HOX11L2/TLX3 is transcriptionally activated through T-cell regulatory elements downstream of BCL11B as a result of the t(5;14)(q35;q32)

Maturation of immunoglobulin (Ig) and T-cell receptor (TCR) genes is associated with somatic V(D)J recombinase rearrangements occurring during lymphoid development.¹  A protein complex containing both ubiquitously expressed and lymphoid-specific proteins (RAG1, RAG2, and TdT) catalyzes this process. The complex interacts with recombination sequences (RSs) that limit the segments to be rearranged. During the recombination process, the extremities are nibbled, and additional nucleotides (Ns) are added without template by the TdT.¹  During this process, which includes double-strand DNA break, the Ig and TCR genes are prone to illegitimate rearrangements and chromosomal translocations observed in human T-cell acute lymphoblastic leukemia (T-ALL) often involve the site of the TCR genes.²  Molecular analyses of the chromosomal breakpoints of these translocations often showed hallmark of recombinase activity and allowed the characterization of numerous T-ALL oncogenes.

Many of the oncogenic events specific for T-cell leukemogenesis result in the activation of transcription factors, such as HOX11/TLX1, LMO2, or TAL1/SCL.^3,4  A frequent event in T-ALL is the occurrence of a 90-kb deletion of chromosome 1 sequences, juxtaposing the first exon of the SIL gene, to the 5′ part of the TAL1 gene.⁵  This rearrangement is observed in up to 20% of childhood T-ALL samples, and not in other ALL subtypes. The fusion points of the deletion bear the hallmarks of recombinase activity, which is therefore thought to be responsible for this rearrangement and provides the grounds for its specific association with T-ALL. Illegitimate recombinase activity is also responsible for at least some of the deletion affecting the P16-ARF tumor-suppressor locus observed in a T-ALL.⁶ 

Ectopic expression of the homeobox transcription factor TLX3 was recently reported in childhood T-ALL, virtually always in association with genomic rearrangement of its locus.^7-10  It is mainly associated with a cryptic t(5;14)(q35;q32) chromosomal translocation that recombines the RANBP17-TLX3 and BCL11B loci.^11,12  To uncover the mechanisms responsible for the translocation and ectopic expression of TLX3, and the reason for the specific association between t(5;14) and T-ALL, we analyzed the nucleotide sequence at the breakpoint of 8 examples of t(5;14). In addition, we identified cis-transcriptional regulatory elements lying downstream of BCL11B, which could be involved in the up-regulation of TLX3 following the translocation.

Samples were obtained with informed consent, in accordance with the Declaration of Helsinki. Approval for this study was given by the INSERM review board. The hematologic and immunophenotypic characteristics and conventional karyotypes of the 8 T-ALL patients with t(5;14)(q35;q32) are listed in Table S1, available on the Blood website (see the Supplemental Materials link at the top of the online article). The procedure for DNAse1 experiments is described in Figure S2. DNA analyses were performed using standard techniques. DND41, HPB-ALL, and Jurkat are T-ALL–derived cell lines; DND41 and HPB-ALL bear variant t(5;14).^12,13  The primers used are listed in Table S2. Inverse polymerase chain reaction (PCR) was performed as described previously,¹²  and classical PCR was then used to confirm breakpoint location and sequence. Sequence analyses were performed locally or at http://genome.ucsc.edu or http://www.ncbi.nlm.nih.gov. Appropriate fragments were cloned in pGL3 backbone (Promega, Madison, WI). Cells (2.5 × 10⁵) were transfected with 0.5 to 2 μg reporter gene constructs using Lipofectine (Invitrogen, Frederick, MD), and the luciferase activity was determined using luciferase assay system (Promega) after 24 hours. A representative experiment is shown, but every experiment was repeated at least 3 times in duplicate. The results were similar with or without normalization (using a thymidine kinase promoter–driven β-galactosidase reporter).

Approximate mapping of the breakpoints was done by fluorescence in situ hybridization (FISH), analyses and more precise localization was achieved by Southern blotting analysis using single-copy probes (data not shown and Figure 1A). The translocation breakpoints were amplified from patient's genomic DNA using an inverse PCR approach and analyzed by direct sequencing. Both translocation breakpoints (der5 and der14) could be analyzed in 7 cases and the fusion points exhibited complex features in only patient 1 (Figure S1). Most of the fusion points exhibited N nucleotides. In patients 3 and 7, no N nucleotides were observed, suggesting that the translocation could have occurred in a cell that did not express the TdT. Of importance, no convincing RS-like sequences were observed at the translocation breakpoints, suggesting that the recombinase activity is not involved in the occurrence of the translocation.

