Evolutionary conservation of major histocompatibility complex-DR/peptide/T cell interactions in primates

Strategies for tolerance induction in nonhuman primates

Organ transplants in nonhuman primates provide a model which closely simulates the biological conditions of human organ transplantation, due to similarities between human and primate MHC (class I and II) structure and expression. Several strategies for tolerance induction have been developed in nonhuman primate models. These include targeting the T cell receptor or costimulatory molecules and the generation of mixed chimerism. Tolerance can be reliably induced in several such models, although none with 100% success.

The common marmoset: a new world primate species with limited Mhc class II variability

The common marmoset (Callithrix jacchus) is a New World primate species that is highly susceptible to fatal infections caused by various strains of bacteria. We present here a first step in the molecular characterization of the common marmoset's Mhc class II genes by nucleotide sequence analysis of the polymorphic exon 2 segments. For this study, genetic material was obtained from animals bred in captivity as well as in the wild. The results demonstrate that the common marmoset has, like other primates, apparently functional Mhc-DR and -DQ regions, but the Mhc-DP region has been inactivated. At the -DR and -DQ loci, only a limited number of lineages were detected. On the basis of the number of alleles found, the -DQA and -B loci appear to be oligomorphic, whereas only a moderate degree of polymorphism was observed for two of three Mhc-DRB loci. The contact residues in the peptide-binding site of the Caja-DRB1*03 lineage members are highly conserved, whereas the -DRB*W16 lineage members show more divergence in that respect. The latter locus encodes five oligomorphic lineages whose members are not observed in any other primate species studied, suggesting rapid evolution, as illustrated by frequent exchange of polymorphic motifs. All common marmosets tested were found to share one monomorphic type of Caja-DRB*W12 allele probably encoded by a separate locus. Common marmosets apparently lack haplotype polymorphism because the number of Caja-DRB loci present per haplotype appears to be constant. Despite this, however, an unexpectedly high number of allelic combinations are observed at the haplotypic level, suggesting that Caja-DRB alleles are exchanged frequently between chromosomes by recombination, promoting an optimal distribution of limited Mhc polymorphisms among individuals of a given population. This peculiar genetic make up, in combination with the limited variability of the major histocompatability complex class II repertoire, may contribute to the common marmoset's susceptibility to particular bacterial infections.

Characterization of the Peptide Binding Motif of a Rhesus MHC Class I Molecule (Mamu-A*01) That Binds an Immunodominant CTL Epitope from Simian Immunodeficiency Virus

The majority of immunogenic CTL epitopes bind to MHC class I molecules with high affinity. However, peptides longer or shorter than the optimal epitope rarely bind with high affinity. Therefore, identification of optimal CTL epitopes from pathogens may ultimately be critical for inducing strong CTL responses and developing epitope-based vaccines. The SIV-infected rhesus macaque is an excellent animal model for HIV infection of humans. Although a number of CTL epitopes have been mapped in SIV-infected rhesus macaques, the optimal epitopes have not been well defined, and their anchor residues are unknown. We have now defined the optimal SIV gag CTL epitope restricted by the rhesus MHC class I molecule Mamu-A*01 and defined a general peptide binding motif for this molecule that is characterized by a dominant position 3 anchor (proline). We used peptide elution and sequencing, peptide binding assays, and bulk and clonal CTL assays to demonstrate that the optimal Mamu-A*01-restricted SIV gag CTL epitope was CTPYDINQM181–189. Mamu-A*01 is unique in that it is found at a high frequency in rhesus macaques, and all SIV-infected Mamu-A*01-positive rhesus macaques studied to date develop an immunodominant gag-specific CTL response restricted by this molecule. Identification of the optimal SIV gag CTL epitope will be critical for a variety of studies designed to induce CD8+ CTL responses specific for SIV in the rhesus macaque.

Liposome-mediated peptide loading of MHC-DR molecules in vivo

Amino acid residues 3-15 of mycobacterial HSP60 define a dominant T-cell epitope for HLA-DR3+ve humans and Mamu-DR3+ve rhesus monkeys. Our results show that Mamu-DR3 molecules on PBMC can be efficiently loaded in vivo with the above-mentioned peptides when they are intravenously injected encapsulated in liposomes, but not in the free form. Mamu-DR3 loading is abolished by encapsulation of a nonstimulatory peptide. These results have implications for the delivery of therapeutic peptides in vivo.

