Recently, we and others described a new chromosomal rearrangement, that is, inv(7)(p15q34) and t(7;7)(p15;q34) involving the T-cell receptor beta (TCRβ) (7q34) and the HOXA gene locus (7p15) in 5% of T-cell acute lymphoblastic leukemia (T-ALL) patients leading to transcriptional activation of especially HOXA10. To further address the clinical, immunophenotypical and molecular genetic findings of this chromosomal aberration, we studied 330 additional T-ALLs. This revealed TCRβ-HOXA rearrangements in five additional patients, which brings the total to 14 cases in 424 patients (3.3%). Real-time quantitative PCR analysis for HOXA10 gene expression was performed in 170 T-ALL patients and detected HOXA10 overexpression in 25.2% of cases including all the cases with a TCRβ-HOXA rearrangement (8.2%). In contrast, expression of the short HOXA10 transcript, HOXA10b, was almost exclusively found in the TCRβ-HOXA rearranged cases, suggesting a specific role for the HOXA10b short transcript in TCRβ-HOXA-mediated oncogenesis. Other molecular and/or cytogenetic aberrations frequently found in subtypes of T-ALL (SIL-TAL1, CALM-AF10, HOX11, HOX11L2) were not detected in the TCRβ-HOXA rearranged cases except for deletion 9p21 and NOTCH1 activating mutations, which were present in 64 and 67%, respectively. In conclusion, this study defines TCRβ-HOXA rearranged T-ALLs as a distinct cytogenetic subgroup by clinical, immunophenotypical and molecular genetic characteristics.
T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive malignancy of immature T cells characterized by high numbers of bone marrow and circulating blast cells, enlargement of mediastinal lymph nodes and often central nervous system involvement.1 T-ALL accounts for approximately 15% of pediatric and 25% of adult ALL cases. During the past 20 years, a large number of genes involved in the pathogenesis of T-ALL have been identified by molecular characterization of recurrent chromosomal aberrations and cryptic alterations.2 Several oncogenes initially identified by rare genetic alterations were shown to be functionally activated in subsets of T-ALLs such as LMO1, LMO2, LYL1, NOTCH1 in the absence of the corresponding genetic aberration. Recently, we3 and others4 described a new recurrent chromosomal aberration in a subgroup of T-ALLs affecting the TCRβ (7q34) and HOXA (7p15) loci. This group of patients showed a significant upregulation of particular HOXA cluster genes, under the influence of regulatory sequences embedded in the TCRβ locus. In this study,3 especially HOXA10 showed a significant higher expression level in the TCRβ-HOXA rearranged subgroup compared to non-TCRβ-HOXA rearranged T-ALLs. Ectopic expression of HOXA cluster genes was already described in other cytogenetic subgroups of T-ALL, that is, MLL5, 6 and CALM-AF10 rearranged cases,4, 7, 8 further underlining the importance of HOXA genes in T-cell oncogenesis. The mechanisms underlying this overexpression are however different. Human leukemias and cell lines carrying MLL rearrangements show upregulated expression of especially 5′ HOXA cluster genes, that is, HOXA5, HOXA7, HOXA9, HOXA10 and HOXA11.5, 9, 10 In contrast, 3′ HOXA genes showed low levels or no expression.11 The transcriptional activation of HOX cluster genes by MLL was shown to be dependent on histone H3 methylation of HOX genes, with HOXA9 and HOXA7 proven to be the direct targets.12, 13, 14, 15 Similarly, T-ALLs carrying the CALM-AF10 rearrangement show an elevated expression of especially HOXA5, HOXA9, HOXA10 and BMI1,7 suggesting that these two aberrations activate common oncogenic pathways. However, as no obvious DNA binding domain exists in CALM, the mechanism of HOXA gene activation must be different. Previous gene expression analysis on a large group of T-ALLs already showed the existence of a HOXA-expressing subgroup, consisting of MLL, CALM-AF10 and TCRβ-HOXArearranged cases.4 Interestingly, these data pointed to the expression of a specific short HOXA10b transcript in the TCRβ-HOXA rearranged cases, which was absent in other T-ALLs. Given these findings and in order to make a comprehensive picture of patients carrying the TCRβ-HOXA rearrangements, we analyzed additional T-ALL patients using fluorescent in situ hybridization (FISH) and real-time quantitative reverse transcriptase-PCR (RT-PCR) for expression of the long HOXA10 transcript. Cases showing overexpression of the long transcript of HOXA10 were subsequently analyzed for expression of the short HOXA10b transcript, which was shown to be specifically expressed in TCRβ-HOXA rearranged cases. This study described the largest group of TCRβ-HOXA rearranged T-ALLs identified so far and interestingly showed one patient carrying a triplication of a TCRβ-HOXA on a ring chromosome 7, pointing to an additional mechanism of transcriptional activation of HOXA cluster genes. Moreover, all 14 TCRβ-HOXA+ patients showed absence of additional molecular–cytogenetic alterations like SIL-TAL1, HOX11, HOX11L2, CALM-AF10andNUP214-ABL1, providing further evidence for a distinct cytogenetic entity. In contrast, deletions of 9p21 harboring the tumor suppressor genes CDKN2A and CKDN2B and NOTCH1 activating mutations were present in 64 and 67% of TCRβ-HOXA+ patients, pointing to a multistep oncogenesis in this cytogenetic T-ALL subgroup. Of further interest is the finding of the highly clustered breakpoints in three TCRβ-HOXA-positive cases and all three showed breakpoints in intron 1A from HOXA9 and JB2.7–JB2.1 segments of the TCRβ gene.
