T cell receptor binding kinetics and special role of Valpha in T cell development and activation.

The kinetics of the interaction between T cell receptor (TCR) and major histocompatibility complex (MHC) has an important role in determining thymocyte-positive and -negative selection in the thymus, as well as in T cell activation. The alpha chain of the TCR is the major player in determining how the TCR fits onto the MHC ligand, and thus has a major role in determining whether a T cell develops as class I or class II restricted. In this article, we summarize recent data from our laboratory and others on the role of polymorphism in the Valpha combining site in determining MHC class restriction, and on kinetic parameters in thymocyte selection.

Reciprocal expression in CD4 or CD8 subsets of different members of the V alpha 11 gene family correlates with sequence polymorphism.

Previous staining studies with TCR V alpha 11-specific mAbs showed that V alpha 11.1/11.2 (AV11S1 and S2) expression was selectively favored in the CD4+ peripheral T cell population. As this phenomenon was essentially independent of the MHC haplotype, it was suggested that AV11S1 and S2 TCRs exert a preference for recognition of class II MHC molecules. The V alpha segment of the TCR alpha-chain is suggested to have a primary role in shaping the T cell repertoire due to selection for class I or II molecules acting through the complementarity determining regions (CDR) 1 alpha and CDR2 alpha residues. We have analyzed the repertoire of V alpha 11 family members expressed in C57BL/6 mice and have identified a new member of this family; AV11S8. We show that, whereas AV11S1 and S2 are more frequent in CD4+ cells, AV11S3 and S8 are more frequent in CD8+ cells. The sequences in the CDR1 alpha and CDR2 alpha correlate with differential expression in CD4+ or CD8+ cells, a phenomenon that is also observed in BALB/c mice. With no apparent restriction in TCR J alpha usage or CDR3 alpha length in C57BL/6, these findings support the idea of V alpha-dependent T cell repertoire selection through preferential recognition of MHC class I or class II molecules.

Thymic skewing of the CD4/CD8 ratio maps with the T-cell receptor alpha-chain locus.

The thymic preference for CD4+ T cells over CD8+ T cells is often attributed to a default pathway favouring CD4+ T cells or to homeostatic mechanisms. It is also clear, however, that T-cell receptor (TCR) preferences for major histocompatibility complex (MHC) class I versus class II binding will strongly influence an individual clone's skewing to the CD4 or CD8 subset. The variable region of each TCR alpha chain (V alpha) studied to date is found to be overrepresented in either CD4+ or CD8+ cells, suggesting that each V alpha element can interact more favourably with either MHC class I or class II molecules. Indeed, TCRs appear to have an intrinsic ability to interact with MHC molecules, and single amino acid residues present in germline-encoded complementarity determining region 1 (CDR1) and CDR2 of the V alpha element can be responsible for determining MHC specificity. Interestingly, the degree of CD4/CD8 skewing is variable among different mouse strains and in human populations. Here, we have shown that polymorphism in CD4/CD8 skewing between B6 and BALB/c mice is determined by the stem cell genotype and not by environmental effects, and that it maps in or near the TCR alpha-chain complex, Tcra. This was confirmed by comparing Tcra(b) with Tcra(a) or Tcra(c) haplotypes in congenic mice. We propose that the array of V alpha genes in various Tcra haplotypes exerts influence over the proportion of CD4 and CD8 subsets generated and may account in part for the observed thymic skewing. Thus, while it has been suggested that the TCR genes have been selected by evolution for MHC binding, our results further indicate selection for class II MHC preference.

Polymorphism within a TCRAV family influences the repertoire through class I/II restriction.

