Infant acute leukemia occurs within the first year of life and displays distinct biologic and clinical features with a poor prognosis.1–4 Around 70% to 90% of patients harbor rearrangements of the mixed-lineage leukemia (MLL) gene at the chromosomal region 11q23.5,6 The acute lymphoblastic leukemia (ALL) subtype of infant acute leukemia predominantly includes the t(4;11) (q21;q23) region with the MLL/AF4 fusion gene, followed by the t(11;19) (q23;p13) region with the MLL/ENL fusion gene and the t(9;11)(p22;q23) region with the MLL/AF9 fusion gene.7,8 Recent gene expression profiling studies indicate that MLL-rearranged (MLL-R) ALLs are arrested at an early lymphoid progenitor stage of development.9,10 Among these highly distinct expression patterns in MLL-R, a total lack of CD10 expression is a hallmark and provides a predictable signature of MLL rearrangements.9,11
CD10, the neutral endopeptidase 24.11, is a regulator of B-cell growth and proliferation.12 The enzyme degrades a number of bioactive peptides with various functions that depend on the cell type or tissue of origin. CD10 is expressed in a biphasic pattern that correlates with specific stages of B-cell development. CD10 is detected on precursor B (pro-B and pre-B) cells and lymph node germinal center B cells but is not detected on lymphoid progenitor cells, early pro-B cells, mature B cells (naive B cells and mantle cells), or plasma cells.13 This biphasic pattern is generally maintained in malignant counterparts.13
Immunoglobulin (Ig) gene rearrangements are regarded as a molecular fingerprint of B-cell development.14 CD10 is believed to be expressed during the first stages of Ig heavy chain (IgH) rearrangement and can be coexpressed with surface μ-Vpre-B or μ-λ5 pre−B-cell receptor.15 Several lines of evidence indicate a high frequency of complete IgH rearrangements (VDJH) in CD10-negative pro-B ALL.16 Ig gene rearrangement patterns were related to the presence and type of MLL rearrangement but not to the age at diagnosis or to ALL-CD10 expression.16,17 Discordance between the genotype and the immunophenotype of infant ALL (iALL) with respect to cellular maturity suggested an aberrant process in the steps of immunophenotypic differentiation or a secondary down-regulation of CD10 expression associated with MLL rearrangements.
Methylation of promoter-associated cytosine-phosphate-guanine (CpG) islands is an epigenetic modification of DNA and has been associated with gene silencing and malignant transformation.18,19 An inverse correlation was previously observed between methylation of the CD10 promoter and CD10 expression in adult20 and pediatric ALL21 as well as in human prostate cancer.22 However, data on the frequency of methylation in the various types of ALL, particularly in iALL with MLL-R, remain scarce. We therefore performed methylation analysis of the entire promoter region of the CD10 gene in iALL to investigate epigenetic mechanisms responsible for CD10 negativity. We describe dense methylation of the CD10 promoter with particular emphasis on the methylation of transcription factor binding sites within the CD10 promoter (type 1 and 2 regulatory regions) in CD10-negative MLL/AF4 iALL carrying a productive rearrangement of the VDJH allele.
MATERIALS AND METHODS
Peripheral blood or bone marrow samples from MLL/AF4 infant leukemia (3 samples), acute myeloid leukemia (AML) (2 samples, AML1/ETO and MLL/AF9, respectively) and a sample of infant pre-B ALL (MLL-germline, MLL-G) were obtained after informed parental consent and with approval of the Ethics Committee of the Kanazawa University Hospital. The pre-B ALL cell line (CL; HAL-01) was from RIKEN Bank (Tsukuba, Ibaraki, Japan). Mononuclear cells were isolated from peripheral blood or bone marrow samples by Ficoll-Paque density gradient centrifugation. DNA was extracted and purified according to the manufacturer's instructions using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany).
