Abstract
Epidemiological studies have demonstrated an association between different levels of air pollution and various health outcomes including mortality, exacerbation of asthma, chronic bronchitis, respiratory tract infections, ischaemic heart disease and stroke. Of the motor vehicle generated air pollutants, diesel exhaust particles account for a highly significant percentage of the particles emitted in many towns and cities. This review is therefore focused on the health effects of diesel exhaust, and especially the particular matter components.
Acute effects of diesel exhaust exposure include irritation of the nose and eyes, lung function changes, respiratory changes, headache, fatigue and nausea. Chronic exposures are associated with cough, sputum production and lung function decrements. In addition to symptoms, exposure studies in healthy humans have documented a number of profound inflammatory changes in the airways, notably, before changes in pulmonary function can be detected. It is likely that such effects may be even more detrimental in asthmatics and other subjects with compromised pulmonary function.
There are also observations supporting the hypothesis that diesel exhaust is one important factor contributing to the allergy pandemic. For example, in many experimental systems, diesel exhaust particles can be shown to act as adjuvants to allergen and hence increase the sensitization response.
Much of the research on adverse effects of diesel exhaust, both in vivo and in vitro, has however been conducted in animals. Questions remain concerning the relevance of exposure levels and whether findings in such models can be extrapolated into humans. It is therefore imperative to further assess acute and chronic effects of diesel exhaust in mechanistic studies with careful consideration of exposure levels. Whenever possible and ethically justified, studies should be carried out in humans.
The authors were supported by the Swedish Heart and Lung Foundation, The Medical Research Council (project No 9071), the Swedish National Road Administration (Vägverket) and The Foundation for Health Care and Allergy Research (Vårdal).
There is growing international concern regarding the adverse health effects of air pollution. Pollution is becoming an important public health problem and political issue, due to the rapid growth in world population and the increasing world-wide migration from rural to urban areas 1. Recent United Nations estimates have indicated that 47% of the global population is living in urban areas. This urbanization has brought with it an increased need for transportation and hence an increase in motor vehicle generated air pollutants. However, a large number of epidemiological studies from different parts of the world have consistently identified an association between ambient levels of air particles and various health outcomes, including mortality, exacerbation of asthma, chronic bronchitis, respiratory tract infections, ischaemic heart disease and stroke 2. Indeed, the United Nations Environment Programme has identified particulate matter pollution as the most serious air pollution problem faced by many cities 3. It is therefore important to evaluate the health effects of motor vehicle generated pollutants in mechanistic studies, in order to understand the relevance of associations found in the epidemiological studies. This overview is based on a literature search including papers published up to the first quarter of the year 2000. Much of the research on the adverse effects of diesel exhaust, both in vivo and in vitro, has been conducted in animals. Such experimental studies are reviewed critically and the findings are compared with those in human studies.
Diesel emission
Of the motor vehicle generated air pollutants, diesel exhaust particles (DEPs) account for a highly significant percentage of the particles emitted in many towns and cities 3, 4. Complete combustion of diesel fuel produces water and carbon dioxide, but use of diesel in motor vehicles normally results in incomplete combustion and the formation of various gases, liquids and solid particles. Compared with petrol engines, diesel engines produce far less carbon monoxide, but give rise to a greater amount of nitrogen oxides and aldehydes, which are particularly prone to cause irritation of the upper respiratory tract. Diesel engines also produce submicron soot particles that are believed to mediate several of the observed adverse effects. It has been estimated that the particulate emission from diesel engines per travelled distance is over 10 times higher than the emission from petrol engines of equivalent power running on unleaded petrol, and over 100 times higher than that from petrol engines fitted with catalytic converters 5, 6.
The dose of particle deposited in the lung depends on their concentration in the inhaled air and their size. Particles with a diameter <5 μm 7, 8 reach the alveoli and are deposited there, whereas particles larger than 5 μm only reach the proximal airways and are eliminated by mucociliary clearance. Previous human studies using radioactive particles have demonstrated that 83% of particles with a mass median diameter of 2.5 μm are deposited in the lung, whereas only 31% of particles of 11.5 μm are deposited 9. Recent electron microscopy studies suggest that over 80% of DEPs have a size ≤0.1 μm; DEPs consist of a carbonaceous core similar to carbon black, onto which an estimated 18,000 different high-molecular-weight organic compounds are adsorbed 2. Diesel exhaust (DE), in addition to DEPs, contains a complex mixture of gases such as carbon monoxide (CO), nitric oxides (NO, NO2), sulphur dioxide (SO2), hydrocarbons, formaldehyde, transition metals and carbon particles 10. Regarding the health effects, recent focus has been on ultrafine particles (diameter <0.05–0.10 μm). They are highly reactive and are present in large numbers in the urban environment. They are able to penetrate the epithelium and vascular walls and enter the bloodstream. It has been hypothesized that ultrafine particles account for the systemic effects of DEPs, such as increased carcinogenicity 11, potentiation of autoimmune disorders 12, alterations in blood coagulability and increased cardiovascular disorders 13, 14.
Air pollution as a cause of increased asthma and allergy?
Parallel to the increase in air pollution, there has also been a rapid increase in the global incidence of allergic diseases such as asthma and rhinitis in the last two decades, which cannot be attributed to genetic changes, and is assumed to be related to changes in environmental factors 15. Observations in Japan have suggested that children living close to roads with heavy traffic are more likely to develop allergies 15. Recent epidemiological data support the theory that atopic children may constitute a group of individuals that run a heightened risk of developing negative health effects following exposure to airborne particles 16. The existence of such sensitive subgroups amongst the general population would seem to deserve particular attention for risk assessment. This review has identified several effects of DEPs on immunological or inflammatory systems that may potentially have particular relevance for a role of chronic diesel exhaust exposure in the pandemic of allergic disease.
Irrespective of whether or not vehicle generated pollutants contribute to the increased numbers of sensitized individuals in urban areas, patients with airway diseases such as asthma have been found more adversely affected than the normal population to inhalation of air pollution components 17, 18 and in particular acid aerosols 19. Various studies have shown that overall lung deposition is increased in patients with obstructed airways 20–22. For example, a 30% reduction in airway cross-sectional area results in a deposition increase in the bifurcating airways of >100% 2, 23. There is also an interesting case-report of DEPs alone causing asthma 24; three nonsmoking railroad workers, without any previous history of asthma, developed persistent asthma after acute exposure to excessive levels of DE. This was registered in a situation where two locomotive units were coupled together and the crew riding in the second locomotive unit were exposed for 2–5 h to significant levels of DE. All three subjects developed asthma, which persisted 1–3 yrs after exposure. Although the actual levels were never measured, it is likely that such excessive exposure rarely occurs even in occupational situations.
Experimental studies of diesel emissions
Human subjects
Exposure challenge studies
Experimental human exposure studies have mainly been carried out using exposure chamber set-ups with controlled DE challenges. It is critical to ensure that the method is designed so as to maintain a certain relationship between the particulate and gaseous components and to obtain particles of the same size and chemical properties throughout the exposure series. A unique and carefully validated system for exposures has been particularly useful 25, 26. The effects of such diesel exposures have been evaluated in humans using symptom questionnaires, lung function measurements and bronchoscopy with biopsy sampling and airway lavage.
In one study, healthy volunteers were exposed to DE, with a NO2 concentration of 1.5 parts per million (ppm). Bronchoalveolar lavage (BAL) 18 h after DE exposure revealed a significant decrease in the total number of metachromatic cells (mast cells) in the bronchial portion and a significant increase in neutrophils in the bronchoalveolar portion. An increase in the CD4+/CD8+-ratio was found in the bronchoalveolar portion, along with a reduced phagocytosis rate by alveolar macrophages in vitro 26.
In another study on healthy volunteers, symptoms and lung function responses to DE were assessed. All the exposed subjects reported an unpleasant smell and eye irritation but there was no alteration in the lung function tests measured as forced expiratory volume in one second (FEV1) 25. Two other studies 21, 26, investigated whether the use of a particle trap on the tail pipe of an idling diesel engine would reduce the DE-induced effects on symptoms, lung function and airway inflammation, compared to effects induced by unfiltered DE. Exposure to DE without a filter caused increases in symptoms and airway resistance as well as airway inflammation with BAL neutrophilia. Macrophage phagocytosis was reduced. The particle trap reduced the number of particles by 46% but there was no significant difference in symptoms, lung function or in BAL neutrophil numbers after DE with or without the trap 26. Therefore, for these effects, the relative importance of DEPs and other components of DE has yet to be established.
In a recent study 27, filters intended for use in the air intake into the passenger compartment of vehicles were tested for their ability to prevent DE effects. Thirty-two healthy nonsmoking subjects were exposed for 1 h in a specially designed exposure chamber, once to air and once to unfiltered DE and subsequently to DE filtered with four different air intake filters. The exposure level was 300 μg of particles with a 50% cut-off aerodynamic diameter of 10 μm per cubic metre (PM10 300 μg·m−3). The study included measurements of lung function, symptoms and nasal responses. It proved possible to distinguish differences in efficacy between the filters, all of which gave air quality superior to the unfiltered exhaust. While no acute effects were seen on nasal lavage, rhinometry and lung function (measured as FEV1 and FVC), there were major effects on symptoms. The use of a particle filter in combination with an active charcoal filter was demonstrated to give significantly better results than the other filters.
The effect of a high ambient concentration of DE (300 μg·m−3) on various airway parameters including cells and soluble components was assessed 28. The results showed an increase in neutrophils, mast cells, CD3+, CD4+ and CD8+ T-lymphocytes in the airway mucosa, along with upregulation of adhesion molecules intracellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1 in the vascular endothelium 6 h after exposure to DE. In addition, increased numbers of cells expressing leukocyte function associated antigen (LFA)-1 (the ligand for ICAM-1) were found. This inflammatory response was an order of magnitude greater than the effects documented after allergen challenge in atopic asthmatics 29, 30 indicating a pronounced signal for inflammatory cell recruitment as a response to DE exposure. This upregulation of endothelial and leukocyte adhesion molecules provides a mechanism for the influx of inflammatory cells into the airways. Furthermore, immunohistochemical staining for cytokines has shown enhanced expression of interleukin (IL)-8 and growth related oncogene (GRO)-α in the airway epithelium, which can also play a role in the recruitment of inflammatory cells after exposure to DE 31. The increase in neutrophils in the airway mucosa following exposure to DE was also evident in the bronchial wash, whereas the increase in the number of submucosal mast cells did not correspond to any changes in the number of metachromatic cells (mast cells) in the lavages. However, an elevated concentration of methyl-histamine detected in the BAL after exposure to DE supports increased degranulation of mast cells 28. In contrast to the pronounced inflammatory response detected in the airways, lung function parameters were found unaffected following exposure to DE 28. Consequently, lung function measurements alone cannot be used to exclude adverse air-pollution-associated airway responses.
These pronounced airway inflammatory responses were detected at a fairly high (300 μg·m−3) concentration of DE. In order to evaluate if exposure to a lower concentration (more similar to relevant exposure concentrations in ambient air), would induce a similar response, healthy and asthmatic subjects have been exposed to DE with a PM10 concentration of 100 μg·m−3 for 2 h. Bronchoscopy with biopsy sampling, bronchial wash and BAL was performed 6 h after the end of the exposure. Data from this study are forthcoming.
Studies of induced sputum have also been used to evaluate DE effects on the human airways 32. Sixteen healthy nonsmoking subjects were exposed to air and DE at a particle concentration of 300 μg·m−3 for 1 h. Sputum induction was performed 6 and 24 h after each exposure. Six hours after exposure to DE, a significant increase was found in neutrophil percentage of total cells in sputum, together with an increase in the concentration of IL-6 and methyl-histamine, compared to control air exposures.
In summary, these data confirm that the acute mediator and cytokine responses, together with the enhanced expression of the vascular adhesion molecules in the airway mucosa, may represent an early stage in the inflammatory response following exposure to DE, and be of importance in the development of the DE induced airway inflammation. The results further suggest that the DE concentration is an important factor to take into account when evaluating the time course of DEPs-induced airway inflammation.
