Molecular dissection of t(11;17) in acute myeloid leukemia reveals a variety of gene fusions with heterogeneous fusion transcripts and multiple splice variants.

The majority of translocations that involve the long arms of chromosomes 11 and 17 in acute myeloid leukemia appear identical on the cytogenetic level. Nevertheless, they are diverse on the molecular level. At present, two genes are known in 11q23 and four in 17q12-25 that generate five distinct fusion genes: MLL-MLLT6/AF17, MLL-LASP1, MLL-ACACA or MLL-SEPT9/MSF, and ZBTB16/PLZF-RARA. We analyzed 14 cases with a t(11;17) by fluorescence in situ hybridization and molecular genetic techniques and determined the molecular characteristics of their fusion genes. We identified six different gene fusions that comprised seven cases with a MLL-MLLT6/AF17, three with a MLL-SEPT9/MSF, and one each with MLL-LASP1, MLL-ACACA, and ZBTB16/PLZF-RARA fusions. In the remaining case, a MLL-SEPT6/Xq24 fusion suggested a complex rearrangement. The MLL-MLLT6/AF17 transcripts were extremely heterogeneous and the detection of seven different in-frame transcript and splice variants enabled us to predict the protein domains relevant for leukemogenesis. The putative MLL-MLLT6 consensus chimeric protein consists of the AT-hook DNA-binding, the methyltransferase, and the CXXC zinc-finger domains of MLL and the highly conserved octapeptide and the leucine-zipper dimerization motifs of MLLT6. The MLL-SEPT9 transcripts showed a similar high degree of variability. These analyses prove that the diverse types of t(11;17)-associated fusion genes can be reliably identified and delineated with a proper combination of cytogenetic and molecular genetic techniques. The heterogeneity of transcripts encountered in cases with MLL-MLLT6/AF17 and MLL-SEPT9/MSF fusions clearly demonstrates that thorough attention has to be paid to the appropriate selection of primers to cover all these hitherto unrecognized fusion variants.

NUP98 is fused to topoisomerase (DNA) IIbeta 180 kDa (TOP2B) in a patient with acute myeloid leukemia with a new t(3;11)(p24;p15).

PURPOSE: The nucleoporin 98 kDa (NUP98) gene has been reported to be fused to 17 different partner genes in various hematologic malignancies with 11p15 aberrations. Cytogenetic analysis of an adult de novo acute myelogenous leukemia (M5a) revealed a t(3;11)(p24;p15), suggesting rearrangement of NUP98 with a novel partner gene. EXPERIMENTAL DESIGN: Fluorescence in situ hybridization (FISH) was used to confirm the involvement of NUP98 in the t(3;11)(p24;p15). Selection of possible NUP98 partner genes was done by computer-aided analysis of the 3p24 region using the University of California Santa Cruz genome browser. Fusion gene-specific FISH and reverse transcription-PCR analyses were done to verify the presence of the new NUP98 fusion. RESULTS: FISH analysis using a NUP98-specific clone showed a split signal, indicating that the NUP98 gene was affected by the translocation. Of the genes localized at 3p24, TOP2B was selected as a possible fusion partner candidate gene. Dual-color fusion gene-specific FISH and reverse transcription-PCR analysis verified that NUP98 was indeed fused to TOP2B. In addition to reciprocal NUP98-TOP2B and TOP2B-NUP98 in-frame fusion transcripts, an alternatively spliced out-of-frame TOP2B-NUP98 transcript that resulted in a premature stop codon was detected. Analysis of the genomic breakpoints revealed typical signs of nonhomologous end joining resulting from error-prone DNA repair. CONCLUSIONS: TOP2B encodes a type II topoisomerase, which is involved in DNA transcription, replication, recombination, and mitosis, and besides TOP1, represents the second NUP98 fusion partner gene that belongs to the topoisomerase gene family. This finding emphasizes the important role of topoisomerases in malignant transformation processes.

Screening for NUP98 rearrangements in hematopoietic malignancies by fluorescence in situ hybridization.

