Letters to the Editor 743 REFERENCES 8 Pekarsky Y, Palamarchuk A, Maximov V, Efanov A, Nazaryan N, Santanam U et al. 1 Sgambati MLM, Devesa S. Chronic Lymphocytic Leukemia, Epidemiological, Famil- Tcl1 functions as a transcriptional regulator and is directly involved in the iar, and Genetic Aspects. Bruce Cheson, Ed Marcel Dekker, Inc: New York, 2001, pp pathogenesis of CLL. Proc Natl Acad Sci USA 2008; 105: 19643–19648. 33–62. 9 Harper JW, Elledge SJ, Keyomarsi K, Dynlacht B, Tsai LH, Zhang P et al. Inhibition 2 Fabbri G, Rasi S, Rossi D, Trifonov V, Khiabanian H, Ma J et al. Analysis of the of cyclin-dependent kinases by p21. Mol Biol Cell 1995; 6: 387–400. chronic lymphocytic leukemia coding genome: role of NOTCH1 mutational 10 Mullican SE, Zhang S, Konopleva M, Ruvolo V, Andreeff M, Milbrandt J et al. activation. J Exp Med 2011; 208: 1389–1401. Abrogation of nuclear receptors Nr4a3 and Nr4a1 leads to development of acute 3 Puente XS, Pinyol M, Quesada V, Conde L, Ordonez GR, Villamor N et al. Whole- myeloid leukemia. Nat Med 2007; 13: 730–735. genome sequencing identifies recurrent mutations in chronic lymphocytic 11 Tanoue T, Moriguchi T, Nishida E. Molecular cloning and characterization of a leukaemia. Nature 2011; 475: 101–105. novel dual specificity phosphatase, MKP-5. J Biol Chem 1999; 274: 19949–19956. 4 Balatti V, Bottoni A, Palamarchuk A, Alder H, Rassenti LZ, Kipps TJ et al. NOTCH1 12 Stawowczyk M, Van Scoy S, Kumar KP, Reich NC. The interferon stimulated gene mutations in CLL associated with trisomy 12. Blood 2012; 119: 329–331. 54 promotes apoptosis. J Biol Chem 2011; 286: 7257–7266. 5 Aster JC, Blacklow SC, Pear WS. Notch signalling in T-cell lymphoblastic 13 Huang RP, Fan Y, de Belle I, Niemeyer C, Gottardis MM, Mercola D et al. Decreased leukaemia/lymphoma and other haematological malignancies. J Pathol 2011; 223: Egr-1 expression in human, mouse and rat mammary cells and tissues correlates 262–273. with tumor formation. Int J Cancer 1997; 72: 102–109. 6 Koch U, Radtke F. Notch in T-ALL: new players in a complex disease. Trends 14 Shields JM, Christy RJ, Yang VW. Identification and characterization of a gene Immunol 2011; 32: 434–442. encoding a gut-enriched Kruppel-like factor expressed during growth arrest. J Biol 7 Palomero T, Lim WK, Odom DT, Sulis ML, Real PJ, Margolin A et al. NOTCH1 Chem 1996; 271: 20009–20017. directly regulates c-MYC and activates a feed-forward-loop transcriptional 15 Xiao S, Li D, Zhu HQ, Song MG, Pan XR, Jia PM et al. RIG-G as a key mediator of the network promoting leukemic cell growth. Proc Natl Acad Sci USA 2006; 103: antiproliferative activity of interferon-related pathways through enhancing p21 18261–18266. and p27 proteins. Proc Natl Acad Sci USA 2006; 103: 16448–16453.

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

Imatinib-dependent tyrosine phosphorylation profiling of Bcr-Abl- positive chronic myeloid leukemia cells

