ETV6 Is the Target of Chromosome 12P Deletions in T(12;21) Childhood Acute Lymphocytic Leukemia

Home , Chromosome 12, ETV6

ETV6 is the target of chromosome 12p deletions in t(12;21) childhood acute lymphocytic leukemia H Cave´ 1, V Cacheux2, S Raynaud3, G Brunie1, M Bakkus4, P Cochaux5, C Preudhomme6, JL Laı¨7, E Vilmer8 and B Grandchamp1,9

1Laboratoire de Biochimie Geńe´tique, 2Laboratoire de Biologie du De´veloppement, 8Service d’He´matologie, Hoˆpital Robert Debre´, Paris; 3Laboratoire de Geńe´tique Molećulaire des Cancers Humains, CNRS URA 1462, Nice, France; 4Faculte´ de Me´decine VUB, Bruxelles; 5Universite´ libre de Bruxelles, De´partement de Geńe´tique Me´dicale, Hopital Erasme, Bruxelles, Belgium; 6Institut de recherche sur le cancer, Lille; 7Laboratoire de cytogeńe´tique, CHRU, Lille; and 9INSERM U409, Faculte´ de Me´decine Xavier Bichat, Paris, France

The presence of ETV6 deletions was investigated in 215 chil- of t(5;12)10 and was subsequently found to be fused to a dren with acute lymphoblastic leukemia (ALL) using the loss number of different partners as a result of various leukemia- of heterozygosity (LOH) approach. We used four intragenic or 11–13 juxtagenic microsatellite markers to detect allelic deletions. In associated translocations. In pediatric B lineage ALL, this series of unselected patients, LOH of ETV6 markers was t(12;21)(p13;q22), which leads to the formation of a ETV6- found in 23% of cases (6% of T-ALL and 26% of B lineage ALL) AML1 fusion gene on the der(21) chromosome,14,15 is found confirming that chromosome 12p12-13 deletions represent a with a high frequency (16–28%).16–18 Interestingly, we have major genetic alteration in childhood ALL, frequently missed by previously shown that t(12;21) and deletions of the non-trans- cytogenetic analysis. The presence of a t(12;21)(p13;q22) was located chromosome 12 were frequently associated in these studied by RT-PCR and/or FISH in a total of 134 patients (125 B lineage ALL, nine T-ALL) including 42 cases with LOH. Thirty- children and we found an intragenic deletion of ETV6 in one four out of 44 patients (77%) for whom a t(12;21) was observed patient with a t(12;21) suggesting that ETV6 was the target of displayed LOH of the ETV6 markers. When associated with a the deletions. In our initial series, 14 out of 16 patients with t(12;21), ETV6 is very likely to be the target of deletions as indi- t(12;21) had lost the non-translocated allele of ETV6.6 These cated by the detection of intragenic deletions in three patients. results contrast with those of other studies using Southern blot Although deletion of ETV6 and t(12;21) were associated in most analysis that reported a much lower frequency of ETV6 patients, in eight cases (six B lineage and two T-ALL) LOH was 16,17 detected at the ETV6 locus without ETV6-AML1 hybrid RNA. deletions in patients with a t(12;21). Besides, while FISH studies conducted in five of these eight patients con- t(12;21) is specifically found in children with B lineage ALL, firmed the absence of translocation involving ETV6. In such 12p deletions have also been detected in patients with other patients, the other allele of ETV6 could be disrupted by either types of malignancies19 including T-ALL,20 and it is unclear a small deletion, a point mutation, or an epigenetic modification whether or not ETV6 is the target of 12p deletions in all cases and it will be of interest to study the structure and expression or in a specific subset of patients. Indeed, a recent study of of the remaining allele of ETV6 in these cases. Alternatively, a tumor suppressor gene located close to ETV6 and CDKN1B LOH for 12p markers in childhood ALL suggested that ETV6 could be the target of deletions. might not be the target of all 12p deletions, since a second Keywords: ETV6; ETV6-AML1; t(12;21); loss of heterozygosity; region of deletion that did not include ETV6 was found near deletion; acute lymphoblastic leukemia CDKN1B.4 The aims of the present study were to establish the frequency of ETV6 deletions in a large series of