Acquired balanced chromosomal translocations provide crucial diagnostic and prognostic data in cancer patients. They are probably pathogenetically significant as initiating events if present as the only cytogenetic changes 1. These rearrangements often result in the formation of a fusion gene, encoding a chimeric oncogenic protein, or in the deregulation of the expression pattern of genes flanking the breakpoint regions by promoter swapping 123. Overall, gene deregulation can be accomplished through position effects, such as a gene juxtaposition close to an active chromatin domain. More rarely, translocations may result in gene downregulation because of hypermethylation of CpG islands at the breakpoint 4, or in the formation of chimeric transcripts that do not encode for a protein with oncogenic potential 5.

T-ALL, as other hematological malignancies, is associated with a number of chromosomal abnormalities, resulting either in a position effect or in a gene fusion 6.

Here we report the cloning of a novel balanced t(2;11)(q11.2;p15.1) translocation, found as the sole cytogenetic abnormality in the BM of a childhood T-ALL case, accompanied by the ectopic expression of the 3' portion of EST DA926692.

A 14-year-old male patient was referred to our hospital for rhinolalia. Total body CT disclosed a diffuse enlargement of the rhinopharynx vault, lymphoid hyperplasia of the Waldeyer's tonsillar ring, and a mediastinal enlargement. BM aspiration showed 90% of blast cells. Flow cytometry revealed that blast cell population was positive for CD 45, 2, 5, 7, and cytoplasmic CD3, while negative for CD1a and surface CD3. These findings were consistent with a diagnosis of early T-ALL, thus the patient started treatment based on AIEOP-BFM ALL 2000 protocol 7.

At the end of induction therapy, the patient was categorized as "high risk" due to incomplete remission, as per protocol stratification criteria. After consolidation therapy, the BM aspirate showed no haematological remission, so treatment continued following the AIEOP LLA REC 2003 protocol and complete hematological remission was obtained.

One year from diagnosis, the patient underwent HLA-identical hematopoietic stem cell transplantation from an unrelated donor. He is now doing well and is still in hematologic remission.

G-banding analysis of bone marrow metaphases at onset revealed the karyotype 46,XY,t(2;11)(q11.2;p15.1) [27] 8 (Fig. 1A). FISH analysis excluded the presence of known rearrangements (data not shown), and revealed that the chromosome 11 breakpoint was encompassed by the BAC clone RP11-320K10 (chr11:18,229,291–18,413,758) and the fosmid G248P8335B2 (Fig. 1B), harboring the 5' UTR regions of two genes, HPS5 and GTF2H1, located 422 bp apart and in opposite transcriptional orientation (Fig. 1F).

The break on chromosome 2 was mapped within the interval chr2:96,884,868–97,812,993, delimited by the BACs RP11-1141J14 (chr2:96,884,868–97,038,565) and RP11-716G6 (chr2:97,631,541–97,812,993) at 2q11.2, yielding FISH signals on der(2) and der(11), respectively (Figs. 1B,C). It was not possible to further refine this breakpoint region by large-insert probes, as it is almost entirely composed of segmental duplications. We identified two small regions devoid of duplicons: chr2:97,284,347–97,296,849 (12,503 bp) and chr2:97,379,383–97,391,787 (12,405 bp) (Figure 1E). We obtained a long-range PCR product from the latter interval, that appeared to encompass the breakpoint on chromosome 2 (Figure 1D,E). This conclusion, however, was regarded with caution, since the region is rich in copy number variants (UCSC, Structural Variation track) providing a possible alternative explanation for the FISH results. The amplified interval does not contain known genes, although a number of ESTs of unknown function are annotated (Fig. 1E).

Vectorette-PCR using a reverse primer (hpsgtf-R, Table 1), specific for the genomic segment between genes HPS5 and GTF2H1, produced a fragment of approximately 800 bp. The sequence (Accession No. EU617360) showed that chromosome 11 at nt 18,300,006 (HPS5 intron 1) was fused with chromosome 2 at nt 97,384,504 (EST DA926692 exon 2) (Fig. 1G). Unexpectedly, sequences from chromosomes 2 and 11 had the same orientation.

The same approach was used to clone the reciprocal genomic fusion DA926692/HPS5. Vectorette PCR with a forward primer (hps5-F, Table 1) designed within HPS5 intron 1 yielded a fragment of 1,050 bp. The sequence (Accession No. EU617359) revealed that chromosome 2, at nt 97,384,507 (DA926692 exon 2), was fused with chromosome 11 at nt 18,299,998 (HPS5 intron 1) (Fig. 1H). Again, sequences from the two chromosomes had the same orientation. These observations clearly suggested that one of the two sequences had undergone a submicroscopic inversion before translocation. As a consequence, DA926692 and HPS5 were brought, by the inversion, into the same orientation.

5'- and 3'-RACE-PCR as well as RT-PCR experiments, aimed at cloning possible HPS5/DA926692 chimeric transcripts, failed.

Further RQ-PCR analyses aiming at quantifying the expression levels of genes flanking the breakpoints, i.e. HPS5 (Fig. 2A) and GTF2H1 (data not shown) on chromosome 11, and ANKRD36, FAHD2B, KIAA1641, ZAP70 on chromosome 2, revealed no misexpression in the present case compared to other T-ALL cases without the t(2;11) translocation (data not shown).

