No associations of the mineralocorticoid and glucocorticoid receptor genes with asthma

To the Editors:

Asthma is a multifactorial disease. Although a number of host and environmental risk factors have been identified in the past decades, these cannot fully explain the prevalence of asthma. Previous studies have shown that psychosocial stress confers a risk for the development of asthma [1].

Stress activates the hypothalamus–pituitary–adrenal (HPA) axis, which leads to secretion of cortisol. Cortisol in turn influences the activity of many systems in the human body and shifts, among others, the T-helper type (Th1/Th2) balance of the peripheral blood mononuclear cells towards a predominantly type 2 response. Cortisol binds to the high affinity mineralocorticoid receptor (NR3C2) and the low affinity glucocorticoid receptor (NR3C1). Single nucleotide polymorphisms (SNPs) in these receptors have been associated with basal cortisol and cortisol responses to stress [2]. So far, two studies have found an association between SNPs in NR3C1 (i.e. rs6195, rs41423247) and asthma development [3, 4]. However, these results have not been replicated in other populations so far. Additionally, it is unknown whether other functional or tagging SNPs in the NR3C1 are associated with asthma development. In addition, SNPs in the NR3C2 have not been studied before in relation to asthma development. Also it is yet unclear whether exposure to psychosocial stress modifies the effect of SNPs in NR3C2 or NR3C1 on asthma.

We investigated the associations of SNPs in NR3C2 or NR3C1 with asthma in adolescents from the general population and attempted to replicate our findings in an asthma case–control study.

In adolescents from the general population, we have previously shown that exposure to perinatal stress more than doubles the risk of asthma development [5]. Therefore, we tested additionally whether exposure to perinatal stress modifies the effect of SNPs in NR3C2 or NR3C1 on asthma in adolescents.

We genotyped two functional SNPs in NR3C2 (rs5522, rs2070951) and 12 functional or tagging SNPs in NR3C1 (rs4912903, rs6198, rs6196, rs258813, rs33388, rs17100236, rs10482642, rs2963155, rs41423247, rs9324924, rs4244032, rs4607376) in 1,454 adolescents of the prospective TRacking Adolescents' Individual Lives Survey (TRAILS) cohort (48% males; mean±sd age at survey 1, 11±0.6 yrs; survey 2, 14±0.5 yrs; survey 3, 16±0.7 yrs) [6] and in 998 individuals from the Dutch Asthma GWAS study (44% males; mean±sd age 42±12.2 yrs) [7], that acted as replication study. In TRAILS, data on parentally reported asthma was collected at age 11, 14 and 16 yrs via self-reported questionnaires [8]. Asthma was defined as a doctor diagnosis of asthma, and/or symptoms of asthma and/or asthma treatment prescribed by a physician in the past 12 months, before or at the age of 16 years (n=141; 10%). Adolescents not meeting these criteria were defined as not having asthma (n=1,293; 89%). Information about asthma was missing in 20 adolescents (1%). The replication study consisted of 529 asthmatics and 469 nonasthmatics; asthma was defined as a doctor's diagnosis of asthma, the presence of asthma symptoms and bronchial hyperresponsiveness [7]. In TRAILS, perinatal stress was defined as in utero exposure to the mother's self-reported maternal psychological problems and/or self-reported maternal postnatal depression (n=67 adolescents; 5%) [9].

SNPs were analysed in an additive genetic model for the effect on asthma using logistic regression models. Interactions between genotype and perinatal stress were tested in TRAILS by introduction of perinatal stress and the interaction term of genotype times perinatal stress into the model.

None of the SNPs in NR3C2 or NR3C1 were significantly associated with asthma in adolescents. We also found no association between these SNPs in NR3C2 or NR3C1 and asthma in the replication study (table 1).

We observed no effect of perinatal stress on the association between SNPs in NR3C2 or NR3C1 and asthma in adolescents (i.e. no interaction between perinatal stress and SNPs).

Previous studies have shown that SNPs in NR3C1 (i.e. rs6195, rs41423247) were associated with asthma [3, 4]. Our results do not support the findings of these previous studies, specifically not those suggesting that a mutation (G/C) within rs41423247 in NR3C1 would have a protective effect on asthma development [4]. However, this association was shown in merely one study including only 59 adults with asthma and 70 healthy adults. Since we found neither an association between rs41423247 and asthma in 1,434 adolescents (OR 0.99, 95% CI 0.76–1.28) nor in 998 adult individuals (OR 0.96, 95% CI 0.80–1.15), we feel that the findings from the study of Pietras et al. [4] may be due to chance.

Although we found no association between SNPs in the glucocorticoid or mineralocorticoid receptor and asthma, this does not exclude a role of glucocorticoid or mineralocorticoid receptor activity in asthma development, since this activity is influenced by multiple other mechanisms, such as chaperone proteins that modulate the translocation of these receptors and post-translation changes that modify the conformation of these receptors. In addition, methylation of both receptors can also influence whether cortisol levels are associated with inflammatory processes [10].

In conclusion, we show suggestive evidence that SNPs in NR3C2 or NR3C1 are not associated with asthma development. We also find no indications that perinatal stress would affect the role of these genes on asthma development. These results indicate that, while these genes are biologically plausible candidate genes for asthma development in relation to perinatal stress, they cannot be linked to the disease.

We are grateful to all adolescents, their parents and teachers who participated in this research and to everyone who worked on this project and made it possible.

• ©ERS 2012