Introduction

Childhood acute lymphoblastic leukemia (ALL) is a heterogenous group of leukemias with a predominance of the B-cell precursor phenotype (Pui, 1995). A minority of these leukemias is associated with a translocation involving the MLL gene (also called ALL-1or HRX), which has been identified in the 11q23 translocation (Gu et al., 1992; Tkachuk et al., 1992). Chromosomal abnormalities involving 11q23 were found in both ALL and acute myeloid leukemia (AML) (Rowley, 1998; Ferrando and Look, 2000; Hayashi, 2000). MLL gene rearrangement, which is found in the majority of infants (Rubnitz et al., 1994; Taki et al., 1996) and therapy-related leukemias (Hunger et al., 1993; Super et al., 1993), is strongly associated with a poor outcome in infants with ALL compared to older children with ALL, or with AML. It occurs in 60–70% of infants with ALL and in 10% of cases in older children and adults (Pui, 1995; Rowley, 1998; Hayashi, 2000). The MLL is translocated to more than 40 different chromosomal loci (Rowley, 1998; Hayashi, 2000). Up to now, more than 30 partner genes for MLL have been cloned from leukemic cells with varieties of 11q23 translocations (Hayashi, 2000). t(4;11)(q21;q23) mostly occurs in infant ALL and confers a dismal prognosis in this age group (Pui et al., 2002). Two AF4-related genes have been identified, LAF4 (Ma and Staudt, 1996) and FMR2 (Gecz et al., 1996; Gu et al., 1996). Recently, we identified that AF5q31 is a fusion partner of the MLL in infant ALL with ins(5;11)(q31;q13q23) (Taki et al., 1999). In this study, we analysed a pediatric ALL with t(2;11)(q11;q23) and identified a fusion of the LAF4 and MLL genes.

Results

Rearrangement of the MLL gene in t(2;11)-childhood ALL

A 2-year-old girl was diagnosed as having B-precursor ALL. Cytogenetic analysis of this patient at diagnosis showed to be 47, XX, +X, t(2;11)(q11;q23) in 17 of 20 bone marrow cells (Figure 1a) and expressed CD10 and CD19. We identified the chromosomal breakpoint within the breakpoint cluster region of the MLL gene at 11q23 by Southern blotting using a cDNA probe (probe x) spanning exons 5–11 in the MLL locus (Figure 1b). To date, two AF4-related genes, AF4 and AF5q31, have been identified as fusion partners of the MLL gene in ALL with 11q23 translocations and another AF4-related gene, LAF4, was located on chromosome 2q11.2–12. Therefore, we highly suspected that the partner gene of the MLL in t(2;11) could be the LAF4 gene.

Figure 1.

Identification of the MLL-LAF4 fusion transcripts. (a) Partial karyotype of chromosomes 2 and 11. (b) Rearrangements of the MLL gene by Southern blot analysis with BamHI digestion. The rearranged band is indicated by an arrow. N, normal peripheral lymphocyte. Pt, leukemic cells from patient. (c) Detection of the MLL-LAF4 chimeric transcripts by RT-PCR using ALL-5S and LAF4-2AS. The MLL-LAF4 fusion transcripts were amplified (lane1), while LAF4-MLL fusion transcripts were not amplified (lane3) by nested RT-PCR using LAF4-2S and ALL-9A for first PCR, and LAF4-4-5S and ALL-8A for second PCR. Lane 2, water. M, size marker

Full figure and legend (70K)

Detection of three MLL-LAF4 fusion transcripts

We performed reverse-transcription–polymerase chain reaction (RT–PCR) analysis of RNAs from the patient's leukemic cells, using a sense primer from MLL exon 5 (MLL-5S) and an antisense primer from LAF4 (LAF4-2AS), and obtained two different-sized products from the patient (Figure 1c). These PCR products were subcloned and sequence analysis revealed that they contained 636 bp (type 1), 560 bp (type 2), 522 bp (type 3), 519 bp (type 4), and 446 bp (type 5) products. Schematic representation of the fusion transcripts is shown in Figure 2a. The 636 bp product contained a fusion of a 403 bp sequence of exon 8 in the MLL gene at the 5' region to a 233 bp sequence of exons 6 and 7 in the LAF4 gene at the 3' region of the RT–PCR product. The 560 bp product contained a fusion of a 403 bp sequence of exon 8 in the MLL gene at the 5' region to a 157 bp sequence of exon 7 in the LAF4 gene at the 3' region of the RT–PCR product. Both of these products (types 1 and 2, and also types 4 and 5) had in-frame junctions. Sequence analysis indicated an out-of-frame fusion of MLL and LAF4. We next performed nested RT–PCR to detect reciprocal LAF4-MLL chimeric transcript. However, reciprocal LAF4-MLL fusion transcripts were not generated even by nested RT–PCR (Figure 1c). The fusion point in the chimeric transcript of LAF4 was located within the region homologous to AF4 and AF5q31 (Figure 2b).