Antibody-staining experiments have shown that closely related members of the TCRAV3 family are reciprocally selected into the CD4 or CD8 peripheral T cell subsets. This has been attributed to the individual AV3 members interacting preferentially with either MHC class I or MHC class II molecules. Single amino acid residues present in the complementarity-determining regions (CDR) CDR1alpha and CDR2alpha are important in determining MHC class specificity. We have now extended these observations to survey the expressed repertoire of the AV3 family in C57BL/6 mice. Three of the four expressed AV3 members are preferentially selected into the CD4+ subset of T cells. These share the same amino acid residue in both CDR1alpha and CDR2alpha that differ from the only CD8-skewed member. Preferential expression of an individual AV3 is not caused by other endogenous alpha- or beta-chains, by any conserved CDR3 sequence, or by the usage of TCRAJ regions. This study shows that residues in the CDR1 and CDR2 regions are primary determinants for MHC class discrimination and suggests that polymorphism found within a TCRAV family has an important effect on the overall shaping of the T cell repertoire.

V alpha 3.2 selection in MHC class I mutant mice: evidence for an alternate orientation of TCR-MHC class I interaction.

TCR V alpha elements are expressed preferentially in CD4 or CD8 subsets in a manner that is largely independent of MHC haplotype. It is likely that the V alphas interact preferentially with conserved regions of class I or class II molecules. To investigate the topology of binding of TCR to MHC-peptide complexes, we screened a panel of H-2Kbm mutants for differential V alpha expression. One strain, bm23, showed a consistent alteration in V alpha expression, with increased V alpha 3.2 expression in CD8 peripheral T cells. This overselection is manifest in CD8 single-positive thymocytes and appears to be due to enhanced positive selection on Kbm23. There is no apparent effect of V beta elements. The Kbm23 molecule is unique compared with Kb and the other Kbm molecules at residue 75 on the helix of the alpha 1 domain, suggesting an interaction between V alpha 3.2 and the alpha 1 helix at this point. Such an interaction is inconsistent with the orientation of TCR and MHC defined in two crystal structures, but is consistent with an orientation where the TCR is rotated by 180 degrees relative to MHC.

Control of MHC restriction by TCR Valpha CDR1 and CDR2.

Individual T cell receptor (TCR) Valpha elements are expressed preferentially in CD4 or CD8 peripheral T cell subsets. The closely related Valpha3.1 and Valpha3.2 elements show reciprocal selection into CD4 and CD8 subsets, respectively. Transgenic mice expressing site-directed mutants of a Valpha3.1 gene were used to show that individual residues in either the complementarity-determining region 1 (CDR1) or CDR2 were sufficient to change selection from the CD4 subset to the CD8 subset. Thus, the germline-encoded Valpha elements are a major influence on major histocompatibility class complex (MHC) restriction, most likely by a preferential interaction with one or the other class of MHC molecule.

Selection of TCR V alpha by MHC class II predicts superantigen reactivity.

Recognition of superantigens by T cells predominantly involves the TCR V beta region. The contribution of reactivity from the non-V beta portion of the TCR remains less clear. We have investigated the V alpha repertoire of T cells bearing one V beta element responding to different superantigen or polyclonal stimuli. The data indicate that the V alpha chain is not directly involved in superantigen recognition, since unrelated superantigens do not stimulate cells bearing different V alpha elements. Instead, analysis of V alpha elements used preferentially in the CD4+ or CD8+ cells, shows that the V alpha-regions predominating in the superantigen response are the same V alphas that are positively selected into the CD4+ lineage. Thus the ability of a V alpha-region to either aid or hinder the interaction with class II predicts its frequency in a superantigen-responding population of T cells.

A HLA class I cis-regulatory element whose activity can be modulated by hormones.

