Filaggrin gene mutations and new SNPs in asthmatic patients: a cross-sectional study in a Spanish population

In this study, the R501X and 2282del4 mutations were not identified as being associated with any of the patients’ clinical and biological characteristics. It is noteworthy that the allele frequencies of these mutations in our population were lower than those obtained in European population studies [10, 21]. This could be due to the inclusion criteria of the study, but may also indicate particular aspects of the Spanish population.

In this study we identified for the first time three polymorphisms that had not been previously described in the filaggrin gene (1594C  T, 1607T  A and 1747C  A). It is very interesting to note that one of the polymorphisms identified in this work, 1741A  T, showed an association with non-allergic asthma and with a greater severity of it. In our series, none of the patients with personal history of atopic dermatitis were carriers of any of the 12 identified SNPs.

Mutations in the filaggrin gene have shown great importance in ichthyosis vulgaris [8] and atopic dermatitis [9–19], but no clear relationship to asthma has been deduced [10–26]. There is a considerable controversy about the possible association with asthma and it is unclear whether the association detected in some studies is due to the presence of asthma or the influence of eczema or atopic dermatitis in individuals who have asthma associated with these skin diseases, as two meta-analyses concluded [27, 28].

We found statistically significant differences in SNP 1741A  T when analysing the spirometry values: patients carrying the T allele presented lower FVC, FEV1 and PEF values than those carrying the A allele. We also found statistically significant differences when analysing the treatment step required by patients according to GINA guidelines, requiring a step higher than 3 the patients carrying the T allele. Some mutations in the FLG gene have been associated with the presence of asthma (in particular allergic asthma and asthma with atopic dermatitis). Despite having been described in the databases this SNP has not been studied in populations of patients with atopic dermatitis or asthma. Interestingly, in our population it was associated to non-allergic asthma. Gene expression studies indexed in GeneCards^® (http://www.genecards.org) show that the FLG gene expression was observed in lung and bronchial mucosa, despite of the lack of protein detection in such mucosa [7]. In addition, filaggrin is expressed in oral mucosa, where it is assumed that it also contributes to the barrier function [3, 4]. It has also been shown to be expressed in the nasal vestibule up to the transitional epithelium [5] and, very recently, in the nasal mucosa [6]. The link between upper and lower airways and the relation between asthma and nasal polyposis are well known [35] and this could be related to the association found between 1741A  T SNP and non-allergic asthma, as 2 of the 4 non-allergic asthmatic patients carrying the SNP had chronic rhinosinusitis. Nevertheless, in our study none of them had a personal history of nasal polyps, although this entity is frequently infradiagnosed. The epithelium plays an important role in the inception and development of asthma, particularly in non-allergic asthma, through the activation of type 2 innate lymphoid cells (ILC2) [36] and SNPs in filaggrin could compromise skin barrier function, increase penetration of allergens, reduce inflammatory threshold levels to skin irritants and allergens and increase the risk of sensitization [37]. It is tempting to speculate that microbes, pollutants, tobacco smoke or other factors could easier cross the epithelial barrier and induce the inflammatory process in allergic asthma but maybe also in non-allergic asthma.

This SNP generates a change from threonine to serine in the filaggrin protein. While the predicted level of pathogenicity for this SNP is not very high, we have identified that this change is found in a pattern of myristoylation of the protein. Myristoylation of proteins occurs when there is a transfer of a myristoyl group to the glycine amino group of a specific protein sequence, using Myr-CoA as substrate. It is assumed that myristoylation is an irreversible process that occurs co-translationally. In fact, myristoylation of proteins increases their hydrophobicity and thus could have a functional impact not described to date. More analysis of the protein produced in patients carrying the T allele are needed.

Nine other polymorphisms have been described; three of them (1594C  T, 1607T  A and 1747C  A) have been described for the first time in this study. Nevertheless, no clinical associations were identified, perhaps due to the low frequency of occurrence, at least in our population. Considering the scarce representation of FLG mutant alleles a combined variable using all the nonsynonymous SNPs was created. This approach has been used in the field of FLG mutations due to their low frequency. A significant association was detected between carrying mutant alleles and worse respiratory function values in non-allergic asthma.

We think it is important to point out that our study was prepared from the outset to assess the influence of these mutations and SNPs in the filaggrin gene in asthma and potential importance of atopy in this association. Therefore, we included asthmatic patients divided into two groups with well-differentiated AA and NA phenotype and solid clinical classifications, and a control group rigorously selected without personal or family history of atopy. Our results are consistent with several studies in other populations that question the association of mutations in the FLG gene and asthma [10, 14, 17, 20], but interestingly in our study the association seems to be with non-allergic asthma without eczema or atopic dermatitis. In this sense, it could be interesting to interrogate possible differences regarding filaggrin expression in the bronchial mucosa between allergic and non allergic asthma to assess implication of SNPs in the filaggrin gene.

In our population, we detected no association for any of the variants of the FLG gene analysed with atopic dermatitis or with any other clinical variable, possibly because our patients, unlike those of other studies, were selected due to presenting asthma and not dermatitis as initial clinical variable, making them a distinct population. In this respect this result would indicate the existence of distinct genetic associations for both diseases and again could explain the contradictory results obtained in the literature.

In conclusion, in our population we found no significant differences with null-mutations R501X and 2282del4, but we did find significant differences with SNP previously described in the FLG gene but which has not been studied in patient populations. In addition, we have first identified three polymorphisms that had not been previously described in FLG gene (1594C  T, 1607T  A and 1747C  A).