The immunophenotype of leukemic cells was assessed by flow cytometry using a dual-laser FACSCalibur (Becton Dickinson, San Diego, CA). The resulting data were analyzed using the CELL Quest software (Becton Dickinson). Immunophenotypic subclassification was done according to the guidelines of the European Group for the Immunological Characterization of Leukemias: Pro-B ALL, CD19+/CD10−/cytoplasmic (cy) IgM−/surface (s) IgM−; common ALL, CD19+/CD10+/cyIgM−/sIgM− and Pre-B ALL, CD19+/CD10+/cyIgM+/sIgM−. Criteria for immunophenotypic marker positivity were: expression on ≥20% or ≥10% of blasts for cell surface and cytoplasmic markers, respectively.
Detection of Rearranged Leukemia-associated Genes
The presence of t(4;11)(q21;q23) with the MLL/AF4 fusion gene, t(9;11)(p22;q23) with the MLL/AF9 fusion gene and t(8;21)(q22;q22) with the AML1/ETO fusion gene was demonstrated by one or more of the following methods: metaphase cytogenetic analysis, polymerase chain reaction (PCR) with reverse transcription and fluorescence in situ hybridization as part of the diagnostic work-up.
PCR and Sequence Analysis of IgH Gene Rearrangements
PCR amplification of IgH-VDJ rearrangements was performed in a Thermocycler (GeneAmp PCR system 9700, Applied Biosystems) using a series of 7 VH family FR1 consensus primers and a JH consensus primer by modification of a method previously described.23 Clonal PCR products were purified using QIAquick gel extraction kits (QIAGEN, Hilden, Germany), directly sequenced using the BigDye-Terminator v3.1 cycle sequencing kit and were analyzed on a 3100 sequencer (Applied Biosystems). VH, DH, JH segments and their family members were identified with the closest matching known human germ line genes using the IGBlast search (http://www.ncbi.nlm.nih.gov/igblast/, National Center for Biotechnology Information, Bethesda, MD). The criteria used for DH gene determination were: a minimal homology of 6 matches in a row or 7 matches interrupted by 1 mismatch.
Bisulfite Modification of DNA, PCR Amplification, and Sequence Analyses of the CD10 Promoter Regions
All methylation studies were done by bisulfite modification of DNA, which converts all unmethylated CpG sites to uracil-phosphate-guanosine, leaving methylated CpGs intact. Bisulfite treatment of genomic DNA was done using MethylEasy Xceed (Human Genetic Signatures, North Ryde, Australia). The DNA was then amplified with the following gene-specific primers: Type 1-1, sense 5′-TTTATAGGGGAGAGATGGAG-3′; antisense 5′-ACCAAAAAAAAAAACAATTCC-3′; Type 1-2, sense 5′-TTTTTGGTTTTGTAGTGGT-3′; antisense 5′-CAATCAAACTTCACCACTTA-3′; Type 2-1, sense 5′-TTYGGTTTTAGTTTGGAGTT-3′; antisense 5′-CCCTTTAAACCTTTCTCCCT-3′; Type 2-2, sense 5′-TGTGGGTGTGGGTTGGGGGATG-3′; antisense 5′-CCCTCCCCATGCCCATCCCAC-3′. Following sequencing of the amplified DNA, methylated CpGs were identified by visual inspection of the sequencing traces in the electropherograms and by comparison with sequencing traces of the CD10 gene sequence relative to the reference GenBank sequence (accession number X79438).
Clinical Features, Immunophenotype, and MLL Rearrangements of the Leukemic Samples
The clinical features of 7 samples and the characteristics of their leukemic cells are described in Table 1. Four of the 5 samples (<12 mo of age) of iALL (iALL-1 to iALL-4, and iAML) and a pediatric AML (pAML) sample (5 y of age) were studied at diagnosis. A case with congenital ALL was analyzed at day 1 of age for initial diagnosis (iALL-1) and at 15 months of age for relapse (iALL-2). A pre-B CL established from a 17-year-old female with pre-B ALL was also studied. Three of the 4 iALL samples (iALL-1 to iALL-3) were t(4;11)-positive ALL with the MLL/AF4 fusion gene and 1 sample (iALL-4) showed a normal karyotype. The iAML and pAML samples were t(9;11)-positive with the MLL/AF9 fusion gene and t(8;21)-positive with the AML1/ETO fusion gene, respectively. CD10 expression, determined by the flow cytometric immunophenotyping, was negative (≤20%) in all 3 MLL/AF4-positive ALLs (iALL-1 to iALL-3) with a pro-B ALL immunophenotype (CD19+/CD34+/cyIgM−/sIgM−). In contrast, MLL-G ALLs (iALL-4 and pre-B CL) expressed CD10 and the common ALL immunophenotype (CD19+/CD34−/cyIgM−/sIgM−).
