1426
Chem. Res. Toxicol. 2007, 20, 1426–1433
Biological Effects of Particles from the Paris Subway System Rafik Bachoual,†,‡ Jorge Boczkowski,†,§ Delphine Goven,† Nadia Amara,† Lyes Tabet,† Dinhill On,† Véronique Leçon-Malas,| Michel Aubier,†,§ and Sophie Lanone*,† Inserm, U700 UniVersité Paris 7 Denis Diderot, site Bichat, Paris, France, Faculté des Sciences de Gabès, Tunisia, and CIC 007 and Biochimie B, Assistance Publique-Hôpitaux de Paris, Hôpital Bichat, Paris, France ReceiVed March 28, 2007
Particulate matter (PM) from atmospheric pollution can easily deposit in the lungs and induce recruitment of inflammatory cells, a source of inflammatory cytokines, oxidants, and matrix metalloproteases (MMPs), which are important players in lung structural homeostasis. In many large cities, the subway system is a potent source of PM emission, but little is known about the biological effects of PM from this source. We performed a comprehensive study to evaluate the biological effects of PM sampled at two sites (RER and Metro) in the Paris subway system. Murine macrophages (RAW 264.7) and C57Bl/6 mice, respectively, were exposed to 0.01–10 µg/cm2 and 5–100 µg/mouse subway PM or reference materials [carbon black (CB), titanium dioxide (TiO2), or diesel exhaust particles (DEPs)]. We analyzed cell viability, production of cellular and lung proinflammatory cytokines [tumor necrosis factor R (TNFR), macrophage inflammatory protein (MIP-2), KC (the murin analog of interleukin-8), and granulocyte macrophagecolony stimulating factor (GM-CSF)], and mRNA or protein expression of MMP-2, -9, and -12 and heme oxygenase-1 (HO-1). Deferoxamine and polymixin B were used to evaluate the roles of iron and endotoxin, respectively. Noncytotoxic concentrations of subway PM (but not CB, TiO2, or DEPs) induced a time- and dose-dependent increase in TNFR and MIP-2 production by RAW 264.7 cells, in a manner involving, at least in part, PM iron content (34% inhibition of TNF production 8 h after stimulation of RAW 264.7 cells with 10 µg/cm2 RER particles pretreated with deferoxamine). Similar increased cytokine production was transiently observed in vivo in mice and was accompanied by an increased neutrophil cellularity of bronchoalveolar lavage (84.83 ( 0.98% of polymorphonuclear neutrophils for RER-treated mice after 24 h vs 7.33 ( 0.99% for vehicle-treated animals). Subway PM induced an increased expression of MMP-12 and HO-1 both in vitro and in vivo. PM from the Paris subway system has transient biological effects. Further studies are needed to better understand the pathophysiological implications of these findings. Introduction In many large cities, the subway system is an important source of atmospheric pollution; the atmospheric concentration of particulate matter (PM)1 with an aerodynamic diameter of less than 10 µm (PM10) can be as high as 1000 µg/m3 (1, 2), which is much greater than the 50 µg/m3 daily ambient air recommended limit (3). The potential health effects of such emissions are important to evaluate because, in France, for example, the Paris subway system hosts more than 1 million commuters daily (http://www.ratp.fr). A large number of epidemiologic studies have shown that PM10 is associated with adverse cardio-respiratory effects (4, 5). PM10 is a complex mixture of carbon, sulfates, terrigenous particles from erosion and scrap iron, and chemical and * To whom correspondence should be addressed: Sophie Lanone, Inserm U700, Faculté de Médecine Paris 7, site X. Bichat, BP416, 75870 Paris Cedex 18, France. Phone: (33 1) 44856248. Fax: (33 1) 44856257. E-mail:
[email protected]. † Inserm. ‡ Faculté des Sciences de Gabès. § CIC 007, Assistance Publique-Hôpitaux de Paris. | Biochimie B, Assistance Publique-Hôpitaux de Paris. 1 Abbreviations: PM, particulate matter; MMP, matrix metalloprotease; TIMP, tissue inhibitor of MMP; CB, carbon black; DEPs, diesel exhaust particles; HO-1, heme oxygenase-1; LDH, lactate dehydrogenase; BAL, bronchoalveolar lavage; MIP-2, macrophage inflammatory protein; TNF, tumor necrosis factor; IL, interleukin; GM-CSF, granulocyte macrophagecolony stimulating factor; TEOM, tapered element oscillating microbalance; PIXE, particle-induced X-ray emission; DPL, dipalmitoyl lecithin.
