Role of Recurrent Hydrophobic Residues in Catalyzing Helix Formation by T Cell-Presented Peptides: a Thesis

Abstract

The overall objective of this study was to understand the mechanisms that control antigen processing and binding of peptides to major histocompatibility complex (MHC) molecules. Towards this goal I investigated (a) the structural features of T cell-presented peptides with a focus on the role of recurrent hydrophobic residues in catalysis of helix formation by these peptides and (b) the biochemical events that determine the fates of the invariant chain molecule (Ii) in its various post-translational processing pathways. In the structural studies, I tested the hypothesis that the recurrence of hydrophobic amino acids in a polypeptide at positions falling in an axial, hydrophobic strip if the sequence were coiled as an α-helix can lead to helical nucleation on a hydrophobic surface.For a series of HPLC-purified peptides, including some T cell-presented peptides varying considerably in primary sequence, percentage helicity in the presence of lipid vesicles correlated with strip-of-helix hydrophobicity index (SOHHI), as shown by circular dichroism (CD) analysis. A prototypic helix peptide PH-1.0 (LYQELQKLTQTLK) was designed with a strong axial hydrophobic strip of 4 leucine residues. PH-1.0 formed about 38% helical structure in 10 mM phosphate buffer at pH 7.0 with di-O-hexadecyl phosphatidylcholine (DHPC) lipid vesicles, but no helical structure was detected when the peptide was in phosphate buffer alone. The helix-forming tendencies of 9 analogs of PH-1.0 with one or two amino acid variations from the parent peptide were examined in the presence of lipid vesicles and the results showed that (a) decreasing the strip-of-helix hydrophobicity by substituting even one of the four leucine residues in the axial hydrophobic strip with a less hydrophobic threonine residue reduced the helix-forming tendency of a peptide in the presence of lipid vesicles; (b) the placement of recurrent hydrophobic residues within the axial hydrophobic strip appeared to be critical for a peptide to be induced to form an α-helix by a hydrophobic surface; (c) there was an orientation preference for these peptides to interact with lipid vesicles and to form helical structure; (d) extra hydrophobic residues in other positions appeared to compete with the hydrophobic residues within the axial hydrophobic strip for interaction with the lipid vesicles and therefore to decrease the helix-forming tendency of peptides. For the biochemical studies of the function of Ii, a 17-residue peptide, Ii-3 (Ii 148-164), was synthesized. The CD analysis of Ii-3 showed mainly an α-helical conformation when the peptide was examined in the presence of lipid vesicles. [125I]-labeled Ii-3, after coupling at the N-terminus with a photoactivatable, heterobifunctional crosslinker N-hydroxysuccinimidyl-4-azidobenzoate (HSAB), was able to bind to both α and β chains of class II MHC molecules, indicating that this region of Ii might cover the desetope of class II MHC molecules from the time of their synthesis until their charging with foreign peptides at an endosomal compartment. The biosynthesis of a chondroitin sulfate proteoglycan-form of Ii (CS-Ii) was examined in a class II MHC-negative cell line P3HR-1. [35S]sulfate-labeled microsomal membrane proteins of P3HR-1 were immunoprecipitated with anti-Ii monoclonal antibody and the results of SDS-PAGE analysis demonstrated that P3HR-1 could process Ii to CS-li in the absence of class II MHC molecules and the chondroitin sulfate identity of this molecule was confirmed by chondroitinase-ABC treatment. We conclude that there might be a class II MHC-independent pathway to process Ii to a chondroitin sulfate proteoglycan form as compared to the pathway in which Ii was associated with class II MHC and later cleaved by proteases residing in the endosomal compartment. In an effort to demonstrate in vitro that the class II MHC-associated Ii was eventually dissociated from class II MHC molecules by a proteolytic cleavage process, it was found that cathepsin B could completely remove Ii without damage to class II α and β chains. In order to identify those cleaved Ii fragments, three polyclonal anti-Ii peptide sera were produced by immunizing rabbits with keyhole limpet hemocyanin (KLH)-conjugated Ii peptides. Anti-Ii (146-169) was shown to be able to precipitate a p18 molecule only in cells expressing Ii. Anti-Ii (148-164 )and anti-Ii(78-92) were specific for their respective antigenic peptides as tested by enzyme-linked immunosorbent assay (ELISA).