Portal de Periódicos da CAPES

Interactions between genes and environment in the development of asthma

2001; Wiley; Volume: 56; Issue: 4 Linguagem: Inglês

10.1034/j.1398-9995.2001.00135.x

1398-9995

Cecilia María Patino, F. D. Martinez,

Pediatric health and respiratory diseases

Allergic diseases in general and asthma in particular have become an increasing problem for public health, especially in developed countries. Epidemiologic evidence suggests that the prevalence of asthma has increased significantly, especially among children (1, 2), and it is now the most frequent cause of chronic symptoms in the pediatric age (3). Some authorities questioned the validity of the first reports that suggested that the prevalence of asthma was increasing considerably in the population as a whole. The main argument was that the reported increases were due to a greater awareness among physicians and caregivers and among the population in general of the importance of asthma as a cause of chronic respiratory symptoms, especially among children (4). This was a plausible argument, because it is still true today that many children with asthma receive other diagnoses (wheezy bronchitis, spastic bronchitis, etc.). However, recent surveys that have included objective measures of risk factors for asthma, such as allergic sensitization and bronchial hyperresponsiveness (5), confirm that only a small proportion of the observed increases in the prevalence of asthma are due to a shift in diagnostic labeling. The same can be said for other diseases associated with IgE-mediated immune responses such as allergic rhinitis (6). There is widespread consensus today among the experts that these increases are real. Identifying the factors that have caused these increases and developing new preventive strategies that will address this epidemic are significant challenges for both the scientific community and public health authorities. The observed increases in the prevalence of asthma can-not be due to direct, qualitative, or quantitative changes in genetic susceptibility in the population as a whole. Genetic susceptibility is due to the presence of polymorphisms in genes that determine that certain individuals are at increased risk of developing those diseases. Several generations are needed for changes in the prev-alence of certain genetic variants to occur in the popu-lation in a way that will change genetic predisposition to conditions related to those variants. Since the changes in the prevalence of asthma and allergies have occurred in the span of not more than one or two generations, it can be concluded that these changes are due to modifications in the pattern of exposure to various environmental factors in large portions of the population. Nevertheless, the distribution of genetic variants in the population may have played a role in determining the recent changes in asthma prevalence. It is plausible to surmise that the environmental modifications that are causing the increases in asthma are more likely to exert their influences in individuals who have certain genetic backgrounds. The genetic variations (or combinations thereof) that predispose to the emerging cases of asthma surely were present in the population long before the current epidemics started. In the absence of certain environmental influences, these variations did not lead to the development of clinical illnesses in the past. In other words, complex diseases such as asthma and allergies are almost invariably the result of interactions between different sets of genes and environmental influences that, acting together with those genetic variations, make these complex diseases more likely. The above discussion has a very important corollary that needs to be taken into account in studying gene–environment interaction in the development of asthma. The environmental factors that are determining the increases in asthma prevalence may be different from those that determine the development of the same disease among individuals who would have had it even if these new environmental factors (whatever they may be) had not become more prevalent during the last decades. Moreover, the genetic factors that determine these more "endemic" forms of asthma and allergies are also likely to differ from those that determine the more "epidemic" forms. Christie et al. (7) have recently provided indirect evidence to support the contention that these new "epidemic" forms of asthma may have a different pathogenesis from that of "endemic" asthma. They found that the asthma risk associated with having at least one parent with the disease was significantly lower in their data than what had been reported in similar studies in the past. They speculated that the cases of asthma that are responsible for the increases in the prevalence of the disease may be occurring in individuals who would have been at low risk of developing the disease one or two generations earlier. The above arguments point to the complexity of the interactions that may determine asthma and allergies in different individuals. Whereas some genetic variants may increase the risk of asthma in individuals exposed to a very specific set of environmental determinants, other variants may determine the disease in other individuals who may have had a different environmental history. It is in this complex framework that one has to discuss and understand the potential role of different environmental exposures in determining the recent increases in asthma prevalence. Once it was established that these increases were real, the debate in the scientific community was centered on their causes. Given the complexities described above, it was natural to expect that no clear consensus would emerge from these discussions. Simply stated, it is most likely that more than one cause may explain