Here, we focused on SNPs previously described to be associated with ADEH + and located in key AD genes, namely STAT6, IFNG, IFNGR1, IRF2, and TSLP. These genes are involved in key immunological events during AD, such as IL-4 and IFNG expression and signal transduction, Th1/Th2 balance, or isotype switching 20. An impaired expression of any of these factors may affect susceptibility to viral infections 10,21–23.
rs2416259, a SNP located in an intron of TSLP, showed a deviation from the HWE in the case group (ADEH+), which is assumed to be a symptom of disease association (reviewed in 17). rs2416259 has been associated with EH and AD severity 6. Although this SNP has not been described as deviating from HWE, previous studies have not analyzed HWE separately in the different patient groups. TSLP encodes for two different isoforms, a long-form (lf) and a short-form (sf)TSLP of 63 amino acids. While lfTSLP is increased in AD lesions and enhances Th2 responses by stimulating DCs to induce Th2 cells, sfTSLP levels are decreased (reviewed in 24). Upregulation of lfTSLP and downregulation of sfTSLP might facilitate HSV-1 infections as sfTSLP acts as an antimicrobial peptide and protects against HSV-1 infections (Zeitvogel et al. manuscript in preparation).
Interferon regulatory factor 2 (IRF2) is a member of a family of transcription factors involved in the modulation of cellular responses to IFNs and viral infection. It is induced by IFNγ and acts as an antagonist of IRF1, thereby blocking IFN-mediated signaling 25. Irf2 knockout mice show a defect in Th1 cell development and develop an inflammatory skin disease with histological evidence of epidermal thickening and keratinocyte proliferation similar to AD 26. The intron variant rs111342852 was associated with a reduced IFNG expression 14, which may explain an increased susceptibility to HSV-1 infections in carriers of this variant. We analyzed two SNPs of IRF2, rs11132242 and rs1342852. While the minor allele of rs11132242 was protective, rs1342852 showed no disease association in our cohort. These two SNPs, rs11132242 and rs1342852, were previously examined in an American cohort, one of which found a significantly increased risk of ADEH + associated with the minor alleles, whereas the second study could not confirm these results 11,14.
STAT6 is a key transcription factor of the type 2 dominated inflammatory micro milieu of AD 20,27. It plays an important role in isotype switching to IgE synthesis and suppresses various antimicrobial peptides as well as IFNG expression. Both effects of activated STAT6 might increase the susceptibility to viral infections 10,21–23. Here, we investigated the genetic association of rs167769, a 5’ UTR variant of STAT6, which could affect transcript levels, mRNA translation, and, thus, STAT6 protein levels. Contrary to a previous study in an American cohort that reported the minor allele (T) at rs167769 to be protective against EH 10, in our European cohort, the minor allele was associated with an increased risk.
We found no significant associations between the investigated IFNG or IFNGR1 SNPs and ADEH + compared to ADEH-. This finding contrasts the observation made in a European-American cohort where Leung et al. reported that carriers of the IFNG haplotype AGTA of the four SNPs rs2069705, rs2069718, rs2069727, rs2430561 have a higher risk of developing EH 7. Moreover, they showed that two of these SNPs, rs2069727 (genotype CC) and rs2430561 (genotype AA), are functional, resulting in a reduced IFNG expression upon HSV-1 infection.
Although no physically linked LD exists between the three genes IFNGR1, IFNG, and STAT6, as IFNGR1 is located on chromosome 6 while STAT6 and IFNG are both on chromosome 12, the data suggest a gametic LD and epistatic effects. Epistatic effects are very likely and align with previous reports showing that the interactions between STAT6, IFNG, and IFNGR1 regulate each other's phenotypic expression and influence the development and course of AD 28–30.
In conclusion, we confirmed an association of rs2416259 (TSLP), rs167769 (STAT6), and rs11132242 (IRF2) in our European cohort for ADEH+. However, our findings are inconsistent with previous reports as the risk alleles of rs167769 and rs11132242 are contradictory. We could not confirm an association in our European cohort for several loci previously described in American subjects. In addition to the different ethnicities, the study's size and the subjects' selection could also be reasons for the different results. The small number of cases is a limitation of our study. However, in contrast to several previous studies, we carefully paired our cohort to an age-, sex- and SCORAD-matched control group and thus corrected our results for AD severity. Several SNPs have been associated with AD severity, which can lead to false positive test associations if the study did not use carefully matched control groups. Larger studies are still needed to substantiate the associations and to clarify the functional impact of disease-associated SNPs on susceptibility to HSV-1 infection in AD patients. Nevertheless, the genes TSLP, STAT6, and IRF2 were associated with EH across different cohorts and ethnicities, reinforcing these genes’ importance for EH. A better knowledge of the genetic factors predisposing to EH may allow the identification of patients at increased risk and, thus, specific patient education and even personalized early intervention.