biological substances, such as allergens and endotoxins, that can interact with and adsorb on the PM10 (3). The exact composition of PM10 depends on the source of emission, and therefore, PM10 is as various as its different sources and potentially different effects (6–8). PM10 can easily deposit in the airways, and smaller particles such as PM2.5 can deposit in the alveoli (3, 9); inflammation and oxidative stress are considered key mechanisms of their adverse effects (10, 11). Lung inflammation is characterized by the local recruitment of inflammatory cells such as neutrophils and macrophages (12), which leads to increased production of inflammatory cytokines and chemokines such as tumor necrosis factor R (TNFR), interleukin (IL)-1, -6, and -8, and granulocyte macrophagecolony stimulating factor (GM-CSF) (13, 14). Inflammatory cells are also a source of proteases such as matrix metalloproteases (MMPs), which have elastolytic and collagenolytic activities and are important players in lung structural homeostasis (15, 16). Finally, inflammatory cells, as an important source of oxidants, could participate in the development of oxidative stress, which occurs when the endogenous antioxidant systems are overwhelmed by the amount of oxidants present in a biological system. The expression of the inducible isoform of heme oxygenase-1 (HO-1), which has antioxidant properties, is a very sensitive marker of oxidative stress, since its level increases under oxidative stress (17). To date, the biological effects of only two subway systems (London and Stockholm) have been studied (2, 18, 19). These studies showed that a short exposure (4–8 h) of a human
10.1021/tx700093j CCC: $37.00 2007 American Chemical Society Published on Web 09/21/2007
Biological Effects of Paris Subway Particles
epithelial lung cancer cell line (A549) to noncytotoxic concentrations of PM (50 µg/mL and 40 µg/cm2 for the London and Stockholm systems, respectively) induces a genotoxic response and increases IL-8 production. At the same concentrations, PM samples from both subway systems were shown to produce oxidative damage to DNA. The effects of PM from both subway systems were greater than those of the respective reference particles (TiO2 and PM sampled from the street). A subset of experiments in which human blood monocyte-derived macrophages were exposed to PM from the Stockholm system showed only a slight increase in TNFR production and no increase IL-6 or IL-8 production, although exposure to street PM induced substantial production of those three cytokines (18). However, the concentration used (50 µg/mL) was considered cytotoxic for macrophages, making the relevance of the last data questionable. Therefore, given this limited amount of data, which shows a greater biological effect of subway than reference PM under some conditions, the aim of our study was to perform a comprehensive examination of the biological effects of PM10 sampled at two sites in the Paris subway system (RATP) on a murine macrophage cell line (RAW 264.7). We chose two RATP sites differing in exploitation materials (an RER site, with iron wheels and composite brakes, and a Metro site, with pneumatics and wooden brakes), possibly leading to different compositions of PM10 and thus different biological effects. We analyzed the consequences of exposure to PM on cell viability, inflammatory cytokine production, and mRNA and/or protein expression of MMPs, their inhibitors, tissue inhibitors of metalloprotease (TIMPs), and HO-1. The respective roles of iron and endotoxin content of the PM were evaluated. Finally, the relevance of our in vitro results was evaluated in vivo in mice after intratracheal administration of the PM. Subway PM10 was compared to the following reference materials: carbon black (CB), a negative control for the carbonaceous portion of the PM; TiO2, which contains metallic particles; and diesel exhaust particles (DEPs), an important component of urban-pollution PM (20, 21).
