Research background
Molecular mimicry between microbial and self peptides has been proposed for decades as a route by which infection or commensal colonization might initiate autoimmunity, but convincing examples that pair a defined human autoreactive T-cell receptor with a defined bacterial ligand have been comparatively rare. SLPRLFPLL is one such ligand.
The peptide derives from the ATP-binding protein of a glutathione ABC transporter of Klebsiella oxytoca, encoded by gsiA. Dolton and colleagues arrived at it by screening a combinatorial peptide library against the 4C6 T-cell receptor, which was obtained in the setting of type 1 diabetes and recognizes the HLA-A*24:02-restricted preproinsulin 3-11 epitope LWMRLLPLL, then searching sequence databases for natural matches to the preferred motif. In functional assays the bacterial sequence proved almost 30,000 times more potent than the natural self peptide for this receptor, a difference far larger than is usually seen between a self antigen and a cross-reactive foreign one.
Importantly, the study did not rely on synthetic peptide pulsing alone. Bacterial genes were introduced into target cell lines by lentiviral transduction, demonstrating that the sequence can be generated and presented from full-length protein through normal cellular processing rather than only when supplied exogenously. Alignment against the preproinsulin register shows the two peptides share a central PRLF-containing core and comparable hydrophobic anchors while differing at several solvent-exposed positions, which is the pattern expected of genuine degenerate recognition rather than sequence identity. The findings have been discussed in accompanying commentary as an argument that microbial mimics can present the autoreactive repertoire with agonists considerably stronger than the self antigen itself.
LT9750 is supplied for research use as a defined cross-reactive ligand for comparison against LT9748 and the screened superagonist LT9749. Potency measured with the 4C6 receptor is specific to that receptor and should not be generalized to other HLA-A*24:02-restricted T cells, nor interpreted as evidence of causation in human disease.
Research applications
- Molecular mimicry studies pairing defined bacterial ligands with human autoreactive T-cell receptors
- Potency comparisons and dose-response titration against the natural preproinsulin epitope
- HLA-A*24:02 multimer assembly for detection of cross-reactive CD8+ T-cell populations
- Cross-reactivity panels assembled to quantify degenerate peptide recognition
- Antigen-processing controls alongside cells transduced with the source bacterial gene
- Microbiome and infection-related studies of autoimmune T-cell priming
- Peptide-MHC binding and stability measurement for a bacterial ligand of a susceptibility allele
References
- Dolton G, Bulek A, Wall A, et al. HLA A*24:02-restricted T cell receptors cross-recognize bacterial and preproinsulin peptides in type 1 diabetes. Journal of Clinical Investigation (2024) 134:e164535. doi:10.1172/JCI164535
- Microbial mimics supersize the pathogenic self-response. Journal of Clinical Investigation (2024), commentary. JCI 2024 commentary
- Human islet T cells are highly reactive to preproinsulin in type 1 diabetes. PNAS (2021) 118:e2107208118. doi:10.1073/pnas.2107208118
- Circulating preproinsulin signal peptide-specific CD8 T cells restricted by the susceptibility molecule HLA-A24 are expanded at onset of type 1 diabetes and kill beta cells. Diabetes (2012) 61:1752-1759. Diabetes 2012;61:1752
The 4C6 cross-reactivity panel and related peptides
For research use only. Not for use in diagnostic or therapeutic procedures, or for human or veterinary use.
