BRIEF COMMUNICATION Deletions of Chromosome 21 Restricted to the Leukemic Cells of Children with Down Syndrome and Leukemia HM Kempski1, JM Chessells2 and BR Reeves1

Home , Chromosome 21

BRIEF COMMUNICATION Deletions of chromosome 21 restricted to the leukemic cells of children with Down syndrome and leukemia HM Kempski1, JM Chessells2 and BR Reeves1

1LRF Centre for Childhood Leukaemia, Department of Molecular Haematology, Institute of Child Health, London; and 2Department of Haematology and Oncology, Great Ormond Street Hospital for Children NHS Trust, London, UK

Down syndrome (DS) is associated with an increased risk of Probes developing hematological malignancies, but the basis for this predisposition is so far unknown. Using fluorescence in situ hybridization with a panel of locus-specific probes on normal A whole chromosome paint and alpha-satellite probe for chro- and leukemic metaphases, we have found long-arm interstitial mosome 21 and a locus-specific probe for D21S19 were deletions of one of the chromosome 21s in the leukemic cells obtained from Cambio (Cambridge, UK). Further probes for from five patients with DS and leukemia. This finding provides the loci D21S65, D21S55, and a cosmid contig for the loci strong evidence for a gene or genes present on chromosome D21S1219 and D21S1220 were obtained from Appligene 21 having an important function in the development of leukemia Oncor (Co. Durham, UK). A YAC probe C4C105 covering the in individuals with Down syndrome. ′ Keywords: Down syndrome; leukemia; chromosome 21 deletions 3 end of the AML1 gene was provided by Marie-Laure Yaspo and Franco Calabi (Institute of Child Health, London, UK). Figure 1 shows a map of chromosome 21, indicating the rela- tive positions of the locus-specific probes used in this study. Introduction

Individuals with trisomy 21, Down syndrome (DS), have an Fluorescence in situ hybridization (FISH) increased risk of developing leukemia,1 but the basis for this association is uncertain. Recognition of non-random chromo- FISH was carried out on metaphases prepared from the same some abnormalities has been instrumental in identifying genes samples used for conventional cytogenetic analysis. Probes involved in the development of several different types of leu- were hybridized in pairs (one biotin-labeled, the other digoxi- kemia but up to now no acquired chromosome abnormalities genin labeled) and for each patient a hybridization experiment have been found to be particularly associated with leukemias using the complete panel of probes was processed at the same arising in DS patients.2,3 time. Control metaphases from a normal individual were also In the course of conventional G-band analysis of a series of hybridized as an internal control for hybridization efficiency. DS patients with leukemia, we noticed that in two cases one The whole chromosome 21-specific paint, alpha-satellite copy of chromosome 21 in the leukemic cells appeared shorter in the long-arm. Fluorescence in situ hybridization (FISH) with a whole chromosome paint for 21 failed to dem- onstrate evidence of translocation, but using a panel of locus- specific probes we have found interstitial deletions in both cases and in the malignant cells from three additional cases in which no deletion was suspected. This finding suggests an important role for gene(s) on chromosome 21 in the genesis or progression of the leukemic state in patients with DS and leukemia.

Materials and methods

Patients

Five children with DS and leukemia, one with acute lymphoblastic leukemia (ALL) and four with acute myeloid leukemia (AML), were selected for this study on the basis of avail- ability of sufﬁcient good-quality metaphases (see below). Bone marrow chromosomes obtained at the time of diagnosis of leukemia were processed by standard methods. Karyotypes are described according to ISCN (1995).4

Correspondence: HM Kempski, LRF Centre for Childhood Leukaemia, Department of Molecular Haematology, Institute of Child Health, 30 Guilford Street, London, WC1N 1EH, UK Figure 1 Ideogram of chromosome 21 showing positions of loci Received 31 March 1997; accepted 10 July 1997 tested in this study. Brief communication HM Kempski et al 1974 probe and locus-specific probes for D21S65, D21S55, D21S19, D21S1219 and D21S1220 were prepared according to the manufacturer’s instructions. C4C10 was nick-translated using Biotin-14 dATP (Gibco BRL, Renfrewshire, UK). Slides were denatured at 70°C in 70% formamide, 2 × SSC for 2 min and dehydrated in an ethanol series. Probes were then placed on the slides, sealed and incubated overnight at 37°C. All post-hybridization washes for the whole chromosome paint, alpha-satellite probe and commercial locus-specific probes were performed according to procedures recommended by the manufacturers. C4C10 was washed at 43°C for 15 min in 60% formamide, 2 × SSC and for a further 15 min in 2 × SSC. Signal detection with fluorescein avidin- and biotinylated anti-avidin-labeled antibodies (Vector, Peter- borough, UK) and anti-digoxigenin-rhodamine conjugate (Boehringer, Lewes, UK) were performed according to proto- cols recommended by the manufacturers. Cells were viewed using a Zeiss Axioskop microscope equipped with a CCD camera (Photometrics, Tucson, AZ, USA) connected to a Mac- intosh Quadra Computer. Image analysis was effected using a SmartCapture software (Digital Scientific, Cambridge, UK).