children with B and T lineage ALL, to determine the occurence of small Introduction deletions involving specifically ETV6 that may incriminate this gene as the target gene of the deletions, and to correlate the Abnormalities of the short arm of chromosome 12 are quite presence of these deletions with the presence of t(12;21). We common in various hematological malignancies including used four intragenic or juxtagenic microsatellite markers to acute myeloid leukemia, myelodysplastic syndrome, myelo- detect allelic deletions of ETV6, and the t(12;21) translocation proliferative disease and acute lymphoblastic leukemia (ALL).1 was sought by FISH and/or RT-PCR of the ETV6-AML1 Chromosome 12p alterations consist of both deletions and chimeric transcript. translocations. Recent studies using microsatellite markers have shown that loss of heterozygosity (LOH) in the chromosomal region 12p12-13 is found in 20–40% of cases in child- Patients and methods hood ALL.2–4 This suggested that a tumor suppressor gene (TSG) might be located in this region whose inactivation Patients would participate in leukemogenesis.5 LOH mapping and FISH analysis indicated that, in the majority of patients, the We studied 215 children (aged 3 months to 15 years) with deletion contained two genes, the CDKN1B gene (previously ALL (183 B lineage ALL, 32 T-ALL). Eight of these children named KIP1) and the ETS-variant gene 6 (ETV6, previously were studied both at diagnosis and at marrow relapse. Chil- named TEL).3,4,6–9 CDKN1B encodes the cyclin-dependent dren were consecutively diagnosed and unselected, except kinase inhibitor p27 that is a regulator of the cell cycle and that inclusion required the presence of at least 70% of blasts therefore a candidate TSG, but several studies concluded that in marrow at the time of diagnosis or relapse (13 children the remaining allele of this gene is not mutated in patients have been excluded because of a low percentage of blasts in with one deleted allele.3,4 ETV6 was cloned at the breakpoint marrow). ALL were classified on the basis of lymphoid mor- phologic appearance and immunophenotype as assessed by flow cytometry. ALL were defined as belonging to the T lin- Correspondence: H Cave´, Laboratoire de Biochimie Geńe´tique, Hoˆpi- eage based on the presence of at least two of the T cell anti- tal Robert Debre´, 48 boulevard Se´rurier, 75019 Paris, France gens CD2, CD5 and CD7. The diagnosis of B precursor ALL Received 20 March 1997; accepted 17 June 1997 was based on the expression of B cell-associated antigens ETV6 deletions in childhood B lineage ALL H Cave´ et al 1460 (CD19+, CD22±, CD10±) and the lack of surface membrane lows: 10 s denaturation at 94°C, 10 s annealing extension at immunoglobulins (Ig). Informed consent was obtained from 68°C. Amplification was controlled by electrophoresis in 2% the patients, their parents, or both, as appropriate. agarose and ethidium bromide staining. ETV6-AML1 and AML1-ETV6 PCR gave rise to a 103 bp and a 199 bp fragment, respectively. The quality of the cDNAs was checked by ampli- Preparation of mononuclear cell lysates fying a 155 bp sequence of the porphobilinogene deaminase cDNA using primers P-PBGD (5′-AGCGGAGCCATG- Bone marrow (BM) was collected on EDTA before induction TCTGGTAACGGCAATG-3′)and N-PBGD (5′-AGGCCAGGG- therapy and during clinical remission between 6 and 24 TACGAGGCTTTCAATGTTG-3′) under the same PCR con- months of therapy according to the EORTC 58 881 follow- ditions as described above but with an annealing extension up protocol. Mononuclear cells were separated by gradient temperature of 60°C. centrifugation over Ficoll (Lymphoprep; Pharmacia, Uppsala, Specificity of the amplification was tested by Southern blot Sweden), washed in 0.9% NaCl, counted, and aliquots of cells hybridization to radio-labeled junctional probes for ETV6- were stored at −80°C. Erythrocytes were lysed by resus- AML1 (5′-AGAATAGCAGAATGCATACT-3′).15 pending the cells in a solution containing 10 mm Tris-HCl (pH 7.5), 1% Triton-100 (Sigma, St Louis, MO, USA), 5 mm

MgCl2 and 0.32 mm saccharose. After centrifugation, mono- Results nuclear cells were lysed in a buffer containing 0.45% Tween 20 (Difco, Detroit, MI, USA), 1.45% Nonidet P40 (Sigma), Frequency of monoallelic ETV6 deletion in childhood

2.5 mm MgCl2 and 10 mm Tris-HCl (pH 7.5) and incubated ALL for 2 h at 60°C in the presence of 1 mg/ml proteinase K. Vol- umes were adjusted to obtain 6 ng of DNA per ␮l of lysate. The presence of allelic deletions of ETV6 was investigated in After a 10-min incubation at 95°C, the lysates were stored at 215 children with ALL using the LOH approach. Partial results −20°C until analysis. concerning 66 of these children (including 16 with LOH) have been previously reported.6 The polymorphic markers used for this screening were D12S1697, D12S89, CA159A7 and Search for loss of heterozygosity D12S98. Markers D12S89 and CA159A7 are located within the ETV6 gene. D12S89 has been previously shown to map Seven simple tandem repeat (STR) markers were used that immediately upstream of exon 2 while CA159A7 maps in the span the chromosome bands 12p12-13. These markers are 5′ region of exon 1B.21 D12S1697 is located 7 kb upstream ordered from telomere to centromere: D12S1697, D12S89, of exon 1A of ETV6, and D12S98 is about 50 kb downstream CA159A7, D12S98, D12S1581, D12S1580 and D12S320.3,21 of exon 8.21 Loss of heterozygosity was evidenced by comparison of Percentages of LOH observed for each marker are reported allelic patterns of amplified DNA on the diagnosis sample in Table 1. Among the 215 patients studied, 210 (98%) were (blasts) and on a remission sample (normal cells) from each informative for at least one intragenic marker of ETV6. patient. Microsatellite typing was achieved using PCR and a The rate of LOH was similar with all four markers, and a fluorescent automated laser DNA sequencer, as previously percentage of about 18% of LOH among informative cases described.3 Primer sequences and percentage of heterozygos- is detected when isolated markers are considered (Table 1). ity were obtained from the Genome Data Base (Welch Medi- Compilation of the results obtained with the four markers cal Library, Johns Hopkins University, Baltimore, MD, USA). allowed the detection of LOH for at least one marker in 50 out of 215 patients (23%). Although deletional events were detected in ALL from both lineages, the observed frequency FISH was significantly higher in B lineage ALL (26%) than in T-ALL (6%) (␹2 test significant at P Ͻ 0.01) (Table 1). In eight patients Probes used for in situ hybridization consisted of YAC958b8, (two patients with 12p alterations and six without 12p Cos50F4 and whole chromosome painting probes for chromo- alteration), results obtained at diagnosis and at marrow relapse some 12 (Coatasome, Oncor; Dako, Trappes, France). These were compared and no difference was observed. probes and FISH conditions were as previously decribed.6,22 The presence of a t(12;21)(p13;q22) was studied in a total of 134 patients (125 B lineage ALL, nine T-ALL) including 42 cases with LOH. The study was conducted by RT-PCR alone RT-PCR analysis (n = 68), FISH alone (n = 27) or both techniques (n = 39), depending on the material available. RNA was extracted from blasts and RT was carried out with 1 ␮g of RNA using random priming as previously described.23 One tenth of the reaction was submitted to PCR amplification Deletion of ETV6 in patients with a t(12;21) with various combinations of primers: ETV6-AML1 was amplified using P-TEL (5′-GAAGAGCACGCCATGCCCATTGGGA- The t(12;21) was never detected in T-ALL but was detected in GAATA-3′ nucleotides 1000–1029)10 and N-AML1 (5′- 44 patients with B lineage ALL (Table 2). In one patient, RT- GCGGCGTGAAGCGGCGGCTCGTGCT-3′: complementary PCR yielded an ETV6-AML1 product shorter than the usual to nucleotides 700–724).24 transcript that did not hybridize to the junctional probe. PCR were performed in 25 ␮l reaction mixtures containing Sequencing of this fragment showed that it consists of a ETV6-

10 mm Tris-HCl (pH 9.0), 50 mm KCl, 2.5 mm MgCl2, 200 ␮m AML1 transcript lacking the second AML1 exon (39 bp), poss- of each dNTP, 10 pmoles of each primer, and 0.25 U of Taq ibly because of an alternative translocation breakpoint on polymerase (Appligene, Strasbourg, France). After a 5-min AML1. Thirty-four patients (77%) for whom a t(12;21) was denaturation step at 94°C, 40 cycles were performed as fol- detected displayed LOH at the ETV6 locus. In 10 patients, no ETV6 deletions in childhood B lineage ALL H Cave´ et al 1461 Table 1 Loss of heterozygosity observed at the ETV6 locus in childhood ALL

No. of No. of Observed No. of No. of patients informative heterozygosity cases with cases with studied cases % LOH LOH (single (compilation) marker) (%) (%)

Total ALL (n = 215) D12S1697 139 122 88 22 (18) 50 (23) D12S89 212 180 85 35 (19) CA159A7 188 150 80 28 (18.5) D12S98 146 90 62 17 (19) B lineage ALL (n = 183) D12S1697 132 116 21 (18) 48 (26) D12S89 182 155 34 (22) CA159A7 159 127 26 (20.5) D12S98 139 85 17 (20) T-ALL (n = 32) D12S1697 7 6 1 2 (6) D12S89 30 25 1 (4) CA159A7 29 23 2 (9) D12S98 7 5 0

Table 2 Correlation between the occurrence of a t(12;21) and the (two cases), CA159A7 (one case), D12S98 (one case). Three deletion of the ETV6 gene in 134 cases of childhood ALL patients had intragenic deletions and although all deletions involved ETV6, no region of overlap was shared by all t(12;21) patients (Figure 1).

Present Absent

Present 34 8a Deletion of ETV6 without demonstrable t(12;21) (23%) (8%) LOH of ETV6 No evidence of t(12;21) was found in eight out of 42 cases markers with LOH for ETV6 markers using RT-PCR and/or FISH b Absent 10 82 (Table 2). Six of these patients had B lineage ALL and two (8%) (61%) patients had T-ALL. Chromosome 12 painting was performed a in four cases (three B lineage ALL and one T-ALL) and was Eight patients had B lineage ALL and two patients had T-ALL. normal in three cases while in one case an add(12p11) was bThree of these children where studied by FISH and presented ETV6 deletion in a sub-clone of cells with t(12;21). detected suggesting that the allelic loss of ETV6 resulted from an unbalanced translocation of chromosome 12 (Figure 2). In order to specify the extent of the 12p deletion in these deletion of ETV6 was detected when studied with polymor- patients, additional markers spanning the 12p12-13 chromo- phic markers (Table 2). somal bands were tested (from telomere to centromere: Seven patients with a t(12;21) had a deletion with at least D12S77, D12S1581,D12S1580 and D12S320). Deleted areas one border lying within the ETV6 gene (Figure 1). Four of were large in most patients and included all or most of the these patients displayed LOH for a single marker, D12S89 markers tested, which encompass CDKN1B (Figure 2, patients 8–10 and 12–15).

Discussion

The frequency of 12p deletion evidenced by cytogenetics is 5% in childhood ALL.25,26 In our series of unselected childhood ALL, LOH of ETV6 markers was found in 23% of cases (6% of T-ALL and 26% of B lineage ALL) conﬁrming that this represents a major genetic alteration in childhood ALL, frequently missed by cytogenetic analysis. The frequence of LOH involving marker D12S89 is 19% among 180 informative cases in our series which is intermediate between the results of Stegmaier et al3 and Takeuchi et al4 who reported LOH frequencies of 14% (out of 63 patients) and 26% (out of 82 patients), respectively. The difference could result from a dif- Figure 1 Schematic representation of LOH in patients with t(12;21) who display a small deletion or a breakpoint located within ference in proportion of T-ALL or in the age of patients ETV6.( ) Heterozygote; ( ) non-informative; ( ) loss of hetero- between series. For example, Stegmaier’s series included zygosity. patients aged 15 to 18.2 The prevalence of LOH in B lineage ETV6 deletions in childhood B lineage ALL H Cave´ et al 1462

Figure 2 Schematic representation of LOH and cytogenetic data in patients without a demonstrated t(12;21). ( ) Heterozygote; ( ) non- informative; ( ) loss of heterozygosity. ND, not done.

ALL was much higher than in T-ALL. This is related to the deletions of ETV6 have been observed (Figure 1, patients 1, frequent association of ETV6 deletions with t(12;21). 2, 3 and 7). This makes it very unlikely that the target is a The presence of a t(12;21)(p13;q22) was detected in 44 of gene present in an intron of ETV6. It is worth noticing that 125 patients with B lineage ALL. No t(12;21) was detected in FISH analysis using only one cosmid (eg cos50F4) would have T-ALL, in agreement with previous reports.16 Thirty-four missed 8% of ETV6 deletions. Interestingly, Kobayashi et al28 patients (77%) for whom a t(12;21) was observed displayed recently reported a case of ALL with a t(12;21) associated with LOH of the ETV6 markers. This underscores the frequency of a t(3;12) disrupting the other ETV6 allele. In this case, the deletions associated with t(12;21) because three additional t(3;12) may inactivate the second ETV6 allele and be func- patients with t(12;21) had a deletion only in a subclone of tionally similar to deletions commonly observed in most of blasts (data not shown), a finding that has been previously the patients with t(12;21). decribed,6,27 and because a small deletion or another inactiv- Chromosomal translocations generally participate in the ating event cannot be excluded. Some 12p deletions have tumoral phenotype via the production of an oncogenic fusion been described that affect the der(12)t(12;21) chromosome protein or the overproduction of an oncogene.29 Fusions pro- and are likely to result from unbalanced translocations.28 In teins involving ETV6 that have been identified so far are our series, deletions always concerned the non-translocated believed to be oncogenic. The ETV6-PDGFR␤ fusion tran- chromosome 12 homologue. This was either assessed directly script resulting from t(5;12) was tested for transforming activity by FISH or indirectly from the presence of a ETV6-AML1 tran- in cultured mammalian cell lines. ETV6-PDGFR␤ confers fac- script in patients with LOH for ETV6 markers encompassing tor-independent growth to the interleukin 3-dependent cell the translocation breakpoint. line Ba/F3.30 Identical results were obtained for ETV6-ABL.30 The frequency of the association translocation/deletion Although there is no experimental demonstration of the reported here contrasts strikingly with reports from other oncogenic properties of ETV6-AML1, several lines of evidence investigators. Evidence for deletion of the non-translocated argue for such a role. The translocation breakpoint is highly allele of ETV6 was found in only seven out of the 35 patients conserved, ETV6-AML1 transcript is always present, in con- with t(12;21) (20%) studied by Shurtleff et al17 and in two out trast with the reciprocal transcript which is only found in of the eight patients (25%) studied by Romana et al.16 Techni- about half of the cases,6 and the ETV6-AML1 transcript leads cal differences may partly account for this apparent discrep- to the production of a chimeric protein detectable by Western ancy. For example, the presence of residual normal cells in blot (data not shown). blastic samples makes it very difficult to assess the deletions In addition to the production of the ETV6-AML1 fusion pro- using Southern blot. The use of an automated fluorescent tein, t(12;21) inactivates one allele of ETV6 and this may also sequencer to analyze PCR products permits an accurate deter- be significant for the oncogenic process. The function of ETV6 mination of allelic ratios and enables a reliable detection of is still unknown. The member of the ETS family with the most allelic imbalance even in the presence of some normal cells. homology to ETV6 is the drosophila gene YAN/POK, a tran- In addition, about 25% of ETV6 deletion involves only a part scriptional repressor inactivated by the ras1/MAPK signal. of this gene. Small deletions might not have been covered by YAN/POK inactivation results in overproliferation of cells from the probes these authors used for Southern blotting. different tissues.31,32 ETV6 could act as a TSG, with t(12;21) When associated with a t(12;21), ETV6 is very likely to be disrupting one copy of ETV6 and the deletion being the the target of deletions as indicated by the detection of intra- second inactivating event according to the Knudson hypoth- genic deletions in a number of our patients and the constant esis.33 The generation by t(12;21) of a fusion protein likely to involvement of ETV6 in deletions. Moreover, non-overlapping be oncogenic is not in contradiction with the assumption that ETV6 deletions in childhood B lineage ALL H Cave´ et al 1463 ETV6 is a TSG. The two-step alteration of ETV6 observed in 2 Stegmaier K, Pendse S, Barker GF, Bray-Ward P, Ward DC, childhood ALL may somehow be analogous to what has been Montgomery KT, Krauter KS, Reynolds C, Sklar J, Donnelly M, reported for ABL during the progression of chronic myelogen- Bohlander SK, Rowley JD, Sallan SE, Gilliland DG, Golub TR. Fre- quent loss of heterozygosity at the TEL gene locus in acute lym- ous leukemia (CML). The first and major event is the formation phoblastic leukemia of childhood. Blood 1995; 86: 38–44. of the BCR-ABL gene whose oncogenic properties have been 3 Cave´ H, Ge´rard B, Martin E, Guidal C, Devaux I, Weissenbach J, extensively documented. In some patients, the blast crisis is Elion J, Vilmer E, Grandchamp B. Loss of heterozygosity in the accompanied by the loss of the non-translocated allele of ABL chromosomal region 12p12-13 is very common in childhood that results in the complete absence of a normal ABL pro- acute lymphoblastic leukemia and permits the precise localization tein.34 In view of recent data that demonstrate that c-ABL is of a tumor-suppressor gene distinct from p27KIP1. Blood 1995; 34,35 86: 3869–3875. involved in the growth arrest response to DNA damage, 4 Takeuchi S, Bartram CR, Miller CW, Reiter A, Seriu T, Zimmerann the inactivation of the second ABL allele may be considered M, Schrappe M, Mori N, Slater J, MiyoIshi I, Koeffler HP. Acute a loss of function of a TSG that contributes to the disease pro- lymphoblastic leukemia of childhood. Identification of two distinct gression. Whether or not such a model would be relevant for regions of deletion on the short arm of chromosome 12 in the ETV6 in ALL requires elucidation of the normal ETV6 func- region of TEL and KIP1. Blood 1996; 87: 3368–3374. tions as well as direct experimental support for the oncogenic- 5 Lasko D, Cavenee W, Nordenskjo¨ld M. Loss of constitutional het- ity of the ETV6-AML1 protein. Alternatively, the demon- erozygosity in human cancer. Ann Rev Biochem 1991; 25: 281– 30 314. stration by Golub et al that ETV6 can homodimerize or 6 Raynaud S, Cave´ H, Baens M, Bastard C, Cacheux V, Grosgeorge heterodimerize via its HLH domain to ETV6-ABL, a fusion pro- J, Guidal-Giroux C, Guo C, Vilmer E, Marynen P, Grandchamp B. tein resulting from the (9;12) translocation, raises the interest- The 12;21 translocation involving TEL and deletion of the other ing possibility that the protein encoded by the normal allele TEL allele: two frequently associated alterations found in child- of ETV6 could interfere with the fusion protein and buffer its hood acute lymphoblastic leukemia. Blood 1996; 87: 2891–2899. oncogenic activity. Accordingly, disruption of ETV6 could 7 Wlodarska I, Marynen P, La Starza R, Mecucci C, Van Den Berghe H. The ETV6, CDKN1B and D12S178 loci are involved in a seg- represent a second step in tumor progression, confering an ment commonly deleted in various 12p aberrations in different additional growth advantage on a cell bearing a t(12;21). hematological malignancies. Cytogenet Cell Genet 1996; 72: Although deletion of ETV6 and the presence of t(12;21) 229–235. were associated in most of our patients, eight patients includ- 8 Sato Y, Suto Y, Pietenpol J, Golub TR, Gilliland DG, Davis EM, ing two cases with T-ALL displayed LOH at the ETV6 locus Le Beau MM, Roberts JM, Vogelstein B, Rowley JD, Bohlander SK. without evidence for t(12;21). In cases without t(12;21), TEL and KIP1 define the smallest region of deletions on 12p13 in whether or not ETV6 is the target of the deletions remains hematopoietic malignancies. Blood 1995; 86: 1525–1533. 9Ho¨glund M, Johansson B, Pedersen-Bjergaard J, Marynen P, Mitel- questionable. The other allele of ETV6 could be inactivated man F. Molecular characterization of 12p abnormalities in hema- in these patients by either a small deletion, a point mutation, tologic malignancies: deletion of KIP1, rearrangement of TEL, and or any other inactivating event and it will be of interest to amplification of CCND2. Blood 1996; 87: 324–330. study the structure and expression of the remaining allele of 10 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF ETV6. Alternatively, a gene different from ETV6 may be the receptor beta to a novel ets-like gene, TEL, in chronic myelomono- target of deletions in these patients. Takeushi et al4 recently cytic leukemia with t(5;12) chromosomal translocation. Cell 1994; 77: 307–316. reported the existence of a second deletional area in12p12- 11 Buijs A, Sherr S, van Baal S, van Bezouw S, van der Plas D, van 13 very close to but distinct from the ETV6 gene. When Kessel GA, Riegman P, Lekanne Deprez R, Zwarthoff E, Hage- determined according to the actual order of the polymorphic meijer A, Grosveld G. Translocation (12;22) (p13;q11) in myelo- markers, this second region is flanked by marker D12S358 on proliferative disorders results in fusion of the ETS-like TEL gene on the telomeric side and D12S320 on the centromeric side, an 12p13 to the MN1 gene on 22q11. Oncogene 1995; 10: 1511– area deleted in all our patients without t(12;21). Thus, our 1519. present data are compatible with the involvement of a gene 12 Papadopoulos P, Ridge SA, Boucher CA, Stocking C, Wiedemann 4 LM. The novel activation of ABL by fusion to an ETS-related gene, located within the second region defined by Takeushi et al. TEL. Cancer Res 1995; 55: 34–38. 13 Wlodarska I, Mecucci C, Marynen P, Guo C, Franckx D, La Starza R, Aventin A, Bosly A, Martelli MF, Cassiman JJ, Van den Berghe Acknowledgements H. TEL gene is involved in myelodysplastic syndromes with either the typical t(5;12)(q33;p13) translocation or its variant t(10;12)(q24;p13). Blood 1995; 85: 2848–2852. This work was supported in part by grants from the Ligue 14 Golub TR, Barker GF, Bohlander SK, Hiebert SW, Ward DC, Bray- Nationale Contre le Cancer, the Association pour la Rec- Ward P, Morgan E, Raimondi SC, Rowley JD, Gilliland DG. Fusion herche contre le Cancer and the De´le´gation a` la Recherche of the TEL gene on 12p13 to the AML1 gene on 21q22 in acute Clinique de l’Assistance Publique-Hoˆpitaux de Paris (CRC lymphoblastic leukemia. Proc Natl Acad Sci USA 1995; 92: No. 950262). We thank members of the EORTC-CLCG for 4917–4921. providing us with samples from their patients: Drs M Fournier 15 Romana SP, Mauchauffe M, Le Coniat M, Chumakov I, Le Paslier D, Berger R, Bernard OA. The t(12;21) of acute lymphoblastic and B Nelken (Lille), N Dastugue and A Robert (Toulouse), C leukemia results in a tel-AML1 gene fusion. Blood 1995; 85: Bastard (Bois-Guillaume), E Lebrun and P Boutard (Caen), R 3662–3670. Garand, D Talmant, F Bugnon and F Mechinaux (Nantes), E 16 Romana SP, Poirel H, Leconiat M, Flexor MA, Mauchauffe M, Jon- Racadot and E Plouvier (Besançon), A Ferster and J Otten veaux P, Macintyre EA, Berger R, Bernard OA. High frequency (Bruxelles), O Fenneteau, B Lescoeur and M Duval (Paris). of t(12;21) in childhood B-lineage acute lymphoblastic leukemia. Blood 1995; 86: 4263–4269. 17 Shurtleff SA, Buijs A, Behm FG, Rubnitz JE, Raimondi SC, Hancock ML, Chan GC, Pui CH, Grosveld G, Downing JR. TEL/AML1 fusion References resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and defines a subgroup of patients with an excel- 1 Johansson B, Mertens F, Mitelman F. Cytogenetic deletion map of lent prognosis. Leukemia 1995; 9: 1985–1989. hematological neoplasms: circumstantial evidence for tumor sup- 18 Cayuela JM, Baruchel A, Orange C, Madani A, Auclerc MF, Daniel pressor loci. Genes Chromosom Cancer 1993; 8: 205–218. MT, Schaison G, Sigaux F. TEL-AML1 fusion RNA as a new target ETV6 deletions in childhood B lineage ALL H Cave´ et al 1464 to detect minimal residual disease in pediatric B-cell precursor 27 Romana SP, Le Coniat M, Poirel H, Marynen P, Bernard O, Berger acute lymphoblastic leukemia. Blood 1996; 88: 302–308. R. Deletion of the short arm of chromosome 12 is a secondary 19 Mitelman F, Kaneko Y, Berger R. Report of the committee on chro- event in acute lymphoblastic leukemia with t(12;21). Leukemia mosome changes in neoplasia. Human Gene Mapping. Johns 1996; 10: 167–170. Hopkins University Press: Baltimore, 1993, pp 773–812. 28 Kobayashi H, Satake N, Maseki N, Sakashita A, Kaneko Y. The 20 Raimondi SC. Current status of cytogenetic research in childhood der(21)t(12;21) chromosome is always formed in a 12;21 translo- acute lymphoblastic leukemia. Blood 1993; 81: 2237–2251. cation associated with childhood acute lymphoblastic leukemia. 21 Baens M, Peeters P, Guo C, Aerssens J, Marynen P. Genomic Br J Haematol 1996; 94: 105–111. organization of Tel: the human ETS-variant gene 6. Genome Res 29 Rabbitts TH. Chromosomal translocations in human cancer. Nat- 1996; 6: 404–413. ure 1994; 372: 143–149. 22 Raynaud SD, Bekri S, Leroux D, Grosgeorge J, Klein B, Bastard C, 30 Golub TR, Barker GF, Stegmaier K, Gilliland DG. Involvement of Gaudray P, Simon MP. Expanded range of 11q13 breakpoints with the TEL gene in hematologic malignancy by diverse molecular differing patterns of cyclin D1 expression in B-cell malignan- genetic mechanisms. Curr Top Microbiol Immunol 1996; 211: cies.Genes Chromosom Cancer 1993; 8: 80–87. 279–288. 23 Malinge MC, Mahon FX, Delfau MH, Daheron L,Kitzis A, Guilhot 31 Rebay I, Rubin M. Yan functions as a general inhibitor of differen- F, Tanzer J, Grandchamp B. Quantitative determination of the tiation and is negatively regulated by activation of the Ras1/MAPK hybrid Bcr-Abl RNA in patients with chronic myelogenous leu- pathway. Cell 1995; 81: 857–866. kaemia under interferon therapy. Br J Haematol 1992; 82: 701– 32 O’Neill E, Rebay I, Tjian R, Rubin G. The activity of two ETS- 707. related transcription factors required for drosophila eye develop- 24 Miyoshi H, Shimizu K, Kozu T, Maseki N, Kaneko Y, Ohki M. t(8;21) breakpoints on chromosome 21 in acute myeloid leukemia ment are modulated by the Ras/MAPK pathway. Cell 1994; 78: are clustered within a limited region of a single gene, AML1. Proc 137–147. Natl Acad Sci USA 1991; 88: 10431–10434. 33 Knudson AG. Antioncogenes and human cancer. Proc Natl Acad 25 Raimondi SC, Williams DL, Callihan T, Peiper S, Riviera GK, Mur- Sci USA 1993; 90: 10914–10921. phy SB. Non-random involvement of the 12p12 breakpoint in 34 Sawyers C, McLaughlin J, Goga A, Havlik M, Witte O. The nuclear chromosomal abnormalities of childhood acute lymphoblastic tyrosine kinase c-Abl negatively regulates cell growth. Cell 1994; leukemia. Blood 1986; 68: 69–75. 77: 121–131. 26 Groupe Français de Cytogeńe´tique He´matologique. Collaborative 35 Yuan ZM, Huang Y, Whang Y, Sawyers C, Weichselbaum R, Sur- study of karyotypes in childhood acute lymphoblastic leukemias. ender K, Kufe D. Role for c-Abl tyrosine kinase in growth arrest Leukemia 1993; 7: 10–19. response to DNA damage. Nature 1996; 382: 272–274.