Nasal challenge studies
DEPs have been shown to potentiate immunoglobulin-E (IgE) production in human respiratory mucosal membranes. Diaz-Sanchez 33 performed a study in which nasal DEP challenges at various doses were used to investigate the effect on localized immunoglobulin production. Four days after challenge with 0.3 mg DEPs a significant increase in nasal IgE, but not in other immunoglobulin classes, was detected. There was also an increase in the number of IgE secreting cells in nasal lavage, but no increase in immunoglobulin-A (IgA) secreting cells.
Human nasal provocation studies have shown that DEPs can act as an adjuvant to allergen. Nasal challenge was performed with DEPs (0.3 mg), the ragweed allergen Amb a I, or both, in a group of ragweed-sensitive subjects 34. Ragweed challenge alone demonstrated an increase in IgE and immunoglobulin-G4 (IgG4) as well as ragweed-specific IgE in the lavage fluid. However, after challenge with allergen and DEPs there was a sixteen-fold increase in ragweed-specific IgE. DEPs alone increased total IgE, but in combination with allergen there was an increase in antigen-specific IgE and in expression of Th0 and Th2-type cytokines (IL-4, IL-5, IL-6, IL-10 and IL-13). It is currently believed that such deviations of the immune system may have a crucial role for the development of an atopic response 35. These studies suggest that DEPs can enhance B-cell differentiation. By initiating and elevating IgE production, DEPs may theoretically play a role in the increased prevalence of allergic disease.
Nasal challenge with a combination of DEP and allergen has been shown to induce larger ragweed-specific IgE and IgG4 responses compared with DEP alone but with similar total IgE levels 36. There was also a change in the cytokine pattern, favouring allergic sensitization. The authors propose that synergism between DEP and natural allergens is a key feature in increasing allergen induced respiratory allergic disease.
There is also evidence of DEPs inducing IgE isotype switching. Fujieda et al. 37 exposed human volunteers, by nasal challenge, to DEPs together with ragweed allergen. There was a local increase in IgE production, cytokine production and an even greater increase in mucosa-specific ragweed IgE production. The isotype switching occurred only when DEPs and allergen exposure was combined. The findings suggest that increasing environmental DEPs with the same environmental load of allergen could be a factor behind the increasing sensitization and prevalence of allergic asthma.
Healthy, nonsmoking human volunteers have been exposed to DEPs by intranasal instillation and cytokines in nasal lavage were estimated after 18 h by an indirect approach (messenger riboneucleic acids mRNA) 38. Before challenge, most subjects had detectable mRNA levels of only a few cytokines (Interferon IFN-γ, IL-2 and IL-13), while after challenge with 0.3 mg DEPs, these three and a number of additional cytokines (IL-4, IL-5, IL-6 and IL-10) were seen in increasing levels. Increase in such nasal cytokine expression after DEP exposure could again contribute to enhanced local IgE production.
Taken together, these studies suggest that DEPs have the ability to induce IgE responses directly by acting on B-cells and indirectly by enhancing the opportunity for cytokine production. DEPs in combination with allergens are able to generate a Th2-type cytokine response and thus favour the production of allergen-specific IgE. It has therefore been proposed that DEPs can enhance B-cell differentiation and, by initiating and enhancing IgE production, they may be of importance in the increased incidence of allergic airway diseases 34.
Animals
Some general points must be raised regarding animal studies. Apart from the possibility that mechanisms may be very different from those in man, studies using radioactive particles have also demonstrated that there is a large difference in the dosimetry of the small airways of rodents compared to humans. Therefore, care must be taken when extrapolating animal data to humans. Certainly, for ozone at least, it has been shown that in order to elicit the same response in rodents as in humans, several times higher concentrations are required 39. In addition, in many studies on animals, the doses of DE are much higher than those humans are exposed to in daily life. Measurements in the Scandinavian countries have shown that the average 24-h particulate matter concentration varied from 30–150 μg·m−3 total suspended particles (TSP). However, in certain industrial areas, concentrations of up to 1,500 μg·m−3 have been measured. In Stockholm for example, 1-h mean values 300–500 μg·m−3 have been found at kerbsides and maximum levels of almost 600 μg·m−3 have been reached in a tunnel in the centre of the city 40.
Exposure studies in rats
Long-term studies looking at the effects of DE exposure in rats have demonstrated increased accumulation of particles and aggregates of particle laden macrophages in the alveoli and peribronchial interstitial tissues, as well as local inflammation, epithelial proliferation, fibrosis and emphysematous lesions.
For example, rats were exposed to DE for 30 months 41 and examined at 6 month intervals by electron microscopy. Changes consistent with anthracosis were seen after 6 months exposure at a concentration of 1,000 μg·m−3. There were foci of diesel particle laden alveolar macrophages. DEPs were found in type 1 epithelial cells and there was also hypertrophy and proliferation of type 2 cells. An infiltration of particle laden macrophages, neutrophils, mast cells and plasma cells in the interstitium of the alveolar septa was also seen. The most prominent changes, however, were focal shortening of cilia and the protrusion of nonciliated cells.
Mohr et al. 42 compared the effects of inhaled DE with those of coal oven gas mixed with pyrolyzed pitch. Three groups of rats were exposed to clean air, filtered DEPs or unfiltered DEPs (4,000 μg·m−3). Another group was exposed to coal oven flue gas mixed with pitch fumes pyrolyzed under nitrogen. Most of the DEP-exposed animals had deposits consisting of large amounts of carbonaceous particles phagocytozed by alveolar macrophages and this was associated with severe chronic inflammation, alveolar septal thickening, bronchiolo-alveolar hyperplasia and alveolar lipoproteinosis. Rats exposed to coal oven flue gas had much less severe inflammatory changes.
In a study exposing rats to increasing concentrations of DEPs (350–7,000 μg·m−3), for up to 24 months 43, a progressive increase in the lung burden of particles was seen at the higher DEP concentrations. Tracheal mucociliary transport was not affected, but there was a significant prolongation of long-term pulmonary clearance half-times in the two groups that were exposed to the highest concentrations of DEPs.