Figure 2.

(a) Different fusion transcripts of MLL-LAF4 are schematically depicted. The gray/dotted boxes denote predicted MLL exons and the white boxes represent predicted LAF4 exons. ALL-5S and LAF4-2AS indicate the primers used for RT–PCR. (b) Schematic illustration of LAF4, AF4, and AF5q31 proteins. Arrows indicate the fusion points with MLL. The horizontal scale shows the amino-acid positions based on the LAF4 protein sequence

Full figure and legend (69K)

FISH analysis of the patient's leukemic cells with t(2;11)

Fluorescence in situ hybridization (FISH) analysis using an LAF4-specific BAC clone showed the hybridization of the clone to both the der(2) and der(11) chromosomes, confirming that LAF4 was split in the t(2;11) (Figure 3).

Figure 3.

FISH analysis of the leukemic metaphase. Split signals (arrows) of a BAC clone RP11-436F6 containing LAF4 are observed on the boundary between painted (arrows) and unpainted regions of der(2)t(2;11) and der(11)t(2;11). The intact RP11-436F6 signal was observed on the normal chromosome 2 (arrowhead)

Full figure and legend (130K)

Expression of the LAF4 gene in normal human tissues

To examine the expression of the LAF4 gene, we performed Northern blot analysis on poly (A)⁺ RNA from various human tissues. Expression of the 8.5-kb transcript was detected in the adult heart, brain, and placenta, and all fetal tissues examined, particularly the fetal brain at a high level (Figure 4a, b).

Figure 4.

Northern blot analysis of RNAs from fetal (a) and adult (b) human tissues. (c) Expression of the LAF4, AF4, AF5q31 genes in varieties of leukemic cell lines and EBV-B cell lines detected by RT–PCR. Primers used were LAF4-2S and LAF4-1AS for LAF4, AF4-1S and AF4-1A for AF4, and 5-4S and 5-5A for AF5q31, respectively. Lanes 1–3, EBV-B cell lines; Lanes 4–6, B precursor (B-pre) ALL cell lines; Lanes 7–9, B-ALL cell lines; Lanes 10–12, T-ALL cell lines; Lanes 13–15, CML cell lines; Lanes 16–17, AML cell lines

Full figure and legend (142K)

Expression of the LAF4 gene in various leukemic cell lines

We analysed expression of the LAF4 gene in 34 leukemic and five EBV-B cell lines by RT–PCR. LAF4 expression was found in all ALL cell lines, four (80.0%) of five EBV-transformed B lymphocyte (EBV-B) cell lines, four (57.1%) of seven AML cell lines, and one (20.0%) of five CML cell lines. To compare the LAF4 expression with that of AF4 and AF5q31, the AF4 and AF5q31 expressions were also examined by RT–PCR. AF4 and AF5q31 were expressed in all cell lines examined (Figure 4c).

Discussion

In this study, we identified that an AF4-related gene, LAF4, was fused to MLL in pediatric ALL with t(2;11)(q11;q23). In the patient's leukemic cells, only an MLL-LAF4 fusion transcript was detected, but no reciprocal LAF4-MLL transcript, suggesting that the 5'-MLL-LAF4-3' fusion transcript is critical for leukemogenesis as previously described (Taki et al., 1999; Ono et al., 2002).

t(4;11) creating MLL-AF4 is most frequent in ALL with 11q23 translocations, and is associated with an extremely poor prognosis (Pui et al., 2002). Patients with t(4;11) are characterized by very young age, hyperleukocytosis, early pre-B phenotype (CD10-/CD19+), and poor treatment outcome for infants and patients aged more than 10 years (Heerema et al., 1999), although the relation between fusion partner of MLL gene and prognosis remains controversial (Pui et al., 2002). The clinical features of our patient, who was 2 years old and whose leukemic cells expressed CD10 and CD19, are not typical for ALL with MLL gene rearrangement. It was reported that differences in the chromosomal breakpoints of the MLL gene between infants and children or adults with t(4;11)-ALL suggest different mechanisms for the development of these instances of ALL and, thus, their different biologic functions (Reichel et al., 2001). Further accumulation of these patients is needed to clarify the significance of t(2;11), CD10 positive expression, and non-infant ALL with MLL gene rearrangement.

