Interindividual conservation of T-cell receptor beta chain variable regions by minor histocompatibility antigen-specific HLA-A*0201- restricted cytotoxic T-cell clones

Minor histocompatibility antigens (mHags) are involved in the induction of graft-versus-host disease (GVHD) after HLA-identical bone marrow transplantation. Previously, we isolated a series of HLA-A*0201- restricted cytotoxic T-cell (CTL) clones specific for the same mHag HA- 1 from peripheral blood of three unrelated patients who were suffering from GVHD. We have now analyzed the composition of the T-cell receptor (TCR) V regions of 12 of these mHag HA-1-specific HLA-A*0201-restricted CTL clones by DNA sequencing of the alpha and beta chains. Of these 12 clones, derived from three unrelated individuals, five independent TCR alpha V- and beta V-region sequences were established. The TCR alpha chains were composed of varying TCR alpha V and TCR alpha J genes with no obvious similarities in structure in the N regions. However, the TCR beta chains all used the TCR beta V6S9 gene segment, and showed remarkable similarities within the N-D-N regions; ie, three independent beta-chain sequences (originating from donors Ha and Gy) shared a leucine/valine amino acid pair, whereas the other two (originating from donors Ha and Wi) shared a serine/threonine pair, all at positions 99 and 100 of the TCR beta V region. In conclusion, the TCR analysis of HA- 1 mHag-specific CTL clones has shown that the HA-1 mHag/HLA-A*0201 complex selects for highly similar TCR beta V regions.     

Two distinct alpha beta T-cell lineages can be distinguished by the differential usage of T-cell receptor V beta gene segments.

Avian T cells can be divided into three subpopulations based on their expression of distinct T-cell receptors (TCR1, TCR2, and TCR3), ontogeny, and tissue distribution. The TCR1 cells appear to be the equivalent of mammalian gamma delta cells, but the derivation of cells expressing TCR2 and TCR3 has been unclear. Here we report that chickens contain two families of TCR beta variable (V) gene segments, V beta 1 and V beta 2. Furthermore, TCR2 and TCR3 represent subsets of alpha beta cells that are defined by mutually exclusive usage of these two families of V beta gene segments. Sequence comparisons of V beta 1 and V beta 2 with mammalian TCR beta V segments reveal that V beta 1 gene segments encode the conserved amino acids used to define the mammalian V beta consensus subgroup I, while V beta 2 encodes the amino acids used to define the mammalian V beta subgroup II. Although the beta chains of TCR2 and TCR3 cells are encoded by the same diversity (D), joining (J), and constant (C) region segments, V beta 1 gene segments undergo rearrangement before V beta 2 gene segments during T-cell development. This may result from the fact that TCR2 cells undergo V-DJ joining by deletional rearrangement, whereas TCR3 cells undergo V-DJ joining by inversional rearrangement. These data suggest that the TCR alpha beta cells can be divided into two distinct and evolutionarily conserved lineages based on V beta gene segment usage. The clear-cut separation of these lineages in the chicken may help to define their immunologic role.     

T-cell receptor variable region gene usage by CD4+ and CD8+ T cells in bronchoalveolar lavage fluid and peripheral blood of sarcoidosis patients.

Sarcoidosis is a chronic noncaseating granulomatous disease of unknown etiology. An accumulation of CD4+ T cells in the alveolar space of the lungs is a characteristic feature of the disease. We have in this study analyzed T-cell receptor (TCR) variable region (V) gene usage by CD4+ and CD8+ lung and peripheral blood T cells of 29 sarcoidosis patients and 15 control subjects. In the patient group, we found a 100% positive correlation between TCR V alpha 2.3+ CD4+ lung T-cell expansions and the expression of the HLA-DR3(17),DQ2 haplotype. The remaining TCR V alpha/V beta gene products analyzed in this study--V alpha 12, V beta 2, V beta 3, V beta 5.1, V beta 5.2/5.3, V beta 5.3, V beta 6.7, V beta 8.1, and V beta 12--were in general normally expressed by CD4+ T cells, although some of them were used to a significantly higher or lower degree by lung T cells compared to peripheral blood T cells. We also performed repeated TCR V gene analyses on some HLA-DR3+ patients and found an association between the ratio bronchoalveolar lavage fluid/peripheral blood V alpha 2.3+ CD4+ T cells and clinical signs of disease activity. Finally, when analyzing TCR V gene usage by CD8+ bronchoalveolar lavage fluid and peripheral blood T cells, a normal V alpha 2.3 usage was found in all cases, but lung-restricted T-cell expansions using other TCR V gene segment products were identified.     

