© 2006 American Thoracic Society doi: 10.1164/rccm.2604006
A Glitch in the Switch?Of Endotoxin, CD14, and AllergyUniversity Children's Hospital, Ludwig Maximilians University Munich, Germany Genetic susceptibility and environmental factors cause complex diseases. What is now common knowledge had been suggested some decades ago when early segregation analyses had hinted toward a gene-by-environment interaction in the development of asthma and allergies. Thus, it was hypothesized that certain environmental factors may induce disease in individuals especially susceptible to these stimuli due to their genetic make-up. But where in the development of allergies might these gene-by-environment interactions occur and how? A number of hypotheses have been put forward, and some of these are more obvious than others. It was suggested that air pollution, and specifically active and passive smoke exposure, may have even more devastating effects in people lacking certain defense mechanisms and enzymes against toxins due to genetic variation. These interactions, both statistically significant and biologically plausible, were found repeatedly in different populations and studies (e.g., for the interaction between tobacco smoke exposure and the lack of glutathione S transferases in respiratory diseases) (1–4). Even more intriguing (but less obvious), protection against the development of asthma and allergy may work through gene-by-environment interactions. In this issue of the Journal (pp. 386–392), Simpson and colleagues (5) investigate such a potential interaction from the area of innate immunity. CD14 is a premium candidate for gene-by-environment studies as it is a key player in innate immune mechanisms, putatively interacting with a wide range of pathogens, including endotoxin. Also, a functional polymorphism in the promoter region of the gene has been identified, which leads to changes in transcription factor binding (6) as well as attenuated expression of soluble CD14 (sCD14) (7). CD14 is predominantly expressed on monocytes and neutrophils, but it is not known if the CD14 promoter polymorphism also affects the expression of the cell-bound molecule. Numerous genetic association studies have been performed trying to replicate the initial findings that the T allele at position –159 (alternatively numbered –260) of the CD14 promoter correlates with decreased total and specific serum IgE levels (7). However, replication results were conflicting. The same (T) allele that was associated with decreased serum IgE levels in the initial study, and in other studies (8) was also found to be linked to more allergy and higher serum IgE (9). A pooled analysis (10), as well as our own large cross-sectional study in approximately 4,000 individuals, came to the conclusion that the CD14 promoter polymorphism may not influence IgE levels or atopy (11). Alternative explanations for these diverse findings were suggested, known as the "endotoxin switch hypothesis" (12). Exposure to a certain antigen concentration may result in Th1 (antigen tolerance) or Th2 (allergic sensitization) responses depending on the amount of concomitant microbial stimulation. The increased potential to produce CD14 due to the promoter T allele at position –159 shifts the response to microbial stimuli from Th2 toward Th1, so that even low levels of microbial stimuli may be protective. In this scenario, the T allele would protect against allergy. According to the hypothesis, however, at very high microbial exposure levels, a second switch would occur from Th2 toward Th1 and again earlier in carriers of the T allele (see also Figures 1 and 2 in Reference 12). Thus, in the setting of high microbial exposure, carriers of the T allele would have more allergies, as observed in the traditionally farming Hutterites (9). The results presented by Simpson and colleagues in this issue seem to confirm the hypothesis. Also in this study from Manchester, the T allele is associated with more atopic sensitization and more atopic eczema at high levels of endotoxin exposure compared with the C allele. The same kind of association has been observed in a recently published article of a study of farmers' and nonfarmers' children from rural alpine areas (13). However, although that report found different effects of the CD14 polymorphism (in accordance with the endotoxin switch hypothesis) at the extreme ends of endotoxin exposure, Simpson and colleagues' data suggest that these diverse effects of the CD14 promoter polymorphism may already be observed in the common range of microbial exposure found in children living in urban areas. Indeed, when absolute levels of endotoxin measured in the beds of children are compared between these two studies, almost all children of the Manchester study fall in the lowest tertile of the ALEX study (13), rather challenging than replicating the results by Eder and colleagues (13). Although the endotoxin switch hypothesis is intriguing, a number of open questions remain: The T allele at –159 increases the transcription of the gene in vitro as shown by luciferase assays (6) and the expression of soluble CD14 in vivo in epidemiologic studies (7, 11). Also, in the study by Simpson and colleagues, carriers of the T allele had increased sCD14 levels, but the study could not find an association between increased sCD14 levels and decreased IgE levels. Thus, the question of how the –159 polymorphisms, if not by CD14 expression, may influence IgE levels remains unanswered. A further open question is which ligand interacts with CD14 to induce this protective effect, as endotoxin may not be the causative agent for the protective effect but only a surrogate for other, yet unknown immuno modulators (14). So far, the CD14 data are suggestive of a gene-by-environment interaction along the lines of the switch hypothesis, but this has not yet been proven biologically and may not fit all aspects of epidemiologic results on CD14. Thus, the work presented in this issue may be an important trigger for further research into the area of gene-by-environment interaction where large-scale studies, advanced assessment of environmental exposure, and experimental investigations of interactions are needed. Only then will valid public health recommendations seem possible. FOOTNOTES Conflict of Interest Statement: M.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