Experimental Procedures Chemicals and Reagents. Culture media, supplements, and fetal calf serum were obtained from Life Technologies SARL (Cergy Pontoise, France). Polymixin B and deferoxamine mesylate were procured from Sigma-Aldrich (Saint-Quentin Fallavier, France). Particles. Subway particles were collected November 13–28, 2003, using a 1 m3/h volume sampler machine (Partisol Plus, Ecomesure, Janvry, France) equipped with a PM10 selective inlet head. Particles were sampled at two sites in the Paris subway system (http://www.ratp.fr): the Nation RER A platform (RER) and the Châtelet arrival platform, Metro line 11 (Metro). PM10 was recovered on a Pallflex filter (Teflon-coated fiberglass, 47 mm diameter) and detached from the filter by sonication in water. The recovered suspensions were frozen in liquid nitrogen and lyophilized at -80 °C as previously described (22). PM10 concentrations on station platforms were determined using a tapered element oscillating microbalance (TEOM) (Ecomesure 1400A), elemental compositions using the particle-induced X-ray emission (PIXE) method (23), and size distributions using a GRIMM meter (ICS). Endotoxin levels were determined using the Limulus amoebocyte lysate (LAL) test, as previously described (6). CB particles (95 nm in diameter, FR103) were obtained from Degussa (Frankfurt, Germany). TiO2, 150 nm in diameter, was acquired from Huntsman Tioxide Europe S.A.A. (Calais, France). DEPs (SRM 1650) were purchased from the National Institute of Standards and Technology (Gaithersburg, MD). In Vitro Studies. 1. RAW 264.7 Macrophage Culture. RAW 264.7 murine macrophages were purchased from the American Type
Chem. Res. Toxicol., Vol. 20, No. 10, 2007 1427 Table 1. Primer Sequences for Q-PCR Analysis gene of interest MMP-2 MMP-9 MMP-12 TIMP-1 TIMP-2 HO-1 RpL13
type
sequence
sense antisense sense antisense sense antisense sense antisense sense antisense sense antisense sense antisense
5′-CACACCAGGTGAAGGATGTG-3′ 5′-AGGGCTGCATTGCAAATATC-3′ 5′-TTCTCTGGACGTCAAATGTGG-3′ 5′-CAAAGAAGGAGCCCTAGTTCAAGG-3′ 5′-AAGCAACTGGGCAACTGGACAACTC-3′ 5′-TGGTGACAGAAAGTTGATGGTGGAC-3′ 5′-ATATCCACAGAGGCTTTCCATG-3′ 5′-CATGGGTTCCCCACGAATC-3′ 5′-GGCAGAGCTTTATCACTAACAA-3′ 5′-GAACATTCACAGGCTACCAGAC-3′ 5′-CACGCATATACCCGCTACCT-3′ 5′-CCAGAGTGTTCATTCGAGCA-3′ 5′-GTGGTCCCTGCTGCTCTCAA-3′ 5′-CGATAGTGCATCTTGGCCTTTT-3′
Culture Collection (Manassas, VA). Cells were seeded at 40000 cells/mL and cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% heat-inactivated fetal calf serum and antibiotics (streptomycin, 10 mg/mL; penicillin G, 10000 IU/mL; and amphotericin B, 25 µg/mL) in a humidified atmosphere of 5% CO2/95% air at 37 °C until they reached 80–90% confluence. Following this, they were washed twice with phosphate-buffered saline and incubated in serum-free medium in the presence of the different pollutants (RER PM10, Metro PM10, CB, TiO2, and DEPs). Stock suspensions of particles in a solution of 0.04% dipalmitoyl lecithin (DPL) (Sigma) in distilled water were sonicated for 15 min in an ultrasonic water bath under ice. Particle concentrations were 0.01, 0.1, 1, and 10 µg/cm2 (concentrations are expressed in µg/ cm2 because particles rapidly sediment in cell culture). In all experiments, 10 µg/cm2 was equivalent to 50 µg/mL. This dosage was chosen to allow a comparison of our results to those of the London and Stockholm studies (2, 18, 19). Supernatants were recovered 3, 8, and 24 h after the beginning of the experiment and kept at -80 °C until needed for further analysis. Cells at the 3 and 8 h time points were harvested in RA1 reagent from the NucleoSpin RNA II kit (Macherey-Nagel). In a subset of experiments, particles were incubated overnight at 37 °C in the presence of 2 mM deferoxamine mesylate or 10 µg/mL polymixin B (24, 25). Each solution was then centrifuged (3000 rpm for 5 min), and the pellet was resuspended in 0.04% DPL. 2. Optical Microscopy. Cells were seeded at 40000 cells/mL, exposed for 24 h to 10 µg/cm2 PM, fixed in 4% paraformaldehyde, and stained with Harris hematoxylin–phloxin. 3. Cellular Viability. Cellular viability was assessed using two methods: total cell count followed by trypan blue exclusion and lactate dehydrogenase (LDH) release in the medium (Roche Diagnostics), following the manufacturer’s instructions. 4. Cytokine Assay. Amounts of macrophage inflammatory protein (MIP-2), TNFR, KC (the murin analog of IL-8), and GMCSF in cell-culture supernatants were measured by ELISA (R&D Systems, Lille, France), following the manufacturer’s instructions. 5. Reverse Transcription and Quantitative PCR (Q-PCR). Total RNA of RAW 264.7 cells exposed for 3 or 8 h was isolated using the NucleoSpin RNA II kit, following the manufacturer’s instructions. The amount of total RNA in aqueous solution was determined by absorbance at 260 nm. Equal amounts of total RNA were reverse-transcribed by use of a random hexamer and MMLV reverse transcriptase (Promega, Charbonnières, France). The cDNA obtained was amplified using the SybrGreen JumpStart Taq Ready Mix detection kit (Sigma), and Q-PCR was carried out using the PCR Mx3000P apparatus (Stratagene, La Jolla, CA). In all assays, cDNA was amplified using a standardized program (2′ JumpStart Taq Polymerase activation step at 94 °C; 40 cycles of 30 s at 94 °C and 1 min at 60 °C). All assays were performed in a final volume of 20 µL, and primers were used at a final concentration of 0.33 µM. Primer sets are shown in Table 1. The expression of the gene of interest was reported as the ratio to RpL13 expression.