Analysis

Only metaphases showing clear signals and no/very low back- ground were assessed in order to eradicate artifactual misin- terpretation. In each case and for each probe, 10 normal and 25 leukemic metaphases were evaluated by visual assessment and measurement of signal intensity/size using the Graph polygon program in SmartCapture software. In all cases, some probes were found to be partially deleted (ie had reduced signal intensities). In each instance, the ratios of reduced/non- reduced signals remained constant to within a 10% value for each patient, although the sizes of the partial deletions varied from case to case. Examples of graph polygons illustrating normal, reduced and absent signals from case 2 are shown in Figure 2.

Southern blot analysis

Southern blot hybridization was carried out using a composite 21A/21b probe, identifying both the 11 kb BamHI fragment harbouring AML1 exon 5 and the 20 kb BamHI fragment harbouring AML1 exon 6 as previously described.6

Results

Conventional cytogenetic analysis

Clinical and cytogenetic results are shown in Table 1. In each Figure 2 Ratio proﬁles generated using the Graph polygon Smart- case, metaphases with the normal constitutional karyotype capture software showing examples of results from case 2 for + (a) probes D21S19 (biotin-labeled) and D21S1219/D21S1220 (47,XX or XY, 21c) were found, in addition to leukemic cells (digoxigenin-labeled) showing equal signal intensities for both probes, with clonal chromosome abnormalities. These abnormalities (b) probes D21S65 (biotin-labeled) and D21S1219/D21S1220 were used as markers of the leukemic cells in the FISH (digoxigenin-labeled) showing a reduced signal for D21S65, and analysis. (c) probes C4C10 (biotin-labeled) and D21S1219/D21S1220 (digoxigenin-labeled) showing an absent signal for C4C10.

FISH probes we found long-arm interstitial deletions of one chromosome 21 in the leukemic cells from each patient. In Using a whole chromosome paint, we found no evidence for patients 2 and 3, the deletions were detectable in some G- translocations involving chromosome 21 in the normal or leu- banded metaphases found in their leukemic clones. In no kemic cells from any case. However, using the locus-speciﬁc instance was a deletion found in the non-leukemic cells. The Brief communication HM Kempski et al 1975 Table 1 Clinical and cytogenetic data for ﬁve patients with Down syndrome and leukemia

Case Age Sex Diagnosis Outcome Karyotype (years) (years)

1 3.3 F ALL/pre-B A 2.1 47,XX,+21c [5] 48,XX,+X,+21c [25] 2 2.0a F AML/M7 D 1.5 47,XX,+21c [2] 47,XX,del(6)(q13q21),+21c [19] 3 1.3 M AML/M7 A 0.4 47,XY,+21c [2] 47,XY,del(6)(q15q21),del(13) (q21q21.1),+21c [23] 4 1.1 F AML/M7 A 0.8 47,XX,+21c [6] 48,XX,+2,t(4;13)(q21;q34), ins(11;4)(q23;q21q25),+21c [24] 5 1.3 M AML/M7 D 0.9b 47,XY,+21c [20] 47,XY,t(1;9)(p3?4;q34),+21c [10] aDiagnosed as transient abnormal myelopoiesis (TAM) aged 2 months. bAccidental death from unrelated cause. A, alive; D, deceased.

