Environ. Sci. Technol. 2008, 42, 1761–1765
Cytotoxicity and Inflammatory Potential of Soot Particles of Low-Emission Diesel Engines D A N G S H E N G S U , * ,† ANNALUCIA SERAFINO,‡ JENS-OLIVER MÜLLER,† R O L F E . J E N T O F T , † R O B E R T S C H L Ö G L , * ,† AND SILVANA FIORITO‡ Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, D-14195 Berlin, Germany and Institute of Neurobiology and Molecular Medicine, National Research Council (CNR), Via Fosso del Cavaliere 100, 00133 Rome, Italy
Received July 5, 2007. Revised manuscript received November 27, 2007. Accepted December 3, 2007.
We evaluated, in vitro, the inflammatory and cytotoxic potential of soot particles from current low-emission (Euro IV) diesel engines toward human peripheral blood monocytederived macrophage cells. The result is surprising. At the same mass concentration, soot particles produced under lowemission conditions exhibit a much higher toxic and inflammatory potential than particles from an old diesel engine operating under black smoke conditions. This effect is assigned to the defective surface structure of Euro IV diesel soot, rendering it highly active. Our findings indicate that the reduction of soot emission in terms of mass does not automatically lead to a reduction of the toxic effects toward humans when the structure and functionality of the soot is changed, and thereby the biological accessibility and inflammatory potential of soot is increased.
1. Introduction Since the implementation of the 1970 Clean Air Act in the United States of America, progress has been made in the reduction of exhaust gas and soot emissions of light-duty and heavy-duty vehicles (passenger cars and trucks). Particulate standards for diesel engines were introduced in 1982 and were tightened in 1991, 1994, and 1998 (1). The European Union followed with emission standards for heavy-duty diesel engines in 1992 (Euro I), and in stiffer form in 1998 (Euro II), 2000 (Euro III), and in October 2005 (Euro IV) (1). All major automobile companies have developed low-emission engines as well as filters for soot particles. Research and development strategies have focused on the reduction of soot emission yet have neglected the question of how changes in soot quality may change its effect on human health. Hence, the question is: does the low-emission engine Euro IV soot pose the same health risk per unit mass as the soot produced from old engines? The cytotoxicity and inflammatory potential of soot nanoparticles (NPs) can be assessed by in vitro studies. Macrophages constitute the primary cellular effectors of the immune response, playing a pivotal role in the detection of * Address correspondence to either author. E-mail: dangsheng@ fhi-berlin.mpg.de (D.S.S.) and
[email protected] (R.S.). † Fritz Haber Institute of the Max Planck Society. ‡ Institute of Neurobiology and Molecular Medicine. 10.1021/es0716554 CCC: $40.75
Published on Web 01/25/2008
2008 American Chemical Society
all foreign bodies. These cells are ubiquitously present in the mucosal and submucosal tissues (especially in the bronchial and alveolar membrane), and human macrophage primary cultures in vitro can provide a model of potential effects upon in vivo inhalation of the soot NPs. When these cells come in contact with particles or pathogens, they become activated and secrete a variety of chemical mediators of inflammation, very aggressive against foreign molecules or particles. Currently, the toxicity of NPs is a hot research topic because the increasing production of nanomaterials is likely to significantly enhance the exposure of humans to NPs (2–4). However, the research in the field of nanotoxicology is still at its infancy. The parameters that determine the toxicity of NPs are not known in any detail, as one can tell from the large number of review articles published recently on the topic (5). The parameter most frequently used as a measure of dose is the surface area. However, lung inflammation studies involving instillation of different types of carbon NPs in mice have revealed a much more complex situation: particles prepared by different techniques exhibit significant differences in surface toxicity (5). The purpose of this study was to compare the cytotoxicity and the inflammatory response, in vitro, of human monocytederived macrophage cells (MDMs) to a Euro IV test heavyduty diesel engine soot and to soot from an old diesel engine and to relate the results to the microstructure of these particles, previously determined in detail by means of highresolution transmission electron microscopy and other methods of NP characterization.