FN18-CRM9 immunotoxin promotes tolerance in primate renal allografts

BACKGROUND

Transplant tolerance, rather than immunity, may be favored in the setting of a lower mature lymphoid mass in the recipient induced by anti-T cell agents. A novel immunosuppressive agent, FN18-CRM9, known to specifically kill T cells with great potency, was evaluated in a transplant model.

METHODS

In order to ablate recipient T cells, the immunotoxin FN18-CRM9 was administered to rhesus monkey recipients of MHC-mismatched renal allografts. Donor lymphocytes were injected intrathymically into some animals.

RESULTS

All monkeys with T-cell depletion by immunotoxin had prolonged allograft survival, and tolerance confirmed by skin grafting has been confirmed in five of six long-surviving recipients.

CONCLUSIONS

In this clinically relevant model, profound but transient T-cell depletion by a single agent substantially promotes tolerance.

Diversity and evolution of the DRB1*03 family: description of DRB1*03022,*0307,*0308

Three previously unreported DRB1*03 alleles are described, adding to the diversity of the DRB1 family of alleles. DRB1*03022 contains a silent substitution at codon 77. DRB1*0307 differs from DRB1*03011 by a substitution at codon 26 resulting in a predicted change from tyrosine to phenylalanine. DRB1*0308 is almost identical to DRB1*03011 differing at codon 58 which specifies the glutamic acid residue commonly found in DRB1*11 alleles. The new alleles (DRB1*03022,*0307,*0308) may have arisen by gene conversion-like events and add to the increasing complexity of the HLA system.

The impact of DR3 microvariation on peptide binding: the combinations of specific DR beta residues critical to binding differ for different peptides

HLA-DR molecules are a group of highly polymorphic glycoprotein heterodimers that present peptide antigens to T lymphocytes for immune surveillance. To assess the significance of limited polymorphism on the functional differentiation of DR molecules, the binding of several immunogenic peptides to the DR3 microvariants [DR(alpha, beta 1*0302) and DR(alpha, beta 1*0301)] and to mutants of these DR3 molecules was examined. This analysis has shown that each residue (DR beta 26, DR beta 28, DR beta 47, and DR beta 86), which differentiates these two DR3 molecules, contributes to their functional distinction and that the relative contribution of each residue varies for different peptide/DR3 complexes. For example, DR beta 28 and DR beta 86 controlled the mycobacterium tuberculosis 65-kD heat shock protein peptides 3-13 and 4-15 (HSP) binding specificity to DR (alpha, beta 1*0301). [HSP does not bind to DR(alpha, beta 1*0302)], whereas DR beta 26, DR beta 28, and DR beta 86 controlled the influenza hemagglutinin peptide 306-318 (HA) binding specificity to DR(alpha, beta 1*0302). [HA does not bind to DR(alpha, beta 1*0301).] In comparison, DR beta 86 alone controlled the binding level difference of sperm whale myoglobin peptide 132-151 (SWM) and of myelin basic protein peptide 152-170 (MBP) [both bind to DR(alpha, beta 1*0301) at levels five times greater than to DR(alpha, beta 1*0302)] to the DR3 molecules. Although not critical, additional DR beta residues influenced the binding level of individual peptides of each of the DR3 molecules and, again, the combinations of these residues differed for different peptide/DR3 complexes. These data showed that individual DR residues vary in their relative contribution to the interaction between a specific DR molecule and different peptides and that limited polymorphism can create substantial differences in the peptide binding profiles among DR molecules.

The presentation of a hepatitis C viral peptide by distinct major histocompatibility complex class I allotypes from two chimpanzee species

A cytotoxic T lymphocyte (CTL) line, derived from the liver of a common chimpanzee (Pan troglodytes) with hepatitis C, specifically recognized a hepatitis C viral 9-mer peptide (KHP-DATYSR in single-letter amino acid code) bound by the major histocompatibility complex (MHC) class I molecule, Patr-A04. This same CTL line also recognized the identical peptide bound by a structurally different class I molecule, Papa-A06, derived from the separate chimpanzee species, Pan paniscus or pygmy chimpanzee. These class I allotypes differ by six amino acids but, in spite of the structural differences, share the same antigen-presenting function. This is the first observation of antigen presentation to a given T cell receptor by different MHC class I allotypes from separate species.