LAF4 originally was identified as a lymphoid-restricted nuclear protein that showed strong sequence similarity to AF4 from a subtracted cDNA library (Ma and Staudt, 1996). LAF4 protein is a member of a family of proteins including AF4, FMR2, and AF5q31, which have various similarities: size of transcripts, protein size of 1300 amino acids, and nuclear localization of the proteins (Gecz et al., 1996; Gu et al., 1996; Ma and Staudt, 1996). Functional studies of LAF4 and AF4 have shown that these proteins have the potential to achieve the transcription of a receptor gene in an in vivo transcription activation assay (Ma and Staudt, 1996). The role of LAF4 in normal development has not been elucidated, although mouse Af4 was found to be critical for normal lymphocyte development by knockout mouse analysis (Isnard et al., 2000). AF4 and AF5q31 were also found to be involved in ALL, as fusion partners of the MLL gene. The breakpoint of LAF4 was located within the region homologous to the transactivation domain of AF4 and AF5q31. The transcriptional activation domains of LAF4, AF4, and AF5q31 were retained in the MLL-LAF4, MLL-AF4, and MLL-AF5q31 fusion proteins and, thus, may functionally replace the normal MLL activation domain. It is certainly conceivable that the replacement of the MLL activation domain with the activation domains of LAF4, AF4, and AF5q31 could result in aberrant regulation of certain key target genes (Ma and Staudt, 1996). These findings suggest that these three fusion proteins, MLL-LAF4, MLL-AF4, and MLL-AF5q31, have a similar structure and may contribute to leukemogenesis through a similar mechanism.

LAF4 expression was previously analysed in only mouse tissues and a small number of human and mouse leukemic cell lines (Ma and Staudt, 1996). We analysed LAF4 expression in normal human tissues and compared it with the result in mouse tissues. LAF4 was expressed in the adult heart, brain, and placenta, and the fetal brain, lung, liver, and kidney, particularly the fetal brain at a high level, in human tissues, although Laf4 was reported to be expressed at higher levels in lymphoid tissues and at lower levels in the brain and lung in mouse tissues, suggesting that LAF4 plays different roles in humans and mice. We also compared the expression pattern of LAF4 with that of AF4 and AF5q31 in normal human tissues. In our previous study, the expression pattern of AF4 and AF5q31 was very similar in human adult tissues, but different in fetal tissues (Taki et al., 1999). AF4 and AF5q31 expression was detected in the adult heart, placenta, skeletal muscle and pancreas however, LAF4 was not detected in the skeletal muscle or pancreas (Figure 4b). The expression pattern of AF4 and AF5q31 was quite different from that of LAF4, suggesting that LAF4 may play a different role in normal development from AF4 and AF5q31.

LAF4 was reported to be predominantly expressed in pre-B cells, intermediately in mature B cells, and absent in plasma cells in human and mouse lymphoid cell lines by Northern blot analysis, thus suggesting that LAF4 plays a potential regulatory role in early lymphoid development (Ma and Staudt, 1996). Although our result by RT–PCR analysis is partially different from that by Northern analysis, the expression of the LAF4 gene was low or undetectable in CML, AML, and EBV-B cell lines, and highest in lymphoid cell lines. On the other hand, AF4 and AF5q31 were expressed at similar levels in all leukemic cell lines examined. These findings suggested that LAF4 plays a role in the early lymphoid lineage, although AF4 and AF5q31 are involved more widely. Further functional analyses of the three AF4-related proteins may provide new insights into the pathogenesis of an important class of ALL with MLL-AF4 family members.

Materials and methods

Patient

A 2-year-old girl, with anemia (Hb 5.5 g/dl), low platelet count (13 000/l), and low leukocyte count (2800/l) containing 48% blasts in peripheral blood, was diagnosed as having ALL in December 2001. Leukemic cells expressed CD10, CD19, CD34, CD58, TdT, and HLA-DR in bone marrow cells and were cytogenetically characterized as 47, XX, +X, t(2;11) (q11;q23). She achieved a complete remission with chemotherapy and received allogeneic haemopoietic stem cell transplantation in July 2002. She has been in a complete remission for 11 months.