Materials and methods
Diagnostic bone marrow or pleural fluid samples from T-ALL patients were collected retrospectively at different cytogenetic centers. The only inclusion criterion was the diagnosis of T-ALL and the availability of fixed cells for FISH (n=424) and/or RNA or frozen cells for real-time quantitative RT-PCR (n=170 of 424). Besides newly diagnosed T-ALLs, these series of patient samples includes 229 cases analyzed in the HOX11L2 study of the Groupe Francophone de Cytogénétique Hématologique (GFCH),16 patients analyzed in the first study,3 that is, 94 patients analyzed with FISH and 26 with real-time quantitative PCR, and patients of another study,4 that is, 92 patients analyzed by FISH and 21 with real-time quantitative PCR. This total series of patients included 50% children and adults. Diagnosis of T-ALL was made according to the morphological and cytochemical criteria of the French–American–British classification17 and by immunophenotyping.18
Immunophenotypical analyses were carried out in the respective centers according to established protocols. Blast cells were analyzed for forward/side scatter and fluorescence by BD FACS Calibur using monoclonal antibodies directed against CD34, CD33, CD13, CD2, CD3, CD5, CD7, CD1a, TdT, CD10, CD4, CD8, TCRαβ, TCRγδ, CD19 and CD20.
Diagnostic specimens were cultured and harvested for cytogenetic analysis according to established methods. Chromosome slides were G- or R-banded. Chromosomal aberrations are described according to the guidelines of an International System for Human Cytogenetic Nomenclature (ISCN 1995).19
Fluorescence in situ hybridization
RPCI-11 (Human BAC Library) clones were selected using the bioinformatics resources available at NCBI (http://genome.ucsc.edu) and Ensembl Genome Browser (http://www.ensembl.org/). Clones were provided by the Welcome Trust Sanger Institute (Cambridge, UK) and Invitrogen (Paisley, Scotland).
Disruption of the TCRβ or HOXA gene locus was assessed in all 424 cases by dual-color FISH with TCRβ or HOXA flanking probes. Clones for the TCRβ and HOXA gene locus applied in the present study are shown in Figure 2. DNA isolation of bacterial artificial chromosome (BAC) clones, labelling and FISH were performed as described previously.20 Amplification of NUP214-ABL1 was investigated by the LSI BCR/ABL1 dual-color, ES probe (Vysis, Abbott, Ottignies, Belgium) in 137 samples from which cytogenetic cell suspension was left out. Disruption of the MLL genomic locus was assessed by FISH with the commercial LSI MLL dual-color, break-apart probe (Vysis, Abbott, Ottignies, Belgium) in all patients (n=43) showing overexpression of HOXA10. Deletion of 9p21 was detected using the LSI p16 (9p21)/CEP 9 dual-color probe (Vysis, Abbott, Ottignies, Belgium) in all TCRβ-HOXA rearranged cases.
RNA isolation, cDNA synthesis and real-time quantitative PCR
Frozen cells or RNA were available for expression analysis in 170 of 424 T-ALL patients. These included all TCRβ-HOXA rearranged cases (n=14). All human samples were obtained according to the guidelines of the local ethical committees. RNA isolation from frozen cells was performed using TRIzol (Invitrogen, Merelbeke, Belgium) and the RNeasy mini kit (Qiagen, Hilden, Germany) or RNA Plus (Appligene, Illkirch, France) for detection of the NUP214-ABL1 amplification at different laboratories. DNase pretreatment, cDNA synthesis and SYBR green real-time quantitative PCR were performed for HOXA10 and HOXA10b expression as described previously.3 Reactions were performed on an ABI Prism 5700 sequence detector (Applied Biosystems, Foster City, CA, USA). Real-time quantitative RT-PCR data analysis and expression normalization were performed using three internal control genes with the qBase data analysis software (Hellemans et al., in preparation; medgen.ugent.be/qbase). Expression analysis for the full-length transcript of the HOXA10 gene (Fw: 5′-IndexTermGAGAGCAGCAAAGCCTCGC-3′; Rev: 5′-IndexTermCCAGTGTCTGGTGCTTCGTG-3′) was performed in all 170 T-ALL patients. Cases showing ‘overexpression’ of this transcript were subsequently tested for the expression of the short HOXA10b transcript (Fw: 5′-IndexTermGCACTTCCGATCAATGTCAA-3′; Rev: 5′-IndexTermAGCGAACAAAGGCCAAGTT-3′), or as described.4 Cytogenetically proven TCRβ-HOXA rearranged cases were analyzed for the expression pattern of all HOXA cluster genes with primers used as described.3
The presence of a NUP214-ABL1 fusion was assessed by quantitative real-time RT-PCR using the fluorescent TaqMan methodology. Four NUP214 primers were used (X23, X29, X31, X34) (E Delabesse, personal communication) in combination with ABL primer (ENR 561) and probe (ENP541), previously designed for BCR-ABL transcript quantitation.21 This PCR reaction can detect all NUP214 breakpoints that have been described so far.22
Long-distance PCR and cycle sequencing
For 10 patients showing a TCRβ-HOXA rearrangement (eight inv(7)(p15q34), one t(7;7)(p15;q34) and one triplication on a ring chromosome), DNA was extracted using the QIAmp DNA mini kit (Qiagen, Hilden, Germany). Long-distance PCR was performed using the iProof High-Fidelity PCR Kit (BioRad, Nazareth, Belgium), according to the manufacturer's instructions. Forward and reverse primers were selected in the HOXA9 and TCRβ, Dβ1 gene, respectively, as breakpoints of four patients were suggested to be located between the HOXA9 and HOXA10 gene on der (7p) and upstream from the Dβ1 locus of the TCRβ gene on der (7q), respectively.4 Subsequently, fragments were sequenced using the ABI Prism BigDye Terminator v3.0 Ready Reactions Cycle Sequencing Kit and analyzed on an ABI3730 XL Sequence detector.
