Letters to the Editor 406 Acknowledgements

This study was supported by University of Bologna, BolognaAIL and Carisbo Foundation. PC and MM are recipients of grants provided by the Istituto di Radioterapia ‘Luigi Galvani’, University of Bologna-Medical School. P Corrado1, M Mancini1, G Brusa1, S Petta1, G Martinelli1, E Barbieri2 and MA Santucci1 1Dipartimento di Ematologia e Scienze Oncologiche ‘Lorenzo e Ariosto Sera`gnoli’, University of Bologna-Medical School, Bologna, Italy and 2Dipartimento di Scienze Radiologiche ed Istopatologiche L Galvani, University of Bologna-Medical School, Bologna, Italy E-mail: [email protected]

References

1 Calnan DR, Brunet A. The FoxO code. Oncogene 2008; 27: 2276–2288. 2 Tsai K-L, Sun Y-J, Huang C-Y, Yang M-C H, Hsiao C-D. Crystal Figure 2 Post-translational modifications of FOXO3a following structure of the human FOXO3a-DBD/DNA complex suggests the p210BCR-ABL tyrosine kinase (TK) inhibition in response to imatinib effects of post-translational modification. Nucleic Acids Res 2007; mesylate (IM). FOXO3a expression, phosphorylation at Ser/Thr 35: 6984–6994. residues and acetylation at Lys residues were investigated in the 3 Vogt PK, Jiang H, Aoki M. Triple layer control. Phosphorylation, nuclear compartments of clone 3B kept at the permissive temperature acetylation and ubiquitination of FOXO proteins. Cell Cycle 2005; BCR-ABL for p210 TK (33 1C) at different periods of exposure to IM (1 mM). 4: 908–913. FOXO3a acetylation at Lys residues in the cytoplasmic compartments 4 Matsuzaki H, Daitoku H, Hatta M, Aoyama H, Yoshimochi K, of clone 3B was also investigated. p300 levels and AKT activating Fukamizu A. Acetylation of Foxo1 alters its DNA-binding ability and phosphorylation at Ser473 were assayed by labelling IP with anti-p300 sensitivity to phosphorylation. Proc Natl Acad Sci USA 2005; 102: and anti-phospho-Ser473 AKT antibodies (from Upstate Biotechnology 11278–11283. and Cell Signaling Technology, respectively). See legend to Figure 1 5 Komatsu N, Watanabe T, Uchida M, Mori M, Kirito K, Kikuchi S for technical details. et al. A member of forkhead transcription factor FKHRL1 is a downstream effector of STI571-induced cell cycle arrest in BCR- ABL-expressing cells. J Biol Chem 2003; 278: 6411–6419. 6 Essafi A, Ferna`ndez de mattos S, Hassen YA, Soeiro I, Mufti GJ, of exposure to IM (Figure 2; data not shown). FOXO3a acetyl- Thomas NS et al. Direct transcriptional regulation of Bim by Foxo3a mediates STI571-induced apoptosis in Bcr-Abl-expressing cells. ation is likely involved in its phosphorylation by AKT, Oncogene 2005; 24: 2317–2329. whose enzymatic activity had already recovered at 24 h of 7 Mahmud DL, G-Amlak M, Deb DK, Platanias LC, Uddin S, IM exposure thereby compensating the persistent inhibition Wickrema A. Phosphorylation of forkhead transcription factors by of p210BCR-ABL TK activity (Figure 2).4 In conclusion, our results erythropoietin and stem cell factor prevents acetylation and their support FOXO3a acetylation as a late event of response interaction with coactivator p300 in erythroid progenitor cells. of chronic myeloid leukemia progenitors to IM eventually Oncogene 2002; 21: 1556–1562. 8 Mancini M, Brusa G, Zuffa E, Corrado P, Martinelli G, Grafone T contributing to the development of a drug-resistant phenotype et al. Persistent Cdk2 inactivation drives growth arrest of BCR-ABL- through mechanisms promoting re-phosphorylation and trans- expressing cells in response to dual inhibitor of SRC and criptional attenuation of FOXO3a. ABL kinases SKI-606. Leuk Res 2007; 31: 979–987.