results are shown in Figure 3 and representative examples in arising in DS patients with those in non-DS individuals reveals Figures 4–7. a number of differences, including age distribution and The breakpoints in the four patients with AML (cases 2–5) phenotypes, but there are no particular patterns of acquired occurred in the interval defined by D21S65 and D21S55, with chromosome abnormalities which set the DS patients apart.2,3 the smallest deletion found in patient 5, involving only locus The neonatal disorder of transient abnormal myelopoiesis D21S55. In the patient with ALL (case 1) two deletions were (TAM), is found almost exclusively in DS or DS mosaics, and found, breakage having occurred in two regions of the long in these cases always involves cells with trisomy 21. TAM usu- arm, the more proximal involving C4C10 and the more distal ally resolves spontaneously but approximately 25% of patients region being between D21S19 and the telomere. may present up to 3 years later with AML.3 The increased risk of hematological malignancy and the involvement of trisomy 21 cells in DS mosaics with TAM has suggested that trisomy Southern blot analysis 21 is a predisposing factor and that other events may be required before leukemia develops.1 In order to rule out an involvement of the AML1 locus at Our finding of chromosome 21 long-arm deletions restric- 21q22, Southern blot analysis was carried out on BamHI ted to the malignant cells of five children with DS, is strong digested DNA from the two patients (cases 3 and 5) from evidence for a gene(s) on the chromosome being important in whom presentation DNA samples were available. No the development of their leukemias. Patient 2 in the present rearrangement was found. study had TAM in the neonatal period but unfortunately no sample was available for chromosome analysis at that time. Loss of heterozygosity, as a result of chromosome loss or Discussion deletion, is considered a hallmark of the involvement of a gene whose loss of function is important in the initiation or It has been recognized for many years that individuals with progression of malignancy.7 In this regard, the case of ALL Down syndrome have an increased risk of developing leuke- arising in a child with DS described in detail by Rogan et al8 1 mias, but not other malignancies. Comparison of leukemias is of interest. Constitutionally, the patient had one paternal and two maternal copies of chromosome 21, with maternal isodisomy for part of the long-arm. In the leukemic clone, the paternal chromosome 21 was lost, reducing to homozygosity all loci distal to D21S262, which maps close to D21S65.9 Such a rearrangement could expose a recessive mutation or gene deletion. Partial deletion of one copy of chromosome 21 in trisomic leukemia cells, such as we have found, could result in a similar situation, if there was homozygosity for a tumori- genic mutation or gene deletion on the remaining chromosomes. Lack of material prevented us from determining the parental origins of the chromosomes in any of our cases, but such an analysis will be a priority in our future studies. The FISH probes we have used so far were chosen to include the region containing genes responsible for many of the phenotypic features of DS10 as well as the AML1 gene which is known to be involved in translocations found in 11–15 Figure 3 Deletion map for 21q loci in leukemic cells from five many myeloid and lymphoid leukemias. The AML1 locus patients with Down syndrome and leukemia. was included in the deleted segments in cases 1–4 and Brief communication HM Kempski et al 1976

Figure 4 (a) Non-leukemic cell from patient 2. Chromosome 21s identiﬁed with an alpha-satellite probe (yellow). Note three signals for locus D21S65 (red). (b) Black and white image of cell shown in (a). Figure 5 (a) Leukemic cell from patient 2 showing reduced signal for D21S65 (arrowed). (b) Black and white image of cell shown in (a). Figure 6 (a) Normal cell from patient 1 with distinct signals for loci D21S19 (yellow) and D21S1219/D21S1220 (red) on each chromosome 21. (b) Black and white image of cell shown in (a). Figure 7 (a) Leukemic cell from patient 1 showing loss of locus D21S19 in one chromosome 21 (arrowed). (b) Black and white image of cell shown in (a).