2. Experimental Section In the following, the soot from a Euro IV test heavy-duty diesel engine will be referred to as EuroIV soot; the soot from an old diesel engine operating at black smoke conditions will be referred to as BS soot. The methods of soot production and collection have been described elsewhere (6). Briefly, the EuroIV soot originated from a modified MAN D0836 LF4V six cylinder engine (6.9 L displacement, 228 kW), with two-stage controlled turbocharging, an externally controlled cooled exhaust gas recirculation, and a common rail injection system. The engine was developed to fulfill the Euro IV emission standard. The engine was set for a NOx emission of 3.3 g/kWh and a PM emission of 50 mg/kWh (European stationary cycle, ESC). The BS soot originated from a D2876 CR engine, operated at 30% load, extra-low rail pressure, and air throttling (blackening number 5). The emission rate of the BS engine is 200–600 mg/kWh. The diesel fuel used for both engines was a standard low-sulfur type, containing 78% paraffin and 22% aromatic hydrocarbons (European Norm 590). All samples were collected directly from the exhaust gas of the engine using a special particle collector that was heated to the exhaust gas temperature at the collection position (200 °C). Transmission electron microscopy, energy-dispersive X-ray spectroscopy, and temperature programmed oxidation studies revealed that EuroIV soot contained about 10% ash from the combusted engine lubricant oil (7). This kind of ash was not found in BS soot. For the in vitro studies, the EuroIV and BS soot was sterilized by heating to 180 °C, washed three times in distilled water, then suspended in PBS at a stock concentration of 1 mg/mL and sonicated for 48 h before the use. Human peripheral blood monocytes were isolated from buffy coats of healthy donors by density gradient centrifugation using lympholyte-H (Cederlane, Hornby, Ontario, Canada). The lymphocytic/monocytic fraction was then VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1761
resuspended in RPMI 1640 medium (Hyclone Laboratories Inc. Logan, Utah) supplemented with 10% (v/v) heatinactivated fetal calf serum (FCS) (Hyclone), L-glutamine (2 mM), penicillin (100 IU/mL), and streptomycin (100 mg/ mL). Cells were seeded on 175 cm2 flasks and were maintained at 37 °C in 5% CO2 to generate adhering macrophages (MDMs). After 1 h of culture, nonadhering cells were removed, and the residual adhering MDMs were maintained in culture for 7 days to obtain partially differentiated macrophages. For in vitro testing, soot particles were added to the MDM culture medium to obtain the desired treatment concentrations. In a preliminary dose–response experiment, performed using the Trypan blue dye exclusion method (8), cells were treated with soot particles at concentrations ranging from 15 to 60 µg/mL. The concentration of 30 µg/mL was found to be at the upper end of the linear dose–response curve (see below) and was selected for all further tests in which the cytotoxicity of the soot particles on MDM cultures was evaluated: (1) by observing the nuclear morphology under a confocal laser scanning microscope (CLSM) and counting the number of apoptotic/necrotic cells stained with propidium iodide (Sigma-Aldrich Co., St. Louis, MO), and (2) by a two-color fluorescence cell viability assay that distinguishes metabolically active cells from injured and dead cells (Live/ Dead Cell Vitality Assay, Molecular Probes, Eugene, OR). The analysis of particles uptake was carried out by CLSM on cells fixed with paraformaldehyde and counter-stained with 1 µg/ mL propidium iodide (PI). All optical observations were carried out using the confocal microscope LEICA TCS SP5 (Leica Instruments, Heidelberg, Germany). The excitation/emission wavelengths employed were 568/590 nm for PI labeling, respectively, whereas the soot particles were visualized by recording the reflected intensity of the laser beam. The cell morphology was visualized by differential interference contrast (DIC). The merged images of the three signals (PI/Refl/DIC) were recorded. To distinguish signals stemming from particles internalized or adhering on the surface of the cells, vertical sections xzy (xz-planes along y-axes) were also acquired for each sample in addition to the horizontal confocal sections xyz (xy-planes along z-axes, see the definition in the schematic diagram in Figure 2c). Morphological changes in MDM culture, indicative of macrophage activation, were examined by scanning electron microscopy (SEM). For SEM observation, MDMs were fixed with 2.5% glutharaldehyde in 0.1 M Millonig’s phosphate buffer (MPB) at 4 °C for 1 h. After washing in MPB, cells were postfixed with 1% OsO4 in the same buffer for another 1 h at 4 °C and then dehydrated using acetone with increasing concentrations. The specimens were then critical-point dried using liquid CO2 and were sputter-coated with gold before examination on a Stereoscan 240 scanning electron microscope (Cambridge Instruments, Cambridge, United Kingdom). The amount of the proinflammatory (IL-1β and IL-6) and anti-inflammatory cytokines (IL-10) secreted by MDMs challenged with the soot particles was assessed by an ELISA testing (TEMA Ricerca) in the supernatants of the cell cultures. Student’s t-test was used for statistical analysis. For each variable, at least three independent experiments were carried out.
