Deletion and Reduced Expression of the Fanconi Anemia FANCA Gene in Sporadic Acute Myeloid Leukemia

Home , FANCA

Leukemia (2004) 18, 420–425 & 2004 Nature Publishing Group All rights reserved 0887-6924/04 $25.00 www.nature.com/leu Deletion and reduced expression of the Fanconi anemia FANCA gene in sporadic acute myeloid leukemia

MD Tischkowitz1, NV Morgan1, D Grimwade1,2, C Eddy1, S Ball3, I Vorechovsky4, S Langabeer2, R Sto¨ger1, SV Hodgson1 and CG Mathew1

1Department of Medical and Molecular Genetics, Division of Genetics and Development Guy’s, King’s and St Thomas’ School of Medicine, King’s College London, Guy’s Hospital, London, UK; 2Department of Haematology, University College London, London, UK; 3Department of Haematology, St George’s Hospital, London, UK; and 4Human Genetics Division, School of Medicine, Southampton University Hospital, Tremona Road, Southampton, UK

Fanconi anemia (FA) is an autosomal recessive chromosomal Since the incidence of AML is highly elevated in FA patients, instability disorder caused by mutations in one of seven known it is possible that inherited or acquired mutations in the FA genes genes (FANCA,C,D2,E,F,G and BRCA2). Mutations in the FANCA gene are the most prevalent, accounting for two-thirds contribute to the etiology of sporadic AML. Epidemiological of FA cases. Affected individuals have greatly increased risks studies assessing incidence of malignancies (including AML) in of acute myeloid leukemia (AML). This raises the question as to FA families have failed to demonstrate an increased incidence in whether inherited or acquired mutations in FA genes might be heterozygous individuals.14,15 However, it is conceivable that involved in the development of sporadic AML. Quantitative genomic instability may result from a second somatic mutation fluorescent PCR was used to screen archival DNA from in an FA carrier or from one or two somatic mutations in a sporadic AML cases for FANCA deletions, which account for 16 40% of FANCA mutations in FA homozygotes. Four hetero- progenitor cell resulting in malignancy. Another chromosome zygous deletions were found in 101 samples screened, which is instability syndrome, ataxia telangiectasia (AT) has been 35-fold higher than the expected population frequency for implicated in sporadic leukemias; the AT gene, ATM,is germline FANCA deletions (Po0.0001). Sequencing FANCA in frequently deleted or mutated in T-cell prolymphocytic leuke- the AML samples with FANCA deletions did not detect mia,17–19 and ATM mutations have also been detected in mutations in the second allele and there was no evidence of sporadic non-Hodgkin’s lymphomas18 and B-cell chronic epigenetic silencing by hypermethylation. However, real-time 20 quantitative PCR analysis in these samples showed reduced lymphocytic leukemia. Moreover, heterozygous carriers of expression of FANCA compared to nondeleted AML samples AT mutations have a phenotype at the gene expression level that 21 and to controls. These findings suggest that gene deletions and differs from control individuals. Thus, heterozygosity for an reduced expression of FANCA may be involved in the promo- ATM mutation may predispose to these malignancies. tion of genetic instability in a subset of cases of sporadic AML. Current evidence regarding the role of FA mutations in Leukemia (2004) 18, 420–425. doi:10.1038/sj.leu.2403280 sporadic hematological malignancies including AML is limited; Published online 29 January 2004 Keywords: Fanconi anemia; acute myeloid leukemia; DNA repair; an analysis of 79 cases of adult AML for FANCA point mutations gene deletions found 13 exonic nucleotide changes, none of which led to truncation of FANCA.22 Five changes were novel missense mutations leading to nonconservative amino-acid substitutions Introduction in four cases, none of which was present in 79 normal controls. One additional missense mutation A3982G (causing the amino- acid change T1328A) had previously been reported to be Fanconi anemia (FA) is an autosomal recessive DNA disorder 23 characterized by hypersensitivity to DNA crosslinking agents.1 pathogenic, and the other seven nucleotide changes were Affected individuals have characteristic congenital abnormal- polymorphisms. A recent analysis of FANCC in 97 cases of ities, develop bone marrow failure at an early age, and have a sporadic pediatric AML found no evidence of known pathogenic greatly increased risk of developing myelodysplasia (MDS), mutations, although one polymorphic variant was present at 2–4 four-fold greater frequency in the AML cases compared to acute myeloid leukemia (AML) and solid malignancies. A 24 25 report on 754 FA patients diagnosed over a 20-year period controls. Xie et al analyzed cellular lysates from 10 sporadic showed that the risks of developing bone marrow failure and AML cell lines for FANCA, C, F and G protein expression using MDS or AML by age 40 years were 90 and 33% respectively.3 A Western blotting. Aberrant FA protein profiles, including the recent survey of cancer incidence in 145 FA patients reported absence or reduced presence of protein, were observed in five nine cases of myeloid leukemia with a median age-of-onset of out of the 10 cell lines. In cell lysates from primary AML blasts, 11.3 years, and a ratio of observed to expected cases of 785.4 11 out of 15 cases also had aberrant profiles. Further evidence Complementation analysis by cell fusion and correction of that disruption of the FA pathway may be implicated in the etiology of some sporadic AML cases is provided by Lensch crosslinker hypersensitivity has delineated at least eight com- 26 plementation groups; FA-A, B, C, D1, D2, E, F, G,5 six genes et al, who reported a case of acquired cytogenetic instability have been cloned, FANCA, C, D2, E, F, G,6–12 and a recent in a 68-year-old man with AML and demonstrated FANCA report has shown biallelic inactivation of BRCA2 in FA-D1 cell dysfunction in tumor cells, while lymphoblastoid cells from the lines.13 same individual had normal FANCA function. The high incidence of AML in FA homozygotes and the evidence for deficiency of FA proteins and their complexes in Correspondence: Dr MD Tischkowitz, Clinical Genetics Unit, Institute sporadic AML,25,26 have led us to examine the role of of Child Health, 30, Guilford Street, London WC1N 1EH, UK; Fax: þ 44 20 78138141; E-mail: [email protected] inactivation of the FA pathway in the development of sporadic Received 17 July 2003; accepted 4 December 2003; Published online AML. We selected the FANCA gene for analysis since mutations 29 January 2004 in this gene account for 70% of