Environ. Sci. Technol. 2007, 41, 2622-2629
Toxicity of Parked Motor Vehicle Indoor Air J E R O E N T . M . B U T E R S , * ,† WOLFGANG SCHOBER,† JAN GUTERMUTH,† THILO JAKOB,‡ ANTONIO AGUILAR-PIMENTEL,§ JOHANNES HUSS-MARP,† CLAUDIA TRAIDL-HOFFMANN,† SABINE MAIR,| STEFAN MAIR,| FLORIAN MAYER,| KLAUS BREUER,| AND HEIDRUN BEHRENDT† Division of Environmental Dermatology and Allergy GSF/TUM, ZAUM-Center for Allergy and Environment, Technical University Munich, Munich, Germany, Clinical Research Group Allergology, Department of Dermatology, University Freiburg, Germany, Division of Molecular and Clinical Allergotoxicology, Department of Dermatology and Allergy, Technical University Munich, Munich, Germany, and Fraunhofer-Institute for Building Physics IBP, Branch Institute Holzkirchen, Valley, Germany
The interior of motor vehicles is made of a wide variety of synthetic materials, which emit volatile organic compounds (VOC). We tested the health effects of emissions from vehicles exposed to “parked in sunshine” conditions. A new and a 3 year old vehicle with identical interior were exposed to 14 000 W of light. Indoor air was analyzed by GC-MS. Toxicity of extracts of indoor air was assayed in human primary keratinocytes, human lung epithelial A549 cell line, and Chinese hamster V79 lung fibroblasts. In addition, toxicity after metabolic activation by CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2B6, and CYP2E1 was assayed. The effect on type I allergic reaction (IgE-mediated immune response), type IV allergic reaction (T-cell mediated immune response), and irritative potential was evaluated also. A total of 10.9 and 1.2 mg/m3 VOC were found in new and used motor vehicle indoor air, respectively. The major compounds in the new vehicle were o,m,p-xylenes, C3 and C4-alkylbenzenes, dodecane, tridecane, and methylpyrrolidinone. In the used vehicle they were acetone, methylpyrrolidinone, methylcyclohexane, acetaldehyde, o,m,p-xylenes, ethylhexanol, and toluene. No toxicity was observed in any cell line with or without metabolic activation. Neither did we find an effect on type IV sensitization or an irritative potential. A slight but statistically significant aggravating effect on IgE-mediated immune response of only the new vehicle indoor air was determined (p < 0.05). The IgE-response modulating effect of indoor air might be relevant for atopic individuals. Else no direct * Corresponding author phone: (+) 49-89-41403487; fax: (+) 4989-41403453; e-mail:
[email protected]. Corresponding author address: ZAUM-Center for Allergy and Environment, Technical University Munich, Biedersteiner Strasse 29, 80802 Munich, Germany. † Division of Environmental Dermatology and Allergy GSF/TUM, ZAUM-Center for Allergy and Environment, Technical University Munich. ‡ University Freiburg. § Department of Dermatology and Allergy, Technical University Munich. | Fraunhofer-Institute for Building Physics IBP. 2622
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 7, 2007
toxicity, no toxicity after metabolic activation by cytochrome P450, and no irritative or type IV sensitizing potential of motor vehicle indoor air were found, neither from the new nor used vehicle. Our investigations indicated no apparent health hazard of parked motor vehicle indoor air.
Introduction Humans are exposed to components found in indoor air of motor vehicles, sometimes for prolonged periods. Due to ventilation during driving indoor pollutants are diluted (1, 2), and outside pollutants enter the motor vehicle indoor environment. Air pollution inside driving motor vehicles was often studied (3-6). However, little is known about the health effects of emissions from the interior of cars itself (7, 8). These emissions can be high, especially in new cars or after increased temperatures, as are found after “parked in sunshine” conditions (9, 10). Indoor air components influence the well being of exposed individuals at the work place, and maximal work place concentrations are regulated (11). However, no regulation limits the concentration of components in motor vehicle indoor air (7). In the interior of motor vehicles many novel materials are used which emit volatile organic compounds. Studies on the quantitation of these emissions are reported (2, 9, 10), but no study investigated the potential adverse effects of these emissions found in motor vehicle indoor air. We therefore sampled the indoor air of a new and a used motor vehicle under “parked in sunshine” condition as a “worst case” scenario for high release of volatile organic compounds from the interior of motor vehicles and assayed the direct toxicity, toxicity after metabolic activation, and effects on the immune system of extracts of motor vehicle indoor air.