DEPs cause dust overloading and impairs pulmonary clearance at high doses in both rat and man. There appears to be a threshold above which particle retention and inflammation occurs. It has been calculated, mostly on the basis of evidence from animal studies, that the threshold is ∼500 μg·g−1 lung tissue 44. However, it is difficult to assess how this correlates to levels in the inhaled air, and there may be important interactions between the actual concentration and the duration of the exposure. After acute exposure to a high concentration of DEPs (5,700 μg·m−3 for 3 days) the particles were found to be eliminated efficiently 45. There was a rapid initial increase in the elimination followed by a decline in macrophage burden. A chronic low dose (50 μg·m−3 of DEPs; exposure for 52 weeks) was also cleared relatively efficiently. More than 80% of the inhaled DEPs had been eliminated 1 yr after the exposure. However, comparison showed that animals exposed to the higher concentrations cleared more of the total lung burden than those exposed chronically to low concentrations. At the longer low concentration exposure, the macrophage burden was relatively greater and associated with a steady development of lung maculae, suggesting that continuous low dose exposure may be more detrimental than acute high dose exposure.
Taking these experimental findings into account, a mathematical model for the prediction of lung burden and alveolar clearance in rats has been constructed 46. At low lung burdens, the alveolar clearance rate of diesel soot was calculated to be constant and due to macrophage elimination via the mucociliary transport system, whereas at high lung burden, the alveolar clearance appears to be determined principally by the capacity for transport to the lymphatic system. It should be recognized however that the elimination of particles from central and peripheral airways may be vastly different and partly dependent upon the tidal airflow during the exposure. Recent data in humans suggest that elimination from the peripheral airways may be considerably slower than previously thought 47.
Exposure studies in cats
A group of cats was exposed to DEPs for >2 yrs 48. Their exposure was divided into two periods. During the first period, which lasted 61 weeks, 6,000 μg·m−3of DEPs was used. At this point, no changes in the lungs were evident. During the following 62 weeks, a higher concentration of 12,000 μg·m−3 DEPs was used and after the 2 yrs, a pattern of restrictive lung disease had developed.
In another similar study, cats were exposed to DEPs and NO2 for over 2 yrs. After 62 weeks, the concentration of DEPs had increased from 6,340–11,700 μg·m−3 and the concentration of NO2 from 2.68–4.37 ppm 49. Morphological changes were seen, mainly in the proximal acinar regions of lungs, with peribronchial fibrosis, bronchiolar metaplasia, increased numbers of lymphocytes, fibroblasts and interstitial macrophages containing DEPs. The study indicated a persistent fibrogenic effect on the proximal acinar region of the lungs following long-term DE.
Intratracheal exposure studies of diesel exhaust particles in mice
In a study of in vivo toxicity of DEPs in the ICR mouse strain 50, an acute intratracheal instillation of DEPs (400–1,000 μg·mouse−1) was found to cause severe lung injury and high mortality. The cause of death was pulmonary oedema mediated by endothelial cell damage. The toxicological effect and the increased mortality were to a great extent prevented by pretreatment with the oxygen radical scavenger superoxide dismutase (SOD), supporting the hypothesis that DEP toxicity is connected to production of the radical superoxide O2−1 leading to endothelial cell damage.
Studies of developing lung in the rat
Studies on humans in different geographical areas have raised the concern that children and the elderly may be particularly susceptible to the harmful effects of air pollution. In support of this, there have also been studies indicating the possibility of impaired lung development in animals exposed to oxidant gases or inhaled toxicants early in life. In one study which aimed to investigate this, rats were exposed to DE (3,500 μg·m−3), NO2 (9.5 ppm), or to air as a control 51. One group, which represented a developing lung model, was exposed first in utero (by exposing the mother from conception and throughout gestation) and then from birth up to 6 months of age. Another group, representing an adult model, was exposed between the ages of 6 and 12 months. It was found that DEPs altered the airway fluid constituents and tissue collagen in both groups. Interestingly, in the adult group, there was a six-fold increase in neutrophils as well as increased cellularity in the lung-associated lymph nodes, delayed clearance of particles and an increase in lung weight. However, none of these changes were seen in rats exposed during development. In adult rats there was also a focal aggregation of soot-laden alveolar macrophages, but only scattered individual macrophages were found in the young rats. The authors concluded that there was no evidence for developing rats being more susceptible to the toxic effects of NO2 or DE; if anything, the data would indicate that developing rats may be less sensitive.
Studies in animals with experimental lung disease
Epidemiological studies have indicated that subjects with pre-existing lung disease may be more susceptible to episodic high levels of airborne pollutants than normal subjects 52. This has been studied in a rat model 53, in which pulmonary emphysema was induced in rats by intratracheal instillation of the proteolytic enzyme elastase, and manifested as enlarged alveoli, alveolar ducts and ruptured alveolar septa. These structural changes were not by themselves associated with inflammation or alterations of bronchioles. The emphysematous rats and a group of control rats were then exposed for 24 months to DE (3,500 μg·m−3), NO2 (9.5 ppm), or air as control. Different parameters were measured such as lung burden of diesel soot particles, respiratory function, BAL fluid composition, clearance of radiolabelled particles, pulmonary immune response, lung collagen, excised lung weight and volume, and histopathology. The hypothesis was that the effects of pre-existing emphysema and long-term DE exposure on lung function and morphology were additive. The results, however, showed that rats with experimentally induced emphysema were no more susceptible to inhalation of NO2 or DE than control rats. In fact, fewer soot particles accumulated in the emphysematous lungs 53, 54.
Intraspecies comparative studies
Morphological changes have also been examined in a comparative study of Cynomolgus monkeys and rats. Both animal species received four different exposures for 24 months: DEPs (2,000 μg·m−3), coal dust (2,000 μg·m−3), a combination of DEPs and coal dust (1,000 μg·m−3 + 1,000 μg·m−3) or ambient air 55. It was found that monkeys retained relatively more particulate matter than rats. The sites of particle retention were the same for DEPs, coal dust and the DEPs/coal dust combination. Rats retained more material in the lumen of alveolar ducts and alveoli, whereas monkeys retained more in the interstitium. Rats showed significant alveolar epithelial hyperplasia, inflammation and septal fibrosis. In contrast, such morphological changes were not seen in the monkeys. The results indicate that particle retention patterns and tissue reactions in rats exposed to DEPs may not be predictive of the reactions in primates 56; primates may retain more DEPs, but may also be less sensitive to the harmful effects.