RT-PCR detection assays
A standard RT-PCR protocol was used to detect the CALM-AF10 fusion in two separate RT-PCR assays using one CALM forward primer (5′-IndexTermGCAATCTTGGCATCGGAAAT-3′) and either AF10 reverse primer AF10AS559 (5′-IndexTermCGATCATGCGGAACAGACTG-3′) or AF10AS1002 (5′-IndexTermGCGCTTCAATGATCCAGATATAGAG-3′) as described.8
HOX11L2 was detected using a standard RT-PCR protocol with forward primer (5′-IndexTermGCGCAT CGGCCACCCCTACCAGA-3′) and reverse primer (5′-IndexTermCCGCTCCGCCTCCCGCTCCTC-3′), according to Bernard et al.25
Detection of the SIL-TAL1 fusion was performed using a multiplex RT-PCR for simultaneous screening of multiple chromosomal aberrations in acute leukemia (HemaVision, BioRad). Primers used were SIL prim (5′-IndexTermCGACCCCAACGTCCCAGAG-3′) and TAL1 prim (5′-IndexTermCGGTCATCCTGGGGC ATATTT-3′) and nested primers SIL nest (5′-IndexTermCCCGCTCCTACCCTGCAAAC-3′) and TAL1 nest (5′-IndexTermAGACCGGCCCCTCTGAATAG-3′).26
NOTCH1 mutation detection
Mutation detection of the NOTCH1 receptor was performed on genomic DNA in 12 out of 14 TCRβ -HOXA-positive cases from which DNA was available. PCR amplification of exons 26, 27 of the HD (heterodimerization) domain and of the PEST (proline, glutamate, serine and threonine) domain encoding region of exon 34 was performed with the following primer pairs: ex26F (5′-IndexTermTGAGGGAGGACCTGAACTTG-3′) and ex26R (5′-IndexTermTGGAATGCTGCCTCTACTCC-3′); ex27F (5′-IndexTermGTTGGTGGGTATCTGGGATG-3′) and ex27R (5′-IndexTermCGGAGTGCCATTCAGAAAAT-3′); and ex34F (5′-IndexTermCCATGGCTACCTGTCAGACG-3′) and ex34R (5′-IndexTermTGGCTCTCAGAACTTGCTTGT-3′). Subsequent sequencing of PCR products was performed with primers ex26seq1 (5′-IndexTermGAGGGCCCAGGAGAGTTG-3′) and ex26seq2 (5′-IndexTermCACGCTTGAAGACCACGTT-3′) for exon 26; ex27seq1 (5′-IndexTermCGGGGGAGGAGGAAG-3′) and ex27seq2 (5′-IndexTermCTGCAGGCAGAGCCTGTT-3′) for exon 27; and ex34F, ex34R and ex34seq1 (5′-IndexTermGCTGCACAGTAGCCTTGCTG-3′) for exon 34.
Statistical analysis was performed using SPSS Software (SPSS Inc., Chicago, IL, USA) version 12.0. The non-parametric Mann–Whitney U-test (two-tailed) was used to evaluate the significance of difference in mean expression levels between the patients’ subgroups (TCRβ-HOXA rearranged versus TCRβ-HOXA non-rearranged patients) for the different HOXA cluster genes. Differences in the expression level of a gene were considered statistically significant if P-value <0.05.
FISH analysis was performed in 424 T-ALL cases and real-time quantitative RT-PCR in 170 of these patients. These series of patient samples includes the patients analyzed in the first two studies, that is, 94 patients analyzed with FISH and 26 with real-time quantitative PCR,3 and 92 patients analyzed with FISH and 21 with real-time quantitative PCR.4 This study also includes 229 patients studied in the HOX11L2 study of the GFCH.16
Incidence and immunophenotypical and molecular features of TCRβ-HOXA rearranged T-ALL
The present study performed on 424 T-ALLs identified five patients carrying TCRβ-HOXA rearrangements in addition to the first two studies,3, 4 which brings the total to 14 cases (3.3% of T-ALLs studied): nine inv(7), two t(7;7) and one TCRβ-HOXA triplication on a ring chromosome. For the remaining two cases, no mitoses were available, which hampers a distinction between an inv(7) or a t(7;7). The median age at diagnosis was 25.7 years (range 9–49 years) and affected both men and women (M/F: 8/6), whereas children were less affected (n=5 of 14, <18 years). All but three (n=11 of 14) TCRβ-HOXA T-ALLs showed a typical CD2-negative immunophenotype, besides CD4 single positivity and low or lack of TCRαβ of TCRγδ surface expression, pointing to an immature stage of maturation arrest. Remarkably, none of these TCRβ-HOXA T-ALLs showed additional molecular–cytogenetic aberrations commonly observed in T-ALL (HOX11, HOX11L2, CALM-AF10, NUP214/ABL1, SIL-TAL1), except for 9p21 deletions, which were present in 64% (9/14) of TCRβ-HOXA-positive cases, most of which were mono-allelic deletions. NOTCH1 activating mutations were present in 67% (8/12) of the TCRβ-HOXA T-ALLs, mostly affecting exon 26 of the HD domain and included deletions, insertions and missense mutations. PEST domain mutations were present in two cases, both were point mutations that created a premature stop codon, whereas exon 27 of the HD domain carried a missense mutation in a single patient. One patient showed both a PEST domain and an HD domain exon 26 mutation.
Karyotypic findings in TCRβ-HOXA-positive T-ALL patients
Conventional karyotyping showed clonal aberrations in seven out of 14 TCRβ-HOXA rearranged cases. Of these seven patients, an inv(7)(p15q34) was cytogenetically found in all metaphases in two patients (case no. 2 and no. 4; Figure 1), whereas two other patients showed only a del(7)(p15). Taken together, a chromosome 7 aberration was suggested in four out of 14 cases. Interestingly, case no. 10 harbored a monosomy 7 and a ring chromosome, which was shown to be derived from chromosome 7 upon further analysis. Only few additional chromosomal abnormalities were found in five patients and included add(6)(qter) (case no. 2), del(6)(q14) (case no. 5), add(5)(q31) (case no. 8), del(9)(p21) (case no. 5), del(9)(p12p24) (case no. 4) and +11 and +21 (case no. 3).
TCRβ-HOXA triplication as an alternative mechanism for HOXA upregulation
Based on real-time quantitative RT-PCR results for HOXA10 and HOXA10b expression in 170 T-ALLs, we could identify five additional TCRβ-HOXA rearranged cases. Among these, there was a patient showing an interesting hybridization pattern, using FISH with TCRβ and HOXA flanking probes: the TCRβ (RP11-1220K2) and HOXA (RP5-1103I5) proximal flanking BACs showed 3–6 signals in 98% of interphase cells with deletion of both the TCRβ (RP11-556I13) and HOXA (RP1-167F23) distal flanking BACs (Figure 2). The majority of metaphases showed a triplication of these proximal flanking BACs on a ring/marker chromosome, which was already detected by karyotypic analysis (see Table 1). Further FISH analysis, using a whole chromosome paint probe, identified the ring chromosome as chromosome 7 (data not shown). Given the abnormal hybridization pattern using TCRβ and HOXA flanking FISH clones, additional FISH was performed with BACs covering the HOXA and TCRβ gene locus (Figure 2) and revealed the same hybridization pattern for clones RP11-1025G19, RP11-1132K14 (HOXA covering) and RP11-784K24, RP11-701D14 (TCRβ covering), that is, 3–6 fusion signals in both the nuclei and on the ring chromosome, whereas telomeric HOXA covering FISH clones showed a deletion in the majority of cells (RP11-1036C18, RP11-163M21). Finally, a FISH analysis using a combination of both proximal TCRβ and HOXA flanking FISH probes confirmed the juxtaposition and triplication of these two loci on the ring chromosome 7.
TCRβ-HOXA rearranged T-ALL specifically express HOXA10b
Real-time quantitative PCR for HOXA10 expression in 170 T-ALL cases revealed an upregulated expression in 43/170 (25.2%) of cases, whereas only 14 carried the TCRβ-HOXA rearrangement (8.2%). Upregulation of HOXA10 expression was defined as expression of more than the mean expression level of all samples analyzed. Given the fact that within the TCRβ-HOXA-positive group, there was a large fluctuation of expression levels with some samples showing very low expression (similar to thymocytes), we put the threshold for positivity sufficiently low, in order to avoid missing any true positive cases. Consequently, it can be expected that some of the cases included as positive represent normal HOXA10 expression reflecting their stage of differentiation arrest. Five CALM-AF10-positive patients and two MLL rearranged patients, which are known to be HOXA-expressing T-ALL subtypes, were also included in this series. Interestingly, one patient showing HOXA10 overexpression carried an NUP214/ABL1 fusion. However, for the remaining 21/170 patients showing elevated HOXA10 expression, there was a lack of evidence for one of the genetic alterations mentioned above. In 30 out of the 43 patients showing HOXA10 overexpression, material was available to detect the presence of the HOXA10b short transcript. Interestingly, 16 of these 30 patients tested showed expression of HOXA10b, whereas expression was absent in the remaining T-ALL patients. Remarkably, these 16 HOXA10b-positive patients included all 14 TCRβ-HOXA rearranged cases and only two additional T-ALL patients lacking this rearrangement or other chromosomal defects. These data demonstrate that overexpression of this transcript is almost typically found in the TCRβ-HOXA rearranged cases.
HOXA expression profiling of TCRβ-HOXA rearranged T-ALL
Gene expression of the different HOXA cluster genes (HOXA1, HOXA2, HOXA3, HOXA4, HOXA5, HOXA6, HOXA7, HOXA9, HOXA10, HOXA11 and HOXA13) was measured by real-time quantitative RT-PCR in 10 TCRβ-HOXA rearranged T-ALLs and in 21 non-TCRβ-HOXA rearranged cases. As expected, TCRβ-HOXA rearranged cases showed an upregulation of especially 5′ HOXA cluster genes compared to non-TCRβ-HOXA cases (Figure 3). However, upregulation of individual HOXA cluster genes was statistically significant only for genes HOXA3 (P=0.007), HOXA9 (P=0.0002), HOXA10 (P=0.001) and HOXA11 (P=1.05 × 10−5). Interestingly, case no. 10 carrying the TCRβ-HOXA fusion on a ring chromosome 7 showed an expression pattern of HOXA cluster genes more in favor of the 3′ located HOXA cluster genes (HOXA1, HOXA2, HOXA3) than 5′ located genes HOXA9, HOXA10, HOXA11 and HOXA13. This might suggest a different breakpoint localization compared to breakpoints clustered in the HOXA9 gene in three other TCRβ-HOXA rearranged cases (see below).
Clustered breakpoints in TCRβ-HOXA rearranged T-ALL cases
Long-distance PCR using HOXA9 forward and TCRβ-Dβ1 reverse primers as suggested by Soulier et al.4 was successful in three out of 10 patients. This revealed amplicons with various lengths for each patient: patient 2, 2–3 kb; patient 8, 6–7 kb; and patient 9, 4–5 kb (not shown). Sequencing analysis showed that all three patients had breakpoints within a 2.9 kb region of intron 1A (total length: 3.8 kb) of the HOXA9 gene and within a 1.1 kb region of the TCRβ gene. Breakpoints within the TCRβ gene were located at the JB segments: JB2.7 (case no. 2), JB2.5 (case no. 8) and JB2.1 (case no. 9) (Figure 4). RSS sequences were found at the TCRβ, JB segments at 50–100 bp upstream from the respective breakpoints, suggesting aberrant VDJ recombination as possible mechanism leading to this translocation. The fact that this analysis was unsuccessful in the remaining seven patients, including the TCRβ-HOXA triplication, might suggest the existence of other breakpoints located 3′ of HOXA9 gene or alternative TCRβ breakpoints.
In this study, we report the findings of a retrospective screening of 424 T-ALL patients in search for specific clinical and biological characteristics of T-ALLs carrying a TCRβ-HOXA rearrangement, which we and others4 recently described as a new cytogenetic entity.3 This large-scale study revealed a slightly lower incidence of the abnormality than previously assumed, that is, 3.3% (14/424) compared to our first report (5%).3 The median age at diagnosis was situated in the third decade (25.9 years; range 9–49 years) and affects both men and women (M/F: 8/6). Clinical findings were not significantly different from T-ALL in general.27 The number of patients collected so far (n=14) is rather small to make conclusions regarding survival. Nonetheless, four out of 14 patients deceased 24 (n=3) and 48 (n=1) months after diagnosis, which is comparable to the overall survival in T-ALL.
The reported typical immunological profile of TCRβ-HOXA rearranged cases (CD2−, CD4+, CD8−) was confirmed in all but three cases. T-ALLs used to be classified according to the European Group for the Immunological Characterization of Leukemias (EGIL) classification in T1 and T2 (immature T-ALL) and T3 and T4 (mature T-ALL), largely depending on the expression of CD1a.18 With this classification, the TCRβ-HOXA+ T-ALLs could be assigned to the group of mature T-ALLs (T3–T4). Recently, a TCR-based classification of T-ALLs was described, which demonstrated that T-ALLs largely reproduce normal thymic development and allowed separation of cases into TCRαβ+, TCRγδ+ T-ALLs and immature/uncommitted, TCR and cytoplasmic TCRβ-negative cases.28 Furthermore, the authors demonstrated that specific oncogenetic subclasses of T-ALL were associated with a specific, age-independent stage of maturation arrest. In line with this classification, it seems that the oncogenic pathways leading to the TCRβ-HOXA rearrangements are mostly situated at the immature stage of thymic development based on the lack of expression of surface TCR.
Interestingly, the 14 TCRβ-HOXA rearranged cases failed to show additional molecular/cytogenetic features frequently found in T-ALL such as HOX11, HOX11L2, SIL-TAL1 deregulation or CALM-AF10 rearrangements and NUP214-ABL1 amplification. Recently, the first T-ALL patient carrying a TCRδ-HOXA rearrangement was described and most interestingly, this case carried a CALM-AF10 aberration,29 raising questions regarding mechanistic or oncogenic synergy between CALM-AF10 and HOXA rearrangements. The lack of additional molecular/cytogenetic features in the TCRβ-HOXA rearranged cases is in contrast to other cytogenetic subgroups like NUP214-ABL1 amplified T-ALL, which is associated with HOX11, HOX11L2 upregulation, but further suggests that the TCRβ-HOXA rearranged T-ALLs have a unique oncogenic pathway not shared with other known oncogenic events. However, deletions of 9p21 harboring the tumor suppressor genes CDKN2A (encoding p14 and p16) were present in nine out of 14 cases, suggesting a multistep pathogenesis with deletion of a tumor suppressor gene acting in concert with activation of HOXA proto-oncogenes. Similarly, activating NOTCH1 mutations were present in as much as 67% (8/12) of TCRβ-HOXA-positive T-ALLs, which is in line with previous findings in T-ALL.30 These findings further suggest that HOXA oncogenes might cooperate with NOTCH1 in T-ALL pathogenesis.
Expression profiling of 10 TCRβ-HOXA-positive cases confirmed our previous findings, that is, a significant upregulation of HOXA cluster genes HOXA3 (P=0.007), HOXA9 (P=0.0002), HOXA10 (P=0.001) and HOXA11 (P=1.05 × 10−5), whereas another study found different levels of upregulation of all HOXA genes in this subgroup.4 This discordance might be due to a different approach towards normalizing gene expression data and lack of statistical analysis. Interestingly, Soulier et al.4 pointed to the presence of a specific short alternative HOXA10b transcript, which was exclusively present in the TCRβ-HOXA rearranged cases. Real-time quantitative RT-PCR for expression of HOXA10b revealed a specific expression in all TCRβ-HOXA-positive cases (n=14), whereas MLL+ and CALM-AF10+ cases lacked expression of this alternative transcript. This finding further suggests that the TCRβ-HOXA subgroup of T-ALL has at least a specific oncogenic pathway not shared with other cytogenetic subgroups.
A hallmark of homeobox gene expression seems to be a high frequency of alternative splicing events leading to transcripts that would encode partial homeobox proteins lacking either the homeodomain or transcriptional regulatory domains, or containing alternative putative regulatory regions.31, 32 Alternatively, spliced homeobox-containing cDNAs from the HOXA10 gene were cloned first from two myeloid leukemia cell lines32 and shared the homeodomain and 3′ flanking regions but had unique 5′ flanking regions (Figure 5). Expression of the full-length transcript was detected predominantly in cell lines with a myeloid phenotype, whereas the short transcript was the major transcript in a B-cell line (CESS) and a T-cell line (MOLT). The splicing of HOXA10 in normal bone marrow and primary samples of myeloid leukemias seemed to be different from that observed in leukemic cell lines,33 with the full-length HOXA10 transcript being the predominant transcript in this group, whereas immortalized cell lines contained the additional short transcript. However, in this study, no HOXA10 expression was detected in cell lines or primary samples with lymphoid or erythroid features, whereas we found HOXA10 expression in 25.2% of primary T-ALLs and in one T-ALL cell line (RPMI 8402). These conflicting results could be due to the small number of T-ALL samples analyzed (n=6) and the lower sensitivity of the methods used (RNase protection assay) in this study. The functional role of the N-terminal region of homeobox proteins is largely unknown. However, Zappavigna et al.34 reported the first evidence that sequences in the N-terminal region of a HOX protein influence transcriptional activity.