the changes in prevalence observed during the last decades. The epidemiologic observations made many years ago that asthma is strongly correlated with the frequency of allergic sensitization to common aeroallergens suggested to some authors that exposure to these allergens could be an important risk factor for the inception of the disease (8). Moreover, a group of very influential researchers postulated that the recent increases in asthma prevalence could have been caused by an increase in the level of exposure to certain indoor aeroallergens (5). These authors reasoned that changes in the structure of modern homes had rendered them increasingly more impermeable to the outdoor environment, making the inhabitants of these homes significantly more exposed to indoor allergens. In addition, it was suggested that the use of certain materials for home construction and in furniture stimulated the growth of indoor allergen sources such as house-dust mites. Moreover, it was postulated that the modern tendency to keep cats and dogs inside homes could also increase the risk of sensitization to allergens produced by these pets. Since most of the epidemiologic studies of asthma have been performed in coastal areas, it was quite natural to conclude that allergens present in these areas were the "cause" of the disease in subjects who became sensitized to these allergens, and that it was thus the increased exposure to the allergens themselves that had caused the increases in asthma prevalence. This was a very influential idea, and it prompted several groups to design and conduct studies in which the main hypothesis was that, if levels of indoor allergens could be markedly decreased, sensitization to these allergens could be prevented and, consequently, asthma could also be prevented (9). Although not all the studies of this type were completed at the time of this writing, the results of those available do not confirm this hypothesis. Surveys in the Isle of Wight, England, for example, showed that children raised in environments with low exposure to allergens were less sensitized to these allergens by the age of 4 years than controls, but the prevalence of asthma at that age was not different in the two groups (10). This study has been criticized because not only were efforts made to decrease exposure to aeroallergens in the intervention group, but also measures were taken to stimulate breast-feeding and other actions that were also supposed to prevent the development of asthma. These extra measures, however, should have increased the differences between children in the intervention group and in the control group, but this was not observed. More recently, observational studies by Wahn and coworkers in Berlin, Germany, have contributed decisive new information regarding this issue. These authors measured the concentration of antigens of house-dust mites in the homes of children born and raised in that city, and, subsequently, determined the frequency with which enrolled children became sensitized to house-dust mites and developed asthma symptoms during the first 7 years of life (11). The results demonstrated that, much as in the Isle of Wight, there was a direct relation between exposure to house-dust mites and sensitization to their allergens. However, there was no relation between exposure to house-dust mites and prevalence of asthma by the age of 7 years. Not only did these results make it improbable that exposure to domestic allergens is a direct cause of asthma, but they also made implausible the hypothesis that the recent increases in asthma prevalence are due to an increase in allergen exposure. Moreover, recently published data suggest that there has not been a significant increase in indoor allergen concentrations during the last 30–40 years, in a time interval during which asthma prevalence has more than doubled (12). Finally, the observation that in inland arid regions, where indoor allergens such as those of house-dust mites are infrequent, asthma prevalence is not lower than in the coastal, more humid regions (13) has suggested that the relation between exposure to common allergens and asthma cannot be explained simply as a cause and effect association. It is interesting to stress here that in arid inland zones the allergens that have been most frequently associated with asthma are those of fungi, especially the genus Alternaria (14). In the arctic zones of Nordic countries such as Sweden, where, apparently, neither house-dust mites nor fungi are present to any large extent, the prevalence of asthma is still similar to that of the more temperate regions of those same countries (15). Up to 40% of school-aged asthmatic children are sensitized to common indoor aeroallergens (such as those of cats and dogs) in these areas, a prevalence that is significantly lower than that observed in regions where up to 80–90% of all asthmatic children are sensitized. Interestingly, in these arctic regions of Sweden, circulating IgE levels were significantly higher among skin-test-negative children with asthma than among skin-test-negative children without asthma (15). All this evidence suggests a different explanation for the increases in the prevalence of asthma. It has been proposed that childhood asthma is not caused by exposure or sensitization to any specific allergen (16). Nobody can deny that, in individuals who are sensitized to certain allergens, exposure to these allergens can trigger asthma symptoms. However, the fact that asthmatic individuals become sensitized to the allergens that their immune system is in contact with in the environment where these individuals were raised suggests that future asthmatics have an alteration in the immune response that is not limited to any specific antigen. Recent data on the characteristics of the immune response during