Precision medicine in rheumatoid arthritis: role of alternative splicing in methotrexate response

After specifying the content of the thesis chapters (chapter 1), the main research topic that is investigated in the thesis is outlined in chapter 2; i.e., the role of the enzyme folylpolyglutamate synthetase (FPGS) in the polyglutamylation process of methotrexate (MTX) - the anchor drug in rheumatoid arthritis (RA) treatment - in white blood cells of RA patients. The formation and intracellular accumulation of MTX-polyglutamates (MTX-PGs) is crucial for prolonged intracellular retention and conferring enhanced inhibitory effects on key enzymes in folate and purine metabolism. In RA, studies so far involved analyses of MTX-PG levels in red blood cells (RBCs) as a potential indicator for therapy response. However, immune cells in circulation and inflamed synovium are more representative for disease pathogenesis, thus examination of FPGS activity and MTX-PG accumulation in these cells, beyond RBCs, warrant investigation. The development and validation of a sensitive LC-MS/MS-based assay to measure FPGS catalytic activity in small numbers of blood cell samples is presented in chapter 3. The assay was applied to determine FPGS activity in peripheral blood mononuclear cells (PBMCs) of RA patients and clinical acute leukemic specimens. FPGS activity was markedly higher (50-100-fold) in leukemia samples as compared to PBMCs of RA patients. Next, we researched the impact of aberrant pre-mRNA splicing of FPGS in whole blood of 36 RA patients at baseline for MTX-responsiveness after 3- and 6-months therapy (chapter 4). Particularly, the expression of a FPGS splice variant, a partial retention of FPGS intron 8 (8PR) over wild type (WT) FPGS, was explored. Intriguingly, higher baseline ratios of 8PR/WT were significantly associated with higher DAS44 scores at 3 and 6 months MTX therapy. These were the first data in RA linking aberrant FPGS splicing to reduced MTX-responsiveness. Multiple computational tools to investigate splicing are available for researchers, but their mutual performance is unclear. The performance of 3 splicing tools used for detection of splice events from RNA-seq data sets was compared in chapter 5. These tools (rMATS, MISO, and SUPPA2) had similar performance on a computational level (job time, CPU usage, RAM usage, benefits of parallelization) as well as in splice event detection. Both rMATS and SUPPA2 outperformed MISO in terms of performance, with markedly lower CPU jobtimes, CPU load averages and RAM usage. In addition, splicing tool output comparisons showed high concordance between all tools with the exception of comparisons of rMATS with MISO or SUPPA2 for retained intron (RI) events. (R2 > 0.59). Whether FPGS splicing is a drug (MTX)-induced phenomenon or also proceeds as part of immune cell differentiation and activation, was investigated in chapter 6. The latter appears to be true as, strikingly, during ex vivo skewing of monocytes to either M1-type (pro-inflammatory) macrophages and M2-type (anti-inflammatory) macrophages, a 6-fold decrease in the ratio of FPGS 8PR/WT was noted along with the same level increase in the activity of FPGS. Differential gene expression profiles of M1- and M2-type macrophages w/o exposure to low-dose (50 nmol/L) MTX revealed differences in expression of folate metabolism genes and in MTX-related and metabolic pathways. Results underscore a role for epigenetic alterations in M2-type macrophages during MTX treatment, not in M1-type macrophages. Lastly, in a metabolomics study of baseline plasma samples of 41 RA MTX responders and 41 RA MTX-non-responders after 3 months, increased levels of metabolites of the glycolytic pathway and amino acid metabolism were observed in insufficiently-responding patients (chapter 7). This data needs further confirmation in larger data sets of patient samples, but this plasma metabolome feature holds promise as potential biomarker for MTX (non)-response in RA. Altogether, these studies contribute to improve personalized/precision medicine for MTX in RA treatment.