Air pollution and Dementia: Exploring genetic, cardiovascular, and epigenetic moderators and mechanisms

Background: Dementia is a major, growing public health priority. A person's genetic makeup can place them at higher risk of dementia, but there are also important environmental predictors that can increase risk. Air pollution is one of these, with evidence converging worldwide on the negative effects of pollution on brain health. These population studies have been extremely large, sampling millions of people, but they ignore a person's genetic predisposition. It is possible that pollutants interact with a person's genetic risk of dementia: greater air pollution might increase one's genetic risk, and one's genetic makeup can imply greater susceptibility or resilience to environmental insults. Air pollution also predicts cardiovascular disease (CVD), which in turn predicts dementia, and it is plausible that CVD itself and/or CVD risk factors mediate the observed association between pollutants and dementia, as indicated in prior work. It is also possible that air pollution affects brain health through altering of biological mechanisms such as accelerated telomere (nucleotides that protect the end of chromosomes) shortening and altered DNA methylation (which can change gene expression), both of which are strongly associated with ageing.

Aims: The overarching aim of the project is to study the association between air pollution exposure across the lifespan and risk of dementia/cognitive decline. The aims include :

1) quantify the effects of gene by environment (air pollution) interaction on dementia. This will be done by a) applying classical twin design to data from the Swedish Twin Registry (STR), modelling gene by environment interactions to study whether genetic and environmental variance in dementia differ by levels of exposure to air pollution, and b) modelling a polygenic risk score for dementia in interaction with air pollution in the Lothian Birth Cohorts (LBCs), UK Biobank and STR, to specifically model interaction between air pollution and genetic risk of dementia on dementia status (UK Biobank, STR) and cognitive decline (LBCs where dementia analysis will be less powerful).

2) test whether air pollutants are causally related to dementia. This will be tested using a discordant MZ co-twin control design in the STR (N=1,214 pairs). In this design, MZ twins are perfectly matched for their genes and early environments, enabling comparison of their environmental exposure in adulthood in relation to their dementia status. A causal model would be supported if the twin with greater pollution exposure is the dementia case.

3) test if CVD mediates the effect of air pollutants on dementia. This will be examined by incorporating diagnoses of CVD in mediation models of air pollution and dementia in the LBCs and the STR while accounting for genetic risk of CVD and dementia.

4) test whether increased air pollutant exposure (over the lifetime in the LBCs and in the STR) is related to shorter telomere length and to methylation probes across the genome, and if so, whether these mediate the relationship between the air pollutant exposure and dementia. Uniquely, methylation data are available longitudinally (up to 20 years), so that we can also link them with longitudinal measures of air pollution.