The genetic and life course epidemiology of familial adiposity

• Main Author: Bond, Thomas Andrew Melluish
• Other Authors: Jarvelin, Marjo-Riitta, Lewin, Alexandra, Gunter, Marc, O'Reilly, Paul, Medical Research Council (Great Britain)
• Format: Doctoral or Postdoctoral Thesis
• Language: unknown
• Published: Epidemiology and Biostatistics, Imperial College London 2020
• Subjects: Northern Finland
• Online Access: http://hdl.handle.net/10044/1/99967; https://doi.org/10.25560/99967

The developmental overnutrition hypothesis proposes that prenatal exposure to maternal obesity causes increased risk of obesity and cardiometabolic disease in the offspring in subsequent adult life. Maternal body mass index (BMI) before or during pregnancy is positively associated with offspring adiposity from birth to adulthood, and with adult cardiometabolic disease incidence and mortality, but whether these associations are causal remains uncertain. This thesis aimed to investigate whether greater maternal BMI before or during pregnancy causes greater offspring adiposity in childhood and adolescence, and whether maternal BMI is associated with an adverse offspring cardiometabolic risk factor profile in adulthood. I analysed data from five European prospective birth cohorts: the Northern Finland Birth Cohorts (NFBCs) 1966 and 1986, the Avon Longitudinal Study of Parents and Children (ALSPAC), Born in Bradford (BiB) and Generation R, as well as the UK Biobank. I applied polygenic risk scoring (PRS), intergenerational Mendelian randomization (MR), bivariate Genomic Restricted Maximum Likelihood implemented in the GCTA software package (bivariate GCTA-GREML) and maternal GCTA-GREML. In NFBC1966, greater maternal BMI was associated with greater offspring adiposity and insulin resistance in adulthood, but these associations were somewhat attenuated on adjustment for a PRS partially capturing the offspring’s genetic predisposition to increased BMI. In ALSPAC and BiB, MR analyses suggested that maternal BMI does not have a large causal effect on offspring adiposity in late childhood and adolescence. Bivariate GCTA-GREML analyses in five cohorts showed that imputed offspring single nucleotide polymorphisms (SNPs) explained up to half of the phenotypic covariance between maternal BMI and offspring child and adolescent adiposity. Maternal GCTA-GREML analyses in ALSPAC and BiB showed that genetic confounding (the direct effects of maternal alleles inherited by the offspring) is an important explanation for this. My ...