It is generally accepted that strong association exists between particular MHC class II molecules and autoimmune diseases, but the mechanisms underlying this association are unknown. The genetic link between the HLA-DQ2 allele and celiac disease and type 1 insulin-dependent diabetes has been appreciated long time ago, the explanation for this association, however, remains unknown. MHC II molecules are cell surface proteins that are able to present antigenic peptides to CD4+ T cells. The loading of peptides into MHC II is regulated by two intracellular proteins called invariant chain (Ii) and HLA-DM. Ii stabilizes newly synthesized MHC II and facilitates the movement of MHC II from the ER to endosomes, where Ii is broken down, leaving a small fragment called CLIP in the peptide-binding region. The major function of HLA-DM is to remove CLIP exchanging for an antigenic peptide from endosomes and edit peptide repertoire presented by class II. Our previous data show that HLA-DQ2 is associated with two different CLIP peptides: the traditional CLIP1 and the unusual CLIP2.
In aim 1, I will determine whether this unusual association is initiated within the full-length Ii in ER or in endosomes after some proteolysis of Ii. To answer this question, I will introduce mutations into Ii that eliminate CLIP1 or CLIP2 or both binding to DQ2. The predominant peptide species associated with DQ2 in ER will then be identified by both pulse chase immunoprecipitation and MALDI-TOF/TOF MS. The results will tell us whether both registers are used for Ii binding in the ER. If not, the implication will be that the CLIP2 register is generated after endosomal processing of Ii and likely by sliding of CLIP in the groove to the CLIP2 register. Our previous studies also indicate the interaction between HLA-DQ2 and HLA-DM is weaker compared to HLA-DR3 and HLA-DQ1. To further characterize DQ2-DM interaction, I will elucidate the structural basis of the DQ2 resistance to DM effect using mutational analysis in Aim 2. Fourteen different mutations will be introduced into DQ2 based on previous studies of DM interaction with other MHC class II alleles, which are expected to improve DM-DQ2 interaction. Both DQ2 WT and mutants will be stably expressed in both DM+ and DM- lymphocytes. Additionally, soluble DQ2 WT or mutants will be expressed in and purified from insect cells. The effects of mutations on DQ2 function will be investigated by measuring 1) surface expression of CLIP;2) efficiency of CLIP release from DQ2;3) peptide exchange rate of DQ2;4) binding affinity of gliadin peptides to DQ2;5) DM-meditated conformational change in DQ2;6) binding affinity between DM and soluble DQ2. If none of the above mutations will improve DQ2-DM interaction, random mutagenesis of DQ2 will be employed to identify mutations increasing DQ2 sensitivity to DM function. The identification of mutations in DQ2 will be ultimately used to determine the role of DM-DQ2 interaction in T cell selection in transgenic mice.
Genetic and epidemiological data have revealed that the strong genetic link between HLA-DQ2 and celiac disease and type 1 insulin-dependent diabetes, but the mechanisms underlying this association are unknown. I propose to elucidate the mechanisms that HLA-DQ2 predisposes to autoimmune diseases by exploring the defective interaction between DQ2 and DM (and DQ2's association with unusual invariant chain peptides), both of which play central role in regulating antigen presentation by MHC class II. Our results will provide important mechanistic insights into the unique features of DQ2 interaction with the antigen presentation machinery, which may ultimately lead to improved understanding of the association of DQ2 with autoimmunity and may suggest new therapeutic approaches for DQ2-associated autoimmune disorders.