appeared to be intact in case 5. Southern blot studies were close to D21S559 were not tested. No rearrangements of only possible in cases 3 and 5 and no rearrangements were AML1, ERG or ETS2 were found in the case of AML with pen- found. tasomy 21 studied by Rogan et al.8 Two other possible candidate genes, ERG, which is The deletions we have found were variable in extent, and involved in 16;21 translocations16 and ETS2, which both map in future studies we hope to define the deleted segment more Brief communication HM Kempski et al 1977 closely and seek to identify candidate genes. Patient 5 was Antonarakis SE. Duplication and loss of chromosome 21 in two the most informative, having a partial deletion involving the children with Down syndrome and acute leukemia. Am J Hum D21S55 locus, with the probability of the gene(s) of interest Med Genet 1995; 59: 174–181. 9 Shimizu N, Antonarakis SE, Van Broeckhoven C, Patterson D, Gar- lying between this and the AML1 locus. diner K, Nizetic D, Creau N, Delabar J-M, Korenberg J, Reeves R, Doering J, Chakravati A, Minoshima S, Ritter O, Cuticchia J. Report of the fifth international workshop on human chromosome 21 mapping 1994. Cytogenet Cell Genet 1995; 70: 147–182. Acknowledgements 10 Korenberg JR, Chen X-N, Schipper R, Sun Z, Gonsky R, Gerwehr S, Carpenter N, Daumer C, Dignan P, Disteche C, Graham JM, We thank Dr Franco Calabi for the C4C10 probe and Southern Hugdins L, McGillivray B, Miyazaki K, Ogasawara N, Park JP, blot analysis and Professor Paul Brickell for helpful comments Pagon R, Pueschel S, Sack G, Say B, Schuffenhauer S, Soukup S, Yamanaka T. Down syndrome phenotypes: the consequences of on the manuscript. This work is supported by the Leukaemia chromosomal inbalance. Proc Natl Acad Sci USA 1994; 91: Research Fund (UK). 4997–5001. 11 Erickson P, Gao J, Chang K, Look T, Whisenant E, Raimondi S, Lasher R, Trujillo J, Rowley J, Drabkin H. Identification of breakpoints in t(8;21) acute myelogenous leukemia and isolation of a References fusion transcript AML1/ETO, with similarity to Drosophila seg- mentation gene runt. Blood 1992; 80: 1825–1831. 1 Fong C, Brodeur GM. Down’s syndrome and leukemia: epidemi- 12 Nisson P, Watkins P, Sacchi N. Transcriptionally active chimeric ology, genetics, cytogenetics and mechanisms of leukemogenesis. gene derived from the fusion of the AML1 gene and a novel gene Cancer Genet Cytogenet 1987; 28: 55–76. on chromosome 8 in t(8;21) leukemic cells. Cancer Genet Cyto- 2 Pui C-H, Raimondi SC, Borowitz MJ, Land VL, Behm FG, Pullen genet 1992; 63: 81–89. DG, Hancock ML, Shuster JJ, Steuber CP, Crist WM, Civin CI, Car- 13 Nucifora G, Begy CR, Kobayashi H, Roulston D, Claxton D, Peder- roll AJ. Immunophenotypes and karyotypes of leukemic cells in son-Bjergaard J, Parganas E, Ihle JN, Rowley JD. Consistent children with Down syndrome and acute lymphoblastic leukemia. intergenic splicing and production of multiple transcripts between J Clin Oncol 1993; 11: 1361–1367. AML1 at 21q22 and unrelated genes at 3q26 in (3;21)(q26;q22) 3 Avet-Loiseau H, Mechinaud F, Harousseau J-L. Clonal hemato- translocations. Proc Natl Acad Sci USA 1994; 91: 4004–4008. logic disorders in Down syndrome. A review. J Pediatr Hematol 14 Golub TR, Barker GF, Bohlander SK, Hiebert SW, Ward DC, Bray- Oncol 1995; 17: 19–24. Ward P, Morgan E, Raimondi SC, Rowley JD, Gilliland DG. Fusion 4 ISCN (1995). An International System for Human Cytogenetic of the TEL gene on 12p13 to the AML1 gene on 21q22 in acute Nomenclature. Mitelman F (ed). S Karger: Basel, 1995. lymphoblastic leukemia. Proc Natl Acad Sci USA 1995; 92: 5 Patterson D. Report of the second international workshop on 4917–4921. human chromosome 21 mapping. Cytogenet Cell Genet 1991; 57: 15 Romana SP, Mauchauffe M, Le Coniat M, Chumakov I, Le Paslier 168–174. D, Berger R, Bernard OA. The t(12;21) of acute lymphoblastic leu- 6 Tighe JE, Daga A, Calabi F. Translocation breakpoints are clustered kemia results in a TEL/AML1 gene fusion. Blood 1995; 85: on both chromosome 8 and chromosome 21 in the t(8;21) of acute 3662–3670. myeloid leukemia. Blood 1993; 254: 592–596. 16 Ichikawa H, Shimizu K, Hayashi Y, Ohki M. An RNA-binding pro- 7 Solomon E, Borrow J, Goddard AD. Chromosome aberrations and tein gene, TLS/FUS, is fused to ERG in human myeloid leukemia cancer. Science 1991; 254: 1153–1160. with t(16;21) chromosomal translocation. Cancer Res 1994; 54: 8 Rogan PK, Close P, Bovin J-L, Siep JR, Gannutz L, Ladda RL, 2865–2868.