CD3+ leukemic large granular lymphocytes utilize diverse T-cell receptor V beta genes

CD3+ large granular lymphocyte (LGL) leukemia is a disease of unknown etiology characterized by clonal proliferation of T cells that usually express T-cell receptor (TCR) alpha beta heterodimers. The purpose of this study was to identify the variable (V), joining (J), and diversity (D) region TCR beta-chain genes expressed by CD3+ LGL leukemic cells in an attempt to gain insights into the etiology of this disorder. Twelve patients with LGL leukemia were studied, including seven with both LGL leukemia and rheumatoid arthritis (RA). RA is also a disease of unknown etiology that occurs frequently in patients with LGL leukemia. Clonally expanded T cells that express specific TCR V beta genes have been identified in fluid and tissue specimens from the joints of patients with RA. In this study, V beta expression was determined by PCR using a panel of 22 unique V beta primers to amplify cDNA prepared from peripheral blood mononuclear cells (PBMC). A dominant V beta gene product was readily apparent in all patients. To confirm that the dominant V beta gene originated from a clonal expansion, DNA fragments corresponding to the dominant V beta genes were subcloned into plasmids and independently isolated recombinants were sequenced. V-D-J region sequences that occurred repeatedly indicated clonality. The V beta and J beta genes expressed by the leukemic cells showed a pattern of distribution that followed the frequency with which these genes are represented in the peripheral blood. The residues corresponding to the third complementarity-determining region of the TCR beta chain were different in all cases. A specific pattern of VDJ usage was not identified for those patients with both LGL leukemia and RA; however, utilization of V beta-6 by LGL clones (N = 3) was observed only in the setting of RA. These data suggest that leukemic CD3+ LGL cells have been clonally transformed in a random fashion with respect to the TCR beta chain.     

T cells expressing the V beta 1 T-cell receptor are required for IgA production in the chicken.

While alpha beta T cells in mammals may express one of many variable (V) gene families in the beta locus, chickens have only two V beta gene families. The avian V beta 2+ T cells are recognized by the T-cell receptor 3 (TCR3) monoclonal antibody and V beta 1+ T cells are recognized by the TCR2 antibody, which we used to selectively suppress development of V beta 1+ T cells in order to examine their functional role. Suppression was accomplished by multiple injections of anti-TCR2 antibodies beginning in embryonic life and perpetuated by thymectomy 8 days after hatching. Young birds thus depleted of V beta 1+ T cells had greater than normal numbers of V beta 2+ T cells and appeared as healthy as thymectomized and untreated controls. While production of IgM and IgG antibodies was unimpaired, IgA antibody production was severely compromised in the V beta 1-depleted birds. The levels of secretory IgA in bile and lung lavage fluid were reduced 1000- to 10,000-fold and secretory IgA antibodies were not produced in response to mucosal immunization. B-cell production of IgA antibodies thus appears to require T cells expressing the V beta 1 genes, whereas T cells that express the V beta 2 genes lack this capacity.     

Evolutionary conservation of antigen recognition: the chicken T-cell receptor beta chain

T cells play important regulatory roles in the immune responses of vertebrates. Antigen-specific T-cell activation involves T-cell receptor (TCR) recognition of a peptide antigen presented by a major histocompatibility complex molecule, and much has been learned about this antigen-recognition process through structural and genetic studies of mammalian TCRs. Although previous studies have demonstrated that avian T cells express cell-surface molecules analogous to the mammalian TCR heterodimers, TCR genes have not been identified in nonmammalian species. We now report the cloning of a cDNA that encodes the beta chain of the chicken TCR. Southern blot analysis using this TCR beta cDNA probe demonstrated that the chicken TCR beta locus was clonally rear-ranged in chicken T-cell lines. TCR beta mRNA was expressed in cells isolated from the thymus but not in cells from the bursa of Fabricius where B cells are generated. Sequence analysis of six additional TCR beta cDNAs suggested the existence of at least two variable (V) region families, three joining (J) elements, and single diversity (D) and constant (C) elements. As in mammals, considerable nucleotide diversity was observed at the junctions of the variable, diversity, and joining elements in chicken TCR beta cDNAs. Genomic V beta and J beta elements were also cloned and sequenced. Both elements are flanked by classical heptamer/nonamer recombination signal sequences. Although the chicken and mammalian TCR beta chains displayed only 31% overall amino acid sequence identity, a number of conserved structural features were observed. These data indicate that (i) the chicken TCR beta repertoire is generated by combinatorial and junctional diversity and (ii) despite divergent evolution at the level of nucleotide sequence, important structural features of the TCR beta polypeptide are conserved between avian and mammalian species.     

