The Paired Box Domain Gene PAX5 Is Fused to ETV6/TEL in an Acute Lymphoblastic Leukemia Case

Introduction

The PAX5 gene encodes the B-cell-specific activator protein, a transcription factor expressed in the early stages of B-cell differentiation that acts as a critical determinant of commitment to the B-lymphocyte pathway (1) . The B-cell-specific activator protein binds to DNA and functions both as a transcriptional activator and a repressor, because it positively controls CD19, mb-1 (Iga), N-myc, and Lef-1 expression and negatively regulates PD-1 transcription (2) . The PAX5 gene was localized by FISH 2 at 9p13 (3) and retains a high degree of homology between human and mouse. It was found juxtaposed to the IgH gene in cases with t(9;14)(p13;q32), a rare but recurring translocation found in a subset of B-cell non-Hodgkin’s lymphoma (reviewed in Ref. 4 ). Although breakpoints on both PAX5 and IgH genes were variable, t(9;14)(p13;q32) consistently left the PAX5 coding region intact, most likely resulting in deregulated expression of the gene product because of the proximity of IgH enhancers.

We previously reported a case of ALL in an adult patient carrying a t(9;12)(q11;p13) translocation (5) . FISH and Northern blot analyses provided direct evidence that the ETV6/TEL gene was rearranged in this case. The TEL gene encodes a member of the Ets family of transcription factors and is located on 12p13 (6) . It is characterized by a HLH domain at the NH₂-terminal region and an Ets-family DNA binding domain at the COOH terminus. TEL was found rearranged with platelet-derived growth factor-β receptor in a case of chronic myelomonocytic leukemia associated with a t(5;12) translocation (6) . Extensive analyses demonstrated that TEL is frequently rearranged in various hematological malignancies, with several partner genes (reviewed in Ref. 7 ). Interestingly, two mechanisms of transformation have been found associated to TEL fusions; in most cases, the fusion protein contains the HLH domain fused to phosphotyrosine kinase domains (Ref. 8 ; reviewed in Ref. 9 ), resulting in dimerization and constitutive kinase phosphorylation. By contrast, although TEL has been found fused to several transcription factors (reviewed in Ref. 9 ), only two cases have been described to involve the TEL DNA binding domain into the fusion protein, MN1-TEL (10) and BTL-TEL (11) . For MN1-TEL, it has been demonstrated recently that the fusion protein acts as an aberrant transcription factor (12) , and the same mechanism can be speculated for BTL-TEL.

We present here the molecular characterization of the previously reported case carrying a t(9;12)(q11;p13). By 5′ RACE PCR, a novel chimeric transcript was identified between the 5′-terminal region of PAX5 and most of the TEL gene. This is the first report of PAX5 rearrangement in a human malignancy resulting in a chimeric transcript.

Patient and Methods

Patient.

The main clinical and cytogenetic features of the patient have been described by Tosi et al. (5) .

FISH Analysis.

FISH analysis was performed on bone marrow metaphases from archival methanol:acetic acid-fixed chromosome suspensions stored at −20°C. Probes were labeled with biotin-16-dUTP by nick translation and hybridized as described previously (5) . Results were analyzed, and images were captured using a cooled CCD imaging system (Applied Imaging, Newcastle upon Tyne, United Kingdom). Probes used for FISH were YAC 964c10, containing the entire TEL gene (Centre d’Etude du Polymorphisme Humain, Paris, France), and cosmids from the TEL gene, 50F4 (exon 2) and 2G8 (exon 3), kindly provided by Dr. Marynen (Center for Human Genetics, University of Leuven, Leuven, Belgium).

RNA Isolation.

Total RNA was extracted by the standard guanidinium thiocyanate-phenol-chloroform extraction method. One μg of total RNA was reverse transcribed with Superscript II reverse transcriptase (Life Technologies, Inc.) in 20-μl final volume.

5′ RACE Analysis.