BACKGROUND AND OBJECTIVES: The aim of this study was to determine the incidence of rearrangements of NUP98 (the gene coding for nucleoporin 98kDa protein) in childhood acute myeloid leukemia (AML) and selected patients with 11p13-15 rearrangements. This aim was achieved using a fluorescence in situ hybridization (FISH) assay that allows the detection of NUP98 aberrations independently of the partner gene involved. DESIGN AND METHODS: Screening of 59 consecutive patients enrolled in the Austrian AML-BFM93 clinical trial was performed by dual-color FISH. In addition, 14 selected cases with various hematologic malignancies and 11p13-15 aberrations were analyzed. NUP98-positive cases were further investigated by fusion gene-specific FISH and reverse transcription polymerase chain reaction assays. RESULTS: Among the 59 AML patients, one NUP98-NSD1 positive case (1.7%) was detected. Among the 14 selected patients, five new NUP98 positive cases were determined. Two cases showed an inv(11)(p15q22)/NUP98-DDX10 fusion, one each displayed a t(5;11)(q35;p15)/NUP98-NSD1 and a t(11;20)(p15;q12)/NUP98-TOP1 fusion, and one case with a putative new fusion partner gene at 3p24 was identified. INTERPRETATION AND CONCLUSIONS: The observed frequency of 1.7% confirmed the low incidence of NUP98 rearrangements in childhood AML. The low occurrence of NUP98 rearrangements in selected samples with 11p13-15 alterations suggests the existence of variable chromosomal breakpoints and affected genes in this region. The identification of a new NUP98 fusion partner region confirms the evident promiscuity of NUP98. Thus, analysis of NUP98 aberrations by FISH seems to be the method of choice for determining the presence of these genetic lesions in unselected patients, and to confirm the involvement of NUP98 in cases with 11p15 aberrations.

Deregulation of the carbohydrate (chondroitin 4) sulfotransferase 11 (CHST11) gene in a B-cell chronic lymphocytic leukemia with a t(12;14)(q23;q32).

The t(12;14)(q23;q32) breakpoints in a case of B-cell chronic lymphocytic leukemia (B-CLL) were mapped by fluorescence in situ hybridization (FISH) and Southern blot analysis and cloned using an IGH switch-gamma probe. The translocation affected a productively rearranged IGH allele and the carbohydrate (chondroitin 4) sulfotransferase 11 (CHST11) locus at 12q23, with a reciprocal break in intron 2 of the CHST11 gene. CHST11 belongs to the HNK1 family of Golgi-associated sulfotransferases, a group of glycosaminoglycan-modifying enzymes, and is expressed mainly in the hematopoietic lineage. Northern Blot analysis of tumor RNA using CHST11-specific probes showed expression of two CHST11 forms of abnormal size. 5'- and 3'-Rapid Amplification of cDNA Ends (RACE) revealed IGH/CHST11 as well as CHST11/IGH fusion RNAs expressed from the der(14) and der(12) chromosomes. Both fusion species contained open reading frames making possible the translation of two truncated forms of CHST11 protein. The biological consequence of t(12;14)(q23;q32) in this case presumably is a disturbance of the cellular distribution of CHST11 leading to deregulation of a chondroitin-sulfate-dependent pathway specific to the hematopoietic lineage.

Quantitative comparison of the expression of EVI1 and its presumptive antagonist, MDS1/EVI1, in patients with myeloid leukemia.

The EVI1 gene in chromosome band 3q26 exhibits a number of properties consistent with a role as an oncogene, and its expression is activated in most myeloid leukemia patients with, as well as in a minority of patients without, 3q26 rearrangements. A splice variant of this gene, MDS1/EVI1, acts as its antagonist at least in some tissue culture assays. We established real-time quantitative reverse transcriptase polymerase chain reaction (RTQ-RT-PCR) assays for these mRNA variants to compare their expression levels in a quantitatively reliable manner. EVI1 was overexpressed to highly variable extents in all patients with, as well as in 14% of patients without, 3q26 rearrangements. In some of these samples, MDS1/EVI1 was also transcribed at elevated levels compared to those of healthy controls. However, although the induction of MDS1/EVI1 was comparable to, or higher than, that of EVI1 in three of five samples with a normal EVI1 locus, this was true for only two of 13 patients with a 3q26 aberration. We further provide preliminary evidence that the RTQ-RT-PCR assay may be useful for disease monitoring in patients overexpressing EVI1.