Pollution and the immune response: atopic diseases – are we too dirty or too clean?

The effect of pollutants on the immune system has filled tomes and is almost always controversial whether the topic is disinfectants in drinking water or particulates and lung cancer. Even whether outdoor and indoor air quality is improving or declining has been a source of considerable investigation, consultation and often litigation. Few areas, however, yield so much uncertainty and debate as the role of airborne pollutants in atopy. Atopy is defined as an immunoglobulin E- (IgE) mediated allergic response to environmental antigens and can be manifested as rhinitis, asthma or atopic dermatitis. While there is some debate over whether there has been a real increase in atopy over the last decades, it is indisputable that there has been a dramatic change over the last 200 years. While isolated cases of anaphylaxis had been noted before, allergies were unheard of until the early 19th century. Only 54 years after the first description of hay fever in England in 1819, it was considered an epidemic and a clear association with urbanization had been reported. 1 What aspect of urbanization is involved has been much debated. There has been much discussion amongst both experimental immunologists and epidemiologists over whether pollutants are responsible for increased allergic sensitization and/or severity.2,3 In the last few years new studies have supported an alternative explanation, namely the decline in infectious diseases in industrialized countries.4,5 Does this ‘hygiene theory’ toll the death knell for the ‘pollution theory’? In this review I will argue that there is a compelling case to be made for both theories and that even this is not the entire picture. Some studies have linked the rise of atopy with an increase in living standards and immunization programs in industrialised countries.6,7 This has led to the theory that, while the normal response to childhood infections is the deviation of the immune system to a T helper (Th) 1-type cytokine response, the absence of these infections in ‘Western’ society has resulted in the predominance of a Th2-type response.5,8,9 The Th2 cytokine environment is known to promote IgE production and atopic diseases. 10 In support of this theory, some epidemiological studies have shown an inverse association between the risk of atopy and sibling number, 11 birth order, 12 and nursery school attendance, 13 all factors that will increase the risk of infections in infancy. Perhaps the most cited evidence has come from a study looking at Japanese schoolchildren vaccinated against Mycobacterium tuberculosis. 14 There was an inverse relation between tuberculin responses and asthma, rhinitis and eczema. Additionally, serum IgE levels, both total and allergen-specific, were also lower in children with positive tuberculin responses. 14 While some studies looking at measles 15 or hepatitis A virus 16 infections have supported the conclusion that infections protect against atopy, others have shown an association between asthma or allergic disease and M. tuberculosis infection only in women but not in men 17 and yet others have failed to show similar associations between atopy and measles 18 or bacillus Calmette–Guérin (BCG) vaccination. 19 An alternative explanation has been proposed that mycobacteria do not lower the risk of atopy, but that atopics have an impaired ability to make a Th1 response to mycobacteria. 20 Yet, in experimental animal models, immunization with mycobacteria or other infectious agents will result in the promotion of a Th1 response and the inhibition of Th2 responses.21,22 As a cautionary note, it should be mentioned that, in the UK and the USA the greatest increase in atopy prevalence has come not in the last decades but in the years between 1830 and 1900. Deaths from infectious diseases have declined markedly in these countries only during the 20th century. By the turn of the century improvements in hygiene and immunizations regime were still limited and the life expectancy was around 47 years, only 20 years more than the Greco–Roman era, 2000 years earlier. 23 In 1900, 30·4% of all deaths in the USA occurred among children aged less than 5 years; in 1997, that percentage was only 1·4%. 24 There is broad agreement that the increased prevalence of atopy is due to the skewing of the immune system to a Th2 environment. While the ‘hygiene theory’ is an attractive explanation, other possibilities remain. Some have argued that the shift is due to diets,25,26 others that it is because of the adjuvants used in immunizations, 27 others that high growth rates in successful pregnancies select for a Th2-type environment, 28 and, as we shall see, certain pollutants can interact with allergen to promote Th2 cytokines. The most cited argument against a role for pollutants in asthma and allergy onset is the comparative studies between the former nations of the now reunited Germany.29,30 Despite increased particles and SO2 levels in East Germany, allergic airway disease was significantly higher in West Germany. Additionally, prevalence of positive skin-prick tests (atopy) in an East German city (Leipzig) were half that of one in West Germany (Munich). Similar results have also been observed in comparisons between Sweden and Estonia. 