all cases of FA. The mutation FANCA gene deletions in AML MD Tischkowitz et al 421 profile of FANCA is highly heterogeneous with over 100 the FANCA CpG island. Following sodium bisulfite conversion, different mutations described,23,27–29 and there is a 40% PCR reactions to amplify the FANCA CpG island sequence were incidence of large deletions, which remove 1–43 exons from performed on a Techne Genius thermocycler in oil-free 25 ml the gene.28,30 Screening for these mutations thus accounts for reactions. A measure of 12.5 ml of sodium bisulfite-converted nearly 30% of all FA mutations across all complementation DNA was added to 12.5 ml of the reaction mix containing 2 Â groups. Taq Polymerase buffer (67 mM Tris-HCl, pH 8.8, 16.6 mM (NH4)2SO4, 1.5 mM MgCl2, 170 mg BSA/ml), 400 mM each dNTP (Pharmacia) and 0.4 mM of each primer (1F, 50 TAGGT- Materials and methods TAGTTTGGAATTTTTGGG 30;1R50 AGAGAAAAAATG- GAAAAAAAA 30). After an initial denaturation of 5 min at Samples 941C, a touch-down protocol was used: five cycles of 941C for 20 s, 601C for 30 s (reducing by 11C/cycle) and 721C annealing Samples were obtained from tissue banks at the Departments of for 40 s, followed by a further 30 cycles of 941C for 20 s, 551C Hematology of University College Hospital, London, St for 30 s, 721C annealing for 40 s (increasing by 2 s with each George’s Hospital, London and the Karolinska Institute, Sweden. cycle), and a final extension step of 5 min at 721C. For the A total of 101 samples were analyzed. A total of 22 samples second (seminested) PCR, 0.5 ml of the first PCR end-product (22%) were selected for monosomy 7, since this is one of the was transferred to a new 24.5 ml PCR mixture with the same final commonest recurring cytogenetic findings in AML arising on a concentration of reagents as above, but using a different reverse background of FA,2 and the rest were randomly selected. The primer (2R, 50 GGAAATGAATGGGGTTGGTAATG 30). The median age was 58 years (range 16–90 years). FAB group was: conditions for this PCR were the same as above, but without the M1 ¼ 3, M2 ¼ 19, M3 ¼ 15, M4 ¼ 23, M5 ¼ 10, M6 ¼ 3, M7 ¼ 1, initial 5-cycle touch-down step. unknown ¼ 27. Risk classification based on cytogenetic analy- The PCR product was purified and sequenced using a sis31 was: adverse ¼ 25, intermediate ¼ 32, favorable ¼ 16, commercial sequencing kit (Applied Biosystems, Warrington, unknown ¼ 28. The study had ethical approval (MREC 99/7/45). UK) according to the manufacturer’s instructions. The reverse primer was the same as that used for the second PCR, with a 0 T different forward primer (2F, 5 GTAGG( /C)GTATTTTTTAG- Quantitative fluorescent PCR analysis of FANCA gene GATTAATATG 30). The resulting amplicon had an expected dosage length of 221 bp and was predicted to contain 24 CpG dinucleotides. The product was cleaned and analyzed using The FANCA multiplex quantitative fluorescent PCR (QF-PCR) an ABI 310 Genetic Analyzer (Applied Biosystems, Warrington, gene dosage assay used has been previously described by UK). Sequence data were analyzed using Sequence Navigator Morgan et al.28 This multiplex amplifies five exons (5, 11, 17, software (Applied Biosystems, Warrington, UK), specifically 21, and 31) and uses exon 1 of the myelin protein zero (MPZ) assessing each CpG island to determine whether a thymine was gene as a non-FA gene control. An additional biplex assay that present (indicating that the original cytosine residue was amplifies a 352 bp fragment encompassing both exons 19 and unmethylated), or whether a cytosine was present (indicating 20 of FANCA and exon 1 of MPZ was also designed (forward 50 methylation). A cell line with extensive methylation of the GAAACACCGGTCACCGTCTGTG 30, reverse 50 AGATCCAC- FANCF promoter was used as a control for the lack of GATTCTTCGCATTGTC 30). PCRs were performed in 25 ml conversion of methylated CpGs by sodium bisulfite. reactions with 125 ng DNA, 1 Â Taq Polymerase buffer, 200 mM each dNTP and 0.2 mM of each primer. After an initial denaturation at 941C for 3 min 1.5 U Taq DNA Polymerase Real-time quantitative RT-PCR (RQ-PCR) (Promega, UK) was added (hot start), followed by 22 cycles of 941C for 1 min, annealing at 571C for 1 min, and extension for In order to determine whether FANCA mRNA expression was 2 min at 721C with a final extension for 5 min at 721C. Only 22 altered in AML samples with heterozygous FANCA deletions, an PCR cycles were performed to keep the reaction within the RQ-PCR assay was developed using the Taqman technology and exponential phase.28,32 The fluorescent products were quanti- an ABI 7700 (Applied Biosystems, Warrington, UK) with fied using an ABI 310 (Applied Biosystems, Warrington, UK) and methods previously described.34 Expression of the FANCA exon compared to MPZ. Values were compared to at least two normal 19/20 transcript was compared with two different endogenous control samples and a peak area ratio was calculated for each control genes, ABL and b2-microglobulin (B2M), using pub- sample. In samples where the deletion included exon 31, a lished primer and probe sets. Expression of these control genes further QF-PCR biplex assay that amplified FANCA exons 42 has been documented as comparable between normal hemo- and 43 in a single 468 bp fragment (forward 50 GCACAGCATG- poietic cells and acute and chronic leukemias.35 Primers and CAGACTATGG 30; reverse 50 GGCAGGTCCCGTCAGAAGA- probe sequences for the FANCA amplicon including exons 19 GAT 30) and MPZ was carried out to determine the 30 extent of and 20 were: forward primer, 50-CGTCATGGAGGCCAGCA- the deletion. Samples in which deletions were found were typed TAT-30, and reverse primer 50-TGGCAGTAGGTGGAGTAGA- with microsatellite markers, which flanked and were tightly GAG-30, and probe 50-FAM-CCCGCCCTGCTCACACCTCG- linked to FANCA. Sequencing of the 43 exons of FANCA was TAMRA-30. Plasmid standards for each target gene were undertaken to screen for a second mutation in the nondeleted generated (FANCA) or obtained commercially (B2M, ABL; allele. Ipsogen, Marseille, France) and serially diluted to encompass the range of expression levels encountered in hemopoietic cells. Gene expression was quantified in terms of normalized copy Sodium bisulfite sequencing number (NCN), which was derived from the relevant standard curve. Samples with control gene expression levels outside the The sodium bisulfite conversion technique described by Clark normal range as defined previously35 were deemed to have et al33 was used to develop an assay for the methylation status of unsatisfactory RNA and were excluded from the analysis.

