Optimization of peptide linker length in production of MHC class II/peptide tetrameric complexes increases yield and stability, and allows identification of antigen-specific CD4+T cells in peripheral blood mononuclear cells.

Reliable, efficient systems for producing soluble HLA-DR molecules, suitable for multimerization and use as staining reagents, have proved elusive. We found that the addition of a flexible linker between peptide and N terminus of the DRB1*0101-chain (Crawford, F., Kozono, H., White, J., Marrack, P. and Kappler, J., Immunity 1998. 8: 675-682.), results in greater in vitro folding efficiency of Escherichia coli-expressed alpha- and beta-chains, and increases both the yield and stability of the DRA1*0101/DRB1*0101/peptide complexes. Although a 10-amino acid linker functioned efficiently for a 20mer epitope from HIV p24, a longer linker was required to produce a DR1 MHC class II tetramer with the influenza hemagglutinin epitope (HA(306-318)). The DR1-HA tetramer was able to stain positively over 98% of a specific clone (HA 1.7) with only a brief 30-min incubation. The tetrameric complexes detected clone cells diluted into PBMC, with high sensitivity, coupled with low background staining in CD4(+) cells. It was possible to detect antigen-specific CD4(+) T cells within a population of PBMC stimulated with the HA peptide. This demonstrates the potential to monitor CD4(+) T cell responses in peripheral blood in a number of clinical scenarios.

Perfusion chromatography for very rapid purification of class I and II MHC proteins.

Major histocompatibility complex (MHC) proteins are surface glycoproteins that are strongly associated with either self or foreign peptides. Their interaction with the T-cell receptor on the T-cells initiates an immune response and help in discriminating between self and non-self, respectively. We describe here a novel means of rapidly purifying human MHC molecules on either small scale or large scale from the cell lysate of lymphoblastoid B cell line and from insect cell culture supernatants by using affinity perfusion chromatography. As representative cases HLA-B2705, a class I MHC molecule, and HLA-DR1, a class II MHC molecule were purified from EBV-transformed human lymphoblastoid B cells, LG2. Soluble HLA-DR1 was also purified from the cell culture supernatant of insect cells. The peptides eluted from the purified HLA-B2705 were pool sequenced and found to have the same motif as has previously been published. This new method provides a very rapid means of purifying MHC protein molecules, applicable to both large scale and small scale purification, which in turn greatly enhances the accuracy of further analysis of the associated peptides through mass spectrometry.

Empty and peptide-loaded class II major histocompatibility complex proteins produced by expression in Escherichia coli and folding in vitro.

The human class II major histocompatibility complex protein HLA-DR1 has been expressed in Escherichia coli as denatured alpha and beta subunits and folded in vitro to form the native structure. DR1 folding yields are 30-50% in the presence or absence of tight-binding antigenic peptides. The protein produced in this manner is soluble and monomeric with the expected apparent molecular weight. It reacts with conformation-sensitive anti-DR antibodies and exhibits peptide-dependent resistance to SDS-induced chain dissociation and to proteolysis as does the native protein. The observed peptide specificity and dissociation kinetics are similar to those of native DR produced in B-cells and finally the protein exhibits circular dichroism spectra and cooperative thermal denaturation as expected for a folded protein. We conclude that the recombinant DR1 has adopted the native fold. We have folded DR1 in the absence of peptide and isolated a soluble, peptide-free alphabeta-heterodimer. The empty DR1 can bind antigenic peptide but exhibits altered far UV-circular dichroism and thermal denaturation relative to the peptide-bound form.

HLA-DR-restricted presentation of purified myelin basic protein is independent of intracellular processing.

Antigen presentation to CD4+ T cells involves intracellular antigen processing and loading of peptides onto newly synthesized major histocompatibility complex (MHC)-class II molecules. Some antigens, such as the lipid-bound, native form of myelin basic protein (LB-MBP) can also be presented by recycling of cell surface MHC class II molecules. The data reported here demonstrate that a preparation of highly purified, delipidated MBP (HP-MBP) follows yet another presentation pathway. Similar to LB-MBP, presentation of HP-MBP to HLA-DR1-restricted T cells was independent of HLA-DM, of newly synthesized proteins, and of invariant chain expression. However, in contrast to LB-MBP, presentation of HP-MBP was also independent of internalization of surface HLA-DR molecules. The different requirements for the presentation of the two molecular forms of MBP were further confirmed by use of the protease inhibitor E64: presentation of LB-MBP but not of HP-MBP was inhibited after treatment of target cells with E64. Furthermore, intact HP-MPB bound to isolated HLA-DR molecules in vitro with an association rate that was considerably faster than that of short peptides. These results show that presentation of HP-MBP is independent of intracellular processing and suggest that it may be presented to T cells by direct binding to surface HLA-DR molecules.

Invariant chain protects class II histocompatibility antigens from binding intact polypeptides in the endoplasmic reticulum.

Unlike class I histocompatibility (MHC) antigens, most newly synthesized MHC class II molecules fail to be loaded with peptides in the endoplasmic reticulum (ER), binding instead to the invariant chain glycoprotein (Ii). Ii blocks the class II peptide binding groove until the class II:Ii complexes are transported to endosomes where Ii is removed by proteolysis, thus permitting loading with endosomal short peptides (approximately 12-25 amino acids). Ligands from which the groove is protected by Ii have not yet been identified; theoretically they could be short peptides or longer polypeptides (or both), because the class II groove is open at both ends. Here we show that in Ii- deficient cells, but not in cells expressing large amounts of Ii, a substantial fraction of class II alpha beta dimers forms specific, SDS-resistant 1:1 complexes with a variety of polypeptides. Different sets of polypeptides bound to H-2Ak, Ek, Ed and HLA-DR1 class II molecules; for Ak, a major species of Mr 50 kDa (p50) and further distinct 20 and 130 kDa polypeptides were detectable. Class II binding of p50 was characterized in detail. Point mutations within the Ak antigen binding groove destabilized the p50:class II complexes; a mutation outside the groove had no effect. A short segment of p50 was sufficient for association with Ak. The p50 polypeptide was synthesized endogenously, bound to Ak in a pre-Golgi compartment, and was transported to the cell surface in association with Ak. Thus, Ii protects the class II groove from binding endogenous, possibly misfolded polypeptides in the ER. The possibility is discussed that polypeptide binding is an ancestral function of the MHC antigen binding domain.

Isolation of HLA-DR1.(staphylococcal enterotoxin A)2 trimers in solution.

Mutational studies indicate that the superantigen staphylococcal enterotoxin A (SEA) has two separate binding sites for major histocompatibility complex (MHC) class II molecules. Direct evidence is provided here for the formation of SEA-MHC class II trimers in solution. Isoelectric focusing separated SEA-HLA-DR1 complexes into both dimers and HLA-DR1.SEA2 trimers. The molar ratio of components was determined by dual isotope labeling. The SEA mutant SEA-F47S, L48S, Y92A, which is deficient in MHC class II alpha-chain binding, formed only dimers with HLA-DR1, whereas a second SEA mutant, SEA-H225A, which lacks high-affinity MHC class II beta-chain binding was incapable of forming any complexes. Thus SEA binding to its MHC receptor is a two-step process involving initial beta-chain binding followed by cooperative binding of a second SEA molecule to the class II alpha chain.

Assembly and peptide binding of major histocompatibility complex class II heterodimers in an in vitro translation system.

In vitro transcription/translation of HLA-DR1 cDNAs in the presence of microsomal membranes was used to study the association of major histocompatibility complex class II molecules with peptide and invariant chain (Ii) in the endoplasmic reticulum (ER). HLA-DR alpha and HLA-DR beta subunits assembled into SDS-unstable heterodimers in the absence of exogenous peptide. The inclusion of synthetic peptides during the alpha/beta assembly process promoted their conversion to SDS-resistant heterodimers. Addition of Ii RNA during the translation of HLA-DR alpha and HLA-DR beta RNAs resulted in the formation of alpha/beta/Ii complexes. Peptide binding by class II molecules was detected even when excess Ii was present during alpha/beta assembly. These findings indicate that peptides can bind alpha/beta heterodimers in the ER microenvironment and suggest that peptides derived from cytosolic proteins that are presented by class II molecules at the cell surface may have bound to HLA-DR in the ER.

Identification of a motif for HLA-DR1 binding peptides using M13 display libraries.

Oligonucleotides encoding peptides known to bind to HLA-DR1 molecules have been inserted into the gene III of filamentous M13 phages. DR1 molecules purified from human lymphoblastoid cell lines could specifically bind to these peptide sequences expressed on the phage surface. A M13 phage peptide library was next constructed and screened with DR1 molecules. After four rounds of selection, more than 80% of the phages were able to bind to DR1. Competition experiments with both isolated phages and corresponding synthetic peptides showed that the binding was specific. Sequence analysis of the peptide encoding region of 60 phages binding to DR1 molecules and comparison with phages of the original library revealed two potential anchor positions. The first was an aromatic residue (Tyr, Phe, or Trp) at the NH2 terminus of the peptide sequences, and the second was located three residues downstream and consisted of Met or Leu. In addition, the negatively charged amino acids Asp and Glu were mostly excluded from the DR1 binding sequences, and the small amino acid residues Gly and Ala were enriched at position 6. As for DR1, this approach should enable one to easily determine the binding motifs of other MHC class II alleles and isotypes. Furthermore, it could have interesting applications in the design of major histocompatibility complex-specific antagonists.

Zinc regulates the function of two superantigens.

Staphylococcal enterotoxins bind with high affinity to class II major histocompatibility complex proteins and subsequently stimulate large numbers of T cells via the V beta portion of the T-cell receptor. Binding of enterotoxin A and enterotoxin E to HLA-DR was completely abolished by low levels of EDTA, whereas binding of toxic shock toxin was unaffected. Addition of Zn2+ to as little as 2 microM excess over EDTA completely reconstituted binding, but Ca2+, Mg2+, Cu2+, Fe2+, and Mn2+ had no effect. The dissociation constant (Kd) of 65Zn2+ binding to a single site on purified enterotoxin A was 2 microM, and addition of purified HLA-DR1 did not alter the Kd, indicating that the binding site was exclusive to enterotoxin A. In the presence of saturating levels of zinc the Kd for enterotoxin A binding to purified HLA-DR1 was 25 nM. Thus, zinc binding is an essential first step in the formation of the major histocompatibility complex binding domain of at least two bacterial superantigens. Given the measured Kd of zinc binding to enterotoxin A, serum levels of free zinc (0.2-1.0 microM) may well regulate the toxic sequelae by these two superantigens.

HLA antigens in Yugoslav population with rheumatoid arthritis.

The antigens of HLA-DR locus were determined in 127 patients with rheumatoid arthritis and in 175 healthy persons. The frequency of HLA-DR1 antigen was discovered in 45% of the patients and in 22.3% of the examinees from the control group (p less than 0.001). The relative risk is 2.84. 69% of the patients were seropositive. The HLA-DR4 antigen was found in 34% of the patients and in 21% of the healthy examinees (p less than 0.05). The relative risk is 1.91. The rheumatoid factor was determined in 70% of the patients with this antigen.