RT–PCR

Total RNA (4 g) was reverse transcribed to cDNA in a total volume of 33 l with random primer by using the Ready-To-Go You-Prime First-Strand Beads (Pharmacia Biotech, Tokyo, Japan). Using 1 l of the cDNA, PCR amplification was performed to confirm the MLL-LAF4 transcripts in 20 1 of reaction mixture consisting of 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl₂, 0.001% (w/v) gelatin, 20 pmol of each primer, 200 M each dNTP, and 1 unit of Taq polymerase (Applied Biosystems). After 35 rounds of PCR (30 s at 94°C, 30 s at 55°C, and 1 min at 72°C), 5 l of the PCR product was electrophoresed in a 3% agarose gel. The sequences of primers used were as follows: ALL-5S, 5'-TACTACAGGACCGCCAAGAA-3'; LAF4-2AS, 5'-TGAGCTGCCTGCTGTTCATT-3'; LAF4-2s, 5'-AAGGCCAAGCTCTCCAAGTT-3'; ALL-9A, 5'-GGAAGGGCTCACAACAGAC-3'; LAF4-4-5S, 5'-GAGGAGAGTAGATCTGGA-3'; ALL-8A, 5'-CTCCCA- TCTCCCACACATTT-3'; LAF4-1AS, 5'-GCTTAAGGTCA- TCTTCCAGC-3'; AF4-1S, 5'-AGAAGGAAAGACGCAACCAG-3'; AF4-1A, 5'-TGCCATGAACCTTGCTGCTT-3'; 5-4S, 5'-CCGGAATGTGCTGCGTATGA-3'; 5-5A, 5'-GGGTCC-TACTGGAGTCCATTT-3'.

Nucleotide sequencing

Nucleotide sequences of PCR products and, if necessary, subcloned PCR products were analysed as described previously (Taki et al., 1999; Ono et al., 2002).

Southern blot analysis

High molecular weight DNA was extracted from the patient's bone marrow cells by proteinase K digestion and phenol/chloroform extraction. DNA (10 g) were digested with BamHI, subjected to electrophoresis on 0.7% agarose gel, and transferred to cDNA probes that were ³²P-labeled by the random hexamor method. A 0.9-kb BamHI fragment (designated probe x) derived from MLL cDNA was used as a probe (Taki et al., 1999).

FISH analysis

FISH analysis of the patient's leukemic cells using BAC clone RP11-436F6 was carried out as described previously (Taki et al., 1999; Ono et al., 2002). We mapped this BAC clone to leukemic cells together with a whole-chromosome painting probe for chromosome 11 (WCP11) (Coatasome 11, digoxigenin-labeled, Oncor).

Cell lines

To analyse the expression of the LAF4 genes as compared to AF4 and AF5q31 genes, we used 39 cell lines as follows (Hayashi et al., 2000; Niitsu et al., 2001; Shibuya et al., 2001; Taketani et al., 2002): three B-precursor ALL cell lines with t(11;19) (KOCL-33, KOPN-1, KOCL-44); five B-precursor ALL cell lines with t(4;11) (MV4;11, RS4;11, KOCL45, KOCL-69, KOCL-58); three B-precursor ALL cell lines (P30/OHK, LC4-1, NALM-26); six B-ALL cell lines (BALM-9, BALM-13, BALM-14, BJAB, DAUDI, NAMALLA); five T-ALL cell lines (RPMI-8402, MOLT-14, KOPT-K1, THP-6, DVD-41); seven AML cell lines (YNH-1, ML-1, KASUMI-3, KG-1, P39/TSU, inv-3, NB4); five CML cell lines (MOLM-1, MOLM7, TS9;22, SS9;22, K-562); five EBV-B cell lines derived from normal healthy adult peripheral lymphocytes.

Northern blot analysis

Multiple human tissue Northern blots (CLONTECH) were hybridized with ³²P-labeled 0.9-kb LAF4 cDNA probe, which covered nucleotides 48–942, 0.6-kb AF4 cDNA probes, which covered nucleotides 476–1117 (Genbank accession number, L13773), and 0.6-kb AF5q31 cDNA probe, which covered nucleotides 1135–1735, respectively.