Leukemia FANCA gene deletions in AML MD Tischkowitz et al 422 Results were deleted, the deletion could have extended beyond the FANCA gene. The two samples with intragenic deletions were FANCA gene dosage analysis and mutation screen either homozygous for all four markers, or other deletions may have been present in the region (summarized in Figure 1, panel Four of the 101 sporadic AML samples that were tested using the B). Clinical and cytogenetic data were available on three FANCA gene dosage assay were found to have heterozygous samples (Table 1) and all three were de novo AMLs; they were deletions in FANCA (see Figure 1, panel A), and these were classiﬁed into FAB groups M1, M2 and M6, and all had complex analyzed further. In samples 63 and 96, one copy of all the karyotypes with monosomy 7. All three samples also had 16q exons tested (5–43) was deleted, and in samples 65 and 83 the abnormalities (FANCA is located on 16q24.3), although none deletions were intragenic, and involved exons 19–21 and 11–21 had cytogenetically apparent chromosome deletions of this respectively. Microsatellite polymorphism analysis was per- region. formed using four markers tightly linked to and ﬂanking the All 43 exons of FANCA were sequenced in the four samples FANCA locus (D16S3074, D16S3026, D16S3121, D16S303) to with deletions to determine whether the remaining allele was investigate the extent of the deletions, and all were homozygous inactivated by other classes of mutations. In addition to the (or hemizygous) for the four markers. This reinforced the coding sequence of each exon, approximately 30 bp of possibility that, for the two samples where all exons tested sequence 50 and 30 to each exon were also analyzed to detect

