Exploratory analysis of four polymorphisms in human GGH and FPGS genes and their effect in methotrexate-treated rheumatoid arthritis patients

Abstract

The enzyme folylpoly-gamma-glutamase synthethase (FPGS) plays an important role in the intracellular polyglutamation of the disease-modifying antirheumatic drug methotrexate (MTX) and the length of the polyglutamated MTX product correlates with the time that MTX resides in the cell. The glutamates are released from MTX by activity of the enzyme gamma-glutamyl-hydrolase (GGH), thereby allowing the efflux of MTX. GGH 452C>T has been associated with decreased catalytic activity and higher accumulation of long-chain MTX-polyglutamate. However, single nucleotide polymorphisms (SNPs) in FPGS and GGH genes have not yet been explored for association with MTX efficacy or toxicity. We selected for SNPs with frequencies higher than 10% or, in case of FPGS 114G>A, causing an amino acid change with no known frequencies. In this study, frequencies of two SNPs in FPGS (1994A>G and 114G>A, rs10106 and rs10760502, respectively) and GGH genes (452C>T and 16T>C, rs11545078 and rs1800909, respectively), were determined using a newly developed method in rheumatoid arthritis patients (n = 352) and in a group of healthy controls (n = 360). Next, the SNPs were associated with response to MTX in rheumatoid arthritis patients treated with MTX monotherapy. In rheumatoid arthritis patients, allele frequencies of FPGS 1994A>G were 0.534 (A) and 0.466 (G), and for FPGS 114G>A 0.714 (G) and 0.286 (A). Allele frequencies of GGH 16T>C were 0.737 (T) and 0.263 (C) and for GGH 452C>T 0.912 (C) and 0.088 (T). No significant differences in allele frequencies between rheumatoid arthritis patients and healthy controls were found. In addition, the SNPs were not associated with good clinical response to MTX. Only patients with the GGH 16C-allele and one or no copies of the GGH 452C-16T haplotype were associated with good clinical improvement at 3 months upon treatment with MTX. No associations with efficacy at 6 months and MTX-induced toxicity were found. Therefore we conclude that despite the positive association of the GGH 16C-allele and one or no copies of the GGH 452C-16T haplotype with good clinical improvement at 3 months upon treatment with MTX, the tested SNPs in GGH and FPGS genes are suggested not to be clinically important for MTX treatment outcome.