IgH Gene Rearrangements (VDJH) in iALL
Sequence analyses of PCR-amplified VDJH segments are summarized in Table 2. In 2 CD10-negative iALLs (iALL-1 and iALL-2) and a CD10-posivitive iALL (iALL-4), 2 major clonal VDJH rearrangements were identified, in which each sequence contained either an out-of-frame joining (a nonproductive rearrangement) or an in-frame joining without a stop codon (a productive rearrangement). In iALL-3, clonal PCR products were not detected after repeated amplification, indicating that the IgH genes remained germline or that there was only DJ joining. These recombination processes were consistent with a hierarchical order of IgH gene rearrangements in normal B precursor cells.14 In addition, in cases with congenital leukemia, biclonality of the leukemic cells was shown by a difference in the usage of VH and DH gene family members in nonproductive rearrangements determined at diagnosis (iALL-1) and relapse (iALL-2). These results indicated that CD10 negativity was not correlated with the developmental status of IgH gene rearrangements in iALL.
Methylation of the CD10 Promoter in iALL
Figure 1 contains electropherograms of base sequences, from nt −211 to nt −177, for 2 representative samples (iALL-2 and pre-B CL) with bisulfite treatment. Cytosine residues other than CpG dinucleotides were all converted to thymine, whereas methylated cytosine residues in CpG dinucleotides were retained as cytosine, indicating that bisulfite treatment was completed. The CD10 gene promoter contains 2 separate regulatory regions that control the transcription of 5′ alternatively spliced CD10 transcripts.24 The type 1 regulatory region contains the exon 1 sequence within nt −821 to nt −326 relative to the major transcription initiation site at the beginning of exon 2. The type 2 regulatory region spans the sequence from nt −326 to nt +105 (Fig. 2).25,26 In addition, 3 functionally active transcription factor binding sites, designated as regions I to III, have been identified in the type 2 regulatory region (Fig. 2B).25 A total of 66 CpG dinucleotides are present within these 2 regulatory regions (24 in the type 1 regulatory region and 42 in the type 2 regulatory region) (Fig. 2 and Table 3). Mapping of the CpG dinucleotides showed that they were densely distributed within the 3′-half portion of the type 1 regulatory region and over the entire type 2 regulatory region (Fig. 2B). At least 33 of these 66 CpG dinucleotides were successfully analyzed by bisulfite sequencing (Table 3 and Fig. 2B). We quantified CpG methylation as the methylated CpG content (%) in a single gene by showing the number of CpG methylation present in the CD10 promoter region. This analysis showed that more than 84% of the identified CpG dinucleotides in all 3 CD10-negative iALL with MLL/AF4 were methylated. Comparison of the 2 regulatory regions revealed no difference in methylation frequency. In contrast, none of the CpG dinucleotides were methylated in the CD10-positive ALL (iALL-4 and pre-B CL) and only 2 of the CpG dinucleotides were methylated in pAML samples (Table 3 and Fig. 2B).
Figure 3 illustrates the CD10 promoter methylation status and transcription factor binding sites in the representative iALL-2 sample.27 The 5′ UTR exon 1 sequence contained multiple putative PU.1 binding sites, consensus ets-binding motifs and a cluster of 4 potential Sp1-binding sites (Fig. 3A). The CpG dinucleotides within the Sp1-binding sites, as well as in the surrounding sites were highly methylated. In the type 2 regulatory region (Fig. 3B), 31 of the 42 CpG dinucleotides were analyzed and all 31 were methylated. Notably, 3 putative Sp1-binding sites, as well as functionally active transcription factor binding sites in regions I to III, contained 12 (39%) of these methylated CpG dinucleotides. These results indicated that both type 1 and 2 regulatory regions in the CD10 gene promoter were highly methylated in CD10-negative iALL with MLL/AF4 and provided structural evidence of dense methylation of the transcription factor binding sites.
We have demonstrated dense methylation patterns in both type 1 and 2 regulatory regions of the CD10 promoter in CD10-negative MLL/AF4 iALL bearing a productive, rearranged VDJH allele. In addition, we are the first to describe the precise methylated sites within the CD10 gene promoter.
Two separate regulatory regions have been identified in the CD10 gene promoter.25–27 The type 1 and 2 regulatory regions share common features characterized by the presence of multiple transcription initiation sites and the absence of classic TATA boxes and consensus initiator elements. The purine-rich type 1 regulatory region includes the 5′ UTR exon 1 sequence with multiple putative PU.1 binding sites and consensus ets-binding motifs.27,28 The distribution map of CpG dinucleotides showed a concentration of methylated CpG dinucleotides within the 3′-half portion of the type 1 regulatory region (Fig. 2A), in which a cluster of 4 potential Sp1-binding sites is located (Fig. 3A). Clustering of Sp1-binding sites and their cooperative activation of transcription has been reported for several genes with a TATA-less promoter.28 Therefore, Sp1-binding sites observed in exon 1 of the CD10 gene are anticipated to contribute to type 1 promoter activity. Indeed, a factor(s) immunologically related to Sp1 specifically binds to the most upstream GC-rich sequence (ggtggg) and is required for optimal promoter activity.26 Bisulfite sequencing revealed that the CpG dinucleotides in the putative Sp1-binding sites were almost all methylated in 3 samples of iALL with MLL/AF4. The sequence upstream of exon 1 that contains multiple putative PU.1 sites and ets-binding motifs also showed a positive influence on type 1 CD10 promoter activity.27 However, it is unlikely that methylation is involved in the action of these regulatory elements as there are no CpG dinucleotides in the consensus sites.
The GC-rich type 2 regulatory region identified in the intron upstream of exon 227 contains numerous CpG dinucleotides (Fig. 2). Strikingly, 31 CpG dinucleotides were distributed in the type 2 regulatory region and all were methylated in a sample of iALL (iALL-2) (Fig. 2). Such high frequency of methylation was consistently seen in additional 2 samples of CD10-negative ALL (Fig. 2). The CD10 type 2 regulatory region contains multiple putative Sp1-binding sites, a potential consensus retinoblastoma control element and an inverted CCAAT box.25,27 Up-regulation of Sp1-mediated transcription by retinoblastoma protein through the retinoblastoma control element has been reported.29 Recently, 3 functionally active transcription factor binding sites, designated as regions I to III, have been identified in the type 2 regulatory region.25 CBF/NF-Y was shown to bind the inverted CCAAT box in region I and is responsible for tissue-specific CD10 expression.25 Bisulfite sequencing indicated complete methylation of the CpG dinucleotides in the 3 putative Sp1-binding sites and within regions I to III in iALL with MLL/AF4 (Fig. 3B). Specific methyl-CpG dinucleotides within the promoter may inhibit trans-factors that preferentially bind to sequences containing CpG dinucleotides by limiting DNA access to the factors or by conformational alteration of DNA structure or they may act as a guide for methyl CpG-binding proteins (MeCP-1 and MeCP-2) that modulate transcriptional activity.30–32 Thus, structural evidence of dense methylation in 2 regulatory regions of the promoter suggests that methylated transcription factor binding sites contribute to CD10 silencing as an epigenetic mechanism. As our study contained no MLL/ENL or MLL/AF9 ALL samples, extended analyses including MLL-R subtypes are required to specify MLL-R involvement in methylation of the CD10 promoter.