Figure 1 FANCA gene dosage results. Panel a: FANCA exon 19/20 gene dosage results. Compared to the normal control sample (top left), in the AML deletion samples the peak area ratios for exon 19/20 to the control exon, MPZ exon 1, are 50%, indicating deletion of one allele. This is also seen in a germline heterozygous deletion from a known FANCA carrier (middle left). Panel b: Schematic diagram of the FANCA gene showing the extent of the detected deletions in the four AML samples.

Leukemia FANCA gene deletions in AML MD Tischkowitz et al 423 Table 1 Summary of features in samples found to have deletions of FANCA

Sample number Clinical details Cytogenetic analysis Deletion

63 Male, 65 years, FAB M2 43, XY, del(5)(q15q3?3), À6, À7, r(7), i(8)(q10), add(16)(q?24), Heterozygous exons 5–43 À17, À22, +der(?)t(?;6)(?;?p1) [12] 65 Female, 45 years, FAB M1 41–45, XX, add(2)(q2?), der(5)t(5;12)(q2?;q1?), À7, ?+10, À12, Heterozygous exons 19–21 add(16)(q2?), À18, À20, ?+22 [cp10] 83 Male, 58 years, FAB M6 44, XY, del(1)(q21q25), add(4)(q2?5), À5, À7, À11, À12, Heterozygous exons 11–21 del(12)(q21q24), add(13)(q?), add(16)(q13), À17, +3mar [20] 96 Female, 69 years N/A Heterozygous exons 5–43 PB ¼ peripheral blood sample; N/A ¼ not available.

Table 2 Summary of sequence variants found in the four AML samples with FANCA exon deletions detected by QF-PCR

Sample(s) Site Mutation Amino-acid change Comment

63, 65, 96 Intron 3 IVS4-24G/C No change 83, 96 Intron 7 IVS7-12A/G No change Known polymorphism G-55.4%, A-44.6% (FA MDB) 83 Exon 9 796A/G T266A Known polymorphism (Savino et al29) 63, 65, 96 Intron 10 IVS10-15A/C No change 83 Intron 13 IVS 13-20 A/G No change Known polymorphism (FA MDB) 96 Exon 16 1501G/A G501S Known polymorphism G-64.3%, A-35.6% (Savino et al29) 83 Exon 26 2426G/A G809D Known polymorphism G-43%, A-57.3% (Savino et al29) 83 Intron 39 IVS39-16C/T No change Known polymorphism C-62.5% T-37.5% (FA MDB) 63, 65, 83, 96 Intron 40 IVS40-C/G No change FA MDB ¼ Fanconi Anemia Mutation Database (http://www.rockefeller.edu/fanconi/mutate/).

possible splice site mutations. Several polymorphisms but no Number of FANCA copies per 106B2M copies inactivating mutations were found in these samples. Details of 1000 the polymorphisms found are given in Table 2. The age range of the four patients with deletions was 45–69 years (Table 1), and as there was no evidence of prior stem cell failure or congenital abnormalities, a primary diagnosis of FA is very unlikely. 100