High-throughput mutational screen of the tyrosine kinome in chronic myelomonocytic leukemia

Leukemia (2009) 23, 406–409; doi:10.1038/leu.2008.187; monocytic leukemia, albeit the result of diverse genetic lesions, published online 10 July 2008 including mutations of N-RAS, K-RAS, NF-1 and PTPN11 (reviewed in Lauchle et al.2). The genetics of CMML and aCML is yet more complex. Various studies reported activating Chronic myelomonocytic leukemia (CMML) and juvenile mutations of N-RAS and K-RAS in 20 to almost 60% of patients, myelomonocytic leukemia are characterized by persistent and mutations in PTPN11 in isolated cases.3,4 NF-1 (neuro- monocytosis and combined dysplastic morphology with features fibromatosis 1) has not been studied extensively, but one study of a myeloproliferative syndrome.1 Together with atypical of patients with MDS (including CMML) and acute myeloid chronic myelogenous leukemia (aCML) and the category of leukemia did not find major rearrangements at the genomic unclassifiable myelodysplastic/myeloproliferative diseases, they level.5 In contrast to juvenile myelomonocytic leukemia, belong to the same WHO diagnostic group.1 Activation of the activating mutations in tyrosine kinases, including PDGFRA/B, RAS signaling pathway is almost universal in juvenile myelo- KIT, FGFR1, FLT3, CSF-1R and JAK2, have also been implicated

Leukemia Table 1 Patient characteristics and mutations

Patient Disease Sex Age at Splenomegaly WBC HGB PLT Karyotype (cytogenetic Point mutation at the respective residue no. diagnosis ( Â 106/l) (g/100 ml) ( Â 106/l) nomenclature) JAK2 K-RAS N-RAS

1 CMML m 71 Yes 7100 8.6 94 000 NA A146T (50) 2 CMML m 66 No 2900 12.9 138 000 46,XY 3 CMML m 65 No 8000 11.4 160 000 46,XY[9]/46,XY,1pÀ[5] 4 CMML f 70 No 24 900 15.5 272 000 46,XX[20] V617F (75) 5 CMML f 66 NA 3700 7.7 63 000 46,XY,À7,Add(18)(p11)[20] 6 CMML m 72 Yes 6200 11.9 74 000 46,XY[20] G12D (35) 7 CMML m 65 No 4600 12.7 177 000 46,XY[20] G12D (16) 8 CMML m 59 NA 10 400 13.4 71 000 NA 9 CMML m 40 Splenectomized 39 700 14.5 404 000 46,XY[20] V617F (100) 10 CMML m 53 NA 13 600 12.7 27 000 46,XY,i(17)(q10)[9/20] 11 CMML m 74 No 17 800 14.1 83 000 46,XY[20] 12 CMML m 59 Yes 22 800 13.8 77 000 NA 13 CMML m 63 No 5000 13.3 64 000 46,XY[20] 14 CMML/AML m NA NA NA NA NA NA 15 CMML m 48 No 2700 11.5 50 000 NA A146T (50) 16 CMML m 65 No 36 600 8.5 52 000 NA G13D (50)