Immunology and allergy
In vivo studies
To study the effect of DEP on the accumulation of particles in lung-associated lymph nodes and the effect on antibody responses after immunization, rats and mice 57 were exposed to varying doses of DEPs (350–7,000 μg·m−3) for 24 months. Control and exposed animals were immunized, by intratracheal instillation of sheep red blood cells, after exposure at 6, 12, 18, and 24 months. Pathological changes were maximal after 18–24 months at 3,500 or 7,000 μg·m−3 exposure levels. Microscopic findings were: increase in particle content of alveolar macrophages; concentration of macrophages closer to terminal bronchioles; wall fibrosis; bronchial metaplasia; and, in some cases, squamous metaplasia. There were cholesterol clefts and increased numbers of interstitial neutrophils and intra-alveolar granular eosinophilic material with entrapped free particles. Lung-associated lymph nodes were blackened and enlarged due to clusters of DEPs. There was increased cellularity in rats exposed to the highest dose. However, the levels of specific antibodies were not significantly changed.
DEPs showed an adjuvant activity on IgE production in mice. Mice injected intraperitoneally with ovalbumin (OA) mixed with DEP showed higher IgE levels than mice immunized with OA alone. In addition, Japanese cedar pollen (JCPA) specific IgE production could be seen in mice immunized with JCPA mixed with DEPs, but not in animals immunized with JCPA alone 58. In another study 59, DEPs and carbon black were instilled intranasally in mice and the animals were then exposed to JCPA. There was a significant adjuvant effect of the particles on JCPA specific IgE and IgG production.
In order to elucidate whether the carbon core of the DEPs or the adsorbed organic substance is responsible for the effect 60, mice were immunized four times with either OA, OA with DEPs or OA with carbon black (carbon black). Specific IgE for OA was then analysed. Both DEPs and carbon black showed an adjuvant activity for specific IgE production after intranasal instillation, indicating that they were both responsible for the effect. There have also been other studies which have demonstrated this effect in mice 61.
A difference in antibody response between acute and chronic exposure to DEP was suggested in a study using combined intratracheal inoculations of DEPs and antigen in mice. Increased airway inflammation, and antigen specific IgG1, were seen after 6 weeks 62, whereas a longer exposure (9 weeks) was required before there was a detectable increase in antigen specific IgE.
When mice were injected intratracheally with OA alone or together with DEPs 63, there was no significant adjuvant activity of IgE production by DEPs (50 μg). However, the degree of eosinophilic inflammation in the airway corresponded to specific IgG1 production. In another study, the combination of sensitization to allergen together with diesel exposure, led to enhanced infiltration of eosinophils and neutrophils and also an increment of goblet cells, together with enhanced airway resistance and IL-5 and IgG1 production, but not IgE production. DE alone did not induce pathological changes 64. Other studies carried out with higher doses of DEPs (300 μg) under similar conditions have demonstrated an increase in specific IgE 65. Together with the aforementioned studies, this suggests that this response may be dose and time dependent. In ICR mice exposed to DEPs by intratracheal instillation, an increase in epithelial eosinophils, lymphocytes and goblet cells was seen, together with increased production of IgG and IgE and the proallergic cytokines IL-2, IL-4, IL-5 and granulocyte macrophage colocny stimulating factor (GM-CSF), but not interferon (IFN)-γ 62. The combination of DEPs and OA has also been shown to increase the bronchoconstriction response to inhaled acetylcholine 66. Furthermore, daily inhalation of DEPs may enhance the allergen response, possibly by increased local expression of proinflammatory cytokines such as IL-5 and GM-CSF 67. In one study, increased IgE production was also seen in mice after intranasal administration of suspended particulate matter (SPM) together with OA at 3-week intervals for a period of 21 weeks 68.
To study the effect of DEPs on allergen-induced bronchial hyperresponsiveness, mice were sensitized to OA and then exposed to DE (3,000 μg·m−3) 69. Three weeks after DEPs exposure, they were challenged with OA. DE exposure, combined with antigen challenge, induced airway hyperresponsiveness and airway inflammation, including increased numbers of eosinophils and mast cells in the lung tissue. DEP exposure alone also increased airway hyperresponsiveness, but there was no eosinophil infiltration. In a similar set-up, diesel inhalations (3,000 μg·m−3) in combination with OA sensitization increased the number of goblet cells in lung tissue, caused increased respiratory resistance and increased immune response measured as specific IgE, IgG1 and IL-5 in the lung tissue 70.
A rhinitis model in guinea pigs was used to study short-term effects of DE (3 h exposure at 1,000 and 3,200 μg·m−3) 71. Following diesel exposure alone, there was no induction of sneezing, nasal secretion or congestion. However, DEPs augmented the sneezing and nasal secretion induced by histamine, but had no significant effect on histamine-induced nasal congestion, suggesting that acute exposure to high levels of DE may cause nasal mucosal hyperresponsiveness, but no overt symptoms of rhinitis. Hiruma et al. 72 also reported effects of DEPs on the nasal mucosa of guinea pigs consistent with the development of nasal hyperreactivity. It was found that diesel exposure led to a dose dependent increase in histamine-induced vascular permeability, an increased eosinophilic infiltration into the epithelial layer and also enhanced nasal reactivity to histamine.
The mechanisms involved in asthma and allergy development after DEPs exposure may include increased penetration of allergen across the respiratory mucosa or direct modulation of the immunological response 73. Particulate air pollution includes particles from diesel and gasoline engine exhaust and biological materials such as plant debris and tyre debris e.g. latex, that may be allergens 74. Some allergens, such as the major grass pollen Lolium perenne -1 (Lol p1) allergen, are also shown to specifically bind to DEPs. This might be a possible mechanism for the triggering of asthma attacks and a theoretical contributing factor to the increase in asthma prevalence 75.
Ormstad et al. 76 have studied the allergen carrying effect of DEP. This was an in vitro study of allergen adsorption to particles such as indoor-suspended particles matter (SPM) and DEPs. They reported that the cat allergen (Fel d1), dog allergen (Can f1) and birch pollen allergen (Bet v1) could all be found on the surface of SPM, whereas house dust mite allergen (Der p1) was not. However, all four allergens were found to be absorbed to DEPs. A chemical characterization of the coating of Birch pollen grains collected during pollen season in the north of Stockholm has been published 77. The greatest portion (80%) consisted of n-alkanes and n-alkenes, but methylketones, ethers, alcohols and amino alcohols were also identified.
In vitro studies
Some in vitro studies have been performed on animal cells, but most of the work has been conducted on cells derived from humans. The eosinophil is a major effector cell in allergic inflammatory disorders. In one study, the effects of DEPs and DEP extract on eosinophil adhesion, survival rate and degranulation were assessed 78. Eosinophils, human mucosal microvascular endothelial cells (HMMECs) and human nasal epithelial cells (HNECs) were preincubated with and without DEP extract. Radiolabelled eosinophils were allowed to adhere to monolayers of HMMECs and HNECs with the degree of reactivity being determined after washing, and the numbers of adherent eosinophils were calculated. There was a significant increase in the adhesiveness of eosinophils to HNECs, but not HMMECs. DEP also induced eosinophil degranulation without changing the eosinophil survival rate. These results would indicate that DEPs play a significant role in the promotion of nasal hypersensitivity induced by enhanced eosinophil infiltration and degranulation.
The cytotoxicity of DEP-induced phagocytosis and the resulting immune response were studied in human bronchial and nasal epithelial cell cultures 79. DEPs exposure induced a time and dose-dependent membrane damage. Transmission electron microscopy showed that DEPs underwent endocytosis by epithelial cells and translocated through the epithelial cell sheet. Flow cytometric measurements allowed establishment of the time and dose dependency of this phagocytosis and its lack of specificity for different particles (DEPs, carbon black and latex were tested). DEPs led to a time-dependent increase in IL-8, GM-CSF and IL-1 beta release. This inflammatory response occurred later than phagocytosis and it appeared to depend on the types of adsorbed compounds, as in this study carbon black had no effect on cytokine release.
Airway epithelial cells play a prominent role in the pathogenesis of respiratory disease. Studies have shown that exposure of nasal or bronchial epithelial cells to DEP results in increased synthesis and release of proinflammatory mediators, cytokines such as IL-6, IL-8 or GM-CSF, and adhesion molecules 80.
In line with the observations during nasal challenge in vivo 36, 38, studies on isolated human B-lymphocytes have demonstrated enhanced IgE synthesis following exposure to DEPs 81. In a study of purified human B-cells stimulated by IL-4, the ability of DEPs to induce IgE production could be mimicked by polyaromatic hydrocarbons (PAH) extracted from DEPs. However, DEP-PAH did not induce IgE production in unstimulated B-cells, indicating that it only enhances ongoing IgE production. Phenanthrene, a major polyaromatic hydrocarbon and an important component of DEPs, has shown the same enhancing effect on IgE production in a human B cell line 82.
The effect of DEP-PAH on the release and mRNA expression of IL-8, monocyte chemotactic peptide-1 (MCP-1) and RANTES (Regulated on activation normal T cell expressed and secreted) was investigated by paripteral blood mononuclear cells (PBMCs) obtained from healthy subjects. The production of protein in supernatants was assessed by enzyme linked immunoabsorbent assay (ELISA), and mRNA production by semiquantitative reverse transcriptase polymerase chain reaction (RT-PCR). There was a dose dependent increase in the secretion of IL-8 and RANTES in response to increasing concentrations of DEP-PAH (range 0.5–50 ng·mL−1). However, there was a significant dose dependent inhibition of MCP-1 secretion. The expression of mRNA coding for IL-8, RANTES and MCP-1 showed variations that parallelled the production of the corresponding proteins. These results suggested that DEP-PAH can modulate chemokine pathways at the transcriptional level 83.
Other studies have been performed using cultured human bronchial epithelial cells (HBEC) exposed to DEPs 84. Exposure of these cells to DEP at 50 μg·mL−1, filtered DEPs solution or DEP at 100 μg·mL−1 attenuated the ciliary beat frequency (CBF) dose dependently and increased the release of IL-8, GM-CSF and soluble ICAM-1 (s/CAM-1). The observations support the hypothesis that DEPs exposure may lead to functional changes and the release of pro-inflammatory mediators with the potential to influence the development of airway disease. Specifically, infiltration of neutrophils and other inflammatory cells would be promoted by the observed changes.
In another report, the effect of DEPs on the CBF, and production of IL-8, GM-CSF, RANTES and sICAM-1 by cultured human bronchial cells was compared between nonatopic, nonasthmatic subjects and atopic patients with mild asthma. Bronchial cells from these two groups were exposed to 10–100 μg·mL−1 DEP for 24 h. The baseline CBF was the same in both groups. There was a significant attenuation in the CBF, in response to increasing DEP levels in both groups, with the largest changes at 100 μg·mL−1. The cell cultures from asthmatics constitutively released significantly greater amounts of IL-8, GM-CSF and sICAM-1 and were the only cultures to release RANTES. In response to 10 μg·mL−1 of DEP, there was a significant increase in the release of IL-8, GM-CSF and sICAM-1 in cells from asthmatics. However, exposure to doses of 50 and 100 μg·mL−1 led to a decrease in the release of IL-8 and RANTES. In contrast, only these higher concentrations of DEPs caused a significant increase in the release of IL-8 and GM-CSF in cells from nonasthmatics. These results suggest that the bronchial epithelial cells of asthmatics are more sensitive than cells from normal subjects to DEPs, and that they are also different with regard to the amounts of proinflammatory mediators released 85.