FANCA promoter methylation analysis

Methylation of cytosine in CpG islands of the promoter region of 10 genes is a well documented mechanism of silencing tumor suppressor genes in cancer36 but has not been studied in FANCA to date. The 50 region of FANCA contains a CpG island that COPY NUMBER extends into exon 1,37 and methylation of this region may 1 provide an alternative mechanism of FANCA gene inactivation. A sodium bisulﬁte conversion and sequencing assay33 was developed to detect methylation of the FANCA CpG island. CONTROLS AML-DEL AML Treatment of DNA with sodium bisulﬁte converts unmethylated cytosines to uracil while methylated cytosines are unchanged; 0.1 after PCR, unmethylated cytosines are thus detected as thymines SAMPLE TYPE on sequencing. This region, which contains 24 CpG dinucleo- Figure 2 NCN values (median and ranges) for FANCA expression tides, was analyzed in the four samples with FANCA deletions to in FANCA deleted and nondeleted AML samples and normal controls. establish whether the remaining allele had been silenced by methylation. There was no evidence of cytosine methylation of FANCA in these four samples, or in peripheral blood lympho- cytes from three normal individuals. was developed to assess gene transcription for FANCA exons 19 and 20, which was a region known to be deleted in all four samples. The assay was carried out in the three samples with FANCA gene expression analysis FANCA deletions from which RNA was available and in peripheral blood lymphocyte or bone marrow RNA derived Since there was no evidence of inactivation of the second allele from six healthy individuals, as well as in bone marrow and in the four samples with FANCA deletions either by a sequence CD34-selected peripheral blood stem cells from three healthy change or by epigenetic mechanisms, FANCA gene expression controls. The results are shown in Figure 2. The normalized was analyzed in these samples. A real-time quantitative assay expression of FANCA was lower in the samples with FANCA