Leukemia (2013) 27, 743–746; doi:10.1038/leu.2012.243 signaling were evaluated in response to Imatinib. Although Imatinib strongly reduced tyrosine phosphorylation, none of the five investigated proteins were degraded (Figure 1a). For qMS Bcr-Abl is the major cause and pathogenetic principle of chronic studies, we treated cells with 0, 1 and 10 mM Imatinib for 4 h as in myeloid leukemia (CML). Bcr-Abl results from a chromosomal previous studies (Figure 1b).5,7 Cell lysates were subsequently translocation that fuses the bcr and abl genes, thereby generating digested with the proteases Lys-C and trypsin, followed by stable a constitutively active tyrosine kinase, which stimulates several isotope dimethyl labeling of the resulting peptides. The labeling signaling networks required for proliferation and survival. regime allows to distinguish three peptide pools (light, Bcr-Abl’s oncogenic properties comprise both a kinase and a intermediate and heavy), which were mixed in equal scaffold protein.1 A number of Bcr-Abl interaction partners and concentrations. Subsequently, tyrosine-phosphorylated peptides downstream effectors have been described, improving our were enriched by immunoprecipitation and analyzed by liquid understanding of the signaling networks deranged in CML. The chromatography–mass spectrometry (LC-MS).6 Analysis of the ‘core interactome of Bcr-Abl’ entails seven major interaction quantitative changes in tyrosine phosphorylation yielded 201 partners: GRB2, Shc1, Crk-l, c-Cbl, p85, Sts-1 and SHIP2.2 unique quantifiable phosphotyrosine peptide triplets belonging to The introduction of the Bcr-Abl tyrosine kinase inhibitor (TKI) 141 proteins (Supplementary Table S1), by far exceeding all Imatinib (Gleevec) has been a landmark in the treatment of CML.3 previous reports.5,7 Of these, 87 peptides showed at least a However, the development of Imatinib resistance poses major twofold downregulation after treatment with Imatinib challenges to the clinical management of CML, and although (Supplementary Table S2 and Figure 1c). second-generation TKIs can block many Imatinib-resistant Imatinib significantly decreased the tyrosine phosphorylation of mutants they are ineffective against the common T315I many peptides originating from Bcr-Abl and its core interactors mutation. An alternative strategy is to circumvent Imatinib (c-Cbl, CrkL and SHIP-2, Figure 1c). Furthermore, several resistance by targeting downstream pathways essential for proteins that have been shown to have pivotal roles in transformation.4 Therefore, it is important to fully understand Bcr-Abl-dependent signaling (Gab1, Gab2, Shc1, Crk, ERK-2, these pathways. Large-scale (phospho)proteomics experiments STAT5A/B and Yes) displayed reduced tyrosine phosphorylation, have addressed the phosphoproteome of Bcr-Abl-positive cells, often on multiple sites. In addition, Src family kinase substrates leading to the identification of an impressive number of serine/ exhibited reduced tyrosine phosphorylation, for example, threonine-phosphorylated sites (for example, Pan et al.5), although Cortactin, Catenin delta-1, nPKC-delta and Paxillin. Finally, Imatinib tyrosine-phosphorylation events remained underrepresented. reduced the tyrosine phosphorylation of several proteins involved However, many of the protein–protein interactions in the in cytoskeletal regulation, such as MEMO1, Intersectin-2, Catenin oncogenic Bcr-Abl network are dependent on tyrosine delta-1, HEPL, GRF-1, Centaurin delta 2 and Plakophilin, which phosphorylation, either directly or indirectly through Bcr-Abl.2 have not been linked to Bcr-Abl signaling. Here, we have enriched and identified phosphotyrosine A motif analysis on the sequences of significantly down- peptides by quantitative mass spectrometry (qMS)6 in order to regulated tyrosine-phosphorylated sites revealed a distinct examine the effect of Imatinib in the CML blast crisis cell line enriched motif, YxxP (Figures 1d and e). In total, 80% (23 out of K562 (Supplementary Methods). First, the phosphorylation 29) of the peptides harboring this phosphorylation motif were status and stability of several key proteins involved in Bcr-Abl significantly downregulated upon Imatinib treatment. The YxxP

Accepted article preview online 27 August 2012; advance online publication, 14 September 2012

& 2013 Macmillan Publishers Limited Leukemia (2013) 718 – 757 Letters to the Editor 744