17 CMML m 72 Yes 27 400 14.3 16 000 46,XY[20] V617F (69) Editor the to Letters 18 CMML m 64 No 5500 12.5 281 000 45,X0,ÀY[20] 19 uMDS/MPD f 39 No 14 600 13.1 203 000 46,XX[20] 20 CMML f 59 Yes 29 100 11.4 69 000 BCR-ABL-negative (FISH) 21 CMML m 53 No 14 300 14.3 98 000 46,XY[20] G12A (40) 22 CMML f 61 NA NA NA NA BCR-ABL-negative (FISH) 23 aCML f 53 NA NA NA NA 47,XX,del(13)1q13q22), +mar(14/22)[25] 24 CMML m 67 NA 5300 11.4 150 000 NA 25 CMML m 82 No 13 400 11.9 73 000 NA 26 CMML m 48 No 11 500 11.4 133 000 46,XY[20] 27 aCML m 68 NA NA NA NA 46,XY[30] 28 aCML f NA NA NA NA NA BCR-ABL-negative (FISH) G12D (36) 29 CMML m 80 NA NA NA NA BCR-ABL-negative (FISH) G12D (51) 30 aCML m 55 Yes 18 000 NA NA XY,del(7)(q23)[61] 31 aCML f NA NA NA NA NA BCR-ABL-negative (FISH) T74P (29) 32 CMML m 68 Yes 38 5000 5.6 453 000 46,XY[20] Abbreviations: aCML, atypical chronic myelogenous leukemia; CMML/AML, the patient was progressing to acute myeloid leukemia at the time of sample collection; f, female; FISH, fluorescence in situ hybridization; m, male; NA, not available; uMDS/MPS, unclassifiable myelodysplastic/myeloproliferative syndrome. Blood parameters refer to the time of diagnosis. The estimated proportion of mutant allele is given in parentheses after the mutation type. Leukemia 407 Letters to the Editor 408 in small subsets of patients.6–11 Mutations in RAS or in tyrosine mutations could reside in tyrosine kinase domains not included kinases primarily drive proliferation, and are collectively in our sequence analysis (for example, the extracellular referred to as ‘type 1’ mutations. Type 1 mutations do not domain), in genes that regulate RAS or tyrosine kinases or they usually coexist within the same leukemic clone, although there could lead to activation by mechanisms not detected by point are occasional exceptions.8,12 The cumulative frequency of mutation screening, such as translocations. Given the limited previously reported mutations in RAS or a tyrosine kinase in size of the cohort under study, it is also possible that we missed CMML and aCML is not known, as no comprehensive mutations that occur at a low or very low frequency. For mutational analysis has been performed in such patients. example, assuming mutual exclusivity of type 1 mutations, the However, given that most of the reported tyrosine kinase probability of detecting at least one mutant case in the 20 mutations are infrequent, it is likely that screening for patients without RAS or JAK2 mutations would be 98.9% for a established mutations in specific kinases and RAS would leave mutation occurring with an incidence of 20%. However, for a a considerable proportion of cases unexplained. We therefore mutation with only 1% incidence, this probability falls to just hypothesized that previously unidentified tyrosine kinase muta- 18.2%. The relative small size of our cohort may also explain tions may be involved in the pathogenesis of CMML and aCML. the lack of detection of any patients with mutations of FLT3 To test this, we performed a comprehensive mutational screen of (reported with a frequency of 3.1%) and CSF-1R (reported with a the tyrosine kinome in a cohort of such patients. frequency of 8%).10 However, although it is clear that our study Thirty-two patients were included in this study. In 26 the is not powered to detect one specific low-incidence mutation, disease was classified as CMML, in 5 as aCML and in 1 as an low-frequency mutations in multiple kinases would not likely unclassifiable myelodysplastic/myeloproliferative disorder have escaped detection. Moreover, our data are consistent with (Table 1). Karyotyping was unavailable for eight CMML patients, two recent studies that failed to detect novel activating reflecting the fact that some samples were acquired up to 20 mutations in tyrosine kinases in large cohorts of patients with years ago. In five patients without available karyotype, BCR-ABL acute myeloid leukemia.13,15 In contrast to these studies, the was excluded by fluorescence in situ hybridization (FISH). There frequency of ‘passenger’ mutations was low (only N502S of was no evidence for a rearrangement of chromosome 5 in any of DDR1 and P607H of EPHA8). Although the reason for this the patients with available cytogenetics. Primers were designed discrepancy is not known, one potential explanation is that to amplify the activation loops and/or juxtamembrane domains CMML may exhibit a lesser degree of genomic instability of all 90 known human tyrosine kinases, the JH2 domains of compared with acute myeloid leukemia. In summary, we find JAK1, JAK2, JAK3 and TYK2, as well as the entire coding no evidence that as yet unreported point mutations in the sequence of H-RAS, K-RAS and N-RAS. To exclude single activation loops, juxtamembrane domains and pseudokinase nucleotide polymorphisms, all candidate mutations were domains of tyrosine kinases are a major pathogenetic compared with public databases and, where appropriate, with mechanism in CMML and aCML. a panel of 96 samples of normal genomic DNA from individuals of similar ethnic background (for a list of primers and conditions, see Loriaux et al.13). Acknowledgements A total of 298 exons were sequenced, with good quality traces (Phred quality score of at least 2014) in 286 exons (96%). For 12 This study was supported in part by NHLBI Grant HL082978-01 exons, we were unable to obtain good quality sequence, despite (MWD), and the Leukemia and Lymphoma Society (MWD). multiple attempts using a variety of primers for amplification JW Tyner1,6, MM Loriaux2,6, H Erickson1, CA Eide1, and sequencing (Supplementary Table 1). We detected JAK2 J Deininger1, M MacPartlin1, SG Willis1, T Lange3, V617F in three patients (9%). These patients have been reported BJ Druker1,4, T Kovacsovics1, R Maziarz1, N Gattermann5 and previously.8 A number of novel sequence variations were MW Deininger1 identified. However, on comparison with a set of 96 DNA 1Division of Hematology and Medical Oncology, Oregon samples from normal individuals, these were all identified as Health and Science University Cancer Institute, Portland, OR, USA; single nucleotide polymorphisms with the exception of one 2 sequence variation in EPHA8 (P607H) (Supplementary Table 2). Department of Pathology, Oregon Health and Science This residue is highly promiscuous across the EPH kinase family University, Portland, OR, USA; 3Department of Hematology/Oncology, University of Leipzig, and various species (Supplementary Figure 1) and therefore Leipzig, Germany; likely also represents a rare polymorphism or a ‘passenger 4Howard Hughes Medical Institute, Portland, OR, USA and mutation’. Thus, we did not detect novel or known mutations in 5Klinik fu¨rHa¨matologie, Onkologie und klinische any tyrosine kinase other than JAK2, including those in which Immunologie, Heinrich-Heine-Universita¨tDu¨sseldorf, mutations were previously reported in CMML (KIT, FLT3, Du¨sseldorf, Germany CSF-1R) (Table 1). We detected RAS mutations in nine patients E-mail: [email protected] 6 (22%), six in K-RAS and three in N-RAS, including two ‘non- These authors contributed equally to this work. canonical’ K-RAS mutations (T74P and A146T), which were found in one and two patients, respectively. Functional References characterization of these mutations is reported elsewhere (manuscript under review) in the context of several novel RAS 1 Vardiman JW, Brunning RD, Harris NL. Chronic myeloproliferative mutations associated with leukemia (Table 1). diseases. In: Jaffe ES, Harris NL, Stein H, Vardiman JW (eds). Our results suggest that point mutations in established Tumors of Haematopoietic and Lymphoid Tissues. IARC Press: mutational hot spots in leukemia-associated tyrosine kinases, Lyon, 2001, pp 15–59. that is, the juxtamembrane domain and the activation loops, 2 Lauchle JO, Braun BS, Loh ML, Shannon K. Inherited predisposi- are rare in patients with CMML and aCML and that as yet tions and hyperactive Ras in myeloid leukemogenesis. Pediatr Blood Cancer 2006; 46: 579–585. unidentified genetic lesions must be responsible in those 3 Hirsch-Ginsberg C, LeMaistre AC, Kantarjian H, Talpaz M, Cork A, patients who do not exhibit mutations in RAS or in any of the Freireich EJ et al. RAS mutations are rare events in Philadelphia other tyrosine kinases previously implicated. Such unidentified chromosome-negative/bcr gene rearrangement-negative chronic

Leukemia Letters to the Editor 409 myelogenous leukemia, but are prevalent in chronic myelomono- 9 Lorenzo F, Nishii K, Monma F, Kuwagata S, Usui E, Shiku H. Muta- cytic leukemia. Blood 1990; 76: 1214–1219. tional analysis of the KIT gene in myelodysplastic syndrome (MDS) 4 Loh ML, Martinelli S, Cordeddu V, Reynolds MG, Vattikuti S, and MDS-derived leukemia. Leuk Res 2006; 30: 1235–1239. Lee CM et al. Acquired PTPN11 mutations occur rarely in adult 10 Padua RA, Guinn BA, Al-Sabah AI, Smith M, Taylor C, Pettersson T patients with myelodysplastic syndromes and chronic myelo- et al. RAS, FMS and p53 mutations and poor clinical outcome monocytic leukemia. Leuk Res 2005; 29: 459–462. in myelodysplasias: a 10-year follow-up. Leukemia 1998; 12: 5 Quesnel B, Preudhomme C, Vanrumbeke M, Vachee A, Lai JL, 887–892. Fenaux P. Absence of rearrangement of the neurofibromatosis 1 11 Reiter A, Sohal J, Kulkarni S, Chase A, Macdonald DH, Aguiar RC (NF1) gene in myelodysplastic syndromes and acute myeloid et al. Consistent fusion of ZNF198 to the fibroblast growth factor leukemia. Leukemia 1994; 8: 878–880. receptor-1 in the t(8;13)(p11;q12) myeloproliferative syndrome. 6 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF Blood 1998; 92: 1735–1742. receptor beta to a novel ets-like gene, tel, in chronic myelo- 12 Deguchi K, Gilliland DG. Cooperativity between mutations in monocytic leukemia with t(5;12) chromosomal translocation. tyrosine kinases and in hematopoietic transcription factors in AML. Cell 1994; 77: 307–316. Leukemia 2002; 16: 740–744. 7 Lee BH, Tothova Z, Levine RL, Anderson K, Buza-Vidas N, Cullen 13 Loriaux MM, Levine RL, Tyner JW, Frohling S, Scholl C, Stoffregen DE et al. FLT3 mutations confer enhanced proliferation and EP et al. High-throughput sequence analysis of the tyrosine kinome survival properties to multipotent progenitors in a murine model in acute myeloid leukemia. Blood 2008; 111: 4788–4796. of chronic myelomonocytic leukemia. Cancer Cell 2007; 12: 14 Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD et al. 367–380. The consensus coding sequences of human breast and colorectal 8 Levine RL, Loriaux M, Huntly BJ, Loh ML, Beran M, Stoffregen E cancers. Science 2006; 314: 268–274. et al. The JAK2V617F activating mutation occurs in chronic 15 Tomasson MH, Xiang Z, Walgren R, Zhao Y, Kasai Y, Miner T myelomonocytic leukemia and acute myeloid leukemia, but not in et al. Somatic mutations and germline sequence variants in the acute lymphoblastic leukemia or chronic lymphocytic leukemia. expressed tyrosine kinase genes of patients with de novo acute Blood 2005; 106: 3377–3379. myeloid leukemia. Blood 2008; 111: 4797–4808.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Rapid diagnosis of ataxia-telangiectasia by flow cytometric monitoring of DNA damage-dependent ATM phosphorylation

