TOXIC EFFECTS OF URBAN AIR AND DIESEL EXHAUST PARTICLES ON THE RESPIRATORY TRACT (PAMTOX)

Project leader: Raimo O. Salonen, National Public Health Institute (KTL), Department of Environmental Health, P.O.Box 95, FIN-70701 Kuopio, Finland, tel. +358-17-201 348, e-mail: Raimo.Salonen@ktl.fi
 
 

PUBLICATIONS

TIIVISTELMÄ SUOMEKSI

Researchers:
Arja Hälinen, National Public Health Institute (KTL), P.O.Box 95, 70701 Kuopio, tel. +358-17-201355, e-mail: Arja.Halinen@ktl.fi
Arto Pennanen, National Public Health Institute (KTL), P.O.Box 95, 70701 Kuopio, tel. +358-17-201355, e-mail: Arto.Pennanen@ktl.fi
Markus Sillanpää, Finnish Meteorological Institute (FMI), Sahaajankatu 20 E, 00810 Helsinki,  tel. +358-9-19295404, e-mail: Markus.Sillanpaa@fmi.fi

Consortium: Urban air particles and environmental health
Financing SYTTY-organisation:  The Academy of Finland
Funding from SYTTY / Total funding of project (€): 197385 / 437871
Person-months of work funded by SYTTY / Total person-months of work: 76 / 120,5

KEY WORDS: Air pollution, ambient air particles, chemical characterisation, cell toxicology,
biomarkers
 

EXTENDED ABSTRACT

1  Introduction

Recent epidemiological studies and health effect assessments have indicated that the most severe adverse health effects of ambient air pollution (increased daily deaths, hospital admissions of cardiorespiratory patients, shortened life expectancy) are consistently associated with inhalable particles (PM10; 50% size cut-off at 10 µm), and possibly even more strongly with fine particles (PM2.5; 50% size cut-off at 2.5 µm) including also diesel exhaust particles (0.02-0.5 µm). Moreover, there have been independent associations of adverse outcomes with ultrafine (PM0.1) particles. The chemical characteristics of these PM fractions and biological mechanisms responsible for the adverse health effects are largely unknown.

The ambient air PM pollution in Finnish cities has large contrasts in different seasons. In mid-winter, the 24-hour concentrations of PM2.5 and PM10 are usually low and the PM originates mainly from regional + long-distance transport and local combustion sources (especially traffic), whereas in springtime, the 24-hour concentrations of PM10 are relatively high and a large proportion of PM originates from resuspension of road dust (sand, asphalt, tyre and stud dust etc.). In recent Finnish epidemiological studies conducted in Kuopio and Helsinki, the short-term concentration variations in all above mentioned PM size fractions (most consistently PM2.5) have been associated with changes in cardiorespiratory functions among susceptible population groups (asthmatic children and adults, elderly subjects with ischemic heart disease).

The main objective of the PAMTOX project has been to introduce a new systematic research strategy for identification of causative constituents in the adverse respiratory cell / tissue responses (proinflammatory activity, cytotoxicity, genotoxicity) to ambient air PM10 and its subfractions. The strategy is based on the introduction of new high-volume sampling techniques that enable a collection of large enough samples of these PM fractions (from tens to hundreds of milligrams per week) to allow for an extensive physicochemical and toxicological characterisation from the same short-term PM samples. The responses to and mechanisms of urban air and diesel exhaust PM have been investigated in the standard murine macrophage cell line RAW 264.7 and the lower respiratory tract of rats. A special emphasis has been made to analyse in vitro the causative role of watersoluble and insoluble PM constituents in the toxic responses.

2  Methods

Urban air particles

Development and calibration of high-volume PM sampling methods: The development and laboratory + field calibration of the new high-volume sampling techniques in association with the PAMTOX project have been made stepwise in close collaboration with the Harvard School of Public Health (Boston, USA) and the Finnish Meteorological Institute (FMI; Helsinki) in 1997-2001: 1) Establishment of a PM sampling station in connection to the Vallila ambient air quality monitoring station of the Helsinki Metropolitan Area Council (YTV) and installation of the very first single-stage, high-volume (1100 L/min) low-cutoff impactor (HVLI) system for large-capacity sampling of ambient air PM10 in January 1999 (Salonen et al., 2000); 2) Modification of HVLI for single-stage large-capacity sampling of ambient air PM2.5 in January 2000; 3) Installation of high-volume (900 L/min) cascade impactor (HVCI) for simultaneous large-capacity samplings of the coarse (PM10-2.5), fine (PM2.5-0.1) and ultrafine (PM0.1) particulate fractions in December 2000; and 4) Final modification and laboratory + field calibration of HVCI in July-December 2001 (Sillanpää et al., 2002, submitted).

