Model for MLL translocations in therapy-related leukemia involving topoisomerase IIβ-mediated DNA strand breaks and gene proximity

Abstract

Topoisomerase poisons such as the epipodophyllotoxin etoposide are widely used effective cytotoxic anticancer agents. However, they are associated with the development of therapy-related acute myeloid leukemias (t-AMLs), which display characteristic balanced chromosome translocations, most often involving the mixed lineage leukemia (MLL) locus at 11q23. MLL translocation breakpoints in t-AMLs cluster in a DNase I hypersensitive region, which possesses cryptic promoter activity, implicating transcription as well as topoisomerase II activity in the translocation mechanism. We find that 2-3% of MLL alleles undergoing transcription do so in close proximity to one of its recurrent translocation partner genes, AF9 or AF4, consistent with their sharing transcription factories. We show that most etoposide-induced chromosome breaks in the MLL locus and the overall genotoxicity of etoposide are dependent on topoisomerase IIβ, but that topoisomerase IIα and -β occupancy and etoposide-induced DNA cleavage data suggest factors other than local topoisomerase II concentration determine specific clustering of MLL translocation breakpoints in t-AML. We propose a model where DNA double-strand breaks (DSBs) introduced by topoisomerase IIβ into pairs of genes undergoing transcription within a common transcription factory become stabilized by antitopoisomerase II drugs such as etoposide, providing the opportunity for illegitimate end joining and translocation.

Fig. 1.

Transcription factory model for the generation of MLL translocations. (A) Transcription units (genes) residing on different chromosomes (red and blue) may associate with a common transcription factory. (B) MLL locus showing breakpoint cluster region (BCR) (C) Ideograms showing locations of genes analyzed by RNA-FISH. Percentages refer to the reported frequency of occurrence of the indicated genes among MLL fusion alleles in pediatric and adult acute leukemias. (D) Representative RNA-FISH images showing KG1 cells simultaneously expressing MLL and AF4 or AF9 alleles; first column, separation >1 μm; second column, separation <1 μm; third column, overlapping. (E–H) Results of RNA-FISH analysis. Values are mean percentages derived from at least three experiments. Significance values refer to t tests for difference between the frequencies of colocalization or juxtaposition of MLL transcripts with AF4/AF9 versus MLL with UGCG. (I) Effect of etoposide on frequency of overlapping or juxtaposed MLL and AF9 DNA-FISH signals in Nalm-6 or Nalm-6^TOP2B−/−. Data are means ±SD. Significance value refers to t tests for the effect of TOP2B status on juxtaposition of MLL and AF9 loci.

Fig. 2.

Topoisomerase IIβ dependence for etoposide-mediated breaks at the MLL locus. (A) DNA-FISH was performed using a probe for the detection of MLL rearrangements, which hybridizes to the region centromeric (5′, green) and telomeric (3′, red) to the MLL BCR. Three representative nuclei are shown, one (labeled br) harboring one normal MLL allele (overlapping red and green spots) and one break (separated red and green spots). Two nuclei with the normal pattern (n) are also present. (B) Frequency of MLL breaks induced in Nalm-6 (WT) and Nalm-6^TOP2B−/− (β^−/−) cells after 72-h exposure to 100 nM etoposide. (C) Frequency of MLL breaks induced in Nalm-6 (WT) and Nalm-6^TOP2B−/− (β^−/−) cells after 72 h exposure to 75 nM NK314.

Fig. 3.

Etoposide-induced genotoxicity is topoisomerase IIβ dependent. Nalm-6 (WT), Nalm-6^TOP2B−/− (β^−/−), or Nalm-6^TOP2A+/− cells were treated with etoposide (100 nM) or solvent alone and were scored for the presence of micronuclei. All significance values were derived using Student’s t test.

Fig. 4.

Topoisomerase IIβ independence of etoposide-induced γH2AX foci. (A) Mean number of γH2AX foci per nuclei of cells treated with 0 or 5 μM etoposide. Foci numbers were counted from at least 30 nuclei per treatment and data are derived from at least three independent experiments. (B) Cells were quantified for γH2AX fluorescence per cell and normalized to the median fluorescence obtained with Nalm-6 cells after etoposide treatment.

Fig. 5.

MLL topoisomerase II ChIP analysis. (A) Map of the MLL locus and BCR with primer pairs used for PCR. (B and C) Sonicated, cross-linked chromatin from control KG1 cells or from KG1 cells treated with 100 μM etoposide was immunoprecipitated with antitopoisomerase IIα (B) or IIβ (C). qPCR was performed using the primers indicated. Data are expressed as percentage input recovered. (D) CTCF chIP, qPCR was carried out as in B. In B–D, experiments were repeated three times and data shown are means ±SEM.