the first years of life have led to new arguments in support of this hypothesis. Longitudinal studies have shown that the development of predominant IgE-mediated responses to local aeroallergens occurs early in life in future asthmatics (17). Moreover, infants who will become sensitized to these aeroallergens show important alterations in immune responsiveness during the first 12 months of life, long before specific IgE against these aeroallergens can be detected in the circulation. These alterations consist of a generalized impairment of cytokine responses to nonspecific stimuli by peripheral blood mononuclear cells (18). Although these alterations affect both cytokines of the T-helper 1 type (Th1) and those of the T-helper 2 type (Th2), it is mainly the former, and especially interferon-gamma (IFN-γ) responses, that are significantly decreased (19). As is well known, IFN-γ inhibits the production of IL-4 and IL-13 by Th2-type cells. These two interleukins are the only molecules capable of providing the signal to B cells to produce IgE. These data would suggest, therefore, that genetic and environmental factors interact in infants and young children to establish the patterns of immune responsiveness that predispose to the development of asthma. It is now well established that genetic determinants play an important role in the susceptibility to the development of asthma. Studies in identical twins have convincingly demonstrated that at least 50% of the susceptibility to asthma is determined by inherited predisposition (20). It has also been clearly established that no single gene or even a small number of genes exert a decisive influence in determining this susceptibility. Asthma is essentially a polygenic disease in which many genetic variants determine small changes in immune responses or in the manner in which the airway responds to the environment (21). As argued earlier, it is quite likely that different sets of genetic variants may interact with different environmental determinants to determine the risk of asthma in different individuals. This is the main source of difficulty when studying the genetic and environmental determinants of asthma: the recent changes in asthma prevalence are most likely due to many different environmental changes, each exerting its own small influence, and these changes may have been active at different ages and in different people. Moreover, environmental factors also interact among themselves and may determine a particular phenotype only if they are present in a certain context. This is what has been called context dependency in the determination of asthma risk (22). It is very difficult to measure simultaneously all the environmental influences that can interact to determine asthma risk, and particularly to measure them at the time in which they are active. It is thus very cumbersome (if possible at all) to design studies of risk factors for asthma that are based exclusively on laborious measurements of suspected environmental factors. The extreme complexity of the environmental determinants of a disease such as asthma has renewed hope that the study of the genetics of the disease may make an important contribution to our understanding of its pathogenesis. By definition, genetic factors are fixed for each individual; therefore, they can be measured with great reliability at any time during the life span. By anchoring studies of the role of environmental determinants of asthma on precise sets of genetic polymorphisms, it may be possible to determine which of these determinants are important in well-defined groups of subjects. We propose a general paradigm that may be useful to all those who are interested in elucidating what causes asthma and other complex diseases such as hypertension, myocardial infarction, diabetes, etc. This paradigm is based on the search for clues in two different and complementary areas. First, we need to use the extraordinary advances in modern genetic technology. Anonymous genetic markers are now available that span the whole human genome, and these markers will allow us to identify the presence of asthma-related genetic variations with an ever increasing degree of precision. Moreover, it is currently not even necessary to know beforehand which genes may be present in areas of the genome where linkage signals reveal the existence of asthma-related genetic variants. There are already several linkage studies of this type in the literature in which different definitions of asthma were used. None of these studies have revealed the presence of major genes for asthma, that is, of genetic variants that explain a substantial proportion of the susceptibility to the disease. However, it is important to stress here that most of these studies have included a relatively small number of sibpairs with asthma, and this is particularly relevant because, as we said earlier, studies of the familial segregation of asthma had already shown it to be a polygenic disease. Frequently, very broad definitions of the disease have been utilized, and it quite likely that, within these definitions, subjects with different forms of asthma were included under the same label. Thus, sibpairs that were supposed to be affected by the same disease may have included individuals whose asthma was determined by different sets of genes. This would obviously dilute the results of any linkage study. Therefore, the levels of statistical significance of most genome-wide linkage studies of asthma have been relatively low. In spite of these difficulties, a recent meta-analysis performed by Morton and coworkers concluded that there is sufficient evidence that genetic variants associated with asthma and asthma-related traits may be present in chromosomes 6, 5, 16, 11, 12, 13, 14, 7, 20, and 10, in that same order from the strongest