3. Results The morphological changes induced by soot NP treatment of human MDM cultures are illustrated in Figure 1. A comparison of representative images of untreated and soot particle-treated MDM cultures obtained by phase contrast and fluorescence microscopy is given in Figure 1a. The images show that both EuroIV and BS soot particles were able to induce macrophage activation, an early phase of the inflammatory reaction. This is revealed by the presence of more developed microvillous structures on the cell surface and by 1762
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 5, 2008
FIGURE 1. Comparison of representative images of untreated (control, Ctr) and particle-treated MDM cultures (30 µg/ml BS or EuroIV soot). (a) Phase contrast and fluorescence microscopy after PI staining of nuclei (red hue). A higher number of apoptotic/necrotic cells (black arrows) is seen to be present in EuroIV soot treated cultures as compared to BS treated cells. Panels (b), (c), and (d) show scanning electron microscopy views of cells. The examples in panel (c) show, at high magnification, that BS and EuroIV soot treatment can induce pronounced morphological changes indicative of macrophage activation. The images in panel (d) show, also at high magnification, a necrotic cell (left) and an apoptotic cell (right); the open arrow points to an apoptotic body detaching from the cell surface. Scale bars: 20 µm. the size of the activated cells, which are typically two times larger than the nonactivated ones (Figure 1, panels b and d). While BS soot particles did not induce significant signs of necrosis or apoptosis, EuroIV soot particles produced extensive damage of cells, as revealed by the appearance of numerous apoptotic and necrotic cells (Figure 1, panels b and d). The analysis of particles uptake by confocal microscopy clearly showed that EuroIV soot particles were more uniformly distributed than the BS soot particles, which were found to aggregate into big clusters (Figure 2, panels a and b). The images also suggest that the EuroIV soot particles were internalized by MDMs in a much larger number than the bigger aggregates of BS NPs. The internalization of a higher amount of EuroIV soot particles may produce more cytotoxic effects in MDMs and may stimulate a more intensive inflammatory reaction as compared to BS soot. Figure 3 shows the results of the dose–response experiment for human MDMs treated with increasing concentrations of EuroIV and BS soot particles, assessed by Trypan blue dye exclusion methods. Linear responses were obtained for both soot particles up to a dose rate of 30 µg/mL, where higher doses result in the saturation of the response, especially for BS soot particles. However, in all doses tested, EuroIV soot induces a higher percentage of dead cells than the BS soot. Figure 4a displays the evaluation of dead cells by live/ dead cell vitality assay and of apoptotic cells in untreated (Ctr) and particles-treated (30 µg/mL BS or EuroIV soot) MDMs. After 24 h of treatment, the EuroIV soot particles induced a significantly higher percentage (p < 0.001) of apoptotic and necrotic cells as compared to the particles from the BS soot. Moreover, EuroIV soot particles were able to stimulate human MDMs to secrete the pro-inflammatory
FIGURE 2. Confocal microscopy images showing the uptake of (a) BS and (b) EuroIV soot particles by human MDMs. The presence of soot particles is visualized by differential interference contrast (DIC), as well as by the intensity of the reflected laser beam (Refl). The cell cytoplasm and nucleus are visualized by the fluorescence signal of PI staining; merged images of the three signals (PI/Refl/ DIC) are also shown. For each cell, the horizontal confocal section along the xyz-axes and the vertical sections along the xzy-axes, obtained as schematized in panel (c), are also reported.
