Letters to the Editor
J Bergeron1,5, E Clappier2, B Cauwelier3, N Dastugue4, C Millien1, E Delabesse1, K Beldjord1, F Speleman3, J Soulier2, E Macintyre1 and V Asnafi1 1 Laboratoire d’He´matologie and INSERM EMI0210, Hoˆpital Necker-Enfants Malades, Universite´ Paris-Descartes, AP-HP, Paris, France; 2 Hematology Laboratory and INSERM U728, Institut Universitaire d’He´matologie, Hoˆpital Saint-Louis and Paris 7 University, Paris, France; 3 Centre for Medical Genetics, Ghent University Hospital, Ghent, Belgium and 4 Laboratoire d’He´matologie, Hoˆpital Purpan, Toulouse, France. E-mail:
[email protected] 5 J Bergeron is a Research Fellow of The Terry Fox Foundation through an award from the Natianal Cancer Institute of Canada References 1 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. 2 Soulier J, Clappier E, Cayuela JM, Regnault A, Garcia-Peydro M, Dombret H et al. HOXA genes are included in genetic and biologic
1187 3
4 5
6
7
8
networks defining human acute T-cell leukemia (T-ALL). Blood 2005; 106: 274–286 [E-pub 2005 Mar 17]. 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. Leukemia 2005; 19: 358–366. 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. 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 [E-pub 2001 Dec 3]. 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 TCR{gamma}{delta} lineage. Blood 2003; 102: 1000–1006. Asnafi V, Beldjord K, Libura M, Villarese P, Millien C, Ballerini P et al. Age-related phenotypic and oncogenic differences in T-cell acute lymphoblastic leukemias may reflect thymic atrophy. Blood 2004; 104: 4173–4180 [E-pub 2004 Mar 30].
Identification of truncated RUNX1 and RUNX1-PRDM16 fusion transcripts in a case of t(1;21)(p36;q22)-positive therapy-related AML Leukemia (2006) 20, 1187–1189. doi:10.1038/sj.leu.2404210; published online 6 April 2006
Particular subtypes of acute myeloid leukemia (AML)s and myelodysplastic syndromes (MDS) are associated with specific recurrent chromosomal anomalies. Among these are translocations involving 21q22, which are commonly observed in both de novo and therapy-related (t-)MDS and AML. The t(8;21)(q22;q22), t(3;21)(q26;q22) and t(16;21)(q24;q22) are most frequently observed, but other variants have been described as well.1,2 The majority of these translocations result in a fusion of the 50 region of the RUNX1 (CBFA2, AML1) gene with the 30 region of either one of its partner genes.1 Here, we report a patient with therapy-related AML and a rare t(1;21)(p36;q22). 30 RACE experiments followed by sequencespecific RT-PCR resulted in the identification of the PRDM16 gene as a novel fusion partner of the RUNX1 gene in this patient. A 47-year-old woman was referred to our hospital with pancytopenia in 2002. In 1999, she had been diagnosed elsewhere with invasive breast cancer and positive axillary lymph nodes without distant metastases, for which she was treated with mastectomy, adjuvant radiotherapy, chemotherapy and tamoxifen. Hematologic examination showed white blood cells (WBC) of 1.1 109/l with 43% polymorphonuclear cells (PMN’s), hemoglobin level of 7.0 mmol/l and a platelet count of 42 109/l. Bone marrow biopsy showed hypoplastic bone marrow with severe myelodysplasia. Bone marrow karyotypes were obtained after short-term cultures, chromosomes were GTG-banded and the karyotype was described according to the International System for Human Cytogenetic Nomenclature (2005); a normal 46,XX[20] karyotype was observed. The patient was treated with an allogenic stem cell transplantation (male donor), which resulted in a complete hematologic response and complete donor chimerism. In July 2004, the patient again
developed a severe pancytopenia. Bone marrow examination showed a hypoplastic bone marrow with myelodysplasia and fluorescence in situ hybridization (FISH) demonstrated a recurrence of acceptor cells (5%). The patient received donor lymphocyte infusion and a stem cell boost from the donor, respectively. Although hematologic parameters improved rapidly, in April 2005 pancytopenia reappeared, and a diagnosis of AML with minimal differentiation was made. Chromosome analysis showed a 46,XX,t(1;21)(p36;q22)[20] karyotype (Figure 1, left panel). FISH analysis with the LSI AML1/ETO Dual Color, Dual Fusion translocation probe (Vysis, Downer’s Grove, IL, USA) revealed three RUNX1 gene signals on the normal 21, der(21) and der(1) chromosomes (Figure 1, right panel). As normal female karyotype was found at diagnosis, these chromosomes were evaluated. Owing to the lack of quality of the metaphases and lack of suitable material for FISH analysis, we cannot exclude the possibility that the cryptic translocation t(1;21)(p36;q22) was missed at diagnosis. In July 2005, bone marrow examination revealed an acute myelomonocytic leukemia and a 46,XX,t(1;21)(p36;q22),del(7)(q?34)[7]/47,idem, þ 13[3] karyotype. The patient died a few months later. An extensive literature search through Medline (http://www.ncbi.nlm.nih.gov/entrez/ query.fcgi) and the Mittelman Database of Chromosome Aberrations in Cancer (http://cgap.nci.nih.gov/Chromosomes/Mitelman) revealed four additional cases (Table 1). This turns the t(1;21)(p36;q22) into a rare but recurrent translocation. The t(1;21)(p36;q22) seems to be associated with a relatively short survival time (6–7 months). (Partial) loss of chromosome 7 was observed in case 2 and in the present case during relapse. Although only present in two cases, it should be noted that chromosome 7 abnormalities are the most commonly observed secondary changes in patients with other 21q22 translocations.1 Two cases were therapy-related and had treatment histories that included topoisomerase II targeting agents. This observation Leukemia
Letters to the Editor
1188 provides further evidence for the previously observed association between RUNX1 translocations and treatment with topoisomerase II inhibitors.2 Aiming at the identification of a presumed chromosome 1p36 fusion gene, we performed RUNX1-primed 30 Rapid Amplification of cDNA Ends (RACE) analysis. Total RNA was extracted from a bone marrow aspirate (April 2005) using a RNA-beebased protocol following the instructions of the manufacturer (Bioconnect, Huissen, The Netherlands). cDNA synthesis was performed in two parallel reactions using both random
8 8 der (1)
der(21) 21 1
der (1)
21 der(21)
Figure 1 Left panel: Partial GTG-banded karyotype of t(1;21)(p36;q22). Right panel: FISH analysis using the AML1/ETO dual-color, dual-fusion translocation probe. Three signals for RUNX1 (red) are observed; on the normal chromosome 21 and on the derivative chromosomes 1 and 21.
hexamers and universal adaptor primer 2 (UAP2)-tailed T(17), in combination with Superscript III reverse transcriptase (Invitrogen). The following 30 RACE-primers were subsequently used: GAGCCCAGGCAAGATGAG (first round PCR, RUNX1 exon 3-specific primer), CTACCGCAGCCATGAAGAAC (second round, RUNX1 exon 4-specific primer), both in combination with universal adaptor primer UAP2. Agarose gel electrophoresis revealed several distinct PCR products (data not shown), which were all excised from gel, cloned and sequenced. Two of the products were composed of RUNX1 sequences (comprising exons 4 (partially), 5 and 6) fused to sequences, which were found to be residing within intron 1 of the PRDM16 gene (Figure 2). The presumed cryptic, PRDM16related exon adds 59 residues to a carboxy-terminal truncated RUNX1 protein. A Pfam search at http://www.sanger.ac.uk/ Software/Pfam/ did not reveal any protein domains encoded by these added residues, and a blastp-search at http:// www.ncbi.nlm.nih.gov/BLAST/ failed to detect any significant homology to any known protein. From these findings we conclude that the intron 1-derived sequences are not likely to result in the addition of any functional protein domain(s), but rather provide a poly-adenylation function to a fraction of the 30 truncated RUNX1 transcripts, thereby stabilizing the truncated RUNX1 transcripts. After identification of PRDM16 sequences, a more sensitive and specific RT-PCR strategy was applied on a mixture of UAP2 and random-primed cDNA, using various combinations of RUNX1 and PRDM16-specific primers including CTACCG
Figure 2 Schematic representation of RUNX1 and PRDM16 (fusion) genes. Upper panel: normal genomic structures of PRDM16 (coding exons indicated in green, noncoding parts in blue), and RUNX1 (coding exons indicated in red, noncoding parts in blue). A cryptic exon, residing within intron 1 of PRDM16, which was identified by 30 RACE analysis, is indicated in green (speckled). Lower panel: structure of RUNX1-PRDM16 fusion transcripts. Please note that the cryptic PRDM16 exon (present in the fusion gene) is not present in the spliced fusion transcript detected by RT-PCR analysis. Positions of gene-specific primers used for 30 RACE and/or RT-PCR analysis are indicated by triangles. Exons are numbered on the basis of consensus gene sequences. Exon sizes are not to scale.
Table 1
Clinical, hematological and cytogenetic features of patients with t(1;21)(p36;q21)
Age
Sex
Diagnose
60 8 72
M F M
AML-M1 RAEBt MDS/AML-M2
72 47
F F
AML-M4 AML-M0
Survival (months)
7 6 6 42 7
Karyotype
RUNX1 split signal (FISH)
Position breakpoint
Previous therapy
46,XY,t(1;21)(p36;q22)[12] 46,XX,t(1;21)(p36;q22), 7,+mar 46,XY,t(1;21)(p36;q22)
Yes ND Yes
Intron 6 ND ND
46,XX,t(1;21)(p36;q22)[20] 46,XX,t(1;21)(p36;q22)[10]
Yes Yes
Intron 6 Intron 6
Chemotherapya None reported Exposure to nuclear explosions None reported Radiotherapy and chemotherapyb
Reference
2 3 4
5
Present case
Abbreviations: AML, acute myeloid leukemia; FISH, fluorescence in situ hybridization; MDS, myelodysplastic syndromes; ND, not determined; RAEBT, refractory anemia with excess blasts in transformation. a Lomustine, vincristine, cyclophosphamide, etoposide. b Fluorouracil, epirubicin, cyclophosphamide. Topoisomere II inhibitors are indicated by boldface type. Leukemia
Letters to the Editor
CAGCCATGAAGAAC (RUNX1 exon 4-specific forward primer) and CGTGTAGGACTTGTGGCAGA (PRDM16 exon 9-specific reversed primer). The results of this experiment confirmed (data not shown) the anticipated RUNX1-PRDM16 fusion transcripts composed of RUNX1 sequences upstream of exon 7, followed by PRDM16 sequences downstream of exon 1 (Figure 2). No fragments resulting from alternative splicing or exon skipping were found among the PCR fragments analyzed. In addition, no reciprocal PRDM16-RUNX1 chimeras were identified, which might indicate that the RUNX1-PRDM16 fusion gene is the most relevant one for inducing leukemia. During the preparation of this manuscript, others independently identified the PRDM16 gene as novel fusion partner of RUNX1 in an AML patient with a t(1;21)(p36;q22) (case 4, Table 1). The PRDM16 gene encodes a zinc finger protein containing two DNA binding domains and a PRDI-BF1 (positive regulatory domain I binding factor 1)/RIZ (retinoblastoma-interacting zinc finger protein) homologous (PR) domain at the N-terminus.6 The evolutionary conserved PR domain is highly homologous to the PR domain as encoded by MDS1/EVI1 gene. Furthermore, the DNA-binding consensus sequence of EVI1 overlaps with that of the PRDM16 protein, suggesting that PRDM16 has a similar DNA-binding capacity and transcriptional regulation activity as EVI1. Moreover, both genes are transcriptionaly activated by chromosome rearrangements involving the RPN1 gene at chromosome 3q21 in the t(1;3)(p36;q21) and inv(3)(q21q26)/ t(3;3)(q21;q26),6 respectively, and both genes can form fusion genes with RUNX1. RUNX1 is essential for hematopoiesis in the fetal liver and myeloid differentiation in adult bone marrow. The gene encodes the CBFa2 subunit of the core binding factor, a heterodimeric transcription factor complex composed of CBFa and CBFb.2 The amino-terminus contains the DNA-binding runt domain, whereas at the carboxy-terminus a transactivation domain is present. As a result of the RUNX1-targeting translocations, the DNA-binding domain becomes separated from the transactivation domain, and is fused to sequences originating from the partner gene. The breakpoints in at least three (case 1, case 4 and present case) of the t(1;21)(p36;q22) patients are localized in intron 6, which is also the affected intron in most of the t(3;21)(q26;q22) patients, while in patients with a t(8;21)(q22;q22) the breakpoint is frequently found within intron 5.2,7 The RUNX1-PRDM16 fusion transcript identified by RACEPCR contains the runt DNA binding domain, but lacks the transactivation domain of RUNX1, fused to sequences derived from intron 1 of PRDM16. The predicted RUNX1PRDM16 chimeric protein contains a limited number of PRDM16 residues fused to RUNX1 in a similar manner to the t(3;21) related RUNX1-EAP and the recently reported RUNX1PRDM16 fusion.5 It is possible that, similar to RUNX1-EAP,8 the truncated RUNX1 gene product may compete with wild-type RUNX1 and act as a dominant-negative inhibitor of the RUNX1 protein. RT-PCR confirmed the presence of a RUNX1-PRDM16 rearrangement. In addition, a second, similar as recently described,5 RUNX1-PRDM16 fusion transcript was identified which contains the almost entire PRDM16 coding sequence. As a result of this chimeric transcript the almost entire PRDM16 protein, which is normally absent in hematopoietic cells, will be expressed under the control of the active RUNX1 promoter. It is likely, that similar has been suggested for patients carrying a t(1;3)(p36;q21), that ectopic PRDM16 expression contri-
butes to the pathogenesis of MDS and AML.6 The RUNX1PRDM16 gene shares extensive homology with RUNX1-MDS1/ EVI1.6 Therefore, it is possible that the RUNX1-PRDM16 fusion protein plays a similar role in the leukemic process in a similar way as has previously been demonstrated for the RUNX1MDS1/EVI1.9 Although relatively rare, the t(1;21)(p36;q22) translocation represents a recurrent translocation targeting the RUNX1 and PRDM16 genes. At least three mechanisms through which the RUNX1-PRDM16 fusion might induce malignant transformation in hematopoietic cells are possible: expression of truncated RUNX1, expression of the RUNX1-PRDM16 fusion gene (both resulting in dominant negative inhibition of RUNX1), and ectopic PRDM16 expression. Identification of recurrent fusion genes will not only add to a better understanding of altered signaling mechanisms triggering oncogenesis, but also will enable specific molecular subtyping and, by doing so, facilitate therapeutic decision making.
1189
M-JPL Stevens-Kroef1, EFPM Schoenmakers1, M van Kraaij2, E Huys1, S Vermeulen1, B van der Reijden2 and A Geurts van Kessel1 1 Department of Human Genetics, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands and 2 Department of Hematology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands E-mail:
[email protected] References 1 Slovak ML, Bedell V, Popplewell L, Arber DA, Schoch C, Slater R. 21q22 Balanced chromosome aberrations in therapy-related hematopoietic disorders: report from an international workshop. Genes Chromosomes Cancer 2002; 33: 379–394. 2 Roulston D, Espinosa III R, Nucifora G, Larson RA, Le Beau MM, Rowley JD. CBFA2 (AML1) Translocations with novel partner chromosomes in myeloid leukemias: association with prior therapy. Blood 1998; 92: 2879–2885. 3 Webb DKH, Passmore SJ, Hann IM, Harrison G, Wheatley K, Chessells JM. Results of treatment of children with refractory anaemia with excess blasts (RAEB) and RAEB in transformation (RAEBt) in Great Britain 1990–99. Br J Haematol 2002; 117: 33–39. 4 Hromas R, Shopnick R, Jumean HG, Bowers C, Varella-Garcia M, Richkind K. A novel syndrome of radiation-associated acute myeloid leukemia involving AML1 gene translocations. Blood 2000; 95: 4011–4013. 5 Sakai I, Tamura T, Narumi H, Uchida N, Yakushijin Y, Hato T et al. Novel RUNX1-PRDM16 fusion transcripts in a patient with acute myeloid leukemia showing t(1;21)(p36;q22). Genes Chromosomes Cancer 2005; 44: 265–270. 6 Mochizuki N, Shimizu S, Nagasawa T, Tanaka H, Taniwaki M, Yokota J et al. A novel gene, MEL, mapped to 1p36.3 is highly homologous to the MDS1/EVI1 gene and is transcriptionally activated in t(1;3)(p36;q21)-positive leukemia cells. Blood 2000; 96: 3209–3214. 7 Nucifora F, Begy CR, Kobayashi H, Roulston D, Claxton D, Pedersen-Bjergaard J et al. Consistent intergenic splicing and production of multiple transcripts between AML1 at 21q22 and unrelated genes at 3q26 in (3;21)(q26;q22) translocations. Proc Natl Acad Sci USA 1994; 91: 4004–4008. 8 Zent CS, Mathieu C, Claxton DF, Zhang D-E, Tenen DG, Rowley JD et al. The chimeric genes AML1/MDS1 and AML1/EAP inhibit AML1B activation at the CSF1R promoter, but only AML1/MDS1 has tumor-promoter properties. Proc Natl Acad Sci USA 1996; 93: 1044–1048. 9 Senyuk V, Chakraborty S, Mikhail FM, Zhao R, Chi Y, Nucifora G. The leukemia-associated transcription repressor AML1/MDS1/EVI1 requires CtBP to induce abnormal growth and differentiation of murine hematopoietic cells. Oncogene 2002; 21: 3232–3240.
Leukemia