31 Yet, these data are not as clear cut as it appears. Even within these studies, within each country, hay fever, atopy and asthma were all higher in industrialized polluted cities than in rural or non-industrial towns. This has led some to conclude that air pollution is an important risk factor but that some undefined feature of Western lifestyle may be even more so. 32 Indeed, since reunification, the prevalence of hay fever and atopy seems to have increased in the former socialist states 33 presumably because they are adopting a more ‘Westernized’ lifestyle. Subsequent studies have compared two polluted counties in East Germany with a neighbouring country without any source of industrial pollution, where lifestyles (including family size and infection rates) are similar in all three counties. 34 Again, children in the polluted counties had higher sensitization rates and specific IgE levels. It is important to note that there seem to be distinctions between pollutants. Certain dusts and SO2 associate strongly with infections and irritant responses, after controlling for as many lifestyle covariates as possible. 35 Conversely, parameters of allergic sensitization correlate significantly with combustion sources such as automobile exhaust. This agrees with epidemiological studies that have shown that in Japan, people living on busy roads lined with cedar trees have higher allergic cedar pollinosis than residents living on similar streets with much less traffic. 36 Similarly, allergic rhinitis is heavier in residential areas with heavy truck traffic 37 and some reports suggest that allergic symptoms correlate with the distance of residences to roads with heavy traffic. 38 The studies cited above have sparked renewed focus on the role of fossil-fuel combustion products on allergic inflammation. Because of their universal nature, indoor concentrations of these products are often equal to or exceed that found in ambient outdoor air. In particular, much of the work in the last decade has concerned diesel exhaust particles (DEP) and the xenobiotic compounds they contain. Consisting of an inert carbon core containing unburned fuel petrochemicals, DEP have been used as model pollutants to understand the mechanisms involved. Many of the chemicals they contain are also present in other pollutants. 39 For example, the prototypic polyaromatic hydrocarbon benzo(a)pyrene found in DEP, 40 is prominent in second-hand smoke, a risk factor for allergic sensitization identified in dozens of epidemiological studies.41,42 In cities such as Los Angeles, 40% of the 10-μm particles (PM10) in the atmosphere are derived from diesel vehicle engines. This has become an area of increasing public health concern as diesel engines have come to dominate the trucking industry in recent decades; for example, 88% of all heavy-duty trucks (> 26 000 lb.) now run on diesel fuel. 43 The rest of this review will use the example of DEP (as this is the best characterized and most studied pollutant of the last few years) to focus primarily on the changes that a pollutant can have on the immune response and especially allergic reactions. Much of the experimental work on pollution and atopy has centred on the ability to alter immunoglobulin production. In particular IgE regulation has been widely studied owing to the strong association of asthma with serum IgE levels and its central role in both the early and late phase of the mucosal response to allergens and the induction of allergic inflammation in humans. In 1986, Muranaka et al. published the first demonstration that DEP could have an adjuvant activity. 44 Instillation of ovalbumin (OA) or Japanese cedar pollen (JCP) mixed with DEP and injected intraperitoneally enhanced the antigen-specific IgE antibody response. Shortly after Takafuji et al. demonstrated an identical effect if an intranasal route of administration was used. 45 Since then, these studies have been validated many times and DEP shown to have similar effects if administered intratracheally or by aerosilization. 46–48 Other studies have focused on IgG1. In mice and other animal models, IgG1 can also cause anaphylactic reactions and is sometimes the major reaginic antibody (e.g. in the guinea pig 49) and may be important in the development of allergic airway inflammation. 50 There is no homologue to IgG1 in humans. Some studies using murine models have failed to show adjuvant activity for IgE but can show DEP can enhance IgG1 responses to allergen.48,51 These differences probably result from differences in doses, route of administration, murine strains and exposure duration. The ability to modulate immunoglobulin production is not confined to DEP. Indeed the ability of both SO2 and ozone to enhance allergen-induced anaphylactic responses in mice was reported 20 years ago. 52 Ozone can also increase the number of IgE-containing cells in the lung following allergen exposure. 53 Guinea pig exposure to SO2 increases anti-OA IgG1 in serum and bronchoalveolar lavage (BAL) fluid. NO2 exposure can augment local IgE (as well as IgA and IgG) to dust-mite antigen in rats. 54 Second-hand tobacco smoke, a pollutant which is often implicated as a risk factor for atopy, has many of the characteristics of DEP (e.g. a particulate nature) and shares many chemicals in common, has recently also been shown to increase IgE to OA in a mouse model. 55 A major advantage of using DEP as a model pollutant has been the ease by which animal models have been transferred into human studies. These studies have categorically shown the potential of DEP to induce and exacerbate IgE responses in vivo in the human upper respiratory tract. 