Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia

Article

Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia Graphical Abstract

Authors Michelle L. Churchman, Maoxiang Qian, Geertruy te Kronnie, ..., Kim E. Nichols, Jun J. Yang, Charles G. Mullighan

Correspondence [email protected] (J.J.Y.), [email protected] (C.G.M.)

In Brief Churchman et al. identify 28 unique germline IKZF1 coding variants in 45 children with acute lymphoblastic leukemia. Many of these variants are not predicted to be damaging using in silico prediction tools, but functional tests reveal that the majority of them have deleterious effects on IKAROS function.

Highlights d

d

d

d

Germline coding IKZF1 variants are present in familial and sporadic B-precursor ALL Most variants affect regions outside of known domains and perturb IKAROS function Germline IKZF1 variants result in aberrant adhesion and bone marrow mislocalization Germline IKZF1 variants result in reduced antileukemic drug sensitivity

Churchman et al., 2018, Cancer Cell 33, 1–12 May 14, 2018 ª 2018 Elsevier Inc. https://doi.org/10.1016/j.ccell.2018.03.021

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Cancer Cell

Article Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia Michelle L. Churchman,1,22 Maoxiang Qian,2,22 Geertruy te Kronnie,3,22 Ranran Zhang,4 Wenjian Yang,2 Hui Zhang,2,4 Tobia Lana,3 Paige Tedrick,5 Rebekah Baskin,5 Katherine Verbist,5 Jennifer L. Peters,6 Meenakshi Devidas,7 Eric Larsen,8 Ian M. Moore,1 Zhaohui Gu,1 Chunxu Qu,1 Hiroki Yoshihara,1 Shaina N. Porter,9 Shondra M. Pruett-Miller,9 (Author list continued on next page) 1Department

of Pathology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA of Pharmaceutical Sciences, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 3Department of Women’s and Children’s Health, University of Padova, 35128 Padova, Italy 4Department of Pediatrics, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, 510120 Guangdong, China 5Department of Oncology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 6Cellular Imaging Shared Resource, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 7Department of Biostatistics, Epidemiology and Health Policy Research, College of Medicine, University of Florida, Gainesville, FL 32610, USA 8Maine Children’s Cancer Program, Scarborough, ME 04074, USA 9Center for Advanced Genome Engineering, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 10Department of Computational Biology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 11Division of Pediatric Hematology-Oncology, New York University, New York, NY 10016, USA 12Department of Pediatrics, Duke University, Durham, NC 27708, USA 13Cook Children’s Medical Center, Fort Worth, TX 76104, USA 14Pediatric Hematology Oncology, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA 15Institute for Genomic Medicine, Nationwide Children’s Hospital, Columbus, OH 43205, USA 2Department

(Affiliations continued on next page)

SUMMARY

Somatic genetic alterations of IKZF1, which encodes the lymphoid transcription factor IKAROS, are common in high-risk B-progenitor acute lymphoblastic leukemia (ALL) and are associated with poor prognosis. Such alterations result in the acquisition of stem cell-like features, overexpression of adhesion molecules causing aberrant cell-cell and cell-stroma interaction, and decreased sensitivity to tyrosine kinase inhibitors. Here we report coding germline IKZF1 variation in familial childhood ALL and 0.9% of presumed sporadic B-ALL, identifying 28 unique variants in 45 children. The majority of variants adversely affected IKZF1 function and drug responsiveness of leukemic cells. These results identify IKZF1 as a leukemia predisposition gene, and emphasize the importance of germline genetic variation in the development of both familial and sporadic ALL.

