Bone marrow ectopic expression of a non-coding RNA in childhood T-cell acute lymphoblastic leukemia with a novel t(2;11)(q11.2;p15.1) translocation

Chromosomal translocations play a crucial role in tumorigenesis, often resulting in the formation of chimeric genes or in gene deregulation through position effects. T-cell acute lymphoblastic leukemia (T-ALL) is associated with a large number of such rearrangements. We report the ectopic expression of the 3' portion of EST DA926692 in the bone marrow of a childhood T-ALL case showing a t(2;11)(q11.2;p15.1) translocation as the sole chromosome abnormality. The breakpoints, defined at the sequence level, mapped within HPS5 (Hermansky Pudlak syndrome 5) intron 1 at 11p15.1, and DA926692 exon 2 at 2q11.2. The translocation was accompanied by a submicroscopic inversion that brought the two genes into the same transcriptional orientation. No chimeric trancript was detected. Interestingly, Real-Time Quantitative (RQ)-PCR detected, in the patient's bone marrow, expression of a 173 bp product corresponding to the 3' portion of DA926692. Samples from four T-ALL cases with a normal karyotype and normal bone marrow used as controls were negative. It might be speculated that the juxtaposition of this genomic segment to the CpG island located upstream HPS5 activated DA92669 expression. RQ-PCR analysis showed expression positivity in 6 of 23 human tissues examined. Bioinformatic analysis excluded that this small non-coding RNA is a precursor of micro-RNA, although it is conceivable that it has a different, yet unknown, functional role. To the best of our knowledge, this is the first report, in cancer, of the activation of a small non-coding RNA as a result of a chromosomal translocation.


Findings
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 [1][2][3]. 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 HLAidentical hematopoietic stem cell transplantation from an unrelated donor. He is now doing well and is still in hematologic remission.
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).
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.  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 Results of the cytogenetic and genomic characterization of the t(2;11) traslocation breakpoints Figure 1 (see previous page) Results of the cytogenetic and genomic characterization of the t(2;11) traslocation breakpoints: A) Karyotype of the case described in the present study. The black arrows point on derivative chromosomes 2 and 11. B) FISH results obtained with fosmids and BAC clones delimiting the breakpoint regions on der(2) and der(11). C) Map of chromosome 2 breakpoint region, according to the latest release of the UCSC Genome Browser (March 2006), delimited by BAC clones RP11-1141J14 (left) and RP11-716G6 (right). RefSeq Genes and CpG islands are represented in blue and green, respectively. Intra-chromosomal duplications are reported at the bottom of the figure. Gray, yellow and orange colours refer to the percentage of sequence similarity of the duplication (respectively 90-98%, 98-99%, and >99%). D) FISH results obtained using the Chr2-97-1 Long-PCR product on the bone marrow of the patient. E) Detailed magnified map of the chromosome 2 breakpoint region, showing the location of the Chr2-97-1 probe. F) Map of the breakpoint region on chromosome 11. G) and H) Partial chromatograms of the junctions sequences on both derivative chromosomes 11 (G) and 2 (H). Inserted nucleotides [three (CAT) and sixteen (GGCGGTATGTCCGTAC) at der(11) and der (2), respectively] at the junctions are underlined in purple. 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

Figure 2
Results of the HPS5 and DA926692 gene expression analysis   (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.
Tissue expression pattern of EST DA926692 Figure 3 Tissue expression pattern of EST DA926692: DA926692 expression analysis was performed using cDNA multiple tissue (Ambion, Milan, Italy) First Choice Total RNA Survey Panel (Catalog No. AM6000) (adipose, bladder, brain, cervix, esophagus, heart, kidney, liver, lung, ovary, placenta, prostate, skeletal muscle, small intestine, spleen, testes, thymus, thyroid, trachea), according to the manufacturer's instructions. We have also tested a pool of tonsil cDNA extracted from three normal individuals. The primer combination used was DA926692ex2anew (F+R). The results, evaluated by RQ-PCR, showed positivity of spleen, ovary, placenta, thyroid, lung, and colon. Compared with the reference ACTB gene expression levels (data not shown), the overall DA926692 expression in the positive tissues was found to be relatively low (2 -ΔCt ≤ 1.8 × 10 -4 ). The strongest expression was seen in the colon.
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/cgibicards/carddisp.pl?gene=LOC643543&search=DA92669 2&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 [9,10]. 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.