High-volume PM sampling campaigns in Helsinki: The ambient air PM10 and PM2.5 samples were collected at 68 m3/h with HVLI during 4-month field campaigns between January and May in 1999 and 2000, respectively. The samplings were usually made in 3 + 4-day periods per week at the Vallila monitoring station located close to downtown Helsinki, at a distance of 15 meters from a busy road (about 13000 vehicles/day). The collected HVLI-PM samples were extracted from the polyurethane foam sampling substrate with 100% methanol that was subsequently evaporated in vacuum at room temperature. The extracted PM10 and PM2.5 masses from individual samples and periods were pooled together into at least two categories per PM fraction formed on the basis of the season and ambient air PM2.5:PM10 concentration ratio (continuous air quality and meteorological data obtained from YTV). The larger pooled PM10 or PM2.5 samples represented different types of ambient air PM pollution situations as follows:

86 Winter-PMs during high PM2.5:PM10 ratio (low resuspension)
87 Spring-PMs during low PM2.5:PM10 ratio (high resuspension)

The pooled HVLI-PM samples were analysed chemically and screened toxicologically in several test systems (Salonen et al. 2002a and 2002b, submitted). In both studies, watersoluble ions were analysed with ion chromatography (IC) and watersoluble elements with inductively coupled plasma mass spectrometry (ICP-MS) at FMI. The genotoxic polycyclic aromatic hydrocarbons (PAHs) were analysed with gas-chromatograph mass-spectrometer single ion monitoring technique (GCMS-SIM) at VTT Chemical Technology (Espoo, FIN). In the latter study, additional low-volume PM2.5 samples were collected to allow for a comparison between the total (ED-XRF; University of Antwerp, B) and watersoluble (ICP-MS) elemental compositions and between the standard low-volume and new HVLI samplings. The HVLI-PM10 and HVLI-PM2.5 samples (30 - 2000 µg per 106 cells) were screened in vitro for production of proinflammatory cytokines (TNF-alpha, IL-6) and nitric oxide (NO), and cytotoxicity (MTT test for functioning mitochondria), in the murine macrophage cell line RAW 264.7 at KTL (24-hour incubation), and for acellular hydroxyl radical production (electromagnetic spin resonance) and DNA damage (8-OH-2-deoxyguanosine) at Institut für umweltmedizinische Forschung (IUMF) an der Heinrich-Heine-Universität (Düsseldorf, D).

Diesel particles

The in-vitro effects of commercially available standard reference (National Institute of Standards & Technology - NIST) diesel particles (DP; 30 - 3000 µg per 106 cells) were tested for cell viability (MTT test) and production of NO, interleukin-1 (IL-1), interleukin-6 (IL-6), interferon-gamma (IFN-gamma) and tumor necrosis factor-alpha (TNF-alpha) by exposing cultured murine RAW 267.4 macrophages for 8-48 hours. In vivo, anaesthetised male rats of Han:Wistar strain (200-300 g) were exposed intratracheally to the same DPs (1 mg or 5 mg/rat) followed by bronchoalveolar lavage (BAL) 24-72 hours after exposure. The BAL fluid was collected and stored on ice until ex-vivo incubation of recovered cells and measurement of supernatant protein concentration (Hälinen et al., 1999).

Other studies

An animal model has been validated for quantitative assessment of bronchial hyperresponsiveness to important co-existing airborne factors (cold air in winter, NO2 in traffic environments and SO2 in industrial environments) in future studies on ambient air PM10 (Hälinen et al., 2000a and 2000b).

3  Results and discussion

The HVLI blanks, prepared and extracted from the collection substrate (polyurethane foam) in the same way as ordinary PM samples after ambient air collection, gave low values in all chemical analyses and toxicological tests.