Junctional region sequences of T-cell receptor beta-chain genes expressed by pathogenic anti-DNA autoantibody-inducing helper T cells from lupus mice: possible selection by cationic autoantigens.

We rescued from the spleens of 10 (SWR x NZB)F1 (SNF1) mice with lupus nephritis the T cells that were activated in vivo and cloned 268 T-cell lines and hybridomas. Only 12% of these T-cell clones had the functional ability to preferentially augment the production of pathogenic anti-DNA autoantibodies. Among these, 16 helper T-cell (Th-cell) clones that were mostly CD4+ and had the strongest autoantibody-inducing ability were analyzed for T-cell receptor (TCR) beta-chain gene usage. Seven of the 16 Th-cell clones expressed beta-chain variable region (V beta) V beta 8 (8.2 or 8.3) genes and three expressed V beta 4, whereas two clones each used a V beta 1 or V beta 2 or V beta 14 gene, suggesting some restriction in TCR gene usage. Although heterogeneous, the V-D-J junctional region sequences of TCR beta-chain genes used by these Th-cell clones invariably encoded one or more negatively charged residues (aspartic or glutamic acid) that had been generated in most cases by unspecified nucleotide (N) additions. Representative pathogenic autoantibody-inducing Th-cell clones could rapidly induce the development of lupus nephritis when injected into young prenephritic SNF1 mice. The pathogenic autoantibody-inducing Th cells expressing the anionic residues in their TCR beta-chain junctions (complementarity-determining region CDR3) were probably selected by some cationic autoantigenic peptide presented by the anti-DNA B cells they preferentially helped. These results offer a clue regarding the nature of the primary autoantigen that may drive the pathogenic autoimmune response in lupus.     

Analysis of T-cell receptor variability in transplanted patients with acute graft-versus-host disease.

T lymphocytes play a pivotal role in graft-versus-host disease (GVHD) and largely contribute to the graft-versus-leukemia (GVL) effect. Most mature T lymphocytes specifically recognize antigens through the alpha/beta T-cell receptor (TCR). Each alpha/beta TCR chain includes a constant region and a variable region, the latter being encoded by V-J alpha or V-D-J beta rearranged gene segments. To better characterize T cells involved in GVHD, V alpha and V beta gene segment usage was analyzed, after cDNA amplification, in peripheral blood mononuclear cells (PBMC) and skin samples from three patients with grade II cutaneous GVHD. At time of GVHD diagnosis (days 11, 22, and 25), when first signs of engraftment were detectable, virtually all V alpha and V beta subfamilies were represented in PBMC RNAs of the three recipients. These results suggest that diversified TCR gene segment expression is observed early after allogenic bone marrow transplantation (alloBMT). Lymphocytes infiltrating GVHD skin also expressed a large series of V alpha and V beta subfamily specificities. However, analysis of the V alpha and V beta amplified products showed substantial differences between PBMC and the skin lymphocyte RNAs. These observations indicate that a large variety of T lymphocytes are present at the disease site, while some of them may be specifically amplified or decreased in response to minor histocompatibility antigens (miHA). Further characterization of the latter T-cell subpopulations should lead to a better understanding of human in vivo responses directed at miHA.     

T-cell receptor alpha-chain variable-region haplotypes of normal and autoimmune laboratory mouse strains.

We used Southern blotting and mRNA analysis to characterize allelic polymorphisms among genes of the T-cell antigen receptor (TCR) alpha-chain variable-region (V alpha) locus in a large panel of normal and autoimmune-susceptible or autoimmune-contributing strains of laboratory mice. Four major V alpha haplotypes were defined on the basis of multiple restriction fragment length polymorphisms for each of nine V alpha subfamily probes used. Southern blotting also revealed haplotype-specific loss of bands within some V alpha subfamilies, consistent with the deletion of particular V alpha genes or sets of genes from haplotype to haplotype. In contrast to the situation in the V beta locus, however, deletion of entire V alpha subfamilies was not observed. The nature of V alpha allelic variability was further explored by using an RNase protection assay to analyze expressed V alpha mRNA sequences in thymocyte RNA. Such analysis revealed both shared and unique patterns of V alpha mRNA expression among the different haplotypes and supported the conclusion that haplotype differences sometimes involve V alpha gene deletions. Interestingly, a disproportionate number of, but not all, autoimmune-susceptible strains, including NZB, SJL, SWR, PL/J, and NOD, share a common V alpha haplotype. The identification of murine TCR V alpha haplotypes should provide a basis for understanding the role of TCR diversity in normal immunoregulatory and immune-response phenomena, as well as autoimmune-disease predisposition.     