Figure 1. Exploring the effect on tyrosine phosphorylation upon Imatinib inhibition of K562 cells. (a) K562 cells were treated with Imatinib or vehicle for 4 h. Western blotting was performed to assess the abundance of several key proteins of Bcr-Abl-dependent signaling, as well as the effect of Imatinib treatment (10 mM) on total phosphotyrosine levels. (b) Overview of the quantitative proteomics workflow. Cells were treated with different doses of Imatinib (0, 1 and 10 mM) for 4 h, followed by cell lysis and protein digestion. The peptides from each Imatinib treatment were then differentially labeled using stable isotope dimethyl labeling. The three differentially labeled digests were combined, followed by simultaneous enrichment of tyrosine-phosphorylated peptides using immobilized phosphotyrosine-specific antibodies. The enriched fraction was analyzed by LC-MS and changes in tyrosine phosphorylation quantified. (c) Quantitative profiles of site-specific tyrosine phosphorylation upon Imatinib inhibition of Bcr-Abl. The changes in tyrosine phosphorylation versus the control are represented on a log scale, marked in red, phosphorylation sites containing the YxxP motif. (d) Peptides showing a more than twofold decrease in tyrosine phosphorylation upon treatment with Imatinib were subjected to motif analysis using the Motif-X algorithm. The YxxP motif shown was significantly overrepresented in the data set. (e) Frequency of the motif YxxP is very prominent in the downregulated phosphotyrosine peptides; 87 of all 201 observed peptides are downregulated (43%), whereas 23 of 29 YxxP-containing peptides are downregulated (79%).

Figure 2. (a) Imatinib treatment of K562 cells leads to destruction of the core Bcr-Abl complex (green; adapted from Brehme et al.2). A small selection of peripheral interactors are depicted in blue. GAB2 (yellow) is used as flag-tagged entry point for interaction analysis upon Imatinib treatment. LC-MS/MS revealed the release of GAB2 from the Bcr-Abl core proteome, except GRB2 (bold red). Dashed lines indicate yet-to-be- explored lost/intact interactions within the Bcr-Abl Core proteome upon Imatinib treatment. (b) Mutation of all regulated GAB2 tyrosine- phosphorylated residues to phenylalanine reveals abolishment of cell periphery localization of both GAB2 and probably also the Bcr-Abl Core interactome.

motif resembles a classic binding site for SH2 domains, and a Most of the detected tyrosine phosphorylation sites on Bcr-Abl consensus target sequence for phosphorylation by Bcr-Abl.8 were completely abolished by Imatinib. Several of these have Peptides found in our screen containing this motif include been linked to the activity of signaling pathways downstream Bcr-Abl (Tyr115 and 128), STAT5A and B (Tyr682 and 699, of Bcr-Abl.9 In contrast, two Bcr-Abl residues (Tyr253 and 257) respectively), several sites in the adaptor proteins Gab1 (Tyr259, were insensitive to Imatinib, suggesting that they may not be 373 and 406) and Gab2 (Tyr266 and 409) and all four essential for downstream oncogenic signaling. phosphotyrosine-containing peptides of the docking protein Several members of the Bcr-Abl core proteome2 are adaptor/ HEPL (Tyr174, 195, 244 and 329), indicating that the latter may scaffold proteins required for the generation of SH2- or PTB- represent a new Bcr-Abl substrate. domain-binding sites that lead to tightly regulated protein

