Introduction: The environmental and physiological exposures can have a significant impact on metabolomics, which reflects eventually on the phenotype. Almost all human diseases and individual traits result from inherited or somatic genomic variations that determine the individuals' phenotype. Those variations relate to a network of biochemical reactions that involve small molecules whose expression and dynamics can be profiled using metabolomics. Developing targeted metabolomics platforms that involve the central and most clinically relevant pathways connected to human genetic disorders have a huge potential of moving the routine genetic diagnosis to a higher level with greater confidence, robustness, and efficiency.
Objective: Dry blood spots (DBS) samples are the most convenient samples for biochemical genetic diagnosis including newborn screening (NBS). In this study, we investigated the impact of profiling a targeted group of endogenous metabolites in DBS using quantitative methods for metabolic pattern biomarker discovery.
Methodology: A targeted metabolomics platform for a group of metabolites (n=350) related to known biochemical genetic diseases was developed using Gas and liquid chromatography-tandem mass spectrometry. The methods were validated for routine analysis, where normal ranges were determined for this group of metabolites using the DBS sample (n=360) classified as normal or abnormal based on routine clinical NBS. Patient samples with different diagnoses were profiled.
Results: These methods cover the most clinically relevant and critical pathways, which was successfully validated for routine analysis with excellent linearity (R2>0.98), inter and intraday validation (CV, 7-12%), and accuracy (80-121%). The targeted metabolites were extracted from different biological materials using different established protocols, where the average extraction recovery was >65% for most metabolites. The metabolomic profiles between healthy and affected individuals were separated with Q2 =0.831 on the principal component analysis (PCA). Various metabolic disorders such as phenylketonuria, tyrosinemias, glutaric aciduria were confirmed. In methylmalonic acidemia, this method showed a distinctive profile with an additional perturbed expression of alpha-KG, biotin, and some amino acids, beyond its known single marker “methyl-malonic acid”. Finally, sorbitol was significantly down-regulated in cystic fibrosis patients leading to the mucoviscidosis seen in their respiratory and gastrointestinal secretions. This finding, in particular, supports the clinical benefit of mannitol inhalation therapy in patients with CF.
Conclusion: Targeted metabolomics can distinguish between patients with known diagnoses using a pattern of metabolites in addition to their known single markers. This approach has the potential to enhance early and fast diagnoses with a high discovery rate. It also can identify biomarkers of potential therapeutic value.