The effect of DEPs has also been studied in three different in vitro systems of human airway epithelial cells: nasal polyps, bronchial cells obtained at lung tumour resection or at autopsy, and the bronchial epithelial cell line BEAS-2B 86. DEP (10–100 μg·mL−1) induced a dose- and time-dependent stimulation of IL-8 and GM-CSF production by all three kinds of epithelial cells. Using double chamber plates, it was shown that the cells could only be stimulated from the apical side. As a control, neither coal nor graphite showed any stimulatory effect, whereas benzpyrene, a constituent of DEPs, did. Thus it appears that human epithelial cell exposure to DEP might stimulate cytokine production, with possible relevance for the allergic inflammation.
In another investigation, the effect of DEPs in doses 40–330 μg·mL−1 was recorded using the human bronchial epithelial cell line, BEAS-2B 87. DEP diameter ranged 25–35 nm and the particles were phagocytized by these cells. There was an increase in IL-6 and IL-8 production (11-fold and 4-fold, respectively). Exposure of these cells to tumour necrosis factor (TNF)-α also stimulated a strong increase in IL-6 and IL-8 production. There was an additive effect on the production of IL-6 and IL-8 after TNF-α priming and subsequent exposure to DEPs, seen only at low doses of DEPs (10–70 μg·mL−1) and TNF-α (0.05–0.2 ng·mL−1).
To elucidate the molecular mechanism of action of DEPs, IL-8 gene expression was studied by northern blot analysis, and run-on transcription assay in human bronchial epithelial cells. Suspended DEPs (1–50 μg·mL−1) increased the steady state levels of IL-8 mRNA. Electrophoretic mobility shift assay (EMSA) demonstrated that DEPs induced increased binding to the specific motif of the nuclear transciption factor κB (NF-κB), but not of transcription factor AP-1. NF-κB is known to stimulate the trancription of genes coding for inflammatory molecules like TNF-α and IL-8. The luciferase reporter gene assay using wild type and mutated NF-κB binding sequences showed that DEP-induced NF-κB activation was involved in IL-8 transcription. These results indicate that DEPs activate NF-κB, which may act as an important mechanism for the increased inflammatory cytokine release 88.
The effects of DEP extract on the expression of histamine H1 receptor (H1R) mRNA and the production of IL-8 and GM-CSF in human nasal and mucosal microvascular endothelial cells have also been investigated. The change in expression of H1R mRNA was evaluated by RT-PCR and southern blot analysis. The amount of IL-8 and GM-CSF was measured by ELISA. It was found that DEP led to a significant upregulation of the H1R gene expression as well as an increase in histamine-induced IL-8 and GM-CSF production 89.
Yang et al. 90 studied the possible role of cytokines in the toxic effect of DEPs on rat alveolar macrophages (AM). The macrophages were incubated with DEPs in different concentrations as well as with methanol, washed DEPs or DEP methanol extracts. High concentrations of DEPs and methanol extracts increased IL-1 secretion by AM, while there was no effect on TNF-α. DEPs inhibited production of IL-1 and TNF-α stimulated by endotoxin (lipopolysaccharide). The results suggest that the pro-inflammatory cytokine IL-1 may play a role in the pulmonary response to DEP inhalation. The suppressive response of AM pretreated with DEPs, to endotoxin stimulation may be a factor contributing to the impairment of pulmonary defence systems after prolonged DEPs exposure.
In one study, the effect of DEPs on isolated tissues and cultured cells from the respiratory tract of guinea pigs was examined 91. DEPs induced a dose dependent relaxation of tracheal smooth muscle and time dependent cytotoxicity on tracheal smooth muscle cells and lung fibroblasts. On the basis of pharmacological interventions, it was suggested that the cytotoxicity of DEPs may be mediated via generation of oxygen radicals. DEPs have also been shown to produce aggressive oxygen radicals in a cell-free in vitro system in the presence of appropriate electron donors 92, which in part may explain the potential toxicity and mutagenicity of DEPs.
The role of endogenous nitric oxide
Nitric oxide (NO) is found in the exhaled air of animals and humans. Increased levels of NO in exhaled air are associated with asthma and airway disease and NO has been observed to suppress Th1 cells leading to a Th2 type response that is associated with allergy 93. The role of NO in asthma-like symptoms induced by DEPs has been studied in mice 94. Repeated intratracheal instillation of DEPs in mice induced a four-fold increase in macrophages, neutrophils, eosinophils and lymphocytes in BAL fluid. DEPs induced a two-fold increase of NO in exhaled air and an increase in staining for the enzyme nitric oxide synthetase (NOS) in the airway epithelium. The increase in respiratory resistance induced by the DEP instillation was suppressed by the NOS inhibitor NG-methyl-L-arginine (L-NMA). These findings in mice suggest that some effects of DEPs may be mediated by endogenous NO. In contrast, Muto et al. 95 found that DEP, like L-NMA, abolished the acetylcholine (Ach)-induced relaxation of aortic rings preconstricted with phenylephrine. NO release from aortic rings in response to Ach was inhibited by DEPs (100 μg·mL−1, 60 min). NO released by the bronchial epithelium in rabbit bronchial strips attenuated the bronchoconstriction induced by Ach and this attenuation was abolished by 60 min preincubation with DEPs (100 μg·mL−1) or L-NMA. Their conclusion was that inhibition of NO release by DEPs may be a part of the observed respiratory effects of DEPs. Clearly further studies are required to establish the significance of these opposite results in two different animal models.
Cardiovascular effects
Epidemiological studies have associated increased mortality in cardiovascular diseases with episodes of heavy air pollution 96, 97. Seaton et al. 14 suggested that the ultra fine particles would induce airway inflammation in susceptible individuals, release of mediators and an increase in blood coagulability. A literature survey has provided some support for the hypothesis of possible association between occupational exposure to dust and increased risk of ischaemic heart disease 98–100. For example, coal miners showed manifestations of pneumoconiosis and increased incidence of ischaemic heart disease. There were also increased levels of fibrinogen in the blood of coal miners with pneumoconiosis, and fibrinogen is a risk factor for ischaemic heart disease. A hypothesis has been put forward that long-term inhalation of particles retained in the lung induces an inflammation which is accompanied by an increase of plasma fibrinogen, leading to elevated risk for blood clotting and ischaemic heart disease 100.