To elucidate the basis of the down-regulation in major histocompatibility complex (MHC) class I gene expression and to identify possible DNA-binding regulatory elements that have the potential to interact with class I MHC genes, we have studied the transcriptional regulation of class I HLA genes in human breast carcinoma cells. A 9 base pair (bp) negative cis-regulatory element (NRE) has been identified using band-shift assays employing DNA sequences derived from the 5'-flanking region of HLA class I genes. This 9-bp element, GTCATGGCG, located within exon I of the HLA class I gene, can potently inhibit the expression of a heterologous thymidine kinase (TK) gene promoter and the HLA enhancer element. Furthermore, this regulatory element can exert its suppressive function in either the sense or anti-sense orientation. More interestingly, NRE can suppress dexamethasone-mediated gene activation in the context of the reported glucocorticoid-responsive element (GRE) in MCF-7 cells but has no influence on the estrogen-mediated transcriptional activation of MCF-7 cells in the context of the reported estrogen-responsive element (ERE). Furthermore, the presence of such a regulatory element within the HLA class I gene whose activity can be modulated by hormones correlates well with our observation that the level of HLA class I gene expression can be down-regulated by hormones in human breast carcinoma cells. Such interactions between negative regulatory elements and specific hormone trans-activators are novel and suggest a versatile form of transcriptional control.

Promotion of tumor growth by transfecting antisense DNA to suppress endogenous H-2Kk MHC gene expression in AKR mouse thymoma.

Many AKR spontaneous thymomas are reported to express different amounts of the major histocompatibility complex class I H-2Kk molecules. Moreover, H-2Kk-deficient AKR tumor cells are found to be more malignant when compared to tumor cells that express abundant levels of the H-2Kk molecules. To corroborate further the role of H-2Kk in tumorigenesis of AKR leukemia, we have, in this study, expressed antisense H-2Kk RNA in a high-H-2Kk-expressing and poorly tumorigenic AKR thymoma cell line 369. The down-regulation of H-2Kk molecules in the transfected 369 clones rendered them more tumorigenic in syngeneic AKR/J mice. The increase in oncogenicity correlates well with a concomitant reduction in their susceptibility to tumor-specific cytotoxic T lymphocytes in vitro. These results suggest the relevance of H-2Kk molecules in the immune surveillance of AKR tumors.

Inactivation of the H-2Klk gene could involve the substitutions of methylated CpGs.

By the isolation of overlapping cosmid clones and 'chromosome walking' studies from the H-2Kk gene, we have obtained cosmid clones encoding the H-2Klk gene from two separate cosmid libraries. The nucleotide sequence of one of the clones was determined. The cloned H-2Klk gene could be transcribed in vitro to give a normal H-2 class I mRNA of 1.7 kb. However, the deletion of four nucleotides in exon 3 of the H-2Klk gene results in a translation termination codon at the beginning of exon 4. In agreement with this, when expressed in human cells, the H-2Klk gene gave a truncated, cytoplasmic polypeptide of Mr 36,000. Therefore, although the H-2Klk gene is homologous to other class I MHC genes in its molecular organization and nucleotide sequence, it is a pseudogene. When compared to the nucleotide sequence of the H-2Kk gene, the H-2Klk gene has undergone many substitutions of methylated CpG residues (meCpG). This represents further evidence to suggest that this gene is inactive.

Cloning and characterization of a polymorphic class I MHC gene in the AKR lymphoma K36.16.

Cancers are the result of somatic heritable changes in certain genes. The AKR leukaemia K36.16 has been extensively studied in our laboratory. When compared to normal AKR thymocytes, the K36.16 tumour cells do not express the H-2Kk antigens and have an unexpected antigenic determinant that could be detected by anti-H-2Dd monoclonal antibodies. To understand the molecular mechanisms that could be responsible for these changes, we have compared the genomic composition of the class I MHC genes in the K36.16 tumour cells to that of normal AKR lymphocytes. A unique polymorphic 2.6-kb Hind III fragment was detected in DNA obtained from the K36.16 tumour cells after hybridization with a 3'-gene-coding H-2 probe. This fragment is not present in DNA of normal AKR lymphocytes. In an effort to further understand the mechanism underlying the nature of this MHC gene polymorphism, we have cloned and sequenced this Hind III fragment. When compared with the reported sequences of a number of mouse class I MHC genes, the nucleotide sequence of this polymorphic Hind III fragment is similar to that of a reported Tla gene.