INTRODUCTION Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy, and despite cure rates approaching 90% re-

mains a leading cause of childhood cancer death (Hunger and Mullighan, 2015). ALL comprises multiple subtypes defined by distinct constellations of somatic structural genomic alterations and sequence mutations that perturb lymphoid maturation,

Significance Somatic IKZF1 alterations are a hallmark of high-risk B-progenitor acute lymphoblastic leukemia (ALL) and correlate with poor response to therapy. We identify germline coding IKZF1 variants in 0.9% of pediatric B-progenitor ALL cases. These variants affect residues throughout the protein with clustering outside of known annotated functional domains, many of which were not predicted to be damaging using in silico prediction tools. However, functional testing in vitro and in vivo revealed that the majority of germline IKZF1 variants have deleterious effects on IKAROS function, including impaired DNA binding and regulation of transcriptional targets, induction of aberrant leukemic cell adhesion, and impaired drug responsiveness. Collectively, these findings provide evidence that these germline alterations predispose to leukemia and potentially influence response to treatment. Cancer Cell 33, 1–12, May 14, 2018 ª 2018 Elsevier Inc. 1

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Gang Wu,10 Elizabeth Raetz,11 Paul L. Martin,12 W. Paul Bowman,13 Naomi Winick,14 Elaine Mardis,15 Robert Fulton,16 Martin Stanulla,17 William E. Evans,2,18 Mary V. Relling,2,18 Ching-Hon Pui,5,18 Stephen P. Hunger,19 Mignon L. Loh,20 Rupert Handgretinger,21 Kim E. Nichols,5,18 Jun J. Yang,2,5,18,* and Charles G. Mullighan1,18,23,* 16McDonnell

Genome Institute, Washington University School of Medicine, St. Louis, MO 63108, USA Hematology and Oncology, Hannover Medical School, Hannover 30625, Germany 18Hematological Malignancies Program, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA 19Department of Pediatrics and Center for Childhood Cancer Research, Children’s Hospital of Philadelphia and The Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA 19104, USA 20Department of Pediatrics, Benioff Children’s Hospital and Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA 94158, USA 21Department of Hematology/Oncology, Children’s University Hospital, 72076 Tuebingen, Germany 22These authors contributed equally 23Lead Contact *Correspondence: [email protected] (J.J.Y.), [email protected] (C.G.M.) https://doi.org/10.1016/j.ccell.2018.03.021 17Pediatric

tumor suppression, cell-cycle regulation, kinase signaling, and chromatin modification (Iacobucci and Mullighan, 2017). Several of these alterations contribute not only to leukemogenesis but also influence the risk of treatment failure and relapse. Notably, subtype-defining rearrangements of KMT2A (MLL), BCR-ABL1 fusion, and activating kinase alterations in Ph-like ALL are associated with poor outcome. Deletion or mutation of the lymphoid transcription factor gene IKZF1 also confers poor prognosis (Mullighan et al., 2009). There is increasing evidence for an inherited predisposition to the development of ALL. Genome-wide association studies have identified common non-coding polymorphisms that modestly influence the risk of ALL (Moriyama et al., 2015). Variants in IKZF1, CDKN2A, CDKN2B, ARID5B, CEBPE, PIP4K2A, and GATA3 exhibit the most reproducible associations with risk of ALL, with differences in strength of association according to genetic subtype of ALL or age at diagnosis (Perez-Andreu et al., 2013, 2015; Trevino et al., 2009). Genomic analysis of familial ALL has identified loss-of-function germline variations in genes encoding the hematopoietic transcription factors PAX5 and ETV6 and the tumor-suppressor gene TP53 (Auer et al., 2013; Duployez et al., 2018; Moriyama et al., 2015; Noetzli et al., 2015; Powell et al., 2013; Shah et al., 2013; Topka et al., 2015; Zhang et al., 2015). While uncommon, these kindreds are particularly informative by identifying rare and highly penetrant ALL risk variants. More importantly, damaging germline variants in these genes (e.g., ETV6 and TP53) have also been identified in a proportion of ALL cases otherwise thought to be sporadic, suggesting that inherited genetic variants have a greater role in ALL predisposition than previously recognized (Moriyama et al., 2015). Common germline non-coding IKZF1 polymorphisms are associated with susceptibility to ALL, and somatic IKZF1 alterations are common in high-risk B cell precursor ALL, particularly BCR-ABL1-positive (Ph+) and Ph-like ALL (Martinelli et al., 2009; Mullighan et al., 2008, 2009; Roberts et al., 2017; Van der Veer et al., 2014). IKZF1 encodes the founding member of the IKAROS family of zinc-finger transcription factors, and is a critical regulator of lymphoid development (Georgopoulos, 2017; Merkenschlager, 2010). IKAROS has six zinc fingers, four N-terminal and two C-terminal, which mediate DNA binding and dimerization, respectively. IKAROS has multiple functions including transcriptional activation and repression, which are mediated in part 2 Cancer Cell 33, 1–12, May 14, 2018