Materials and Methods Extracts of Parked Motor Vehicle Indoor Air. Two motor vehicles of identical brand and assembly (metallic silver vehicles with black leather interior) but of different ages were sampled. The new vehicle was sampled within 1 month after production and 8 km of usage, and the used vehicle was sampled about 3 years after production and 107 000 km of usage. Both vehicles were parked in a laboratory under 28 halogen lamps (14 000 W), split in 4 groups, and positioned about 0.5 m away from the front, back, and side windows and roof to mimic environmental sunshine exposure. Through the driver’s window an inlet was created by which 12 temperature sensors, one humidity sensor, a sampling tube, an air inlet, and a paddle entered the vehicle. The assumed head position of the driver and navigator served as reference, and the illumination was variable adjusted to maintain a temperature of 65 °C throughout the experiment. Cleaned pressured air was passed through an activated charcoal filter, dehumidified by a cold-trap, again filtered through an activated charcoal filter and particle filter, and flow adjusted by mass-flow regulators, and humidity was adjusted to 50% before entering the vehicles. Flow of air entering the motor vehicle was adjusted to 0.75 m3/h. A paddle with charcoal bearings was used to avoid contaminating indoor air by aliphatic hydrocarbons of the grease of normal bearings. Four blades of about 100 cm2 each circulated the air at 72 rpm. Air sampling was set at 0.75 m3/h, an average respiration rate for motor vehicle drivers (12), under permanent monitoring. Sampled air was passed through 4 cold traps at -80 °C. The first trap was empty and served to dehumidify the 10.1021/es0617901 CCC: $37.00
2007 American Chemical Society Published on Web 03/02/2007
FIGURE 1. Cytotoxicity of extracts (DMSO transferred) of indoor air of a new and a used motor vehicle with human primary keratinocytes of a nonatopic individual (graphs A and B) or human epithelial A549 cells (graphs C and D). Cytotoxic endpoints after 48 h incubations were determined with sulforhodamine B (A and C) and Alamar Blue (B and D). Cell density of cells receiving no compounds was put at 100%. For each point the mean of 8 determinations is given. For clarity the standard deviation of the curve with the highest deviation are depicted only. Air eq is liter indoor air equivalents per liter cell culture medium. O1 is new, O2 is used vehicle. air, the second and third traps contained diethyl ether, and the fourth trap acquired escaped vapors. After sampling the diethyl ether and aqueous phases (from freezing of humidity in the sample) of all traps were combined, shaken, and separated in a separating funnel. The diethyl ether phase was dried over anhydrous sodium sulfate and concentrated to 20 mL at 40 °C at ambient pressure in a Laborota 4000 (Heidolph, Kelheim, Germany) and stored at -70 °C until use. Direct Sampling of Parked Motor Vehicle Indoor Air. In addition to the ether extracts, motor vehicle indoor air was analyzed directly by drawing 0.5-1.0 L of air from the center of the car through the sampling tube inserted through the driver’s window and tightened by emission free aluminum adhesive foil onto a Carbotrap adsorber tube (9 × 0.6 cm, Supelco, Taufkirchen, Germany) at 18 °C using a pump at a velocity of 100 mL/min. Analysis of Parked Motor Vehicle Indoor Air Extracts. Motor vehicle indoor air extracts were analyzed by gas chromatography-mass spectrometry (GC-MS, Agilent 6890 Series, Agilent Technologies, Palo Alto, CA, U.S.A.) equipped with a carbon dioxide oven cooling system, a HP-5 capillary column (50 m × 0.32 mm, film thickness 1.05 µm, J&W Scientific, Folsom, U.S.A.), a flame-ionization-detector (FID), and a mass selective detector (MSD 5793, Agilent