Candidate gene study of TRAIL and TRAIL receptors: association with response to interferon beta therapy in multiple sclerosis patients

Abstract

TRAIL and TRAIL Receptor genes have been implicated in Multiple Sclerosis pathology as well as in the response to IFN beta therapy. The objective of our study was to evaluate the association of these genes in relation to the age at disease onset (AAO) and to the clinical response upon IFN beta treatment in Spanish MS patients. We carried out a candidate gene study of TRAIL, TRAILR-1, TRAILR-2, TRAILR-3 and TRAILR-4 genes. A total of 54 SNPs were analysed in 509 MS patients under IFN beta treatment, and an additional cohort of 226 MS patients was used to validate the results. Associations of rs1047275 in TRAILR-2 and rs7011559 in TRAILR-4 genes with AAO under an additive model did not withstand Bonferroni correction. In contrast, patients with the TRAILR-1 rs20576-CC genotype showed a better clinical response to IFN beta therapy compared with patients carrying the A-allele (recessive model: p = 8.88×10(-4), pc = 0.048, OR = 0.30). This SNP resulted in a non synonymous substitution of Glutamic acid to Alanine in position 228 (E228A), a change previously associated with susceptibility to different cancer types and risk of metastases, suggesting a lack of functionality of TRAILR-1. In order to unravel how this amino acid change in TRAILR-1 would affect to death signal, we performed a molecular modelling with both alleles. Neither TRAIL binding sites in the receptor nor the expression levels of TRAILR-1 in peripheral blood mononuclear cell subsets (monocytes, CD4+ and CD8+ T cells) were modified, suggesting that this SNP may be altering the death signal by some other mechanism. These findings show a role for TRAILR-1 gene variations in the clinical outcome of IFN beta therapy that might have relevance as a biomarker to predict the response to IFN beta in MS.

Figure 1. Alignment and secondary structure homology between TRAILR-2 and TRAILR-1.

The high conservation in the amino acid sequence and secondary structure of both receptors has allowed us to use a slightly modified structure of TRAILR-2 in the molecular modelling.

Figure 2. Molecular dynamics simulations of Glu228Ala SNP in TRAILR-1.

A and C corresponds to the superimposed structures of the 50s and 90s loops respectively in the initial (green) and final (orange) model of the wild type of TRAILR-1 (Glutamatic acid at position 228). B and D corresponds to the superimposed structures of the loops in the initial (green) and final (brown) model of the mutant type of TRAILR-1 (Alanine at position 228).

Figure 3. TRAILR-1 expression levels among Responders and Non Responders to IFN beta therapy, in different PBMC subsets.

Expression level is represented as relative expression compared to the reference gene GAPDH, using the ΔΔCt method. TRAILR-1 expression levels between responder and non responder patients to IFN beta therapy in each of the assessed cell subsets were compared by a Mann Whitney test.

Figure 4. TRAILR-1 expression levels in MS patients stratified by the rs20576 genotype, in different PBMC subsets.

Expression level is represented as relative expression compared to the reference gene GAPDH, using the ΔΔCt method. TRAILR-1 expression levels between patients with A/− and CC genotypes were compared by a Mann Whitney test.