FIGURE 3. Dose–response data testing the cytotoxicity of BS and EuroIV particles on human MDMs, assessed by Trypan blue dye exclusion methods. The solid lines reflect a linear dose response. cytokines IL-1β and IL-6, (Figure 4a), whereas the BS soot particles from the old diesel engine did not induce a significant secretion of these pro-inflammatory cytokines (Figure 4b). At present we cannot explain the reason why the BS soot particles seem to inhibit the secretion of proinflammatory cytokines. This topic will be the aim of future studies. We suspect that the difference in the inflammatory potential of soot particles is due to their pronounced dissimilarity in microstructure and reactivity. These aspects have previously been studied in detail by high-resolution TEM (HRTEM) and electron energy loss spectrometry (7, 9, 10). Briefly, the following results were obtained. The mean size of EuroIV soot particles (small nuclei, 10–15 nm;
spherical particles, 18 nm) is much smaller than that of BS soot (35 nm). BS soot (Figure 5a) reveals the common morphology of a spherical secondary structure made from homogeneously sized flat basic structural units (11). EuroIV soot particles exhibit distinctly rough surfaces and strongly bent graphene sheets, many of them having irregular forms (Figure 5b). HRTEM images reveal that the surface of EuroIV soot is often decorated by abundant, very small soot structures (Figure 5c). A recent study by means of quantitative HRTEM analysis (12) has shown that EuroIV soot exhibits a high degree of distortion due to defects such as non-sixmembered rings in the graphitic network. This reactive structure can be expected to have significant consequences for its function in the environment and in technical processes. The olefinic electronic structure and the excess presence of chemically reactive edges tend to destabilize the observed carbon structures, as revealed by temperature-programmed oxidation studies (7, 9). For instance, the combustion rates, an indicator for the chemical reactivity and surface functionalization, are quite different for BS and EuroIV soot. In an oxidative atmosphere (5% O2 in N2), the onset temperature and the temperature of maximum reaction rate are 200 and 70 °C, respectively, lower for EuroIV than for BS soot (7, 9). The large abundance of reactive structural elements in the EuroIV soot facilitates the anchoring of heteroatoms such as hydroxyl groups. X-ray photoelectron spectroscopic measurements (12) revealed that the surface concentration of oxygen in EuroIV soot is as high as 12%, significantly higher than that for BS soot (7.4%). This is in accordance with the results of infrared studies showing that EuroIV soot contains a high concentration of OH groups (7). Such soot particles VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1763
FIGURE 4. (a) Evaluation of dead cells by live/dead cell vitality assay and of apoptotic MDM cells (30 µg/mL BS or EuroIV soot). (b) Effect of BS and EuroIV soot particle treatment on IL-1beta and IL-6 proinflammatory cytokines production by MDMs (30 µg/mL).