Technologies, Palo Alto, CA, U.S.A.). The initial temperature of the GC oven of 20 °C was held for 5 min, then raised at a rate of 5 °C/min to 280 °C, then raised at 20 °C/min to 300 °C, and held for 10 min. Mass spectra were generated at 70 eV by electrical (also called electron impact) ionization. The concentrations of the volatile organic compounds (VOC) were calculated against the external standard toluene. Analysis of Direct Samples of Parked Motor Vehicle Indoor Air. The direct sampled indoor air was analyzed by installing the loaded Carbotrap into an automatic thermal desorption system (ATD 400, Perkin-Elmer, Wellesley, U.S.A.) and analyzing according to ISO 16000-6. Samples were desorbed from the trap in the ATD by heating up to 360 °C
in a stream of helium (33 mL/min) for 10 min. The volatiles were cryofocused in a cold trap, which was piezzo-electrically cooled down to 5 °C. For injection the trap was heated at 350 °C for 3 min, and the volatiles were flushed in a stream of helium onto the gas chromatography column. Analysis was performed by GC-FID-MS (see above). Identification of volatile organic compounds was based on retention time and mass spectra in accordance with authentic reference compounds, IBP-own, Wiley, or NIST-MS-library search. Concentrations were determined by comparing the FID signal area counts with an external toluene standard calibration curve. Analysis of Aldehydes and Ketones in Direct Sampled Parked Motor Vehicle Indoor Air. For analysis of volatile aldehydes and ketones 60 L of in-car-air was drawn on silica cartridges impregnated by dinitrophenylhydrazine (Sep-Pak DNPH cartridges, Waters, Eschborn, Germany). The dinitrophenylhydrazone derivatives of the aldehydes and ketones formed were eluted from the cartridge by 2 mL of acetonitrile. Analysis of the DNPH-derivatives was performed by highperformance liquid chromatography with a diode array detector (HPLC-DAD, 1100 series, Hewlett-Packard/Agilent Technologies, Palo Alto, CA, U.S.A.) on a C18 HPLC column (150 × 3.9 mm, Nova-Pak, Waters, Eschborn, Germany) according to ISO 16000-3. Identification was based on comparison of retention time and UV-spectra with authentic reference compounds, and quantification was done by compound specific calibration using standard solutions of different concentrations. Detection limit was 0.5-3.0 µg/ m3, depending on the compound. Toxicity in Cell Culture. Diethyl ether extracts of motor vehicle indoor air were mixed with an equal volume of DMSO, and ether was evaporated at room temperature under a gentle stream of nitrogen within 15 min. Chinese hamster V79 lung fibroblasts (13), human alveolar epithelial A549 cells (14), and human primary keratinocytes isolated by suction blister VOL. 41, NO. 7, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2623
FIGURE 2. Cytotoxicity after metabolic activation by cytochromes P450 CYP1A1, CYP1A2, CYP1B1, CYP2A6, CYP2B6, and CYP2E1. Cytochromes were expressed in Chinese hamster V79 lung fibroblasts and incubated for 3 days with dilutions of extracts of motor vehicle indoor air dissolved in DMSO or DMSO alone. Toxicity was determined with sulforhodamine B (A), Alamar Blue (B), and CellTiter-Glo (C). Air eq. is liter indoor air equivalents per liter cell culture medium. Each data point represents the mean and standard deviation of 8 replicates. (15) from an healthy male nonatopic individual were incubated with dilutions of motor vehicle indoor air extracts. After incubation toxicity was assessed by sulforhodamine B (Sigma-Aldrich Inc., St. Louis, MO) (16), alamar blue (Serotec Ltd., Oxford, U.K.) (17), or CellTiter-Glo (Promega Inc., Madison, WI) (18). An identical dilution of the solvent DMSO was run in parallel with each assay. Keratinocytes and A549 cells were incubated for 48 h, V79 cells for 72 h. To assess cytotoxicity after metabolic activation a cDNAdirected expression system of cytochromes P450 in V79 cells was used, using the same endpoints as with the other cells. The expressed cytochromes P450 were catalytically active as tested with substrates specific for the individual cytochromes and are being published elsewhere (Schober et al. Toxicology submitted 2006). β-Hexosaminidase Release from the Murine Mast Cell Line L138.8A. The bone marrow derived murine mast cell 2624
