Role of regulatory elements in the genetic determination of VWF level in Von Willebrand disease and healthy individuals

Von Willebrand Factor (VWF) is a protein needed to make the blood clot. VWF is stored and released into the bloodstream by the cells lining the blood vessels (endothelial cells). VWF is also stored in the smallest type of cells in the blood, called platelets.

Low VWF in the bloodstream leads to a bleeding disorder called Von Willebrand disease (VWD), affecting 10,000 people in the UK. It negatively impacts on patients' quality of life. The flipside is that the blood clots too readily resulting in conditions such as deep vein thrombosis (DVT). Studies with a large number of participants have shown that a high level of VWF is a risk factor for DVT. The level of VWF in the blood is for a large part controlled by our genes. Differences in the DNA between individuals are called variants. The variants that lead to VWD are known in every 7-8 out of 10 patients, although in some patients ascribing the variant causing VWD is challenging because of lack of data from enough healthy volunteers. Not all of the variants that explain how VWD is inherited have yet been mapped out by scientists.

Understanding more about VWF inheritance could therefore provide more people with VWD an accurate genetic diagnosis, enabling them to guide their own family members about their chances of being affected. Scientists need also to find out more about which variants are associated with elevated VWF levels as this information could - in the future - potentially be used to guide people, like you about the risk of a DVT.

Reading the DNA code is done by sequencing and the word genomics is the term for the reading and analysis of the entire DNA code (or genome). Cambridge scientists in Professor Ouwehand's team have been at the forefront of this. It is becoming apparent that there are many interactions between different parts of the genome. The genome contains many elements that work like 'switches and dimmers' controlling thousands of our genes ('the light bulbs'). Scientists from the BLUEPRINT project are mapping out the links between the switches and dimmers and the genes they control. They have also done this for the cells in our body that produce VWF and we can now answer the question whether in some VWD patients the gene for VWF does not work because one of the switches or dimmers is faulty.

Work just completed by our scientists has mapped out specific interactions between highly active dimmers called super dimmers for the VWF gene and its neighbour, CD9 (referred to as VWF-CD9). In a study linking blood cell features to variants in 170,000 people, variants were identified in a super dimmer that controls VWF-CD9 in the cells from which platelets originate (megakaryocytes). Because super dimmers can control more than one gene and their effect on the connected genes are not always clear it is important to examine this in relevant types of cells in the laboratory. Using stem cells (which can mature into any type of cell), and 'DNA scissors' to remove the super-dimmer, the stem cells were turned into megakaryocytes which contained the super-dimmer and those that didn't. The megakaryocytes without the super-dimmer had higher levels of VWF and almost no CD9, than those that didn't.

In my research project blood samples from 1000 VWD patients from across seven UK hospitals will be collected and DNA extracted for sequencing. Variants will be looked for in the VWF gene and a closeby region including switches and dimmers. The same will be done for 42,000 volunteers without VWD participating in the 100 000 Genomes Project. I will then work with experts in statistics & computing to analyse for connections between variants in the switches and dimmers for VWF and VWF levels. To confirm the effect of my findings I will use 'DNA scissors' to engineer stem cells with & without the variant in them and mature them not only into megakaryocytes and platelets but also the cells lining blood vessels so I can study how these variants affect VWF production.

In the UK there are ~10,000 VWD patients and they mainly present with mucocutaneous bleeding. The ThromboGenomics (TG) high throughput sequencing (HTS) test, introduced as a diagnostic test in the NHS in 2015 identifies causal variants in VWF exons. There is an assumption, partially supported by TG test results that a fraction of VWF variants, labelled in reference databases as LPV and PV have been erroneously ascribed their label. Therefore, the fraction of type 1 VWD patients in whom a conclusive molecular diagnosis can be reached may be lower than the estimated 75%.

I will use HTS to analyse the 0.5 MB comprising the VWF-CD9 genes in 1,000 VWD patients to identify LPVs in non-coding elements of the locus through comparison with data from 10,000 WGS-analysed controls from the INTERVAL RCT. I will perform a fine-mapping analysis to associate variants in the 0.5 MB and VWF levels. I will then use a new Bayesian algorithm to search for rare variants enriched in the 1,000 VWD patients using 42,000 INTERVAL and 100KGP controls. Erroneous assignment of pathogenicity status to variants will be minimised by having access to NHS record data of the WGS controls, in contrast to the anonymised whole exome sequenced (WES) samples from the ExAC Consortium. There is a considerable chance that a DNA variant present in a regulatory element of VWF and observed in a few or a single VWD patient(s) but absent from the 42,000 controls is not causal of VWD. So putative causal DNA variants will be verified for co-segregation in pedigrees. Those that co-segregate will be selected for functional validation. The variant will be introduced in iPSCs by CRISPR-Cas9. Wild-type and edited iPSCs will be forward programmed into ECs and MKs, in which I will measure VWF transcript and VWF protein levels. The genome-editing is relatively simple for insertions/deletions but challenging for single nucleotide variants. So as a back-up approach iPSC lines will be generated from the relevant patients.

This is a multi-faceted study of von Willebrand factor (VWF), in both those with von Willebrand disease (VWD) and in the healthy population, with data being acquired at a cohort/population and cellular level. There are therefore many potential stakeholders with an interest in the outcomes of this research aside from academic beneficiaries.

These include:

Patients

With VWD
Although VWD is diagnosed by a low plasma VWF level, many patients could benefit from knowledge of an genetic diagnosis if the test was readily available, accurate, and helped explain their clinical symptoms. This is one of the outcomes we aim to achieve by extending the ThromboGenomics (TG) High Throughput Sequencing (HTS) test to those with VWD. If our more detailed analysis of the whole VWF-CD9 locus can help to explain some of the 'missing' heritability of VWD then more patients will be able to be offered tailored genetic counselling.

With venous thromboembolism and/or cardiovascular disease
If the GWAS of the INTERVAL cohort identifies new pQTL associating with VWF levels in the healthy control population, then this could be an area of focus for future epidemiological studies in patients with VTE and in the long-term help identify potential new risk factors for recurrence.

Clinicians

Specialist haematologists in haemostasis and thrombosis
If pathogenic variants (PVs) in regulatory elements causing VWD are identified then this potentially would provide a group of patients to further characterise with case report/case series to identify whether there are any differences between them and others with (type I) VWD.

Genera haematologists
The TG platform has already proven its utility in providing clinicians with a comprehensive genomic test for patients with unidentified bleeding disorders. With its expansion to VWF and the further addition of newly discovered BPD variants it may become a comprehensive test in the work-up of BPD patients in the future. Its reliance on DNA sequencing as opposed to more technically challenging and time critical functional tests such as platelet aggregation tests means that it could become a more readily available tool to the general haematologist in the future.

Surgeons and anaesthetists
The research that my project will generate contributes to a broader aim of improving diagnostic accuracy in the genetic diagnosis of patients with VWD and other inherited BPDs. This will help to reduce the number of patients with an unexplained BPD, which can be challenging to manage in surgical settings.

Professional organisations

United Kingdom Haemophilia Centre Doctors' Organisation (UKHCDO)
In 2014/15, there were 4415 VWD patients with type unreported on the VWD registry out of a total of 10,432. Making the assumption that these are equally distributed across haemophilia centres, this would mean that ~400 out of the 1000 VWD participants recruited across the BioResource have VWD Type Unreported. Although PVs do not map neatly with the type 1/2/3 classification for the majority of type 2 they are well described. Therefore a significant number of Type Unreported participants in my research will be re-classified impacting beneficially on UKHCDO reporting statistics.

British Society of Haematology (BSH)
Detailed evaluation of variants in 1000 VWD patients linked with HPO coding may provide support for more routine testing for genetic variants in the routine diagnosis of VWD, which would be practice-changing and could be reflected in BSH guidelines.

Biotechnology companies
Expanded utilisation of HTS in BPDs will increase revenue for companies offering NGS technology.

Charities and support groups
Examples include: The Haemophilia Society and The Platelet Charity. My research could generate interest in VWD if published in a significantly high-impact journal. This could focus media attention on VWD and be utilised by charities to improve public awareness.