References

1. Ferrando AA and Look AT. (2000). Semin. Hematol., 37, 381–395. | Article | PubMed | ISI | ChemPort |
2. Gecz J, Gedeon AK, Sutherland GR and Mulley JC. (1996). Nat. Genet., 13, 105–108. | Article | PubMed | ISI | ChemPort |
3. Gu Y, Nakamura T, Alder H, Prasad R, Canaani O, Cimino G, Croce CM and Canaani E. (1992). Cell, 71, 701–708. | Article | PubMed | ISI | ChemPort |
4. Gu Y, Shen Y, Gibbs RA and Nelson DL. (1996). Nat. Genet., 13, 109–113. | Article | PubMed | ISI | ChemPort |
5. Hayashi Y. (2000). Semin. Hematol., 37, 368–380. | Article | PubMed | ChemPort |
6. Hayashi Y, Honma Y, Niitsu N, Taki T, Bessho F, Sako M, Mori T, Yanagisawa M, Tsuji K and Nakahata T. (2000). Cancer Res., 60, 1139–1145. | PubMed |
7. Heerema NA, Sather HN, Ge J, Arthur DC, Hilden JM, Trigg ME and Reaman GH. (1999). Leukemia, 13, 679–686. | Article | PubMed | ISI | ChemPort |
8. Hunger SP, Tkachuk DC, Amylon MD, Link MP, Carroll AJ, Welborn JL, Willman CL and Cleary ML. (1993). Blood, 81, 3197–3203. | PubMed | ISI | ChemPort |
9. Isnard P, Core N, Naquet P and Djabali M. (2000). Blood, 96, 705–710. | PubMed | ISI | ChemPort |
10. Ma C and Staudt LM. (1996). Blood, 87, 734–745. | PubMed | ISI | ChemPort |
11. Niitsu N, Hayashi Y, Sugita K and Honma Y. (2001). Br. J. Haematol., 112, 315–326. | Article | PubMed | ISI | ChemPort |
12. Ono R, Taki T, Taketani T, Kawaguchi H, Taniwaki M, Okamura T, Kawa K, Hanada R, Kobayashi M and Hayashi Y. (2002). Cancer Res., 62, 333–337. | PubMed | ISI | ChemPort |
13. Pui CH. (1995). N. Engl. J. Med., 332, 1618–1630. | Article | PubMed | ISI | ChemPort |
14. Pui CH, Gaynon PS, Boyett JM, Chessells JM, Baruchel A, Kamps W, Silverman LB, Biondi A, Harms DO, Vilmer E, Schrappe M and Camitta B. (2002). Lancet, 359, 1909–1915. | Article | PubMed | ISI |
15. Reichel M, Gillert E, Angermuller S, Hensel JP, Heidel F, Lode M, Leis T, Biondi A, Haas OA, Strehl S, Panzer-Grumayer ER, Griesinger F, Beck JD, Greil J, Fey GH, Uckun FM and Marschalek R. (2001). Oncogene, 20, 2900–2907. | Article | PubMed | ISI | ChemPort |
16. Rowley JD. (1998). Annu. Rev. Genet., 32, 495–519. | Article | PubMed | ISI | ChemPort |
17. Rubnitz JE, Link MP, Shuster JJ, Carroll AJ, Hakami N, Frankel LS, Pullen DJ and Cleary ML. (1994). Blood, 84, 570–573. | PubMed | ISI | ChemPort |
18. Shibuya N, Taki T, Mugishima H, Chin M, Tsuchida M, Sako M, Kawa K, Ishii E, Miura I, Yanagisawa M and , Hayashi Y. (2001). Genes Chromosomes Cancer, 32, 1–10. | Article | PubMed | ChemPort |
19. Super HJ, McCabe NR, Thirman MJ, Larson RA, Le Beau MM, Pedersen-Bjergaard J, Philip P, Diaz MO and Rowley JD. (1993). Blood, 82, 3705–3711. | PubMed | ISI | ChemPort |
20. Taketani T, Taki T, Shibuya N, Ito E, Kitazawa J, Terui K and Hayashi Y. (2002). Cancer Res., 62, 33–37. | PubMed |
21. Taki T, Ida K, Bessho E, Hanada R, Kikuchi A, Yamamoto K, Sako M, Tsuchida M, Seto M, Ueda R and Hayashi Y. (1996). Leukemia, 10, 1303–1307. | PubMed | ISI | ChemPort |
22. Taki T, Kano H, Taniwaki M, Sako M, Yanagisawa M and , Hayashi Y. (1999). Proc. Natl. Acad. Sci. USA, 96, 14535–14540. | Article | PubMed | ChemPort |
23. Tkachuk DC, Kohler S and Cleary ML. (1992). Cell, 71, 691–700. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We express our appreciation to Shoko Sohma, Hisae Soga, Yumiko Taketani (University of Tokyo) and Yukiko Konno (Dokkyo University School of Medicine) for technical assistance. We also thank Dr Kanji Sugita (Department of Pediatrics, Yamanashi University School of Medicine, Yamanashi, Japan) for providing ALL (KOCL-33, 44, 45) cell lines, and Dr Yoshinobu Matsuo (Hayashibara Biochemical Laboratories, Inc., Fujisaki Cell Center, Okayama, Japan) for providing varieties of ALL cell lines.