5′ RACE PCR was performed using the SMART RACE cDNA Amplification kit (Clontech, Palo Alto, CA) according to the manufacturer’s instructions. Briefly, first-strand cDNA was reverse transcribed from 1 μg of total RNA with Superscript II reverse transcriptase (Life Technologies, Inc.), according to standard procedures, using the SMART II-specific primer (5′-AAGCAGTGGTAACAACGCAGAGTACGCGGG-3′). 5′ RACE PCR was performed using a primer specific for TEL exon 3, TEL323 (5′-CTGAATGAGGAGATCGATAGCGAA-3′), in combination with Universal primer Mix (Long, 5′-CTAATACGACTCACTATAGGG CAAGCAGTGGTAACAACGCAGAGT-3′ and Short, 5′-CTAATACGACTCACTATAGGGC-3′). A nested PCR was performed using 5 μl of the primary 5′ RACE PCR product, after 1:50 dilution in Tricine-EDTA buffer. The internal primers used were: TEL266 (5′-TGGCCTTAAAGAAAACTCATTT-3′), specific for TEL exon 3, in combination with Nested Universal Primer (5′-AAGCAGTGGTAACAACGCAGAGT-3′).

The whole 5′ RACE PCR product was cloned in a pGEM-T EASY vector (Promega Corp., Madison, WI). Bacterial colonies were screened to contain recombinant plasmids by PCR, using common T7 and SP6 primers.

RT-PCR.

To amplify the breakpoint region, RT-PCR was performed on patient RNA using the PAX308s forward primer (5′-TCAAGCCTGGGGTAATTGGAGGA-3′) on PAX5 exon 3 and the TEL244as reverse primer (5′-AGCCCACTTGAGCCACTGG-3′) on TEL exon 3. Thirty-five cycles of PCR (30 s at 94°C, 30 s at 65°C, and 30 s at 72°C) were performed with 2 μl of cDNA in 50-μl reaction volumes. One unit of Taq DNA polymerase (Roche Diagnostics Corp.) was used.

To amplify the full-length PAX5A-TEL cDNA, RT-PCR was performed on the patient RNA with the PAX1 forward primer (5′-ATGGATTTAGAGAAAAATTATCCGAC-3′) on PAX5 exon 1A and the TEL1811 reverse primer (5′-TCAGCCCGCCTGCTCCGTCACACTTC-3′) on TEL exon 8. Thirty-five cycles of PCR (15 s at 94°C, 30 s at 60°C, and 2 min at 68°C) were performed with 2 μl of cDNA in 50-μl reaction volumes. One unit of Platinum Pfx DNA polymerase (Life Technologies, Inc.) was used. To amplify the PAX5B-TEL fusion, 5′-AATACACTGTAAGCACGACCCGTT-3′ PAX 27ex1B and 5′-CAGCCCACTTGAGCCACTGC-3′ TEL 244As primers were used; thirty-five cycles of PCR (15 s at 94°C, 30 s at 65°C, and 30 s at 72°C) were performed.

Sequencing.

Selected clones were sequenced using vector-specific oligonucleotides T7 and SP6 and the Cy5 Autoread Sequencing kit (Amersham Pharmacia Biotech) with an ALFred DNA Sequencer (Amersham Pharmacia Biotech). A database search was carried out using the BLASTN algorithm on the NIH Blast server.

Results

A preliminary description of the patient analyzed in this study was reported previously (5) . FISH analysis demonstrated that the YAC 964c10, containing TEL, showed a strong signal on the normal chromosome 12 and a consistently less intensive signal on the derivative chromosome 12 (the derivative chromosome 9 was absent in all of the metaphases analyzed). Therefore, the karyotype was refined as 43,XY,−4,−9,der(12)t(9;12)(q11;p13),−19 (5) . In addition, Northern blot analysis using a cDNA probe corresponding to the 3′ region of the TEL gene provided direct evidence that the TEL gene was rearranged. In the present study, FISH analyses on leukemic metaphases of the patient showed fluorescent signals corresponding to TEL cosmid 2 g8 (containing exons 3 and 4; Ref. 13 ) on both the der(12) and the normal chromosome 12 homologue (Fig. 1A) ⇓ . By contrast, fluorescent signals corresponding to TEL cosmid 50F4 (containing exon 2; Ref. 13 ) are only present on the normal chromosome 12 (Fig. 1B) ⇓ . These findings are consistent with a breakpoint location between exons 2 and 3 of the TEL gene.