Positive selection of invariant V alpha 14+ T cells by non-major histocompatibility complex-encoded class I-like molecules expressed on bone marrow-derived cells.

V alpha 14+ T cells are a unique subset expressing an invariant T-cell antigen receptor alpha chain encoded by V alpha 14 and J alpha 281 gene fragments with a 1-nt N region. Most invariant V alpha 14+ T cells develop in extrathymic organs, independent of thymus, and expand at a high frequency in various mouse strains regardless of major histocompatibility complex (MHC) haplotype. In this paper, we show that the positive selection of invariant V alpha 14+ T cells requires a beta 2-microglobulin-associated MHC class I-like molecule not linked to the MHC on chromosome 17. This was determined by linkage analysis on DNA from recombinant mice generated by crossing a C57BL/6 mouse with a wild mouse, Mus musculus molossinus, that is negative for invariant V alpha 14 TCR expression. However, the peptide transporter TAP1 is not necessary for positive selection of invariant V alpha 14+ T cells, indicating the direct recognition of the MHC class I-like molecule without peptide by the invariant V alpha 14 TCR. Further, experiments with bone marrow-chimeric mice show that invariant V alpha 14+ T cells in the periphery are selected by bone marrow cells, suggesting a unique lineage of V alpha 14+ T cells differentiated through a selection process distinct from that of conventional alpha beta TCR+ T cells.     

In vivo clonal dominance and limited T-cell receptor usage in human CD4+ T-cell recognition of house dust mite allergens.

Sensitivity to house dust mite antigens in atopic individuals is a major cause of allergic diseases, ranging from asthma to rhinitis and dermatitis. We have studied the T-cell receptor (TCR) usage of house-dust-mite-specific CD4+ T-cell clones isolated from an atopic individual, by using the anchored polymerase chain reaction, and have analyzed the peripheral TCR repertoire of the same individual. Several T-cell clones had identified TCRs at the sequence level, despite the fact that they had been independently isolated, in some cases, in different years. These data suggest the presence in vivo of long-lived T-cell clones. We have also shown that junctional sequences identical to these clones are present in peripheral blood T cells taken 6 years after the isolation of the T-cell clones. The analysis of TCR genes used by the panel of clones reveals oligoclonality, with the variable (V) region gene segments V alpha 8 and V beta 3 being dominant, although there is minimal conservation of junctional sequences. The results have implications for understanding the TCR recognition of an environmental aeroallergen and the life span of T-cell clones in vivo during a chronic immune response.     

Stages of T-cell receptor protein expression in T-cell acute lymphoblastic leukemia.

In this study five monoclonal antibodies (MoAbs) to T-cell receptor (TCR) proteins (WT31, alpha F1, beta F1, TCR delta-1 and delta TCS-1) were used to identify discrete maturative stages in 40 cases of T-cell acute lymphoblastic leukemia (T-ALL). These MoAbs reacted exclusively with CD3+ T cells and did not label B-lineage and myeloid cells. In 17 of the 40 T-ALL cases studied the leukemic blasts lacked membrane and cytoplasmic TCR chains (group I). In 12 cases cells did not have membrane CD3/TCR but expressed cytoplasmic TCR proteins heterogenously: nine cases had cytoplasmic TCR beta chains (beta F1+, alpha F1-; group II), one case had cytoplasmic TCR alpha chains (alpha F1+, beta F1-; group III), and two cases were labeled by both alpha F1 and beta F1 MoAbs (group IV). The remaining 11 cases were mCD3+: nine were TCR alpha beta+ (group Va) and two exhibited TCR gamma delta (TCR delta-1+, delta TCS-1+; group Vb). The analysis of the TCR beta, -gamma, and -delta gene configurations in 23 of the 40 T-ALLs showed that: (1) the lack of TCR protein expression was due to the lack of TCR gene rearrangements only in one of nine cases; (2) five of five TCR beta+, TCR alpha- cases studied had germline TCR alpha genes (ie, no detectable TCR delta gene deletions); (3) seven of eight cases with TCR delta gene deletions expressed TCR alpha proteins, whereas in 12 of 20 of the T-ALLs with TCR beta gene rearrangements the synthesis of the corresponding protein occurred; only 2 of 16 cases with rearranged TCR delta genes expressed TCR delta chains. The T-ALL categories identified with anti-TCR MoAbs did not have additional characteristic phenotypic patterns and may correspond to the normal stages of T-cell development more precisely than those defined by other differentiation antigens.     