Leukemia (2009) 23, 409–414; doi:10.1038/leu.2008.195; is present as a dimer or higher-order multimer in unstressed cells published online 17 July 2008 with the kinase domain bound to the region surrounding serine (Ser) 1981. Cellular irradiation induces rapid intermolecular autophosphorylation of Ser 1981 and this phosphorylation event Ataxia-telangiectasia (A-T) is an autosomal recessive disorder results in dimer dissociation and initiation of cellular ATM characterized by cerebellar ataxia, telangiectasias, immune kinase activity.5 defects and a predisposition to malignancy. The birth frequency Diagnosis of A-T is based on clinical features combined with of A-T is estimated to be about 1 in 100 000–300 000. The laboratory findings. Several cell biological and molecular A-T heterozygote frequency is estimated to be 0.5–1%. The approaches, such as protein truncation assay, yeast-based patient shows progressive cerebellar ataxia since infancy. protein truncation assay, western blotting, RNA restriction finger Telangiectasia typically develops after 3–5 years of age. Sixty printing assay, and others, have been used to reveal ATM to 80% of patients show immunodeficiency. Twenty to 30% of mutation prior to straightforward genome sequencing.2 How- patients develop a malignancy, mainly leukemia or lymphoma. ever, there is no standard method that could be applied in Laboratory finding shows decreased level of serum IgA, IgG2 clinical diagnosis, genetic counseling or carrier prediction. and IgE. Increased level of serum a-fetoprotein is a specific Here, we describe a novel method for A-T diagnosis using feature. Peripheral circulating lymphocytes show characteristic measurement of phosphorylation status of ATM by flow intra-locus rearrangements involving T-cell receptor and/or cytometry. immunoglobulin loci. Chromosomal breakage and hypersensi- DNA damage response occurring in cells was monitored by tization to ionizing radiation (IR) are a hallmark cell biological measuring the phosphorylation level of ATM at Ser 1981 using feature of the A-T cell. A-T cells also show improper cell cycle the flow cytometry technique. Phosphorylation of ATM at Ser regulation after IR, which has been known as radioresistant 1981 is the hallmark of ATM activation and can be detected DNA synthesis. by the anti-phospho-ATM-specific antibody 10H11.E12 (Cell The responsible gene, ataxia-telangiectasia mutated (ATM), Signaling, Danvers, MA, USA). Epstein–Barr virus (EBV)- contains 66 exons spanning approximately 150 kb of genomic transformed wild-type lymphoblastoid cell line (LCL-WT) was DNA at 11q22.3. cDNA contains 10 140 bp. The ATM protein irradiated or treated with hydrogen peroxide (H2O2) to induce contains 3056 amino acids coded by a 9168-bp open reading DNA damage. Dose-dependent phosphorylation of ATM was 1,2 frame. The ATM gene is also known to be mutated detected 1 h after irradiation or H2O2 treatment by flow secondarily in various hematological malignancies, and is cytometry. An irradiation dose of 0.5 Gy was sufficient to speculated to work as a tumor suppressor gene.3 ATM protein induce phosphorylation of ATM at Ser 1981 and the level of is one of the members of phosphoinositide 3-kinase-related phosphorylation increased in a dose-dependent manner kinases. Exposure of cells to genotoxic stresses such as anti- (Figure 1a). Phosphorylation of ATM was seen in cells treated cancer drugs or irradiation induces DNA double-strand breaks. with 0.1 mM H2O2, and the level of phosphorylation increased 4 The central player in such DNA damage response is ATM. ATM in a dose-dependent manner up to 0.5 mM.

Leukemia