The same approach was used for DA926692. A primer set specific for exon 1, upstream to the breakpoint, revealed no statistically significant difference in expression level compared to controls (Fig, 2B). Interestingly, a primer pair designed in exon 2 (Table 1), downstream to the breakpoint, detected expression only in the t(2;11) case. RT-PCR with the same primer pair revealed a band of 173 pb only in the patient's BM and in the genomic DNA sample, but neither in the T-ALL controls, nor in normal BM or PB (Fig. 2C). These results were confirmed by nested RT-PCR with the primer pair DA926692 ex2a, further validated by the sequencing of PCR products (data not shown). Purity of all RNA samples was tested using ACTB primers in PCR experiments (Fig. 2D).

Next, we investigated the expression pattern of 3'DA926692 by RQ-PCR in 21 adult tissues. The results showed the occurrence of the transcript in 6 of 23 examined tissues (Fig. 3) [normal BM and PB were already shown to be negative (Fig. 2C)].

Notably, in silico translation of the sequence corresponding to the 173 bp PCR product showed that it contained a high density of stop codons resulting in very short ORFs.

We hypothesized that this ncRNA could represent a precursor of an miRNA, but the miRNA prediction table produced by miRRim http://mirrim.ncrna.org/TableS1.html and a sequence similarity search at miRBase http://microrna.sanger.ac.uk/sequences/search.shtml showed no significant matches (data not shown). Similarly, miRNA predictions by ProMiR II http://cbit.snu.ac.kr/~ProMiR2/ and miR-abela http://www.mirz.unibas.ch/ did not retrieve any positive results (data not shown).

Finally, we used the Gene Prediction tracks of the UCSC Genome Browser to discover if DA926692 might be part of a more complex predicted gene. Notably, the N-SCAN PASA-EST track showed that DA926692 and the EST CD110063, located in a double position both proximally (chr2:97,362,909–97,379,287) and distally (chr2:97,391,882–97,408,261) to DA926692 (Fig. 1E), may be parts of the predicted gene chr2.98.005.a (data not shown).

To verify this hypothesis, we checked for chr2.98.005.a gene expression using the primer combination DA926692ex2anewF+CD110063-R in the patient's BM cDNA. No PCR products were recovered.

In conclusion, we describe a t(2;11) translocation in T-ALL not resulting in any HPS5/DA926692 chimeric transcript. The impact on cancer causation of a category of "unproductive" translocations, such as ETV6 gene fusions, remains unclear 5. In such cases, the pathogenetic outcome does not seem to coincide with the formation of a chimeric protein, but perhaps with the production of a truncated form of original proteins (loss of function), or the deregulation of flanking genes.

Here, the only observed gene expression change consisted in the ectopic expression of the 3' portion of DA926692 in the patient's BM (Fig. 2C). One hypothesis is that the juxtaposition, at a distance of only 196 nt, of this genomic segment to the CpG island located upstream to both HPS5 and GTF2H1 (chr11:18,300,202–18,300,814) could have activated its expression in the BM cells of the patient.

DA926692 was isolated from a small intestine cDNA library, described as similar to Ig kappa chain V-I region HK103 precursor http://www-bimas.cit.nih.gov/cgi-bin/cards/carddisp.pl?gene=LOC643543&search=DA926692&suff=txt, but no known function has yet been reported. Its expression in some human adult tissues (Fig. 3) may suggest a potential function of this transcript in the positive samples.

As its function is unclear, it is difficult to speculate on a possible role of this short ncRNA in T-ALL tumorigenesis. As the bioinformatic analysis excluded the possibility that it could behave as a pre-miRNA, we could speculate that this ncRNA may belong to the class of small RNA (20–300 nt), commonly found as transcriptional and translational regulators 910. A variety of complex cellular mechanisms such as gene transcription, chromatin structure dynamics, and others have already been connected to their function 10. Furthermore, changes in expression levels of ncRNAs have been associated with different types of cancer 9.

In this context, we could speculate that the ectopic expression of the 3' portion of DA926692 in chilhood T-ALL could have a pathogenetic role, even if of still unclear significance. It is presently unknown, however, if the described rearrangement represents a primary or a secondary genetic aberration at disease onset. Further studies on similar cases may be extremely important to confirm this new impact of chromosomal translocation, i.e. activation of ncRNA in tumors.

(T-ALL): T-cell Acute Lymphoblastic Leukemia; (TCR): T-Cell Receptor; (CT): Computed Tomography; (BM): Bone Marrow; (PB): Peripheral Blood; (FISH): Fluorescence In Situ Hybridization; (BAC): Bacterial Artificial Chromosome; (HPS5): Hermansky-Pudlak syndrome 5 isoform b; (GTF2H1): general transcription factor IIH, polypeptide 1; (nt): nucleotide; (RQ-PCR): Real-Time Quantitative PCR; (ANKRD36): ankyrin repeat domain 36; (FAHD2B): fumarylacetoacetate hydrolase domain containing; (KIAA1641): hypothetical protein LOC57730; (ZAP70): zeta-chain associated protein kinase 70 KDa; (ORFs): Open Reading Frames; (miRNAs): microRNAs; (ncRNA): non-coding RNA.

The authors declare that they have no competing interests.

MCG and AL designed research and performed molecular analyses. FA and IV contributed in clinical data collection and analysis. SC and AL performed classical cytogenetic analysis and contributed in the development of the study. LI performed molecular cytogenetic analysis. PD performed bioinformatic analysis. IP contributed in molecular data analysis and interpretation of the results. MR and GB gave final approval to the manuscript. CTS supervised the study and drafted the manuscript.