Three types of CD10 transcripts, types 1, 2A, and 2B, have been identified resulting from alternative splicing of 5′ untranslated regions; the type 1 regulatory region promotes type 1 transcripts from exon 1 and the type 2 regulatory region promotes types 2A and 2B transcripts initiated within exon 2 in a tissue-specific manner.27 Type 2 transcripts are more abundant than type 1 transcripts in a variety of cell types, whereas some CD10-positive pre−B-cell ALL CLs have also been shown to contain high levels of type 1 transcripts.27 Although the relative abundance of type1 and type 2 transcripts was not examined in our clinical samples, hypermethylation of both of the regulatory regions is consistent with a total lack of CD10 expression in iALL with MLL/AF4.
Taylor et al13 have shown that the CD10 promoter region is consistently not methylated in normal lymphocytes at early and late stages of maturation regardless of the CD10 expression level, whereas in B-cell malignancies methylation of the CD10 promoter is common at specific stages of maturation and is frequently associated with absence of CD10 expression. These data suggest that the biphasic pattern of CD10 expression in normal differentiation of B cells is controlled by tissue-specific and stage-specific regulatory mechanisms other than DNA methylation. However, CD10 methylation plays a role in lymphoid malignancies by silencing CD10 expression. Further investigations are required for complete elucidation of the role of methylation in CD10 expression because epigenetic modulation involves not only DNA methylation but also histone modifications and chromatin remodeling that interact in an epigenetic network.33,34
In conclusion, our results show structural evidence of DNA promoter methylation suggesting a contribution to CD10 silencing as an epigenetic mechanism. Functional analyses such as transfection of MLL fusion genes are required to confirm the precise involvement of MLL-R in methylation. Further investigation of epigenetic network interactions associated with MLL will provide an additional insight into the aberrant transcriptional program in MLL leukemia.
1. Biondi A, Cimino G, Pieters R, et al. Biological and therapeutic aspects of infant leukemia. Blood. 2000;96:24–33.
2. Lauer SJ, Camitta BM, Leventhal BG, et al. Intensive alternating drug pairs after remission induction for treatment of infants with acute lymphoblastic leukemia: a Pediatric Oncology Group Pilot Study. J Pediatr Hematol Oncol. 1998;20:229–233.
3. Dordelmann M, Reiter A, Borkhardt A, et al. Prednisone response is the strongest predictor of treatment outcome in infant acute lymphoblastic leukemia. Blood. 1999;94:1209–1217.
4. Reaman GH, Sposto R, Sensel MG, et al. Treatment outcome and prognostic factors for infants with acute lymphoblastic leukemia treated on two consecutive trials of the Children's Cancer Group. J Clin Oncol. 1999;17:445–455.
5. Greaves MF. Infant leukaemia biology, aetiology and treatment. Leukemia. 1996;10:372–377.
6. Pui CH, Kane JR, Crist WM. Biology and treatment of infant leukemias. Leukemia. 1995;9:762–769.
7. Huret JL, Dessen P, Bernheim A. An atlas of chromosomes in hematological malignancies. Example: 11q23 and MLL partners. Leukemia. 2001;15:987–989.
8. Meyer C, Schneider B, Jakob S, et al. The MLL recombinome of acute leukemias. Leukemia. 2006;20:777–784.
9. Armstrong SA, Staunton JE, Silverman LB, et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet. 2002;30:41–47.
10. Kohlmann A, Schoch C, Dugas M, et al. New insights into MLL gene rearranged acute leukemias using gene expression profiling: shared pathways, lineage commitment, and partner genes. Leukemia. 2005;19:953–964.
11. Attarbaschi A, Mann G, Konig M, et al. Mixed lineage leukemia-rearranged childhood pro-B and CD10-negative pre-B acute lymphoblastic leukemia constitute a distinct clinical entity. Clin Cancer Res. 2006;12:2988–2994.
12. Cutrona G, Tasso P, Dono M, et al. CD10 is a marker for cycling cells with propensity to apoptosis in childhood ALL. Br J Cancer. 2002;86:1776–1785.
13. Taylor KH, Liu J, Guo J, et al. Promoter DNA methylation of CD10 in lymphoid malignancies. Leukemia. 2006;20:1910–1912.
14. Alt FW, Oltz EM, Young F, et al. VDJ recombination. Immunol Today. 1992;13:306–314.
15. Bene MC, Faure GC. CD10 in acute leukemias. GEIL (Groupe d'Etude Immunologique des Leucemies). Haematologica. 1997;82:205–210.
16. Jansen MW, Corral L, van der Velden VH, et al. Immunobiological diversity in infant acute lymphoblastic leukemia is related to the occurrence and type of MLL gene rearrangement. Leukemia. 2007;21:633–641.
17. Mann G, Cazzaniga G, van der Velden VH, et al. Acute lymphoblastic leukemia with t(4;11) in children 1 year and older: the “big sister” of the infant disease? Leukemia. 2007;21:642–646.
18. Laird PW. Cancer epigenetics. Hum Mol Genet. 2005;14(Spec No 1):R65–R76.
19. Jones PA, Baylin SB. The fundamental role of epigenetic events in cancer. Nat Rev Genet. 2002;3:415–428.
20. Garcia-Manero G, Daniel J, Smith TL, et al. DNA methylation of multiple promoter-associated CpG islands in adult acute lymphocytic leukemia. Clin Cancer Res. 2002;8:2217–2224.
21. Garcia-Manero G, Jeha S, Daniel J, et al. Aberrant DNA methylation in pediatric patients with acute lymphocytic leukemia. Cancer. 2003;97:695–702.
22. Usmani BA, Shen R, Janeczko M, et al. Methylation of the neutral endopeptidase gene promoter in human prostate cancers. Clin Cancer Res. 2000;6:1664–1670.
23. van Dongen JJ, Langerak AW, Bruggemann M, et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia. 2003;17:2257–2317.
24. D'Adamio L, Shipp MA, Masteller EL, et al. Organization of the gene encoding common acute lymphoblastic leukemia antigen (neutral endopeptidase 24.11): multiple miniexons and separate 5′ untranslated regions. Proc Natl Acad Sci USA. 1989;86:7103–7107.
25. Ishimaru F, Mari B, Shipp MA. The type 2 CD10/neutral endopeptidase 24.11 promoter: functional characterization and tissue-specific regulation by CBF/NF-Y isoforms. Blood. 1997;89:4136–4145.
26. Sezaki N, Ishimaru F, Tabayashi T, et al. The type 1 CD10/neutral endopeptidase 24.11 promoter: functional characterization of the 5′-untranslated region. Br J Haematol. 2003;123:177–183.
27. Ishimaru F, Shipp MA. Analysis of the human CD10/neutral endopeptidase 24.11 promoter region: two separate regulatory elements. Blood. 1995;85:3199–3207.
28. Saikawa Y, Price K, Hance KW, et al. Structural and functional analysis of the human KB cell folate receptor gene P4 promoter: cooperation of three clustered Sp1-binding sites with initiator region for basal promoter activity. Biochemistry. 1995;34:9951–9961.
29. Kim SJ, Onwuta US, Lee YI, et al. The retinoblastoma gene product regulates Sp1-mediated transcription. Mol Cell Biol. 1992;12:2455–2463.
30. Boyes J, Bird A. Repression of genes by DNA methylation depends on CpG density and promoter strength: evidence for involvement of a methyl-CpG binding protein. EMBO J. 1992;11:327–333.
31. Majumder S, Kutay H, Datta J, et al. Epigenetic regulation of metallothionein-i gene expression: differential regulation of methylated and unmethylated promoters by DNA methyltransferases and methyl CpG binding proteins. J Cell Biochem. 2006;97:1300–1316.
32. Michelotti GA, Brinkley DM, Morris DP, et al. Epigenetic regulation of human alpha1d-adrenergic receptor gene expression: a role for DNA methylation in Sp1-dependent regulation. FASEB J. 2007;21:1979–1993.
33. Galm O, Herman JG, Baylin SB. The fundamental role of epigenetics in hematopoietic malignancies. Blood Rev. 2006;20:1–13.
34. Santoro R, Grummt I. Epigenetic mechanism of rRNA gene silencing: temporal order of NoRC-mediated histone modification, chromatin remodeling, and DNA methylation. Mol Cell Biol. 2005;25:2539–2546.