Leveraging human genetics and a zebrafish model to investigate novel and repurposed drug targets for aortic valve stenosis

Abstract

Aortic valve stenosis (AVS) is a debilitating valvular heart disease with no approved pharmacotherapies. The overall aim of my thesis was to use interdisciplinary methods, namely genetic and molecular epidemiology in combination with in vivo functional validation in zebrafish, to characterise biological drivers of AVS risk, with an emphasis on drug targets.

In Chapter 2, I used different Mendelian randomization (MR) methods to estimate the genetically predicted effects of established and emerging risk factors on AVS risk, focusing on modifiable risk factors, different measures of adiposity, heart structure and function, and blood cell counts and traits. I also compare the effect of these traits on coronary artery disease (CAD), considering the known similarities in the biological mechanisms underlying the two diseases. This analysis demonstrated serum phosphate, high blood pressure, central obesity and lipid traits should be considered in future risk stratification and prevention strategies. In addition, I identified heart structure and eosinophil counts as traits that need further investigation as potential risk factors.

Given some shared risk factors between AVS and CAD, Chapter 3 focused on using different drug target MR methods to investigate repurposing opportunities in therapies indicated for CAD and conduct a hypothesis-free interrogation of the proteome for novel targets. I then investigated whether colocalising variants were associated with any other phenotypes, informing on further repurposing opportunities or potential side effects. Some antihypertensive and lipid-lowering targets had a causal effect on lower AVS risk. In the hypothesis-free analysis using eight pQTL datasets, I found evidence of a genetically predicted effect for seven proteins on AVS risk, of which four (PLG, ATOX1, LPA, CELSR2) were also supported by genetic colocalisation.

To follow up on these findings, I took forward some genes for in vivo targeting in zebrafish. In Chapter 4, three candidate genes were targeted using CRISPR/Cas9 genome editing in zebrafish with the aims of creating new stable knockout lines for future study of gene role in valvular heart disease and keeping CRISPR/Cas9-injected adults for further characterisation. The high efficiency and germline transmission of CRISPR/Cas9 mutagenesis in this study also showcased the tractability of zebrafish to follow-up on targets from hypothesis-free genetic studies.

The functional and morphological cardiac and valvular phenotype of the CRISPR/Cas9-injected zebrafish lines in adulthood was explored in Chapter 5. Echocardiographic assessment suggested the disruption of pcsk9 may alter valve function in the opposite direction of effect from the phenotype observed in aged wild-type fish. In contrast, no changes to pcsk9 expression or valve morphology were observed.

Finally, in Chapter 6, I undertook an exploration of differentially expressed proteins according to AVS case status or genetic susceptibility using individual-level data from the UKB. Multivariate linear regression showed the proteomic profiles of AVS and CAD patients were strongly correlated, and less so when using a polygenic risk score for each of these diseases. Over- representation and enrichment analyses demonstrated that proteins upregulated in AVS patients were involved in immune, inflammatory, and osteogenic processes. This analysis thus provided a preliminary characterisation of the systemic molecular profile of AVS and potential biomarkers of disease.

Using human genetics and in vivo functional validation, these findings help prioritise novel and emerging drug targets that may yield clinical benefit towards AVS risk. This thesis serves as a case study of the potential of interdisciplinary pipelines in the identification of molecular players in disease risk, enabling the identification and/or validation of drug targets.

Date of Award	4 Feb 2025
Original language	English
Awarding Institution	University of Bristol
Supervisor	Tom R Gaunt (Supervisor), Beck J Richardson (Supervisor) & Tom G Richardson (Supervisor)