Leukemia FANCA gene deletions in AML MD Tischkowitz et al 424 deletions compared to normal controls (Student’s t-test, one- be due to differing proportions of normal and blast cells in each tailed, P ¼ 0.0048) and was also lower compared to nondeleted AML sample. The higher expression of FANCA in some AML AML cases (Student’s t-test, one-tailed, P ¼ 0.036). FANCA samples may be a result of the increased expression of FANCA expression was higher in six out of nine nondeleted AML (and other FA genes) in rapidly dividing tissues.40 samples in comparison with normal controls. Comparable data Taken together, our data indicate heterozygous deletions of were observed using ABL as the control gene (data not shown). the FANCA gene are found in a small but significant proportion of sporadic AMLs. The study by Condie et al22 reported one previously described FANCA mutation and four other non- Discussion conservative amino acid substitutions that were absent in controls. Thus, it is possible that five of 79 adult AML samples The prevalence of FA in the general population is approximately in that series were heterozygous for pathogenic FANCA 3 per million14 and FANCA accounts for around two-thirds of mutations. However, none of the mutations reported in the cases. Using the Hardy–Weinberg equation, one can estimate study by Condie et al22 were truncating. Our study did not test the carrier frequency of FANCA mutations to be 0.28%. for point mutations in the primary mutation screen of the 101 Previous studies have shown that deletions account for 40% of AML samples, whereas Condie et al22 did not screen for large FANCA mutations,28 so the carrier frequency of FANCA deletions. Taken together, these two studies suggest that deletions can be estimated to be 0.11% in the general heterozygous mutation of the FANCA gene occurs in a population. The number of FANCA deletion heterozygotes significant proportion of sporadic AMLs. If the mutations detected in this cohort of sporadic AML patients (4/101) is thus identified in this study are indeed germ line, then heterozygous approximately 35 times the expected frequency of germline FANCA deletions may be considered to be low penetrance FANCA deletion mutations in the general population (exact alleles for conferring AML risk. Evidence for a heterozygote binomial test, Po0.0001). The deletions could be germline or effect has been provided by Djuzenova et al,41 who used an somatic mutations, but this could not be determined as the alkaline single-cell gel electrophoresis (comet) assay to show tumor samples were archival and the patients were no longer that cells from FA heterozygotes that were exposed to X-rays had alive; of the three cases with deletions where clinical details higher initial DNA damage rates and a prolonged DNA repair were available, two (cases 63 and 65) died during induction and half-time compared to control cells. Also, Pearson et al42 found the third (case 83) went into remission for 3 months before that FA heterozygotes had higher mean rates of chromosomal relapsing. aberrations following exposure to diepoxybutane than normal Interestingly, three of the FANCA-deleted samples had controls. These findings suggest that FA heterozygotes may have monosomy 7, which is the most frequently reported cytogenetic an impaired ability to maintain genomic integrity. This hypoth- abnormality in AMLs from cases of FA.38 In addition to esis has not been corroborated by epidemiological studies that monosomy 7, all had loss or structural anomalies of chromo- failed to show an increased incidence of AML in FA some 5 in the context of a highly complex karyotype (see families,14,15 although these were based on small numbers Table 1). Such characteristics are usually associated with and were therefore underpowered. Alternatively, the mutations therapy-related leukemia particularly following exposure to could be somatic with acquired dysfunction of FA proteins alkylating agents.39 It may be that these complex karyotypes are acting as a progression factor in MDS/AML. Haplo-insufficiency specifically associated with FANCA deletions or alternatively of an FA gene in a progenitor cell may promote genomic they may have facilitated detection of FANCA abnormalities in instability, which could result in further mutations and the the neoplastic cells. evolution of a tumorigenic clone.16 In fact, either a germline FA The FANCA deletions could be the result of large chromo- gene mutation or a somatic mutation in a stem cell may somal deletions that may occur in AML, as would seem likely to constitute a progression factor in a pathway that leads in some be the case in samples 63 and 96 where all measured exons cases to chromosomal instability and cancer.16,26 The three were half expected values and the flanking microsatellite cases with FANCA gene deletions in which diagnostic markers were all homozygous. In the other samples, 65 and cytogenetics was available had a very complex karyotype, 83, the deletions were intragenic. which is consistent with this hypothesis. Further studies looking No mutations were detected in the other FANCA allele in at both tumor and germline DNA from sporadic AML cases are these four samples by sequencing all FANCA exons and splice required to determine whether FANCA or other FA gene sites, but other mutations such as those affecting intronic mutations represent germline or somatic mutations. This study splicing enhancers/silencers or creating cryptic splice sites also provides the rationale for a more extensive study; it will be cannot be excluded. Analysis of CpG island methylation in the interesting to establish whether abnormalities of the FA pathway promoter region of FANCA for epigenetic silencing did not are commonly implicated in cases of AML with adverse reveal any evidence of methylation in the AML samples with cytogenetics in which molecular mechanisms are currently FANCA deletions, or in a further 27 AML samples without a poorly understood. FANCA deletion. These data are consistent with the quantitative analysis of FANCA mRNA, which detected reduced FANCA expression in the deleted samples compared to the nondeleted Acknowledgements AML samples and to the controls. If there is inactivation of only one FANCA allele in the deletion samples, one would expect We thank Professor David Linch for supplying samples from the less than a two-fold reduction in gene expression due to the MRC AML trials tissue bank at University College Hospital, presence of some nonmalignant cells. However, the reduction London. These facilities have been supported by the Kay Kendall observed was approximately 2.4-fold relative to normal controls Leukemia Fund, the Medical Research Council and the Leukemia and approximately four-fold relative to nondeleted AML Research Fund of Great Britain. We thank Dr Cathryn Lewis for samples, which suggests that there may have been partial advice on statistical analysis and Grover C Bagby Jr who supplied inactivation of the second FANCA allele in some of the deleted probes for the RQ-PCR assay. MT was funded by Cancer Research samples. Some of the variation in the degree of reduction could UK. DG was supported by the Leukemia Research Fund of Great