Altogether, the Northern blotting analysis (5) , the exclusive presence of the der(12), with TEL oriented Tel-Cen (14) , and the FISH mapping of the breakpoint to between exons 2 and 3, suggested that the hypothetical chimeric transcript would have retained TEL sequences downstream of exon 2. On the basis of this information, 5′RACE was performed on the reverse transcribed total RNA, using nested primers located in TEL exon 3. The whole 5′ RACE-PCR product was cloned, and 17 bacterial colonies were screened by PCR using vector primers T7 and SP6. Twelve of 17 colonies were selected according to the different sizes of the recombinant plasmids on agarose gel, and 3 of them were sequenced with oligonucleotides TEL244as and Nested Universal Primer.

Sequence analyses of PCR products detected an unknown sequence, fused in-frame to the TEL exon 3 sequence. This sequence was proven to be the PAX5 cDNA by BLAST database searching (GenBank accession no. M96944; Fig. 2A ⇓ ). More precisely, the RACE-PCR product resulted in the fusion of exon 4 of PAX5 (nucleotide 550) to exon 3 of TEL (nucleotide 188; Fig. 2B ⇓ ). RT-PCR on diagnostic total RNA from the patient definitively confirmed the presence of both the alternative fusion transcripts between PAX5A or PAX5B and TEL mRNAs, respectively (Fig. 3, A and C) ⇓ . Moreover, RT-PCR analysis on diagnostic total RNA from the patient was performed to amplify the chimeric full-length PAX5A-TEL transcript (Fig. 3B) ⇓ . Only one band, corresponding to the expected full-length transcript, was detected. A 556- or 555-amino acid open reading frame was deduced for the full-length PAX5A-TEL or PAX5B-TEL fusions, respectively (starting at the corresponding PAX5 first ATG and ending at the TEL TGA stop codon). Both of these transcripts would code a M_r 64,000 protein, containing in both cases the “paired box” domain of PAX5 fused to the HLH and Ets-binding domains of TEL.

Discussion

We report here the identification of a novel chimeric transcript between the NH₂-terminal region of the PAX5 gene and most of the TEL gene in an adult patient with ALL. The definition of the chromosome 9 translocation breakpoint reported in the karyotype as q11 does not correspond to the known localization of the PAX5 gene on the 9p13–21 region (3) . This could be explained by a complex rearrangement occurring prior to the translocation, with subsequent juxtaposition of the PAX5 gene in proximity of the q11 region on chromosome 9.

PAX5 has been found previously to be deregulated as a consequence of the t(9;14), a recurrent rearrangement found in small lymphocytic lymphomas of the plasmacytoid subtype and in derived large-cell lymphomas (4) . This translocation juxtaposes the potent enhancer of the IgH gene at 14q32 to a region on 9p13 upstream of PAX5 exon 1A. In this case, it is PAX5 overexpression that contributes to the pathogenesis of the associated lymphomas. This is a common, well-known mechanism by which an oncogene can be activated as a consequence of a translocation in lymphomas (15) .

The present study is the first report of PAX5 rearrangement in a human malignancy resulting in a chimeric transcript, rather than being attributable to PAX5 overexpression. In the present case, the fusion transcript retains the PAX5 paired-box domain region and both the HLH and Ets-binding domains regions of TEL, a gene frequently involved in chromosomal rearrangements associated with hematological malignancies (6 , 7) .

The t(9;12) translocation here described is reminiscent of two different models: (a) the PAX3-FHKR or PAX7-FHKR fusions associated with t(2;13) and t(1;13), respectively, in alveolar rhabdomyosarcomas (reviewed in Ref. 16 ), both involving other members of the PAX gene family; (b) and the MN1-TEL fusion associated with myeloid leukemia (10) .

The PAX3 and PAX7 genes are members (along with PAX5) of a gene family comprising nine members (reviewed in Ref. 17 ). In alveolar rhabdomyosarcomas, breakpoints in PAX3 and PAX7 genes have been found consistently in intron 7 (16) . As a consequence, the PBD, the conserved octapeptide, and the homeobox domain characteristic of the PAX gene family are fused to the COOH-terminal portion of the DNA binding domain and the entire transactivation domain of the FKHR gene. In the fusion products, the PAX domains provide the DNA binding specificity for the fusion transcription factors (16) . The COOH-terminal transactivation domain of FKHR is relatively insensitive to the negative effect of the NH₂-terminal PAX3 modulatory domain, thus explaining how the chimeric protein becomes a potent transcriptional activator. Several studies conducted on the PAX3-FKHR fusion protein have demonstrated that this chimera has a potent transforming activity and it can exert an oncogenic effect through multiple pathways, by influencing either cellular growth, differentiation, and apoptosis (16) . The PAX5-TEL fusion transcript differs from PAX3/7-FKHR in that the breakpoint is located within PAX5 intron 4; thus, only the PAX5 PBD is fused to the TEL moiety. However, an intact PBD was shown to be both necessary and sufficient for the DNA binding capacity of PAX5. More recently, a fusion between PAX8 and the PPARγ1 gene has been found in human thyroid carcinoma (18) . In this case, the PBD and partial homeobox domain of PAX8 were fused to the DNA binding, ligand binding, RXR dimerization, and transactivation domains of PPARγ1. In this model, the PAX8-PPARγ1 fusion protein functions as a dominant-negative suppressor of wild-type PPARγ activities. In addition, it has been postulated that the chimera can also deregulate PAX8 pathways in thyroid cells and contribute novel activities that promote thyroid carcinoma formation.

The PAX5-TEL fusion is similar to the MN1-TEL fusion found in myeloid leukemia (10 , 12) . The sequence at the NH₂-terminal region and the nuclear localization of MN1 are common to many transcription factors. It has been demonstrated recently that the MN1-TEL type I fusion which, in common with PAX5-TEL, retains the HLH TEL domain, has a transforming activity in NIH-3T3 fibroblasts, which depends on the DNA binding activity of TEL and transactivation- and transformation-specific sequences of MN1 (12) . Moreover, it has been determined that MN1 functions as a strong transcriptional coactivator (12) and that fusion of this molecule to TEL changes it from a repressor (19) into a strong transactivating transcription factor (12) . Interestingly, this change is similar to that resulting from the PAX3-FKHR and EWS-FLI1 fusion (16 , 20) . FLI1 is another member of the ETS family of transcription factors which, in common with TEL, retains the DNA binding domain when fused to EWS in Ewing sarcomas. On the basis of the analysis of MN1-TEL, PAX3-FKHR, and EWS-FLI1, it can be suggested that both the strong transactivating sequences added at the NH₂-terminal of the fusion, as well as the addition of sequences with other activities, are important for the oncogenic activity of the chimeric protein. When comparing the PAX5-TEL fusion transcript with the above mentioned models of chimeric transcription factors, it is reasonable to hypothesize that the PAX5-TEL chimeric protein could also act as an aberrant transcription factor, which could affect both the PAX5 and the TEL pathways of transcription modulation.

Two alternative transcripts for PAX5 exon 1 (1A and 1B) have been described previously (21) . Transcription of exon 1A initiates at a single site downstream of a conserved TATA sequence; by contrast, PAX5B transcripts are heterogeneously initiated, in agreement with the absence of any TATA-like sequence. The ratio of PAX5A:PAX5B varied from 6:1 in spleen to 1:1 in the B-mature WEH1–231 cell line. Moreover, it has been demonstrated that the upstream (1A) promoter is predominantly active in B lymphocytes, whereas the downstream promoter is used in all PAX5 expression domains (21) . In the present case, both the alternative PAX5A-TEL and PAX5B-TEL fusion transcripts have been detected by single RT-PCR, with similar efficiency (Fig. 3, B and C) ⇓ . The two resulting fusion transcripts are differentially regulated by the two different PAX5 promoters, and the two putative chimeric proteins are completely different at the NH₂-terminal region. The role of the two putative chimeric proteins must be clarified by functional assays.

Further analyses will be directed to clarify the real incidence of the PAX5-TEL rearrangement on B-cell precursor ALL cases and to analyze the functional activity of the fusion protein. Moreover, the 9p13–21 region is of particular interest because this is a region frequently deleted in leukemia patients (22) . Additional experiments are now in process to test the involvement of the PAX5 gene in cases with del(9p).