by interactions with the nucleosome remodeling deacetylase (Mi-2/NuRD) and SIN3 histone deacetylase complexes, and with Polycomb repressive complex 2 (PRC2) proteins (Hu et al., 2016; Kim et al., 1999; Koipally and Georgopoulos, 2002; O’Neill et al., 2000; Oravecz et al., 2015). The somatic IKZF1 alterations observed in B-progenitor ALL (B-ALL) include deletions that result in haploinsufficiency, intragenic deletions that remove the DNA-binding zinc fingers, and mutations of key DNA-binding residues, each of which result in loss or impairment of DNA binding. Somatic IKZF1 alterations result in the loss of IKAROS activity and the acquisition of stem cell-like features, overexpression of adhesion molecules causing aberrant leukemia cell-cell and cell-stroma interactions, and decreased sensitivity to tyrosine kinase inhibitors (Churchman et al., 2015; Joshi et al., 2014). Recent reports have described the presence of deleterious heterozygous germline IKZF1 variants in the DNA-binding domain in hereditary hypogammaglobulinemia, including common variable and immunoglobulin A (IgA) immunodeficiency (Hoshino et al., 2016; Kuehn et al., 2016). One report also described the occurrence of ALL in 2 of 29 individuals with IKZF1-associated immunodeficiency (Kuehn et al., 2016). Here we investigate the prevalence of germline IKZF1 variation in childhood ALL. RESULTS Identification of Germline IKZF1 Variants The proband of the index family was diagnosed with BCR-ABL1positive B-ALL at the age of 5.1 years and received cytotoxic chemotherapy for high-risk disease. He achieved a morphologic remission after induction chemotherapy but relapsed during maintenance therapy. Following re-induction therapy, the patient received allogeneic hematopoietic stem cell transplantation but subsequently died of transplant-related complications. Candidate gene sequencing of his ALL blasts and a matched remission sample revealed a germline deletion variant (c.del556, or D186fs) in IKZF1 that was predicted to truncate the protein at amino acid 192. Subsequent examination of the IKZF1 status of other members of the index family confirmed transmission of this truncating variant in five subjects spanning three generations, including an uncle (II-2) who died of B-ALL of unknown genetic subtype at the age of 6 (Figure 1A). Variable lymphopenia and low-normal IgG levels were noted in the other

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 1. Germline IKZF1 Variants in Pediatric ALL Patients (A) Pedigree of the index family demonstrating a germline IKZF1 variant (c.del556; p.D186fs) with incomplete penetrance of ALL. An arrow denotes the proband, squares and circles represent males and females, respectively, inset black boxes represent individuals with B-ALL, filled circles indicate low B cell numbers, ‘‘+’’ identifies individuals harboring the heterozygous germline IKZF1 D186fs allele, ‘‘ ’’ identifies individuals with wild-type IKZF1 alleles, and slashes indicate deceased family members. (B) CONSORT diagram of St. Jude Children’s Research Hospital and Children’s Oncology Group ALL patients included in this study. (C) Protein domain plot of IKAROS and the amino acid substitutions predicted to result from the germline IKZF1 variants identified in this study. See also Figure S1 and Tables S1–S4.

cases with the D186fs variant; however, no clinical manifestations of immunodeficiency were observed. To comprehensively examine the prevalence and nature of coding germline IKZF1 variants, we performed targeted sequencing in germline (remission) DNA from 4,963 children enrolled in St. Jude Children’s Research Hospital (SJCRH) and Children’s Oncology Group (COG) frontline clinical trials for newly diagnosed ALL (Figure 1B and Table S1). In this screening cohort, we identified 43 patients who harbored 27 unique nonsilent exonic IKZF1 variants, including 25 missense and two nonsense changes (Figures 1B and 1C; Table 1). All except one IKZF1 variant were observed in patients with B-, not T-lineage, ALL, and all but five were singletons in this ALL cohort. In the Exome Aggregation Consortium (ExAC) cohort that excluded the cases from The Cancer Genome Atlas (TCGA) (n = 53,105) (Lek et al., 2016), 18 of these variants were not observed and the remaining 9 were rare, with a maximum allele frequency of 0.12% (Table 1). Similar results were observed when we used the Genome Aggregation (gnomAD) database (Lek et al., 2016) that included 15,496 whole-genome sequences from unrelated individuals as non-ALL controls. To demonstrate that the mutations were germline and not representative of potential clonal hematopoiesis, we extracted genomic DNA from mesenchymal stromal cells isolated from bone marrow aspirate samples