Characterization of the natural immune response of rhesus monkey CD4+ve T cells to the bacterial antigen streptolysin O (SLO)

Rhesus monkeys show a high proliferative T cell response to the bacterial exotoxin SLO without prior immunization. The present study was undertaken to characterize this naturally present SLO-responsiveness with particular emphasis on CD4+ve reactive T cells. It is demonstrated that the frequency of SLO-reactive cells in the circulation.ranges between 1 in 75 and 1 in 610 CD4+ve T cells as determined with limiting dilution analysis. It is also shown that induction of a good proliferative response requires Mhc-DR matching between T cell and the antigen presenting cells (APC). Stable and DR-restricted SLO-specific CD4+ve T cell lines were generated from CD8 depleted peripheral blood mononuclear cells (PBMC). The SLO-reactive CD4+ve cell lines are tentatively characterized as Th1-like based on the predominant production of interferon-gamma (IFN-gamma) over IL-4, although this seems contradicted by the IL-4 dependent growth of the lines.

HLA-B alleles of the Cayapa of Ecuador: new B39 and B15 alleles

Recent data suggest that HLA-B locus alleles can evolve quickly in native South American populations. To investigate further this phenomenon of new HLA-B variants among Amerindians, we studied samples from another South American tribe, the Cayapa from Ecuador. We selected individuals for HLA-B molecular typing based upon their HLA class II typing results. Three new variants of HLA-B39 and one new variant of HLA-B15 were found in the Cayapa: HLA-B*3905, HLA-B*3906, HLA-B*3907, and HLA-B*1522. A total of thirteen new HLA-B alleles have now been found in the four South American tribes studied. Each of these four tribes studied, including the Cayapa, had novel alleles that were not found in any of the other tribes, suggesting that many of these new HLA-B alleles may have evolved since the Paleo-Indians originally populated South America. Each of these 13 new alleles contained predicted amino acid replacements that were located in the peptide binding site. These amino acid replacements may affect the sequence motif of the bound peptides, suggesting that these new alleles have been maintained by selection. New allelic variants have been found for all common HLA-B locus antigenic groups present in South American tribes with the exception of B48. In spite of its high frequency in South American tribes, no evidence for variants of B48 has been found in all the Amerindians studied, suggesting that B48 may have unique characteristics among the B locus alleles.

Microvariation creates significant functional differences in the DR3 molecules

Two DR3 molecules differ by four amino acids whose side chains point into the DR antigen-binding groove. To begin to assess the role of microvariation on DR3 function, DRB1*0302 residues were replaced with DRB1*0301 residues at beta-chain positions 26, 47, 86, and 47 plus 86. Murine fibroblast cell lines expressing DR(alpha, beta 1*0301), DR(alpha, beta 1*0302), and the four mutant 0302 molecules were examined for alloproliferative DR(alpha, beta 1*0302)-specific TLC stimulation and peptide binding. Changing position 26 had the most profound effect on T-cell recognition (seven of nine TLCs did not respond). Two TLCs did not respond to the mutant 0302V86 molecule and four TLCs that did respond to this mutant lost responsiveness when positions 47 and 86 were mutated together. These data suggest that each of these variant residues, including position 47, influence T-cell recognition. Surprisingly, none of the mutations had an effect on the absolute binding of HA 307-319 (DR[alpha, beta 1*0302] specific) and HSP 3-13 (DR[alpha, beta 1*0301] specific); however, the mutant 0302 molecules changed at position 86 (glycine to valine) consistently bound HA 307-319 at significantly higher levels than DR(alpha, beta 1*0302). These data for position 86 are in contrast to other DR molecules and indicate that peptide contact residues for a specific DR molecule cannot be predicted based on binding results obtained with other DR molecules. These data suggest that each of these variant groove residues, although not accessible to the TCR, contribute to the significant functional differences between the DR3 microvariants through subtle influences on the DR3-peptide complex.