Leukemia (2013) 718 – 757 & 2013 Macmillan Publishers Limited Letters to the Editor 745 interactions.10 Several of these peptides also contain the In summary, to probe the effect of Imatinib on tyrosine kinase consensus sequence YxxP of the enriched phosphorylation motif Bcr-Abl-positive cells, we applied a targeted qMS-based proteo- (Figures 1c and d). As these proteins bind to Bcr-Abl, the known mics approach. We used phosphotyrosine-specific immuno- and novel phosphorylation sites detected here may be putative capturing to systematically screen for Imatinib-dependent tyrosine Bcr-Abl substrates. The small adaptor protein GRB2, a member of phosphorylation events in K562 cells. Our data reveal that several the Bcr-Abl core proteome, has a key role in Bcr-Abl signaling core Bcr-Abl interactors are putative direct substrates. In following through binding to phosphorylated Tyr177 on Bcr-Abl. This is up a novel Imatinib-inhibited target, the adaptor protein Gab2 was required for activation of the Raf-MEK-ERK pathway in leukemic found to regulate localization of the Bcr-Abl core complex to the cells.11 GRB2 interacts with the docking proteins Gab1 and Gab2 plasma membrane in a phosphorylation-dependent manner, thus via GRB2’s SH3 domain and the proline-rich regions in both enabling us to add crucial new dimensions of detail to the Bcr-Abl Gab proteins (Supplementary Figure S1A).12 The observed tyrosine signaling network. phosphorylations of both Gab1 and 2 are downregulated by Imatinib (Supplementary Table S2). Therefore, we quantitatively evaluated the changes in the Gab2 interactome in response to CONFLICT OF INTEREST Imatinib treatment. Parallel FLAG immunoprecipitations from The authors declare no conflict of interest. K562 cells transfected with either a empty FLAG vector or a FLAG-Gab2-encoding construct and treated with or without 10 mM Imatinib were performed. This affinity-based qMS approach ACKNOWLEDGEMENTS revealed a severe alteration in the Gab2 interactome after We like to acknowledge the Fondation Leducq: the Alliance for CamKII signaling in Imatinib treatment (Figure 2 and Supplementary S1B-C and heart disease (CP, AS, AJRH) and the Netherlands Proteomics Centre, embedded in Table S3). Whereas GRB2 remained tightly bound, interactions the Netherlands Genomics Initiative (CP, AS, AJRH), and the Science Foundation between Gab2 and Shp2, as well as with components of the Ireland under Grant No. 06/CE/B1129 (WK) for financial support. We would like to Bcr-Abl core interactome (for example, Bcr-Abl, SHIP-2 and thank Tessa Holyoake (Glasgow) for critical reading of the manuscript and helpful Shc1), were severely disrupted, suggesting that the Gab2 discussions. interactome is dependent on the tyrosine phosphorylation state of this protein. Subsequent immunoprecipitation of endogenous C Preisinger1,2,7, JP Schwarz3, OB Bleijerveld1,2, E Corradini1,2, Gab2 and western blot detection of interactors confirm the PJ Mu¨ller4, KI Anderson3, W Kolch5, A Scholten1,2 and AJR Heck1,2,6 qMS results: Bcr-Abl, Shp2 and the p85 subunit of PI3 Kinase, all 1Biomolecular Mass Spectrometry and Proteomics Group, known Gab2 interactors, fail to coimmunoprecipitate with Bijvoet Center for Biomolecular Research and Utrecht Institute for dephosphorylated Gab2 (Supplementary Figure S1D). Gab2 has Pharmaceutical Sciences, Utrecht University, Utrecht, The Netherlands; been previously shown to coimmunoprecipitate with Bcr-Abl in 2Netherlands Proteomics Centre, Utrecht, The Netherlands; Ba/F3 cells.13 Furthermore, the intracellular localization of Gab2 3The Beatson Institute for Cancer Research, and the closely related Gab1 protein was evaluated by confocal Garscube Estate, Glasgow, UK; microscopy. K562 cells were transfected with GFP-tagged 4Department of Biochemistry, RWTH Aachen constructs of wild-type or mutant Gab1 and Gab2, where all University, Aachen, Germany; observed Imatinib-regulated phosphotyrosine residues were 5Systems Biology Ireland, University College Dublin, changed to phenylalanine. Wild-type GFP-Gab2 smoothly Belfield, Dublin, Ireland and stained the circumference of the cell, whereas the mutant 6Centre for Biomedical Genetics, Utrecht, The Netherlands protein showed a severe alteration in localization with speckled E-mail: [email protected]; [email protected] staining of the cell periphery and cytosolic staining (Figure 2b and 7Current address: Proteomics Facility, Interdisciplinary Centre for Supplementary Figure S2). Similar changes in localization were Clinical Research (IZKF) Aachen, RWTH Aachen University, observed with wild-type and mutant GFP-Gab1. Thus, tyrosine Aachen, Germany. phosphorylation of Gab1 and Gab2 and the subsequent ability to promote the formation of a complex with Bcr-Abl and other core members appear to be required for appropriate membrane REFERENCES localization. These observations strongly advocate an involvement 1 Kolch W, Pitt A. Functional proteomics to dissect tyrosine kinase signalling of Gab2 in the assembly of the Bcr-Abl signaling network at the pathways in cancer. Nat Rev Cancer 2010; 10: 618–629. plasma membrane and the ability of the core proteome to 2 Brehme M, Hantschel O, Colinge J, Kaupe I, Planyavsky M, Kocher T et al. Charting transform cells. the molecular network of the drug target Bcr-Abl. Proc Natl Acad Sci USA 2009; Several affected phosphotyrosine proteins identified in our 106: 7414–7419. screen can be linked to cytoskeletal processes, such as regulation 3 Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S et al. Effects of a of the actin cytoskeleton, cell migration or invasion. Among these, selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive HEPL, a member of the p130CAS family of scaffold proteins that is cells. Nat Med 1996; 2: 561–566. involved in cell migration,14 showed dephosphorylation at four 4 Helgason GV, Karvela M, Holyoake TL. Kill one bird with two stones: potential different sites in response to Imatinib treatment (Figure 1c). efficacy of BCR-ABL and autophagy inhibition in CML. Blood 2011; 118: 2035–2043. 5 Pan C, Olsen JV, Daub H, Mann M. Global effects of kinase inhibitors on signaling Notably, HEPL was identified as a binding partner of Crk-l in K562 2 CAS networks revealed by quantitative phosphoproteomics. Mol Cell Proteomics 2009; cells. The closely related family members p130 (BCAR1) and 8: 2796–2808. HEF1 (CAS-L) have both been identified as binding partners of 6 Boersema PJ, Foong LY, Ding VM, Lemeer S, van Breukelen B, Philp R et al. Crk-l (and other Crk family members) and are downstream In-depth qualitative and quantitative profiling of tyrosine phosphorylation using phosphorylation targets of Bcr-Abl.14 In addition, we also a combination of phosphopeptide immunoaffinity purification and stable isotope identified CD2-associated protein, TRIP-6, Intersectin-2 and dimethyl labeling. Mol Cell Proteomics 2010; 9: 84–99. L-Plastin, all scaffold proteins involved in the modulation of the 7 Goss VL, Lee KA, Moritz A, Nardone J, Spek EJ, MacNeill J et al. A actin cytoskeleton, as being dephosphorylated by Imatinib common phosphotyrosine signature for the Bcr-Abl kinase. Blood 2006; 107: treatment. These findings are in agreement with the literature 4888–4897. 8 Amanchy R, Periaswamy B, Mathivanan S, Reddy R, Tattikota SG, Pandey A. A suggesting that Bcr-Abl affects the cytoskeleton and cell 14,15 curated compendium of phosphorylation motifs. Nat Biotechnol 2007; 25: motility, potentially opening up new avenues of investiga- 285–286. tion to follow up the role of cytoskeletal changes in Bcr-Abl 9 Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous transformation. leukaemia. Nat Rev Cancer 2005; 5: 172–183.

