo¨rster, G., Speit, G., 2001. Toxicity of fibers and particles. Report of the workshop held in Munich, Germany. Inhalation Toxicol. 13, 101 – 119. Gu, Z.W., Zhong, B.Z., Nath, B., Whong, W.-Z., Wallace, W.E., Ong, T., 1992. Micronucleus induction and phagocytosis in mammalian cells treated with diesel emission particles. Mutat. Res. 279, 55 – 60. Gu, Z.W., Zhong, B.Z., Keane, M.J., Whong, W.-Z., Wallace, W.E., Ong, T., 1994. Induction of unscheduled DNA synthesis in V79 cells by diesel emission particles dispersed in simulated pulmonary surfactant. Ann. Occup. Hyg. 38, 345 – 349. Haguenoer, J.M., Shirali, P., Hannothiaux, M.H., Nisse-Ramond, C., 1996. Interactive effects of polycyclic aromatic hydrocarbons and iron oxide particles. Epidemiological and fundamental aspects. Cent. Eur. J. Public Health 4, 41 – 45. Hall, M., Grover, P.L., 1990. Polycyclic aromatic hydrocarbons: metabolism, activation and tumor initiation. In: Cooper, C.S., Grover, P.L. (Eds.), Chemical Carcinogenesis and MutagenesisSpringer, Berlin, pp. 227 – 272. Hughes, N.C., Phillips, D.H., 1993. 32P-postlabelling analysis of the covalent binding of benzo[ghi]perylene to DNA in vivo and in vitro. Carcinogenesis 14, 127 – 133.

P.J.A. Borm et al. / Toxicology and Applied Pharmacology 205 (2005) 157 – 167 Iwai, K., Adachi, S., Takahashi, M., Moller, L., Udagawa, T., Mizuno, S., Sugawara, I., 2000. Early oxidative DNA damages and late development of lung cancer in diesel exhaust-exposed rats. Environ. Res. 84, 255 – 264. Keane, M.J., Xing, S.G., Harrison, J.C., Ong, T., Wallace, W.E., 1991. Genotoxicity of diesel exhaust particles dispersed in simulated pulmonary surfactant. Mutat. Res. 304, 379 – 406. Knaapen, A.M., Seiler, F., Schilderman, P.A.E.L., Nehls, P., Bruch, J., Schins, R.P.F., Borm, P.J.A., 1999. Neutrophils cause oxidative DNA damage in alveolar epithelial cells. Free Rad. Biol. Med. 27, 234 – 240. Knaapen, A., Borm, P.J.A., Albrecht, C., Schins, R.P.F, 2004. Inhaled particles and lung cancer. Part A: mechanisms. Int. J. Cancer 109, 799 – 809. Mauderly, J.L., Jones, R.K., Griffith, W.C., Henderson, R.F., McLellan, R.O., 1987. Diesel exhaust is a pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. Appl. Toxicol. 9, 208 – 211. Randerath, K., Randerath, E., Zhou, G.D., Li, D., 1999. Bulky endogenous DNA modifications (I-compounds)—Possible structural origins and functional implications. Mutat. Res. 424, 183 – 194. Rausch, L.J., Bisinger, E.C., Sharma, A., 2004. Carbon black should not be classified as a human carcinogen based on rodent bioassay data. Regul. Toxicol. Pharmacol. 40, 28 – 41. Sato, H., Sone, H., Sagai, M., Suzuki, K.T., Aoki, Y., 2000. Increase in mutation frequency in lung of Big Blue rat by exposure to diesel exhaust. Carcinogenesis 21, 653 – 661.

167

Schins, R.P.F., 2002. Mechanisms of genotoxicity of particles and fibres. Inhalation Toxicol. 14, 57 – 78. Schins, R.P.F., Duffin, R., Ho¨hr, D., Knaapen, A.M., Shi, T., Weishaupt, C., Stone, V., Donaldson, K., Borm, P.J.A., 2002. Surface modification of quartz inhibits toxicity, particle uptake, and oxidative DNA damage in human lung epithelial cells. Chem. Res. Toxicol. 15, 1166 – 1173. Sun, J.D., Wolff, R.K., Kanapilly, G.M., McClellan, R., 1984. Lung retention and metabolic fate of inhaled benzo(a)pyrene associated with diesel exhaust particles. Toxicol. Appl. Pharmacol. 30, 48 – 59. Van Schooten, F.J., Godschalk, R., Breedijk, A., Maas, L., Kriekm, E., Sakai, H., Wigbout, G., Baas, P., Van ’t Veer, L., Van Zandwijk, N., 1997. 32P-postlabeling of aromatic DNA adducts in white blood cells and alveolar macrophages of smokers; saturation at high exposures. Mutat. Res. 378, 65 – 75. Wallace, W.E., Keane, M.J., Hill, C.A., Xu, J., Ong, T., 1987. Mutagenicity of diesel exhaust particles and oil shale particles dispersed in lecithin surfactant. J. Toxicol. Environ. Health 21, 163 – 171. Wolff, R.K., Bond, J.A., Sund, J.D., Henderson, R.F., Harkema, J.R., Griffith, W., Mauderly, J.L., McClellan, R.O., 1989. Effects of adsorption of benzo(a)pyrene onto carbon black particles on levels of DNA damage in lungs of rats exposed by inhalation. Toxicol. Appl. Pharmacol. 97, 289 – 299. Wong, D., Mitchell, C.E., Wolff, R.K., Mauderly, J.L., Jeffrey, A.M., 1986. Identification of DNA damage as a result of exposure to rats to diesel exhaust. Carcinogenesis 7, 1595 – 1597.