Using a computational human metabolic reconstruction on the study of mucopolysaccharidosis

Glycosaminoglycans (GAG) are the main biomarker used in the diagnosis of mucopolysaccharidosis (MPS). However, during the last years there have been growing interests in the identification of new biomarkers that allow to overcome some of the limitations observed with GAG. In this study we used a computational human metabolic reconstruction (Recon2) to model the metabolic changes observed after silencing of each one of the MPS-related genes, as well as to predict new biomarkers for MPS. Models were done and analyzed using COBRA toolbox through flux balance and variability analysis. Genes associated with reactions that changed after gene silencing were subjected to enrichment analysis. Overall, 895 reactions out of 7441 were commonly impaired in MPS models, which are mainly involved in cellular respiration, mitochondrial process, amino acid and lipid metabolism, and ion exchange processes. Metabolic changes were similar for MPS I and II, and for MPS III A, B and C; while the remaining MPS showed unique metabolic profiles. These results agree with previous reports showing autophagy, cellular and mitochondrial stress on MPS, and improved the prediction previously done using the same stratetgy for GAG degradation pathway. Finally, models were used to predict biomarkers for MPS. Although in-vivo validation of these biomarkers is still needed, these results confirm the potential of the computational human metabolic reconstruction to understand cellular alterations and the prediction of biomarkers for MPS.