to the weakest (23). There is little doubt that we would significantly enhance our understanding of the causes of asthma if we could find the genetic polymorphisms responsible for these linkage signals, and if we could determine the alterations in the genes, or in the proteins codified by these genes, that are induced by these variations. Recently, for example, Graves et al. were able to identify several polymorphisms in the gene for IL-13, located in chromosome 5q (24), and showed these polymorphisms to be associated with an increased risk of having high levels of IgE (25), while others showed that these same polymorphisms were associated with an increased risk of asthma (26). The IL-13 gene is located in close proximity to genetic markers that showed evidence for linkage with circulating levels of IgE in one of the first linkage studies published (by Marsh et al. [27]). Several groups are now attempting to determine the way in which the variants discovered in the IL-13 gene influence the synthesis of IgE and possibly bronchial hyperresponsiveness, especially considering that several experimental studies in animals suggest that IL-13 can directly influence the reactivity of airway smooth muscle (28). Although studies such as those described above may be very useful for our understanding of the pathogenesis of asthma and for the development of new therapeutic instruments, it is unlikely that these studies will explain the increases in the prevalence of asthma or help in the design of new preventive strategies. This conclusion is based on the fact that, on the one hand, genetic variants associated with complex diseases are very common in the population, and it is very likely that they play a role in the complex balances that determine homeostasis. This is true for most common diseases, and it is thus unlikely that gene therapy may be proposed in the near future as a realistic tool for the prevention and treatment of these diseases. On the other hand, no single gene is responsible for the expression of asthma, and it will thus be necessary to study simultaneously genetic variants that increase asthma risk and environmental factors that interact with these variants to determine the asthmatic phenotype. As explained earlier, the final objective of this approach will be to identify specific environmental measures that could be targeted on those subjects who are susceptible to these environmental factors and who have been identified by genetic markers. Several studies published in recent years have identified environmental factors that appear to protect against the development of asthma. These studies may be the basis for promising new strategies that could be based on the paradigm described in the previous paragraph. The relation between asthma and infection has been the matter of much discussion and debate during the last decades. It was thought for many years that viral infections could be one of the causal factors in the development of atopic asthma. This idea was based on the assumption that, since many children who were infected with the respiratory syncytial virus (RSV) developed subsequent recurrent wheezing illnesses, RSV itself might have damaged the lung or in some way altered the immunologic response, thus creating the conditions for the development of asthma. However, recent epidemiologic studies have helped to clarify the relation between infection by RSV and asthma (29). These studies have shown that there is indeed an increased risk of having asthma between the ages of 6 and 11 years in children who had RSV in infancy, but this increased risk is independent of allergic sensitization and tends to decrease with age, becoming nonsignificant during early puberty. Moreover, children who had lower rates of respiratory illnesses associated with RSV infection were not more likely to become sensitized to local aeroallergens than their peers who did not have such illnesses (29). Our concept of the relation between asthma and infection during the first years of life started changing with studies performed in Germany by von Mutius et al. (30). These studies compared the prevalence of asthma and asthma-related phenotypes in children born and raised in two zones of Germany before the fall of the Berlin Wall. The results of these studies are well known: children from East Germany had a higher prevalence of "bronchitis" but significantly less asthma, allergic sensitization, and bronchial hyperreactivity than children in West Germany. Similar studies performed in other countries of Eastern Europe during the same period confirmed these results (31). There has been considerable speculation as to the possible causes of these differences. However, one characteristic that is particularly important in the framework of this discussion is the massive participation of infants in the formerly socialist Eastern European countries in day-care programs (30). These programs allowed the great majority of women in those countries to return to work shortly after the birth of their children. It is well known that contact with other children, particularly in large day-care settings, is associated with a significant increase in the incidence of acute respiratory illnesses (32). More recent studies performed in Western countries have confirmed that children that are taken to day-care facilities during the first months of life do have a greater incidence of acute lower respiratory illnesses during those first months, but they also have a significantly lower risk of having asthma and sensitization to allergens during the school years (33, 34). A decrease in the risk of having asthma and allergic sensitization has also been observed in children who have older siblings at home (34-36). Although it is reasonable to surmise that, both in the case of day-care settings and in