Chicken T-cell receptor beta-chain diversity: an evolutionarily conserved D beta-encoded glycine turn within the hypervariable CDR3 domain.

Unlike mammals, chickens generate an immunoglobulin (Ig) repertoire by a developmentally regulated process of intrachromosomal gene conversion, which results in nucleotide substitutions throughout the variable regions of the Ig heavy- and light-chain genes. In contrast to chicken Ig genes, we show in this report that diversity of the rearranged chicken T-cell receptor (TCR) beta-chain gene is generated by junctional heterogeneity, as observed in rearranged mammalian TCR genes. This junctional diversity increases during chicken development as a result of an increasing base-pair addition at the V beta-D beta and D beta-J beta joints (where V, D, and J are the variable, diversity, and joining gene segments). Despite the junctional hypervariability, however, almost all functional V beta-D beta-J beta junctions appear to encode a glycine-containing beta-turn. Such a turn may serve to position the amino acid side chains of a hypervariable TCR beta-chain loop with respect to the antigen-binding groove of the major histocompatibility complex molecule. Consistent with this hypothesis, the germ-line D beta nucleotide sequences of chickens, mice, rabbits, and humans have been highly conserved and encode a glycine in all three reading frames.     

Conserved structure of amphibian T-cell antigen receptor beta chain.

All jawed vertebrates possess well-differentiated thymuses and elicit T-cell-like cell-mediated responses; however, no surface T-cell receptor (TCR) molecules or TCR genes have been identified in ectothermic vertebrate species. Here we describe cDNA clones from an amphibian species, Ambystoma mexicanum (the Mexican axolotl), that have sequences highly homologous to the avian and mammalian TCR beta chains. The cloned amphibian beta chain variable region (V beta) shares most of the structural characteristics with the more evolved vertebrate V beta and presents approximately 56% amino acid identities with the murine V beta 14 and human V beta 18 families. The two different cloned axolotl beta chain joining regions (J beta) were found to have conserved all the invariant mammalian J beta residues, and in addition, the presence of a conserved glycine at the V beta-J beta junction suggests the existence of diversity elements. The extracellular domains of the two axolotl beta chain constant region isotypes C beta 1 and C beta 2 show an impressively high degree of identity, thus suggesting that a very efficient mechanism of gene correction has been in operation to preserve this structure at least from the early tetrapod evolution. The transmembrane axolotl C beta domains have been less well conserved when compared to the mammalian C beta but they do maintain the lysine residue that is thought to be involved in the charged interaction between the TCR alpha beta heterodimer and the CD3 complex.     

Extensive conservation of alpha and beta chains of the human T-cell antigen receptor recognizing HLA-A2 and influenza A matrix peptide.

The major histocompatibility complex class I molecule HLA-A2.1 presents the influenza A virus matrix peptide 57-68 to cytotoxic T lymphocytes in all individuals with this common HLA type and is among the most thoroughly studied immune responses in humans. We have studied the T-cell receptor (TCR) heterogeneity of T cells specific for HLA-A2 and influenza A matrix peptide using the polymerase chain reaction. The usage of V alpha and V beta sequences seen on these T cells is remarkably conserved as are certain junctional sequences associated with alpha and beta chains. Furthermore, two unrelated HLA-A2 individuals have a similar pattern of TCR usage, implying that this is a predominant response in HLA-A2 populations. Analysis in one individual showed that the conserved TCR V alpha and V beta genes are minor members of the peripheral blood TCR repertoire. The sequences provide important information on the TCR necessary for the final structural analysis of this ternary complex.     

Specific inhibition of cell-surface T-cell receptor expression by antisense oligodeoxynucleotides and its effect on the production of an antigen-specific regulatory T-cell factor.