It has been proposed that the transcriptional activation of TLX3 could result from cis-activation of the gene by a BCL11B transcriptional regulatory element, juxtaposed to TLX3 following the translocation.^12,15  Within the hematopoietic system, expression of the BCL11B gene is restricted to the T-cell differentiation pathway.^7,16  It encodes a transcription factor essential to thymocyte differentiation.¹⁷  The location of the translocation breakpoints in patients 1 and 6 pinpointed to a 120-kb region of chromosome 14 susceptible to contain this putative minimal region. A previously described breakpoint¹⁵  allowed us to narrow down this region to 58kb lying more than 920-kb downstream of BCL11B polyadenylation site. We used DNAse1 hypersensitive experiments to identify active regulatory elements in this region. As shown in Figure 1B, 5 HSSs were observed in T-ALL cell lines (DND41 and Jurkat), but were absent or weaker in the non–T-cell lines Nalm6 and K562 (Figure S2).

To investigate for cis-activating properties, 1.5-kb fragments encompassing the HSSs were subcloned in pGL3-luciferase vector and tested for activation of the SV40 promoter. Only 2 regions (H3 and H4) showed enhancement of SV40 promoter activity in HPB-ALL (Figure 2A). Fine mapping of the HSS and luciferase reporter assays allowed us to map the active sequences within 464-bp (H3*) and 482-bp (H4*) fragments (data not shown and Figure 2B). As expected, cis-activation was observed in T-cell lines, but not in B-cell or myeloid cell lines. We next wanted to ensure the activity of these sequences on the TLX3 promoter. A 504-bp fragment within the 5′ side of TLX3 was subcloned in promoter-less pGL3-luciferase vector (Figures S3 and S4). As shown in Figure 2C, H3* and H4* sequences stimulated the TLX3 promoter, which exhibited only a weak activity on its own in T-cell lines. Taken as a whole, our results identify T-lymphoid–restricted (within the hematopoietic system) cis-activating elements downstream of BCL11B that are juxtaposed to the TLX3 locus as a result of the t(5;14).

Our observations suggest that the recombinase activity is not responsible for the occurrence of the t(5;14) in contrast with the SIL-TAL1 deletion. They rather suggest that the reason for the association between t(5;14) and T-ALL is the juxtaposition of cis-transcriptional regulatory elements that are specifically active during T-cell differentiation. In keeping, murine Bcl11b is very highly expressed at starting from DN2 (and latter) stage of thymic development (data not shown).

The t(5;14) situation is the most frequent example observed in T-ALL that results in bona fide cis-activation due to enhancer juxtaposition. The frequent SIL-TAL1 fusion leads to transcription of TAL1 from the SIL promoter.⁵  The breakpoints of chromosomal translocation affecting the TCRαδ locus frequently affect the 5′ part of the oncogene on the partner chromosome, and result in the transcription from diversity region or cryptic promoters close to the translocation breakpoints.^18-21  We provide here only qualitative notions, and additional work will be necessary to fully characterize the BCL11B gene cis-regulating elements.^22,23  However, it is tempting to speculate that the activity of these transcriptional regulatory regions represents a therapeutic target in t(5;14) patients.

Contribution: X.-Y.S., V.D.-V., I.A.-S., C.L., and I.R.-W. performed research and analyzed the data; P.B., M.L., M.L.-P., and F.M. provided vital reagents; R.B. and S.P.R. analyzed the data and reviewed the paper; O.A.B. and V.P.-L. designed research, analyzed data, and wrote the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

This work was supported in part by INSERM and the Ligue Nationale Contre le Cancer–Comité de Paris (équipe labellisée; O.A.B.). X.-Y.S. was a recipient of successive fellowships from the Fondation pour la Recherche Médicale (FRM), the Association pour la Recherche contre le Cancer (ARC), and the Programme de Recherches Avancées Franco-Chinois.

We thank M. Busson for valuable help.

Dr Su's current address is the Shanghai Institute of Hematology, Rui Jin Hospital, Shanghai, People's Republic of China.