INTEGRATIVE METABOLOMIC AND LIPIDOMIC STUDIES FOR IMPROVED NEONATAL SCREENING OF CYSTIC FIBROSIS

Abstract

Cystic fibrosis (CF) is the most common autosomal recessive disorder in the Caucasian population. It is caused by mutations of the cystic fibrosis transmembrane regulator (CFTR), which controls the transport of chloride and bicarbonate channel as required for fluid excretion and extracellular alkalinization. CF is thus characterized by impaired CFTR function, generating an accumulation of thick mucus, which can cause intestinal obstructions, pancreatic insufficiency, infertility, respiratory infections, and progressive lung damage. Pre-symptomatic diagnosis and early treatment of CF can be achieved via universal newborn screening programs with improved clinical outcomes as compared to symptomatic diagnosis later in life. Nonetheless, there still exists on-going screening challenges, including a high fraction of presumptive screen-positive CF neonates who are unaffected or have an indeterminate diagnosis further complicated by the advent of next generation genome sequencing. Comprehensive metabolite profiling (metabolomics) offers a powerful tool for the discovery of novel biomarkers that can shed new insights into underlying disease mechanisms. With the aim of improving newborn screening for CF, a standardized two-phase liquid extraction procedure was developed to enable the fractionation of metabolites and lipids from a single dried blood spot punch (~3 μL). Polar metabolite filtrates in the aqueous fraction were analyzed using capillary electrophoresis-mass spectrometry (CE-MS) with on-line sample preconcentration under positive ion mode. Lipids were evaluated with an optimized protocol to analyze dried blood spot ether extracts using reversed-phase ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS), where both positive and negative ion mode configurations were used, ensuring the lipidome coverage was maximized. Targeted metabolite and lipid profiling was conducted on a pooled neonatal sample as quality control using CE-MS and UHPLC-MS to classify reproducible, authentic, yet non-redundant molecular features. All of which were annotated by their accurate mass, relative migration time or retention time, and mode of analysis when not detected in blank extracts. This work characterized the dried blood spot metabolome from newborn infants, including metabolites/lipids that differentiate affected CF cases from unaffected carriers and screen-negative controls after covariate adjustments. Unknown metabolites/lipids associated with CF cases will be subsequently characterized by high-resolution tandem mass spectrometry. This work is critical to improve the positive predictive value and cost-effectiveness of CF newborn screening programs and alleviate on-going diagnostic uncertainties.