& 2013 Macmillan Publishers Limited Leukemia (2013) 718 – 757 Letters to the Editor 746 10 Pawson T. Specificity in signal transduction: from phosphotyrosine-SH2 domain 13 Sattler M, Mohi MG, Pride YB, Quinnan LR, Malouf NA, Podar K et al. Critical role for interactions to complex cellular systems. Cell 2004; 116: 191–203. Gab2 in transformation by BCR/ABL. Cancer cell 2002; 1: 479–492. 11 Pendergast AM, Quilliam LA, Cripe LD, Bassing CH, Dai Z, Li N et al. BCR-ABL- 14 Tikhmyanova N, Little JL, Golemis EA. CAS proteins in normal and pathological cell induced oncogenesis is mediated by direct interaction with the SH2 domain of growth control. Cell Mol Life Sci 2010; 67: 1025–1048. the GRB-2 adaptor protein. Cell 1993; 75: 175–185. 15 Preisinger C, Kolch W. The Bcr-Abl kinase regulates the actin cytoskeleton 12 Wohrle FU, Daly RJ, Brummer T. Function, regulation and pathological roles of the via a GADS/Slp-76/Nck1 adaptor protein pathway. Cell Signal 2010; 22: Gab/DOS docking proteins. Cell Commun Signal 2009; 7:22. 848–856.

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

Are ARID5B and IKZF1 polymorphisms also associated with childhood acute myeloblastic leukemia: the ESCALE study (SFCE)?