We have used antisense oligodeoxynucleotides corresponding to genes encoding the variable (V) region of the T-cell receptor (TCR) alpha and beta chains (V alpha and V beta) to control TCR expression in T-cell hybridomas. Two hybridomas, A1.1 and B1.1, recognize a synthetic polypeptide antigen designated poly 18 (poly[Glu-Tyr-Lys-(Glu-Tyr-Ala)5]) together with I-Ad. We have found that TCR function (production of lymphokines in response to antigen) and T3 expression were removed after protease treatment of the cells and were fully recovered 48 hr later. However, when antisense oligodeoxynucleotides corresponding to the appropriate TCR V genes were present after protease treatment, little or no recovery of TCR function or T3 expression was observed. This effect was specific for the TCR V genes utilized by the T cell: antisense oligodeoxynucleotides corresponding to the TCR V regions of A1.1 had no effect on TCR expression in B1.1 and vice versa. Thus, antisense oligodeoxynucleotides can be used to temporarily block expression of a TCR gene in a T-cell hybridoma. This technique was then applied to a paradoxical phenomenon in A1.1 cells. We had observed previously that A1.1 releases an antigen-specific immunoregulatory activity that shows the same antigenic fine specificity as is displayed by the TCR of A1.1. We now report that antisense oligodeoxynucleotides corresponding to the A1.1 V alpha gene blocked the production of this soluble antigen-specific activity by the cell. Antisense oligodeoxynucleotides corresponding to A1.1 V beta, on the other hand, had no effect on the production of this antigen-specific activity. We discuss these observations in the context of recent findings on the nature of T cell-derived antigen-specific regulatory factors.     

Cell-type-specific cDNA probes and the murine I region: the localization and orientation of Ad alpha.

Labeled cDNA probes highly enriched for B-cell-specific (B-T) and T-suppressor-cell-specific (Ts-B) sequences were used to screen a set of genomic cosmid clones spanning 230 kilobases of the murine immune response (I) region. With the B-cell derived probe, four I-region genes were detectable (A beta, E beta, E beta 2, and E alpha), as well as an additional (fifth) region of hybridization. The T-cell probe, prepared from a putative I-J positive suppressor cell hybridoma, was negative in a parallel experiment. A genomic fragment corresponding to the new region of hybridization seen with the B-cell cDNA probe identified a discrete mRNA species in RNA blotting analysis that had a pattern of expression strikingly similar to E alpha mRNA in a variety of lymphoid tumor lines. This fragment was also used to isolate cDNA clones from a library highly enriched (20-40 times) for B-cell-specific sequences (selected cDNA cloning). DNA sequence analysis of one of these cDNA clones indicates that this gene encodes the A alpha molecule of the d haplotype (Ad alpha). These findings establish that the order of the class II genes in the Ia complex is as follows: centromere-A beta-A alpha-E beta-E beta 2-E alpha.     

Rearrangement of the T-cell receptor delta genes in human T-cell leukemias

Two distinct types of T-cell receptors (TCR), designated alpha beta and gamma delta, have been identified on the surface of T cells. In the adult, T cells bearing the gamma delta TCR are a minority and they have the phenotype CD3+, CD4-, CD8-/+. By using appropriate probes, rearrangements of the TCR alpha, beta, and gamma genes have been extensively investigated in a variety of lymphoproliferative disorders. Because the TCR delta gene has been cloned only recently, no comparable information exists with respect to this in human leukemias. We report the analysis of the TCR delta gene configuration in 21 T-cell acute and chronic leukemias, 40 B-cell leukemias, 4 acute myeloid leukemias of difficult classification, and 12 normal controls. The TCR delta genes were structurally modified in all T-cell disorders and in germ-line configuration in all controls and all but one case of non-T-cell leukemias tested. In one case of T-chronic lymphocytic leukemia (CD3+, CD4-, CD8+) we found rearrangement and expression of TCR gamma and delta (but not alpha and beta), suggesting that leukemic transformation took place in a cell bearing a TCR gamma delta rather than a TCR alpha beta. In two cases of pre-T-acute lymphoblastic leukemia, only delta was rearranged out of the three TCR genes tested. This finding is in keeping with the suggestion that the TCR delta gene might be the first to rearrange in T cell ontogeny, and that its mode of rearrangement may play a role in the subsequent choice of the cell between production of a TCR alpha beta or gamma delta. Thus, TCR delta chain gene analysis can provide novel information of the clonal nature of T-cell disorders, particularly if the analysis of the beta and gamma genes has not been helpful.