Leukemia FANCA gene deletions in AML MD Tischkowitz et al 425 Britain, and equipment was provided by the Daniel Ayling 23 Wijker M, Morgan NV, Herterich S, van Berkel CG, Tipping AJ, Fanconi Anemia Trust. Gross HJ et al. Heterogeneous spectrum of mutations in the Fanconi anaemia group A gene. Eur J Hum Genet 1999; 7: 52–59. References 24 Barber LM, McGrath HE, Meyer S, Will AM, Birch JM, Eden OB et al. Constitutional sequence variation in the Fanconi anaemia 1 Auerbach A, Buchwald M, Joenje H. Fanconi anaemia. In: group C (FANCC) gene in childhood acute myeloid leukaemia. Br J Vogelstein B, Kinzler KW (eds). The Genetic Basis of Human Haematol 2003; 121: 57–62. Cancer. New York: McGraw-Hill Inc., 1999, pp 317–332. 25 Xie Y, de Winter JP, Waisfisz Q, Nieuwint AW, Scheper RJ, Arwert 2 Alter BP. Fanconi’s anemia and malignancies. Am J Hematol 1996; F et al. Aberrant Fanconi anaemia protein profiles in acute myeloid 53: 99–110. leukaemia cells. Br J Haematol 2000; 111: 1057–1064. 3 Kutler DI, Singh B, Satagopan J, Batish SD, Berwick M, Giampietro 26 Lensch MW, Tischkowitz M, Christianson TA, Reifsteck CA, PF et al. A 20 year perspective of the International Fanconi Anemia Speckhart SA, Jakobs PM et al. Acquired FANCA dysfunction Registry (IFAR). Blood 2003; 101: 1249–1256. and cytogenetic instability in adult acute myelogenous leukemia. 4 Rosenberg PS, Greene MH, Alter BP. Cancer incidence in persons Blood 2003; 102: 7–16. with Fanconi’s anemia. Blood 2003; 101: 822–825. 27 Levran O, Erlich T, Magdalena N, Gregory JJ, Batish SD, Verlander 5 Joenje H, Oostra AB, Wijker M, di Summa FM, van Berkel CG, PC et al. Sequence variation in the Fanconi anemia gene FAA. Proc Rooimans MA et al. Evidence for at least eight Fanconi anemia Natl Acad Sci USA 1997; 94: 13051–13056. genes. Am J Hum Genet 1997; 61: 940–944. 28 Morgan NV, Tipping AJ, Joenje H, Mathew CG. High frequency of 6 Lo Ten Foe JR, Rooimans MA, Bosnoyan-Collins L, Alon N, Wijker large intragenic deletions in the Fanconi anemia group A gene. Am M, Parker L et al. Expression cloning of a cDNA for the major J Hum Genet 1999; 65: 1330–1341. Fanconi anaemia gene, FAA. Nat Genet 1996; 14: 320–323. 29 Savino M, Ianzano L, Strippoli P, Ramenghi U, Arslanian A, 7 The Fanconi Anaemia/Breast Cancer Consortium. Positional Bagnara GP et al. Mutations of the Fanconi anemia group A gene cloning of the Fanconi anaemia group A gene. Nat Genet 1996; (FAA) in Italian patients. Am J Hum Genet 1997; 61: 1246–1253. 14: 324–328. 30 Levran O, Doggett NA, Auerbach AD. Identification of Alu- 8 Strathdee CA, Duncan AM, Buchwald M. Evidence for at least four mediated deletions in the Fanconi anemia gene FAA. Hum Mutat Fanconi anaemia genes including FACC on chromosome 9. Nat 1998; 12: 145–152. Genet 1992; 1: 196–198. 31 Grimwade D, Walker H, Oliver F, Wheatley K, Harrison C, 9 Timmers C, Taniguchi T, Hejna J, Reifsteck C, Lucas L, Bruun D Harrison G et al. The importance of diagnostic cytogenetics on et al. Positional cloning of a novel Fanconi anemia gene, FANCD2. outcome in AML: analysis of 1,612 patients entered into the MRC Mol Cell 2001; 7: 241–248. AML 10 trial. The Medical Research Council Adult and Children’s 10 de Winter JP, Rooimans MA, van Der Weel L, van Berkel CG, Alon Leukaemia Working Parties. Blood 1998; 92: 2322–2333. N, Bosnoyan-Collins L et al. The Fanconi anaemia gene FANCF 32 Yau SC, Bobrow M, Mathew CG, Abbs SJ. Accurate diagnosis of encodes a novel protein with homology to ROM. Nat Genet 2000; carriers of deletions and duplications in Duchenne/Becker 24: 15–16. muscular dystrophy by fluorescent dosage analysis. J Med Genet 11 de Winter JP, Leveille F, van Berkel CG, Rooimans MA, van Der 1996; 33: 550–558. Weel L, Steltenpool J et al. Isolation of a cDNA representing the 33 Clark SJ, Harrison J, Paul CL, Frommer M. High sensitivity Fanconi anemia complementation group E gene. Am J Hum Genet mapping of methylated cytosines. Nucleic Acids Res 1994; 22: 2000; 67: 1306–1308. 