Leukemia (2013) 27, 746–748; doi:10.1038/leu.2012.244 the cases, the indicator predicted (with 98.2% sensitivity and 94.3% specificity) the CEPH classification, ‘Caucasian derived’, by principal component analysis, from 96 609 SNPs. It was assumed that having Of the average 470 new cases of acute leukemia (AL) registered at least two European-born grandparents was an equally good annually in France among children aged o15 years,1 B90 are proxy for the controls, for whom the SNPs were not available. The cases of acute myeloblastic leukemia (AML). Down’s syndrome 51 AML cases of European descent consisted of 8 M1, 10 M2, 5 M3, 4 and other inheritable predisposing diseases explain only a small M4, 15 M5, 4 M7 and 5 unspecified AML. proportion of AML. Most childhood AML occurs before the age of The allelic dose of imputed SNPs was estimated and included as 2 years, and researches have mainly focused on antenatal or early a continuous variable in the logistic models. Genotypic ORs were childhood exposures.2 Advances in genetic technologies have estimated using the homozygote common genotypes as refer- enabled genome-wide association studies (GWAS) to be under- ences. Haplotypes were also inferred using the Expectation- taken with the aim of identifying common single-nucleotide Maximization algorithm implemented in PLINK, and posterior polymorphism variant alleles associated with small increases in AL haplotype probabilities were used as weights in the logistic models. risk. GWAS on acute lymphoblastic leukemia (ALL) have previously Table 1 presents the associations between AML and the identified SNPs in ARID5B (rs7073837, rs10740055, rs7089424, candidate polymorphisms in IKZF1 and ARID5B. The allelic ORs rs10821936 and rs10994982)3–6 and IKZF1 (rs6964823, rs4132601, for the associations between ARID5B SNPs and AML were greater rs6944602 and rs11978267).3–6 In the present paper, the than one, but the SNPs of ARID5B did not fit co-dominant associations between childhood AML and the polymorphisms models well: carrying the variant alleles of rs7073837 (A), previously identified for ALL in ARID5B and IKZF1 were rs10740055 (C) and rs7089424 (G) (dominant coding) was investigated in a candidate gene approach, using the data from associated with odds ratio (OR) of 2.0 (1.0–4.0), 3.3 (1.3–8.6) the ESCALE case–control study. and 1.9 (1.0–3.6), respectively, whereas the corresponding allelic The ESCALE study has been previously described in detail.7,8 ORs were equal to 1.2 (0.8–1.8), 1.3 (0.9–2.1) and 1.3 (0.9–2.1). Briefly, the cases were children o15 years old diagnosed with AL, AML was inversely related to carrying the variant allele of lymphoma, central nervous system tumor or neuroblastoma in rs4132601 in IKZF1 (OR ¼ 0.5 (0.3–0.9)), on the contrary to what 2003–2004 and identified through the National Registry of has been observed for ALL.3–6 The numbers were small, but Childhood Hematopoietic Malignancies. The study included 101 none of the relationships seems to point specifically to any AML cases, 92% of the 110 eligible cases. A blood sample was cytological subtype of AML (not shown). The associations were not obtained from 89 cases (88% of the included cases), 66 of whom modified after adjustment for parental professional category and for were successfully genotyped. The ESCALE controls were randomly the environmental, immune or lifestyle factors associated with AML selected contemporaneously with the cases from French house- in the present study.8–10 The study of the haplotypes, based on small holds with a landline telephone. Of the 2360 eligible controls numbers, suggested that the association with ARID5B mainly relied identified, 1681 controls (71%) were included, 810 of whom on rs10740055, reinforced with either none or both of the two other provided buccal swab brushes, and 570 were successfully polymorphisms, and showed that the association with IKZF1 was genotyped for candidate SNPs. driven by rs4132601 alone (Supplementary Table 1). The cases were genotyped on an Illumina 370K Quad BeadChip Somatic deletion of IKZF1 is known to be associated with (San Diego, CA, USA) and the controls for 4868 candidate SNPs on survival in childhood ALL.11 In the present study, 1- and 5-year Illumina Infinium iSelect custom beadchips (IntegraGen, Evry, survival were similar between non-genotyped AML cases and France), including rs7073837, rs10740055 and rs7089424 in ARID5B, genotyped AML cases, as well as between non-carriers and carriers and rs6964823, rs4132601 and rs6944602 in IKZF1. Quality control of rs4132601 variant. Therefore, selection of surviving AML case is excluded 109 ESCALE controls with an individual call rate o95% unlikely to explain our results. and 10 AML cases with individual call success o97%, low or high A recent candidate gene study reported a significant association heterozygosity or discrepancies between stated sex and sexual between infant AML and rs11978267 in IKZF1 (allelic OR equal to chromosomes. Mean individual call rates were 99.9% for the cases 1.8 (1.2–2.6)).12 The authors did not investigate rs11978267 in and 98.5% for the controls. To control for potential population IKZF1 and they reported no association between ARID5B SNPs and stratification bias, the analyses were restricted to the 51 cases and infant AML. The sample size of the present study did not allow 442 controls who had at least two European-born grandparents. For restriction of the analysis to the children o1-year-old.

Accepted article preview online 27 August 2012; advance online publication, 11 September 2012

Leukemia (2013) 718 – 757 & 2013 Macmillan Publishers Limited