Identification of HLA-B*27 bound peptides involved in the pathogenesis of Ankylosing Spondylitis and Reactive Arthritis

HLA-B*27 is found in over 90% of patients with Ankylosing Spondylitis and Reactive Arthritis, two of the most common inflammatory arthritic diseases. Yet this immune protein is present in only 5-7% of people without these diseases. HLA-B*27, like other HLA types, is present on almost all cells of the body. The normal function of HLA-B*27 is to bind to foreign peptides that come from infecting microbes, and to stimulate T cells to eliminate the infection. One hypothesis is that sometimes T cells that are stimulated by foreign microbes also react to peptides bound to HLA-B*27 that come from normal, uninfected cells. This is a theory we want to test. In this study, we will take T cells from the joints of patients with inflammatory arthritis and use a powerful, new method to identify all the peptides they respond to. The technique involves putting HLA-B*27 onto yeast cells, each of which can place a different peptide onto the HLA-B*27 protein. We can make 100 million yeast cells, each with a different peptide. Then we will make the receptor proteins, namely the T cell receptors (TCRs), that normally react with HLA-B*27 plus peptide, from the patient's T cells, to select the yeast cells that bind. Finally, we will analyse the HLA-B*27 bound peptides that the TCRs recognise and compare them to all the peptides in the human and microbe databases. The latter have been accumulated from the Human Genome Project and the Microbe Genome Project over the last 10 years. This will allow us to identify both the microbe peptides that first stimulated the T cells, and the peptides from normal cells that they potentially attack to cause arthritis. The ultimate aim of this study is to identify if peptides that bind to HLA-B*27 cause disease. If we are successful we, and others, should be able to make candidate treatments that interfere with that process. This approach to therapy will be highly specific to this type of arthritis and far less likely to cause serious side effects than current treatments.

Although the link between Ankylosing Spondylitis (AS)/Reactive Arthritis (ReA) and HLA-B*27 has been recognised for 40 years, the pathogenic role played by the disease-associated MHC class I subtypes is not understood. Various theories have been proposed, including models implicating the unusual biology of HLA-B*27 that drive the production of aberrant protein forms which contribute to inflammatory disease. The alternative and original, 'arthritogenic peptide' model suggests that disease pathology relates to normal HLA-B*27 forms, where pathogen-specific HLA-B*27-restricted T cells recruited during natural infection subsequently cross-react with self-peptides bound to HLA-B*27 and promote autoimmunity. Recent GWAS studies have re-ignited interest in this latter model, but the lack of powerful techniques to perform peptide-hunting screens have severely hampered progress in this area. In this study, we want to apply newly developed, cutting-edge MHC class I-peptide yeast display technology to re-appraise the 'arthritogenic peptide' model of disease. For the first time, this approach offers a non-bias platform to hunt for peptides that might be linked with disease. Used in conjunction with disease-associated T cell receptor screening tools, the HLA-B*27 yeast platforms will allow high-throughput screening experiments to be performed. Our goal is that this study should dissect the relevance of this model in relation to disease pathology, whilst also driving discoveries amenable to therapeutic development.

A number of different sectors stand to benefit from this work.

It is anticipated that this study could lead to the development of new therapies that improve treatments for SpA. Current treatments for inflammatory arthritis, such as non-steroidal anti-inflammatory agents (NSAIDs) and anti-TNF therapies, can cause side effects; they work broadly and down-stream of the initiating disease trigger. If our research demonstrates that specific T cells recognise peptides that cause autoimmunity, then newer therapies that precisely target these cells could be developed. Improved therapies could also lead to an economic benefit. This would be experienced by pharmaceutical companies who steer the development of these reagents. An economic benefit could also occur in the workplace, with new and improved treatments leading to a reduction in patient morbidity, and therefore absenteeism.

If our work sheds light on the fundamental biology of inflammatory arthritis, not only will this benefit our scientific peers, but it will also improve the accuracy and breath of biological information available to second and third level science/medical students. Findings in this area should add to our overall understanding of general immunology.

Findings from our research could also inform funding bodies, including grant councils, how to best invest future funds to study inflammatory arthritis. Up to three models linking HLA-B*27 to inflammatory disease currently exist - if our research demonstrates that the disease model under investigation in this proposal is primarily involved in disease pathogenesis, then this information could inform subsequent government or charity funding initiatives (for example, via the initiation of funding 'calls' specific to this). Our study could also help guide treatments approved by the NHS, to devise regimens that are most beneficial and cost-effective.