In a recent study combining measurement of air pollution exposure with personal meters and of haematological markers in collected blood, it was found that there was a relationship between exposure to particulate matter, measured as PM10, and changes in haemoglobin concentration, haematocrit (packed cell volume) and red blood cell count 101. There were concomitant decreases in platelet number and fibrinogen levels. By also measuring plasma albumin, the authors concluded that the decrease in haemoglobin was caused by increased peripheral sequestration of red blood cells, rather that generalized haemodilution. The study supports particulate air pollution, or a very closely associated confounding factor, having the potential to affect important cardiovascular phenomena. Studies of acute episodes of increased air pollution have documented effects on plasma viscosity 102, and acute exposure to DE has effects on inflammatory cells in the blood 28. In a time series panel study of particulate air pollution, there was an association between particle levels and pulse, but not with oxygen saturation 103. There was a time lag between exposure and the effect on pulse that the authors interpreted as an indication of lung inflammation with consequent release of mediators and cytokines being the primary event. Some recent studies in the elderly 104–106 have implied an association between increased particulate matter (PM2.5) in ambient air pollution and decreased heart rate variability, suggesting a possible contribution to increased cardiovascular mortality and decreased autonomic control.
The studies of the direct influence of DE on various cardiovascular responses, however, remain very few. Pretreatment of human serum with DEP extracts (500–2,500 μg·mL−1) gave a dose-dependent reduction in complement haemolytic activity of up to 20% 107, and activation of the alternative complement pathway. A direct toxic action of DEPs was examined in a model of isolated atria from guinea pigs 108. DEPs in lower doses (10–500 μg·mL−1) induced a transient but dose-dependent increase in contractile force. DEPs in doses >500 μg·mL−1, only decreased contractile force and induced cardiac arrest. It was concluded that cardiac toxicity contributes to the lung oedema that is known to be one prominent cause of death in DEPs exposed animals. It appears unlikely, however, that inhalation of DEPs by humans could produce the concentrations employed in these particular experiments. The experiments indicating effects of DE on the NO system and oxygen radical formation have been discussed above. All in all, the possible mechanisms involved in the alleged role of DE on various cardiovascular events remain unknown.
Discussion
The epidemiological support for particle effects on asthma and respiratory health is very evident. The experimental studies of DEPs include in vitro models, animal in vivo models, studies of healthy humans and occasional observations in patients. Respiratory, immunological and systemic effects have indeed been documented. The main effects on the respiratory system are summarized in table 1⇓.
Some effects of diesel emission with relevance to the respiratory system
The acute effects include irritation of the nose and eye, lung function changes, airway inflammation, headache, fatigue and nausea. In addition to symptoms, exposure studies in healthy humans have documented a number of profound inflammatory changes in the airways, notably, before changes in pulmonary function can be detected. It is likely that such effects may be even more detrimental in asthmatics and other subjects with compromised pulmonary function.
Chronic exposure to DE induces cough, sputum production and lung function decrements. Pathological and histological findings in the lung after DEPs exposure have mostly been studied in rats and include increases in lung weight, increased numbers of particles in the lung and an increased burden of soot, associated with alveolar infiltration of macrophages, macrophage aggregation, chronic inflammatory responses, proliferation and hyperplasia of alveolar epithelium and type 2 cells, thickening of alveolar septa and wall fibrosis.
Due to the complexity of DE, it is likely that some effects are caused by the gaseous components whereas other effects relate to the particle content. The suggested mechanisms of detrimental actions of particulate matter include oxidative stress and actions of particulate matter content such as metals, hydrocarbons, acids and carbon core. The ultrafine particles are currently suspected of being the most aggressive particulate component of DE. Comparison of DEPs and carbon black in animal inhalation studies show that both induce a reduction in lung function and accumulation of macrophages, suggesting that the toxic effect of DEPs is, in part, coupled to the carbon core. However, much more work is needed to pinpoint the relative role of different components of DE, as well as the interaction between the different components, and other environmental factors.
There are several observations that support the hypothesis that DE is one important factor contributing to the allergy pandemic. For example, DEPs introduced by different routes, intraperitoneally, intranasally or intratracheally, may act as adjuvant to allergen and hence increase the sensitization response. This has been observed both in human and animal studies. IgE production in response to allergen has been shown to be enhanced by DE. DEPs affect human B-cells and may enhance IgE production by several mechanisms. In addition, DEPs show effects on the allergic response that involve inflammatory cells in the respiratory mucosa, such as T cells, mast cells and epithelial cells, and also on local production of various pro-inflammatory cytokines. Another possible mechanism of action of DEPs on allergic responses is to act as a carrier of pollen allergens, allowing enhanced deposition of pollen in the lower airways. Allergens bound to DEPs may trigger asthma attacks and DEP-binding may facilitate penetration of allergen through the airway mucosa.
It is often believed that atopic children are a specially sensitive group, but the few animal studies that have been carried out have not supported the hypothesis that the developing lung 51 or lungs with induced emphysema 53 are more prone to lung injury due to particulate matter. On the basis of published data, there is no single mechanism of action, that can explain the various public health effects of particulate air pollution shown in epidemiological studies. One major shortcoming of many experimental studies relates to their inability to establish whether or not the exposures used for the studies are relevant to the background or peak exposures which may occur in real life, acutely or chronically. In addition, theoretically, both acute exposure to high levels of diesel exhaust particles and chronic exposure to low levels may impair respiratory functions and have various other detrimental effects, but the mechanisms may be different. Such a situation exists in the case of a classical trigger of asthma such as allergen: acute exposure to high allergen doses produces effects that differ from those caused by repeated low dose exposure 109–111. As reviewed, some of the studies of chronic exposure of rats to diesel exhaust particles (vide supra) also highlight such differences between high and low dose exposures. It is therefore imperative to further assess acute and chronic effects of diesel exhaust in mechanistic studies with careful consideration of exposure levels. Whenever possible and ethically justified, such studies should be performed in humans, but animal and cell culture models that are sufficiently predictive and sensitive, may also provide important information on these matters.
- Received October 5, 2000.
- Accepted October 20, 2000.
- © ERS Journals Ltd