Interestingly, we found one patient showing overexpression of both HOXA10 and HOXA10b, which revealed a triplication of HOXA and TCRβ flanking clones with deletion of the distal clones on a ring chromosome 7. In this case, HOXA10 overexpression was not only due to the juxtaposition of these two genes, but also due to gene dosage. Ring chromosome formation may occur by two mechanisms: (1) double-strand breaks in each arm of a chromosome with subsequent fusion of the proximal broken ends; and (2) fusion of dysfunctional telomeres from the same chromosome.35 FISH with telomeric probes showed absence of both 7pter and 7qter in the ring chromosome, suggesting that the first mechanism was responsible for ring formation in this patient. Acquired ring chromosomes have been found in many types of human neoplasia, especially in mesenchymal tumors but infrequent in acute leukemia.35, 36 In this particular patient, the ring chromosome showed little or no size variation, and lacked telomeric sequences but carried multiple3, 4, 5, 6 copies of the TCRβ-HOXA juxtaposition. So far, proto-oncogene amplification by ring chromosome formation was mostly described in solid tumors such as dermatofibrosarcoma protuberans, which is characterized by a reciprocal t(17;22)(q22;q13) or more commonly by supernumerary ring chromosomes containing amplified sequences from chromosomes 17, 22 and 8.37, 38 In leukemia, the mechanism of proto-oncogene activation by amplification on ring chromosomes was never reported in lymphoid leukemia but only demonstrated in rare myeloid leukemia cases. These amplifications on ring chromosomes have been described in the following myeloid leukemia cases: three cases with MLL amplification,39 one with ETV6 amplification40 and two cases with AML1 amplification on ring chromosomes.41 Several amplification mechanisms have been proposed, that is, looping out of extra chromosomal sequences42 without evidence of chromosomal rearrangements, breakage–fusion–bridge cycles that can be triggered by fragile site induction43 and a translocation–deletion–amplification model.44, 45 Most of these mechanisms rely on unequal segregation of chromosome sequences during mitosis.
Based on karyotypic analysis, the TCRβ-HOXA rearrangement could be readily detected in two patients with exclusively abnormal metaphases, whereas two patients showed a del(7)(p15). This suggests that the rearrangement is not fully cryptic, but detection depends on the percentage of good-quality abnormal clonal metaphases within the karyotype and the banding technique (R-banding). Additional karyotypic aberrations were found in five separate cases and included add(6)6(qter), del(6)(q14), add(5)(q31), del(9)(p21), del(9)(p12p24),+11 and +21.
Genomic breakpoints in three out of 10 patients showed a clustered pattern in the TCRβ and HOXA gene locus and were located between the JB2.1 and JB2.7 segments at the TCRβ locus (1.1 kb) and within the intron 1A of the HOXA9 gene (2.6 kb). The fact that this could not be demonstrated in the remaining seven patients, including the TCRβ-HOXA triplication, might suggest the existence of other breakpoints in the HOXA and/or TCRβ regions. RSS sequences were found at the TCRβ, JB segments at 50–100 bp upstream from the respective breakpoints, suggesting aberrant VDJ recombination as possible mechanism leading to this translocation. This is in line with previous reports of genomic breakpoints within TCR genes46, 47 suggesting aberrant VDJ recombination as the most important mechanism leading to these translocations. However, recently, a t(7;14)(p15;q11) involving HOXA genes on chromosome 7p and TCRδ genes at 14q11 failed to show RSS-like sequences on the derivative chromosome 7, suggesting that other mechanisms might be involved in this TCR rearrangement.29
In conclusion, the present study covers the largest group of TCRβ-HOXA rearranged T-ALLs identified so far and summarizes clinical, immunophenotypical and molecular genetic characteristics of this subgroup. Most interestingly, this series includes the first case of oncogene triplication by ring chromosome formation in T-ALL, that is, triplication of the TCRβ-HOXA fusion, probably as a secondary genetic event subsequent to chromosomal rearrangement.
Pui CH, Relling MV, Downing JR . Acute lymphoblastic leukemia. N Engl J Med 2004; 350: 1535–1548.
De Keersmaecker K, Marynen P, Cools J . Genetic insights in the pathogenesis of T-cell acute lymphoblastic leukemia. Haematologica 2005; 90: 1116–1127.
Speleman F, Cauwelier B, Dastugue N, Cools J, Verhasselt B, Poppe B et al. A new recurrent inversion, inv(7)(p15q34), leads to transcriptional activation of HOXA10 and HOXA11 in a subset of T-cell acute lymphoblastic leukemias. 2005; 19: 358–366.
Soulier J, Clappier E, Cayuela JM, Regnault A, Garcia-Peydro M, Dombret H et al. HOXA genes are included in genetic and biologic networks defining human acute T-cell leukemia (T-ALL). Blood 2005; 106: 274–286.
Armstrong SA, Staunton JE, Silverman LB, Pieters R, den Boer ML, Minden MD et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet 2002; 30: 41–47.
Ferrando AA, Armstrong SA, Neuberg DS, Sallan SE, Silverman LB, Korsmeyer SJ et al. Gene expression signatures in MLL-rearranged T-lineage and B-precursor acute leukemias: dominance of HOX dysregulation. Blood 2003; 102: 262–268.
Dik WA, Brahim W, Braun C, Asnafi V, Dastugue N, Bernard OA et al. CALM-AF10+ T-ALL expression profiles are characterized by overexpression of HOXA and BMI1 oncogenes. Leukemia 2005; 19: 1948–1957.
Asnafi V, Radford-Weiss I, Dastugue N, Bayle C, Leboeuf D, Charrin C et al. CALM-AF10 is a common fusion transcript in T-ALL and is specific to the TCRgammadelta lineage. Blood 2003; 102: 1000–1006.
Yeoh EJ, Ross ME, Shurtleff SA, Williams WK, Patel D, Mahfouz R et al. Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell 2002; 1: 133–143.
Rozovskaia T, Feinstein E, Mor O, Foa R, Blechman J, Nakamura T et al. Upregulation of Meis1 and HoxA9 in acute lymphocytic leukemias with the t(4:11) abnormality. Oncogene 2001; 20: 874–878.
Ayton PM, Cleary ML . Transformation of myeloid progenitors by MLL oncoproteins is dependent on HOXA7 and HOXA9. Genes Dev 2003; 17: 2298–2307.
Milne TA, Briggs SD, Brock HW, Martin ME, Gibbs D, Allis CD et al. MLL targets SET domain methyltransferase activity to Hox gene promoters. Mol Cell 2002; 10: 1107–1117.
Nakamura T, Mori T, Tada S, Krajewski W, Rozovskaia T, Wassell R et al. ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation. Mol Cell 2002; 10: 1119–1128.
Santos-Rosa H, Schneider R, Bernstein BE, Karabetsou N, Morillon A, Weise C et al. Methylation of histone H3 K4 mediates association of the Isw1p ATPase with chromatin. Mol Cell 2003; 12: 1325–1332.
Bernstein BE, Humphrey EL, Erlich RL, Schneider R, Bouman P, Liu JS et al. Methylation of histone H3 Lys 4 in coding regions of active genes. Proc Natl Acad Sci USA 2002; 99: 8695–8700.
Berger R, Dastugue N, Busson M, Van Den Akker J, Perot C, Ballerini P et al. t(5;14)/HOX11L2-positive T-cell acute lymphoblastic leukemia. A collaborative study of the Groupe Français de Cytogénétique Hématologique (GFCH). 2003; 17: 1851–1857.
Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposals for the classification of the acute leukaemias. French–American–British (FAB) co-operative group. Br J Haematol 1976; 33: 451–458.
Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A et al. Proposals for the immunological classification of acute leukemias. European Group for the Immunological Characterization of Leukemias (EGIL). Leukemia 1995; 9: 1783–1786.
Mitelman F . An International System for Human Cytogenetic Nomenclature. ISCN: S Karger, Basel, 1995.
Van Limbergen H, Poppe B, Michaux L, Herens C, Brown J, Noens L et al. Identification of cytogenetic subclasses and recurring chromosomal aberrations in AML and MDS with complex karyotypes using M-FISH. Genes Chromosomes Cancer 2002; 33: 60–72.
Gabert J, Beillard E, van der Velden VH, Bi W, Grimwade D, Pallisgaard N et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia – a Europe Against Cancer program. Leukemia 2003; 17: 2318–2357.
Graux C, Cools J, Melotte C, Quentmeier H, Ferrando A, Levine R et al. Fusion of NUP214 to ABL1 on amplified episomes in T-cell acute lymphoblastic leukemia. Nat Genet 2004; 36: 1084–1089.
Ferrando AA, Neuberg DS, Staunton J, Loh ML, Huard C, Raimondi SC et al. Gene expression signatures define novel oncogenic pathways in T cell acute lymphoblastic leukemia. Cancer Cell 2002; 1: 75–87.
Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002; 3: 1–11.
Bernard OA, Busson-LeConiat M, Ballerini P, Mauchauffe M, Della Valle V, Monni R et al. A new recurrent and specific cryptic translocation, t(5;14)(q35;q32), is associated with expression of the Hox11L2 gene in T acute lymphoblastic leukemia. Leukemia 2001; 15: 1495–1504.
Pallisgaard N, Hokland P, Riishoj DC, Pedersen B, Jorgensen P . Multiplex reverse transcription-polymerase chain reaction for simultaneous screening of 29 translocations and chromosomal aberrations in acute leukemia. Blood 1998; 92: 574–588.
Pui CH, Behm FG, Singh B, Schell MJ, Williams DL, Rivera GK et al. Heterogeneity of presenting features and their relation to treatment outcome in 120 children with T-cell acute lymphoblastic leukemia. Blood 1990; 75: 174–179.
Asnafi V, Beldjord K, Boulanger E, Comba B, Le Tutour P, Estienne MH et al. Analysis of TCR, pT alpha, and RAG-1 in T-acute lymphoblastic leukemias improves understanding of early human T-lymphoid lineage commitment. Blood 2003; 101: 2693–2703.
Bergeron J, Clappier E, Cauwelier B, Dastugue N, Millien C, Delabesse E et al. HOXA cluster deregulation in T-ALL associated with both a TCRD-HOXA and a CALM-AF10 chromosomal translocation. Leukemia 2006; 20: 1184–1187.
Weng AP, Ferrando AA, Lee W, Morris JPt, Silverman LB, Sanchez-Irizarry C et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 2004; 306: 269–271.
Shen WF, Detmer K, Simonitch-Eason TA, Lawrence HJ, Largman C . Alternative splicing of the HOX 2.2 homeobox gene in human hematopoietic cells and murine embryonic and adult tissues. Nucleic Acids Res 1991; 19: 539–545.
Lowney P, Corral J, Detmer K, LeBeau MM, Deaven L, Lawrence HJ et al. A human Hox 1 homeobox gene exhibits myeloid-specific expression of alternative transcripts in human hematopoietic cells. Nucleic Acids Res 1991; 19: 3443–3449.
Lawrence HJ, Sauvageau G, Ahmadi N, Lopez AR, LeBeau MM, Link M et al. Stage- and lineage-specific expression of the HOXA10 homeobox gene in normal and leukemic hematopoietic cells. Exp Hematol 1995; 23: 1160–1166.
Zappavigna V, Sartori D, Mavilio F . Specificity of HOX protein function depends on DNA–protein and protein–protein interactions, both mediated by the homeo domain. Genes Dev 1994; 8: 732–744.
Gisselsson D, Hoglund M, Mertens F, Johansson B, Dal Cin P, Van den Berghe H et al. The structure and dynamics of ring chromosomes in human neoplastic and non-neoplastic cells. Hum Genet 1999; 104: 315–325.
Gisselsson D, Pettersson L, Höglund M, Heidenblad M, Gorunova L, Wiegant J et al. Chromosomal breakage–fusion–bridge events cause genetic intratumor heterogeneity. Proc Natl Acad Sci USA 2000; 97: 5357–5362.
Nishio J, Iwasaki H, Ohjimi Y, Ishiguro M, Isayama T, Naito M et al. Supernumerary ring chromosomes in dermatofibrosarcoma protuberans may contain sequences from 8q11.2–qter and 17q21–qter: a combined cytogenetic and comparative genomic hybridization study. Cancer Genet Cytogenet 2001; 129: 102–106.
Nishio J, Iwasaki H, Ishiguro M, Ohjimi Y, Yo S, Isayama T et al. Supernumerary ring chromosome in a Bednar tumor (pigmented dermatofibrosarcoma protuberans) is composed of interspersed sequences from chromosomes 17 and 22: a fluorescence in situ hybridization and comparative genomic hybridization analysis. Genes Chromosomes Cancer 2001; 30: 305–309.
Streubel B, Valent P, Jager U, Edelhauser M, Wandt H, Wagner T et al. Amplification of the MLL gene on double minutes, a homogeneously staining region, and ring chromosomes in five patients with acute myeloid leukemia or myelodysplastic syndrome. Genes Chromosomes Cancer 2000; 27: 380–386.
Andreasson P, Johansson B, Billstrom R, Garwicz S, Mitelman F, Höglund M . Fluorescence in situ hybridization analyses of hematologic malignancies reveal frequent cytogenetically unrecognized 12p rearrangements. Leukemia 1998; 12: 390–400.
Streubel B, Valent P, Lechner K, Fonatsch C . Amplification of the AML1(CBFA2) gene on ring chromosomes in a patient with acute myeloid leukemia and a constitutional ring chromosome 21. Cancer Genet Cytogenet 2001; 124: 42–46.
Ruiz JC, Choi KH, von Hoff DD, Roninson IB, Wahl GM . Autonomously replicating episomes contain mdr1 genes in a multidrug-resistant human cell line. Mol Cell Biol 1989; 9: 109–115.
Ciullo M, Debily MA, Rozier L, Autiero M, Billault A, Mayau V et al. Initiation of the breakage–fusion–bridge mechanism through common fragile site activation in human breast cancer cells: the model of PIP gene duplication from a break at FRA7I. Hum Mol Genet 2002; 11: 2887–2894.
Barr FG, Nauta LE, Davis RJ, Schafer BW, Nycum LM, Biegel JA . In vivo amplification of the PAX3-FKHR and PAX7-FKHR fusion genes in alveolar rhabdomyosarcoma. Hum Mol Genet 1996; 5: 15–21.
Rodley P, McDonald M, Price B, Fright R, Morris C . Comparative genomic hybridization reveals previously undescribed amplifications and deletions in the chronic myeloid leukemia-derived K-562 cell line. Genes Chromosomes Cancer 1997; 19: 36–42.
Marculescu R, Le T, Simon P, Jaeger U, Nadel B . V(D)J-mediated translocations in lymphoid neoplasms: a functional assessment of genomic instability by cryptic sites. J Exp Med 2002; 195: 85–98.
Raghavan SC, Kirsch IR, Lieber MR . Analysis of the V(D)J recombination efficiency at lymphoid chromosomal translocation breakpoints. J Biol Chem 2001; 276: 29126–29133.
This study was supported by the Fonds voor Wetenschappelijk Onderzoek (FWO-) Vlaanderen, Grants G.0106.05 and GOA, grant no.12051203. This paper presents research results of the Belgian program of Interuniversity Poles of Attraction initiated by the Belgian State, Prime Minister's Office, Science Policy Programming. The scientific responsibility is assumed by the authors. BC is supported by the Belgian program of Interuniversity Poles of Attraction. NVR is a Postdoctoral fellow and BV is a Senior Clinical Investigator funded by the FWO-Vlaanderen. We are thankful to Betty Emanuel and Nurten Yigit for excellent technical assistance.
About this article
Cite this article
Cauwelier, B., Cavé, H., Gervais, C. et al. Clinical, cytogenetic and molecular characteristics of 14 T-ALL patients carrying the TCRβ-HOXA rearrangement: a study of the Groupe Francophone de Cytogénétique Hématologique. Leukemia 21, 121–128 (2007). https://doi.org/10.1038/sj.leu.2404410
- TCRβ-HOXA rearrangement
Advances in Biological Regulation (2019)
Updates in the Pathology of Precursor Lymphoid Neoplasms in the Revised Fourth Edition of the WHO Classification of Tumors of Hematopoietic and Lymphoid Tissues
Current Hematologic Malignancy Reports (2018)
Cell Death & Disease (2018)
Experimental Hematology (2013)
The immunophenotype of T-lymphoblastic lymphoma in children and adolescents: a Children's Oncology Group report
British Journal of Haematology (2012)