56 Human atopic volunteers were sprayed intranasally with DEP, with doses ranging up to 0·15 mg into each nostril. 57 Although this dose may seem high, in certain occupational or everyday settings (such as waiting at a bus stop) exposure to DEP can be much higher. DEP exposure resulted in a dose-response associated increase in total IgE production and ε mRNA in nasal lavage fluid. 57 Four days following DEP challenge, total IgE proteins levels found in nasal washes were up to 10 times greater than baseline values, while there was a 25-fold increase in the levels of ε mRNA in cells recovered from the nasal lavage fluid. By day 7, IgE levels returned to baseline. This response was isotype specific as levels of total nasal IgG, IgA, IgM were unchanged by DEP provocation. Cells recovered from nasal washes were used to enumerate immunoglobulin-secreting cells. After DEP challenge the number of IgE-producing cells increased from less than 1 in 2000 000 to more than 1 in 1000 000 of total cells. However, the number of IgA-secreting cells remained unchanged after DEP challenge. 57 Thus DEP seems unlike other pollutants such as NO2 or SO2, which do not produce IgE in the absence of an allergen. 58 Subsequent studies have investigated the ability of DEP to act as an adjuvant in humans in vivo to a co-administered antigen. 59 Ragweed-sensitive subjects were challenged with DEP, ragweed allergen (Amb a I) or both simultaneously. Nasal washes were performed at different days after challenge. At the peak of the response, ragweed-specific IgE was 16-fold higher following ragweed plus DEP challenge compared with ragweed alone. Once again, there was no change in any other immunoglobulin isotype except IgG4, suggesting that DEP acts in an isotype-specific manner. Thus for both total and ragweed-specific responses, only IgE and IgG4 levels were boosted by DEP. This may be because both IgG4 and IgE production are under very similar, although not identical, controls. 60 A popular explanation for the adjuvancy effect of DEP is the physical interaction between pollutants and pollen grains. It has been suggested that pollen from trees growing in cities are more allergenic. 61 Furthermore, purified grass pollen allergen molecules bind to DEP when in Thus it is that DEP may a as a that allergens into the Although this may be the case in some it is probably not the in most of the studies IgE production in can when injected with DEP at from the respiratory tract. can also increase specific IgE production in mice and enhance the local responses in cells of the in mice challenged by OA into the However, the responses by are less than that with DEP. In the chemicals from DEP also can enhance IgE a polyaromatic hydrocarbon present in DEP, tobacco and many other combustion can induce IgE production in mice injected intraperitoneally with The from DEP also increased the IgE production of cells with and in a polyaromatic hydrocarbon in DEP and also found in tobacco smoke, had similar A more explanation for the changes in IgE by pollutants is that they enhance IgE production by on or by the local environment to one for allergic induce a Th2 cytokine A for this be the induction of In humans and mice and can and in association with or immunoglobulin of cells to In Th2 is on the of Thus the induction of by pollutants should result in increased IgE production and the of a Th2 response. pollutants have been shown to enhance production to allergen in and in both animal and human in vivo of DEP with production in was times greater than in mice injected with OA alone. DEP plus higher levels than in the model. Similarly, the demonstrated that intranasal challenge with DEP and allergen increased responses in cells compared with allergen alone. Subsequent studies demonstrated that DEP could have the effects on when used in a more model that of administration 47 Similarly, derived from mice with during antigen sensitization increased antigen Human studies show that intranasal challenge of DEP can augment of many in a manner. levels of mRNA for and all increased in cells recovered from nasal levels, which are to be could be in washes performed 18 after DEP challenge. In two studies in humans and one in demonstrated the of an enhanced Th2 response following challenge with DEP and In the subjects were challenged with ragweed DEP, or to challenge the only in nasal cells for which baseline levels of mRNA were were and challenge with ragweed only and changes in cytokine 59 In challenge with both DEP and allergen resulted in significantly increased of and but of the Th1 In mice intratracheally for DEP with OA an increase in in lung and compared with is a Th2 central to allergic While local levels of and factor also it is that levels of were These studies that the of DEP plus allergen against a Th1 and the a Th2-type cytokine response. In these studies, the induction of Th2 responses in the human and murine models were with increased production of antigen-specific IgE and DEP or induce a production and the Th2 cytokine is still studies have implicated both and In atopic DEP can with doses of which in are to produce In one and cells could be in nasal washes as early as after challenge with DEP plus allergen. At this the of cells in the from to suggesting that T cells were not the source of this early In cells between and of cells. or cells