Edinburgh Imaging

PhD projects 2013 001

Genetic contributions to cognitive ageing & structural brain magnetic resonance imaging phenotypes.


As humans age, specific mental faculties deteriorate even in the absence of dementia. Age related cognitive decline affects quality of life, and has significant implications from a socio-economic perspective; however not everyone declines to equal degrees, at equal rates, or from the same baseline. This PhD examined a large sample of community-dwelling older adults called the Lothian Birth Cohort 1936, most of whom completed an intelligence test at age 11 years, and again around age 73 as part of a detailed assessment that also included detailed brain magnetic resonance imaging (N range = 700-866). I investigated the independent effects of two linked genetic loci which have been associated with greater risk of Alzheimer’s disease – the APOE ε haplotype (commonly ‘genotype’) and a poly-T repeat in the TOMM40 gene. Are 'risk' variants in these loci associated with specific measures of cognitive ageing and brain structure - specifically white matter microstructural integrity, hippocampal volumes, white matter lesions or cerebral microbleeds – in this sample?

Firstly, a pilot study aimed to replicate significant associations between the ADRB2 gene and brain imaging/cognitive phenotypes, that had previously been reported in a smaller subsample of the cohort that had by that time undergone MRI (n = 132). Previously reported significant associations were not significant in the larger, full LBC1936 sample (n = 700-866), but novel significant associations were found (P < 0.05). Specifically, integrity of the left arcuate fasciculus white matter tract significantly mediated part of the association between specific genetic variations at ADRB2, and the Digit Symbol Coding task of information processing speed. These findings indicated that this approach – testing three-way genetic/brain imaging/cognitive associations for mediation - was viable for the main APOE/TOMM40 analyses.

Results in the main APOE/TOMM40 analyses showed that specific variants in the APOE and TOMM40 gene loci were statistically significantly associated (at raw P value <0.05) with white matter tract microstructural integrity, but not white matter lesions, hippocampal volume or cerebral microbleeds. Inconsistencies with previous, positive reports showing significant associations between APOE ε and these latter phenotypes may reflect a degree of type 1 error or more study-specific discrepancies (which are detailed throughout). APOE ε was significantly associated with average scores on a large proportion of cognitive tests, independent of age 11 intelligence (i.e. ‘cognitive ageing’; Deary et al., 2004). These associations were partly – but not completely – mediated by white matter tract microstructural integrity. TOMM40 poly-T repeat genotype was associated with cognitive ageing to a much lesser extent. A range of brain phenotypes may form the anatomical basis for significant associations between APOE genotype and cognitive ageing, among which includes white matter tract microstructural integrity.

  • Genetics
  • Ageing brain
  • MRI
  • Cognitive ability
  • Donald Lyall
  • PhD
Link to University of Edinburgh holdings: