Multiplex PCR for TCR delta rearrangements: a rapid and specific approach for the detection and identification of immature and mature rearrangements in ALL

The preferential occurrence of immature T-cell receptor (TCR) δ rearrangements (i.e. incomplete Dδ2-Dδ3 and Vδ2-Dδ3) in B-cell precursor acute lymphoblastic leukaemia (BCP ALL) and of predominantly mature rearrangements (incomplete Dδ2-Jδ1, complete Vδ1, Vδ2, Vδ3 to Jδ1) in T-lineage ALL prompted us to establish two separate multiplex PCR systems for the identification of clonal TCRδ rearrangements. PCR products of the expected size for the specific rearrangements were detectable from a dilution of 100–1000 clonal cells in 150 000 polyclonal cells. Both multiplex PCR systems were used to analyse samples from 86 childhood BCP ALLs and 30 T-lineage ALLs. The results of the multiplex PCRs were controlled by standard PCR analyses for the individual rearrangements and Southern blots, which were identical. Only immature TCRδ rearrangements were detected in BCP ALL (59%), whereas no rearrangement was found in the remaining BCP leukaemias, thus confirming the exclusive presence of immature TCRδ rearrangements in B-lineage cells. 50% of the T-lineage ALLs contained mature rearrangements, but no immature rearrangements were found. These two multiplex PCR techniques appear to be reliable and fast aids in the analysis of clonal TCRδ rearrangements in ALL.     

Direct sequence analysis of TCR Vδ2–Dδ3 rearrangements in common acute lymphoblastic leukaemia and application to detection of minimal residual disease

Summary. T cell receptor δ chain (TCRδ) gene rearrangements were studied by Southern blot analysis in 36 patients with common acute lymphoblastic leukaemia, including 14 adults and 22 children. The majority of patients (68%) had either a rearrangement or deletion of one or more TCRδ genes. The most frequent rearrangement involved a partial recombination of Vδ2 to Dδ3 (55%). Dδ2–Dδ3 rearrangements were present in five patients (14%). To investigate the TCRδ rearrangement as a tumour marker in minimal residual disease studies, presentation samples from 18 patients were amplified by PCR and directly sequenced. Although the size of the Vδ2–Dδ3 junction varied by only 40 bp, sequence analysis showed extensive diversity. This was derived from four factors: deletion of the 5'end of Dδ3 gene (15/18) and 3'end of Vδ2 gene (16/18); the presence of Dδ2 sequences (6/18); insertion of N nucleotides (15/18); association of P nucleotides with intact Vδ2 and Dδ3 genes (5/18). N nucleotides were the major feature, contributing to 75% of the junction. Dδ1 sequences were not involved. Twenty base oligonucleotide probes, constructed from the junctional sequences, were capable of detecting residual tumour cells at the 10⁻⁴ sensitivity level. Cross hybridization studies confirmed the probes to be clone specific. Longitudinal studies on patients undergoing treatment were capable of detecting tumour in remission samples.     

Multigenetic lesions in infant acute leukaemias: correlations with ALL-1 gene status

The large majority of childhood B-precursor cell acute lymphoblastic leukaemia cases present IgH and TCRδ gene rearrangements. These rearrangements have been widely used as specific markers for monitoring minimal residual disease. However, their prognostic value still remains unclear. In order to determine whether IgH and TCRδ gene rearrangements have any influence on relapse and event-free survival (EFS), we analysed the clinical impact of these genetic characteristics in 51 B-precursor acute lymphoblastic leukaemia patients. 46/51 patients (90.2%) showed IgH gene rearrangements by Southern blot and/or polymerase chain reaction (PCR) analysis. No statistically significant associations were found between IgH gene rearrangement pattern and age, sex, WBC count, immunophenotype, risk factor, relapse or EFS. 27/41 patients (66%) showed Vδ₂Dδ₃ recombination by Southern blot and/or PCR analysis. At a median follow-up of 53 months the estimated 5-year EFS probability was 78 ± 3% for the whole group. The EFS probability among patients with a Vδ₂Dδ₃ recombination pattern in the TCRδ locus was 90 ± 3%, whereas for patients without Vδ₂Dδ₃ recombination was 39 ± 13% ( P  < 0.005).
IgH rearrangement patterns do not appear to influence relapse or EFS probability. However, TCRδ gene rearrangement patterns have a relevant impact on the relapse rate and the EFS probability. Patients with Vδ₂Dδ₃ recombination have better clinical outcome than patients without this recombination, independent of any other prognostic factors.     

Crosslineage TCR delta rearrangements occur shortly after the DJ joinings of the IgH genes in childhood precursor B ALL and display age-specific characteristics

In order to test the hypothesis that the most immature T-cell receptor (TCR) rearrangements occur after the DJ joining of the immunoglobulin heavy chain genes (IgH), we analysed the TCR Vδ2–Dδ3 rearrangements in precursor B-cell leukaemias (PBC ALL) from 25 children younger than 3 years at disease onset and found that most of the junctional regions had N nucleotides inserted. We then selected 14 of these PCB ALLs for DJH (DJ joining of the IgH) characterization. These joining regions showed homology-directed recombination and lack of N regions, indicating absence of terminal deoxynucleotidyl transferase (TdT) activity during their rearrangement. Most leukaemias with a DJH rearrangement without N region have no, or only one, nucleotide in the joining regions of their Vδ2–Dδ3 rearrangements. The N regions of the TCR delta rearrangements displayed 'age-specific' differences: in children younger than 3 years of age the N regions were shorter than in those older than 3 years, and the rearrangements frequently contained complete segments. We conclude that the Vδ2–Dδ3 rearrangement in childhood PCB ALLs is an early event following DJH rearrangement and that it occurs shortly before or after the first hit, leading to malignant transformation.     

NPM/ALK gene fusion transcripts identify a distinct subgroup of null type Ki-1 positive anaplastic large cell lymphomas

The chromosomal aberration t(2;5) resulting in the juxtaposition of NPM and ALK genes is a well-known feature of several Ki-1⁺ anaplastic large cell lymphomas (ALCL) of the T-cell type. However, conflicting results have been reported concerning the presence of this gene rearrangement in other ALCL and Hodgkin's disease (HD), respectively. We performed NPM/ALK RT-PCR on 14 cases of ALCL expressing distinct myelomonocytic markers, e.g. CD11c, CD13, CD14 or CD68, but neither T-cell nor B-cell associated antigens (null cell phenotype). The specific translocation was found exclusively in six childhood tumours previously diagnosed as malignant histiocytosis (MH), whereas all adult lymphomas (three ALCL without characteristics of MH, three secondary ALCL following HD) and two paediatric cases of secondary ALCL following HD did not show NPM/ALK gene fusion products. By Southern blotting, the status of T-cell receptor (TCR) and immunoglobulin heavy chain genes (IgH) were investigated; two patients with initially diagnosed MH had the TCRδ-chain gene rearranged (Dδ2–Dδ3 and Vδ1–Jδ1, respectively). IgH rearrangements were detected in only one patient with secondary ALCL.   Our data indicate a high association of previously diagnosed MH and NPM/ALK gene rearrangements. In one case, this specific translocation was demonstrated at an early stage of development; in another, a mature TCRδ-chain gene rearrangement was detected. These data support the hypothesis of a lymphoid origin of this subgroup of Ki-1 positive ALCL previously diagnosed as MH.     

Sensitive non-radioactive PCR detection of clonal T-cell receptor-gamma gene rearrangements in childhood acute lymphoblastic leukaemia

Rearrangement of the T-cell receptor gamma (TCRγ) gene potentially provides a valuable target for monitoring minimal residual leukaemia by the Polymerase Chain Reaction (PCR). However, existing strategies directed at this locus frequently lack sensitivity or specificity. We describe a novel PCR strategy for improved detection of clonal TCRγ gene rearrangements based on the design of two overlapping clone-specific reverse PCR primers spanning the TCRγ junctional region, which are used sequentially in conjunction with forward V-region specific primers. Unlike other strategies, specificity is generated from the initial and subsequent round of PCR amplification. This non-radioactive technique is highly specific and sensitive, and should prove effective for monitoring minimal residual leukaemia.     

Multiplex PCR for rapid detection of T-cell receptor-gamma chain gene rearrangements in patients with lymphoproliferative diseases

We describe a multiplex polymerase chain reaction (PCR) method suitable for the detection of all T-cell receptor (TCR) γ-chain gene rearrangements in patients with lymphoproliferative diseases. 40 patients with various lymphoproliferative disorders and 40 healthy individuals were tested. Clonal TCRγ rearrangements were identified in all patients with malignant disorders, and in one of 10 cases with established reactive lymphocytosis but not in normal controls. In all individuals testing positive, the patient's specific V and J segment involved in the rearrangement could be determined by simply splitting the multiplex primer mix. Our data show that the multiplex PCR technique enables rapid, simple and sensitive screening for clonal TCRγ chain gene rearrangements.     

Acquired skewing of X-chromosome inactivation patterns in myeloid cells of the elderly suggests stochastic clonal loss with age

More frequent skewing of X-chromosome inactivation patterns (XCIPs) occurs in the white blood cells of elderly females; this study was performed to determine whether this occurs in myeloid or lymphoid lineages. XCIPs were analysed in purified neutrophils and T cells from 80 females  > 75 years and the results were compared with 23 cord blood and 94 younger adult blood samples. The degree of XCIP skewing in cord blood and younger adult blood cells was similar, with 3–4% having  > 90% expression of one allele. Skewing was markedly increased in the neutrophils of elderly females, with 33% having  > 90% expression of one allele ( P < 0.0001). Extreme skewing was present in only 9% of the elderly T-cell samples and no evidence of T-cell clonality was found by PCR analysis of the TCRγ gene. The high level of acquired skewing of the XCIPs in myeloid cells of the elderly suggests that with time there is a change in stem cell usage with stochastic loss of some of the original stem cells. This has major implications for the use of XCIP analysis in the diagnosis of myeloid malignancies in the elderly and for gene therapy into haemopoietic stem cells.     

Minimal residual disease analysis for the prediction of relapse in children with standard-risk acute lymphoblastic leukaemia

We report a largely retrospective analysis of minimal residual disease (MRD) in a cohort of 66 children suffering from acute lymphoblastic leukaemia (ALL). All patients lacked high-risk features at diagnosis, i.e. the presenting white cell count was <50 × 10⁹/l, age 1–16 years and translocations t(9;22) and t(4;11) were not present. All were treated according to either the MRC protocols UKALL X or XI. PCR of IgH, TCRδ and TCRγ gene rearrangements and allele-specific oligoprobing were employed for the detection of MRD. Sensitivity was at least 10⁻⁴ in 78/82 (93%) probes examined.
A total of 33 patients relapsed (seven on therapy and 26 off) and 33 remain in continuing complete remission (CCR) (median follow-up 69 months from diagnosis). Of those who remain in CCR, MRD was present in the bone marrow in 32%, 10% and 0% at 1, 3 and 5 months into therapy respectively. This is in marked contrast to the presence of MRD at these times in 82%, 60% and 41% of patients who relapsed ( P <0.001, P <0.005 and P <0.005). These results provide further evidence of a strong correlation between clearance of MRD early in therapy and clinical outcome in childhood ALL.     

Rearrangements of immunoglobulin and TCR genes in lymphoid blast crisis of Ph + chronic myeloid leukaemia

Clonal rearrangements of immunoglobulin heavy chain genes as well as both T cell receptor (TCR) delta and gamma genes were found in four cases of blast crisis of Ph+ chronic myeloid leukaemia with unequivocal B cell precursor (common) immunophenotype. In one case, the TCR beta chain gene was also rearranged. Although the developmental sequence of TCR delta, gamma and beta rearrangements in T lymphocytes appeared to be respected, a full phenotypic effect, characteristic of T cell was not observed in these otherwise typical 'common' blast cells. Cytogenetic analysis ruled out the occurrence of TCR rearrangement due to structural chromosome changes. A high incidence of unexpected TCR gene rearrangements has been previously reported in the de novo 'common' acute lymphoblastic leukaemias (ALL). Our cases of chronic myeloid leukaemia (CML) in lymphoid blast crisis show that genotypic similarities may exist between these two haematological entities.     

The diversity of rearranged immunoglobulin heavy chain variable region genes in peripheral blood B cells of preterm infants is restricted by short third complementarity-determining regions but not by limited gene segment usage

The immunoglobulin diversity is restricted in fetal liver B cells. This study examined whether peripheral blood B cells of extremely preterm infants show similar restrictions (overrepresentation of some gene segments, short third complementarity-determining regions [CDR3]). DNA of rearranged immunoglobulin heavy chain genes was amplified by polymerase chain reaction, cloned, and sequenced. A total of 417 sequences were analyzed from 6 preterm infants (25-28 weeks of gestation), 6 term infants, and 6 adults. Gene segments from the entire V(H) and D(H) gene locus were rearranged in preterm infants, even though the D(H)7-27 segment was overrepresented (17% of rearrangements) compared to term infants (7%) and adults (2%). CDR3 was shorter in preterm infants (40 +/- 10 nucleotides) than in term infants (44 +/- 12) and adults (48 +/- 14) (P <.001) due to shorter N regions. Somatic mutations were exclusively found in term neonates and adults (mutational frequency 0.8% and 1.8%). We conclude that preterm infants have no limitations in gene segment usage, whereas the diversity of CDR3 is restricted throughout gestation.     

Ig heavy chain gene rearrangements in T-cell acute lymphoblastic leukemia exhibit predominant DH6-19 and DH7-27 gene usage, can result in complete V-D-J rearrangements, and are rare in T-cell receptor alpha beta lineage.

Rearranged IGH genes were detected by Southern blotting in 22% of 118 cases of T-cell acute lymphoblastic leukemia (ALL) and involved monoallelic and biallelic rearrangements in 69% (18/26) and 31% (8/26) of these cases, respectively. IGH gene rearrangements were found in 19% (13/69) of CD3(-) T-ALL and in 50% of TCRgammadelta+ T-ALL (12/24), whereas only a single TCRalpha beta+ T-ALL (1/25) displayed a monoallelic IGH gene rearrangement. The association with the T-cell receptor (TCR) phenotype was further supported by the striking relationship between IGH and TCR delta (TCRD) gene rearrangements, ie, 32% of T-ALL (23/72) with monoallelic or biallelic TCRD gene rearrangements had IGH gene rearrangements, whereas only 1 of 26 T-ALL with biallelic TCRD gene deletions contained a monoallelic IGH gene rearrangement. Heteroduplex polymerase chain reaction (PCR) analysis with VH and DH family-specific primers in combination with a JH consensus primer showed a total of 39 clonal products, representing 7 (18%) VH-(DH-)JH joinings and 32 (82%) DH-JH rearrangements. Whereas the usage of VH gene segments was seemingly random, preferential usage of DH6-19 (45%) and DH7-27 (21%) gene segments was observed. Although the JH4 and JH6 gene segments were used most frequently (33% and 21%, respectively), a significant proportion of joinings (28%) used the most upstream JH1 and JH2 gene segments, which are rarely used in precursor-B-ALL and normal B cells (1% to 4%). In conclusion, the high frequency of incomplete DH-JH rearrangements, the frequent usage of the more downstream DH6-19 and DH7-27 gene segments, and the most upstream JH1 and JH2 gene segments suggests a predominance of immature IGH rearrangements in immature (non-TCRalpha beta+) T-ALL as a result of continuing V(D)J recombinase activity. More mature alpha beta-lineage T-ALL with biallelic TCRD gene deletions apparently have switched off their recombination machinery and are less prone to cross-lineage IGH gene rearrangements. The combined results indicate that IGH gene rearrangements in T-ALL are postoncogenic processes, which are absent in T-ALL with deleted TCRD genes and completed TCR alpha (TCRA) gene rearrangements.     

Human T leukaemia virus type 1 (HTLV-1) specific T-helper cell response: clonal fluctuations and repertoire heterogeneity

The naive T-helper (Th) repertoire specific for HTLV-1 envelope (env) has been examined on antigen specific T-cell lines and clones from non-immune individuals. Clonal heterogeneity was determined by analysing the T-cell receptor (TCR) Vβ gene usage and by sequencing the hypervariable regions of the TCR genes. Fluctuations in the Vβ gene usage were determined by comparing the TCR Vβ gene profiles of T-cell lines at different times. We found that a diverse repertoire for HTLV-1 env could be triggered in vitro . Diverse Vβ genes were used by the same line tested at different times, suggesting that clonal composition of an antigen-specific T-cell line is not constant in vitro . Clones in fact may be up- and down-regulated and clonotypes undetectable at one time point can emerge upon subsequent restimulation. Therefore evaluation of the clonal composition of a T-cell line gives a snapshot of the dominant clones at the time of analysis, and does not tell the whole picture of the antigen-specific ensemble. Furthermore, by sequencing the TCR genes, we identified clones with identical Vβ gene usage which differed in hypervariable regions (CDR3), indicating their derivation from independent precursors and contributing to overall clonal heterogeneity. If these data can be extended to HTLV-1-infected patients studied in vivo , the Th cell repertoire specific for HTLV-1 env may prove very heterogenous, with important implications for vaccine development.     

Immunoglobulin heavy chain gene VH-D junctional diversity at diagnosis in patients with acute lymphoblastic leukemia

Acute lymphoblastic leukemia (ALL) represents the clonal outgrowth of transformed hematopoietic progenitor cells. We have found that blast cells in some cases of B-precursor cell ALL contain Ig heavy chain gene rearrangements with considerable diversity at the junctions of the variable (VH), diversity (D), and joining (JH) regions. This diversity consists of heterogeneous nucleotide sequences at the VH-D and, less frequently, the D-JH junctions. In two cases, different VH segments were attached to the same D-JH rearrangement. In all cases studied there was a much higher than expected frequency of nucleotide sequence changes in the VH segment. At least three mechanisms may produce these changes in different cases: (1) continuing rearrangement of the heavy chain gene, in some cases by VH addition to a preexisting D-JH; (2) VH replacement; and (3) an open-and-shut mechanism. These findings suggest that an active VDJ recombinase system is present at the time of transformation in a high percentage of ALLs. An active recombinase in the rapidly growing leukemic cell population could lead to genomic instability.     

Southern blot patterns, frequencies, and junctional diversity of T-cell receptor-delta gene rearrangements in acute lymphoblastic leukemia

Southern blot analysis of T-cell receptor (TCR)-delta gene rearrangements is useful for diagnostic studies on the clonality of lymphoproliferative diseases. We have developed 18 new TCR-delta gene probes by use of the polymerase chain reaction (PCR) techniques. Application of these probes for detailed analysis of the TCR-delta genes in normal control samples, 138 T-cell acute lymphoblastic leukemias (T-ALL), and 91 precursor B-ALL allowed us to determine the TCR-delta gene restriction map for five restriction enzymes, as well as the Southern blot restriction enzyme patterns of all theoretically possible TCR-delta gene rearrangements. Based on this information, it appeared that 97% of all 213 detected TCR-delta gene rearrangements in our series of ALL could be detected by use of the TCRDJ1 probe and that the majority (76%) of the 213 rearrangements could be identified precisely. In T-ALL, we found a strong preference for the complete rearrangements V delta 1-J delta 1 (33%), V delta 2-J delta 1 (10%), and V delta 3-J delta 1 (7%) and the incomplete rearrangement D delta 2- J delta 1 (11%). In precursor B-ALL, the majority of rearrangements consisted of V delta 2-D delta 3 (72%) and D delta 2-D delta 3 (10%). The junctional diversity of these 6 preferential TCR-delta rearrangements was analyzed and showed an extensive junctional insertion (approximately 30 nucleotides) for complete V delta-J delta rearrangements, whereas incomplete rearrangements had correspondingly smaller junctional regions. The detailed TCR-delta gene restriction map and probes presented here, in combination with the Southern blot patterns of TCR-delta gene rearrangements, are important for TCR-delta gene studies in ALL; all TCR-delta gene rearrangements can be detected and the majority can be identified precisely. Identification of rearrangements is a prerequisite for subsequent PCR analysis of TCR- delta gene junctional regions, eg, for detection of minimal residual disease during follow-up of ALL patients.     

V(H) gene family utilization in different B-cell lymphoma subgroups

V(H) gene family specific polymerase chain reaction (PCR) amplification was performed in 87 B-cell lymphoma samples from 4 different subgroups. No apparent restriction in the VH gene usage was found in follicular lymphomas, lymphoplasmacytoid lymphomas or large B-cell lymphomas, whereas a biased VH1 utilization was shown in patients with chronic lymphocytic leukemia. Eleven of 18 chronic lymphocytic leukemia cases utilized the VH1 gene family, and nucleotide sequencing of the VH1 gene rearrangements revealed that a majority utilized the DP10 (51p1) germline gene, which has been reported to be strongly associated with autoimmune disease. No VH5 or VH6 rearrangements were amplified in the chronic lymphocytic leukemia subgroup, 2 gene families which previously have been found to be over-represented in these patients. In a high proportion (40%) of large B-cell lymphomas, VH gene family-specific PCR failed to amplify any rearrangement. Using primers hybridizing to the framework regions 2 and 3 and Southern blot analysis of the immunoglobulin heavy chain locus, clonal rearrangements were displayed in two-thirds of these PCR negative cases. However, the rearrangement status could not be elucidated in 5 of 35 patients with large B-cell lymphoma.     

Analysis of junctional diversity in the preferential V delta 1-J delta 1 rearrangement of fresh T-acute lymphoblastic leukemia cells by in vitro gene amplification and direct sequencing

To define the junctional diversity of T-cell antigen receptor delta gene rearrangements in fresh T-acute lymphoblastic cells and to correlate cell phenotype with the coding potential of rearrangements, we determined the junctional nucleotide sequences of 13 T-cell antigen receptor delta gene rearrangements involving the preferentially rearranged V (V delta 1) and J (J delta 1) segments using in vitro gene amplification and direct sequencing. We showed that, as in gamma delta+ cell lines, extensive junctional diversity exists in these clones and that this diversity is due both to random nucleotide deletions/additions and to the use of at least two D delta segments. We also showed that a high percentage of these rearrangements are potentially translatable (7:13) and that such functional rearrangements occur in both surface CD3+ and CD3- cells. Comparison of alpha beta versus gamma delta surface expression demonstrates that all CD3+ T acute lymphoblastic leukemias with a functional V delta 1-J delta 1 rearrangement express a surface gamma delta receptor and are recognized by the anti-delta monoclonal antibody delta TCS1, whereas a control CD3+ gamma delta+ leukemic case that had not undergone V delta 1 rearrangement was delta TCS1-. In addition, expression of this monoclonal antibody is not restricted by V gamma or C gamma usage or by the covalent or noncovalent link between gamma and delta chains.     

Low rate of somatic hypermutations characterize progressive B-cell lymphomas

Abstract: Immunoglobulin heavy (IgH) chain gene rearrangements were characterized in 40 samples from 15 patients with B-cell lymphomas at different time points during tumour progression. Using polymerase chain reaction (PCR) amplification and single strand conformation polymorphism (SSCP) analysis of variable heavy (V_H) chain gene segments, we found that 6 cases displayed alterations in their IgH chain rearrangements at relapse. These alterations were mainly observed in follicular or transformed lymphomas, but no association to clinical features was found. Nucleotide sequence analysis revealed a low frequency of mutations in 3 cases, whereas 1 case displayed an extensive mutation rate in a compartment with transformed morphology at relapse. The mutations observed most probably resulted from somatic hypermutations. Further, the mutations were scattered randomly over the V₁₁ gene segment and no significant bias favouring amino acid substitutions was observed in 3 cases, suggesting that the tumour cells had not been subjected to antigen-driven selection. In 1 case, however, the mutation pattern indicated that the tumour cells had been affected by an antigen selection process. In the 2 remaining cases, the original V_HDJ_H rearrangement could no longer be detected by V_H gene family specific PCR at relapse, but using primers specific for the framework region 2 or 3 altered rearrangements were demonstrated, implying that mutations had been introduced in framework region 1. However, the majority of the tumour cell clones analysed were relatively stable during tumour progression, which make them eligible for analysis of minimal residual disease using the V_H gene regions as molecular markers.     

Heterogeneity of the T-cell receptor delta gene indicating subclone formation in acute precursor B-cell leukemias

Precursor B-cell acute lymphoblastic leukemias (B-ALLs) have been shown to be oligoclonal at the Ig heavy-chain (IgH) gene level in up to 40% of cases by Southern blot hybridization. In contrast, oligoclonality as deduced from diversity of T-cell receptor (TcR)-delta gene rearrangements of the immature types (ie, V delta 2-D delta 3, D delta 2-D delta 3) has not been reported, so far. We detected oligoclonality characterized by the coexistence of different junctional regions of identical V delta 2-D delta 3 rearrangements in four childhood precursor B-ALLs. No variation was found in the IgH gene status. Therefore, we define these populations as subclones. Two leukemias displayed the variants in an unequal proportion. In the other two leukemias, for which similar quantities of the coexisting rearrangements were detected, single cell-nuclei polymerase chain reaction (PCR) showed two separate leukemic populations. Subclone formation could not be demonstrated by Southern blot hybridization, but was detectable after PCR amplification of the V delta 2-D delta 3 rearrangement and separation by polyacrylamide gel electrophoresis. The variants arose independently from each other, as deduced from their individual sequences. Using subclone-specific oligonucleotides for hybridization to amplified DNA obtained at diagnosis and during follow- up from bone marrow samples, we demonstrate, (1) specificity of all subclone-deduced probes, (2) that one residual leukemic cell can be detected in 10(4) to 10(5) normal mononuclear cells in a semiquantitative assay, and (3) that none of the subclones persisted after induction therapy. We propose that in a leukemic cell population, TcR-delta gene diversity arises after rearrangements of the IgH genes resulting in apparent clonality at the IgH gene level. However, cells are oligoclonal, if the TcR-delta gene rearrangements are considered. As various subclones may respond differently to chemotherapy, they may hamper the detection of minimal residual disease. Therefore, we use all subclone-specific oligonucleotides for hybridization to amplified DNA from follow-up samples.