2990–2997. 12 de Winter JP, Waisfisz Q, Rooimans MA, Van Berkel CGM, 34 Grimwade D, Outram SV, Flora R, Ings SJ, Pizzey AR, Morilla R Bosnoyan-Collins L, Alon N et al. The Fanconi anaemia group G et al. The T-lineage-affiliated CD2 gene lies within an open gene FANCG is identical with XRCC9. Nat Genet 1998; 20: chromatin environment in acute promyelocytic leukemia cells. 281–283. Cancer Res 2002; 62: 4730–4735. 13 Howlett NG, Taniguchi T, Olson S, Cox B, Waisfisz Q, De Die- 35 Gabert J, Beillard E, Van Der Velden VH, Bi W, Grimwade D, Smulders C et al. Biallelic inactivation of BRCA2 in Fanconi Pallisgaard N et al. Standardization and quality control studies of anemia. Science 2002; 297: 606–609. ‘real-time’ quantitative reverse transcriptase polymerase chain 14 Swift M. Fanconi’s anaemia in the genetics of neoplasia. Nature reaction of fusion gene transcripts for residual disease detection in 1971; 230: 370–373. leukemia – A Europe Against Cancer Program. Leukemia 2003; 17 15 Swift M, Caldwell RJ, Chase C. Reassessment of cancer predis- (12): 2318–2357. position of Fanconi anemia heterozygotes. J Natl Cancer Inst 1980; 36 Baylin SB, Herman JG. DNA hypermethylation in tumorigenesis: 65: 863–867. epigenetics joins genetics. Trends Genet 2000; 16: 168–174. 16 Lensch MW, Rathbun RK, Olson SB, Jones GR, Bagby Jr GC. 37 Ianzano L, D’Apolito M, Centra M, Savino M, Levran O, Auerbach Selective pressure as an essential force in molecular evolution of AD et al. The genomic organization of the Fanconi anemia group myeloid leukemic clones: a view from the window of Fanconi A (FAA) gene. Genomics 1997; 41: 309–314. anemia. Leukemia 1999; 13: 1784–1789. 38 Butturini A, Gale RP, Verlander PC, Adler-Brecher B, Gillio AP, 17 Stoppa-Lyonnet D, Soulier J, Lauge A, Dastot H, Garand R, Sigaux Auerbach AD. Hematologic abnormalities in Fanconi anemia: an F et al. Inactivation of the ATM gene in T-cell prolymphocytic International Fanconi Anemia Registry study. Blood 1994; 84: leukemias. Blood 1998; 91: 3920–3926. 1650–1655. 18 Vorechovsky I, Luo L, Dyer MJ, Catovsky D, Amlot PL, Yaxley JC 39 Pedersen-Bjergaard J, Rowley JD. The balanced and the unba- et al. Clustering of missense mutations in the ataxia-telangiectasia lanced chromosome aberrations of acute myeloid leukemia may gene in a sporadic T-cell leukaemia. Nat Genet 1997; 17: 96–99. develop in different ways and may contribute differently to 19 Stilgenbauer S, Schaffner C, Litterst A, Liebisch P, Gilad S, Bar- malignant transformation. Blood 1994; 83: 2780–2786. Shira A et al. Biallelic mutations in the ATM gene in T- 40 van de Vrugt HJ, Cheng NC, de Vries Y, Rooimans MA, de Groot J, prolymphocytic leukemia. Nat Med 1997; 3: 1155–1159. Scheper RJ et al. Cloning and characterization of murine fanconi 20 Stankovic T, Weber P, Stewart G, Bedenham T, Murray J, Byrd PJ anemia group A gene: Fanca protein is expressed in lymphoid et al. Inactivation of ataxia telangiectasia mutated gene in B-cell tissues, testis, and ovary. Mamm Genome 2000; 11: 326–331. chronic lymphocytic leukaemia. Lancet 1999; 353: 26–29. 41 Djuzenova CS, Rothfuss A, Oppitz U, Speit G, Schindler D, Hoehn 21 Watts JA, Morley M, Burdick JT, Fiori JL, Ewens WJ, Spielman RS H et al. Response to X-irradiation of Fanconi anemia homozygous et al. Gene expression phenotype in heterozygous carriers of ataxia and heterozygous cells assessed by the single-cell gel electrophor- telangiectasia. Am J Hum Genet 2002; 71: 791–800. esis (comet) assay. Lab Invest 2001; 81: 185–192. 22 Condie A, Powles RL, Hudson CD, Sheperd V, Bevan S, Yuille MR 42 Pearson T, Jansen S, Havenga C, Stones DK, Joubert G. Fanconi et al. Analysis of the Fanconi anaemia complementation group A anemia. A statistical evaluation of cytogenetic results obtained gene in acute myeloid leukaemia. Leukemia Lymphoma 2002; 43: from South African families. Cancer Genet Cytogenet 2001; 126: 1849–1853. 52–55.

Leukemia