were at this As the allergic response the source of At the peak of 18 after challenge, for was observed in cells. This study that cells produce the of which the response a Th2 by the of cells in the local In support of this are studies that show that DEP can induce and production from purified cells in et al. have shown that in production can be by the in both and a human but not in suggesting that by of the is Human IgE is by by the which also and major not alter regulation of suggesting that it not increase by with models have demonstrated that the increased allergic antibody production and the of a Th2 response by interaction between allergen and pollutants may also result in enhanced effects on allergic In a of studies, and have shown that in mice challenged intratracheally with DEP and not only were IgG1 and IgE and Th2 higher than in OA mice but also there were clear of increased allergic of the of and was significantly as were in the and these changes significantly with anti-OA IgG1 production. Additionally, of that DEP and was in these mice by respiratory to and was found to increase with DEP The their model to that these results could be by diesel exhaust were with plus OA and to diesel exhaust for allergic airway inflammation and was The number of of of the was on greater in mice DEP plus OA than in OA while were higher. Similarly, and airway were also et al. have shown that in a DEP intranasal mouse against will the responses, suggesting that DEP may induce airway by effect of diesel exhaust on human lower have been to normal and have subjects for 1 to diesel exhaust and performed 24 after exposure. and in the was with increases in and Additionally, by in was Similar results have been observed following NO2 or ozone performed after a exposure to diesel exhaust have shown increases in cells and T and of the molecules and both implicated in allergic In human exposure to diesel exhaust increased concentrations of the in nasal Similarly, in the animal models, the effects of DEP could be by into the As are to be to increased these studies suggest that DEP and other pollutants are by induction of a review of the growing evidence to support this studies to pollutants have been performed on airway responses to allergen has been reported following exposure to the of NO2 and 200 a dose of allergen to induce a in in 1 by compared with exposure to air. in studies have demonstrated that both and cells may and which may be involved in allergic airway inflammation observed in and have demonstrated that SO2 and DEP could all cause of and such as and from human normal cells in have shown that DEP and benzo(a)pyrene can also have similar effects on and cells. had such that the activity of DEP is not due to its particulate The production of and by airway cells can be by the and DEP and DEP can induce in and production in is a of the family of or and is on The family on and and This family is especially important in allergic inflammation as they are involved in of of from and and of and cytokine production. In of has been reported in allergic and in nasal or of allergic subjects following challenge with allergen. et al. have demonstrated that in the from diesel an increase in both and but a in production. Furthermore, from cells with DEP a significantly enhanced activity for and It seems that while DEP with allergen to act on cells and to induce and IgE it can act in the absence of allergen to induce from and cells and promote reactions. these effects are all because of one or whether different chemicals in DEP modulate different on different is still A to the experimental and epidemiological data is to suggest that pollutants may cause of allergic but can not increase the of atopy, that induce the However, at in an experimental human the ability of the model pollutant DEP to cause allergic sensitization has been vivo studies in humans nasal challenge with DEP and demonstrated isotype to IgE in Four days after challenge, of from to ε from nasal lavage could be These results that the of mucosal with DEP and ragweed allergen was of in vivo isotype to IgE in humans with ragweed These results were the first demonstration of in vivo IgE isotype in humans and suggested that mucosal exposure to DEP may to In a the was used to the ability of DEP exposure to result in sensitization of humans by a mucosal IgE response. subjects were an nasal immunization with 1 mg by two nasal with of with resulted in the of and IgA in nasal however, no IgE could be in any the was on different atopic but DEP administered 24 to each IgE was in of subjects Furthermore, the response was characterized by the of a Th2-type cytokine with levels of but not in nasal lavage In subjects these levels were It is that pollutants can a into an atopic however, these studies that DEP, and similar combustion may cause with the to become to allergens to which they may not have become the increased prevalence of atopy observed in industrialized countries in recent While the ‘hygiene theory’ is both a and attractive explanation it seems clear that it is not there is probably not one studies have implicated exposure to yet these studies also show a strong role for pollutants. animal and human models in and in have shown the and mechanisms by which pollutants may modulate the immune response to increase atopic and As lifestyles become more and more similar it is that pollution will become an more important risk factor for atopy.