that of the presence of other children at home, the common "protective" factor was a greater exposure to infectious agents, the specific mechanism by which such exposure could prevent the development of allergies was not well understood until very recently. A new, very valuable source of information has been offered by the finding that, in four different rural areas of Europe (37-40) and in Canada (41), children who live in direct contact with farm animals have a significantly lower risk of allergic sensitization and asthma than children who live in these same zones but without direct contact with animals. More detailed studies showed that one of the characteristics of children who lived in contact with farm animals, and that distinguished them from those who did not, was a much higher exposure to endotoxin (36). Endotoxin, a component of the wall of Gram-negative bacteria, is very abundant in nature, and it is a very accurate indicator of the degree of cleanliness in indoor environments in urban areas (42). Gereda et al. (43) recently showed an inverse relation between the level of indoor endotoxin and the likelihood of sensitization to allergens during the first 24 months of life in infants born and raised in Denver (USA). These authors also showed that there was a direct relation between exposure to endotoxin and the proportion of T-helper cells producing IFN-γ. Concomitantly with reports of the results of studies on the prevalence of asthma in rural zones of Europe, other studies suggested that exposure to domestic animals during infancy was associated with a decreased likelihood of allergic sensitization and asthma (44-46). Although, here again, there could be many explanations for these findings, it is interesting to stress that one of the factors that influences the level of indoor endotoxin is the presence of pets at home (47). These studies thus provide a novel scenario that could explain some of the gene–environment interactions that are important at the beginning of asthma. It is known that endotoxin is a potent inducer of the production of IL-12 by antigen-presenting cells (48), while IL-12 is a strong signal for the deviation of the immune response toward the production of IFN-γ. Studies in experimental animals have shown that, if exposure to endotoxin occurs immediately before or shortly after exposure to an allergen, the IgE-mediated response that usually ensues after exposure to that allergen is very efficiently blocked (49). However, if exposure to endotoxin occurs several days after exposure to the allergen, the latter is associated with a potentiation of the IgE-mediated response. This finding may explain why exposure to endotoxin increases bronchial reactivity in asthmatics (50), and suggests that the response to endotoxin exposure depends on the context in which the exposure occurs, as we explained earlier. The finding that exposure to endotoxin during the first years of life can be an important factor protecting against allergic sensitization and asthma offers a very attractive paradigm for the type of gene–environment interaction studies that we suggested earlier will be crucial for our understanding of the pathogenesis of asthma. Most immune systems in vertebrates have a receptor structure that is exquisitely sensitive to endotoxin. This system is centered on CD14, which is present both in circulation (sCD14) and on the surface of macrophages and mononuclear immune cells (mCD14). Animals in which genetic expression of CD14 has been blocked become insensitive to endotoxin, to the point that these animals are completely resistant to massive doses of endotoxin that are usually associated with 100% mortality in wild-type animals (51). Martinez & Holt postulated that, if exposure to endotoxin is important as a factor protecting against sensitization to allergens, a natural corollary would be that a genetic variant in the CD14 gene could have a role in modulating the immune response (48). A search for polymorphisms was thus conducted in the CD14 gene, located in chromosome 5q, very close to the IL-13 gene and also close to markers that have been shown to be linked to serum IgE levels (27, 52). A variant (a C to T substitution) in the 5′ regulatory region of the gene, at position −159 from the transcription initiation site, was originally reported by Baldini et al. (53). These authors showed that the T allele of this polymorphism was associated with a higher sCD14 level, with lower serum IgE concentrations, and with a larger number of positive skin tests to aeroallergens in atopic subjects. At least two other reports have confirmed these findings, although not all groups have been able to do so (54-57). This should not be surprising because, as we said earlier, the expression of environmental and genetic determinants of complex diseases depends on the context in which these diseases occur. However, the discovery of genetic variants in the receptor system for endotoxin offers a very promising scenario for the design of studies in which innocuous surrogates of endotoxin could be used to prevent the development of asthma in persons who, because they have a certain genetic background, could be more susceptible to exposure to endotoxins or to its innocuous surrogates. Recent advances in our understanding of the gene– environment interactions that determine susceptibility to asthma offer very promising clues to the reasons why asthma has increased dramatically during the last decades. They also offer the blueprint for the development of strategies for the primary prevention of a disease that dramatically affects the quality of life and the well-being of so many adults and children all over the world. This research was supported by National Heart, Lung, and Blood Institute grants HL 66447, HL 56177, HL 61892, HL 64307, and HL 66800.