4. Discussion
FIGURE 5. High-resolution TEM images of (a) BS soot showing almost spherical soot particles, (b) EuroIV soot with core– shelled primary particles showing defective bulk and surface structure, and (c) an expanded section of the image of an EuroIV soot showing small soot structures at the periphery. are hydrophilic and may disperse in aqueous media, whereas particles of industrial carbon black are largely hydrophobic and thus aggregate in water. 1764
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 5, 2008
Our experiments show that EuroIV soot particles are more cytotoxic and have a higher inflammatory potential than soot particles from an old diesel engine. Various parameters such as surface area, number of particles, and joint length have been examined to interpret or quantify the lung inflammatory response to nanoparticle exposure (5). The results presented in this work may suggest that the particle size is a proper dose metric for nanotoxicity because the mean particle size of EuroIV soot is much smaller than that of BS soot. However, we strengthen that the reported effects are more likely due to the strongly defective structure, the high abundance of chemically reactive structural elements (edges), and the presence of surface functional groups on the EuroIV soot particles that enabled them to be phagocytized more readily by MDMs than the larger BS particles. The expected hydrophilic surface chemistry due to attached OH groups should also be considered. These properties allow facile chemical and morphological contact with hydrophilic biomolecules. The OH and olefinic surface chemistry is a novel property of the EuroIV soot as opposed to the BS soot with its smooth, inert surface and its hydrophobic character. All theses aspects need to be taken into account in future health risk assessment. The present results are generally in accordance with the results of recent studies involving C60-fullerenes that were found to exhibit a rather low acute toxicity against human and animal cells in vitro and animal tissues in vivo (13, 14). However, after surface derivatization or functionalization they became cytotoxic (15). Low-emission diesel engines emit a comparatively small amount of soot particulate matter. Our study, however, has shown that, by mass, soot nanoparticles produced under low-emission conditions have a higher cytotoxic and inflammatory potential against human peripheral blood monocytederived macrophage cells than particles from an old diesel engine. The macrophages exposed to soot particles of lowemission engines showed characteristic features of necrosis and degeneration. A high apoptotic cell death rate was observed. This effect is assigned to the functionalized defective surface structure of the low-emission diesel engine soot, rendering it highly active. Moreover, the particles size of EuroIV soot is smaller than that of BS soot that tends to aggregate in bigger clusters. This makes the internalization of a higher amount of EuroIV particles possible, leading to more cytotoxic effects and stimulating a more intensive inflammatory reaction, as compared to BS soot. Our findings imply that a reduction of the emission rate of soot particulates does not automatically lead to a reduction of the toxic effects toward humans if, concurrently, the structure and functionality of the soot changes and therefore the biological accessibility and inflammatory potential of
the soot increases. Fortunately, the microstructural features that aggravate the health risk also lead to a more effective oxidation of soot particles to CO2, provided suitable filtering techniques are applied (16). Hence, the development of filtering technology must be directed toward the removal of ultrasmall particles that, per unit mass, pose a higher risk to the biosphere than the more conventional forms of largeparticle soot.
Acknowledgments This work was part of the project “Katalytisches System zur filterlosen kontinuierlichen Rußpartikelverminderung für Fahrzeugdieselmotoren” supported by the Bayerische Forschungsstiftung, Munich. We are indebted to E. Jacob and D. Rothe, Nürnberg, for access to the motor test equipment and for helpful discussions. We acknowledge multiple discussions with T. Velden. We also acknowledge F. Andreola for the technical assistance in preparing cell cultures and biological tests. We are very grateful to the anonymous reviewer for helpful contributions to the manuscript and the data presentation.
Literature Cited (1) Emission Standards: Summary of worldwide diesel emission standards. http://www.dieselnet.com/standards. (2) Fiorito, S.; Serafino, A.; Andreola, F.; Bernier, P. Effects of fullerenes and single-wall carbon nanotubes on murine and human macrophages. Carbon 2006, 44, 1100–1105. (3) Fiorito, S.; Serafino, A.; Andreola, F.; Togna, A.; Togna, G. Toxicity and Biocompatibility of Carbon Nanoparticles. J. Nanosci. Nanotechnol. 2006, 6, 591. (4) Duffin, R.; Tran, R.; Brown, D.; Stone, V.; Donaldson, K. Proinflammogenic Effects of Low-Toxicity and Metal Nanoparticles in vivo and in vitro: highlingting the role of particles surface area and surface reactivity. Inhal. Toxicol. 2007, 19, 849. (5) Wittmaack, K. Search of the Most Relevant Parameter for Quantifying Lung Inflammatory Response to Nanoparticle Exposure: Particle Number, Surface Area or What. Environ. Health Perspect. 2007, 115, 187.
