NEW YORK (GenomeWeb) – A genome-wide comparative analysis by researchers in Germany, Ireland, and elsewhere has found that while atopic dermatitis and psoriasis share certain associated loci, the alleles at those loci have opposing effects on the disorders.
As the researchers reported in the American Journal of Human Genetics this week, they pulled together genome-wide association and ImmunoChip data from more than 19,000 people into a large meta-analysis. Based on this meta-analysis, the researchers found half a dozen such opposing risk alleles at shared loci as well as independent disease-specific risk alleles. A number of these loci could be traced to the epidermal differentiation complex, the Th2 locus control region, and the major histocompatibility complex.
"Our results show that atopic dermatitis and psoriasis have distinct genetic mechanisms with opposing effects in shared pathways influencing epidermal differentiation and immune response," the University of Dundee's Sarah Brown and her colleagues wrote in AJHG.
Atopic dermatitis — a type of eczema —and psoriasis are both common chronic inflammatory skin conditions. According to the US National Institute of Arthritis and Musculoskeletal and Skin Diseases, atopic dermatitis affects between 9 percent and 30 percent of people in the US, typically beginning in infancy or childhood, while psoriasis affects about 3 percent of the US population.
The two disorders, though, rarely co-occur, the researchers noted.
For this study, Brown and her colleagues amassed genome-wide genotyping data for some 2,260 atopic dermatitis cases, 4,490 psoriasis cases, and more than 12,300 controls from six case-control cohorts from Germany, Ireland, the UK, and the US. They also collected ImmunoChip data for nearly 2,500 atopic dermatitis cases, 3,600 psoriasis cases, and more than 9,000 controls.
They performed a meta-GWAS on each disease and turned to two different approaches — a compare and contrast meta-analysis approach and a transethnic meta-analysis using the MANTRA software — to filter the SNVs and model their effects.
Excluding the MHC region at first, the researchers homed in on some 3,000 SNVs uncovered using both the CCMA and MANTRA approaches, which they then subjected to multinomial regression modeling. For their further analyses, the researchers focused on the loci harboring more than one SNV and whose effect was measured to be in the same direction by both CCMA and multinomial regression modeling.
In their disease-specific meta-analyses, the researchers were able to replicate 14 of the 16 previously identified European loci linked to atopic dermatitis and 43 of the 44 previously identified loci linked to psoriasis in Europeans.
Outside of the MHC region, Brown and her colleagues identified 25 loci that had a genome-wide significant association with either atopic dermatitis or psoriasis, as judged by all three analysis methods. Each co-associated locus had opposing effects, and two of them — located near ANXA6/TNIP1 and PRKRA — hadn't been linked before with both the conditions, the researchers said.
For example, the ANXA6 gene, which is downstream of the rs17728338 locus the researchers identified, had been linked before to psoriasis in European and Chinese populations, though not to atopic dermatitis. This gene encodes a calcium-dependent membrane and phospholipid binding protein, and it is upregulated in atopic skin as compared to controls, as well as in affected skin as compared to unaffected skin. At the same time, it is decreased in psoriasis lesions as compared to healthy skin.
"Clearly, further fine mapping is necessary to identify the causal variant that exerts opposing effects on AD and psoriasis, but we speculate that ANXA6 might be a switch-point differentiating AD from psoriasis that reflects the importance of calcium-dependent effects in keratinocyte differentiation," the researchers said.
Brown and her colleagues also found seven linkage disequilibrium blocks with either disease-specific or opposing signals in the epidermal differentiation region, including an atopic dermatitis-specific locus at FLG, a psoriasis-specific locus at LCE3B-LCE3C, and a locus with opposing effects at RPTN/HRNR/FLG-AS1.
FLG-AS1 expression is increased in psoriasis lesions as compared to healthy skin and decreased in atopic dermatitis lesions, the researchers added.
Meanwhile, in the Th2 locus control region, the researchers found three independent loci — IL13, KIF3A, and SLC22A4 — linked to atopic dermatitis risk, though with no effect on psoriasis risk. A fourth locus with opposing effects mapped to RAD50 and is linked with an increased risk of atopic dermatitis, while being protective against psoriasis. RAD50 mRNA is expressed at statistically significantly higher levels in psoriasis lesions, and, the researchers noted, trended toward reduced expression in atopic dermatitis lesions.
Because of its complexity, the researchers analyzed the MHC region separately. Here, the strongest association they found tagged a well-known psoriasis risk allele, HLA-C*06:02. Conditional analysis also found additional psoriasis risk loci at MICA and HLA-A as well as loci with opposing effects at HLA-C and HLA-DRB1.
Most of the opposing-effect loci the researchers uncovered are linked to pathways involved in adaptive immune functions, they noted.
"This raises the intriguing possibility that the same biological mechanisms might act differentially on AD versus psoriasis," they added.