T cell receptor and peptide-contacting residues in the HLA-DR17(3) beta 1 chain

Previously, we have proposed that the beta 1 residues 9-13, 26, 28 and 86 in HLA-DR17, the most common subtype of DR3, might be critical for the binding of an immunodominant, mycobacterial epitope (peptide 3-13 of the 65-kDa heat shock protein). In order to examine directly (i) which DR17 residues are involved in peptide binding, (ii) whether the same or other DR17 residues are involved in the binding of different peptides, and (iii) whether subtle differences in the mode of peptide binding can influence T cell stimulation, we have now systematically mutated 15 highly polymorphic DR17 beta 1 residues, located in the proposed peptide binding groove of DR17, and examined the effect thereof on binding and presentation of two peptides, hsp65 p3-13 and p56-65 of the 30/31-kDa secreted mycobacterial protein. Mutations in residues 28 (D-->H) and 86 (V-->G) completely eliminated binding of p3-13 and significantly reduced binding of p56-65. A mutation in residue 26 (Y-->F) decreased binding of p3-13 but did not affect binding of p56-65. Substitutions of amino acid residues 28, 67, 71 and 86 in the DR17 beta 1 chain abrogated peptide-specific stimulation of both the p3-13- and the p56-65-specific T cell clones, while specific stimulation by only one peptide was eliminated by substitution at positions 26 and 74 (p3-13) and by substitution of residues 11 and 37 (p56-65). The observation that substitution of several other peptide-contacting DR17 beta 1 chain residues does not significantly affect peptide binding but does affect T cell stimulation, suggests that these substitutions alter the conformation of the bound peptide.

A uniquely high level of recombination at the HLA-B locus

Major histocompatibility complex (MHC) loci are some of the most polymorphic genes in the animal kingdom. Recently, it has been suggested that although most of the human MHC loci are relatively stable, the HLA-B locus can undergo rapid changes, especially in isolated populations. To investigate the mechanisms of HLA-B evolution we have compared the sequences of 19 HLA-B homologues from chimpanzees and bonobos to 65 HLA-B sequences. Analysis of the chimpanzee and bonobo HLA-B homologues revealed that despite obvious similarities between chimpanzee and human alleles in exon 2, there was little conservation of exon 3 between humans and the two chimpanzee species. This finding suggests that, unlike all other HLA loci, recombination has characterized the HLA-B locus and its homologues for over 5 million years.

Gel electrophoretic analysis of rhesus macaque major histocompatibility complex class II DR molecules

Rhesus macaque MHC class II DR molecules were isolated from radiolabeled B-cell line extracts by immunoprecipitation with the mAbs 7.3.19.1 and B8.11.2 and subsequently analyzed by 2D-gel electrophoresis. The B-cell lines used for this study were obtained from monkeys that are homozygous for the Mamu-DR region as defined by serologic techniques. Some of these animals have been selectively bred and originate from consanguineous matings. These analyses show that monkeys with the same allotyping may express different types of DR molecules. As in humans, the number of DR molecules expressed per haplotype is not constant and varies from 1 to 3, depending on the serologically defined Mamu-DR specificity, whereas it has been shown that the number of Mamu-DRB genes present per haplotype varies from 2 to 6. Therefore the present study also demonstrates that some of the rhesus macaque DR regions contain one or more pseudogenes.

The biologic importance of conserved major histocompatibility complex class II motifs in primates

Phylogenetic comparisons of polymorphic second-exon sequences of MHC class II DRB genes showed that equivalents of the HLA-DRB1*03 alleles are present in various nonhuman primate species such as chimpanzees, gorillas, and rhesus macaques. These alleles must root from ancestral structure(s) that were once present in a progenitor species that lived about 35 million years ago. Due to accumulation of genetic variation, however, sequences that cluster into a lineage are generally unique to a species. To investigate the biologic importance of such conservation and variation, the peptide-binding capacity of various Mhc-DRB1*03 lineage members was studied. Primate Mhc-DRB1*03 lineage members successfully binding the p3-13 peptide of the 65-kD heat-shock protein of Mycobacterium tuberculosis/leprae share a motif that maps to the floor of the peptide-binding site. Apart from that, some rhesus macaque MHC class-II-positive cells were able to present the p3-13 peptide to HLA-DR17-restricted T cells whereas cells obtained from great ape species failed to do so. Therefore, these studies open ways to understand which MHC polymorphisms have been maintained in evolution and which MHC residues are essential for peptide binding and T-cell recognition.