21-Hydroxylase-Specific CD8+ T Cells in Autoimmune Addison’s Disease Are Restricted by HLA-A2 and HLA-C7 Molecules

In this paper we describe an improved neural network method to predict T-cell class I epitopes. A novel input representation has been developed consisting of a combination of sparse encoding, Blosum encoding, and input derived from hidden Markov models. We demonstrate that the combination of several neural networks derived using different sequence-encoding schemes has a performance superior to neural networks derived using a single sequence-encoding scheme. The new method is shown to have a performance that is substantially higher than that of other methods. By use of mutual information calculations we show that peptides that bind to the HLA A*0204 complex display signal of higher order sequence correlations. Neural networks are ideally suited to integrate such higher order correlations when predicting the binding affinity. It is this feature combined with the use of several neural networks derived from different and novel sequence-encoding schemes and the ability of the neural network to be trained on data consisting of continuous binding affinities that gives the new method an improved performance. The difference in predictive performance between the neural network methods and that of the matrix-driven methods is found to be most significant for peptides that bind strongly to the HLA molecule, confirming that the signal of higher order sequence correlation is most strongly present in high-binding peptides. Finally, we use the method to predict T-cell epitopes for the genome of hepatitis C virus and discuss possible applications of the prediction method to guide the process of rational vaccine design.

Many biological processes are guided by receptor interactions with linear ligands of variable length. One such receptor is the MHC class I molecule. The length preferences vary depending on the MHC allele, but are generally limited to peptides of length 8-11 amino acids. On this relatively simple system, we developed a sequence alignment method based on artificial neural networks that allows insertions and deletions in the alignment.

Molecular mimicry is one of the leading mechanisms by which infectious or chemical agents may induce autoimmunity. It occurs when similarities between foreign and self-peptides favor an activation of autoreactive T or B cells by a foreign-derived antigen in a susceptible individual. However, molecular mimicry is unlikely to be the only underlying mechanism for autoimmune responses; other factors such as breach in central tolerance, non-specific bystander activation, or persistent antigenic stimuli (amongst others) may also contribute to the development of autoimmune diseases. Host genetics, exposure to microbiota and environmental chemicals are additional links to our understanding of molecular mimicry. Our current knowledge of the detailed mechanisms of molecular mimicry is limited by the issues of prolonged periods of latency before the appearance of disease, the lack of enough statistical power in epidemiological studies, the limitations of the potential role of genetics in human studies, the relevance of inbred murine models to the diverse human population and especially the limited technology to systematically dissect the human T-cell repertoire and B-cell responses. Nevertheless, studies on the role of autoreactive T-cells that are generated secondary to molecular mimicry, the diversity of the T-cell receptor repertoires of auto-reactive T-cells, the role of exposure to cryptic antigens, the generation of autoimmune B-cell responses, the interaction of microbiota and chemical adjuvants with the host immune systems all provide clues in advancing our understanding of the molecular mechanisms involved in the evolving concept of molecular mimicry and also may potentially aid in the prevention and treatment of autoimmune diseases.