Chemical characteristics  of PM10 and PM2.5

There were no major differences between the pooled HVLI Winter-PM10 and Spring-PM10 samples (collected in 1999) with regard to their watersoluble ionic and elemental contents, but individual samples, representing an extreme difference in the PM2.5:PM10 ratio, showed higher watersoluble soil metal contents (Al, Fe, Ca) and lower sulphate content  in spring (Salonen et al., 2000). The pooled HVLI-PM2.5 samples (collected in 2000) showed clearly higher watersoluble soil metal contents and lower anionic contents (SO42-, NO3-) in spring compared to winter. In the ED-XRF and ICP-MS analyses of the reference low-volume PM2.5 samples, the total and watersoluble fraction of soil metals were higher in spring. The PAH contents of the pooled HVLI-PM10 and HVLI-PM2.5 samples were higher in winter.

In-vitro toxicity profiles of PM10 and PM2.5

The pooled HVLI-PM10 samples collected in 1999, and HVLI-PM2.5 samples collected in 2000, showed rather similar in-vitro toxicity profiles and seasonal differences in their responses. The PM samples from both winter and spring induced dose-dependent NO production in murine RAW 264.7 macrophages without major seasonal difference in potency. Winter-PMs were significantly less potent inducers of TNF-alpha production compared to Spring-PMs. Moreover, Winter-PMs caused practically no IL-6 production, whereas Spring-PMs produced partially dose-dependent responses. Winter-PMs and Spring-PMs induced dose-dependent reductions in cell viability without major seasonal difference. With regard to cell viability, and especially cytokine productions, the insoluble fractions of Winter-PMs and Spring-PMs seemed to be responsible for nearly the whole responses. Polymyxin B (antagonist of endotoxin) abolished the IL-6 production induced by Spring-PMs and significantly reduced the TNF-alpha productions induced by both Winter-PMs and Spring-PMs. Deferoxamine (somewhat cytotoxic iron chelator) did not modify these responses. Neither polymyxin B nor deferoxamine modified the PM-induced reductions in cell viability. There were dose-dependent acellular hydroxyl radical production and DNA damage by both HVLI-PM10 and HVLI-PM2.5 without major seasonal difference in potency.

Diesel particles

In 24 and 48-hour incubations, DP caused a dose-dependent NO-production. Cytotoxicity preceded the increases in NO production, as DP 30 µg/106 cells decreased cell viability by 30% in an 8-hour incubation. A single intratracheal instillation of DP 1 mg/rat and 5 mg/rat increased NO production and protein leakage in the lungs at 48-72 hours. This was associated with pulmonary oedema and hemorrhage already at 24 hours. Thus, the DP-induced in-vitro cytotoxicity agreed with the effects on the whole respiratory organ system in vivo.

Implications

The new information about potentially hazardous ambient air PM fractions complements the current exposure and epidemiological data from the population in Helsinki. Thus, it can be used as supportive information from subarctic environments in the health risk assessments of ambient air PM pollution that are going to be conducted in near future by WHO and EC. The PAMTOX study design, HVLI and HVCI calibrations and international collaboration form the basis for a new EU / Quality of Life RTD-project on ‘Chemical and biological characterisation of ambient air coarse, fine, and ultrafine particles for human health risk assessment in Europe’ (PAMCHAR; QLK4-CT-2001-00423) coordinated by KTL in 2002-2004. Since the introduction of the first generation high-volume sampler (HVLI) to PM toxicology by the PAMTOX team and HSPH, several other research institutions in Europe and USA have chosen the same methodological approach (HVLI, HVCI) for their ambient air PM studies.

4  Conclusions

1. The HVLI proved to be a suitable technique for single-stage, large-capacity collection of ambient air PM fractions enabling rather extensive chemical characterisation and in-vitro toxicity testing from the same samples. The laboratory and field calibration of HVCI enables simultaneous large-capacity samplings of several PM10 subfractions. However, additional size-fractionated, low-volume samplings are needed for mass-balance assessment of the chemical constituents in PM10 subfractions (Sillanpää et al., 2000).
2. Winter vs. spring differences in ambient air PM pollution in Helsinki were reflected, not only in PM10, but there were major differences also in the inorganic and organic chemical compositions of its subfraction PM2.5.
3. There was a clear-cut difference between winter and spring in the in-vitro proinflammatory activities of both PM10 and PM2.5. The proinflammatory and cytotoxic activities of these PM fractions during springtime episodes of resuspended road dust were largely mediated via insoluble chemical constituents (possibly organic and/or silica), and the proinflammatory activities were partially mediated via bacterial endotoxin. The in-vivo relevance of the present findings needs to be investigated in future studies.
 
 

[ Projects | Main Page ]