1428 Chem. Res. Toxicol., Vol. 20, No. 10, 2007 6. Immunohistochemical Analysis. Immunohistochemistry was performed as described previously (26). We used a rabbit polyclonal anti-HO-1 antibody (dilution 1:200, Euromedex). Positive cells were revealed by use of the Vectastain ABC-alkaline phosphatase kit system and fast red substrate (Dako). The specificity of immunostaining was assessed by replacing HO-1 antibody with normal rabbit serum (isotype condition). In Vivo Studies. 1. Animal Exposure. The experiments were approved by the local Institutional Animal Care and Use Committee, and the experimental protocol was in agreement with French legal recommendations related to animal studies. Male C57Bl6 mice, aged 7 weeks and weighing 22.3 ( 0.73 g, were purchased from Janvier. Animals were housed in standard wire-topped cages and temperature-controlled units. Food and water were supplied ad libitum. Particles were suspended in sterile saline (0.09% NaCl), and the suspension was sonicated under ice for 3 min in an ultrasonic water bath. Mice were given a cocktail of anesthetics [75 mg/kg ketamin (Virbac Santé Animale) plus 1 mg/kg medetomidine (Pfizer)], and particles (RER PM10, CB, TiO2, and DEPs) were intratracheally administered at 5, 50, or 100 µg/mouse [0.22–4.48 mg/kg of body weight; the highest dose is relevant to the 10 µg/cm2 concentration used in the in vitro studies (27)]. Mice were awakened with an intraperitoneal injection (1 mg/kg) of atipamezol (Pfizer), an antagonist of medetomidine, and sacrificed 8 or 24 h later. 2. Bronchoalveolar Lavage (BAL) and Lung Recovery. Mice were anesthetized with an intraperitoneal injection of 50 mg of urethane (Sigma) and killed by exsanguination. Lungs were lavaged twice with 1 mL of physiological saline, removed from the chest cavity, and immediately frozen at -80 °C for use in further experiments. The lavage fluid (1.8 mL) was immediately placed on ice. Free alveolar cells were recovered from the lavage fluid by centrifugation at 400g for 15 min at 4 °C. The supernatant was used for analysis of total protein concentration (Quick Start Bradford assay, Bio-Rad, Marnes-la-Coquette, France) and expression of cytokines. The cellular pellet was suspended in 150 µL of physiological saline. An aliquot of the cell suspension was then examined using a hemocytometer to evaluate the total white cell number. For differential counts, the cell suspension was cytospun (cytospin-2, Shandon Products Ltd.), fixed in methanol, and stained with Diff Quick solution (Medion Diagnostics, Plaisir, France). Differential cell counts were determined by counting 100 cells under oil immersion (1000×). mRNA expression of MMP-2, -9, and -12, TIMP-1 and -2, and HO-1 was examined by Q-PCR on total lung homogenates, and BAL MIP-2 and TNFR protein content was quantified by ELISA (as described above) 8 h after particle administration. 3. Statistical Analysis. Each reported value is the mean ( SE of at least four different experiments, each performed in triplicate. The data were analyzed by one-way ANOVA or nonparametric tests, as appropriate. For all tests, p < 0.05 was considered significant.
Bachoual et al. Table 2. Subway PM Characteristicsa concentration on station platform (g/m3) elemental composition (mass %) Fe Mn Ca Cu S Si particle size distribution (% of total number of particles) 1 µm diameter
RER
Metro
360.9
67.5
61 7 0.2 0.45 1.95 1.8
41.8
