Characterisation of autoantigen-specific human B cells in central nervous system autoantibody-mediated diseases

Background
Our immune system has evolved to protect us from infections. Often, the most effective protection is via its production of antibodies. Antibodies are made by the immune system's 'B cells'. Antibodies can bind to bacteria and directly clear an infection. However, sometimes, they can - in error - recognise our own body as the invader. These so-called 'autoantibodies' lead to 'autoimmune' diseases which affect ~10% of the population and causes significant disability and sometimes death. So, understanding how the body removes the self-reactive B cells is of major importance to develop better treatments.

Autoimmune conditions can affect almost any tissue in the body but, traditionally, the brain has been considered relatively protected. However, over the last decade, we and others have successfully identified 15 brain proteins which are targeted by autoantibodies. This far outweighs the number of similar autoimmune syndromes in any other branch of medicine. Patients with these autoantibodies often have seizures, memory loss and psychiatric symptoms, together called 'encephalitis'. The number of forms of autoimmune encephalitis (AE) continues to grow annually.

Oxford is the UK's major referral centre for patients with AE. 95% of our patients consent to research involvement, so my team can directly study patient samples and link findings to their symptoms and treatment responses. Indeed, our patients show some improvements with available medications which suppress their immune system. However, 80% either cannot return to work, have relapses or experience unacceptable side effects from their medications.

To improve these outcomes for patients, this project aims to direct the discovery of improved medications. Also, as autoimmune diseases are common and B cells are found in the brain in many neurological conditions, our findings will all inform the biology and care of patients with several other illnesses.


Aims of this project:
1. To identify which B cells are the first to become self-reactive and which are the most self-reactive. This will highlight cells that are ideal targets for future therapies, lay the foundations for Aims 2 and 3, and offer explanations about causation to our patients.

This work will study blood donated by our patients and by healthy people. We will separate blood B cells into populations which represent stages of their development. Then, individual populations can be activated to discover the earliest, and the most potent, producers of the autoreactive antibodies. Comparisons to healthy people will show where the self-reactive B cells are successfully cleared. By repeating these studies after patients are treated, we will understand if current therapies can reset the process to that observed in health.


2. To describe molecules exclusively found on the B cells which make the autoantibodies. Targeting these molecules with medications would simultaneously increase effectiveness and reduce side-effect profiles.

Using findings from Aim 1, the most self-reactive B cell populations in blood will be purified. Also, samples from patient spinal fluid will access self-reactive B cells which circulate and contact their ultimate target - the brain. From both sites, we aim to compare the self-reactive to the non self-reactive cells and identify the distinct range of molecules on the former. Work will occur in collaboration with expert cell profilers at Oxford University and Yale University (USA).


3. To ask if certain autoantibodies are especially potent at inducing disease. The B cells that make these would be even more precise targets for future therapies.

From the B cells purified in Aim 2, we will recover multiple autoantibodies. These will be applied to brain cells 'in a dish' and injected into mice, to identify those which most robustly disrupt these systems. Relating this to data from Aim 2 will reveal the profile of the B cells that made the most potent autoantibodies.

Autoantibody-mediated illnesses are rare throughout medicine. By contrast, over twenty currently exist in neurology. In these conditions, a focused characterisation of the autoantigen-specific B cell lineage offers a highly tractable opportunity to accurately study the underlying human pathology and simultaneously identify highly selective drug targets. As a paradigm, I will use CASPR2-antibody encephalitis as a model disease.

My aims are to:

1. Track the maturation of CASPR2-specifc B-cell lineages across key human lymphoid compartments. I will identify the earliest CASPR2-reactive B cells and their development through lymph nodes into their key site of pathogenic autoantibody production: cerebrospinal fluid (CSF). This will exploit our established methods to isolate individual CASPR2 B cells.

2. Characterise the CASPR2-specific B cells in blood and CSF, to identify molecules which mediate their striking concentration in CSF. Flow cytometry and single cell transcriptomics will identify molecules exclusive to CASPR2-specific B cells. These may represent highly-selective future drug targets and mediate their observed preferential migration to or retention within CSF.

3. Compare the relative pathogenicity of CASPR2-specific monoclonal antibodies (mAbs) derived from CSF B cells in Aim 2. Well-established in vitro and in vivo models will explore the dominant mechanisms of CSF mAbs. This will identify the most pathogenic mAbs as disease biomarkers and their transcriptomes (derived from Aim 2) as even more specific future drug targets.

Overall, these data will molecularly profile a well-defined 'pathway to pathogenicity'. Biologically, they will generalise to identify human autoreactivity checkpoints and mechanisms of B cell access to the CSF. Clinically, the proposal creates a precision medicine paradigm across autoantibody-mediated diseases which will translate to improved treatments for these and other established and putative neuroimmune conditions.