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 7, 2007
line L138.8A (19) was grown in the presence of 0.02 U/mL IL-3. Before an experiment cells were grown overnight in the presence of purified mouse IgE (5 µg/mL, Pharmingen Inc., San Diego, CA). The cells were washed and incubated for 60 min with 0.5 µg/mL anti-IgE (R35-72, Pharmingen) and motor vehicle indoor air extracts in Tyrode’s buffer at 37 °C. β-Hexosaminidase release in the supernatant from 200 000 cells was assessed colorimetric by p-nitrophenol liberation from p-nitrophenol-D-2-acetamido-2-deoxyglucopyranoside (Sigma-Aldrich) at 405 nm. Mast cells treated with 0.5% Triton-X revealed maximal β-hexosaminidase release. IgE/ anti-IgE concentrations were adjusted to yield about 20% of maximal β-hexosaminidase release. This basic activation level served as control value, depicted as 100% in Figure 3 (20). Mouse Local Lymph Node (LLNA) and Ear Swelling Assay. Irritative and sensitizing potential of motor vehicle indoor air was assessed in a modified mouse LLNA and ear-
FIGURE 3. Enhancement of antigen induced β-hexosaminidase release by parked motor vehicle indoor air extracts of a new (A) and a used (B) but otherwise identical vehicles. DMSO was e0.1% v/v in all experiments. Mean ( SD of at least three experiments are depicted. Air eq. is liter indoor air equivalents per liter cell culture medium. β-Hexosaminidase release of incubations with only IgE/anti-IgE were put at 100%. * p < 0.05 swelling test (21). Groups of five 8 week old female BALB/c mice were treated for 3 days with 25 µL of DEE-solvent (dimethylacetamide 40%:diethyl ether 30%:ethanol 30%) on the dorsal side of both ears (negative controls). Mice serving as positive controls received croton oil as irritant or oxazolone (4-ethoxymethylene-2-phenyloxazolin-5-one, Sigma-Aldrich) as a sensitizing agent, respectively. In verum groups, diethyl ether in DEE was replaced by motor vehicle indoor air extract in ether. On day zero, before the first application of the test substances and on day 3, ear thickness was determined using a micrometer (B.C. Ames Co., Waltham, MA). Subsequently, mice were killed, and the ear-draining auricular lymph nodes were excised. Single cell suspensions of lymph node cells were stained with a CD45-FITC labeled antibody (Pharmingen Inc.) for 15 min at 4 °C, and the absolute numbers of CD45+ leukocytes were quantified by flow cytometry (FACS Calibur, Becton Dickinson, NJ) using TrueCount beads according to the manufacturers instructions (BD Biosciences). Lipopolysaccharide (LPS) was determined by the Limulus amebocyte lysate test according the manufacturers instructions (QCL-1000, Cambrex, Baltimore, MD). Statistical Analysis. Differences were analyzed by a paired Students’ t-test after testing for normal distribution with the Shapiro-Wilk test (22). p < 0.05 was considered statistically significant.
Results Sampling and Analysis of Parked Motor Vehicle Indoor Air. Of the new and the used motor vehicle 45.25 m3 and 45.44 m3 indoor air could be sampled in one period of 60.25 and 60.5 h, respectively. A total of 10.929 µg/m3 and 1.235
µg/m3 of volatile organic compounds (VOC) in the new and the used vehicle, respectively, could be determined directly from the inside of the vehicles. Air entering the car contained
