Lung surfactant kinetics for biomarker discovery in pulmonary disease

Pulmonary diseases has long suffered from a lack of clinical biomarkers for disease diagnosis, prognostic prediction and treatment response. Pulmonary surfactant is critically important for optimal lung function. While altered surfactant function, concentration and composition have been reported in patients with various lung diseases, these changes have also yet to be translated into clinical practice. An alternative approach to measurement at a single time point is to quantify time course kinetics of substrate fluxes through metabolic pathways. Consequently, in this thesis, incorporation of the stable isotope-labelled substrate deuterated methyl choline chloride (methylD9choline) into the major surfactant phospholipid phosphatidylcholine (PC), combined with electrospray ionization tandem mass spectrometry (ESI MS/MS), was employed to monitor the metabolic flux of surfactant phospholipid synthesis in vivo. Time series PC data acquired through ESI MS/MS coupled with stable isotope labelling techniques from three studies was analysed using the smoothing spline mixed effect method (SME). This is a pilot study applying time course label enrichment of multiple surfactant PC species as biomarkers to phenotype animal models and patients with various lung diseases.There is no effective software for data processing mass spectrometry analysis of time series substrate incorporation into multiple molecular species products. Consequently, a bioinformatics platform Lipidome Labelling was developed to facilitate data analysis of dynamic labelling studiesand was used for the following biomarker discovery analysis. Time course methylD9 label enrichment of 16 PC species was modelled using SME method for all the subjects of group pairs in three different studies, which makes the foundation of the multivariate time course biomarker discovery analysis for lung diseases. In a study, a Ft statisticwas calculated as the scaled difference between the time course label enrichment of two contrasting groups for a single PC species, and the Ft statistics for all the species were ranked and according to their significance in distinguishing the two groups, for potential biomarker identification. The surfactant PC kinetic biomarker analysis of a mouse model shows that time course label enrichment of multiple PC species significantly differentiated between wild type mice and granulocyte macrophage colony stimulating factor (GM-CSF) beta chain knock out mice. However, this methodology failed to provide useful biomarkers in the clinical conditions examined, preterm infants at risk of neonatal respiratory distress syndrome and adult patients with acute respiratory distress syndrome compared with healthy volunteers. While differences could be determined between grouped data, subject heterogeneity precluded provision of diagnostic individual for individual patients