(6) Jacob, E.; Rothe, D.; Schlögl, R.; Su, D. S.; Müller, J.-O.; Niessner, R. C.; Adelhelm, A.; Messerer, A.; Pöschl, U.; Müllen, K.; Simpson, C. D.; Tomovic, Z. Dieselruss: Mikrostruktur und Oxidationskinetik. In 24. Internationales Wiener Motorensymposium, 15.– 16. Mai 2003, Band 2: Fortschritt-Berichte VDI Reihe 12 Nr. 539; Lenz, H. P. (Hrsg.); VDI-Verlag: Düsseldorf, 2003; pp 19–45. (7) Müller, J.-O.; Su, D. S.; Jentoftzprint, R. E.; Kröhnert, J.; Jentoft, F. C.; Schlögl, R. Morphology Controlled Reactivity of Carbonaceous Materials towards Oxidation. Catal. Today 2005, 102– 103, 259. (8) Detrick-Hooks, B.; Borsos, T.; Rapp, H. J. Quantitative Comparison of Techniques Used to Measure Complement-mediated Cytotoxicity of Nucleated Cells. J. Immunol. 1975, 114, 287. (9) Müller, J.-O.; Su, D. S.; Jentof, R. E.; Wild, U.; Schlögl, R. Diesel Exhaust Emission: Oxidative Behaviour and Microstructure of Black Smoke Soot Particulates. Environ. Sci. Technol. 2006, 40, 1231. (10) Müller, J.-O.; Su, D. S.; Wild, U.; Schlögl, R. Bulk and Surface Structural Investigations of Diesel Engine Soot and Carbon Black. Phys. Chem. Chem. Phys. 2007, 9, 4018. (11) Oberlin, A. High-Resolution TEM Studies of Carbonization and Graphitization. In Chemistry and Physics of Carbon. Thrower, P. Ed.; Dekker: New York, 1989; p 22. (12) Baierl, T.; Drosselmeyer, E.; Seidel, A.; Hippeli, S. The differential cytotoxicity of watersoluble fullerenes. Exp. Toxicol. Pathol. 1996, 48, 508. (13) Yamago, S.; Tokuyama, H.; Nakamura, E.; Kikuchi, K.; Kananishi, S.; Sucki, K.; Nakahara, H.; Enomoto, S.; Ambe, F. In vivo biological behavior of a water-miscible fullerene:14C labeling, absorption, distribution, excretion and acute toxicity. Chem. Biol. 1995, 2, 385. (14) Rancan, F.; Rosan, S.; Boehm, F.; Cantrell, A.; Brellreich, M.; Hirsch, A.; Moussa, F. Cytotoxicity and photocytotoxicity of a dendritic C(60) mono-adduct and a malonic acid C(60) trisadduct on Jurkat cells. J. Photochem. Photobiol. B 2002, 67, 157. (15) Sayes, C. M.; Gobin, A. M.; Ausman, K. D.; Mendez, J.; West, J. L.; Colvin, V. L. Nano-C60 cytotoxicity is due to lipid peroxidation. Biomaterials 2005, 26, 7587. (16) Jacob, E.; D’Alfonso, N.; Döring, A.; Reisch, S.; Rothe, D.; Brück,R.; Treiber, P. PM-KAT: Nichtblockierende Lösung zur Minderung von Dieselruß für EuroIV Nutzfahrzeugmotoren. In 23. Internationales Wiener Motorensymposium, 25.–26. April 2002, Band 2: Fortschritt-Berichte VDI Reihe 12 Nr. 490; Lenz, H.P.(Hrsg.); VDI-Verlag: Düsseldorf, 2002; pp 196–216.
ES0716554
VOL. 42, NO. 5, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1765
