We performed model-building of peptides associated with the NOD mouse MHC class II molecule, I-Ag7. All possible 13mer peptides were derived from the M3 muscarinc receptor protein sequence. The docking of potential peptide fragments from the M3 receptor was guided through comparison with the X-ray structure of the GAD peptide YEIAPVFVLLEYVT and I-Ag7. These potential models were solvated using an in-house algorithm that was developed to examine the role of water in mediating peptide-MHC binding. The energetics of the peptide-MHC-water complexes were evaluated using the AMBER6 forcefield. These calculations were used to determine if the potential antigenic peptides remained bound in the groove (an indication of stability) or whether the peptides tended to show dissociation from the binding groove. In addition to these results, the hydropathy profile of the M3 receptor was used to determine which 13mers were likely to be produced by cathepsin-mediated cleavage in the immune system process, which would trigger the response for SjS.

Lacrimal glands from male and female BALB/c and NOD mice at 1, and 4 months were isolated. For confocal microscopy analysis, glands were cut into small pieces and fixed in 4% paraformaldehyde, 4% sucrose for 3 hours. Samples were cryoprotected overnight in 30% sucrose solution, placed in OCT and snap frozen with liquid N2 before preparation of 5 micrometer sections. Commercially available antibodies were utilized for the M3 muscarinic receptor and lysosomal membranes. Appropriate secondary antibodies conjugated to fluorophores localized the fluorescence associated with these proteins. Dual labeling with other antibodies or affinity label served to localize the following cellular compartment as previously described: apical plasma membrane, basolateral plasma membrane, Golgi apparatus, endosomes, lysosomes, and secretory vesicles. Specimens were imaged using a Zeiss LSM 510 Meta NLO imaging system equipped with visible and IR lasers for standard and multiphoton illumination; signal intensity associated with each protein as well as their colocalization with different compartments were quantified in multiple samples with the software available on this system.