Use this URL to cite or link to this record in EThOS:
Title: Metabolic profiling of the early pregnancy journey
Author: Georgakopoulou, Nancy
ISNI:       0000 0004 7657 8503
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Pregnancy is a dynamic state of maternal immunologic adaptation during which women undergo intense anatomical, physiological and metabolic adaptations to retain conception and accommodate the growing foetus. These adaptations are reflected in the maternal plasma and urine metabolomes, which can be subsequently studied through metabolic profiling. A rich and functional description of an expecting mother's molecular phenotype can be the basis for the stratification of pregnant population and development of interventions to prevent pathological outcomes such as preterm birth. Preterm birth, that is delivery before 37 weeks of gestation, represents one of the most important problems in obstetric practice in the developed world, being the leading cause of death for children under the age of five around the globe. Therefore, the establishment of a descriptive early pregnancy phenotype, predictive of preterm delivery could facilitate novel understanding of the biological mechanisms underlying such pathologies and open prognostic avenues. The work presented in this thesis primarily aims to describe the healthy maternal plasma and urine metabolomes during the early second trimester of pregnancy. Other sources of variation that impact upon the maternal metabolome, besides gestational age, are also identified. Additionally, metabolic perturbations indicative of impending cervical shortening and preterm delivery are also explored in both maternal plasma and urine biofluids. Therefore, a large-scale, longitudinal, exploratory 1H-NMR and RP-ULPLC-MS lipidomic study of early second trimester maternal plasma and urine was performed to accurately define a metabolic map of a 'key' time-window of the early pregnancy journey (i.e., 12-21 weeks). Following data processing, univariate and multivariate data analysis was employed to assess metabolic markers' evolution in relation to the uncomplicated gestation. Gestational age was the major source of variation in the plasma metabolic profile during the studied time-period and characterised by global increases in lipoprotein particle subfractions. Additionally, lipidomic analysis highlighted substantial increases in PC and PE subclasses from 12 to 22 weeks in healthy gestation; whereas lysoPCs and fatty-acyls simultaneously decreased, indicating altered phospholipid metabolism and a shift in energy requirements. Despite high patient-to-patient variability, urine metabolites robustly associated with gestational age were also identified. Also, correlations between matched plasma and urine data enabled the investigation of the gestational age impact upon systemic maternal metabolic phenotypes. Finally, metabolic differences associated mainly with ethnicity but also with maternal age, infant gender and weight at birth were also explored in plasma and urine, indicating that variability due to confounding factors should be considered when investigating the maternal metabolome. In conclusion, a detailed description of plasma and urine metabolic trajectories in the early second trimester were presently described and provide a framework for additional studies of markers of adverse outcomes. In order to observe metabolic perturbations predictive of impending cervical shortening and preterm delivery, a cohort of patients that delivered prematurely (< 37 gestational weeks) was investigated at a time-point conducive to timely stratification and clinical intervention. Dysregulation of lipid metabolism appeared to dominate the plasma metabolic phenotype of women experiencing preterm delivery, which manifested as either increased (i.e., in women delivered between 32+0-36+6 weeks) or decreased (i.e., in in women delivered before 31+6 weeks) plasma lipid levels. Also, gestational age at delivery was associated with differences in microbial-mammalian co-metabolites between the urinary metabolic profiles of women delivering either term or preterm babies. By combining cervical length information and urine metabolic phenotyping, the present study paved the way for the potential development of an improved preterm delivery prognostic score. This thesis suggests that metabolic phenotyping is a key tool to understand the alterations associated with gestational complications. At a time-period during which two partner organisms are interacting for mutual benefit, metabolic phenotyping platforms offer the potential for identification of key molecular markers for early prediction, diagnosis and monitoring of different obstetric conditions. Both the metabolic description of healthy phenotypes and the characterisation of perturbed pathological phenotypes offer an improved understanding of the associated biological mechanisms. This information may permit monitoring and stratification of pregnant cohorts and facilitate informed, objective and targeted clinical intervention.
Supervisor: Bennett, Phillip ; Holmes, Elaine ; Nicholson, Jeremy Sponsor: Imperial College London ; National Institute for Health Research ; Medical Research Council ; SPARKS Children's Medical Research
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral