Principal Investigator: Dr Laura Howe
Department: School of Social and Community Medicine
Institution: University of BristolTags: 10953, cardiovascular disease, Mendelian randomization, obesity, stroke
Dr Stephen Burgess
Collaborating Institutions and Address
University of Cambridge
MRC – grant reference MR/M020894/1
1a: Obesity is a strong risk factor for stroke and heart disease. While preventing obesity is the ultimate goal, this is not always possible. As such, it is important to find ways of preventing the adverse health consequences of obesity. In this project, we will estimate how much of the ill-effects of obesity could be prevented if we intervened on intermediate processes such as blood pressure and lipid levels.
1b: The research will help efforts to prevent the adverse health consequences of obesity. I will do this by helping us to understand the potential benefits of intervening at different steps in the causal pathway from obesity to disease.
1c: We will use a technique called ‘Mendelian Randomization’, which uses genetic data to help us be sure that the effects we are seeing in the data are real and not due to other factors.
1d: We will include the full cohort in our analysis.
PROJECT EXTENSION 1 – APPROVED 11.01.2016:
“ECSOD – Extracellular superoxide dismutase (ECSOD) is the major extracellular scavenger of superoxide and a main regulator of nitric oxide (NO) bioactivity in the blood vessel wall, heart, lungs, kidney, and placenta (Qin et al Transl Research 2008). Involvement this metabolic axis has been implicated in many pathological processes and removal of extracellular superoxide (by ECSOD gene transfer) has emerged as a promising experimental technique to treat vascular disorders associated with increased oxidant stress. Efforts to examine this contribution using applied genetic epidemiological approaches have focused on the coding variant R213G(rs1799895) which is a functional single nucleotide polymorphism causing a substitution of Arginine for Glycine. Heterozygotes for the minor allele have substantially increased plasma SOD3 levels (approximately 10-fold), while homozygotes for the minor allele have dramatically increased plasma SOD3 levels (approximately 40-fold)(Folz et al Hum Mol Genet 1994, Juul et al Am J Respir Crit Care Med 2006). Up until now, reference panels have not carried the functional R213G(rs1799895) variant and as such our work has been limited to looking at bespoke genotyping exercises only and related phenotypes. However, recent developments in the application of whole genome sequencing and reference panel formation (in principle the UK10K project, UK10K Consortium, Walter et al, Nature 2015) have allowed for capture of this variant (which is at around MAF 0.02 in Europeans and imputed with high quality). Consequently, the opportunity to expand a cardiomartabolic phenotype association sweep within the UK Biobank study now exists and we are well placed to execute this quickly with existing resources. This would have a definitive role in the clarification of the role of ECSOD in this axis and more importantly the potential causal role of superoxide in disease.”
PROJECT EXTENSION 2 – APPROVED 15.02.2016:
“We would like to look at the association between smoking and cardiovascular diseases, and the mediators of this association, including body mass index, blood pressure and, when available, biomarkers. We will use Mendelian Randomization to assess these relationships, using separate sets of SNPs for smoking and for each mediator of interest. We anticipate this work leading to a methodological paper about mediation analysis with Mendelian Randomization, as well as one or more applied papers looking at causal pathways from smoking to disease.”
Project extension 20.11.17
“To include pathways linking social exposures (education, Townsend deprivation) to cardiovascular disease. Mediators considered will include BMI, smoking, alcohol use, social support, employment, health service use and medication use.
We will conduct both observational analysis of mediating pathways and Mendelian Randomization approaches, using genetic variants as instrumental variables for the exposures and mediators (where possible). This amendment involves me and co-applicants Alice Carter and Teri North.”
PROJECT EXTENSION – APPROVED 08.12.2017:
Is the bidirectional association between cognitive function and CVD causal?
Observational evidence has shown that cardiovascular disease (CVD), and a more adverse CVD risk factor profile, is associated with lower cognitive function in adulthood. The hypothesis is that greater CVD risk (i.e. elevated blood pressure, lower arterial distensibility and greater atherosclerosis) may be associated with restricted cerebral blood flow and subsequent neuronal death, thereby reducing cognitive function. There is also some evidence that the association goes in the other direction (i.e. lower cognitive function is associated with increase CVD risk), however it is not clear whether this latter association is likely to be causal (potentially mediated by behavioural variables) or a result of confounding by social factors. We would like to examine the bidirectional relationship between CVD risk and cognitive function using Mendelian Randomization as a tool to overcome bias due to confounding and reverse causation. If causal associations are identified between cognitive function and CVD, we will examine the extent to which this is mediated by behavioural factors such as smoking, alcohol use, and physical activity.
Last updated Dec 12, 2017