Differential sensitivity to mutations in a single peptide by two TCRs having identical beta-chains and closely related alpha-chains.

The TCR on CD4 T cells binds to and recognizes MHC class II:antigenic peptide complexes through molecular contacts with the peptide amino acid residues that face up and out of the peptide-binding groove. This interaction primarily involves the complementarity-determining regions (CDR) of the TCR alpha- and ss-chains contacting up to five residues of the peptide. We have used two TCRs that recognize the same antigenic peptide and have identical Vss8.2 chains, but differ in all three CDR of their related Valpha2 chains, to examine the fine specificity of the TCR:peptide contacts that lead to activation. By generating a peptide library containing all 20 aa residues in the five potential TCR contact sites, we were able to demonstrate that the two similar TCRs responded differentially when agonist, nonagonist, and antagonist peptide functions were examined. Dual substituted peptides containing an agonist residue at the N terminus, which interacts with CDR2alpha, and an antagonist residue at the C terminus, which interacts with the CDR3ss, were used to show that the nature of the overall signal through the TCR is determined by a combination of the type of signal received through both the TCR alpha- and ss-chains.

Binding specificity of avian heat shock protein 108.

Chicken heat shock protein 108 (HSP108), the avian homolog of GRP94, was originally isolated from hen oviduct and binds Fe-ovotransferrin (Fe-OTf). The liver is also a rich source, and liver membranes bind Fe-OTf with a KD of 1.7 x 10(-7) M, a value similar to oviduct membranes. A competition assay, based on the binding of 125I-Fe-OTf to liver membranes, was utilized to examine the binding specificity of HSP108. Ovalbumin and avidin competed effectively, with KD's of 1.8 x 10(-7) M and 1.4 x 10(-7) M, respectively. Iron-free OTf bound with a 10-fold higher KD. Egg white lysozyme, chicken IgG, human transferrin, rabbit muscle actin, and porcine insulin do not bind. Neither do denatured ovalbumin or ovalbumin tryptic peptides. Thus, the binding activity of HSP108 is not restricted to Fe-OTf, nor is it universal.

Crystal structure of diferric hen ovotransferrin at 2.4 A resolution.

The three-dimensional structure of diferric hen ovotransferrin has been determined by X-ray crystallography at 2.4 A resolution. The structure was solved by molecular replacement, using the coordinates of diferric human lactoferrin as a search model. Several rounds of simulated annealing and restrained least-squares refinement have resulted in a model structure with an R-factor of 0.171 for the data between 11.0 and 2.4 A resolution. The model comprises 5284 protein atoms (residues 5 to 686), 2 Fe3+, 2 CO3(2)- and 132 water molecules. The overall structure of ovotransferrin is similar to those of human lactoferrin and rabbit serum transferrin, being folded into two homologous lobes, each containing two dissimilar domains with one Fe3+ and one CO3(2)- bound at a specific site in each interdomain cleft. However, the relative orientation of the two lobes, which may be related to the class specificity of transferrins to receptors, is different from either human lactoferrin or rabbit serum transferrin. The angle of the relative orientation in ovotransferrin is increased by 6.8 degrees and 15.7 degrees as compared with to those in rabbit serum transferrin and human lactoferrin, respectively. Interdomain Lys209-Lys301 and Gln541-Lys638 interactions are found near the metal binding site of each lobe. The interlobe interactions and their role in the stabilization of iron binding are discussed.

Different stability of N- and C-domain of diferric ovotransferrin in urea and application to the determination of iron distribution between the two domains.

The study of guanidine-HCl or thermal denaturation of diferric ovotransferrin (Fe2Tf) has revealed a simultaneous unfolding of the two domains of the protein (Ikeda et al. (1985) FEBS Lett. 182, 305-309). In urea denaturation of Fe2Tf, however, two distinct steps of unfolding were observed in the urea concentration range from 4.5 to 9 M at pH 8.0 and 37 degrees C by measuring the residual iron-bound protein (absorbance at 465 nm) and the remaining folded structures (circular dichroism at 222 nm). From a study of urea denaturation of partially iron-saturated Tf whose iron preferentially occupied the N-domain, it was found that the first and the second steps of denaturation corresponded to those of the N-terminal (4.5-6 M urea) and C-terminal domains (over 7 M urea), respectively. The N-domain of Fe2Tf was selectively unfolded in 7 M urea and digested with trypsin to provide an iron-bound C-terminal fragment (42 kDa) in good yield (about 80% of theoretical). The kinetic analysis of the decrease in A465 of Fe2Tf in 9 M urea showed that the N-domain unfolded 3 x 10(2) times faster than the C-domain. With partially iron-saturated Tf, the decrease of A465 in 9 M urea also proceeded in a biphasic manner and the ratio, the decrement in A465 of the rapid phase/the decrement in A465 of the slow phase, gave the value of iron distribution as Fe at the N-site/Fe at the C-site.