obtained at remission in four cases, and performed Sanger sequencing that confirmed the heterozygous germline status of each variant (Figure S1A). Moreover, the allele fraction of all variants was approximately 50%, also consistent with the variants being germline. To explore the possibility that germline structural variants in IKZF1 may also predispose to ALL, we examined germline SNP array data for the cohort, and did not identify any cases with IKZF1 deletions. As SNP array analysis lacks resolution to identify all possible focal structural variants, we also examined germline whole-genome sequencing data performed for 691 children with ALL studied by the SJCRH-Washington University Pediatric Cancer Genome Project. With the exception of two cases with tumor-in-normal contamination, there was no evidence of germline IKZF1 deletion. In total we identified 28 unique germline IKZF1 variants in 45 children with ALL; one frameshift variant (D186fs) in two children with ALL and four unaffected relatives within a kindred, and 27 variants in 43 patients from targeted screening of presumed sporadic cases. These IKZF1 variants were distributed across the gene, particularly in a region proximal to the C-terminal zincfinger dimerization domain that in the mouse has been shown to mediate interactions with SIN3 and HDAC proteins (Koipally and Georgopoulos, 2002) (Figure 1C). This contrasts with the pattern of somatic mutations in B-ALL that cluster in both the Cancer Cell 33, 1–12, May 14, 2018 3

No. of Cases Identified

Allele Frequency in Normal Control (ExAC Excluded TCGA)

Allele Frequency in Normal Control (gnomAD Genomes)

CADD (PHRED)

Polyphen2

SIFT

Functionally Damaging In Vitro/ In Vivo

0

21.2

damaging

tolerated

no

5

18.7

damaging

tolerated

yes

IKZF1 Variant

Position (hg19/chr7)

Reference Allele

Mutant Allele

Mutation Type

Pro18Thr (P18T)

50367245

C

A

missense

1

0

Met31Val (M31V)

50367284

A

G

missense

2

3.1 3 10

5

3.2 3 10

Val52Leu (V52L)

50367347

G

C

missense

1

0

5.8

benign

tolerated

yes

Val53Met (V53M)

50367350

G

A

missense

4

3.0 3 10

4

8.0 3 10

0 4

9.1

probably damaging

tolerated

yes

Arg69His (R69H)

50444276

G

A

missense

1

2.0 3 10

4

6.5 3 10

5

15.7

benign

tolerated

no

3.2 3 10

5

28.4

damaging

tolerated

yes

0

16.5

probably damaging

tolerated

yes yes

Asp81Asn (D81N)

50444311

G

A

missense

1

9.5 3 10

6

Ser105Leu (S105L)

50444384

C

T

missense

1

9.5 3 10

6

Arg162Pro (R162P)

50450301

G

C

missense

1

0

0

27.6

damaging

deleterious

His163Tyr (H183Y)

50450303

C

T

missense

1

0

0

27.5

damaging

deleterious

yes

Asp186fs (D186fs)

50450369

AG

G

frameshift

2

0

0

36.0

NA

NA

yes

Asp252Asn (D252N)

50459465

G

A

missense

3

1.0 3 10

14.8

damaging

tolerated

yes

Ser258Pro (S258P)

50459483

T

C

missense

1

0

0

15.1

benign

tolerated

yes

Met306* (M306*)

50467680

GA

G

nonsense

1

0

0

36.0

NA

NA

yes

Thr333Ala (T333A)

50467762

A

G

missense

1

1.0 3 10

4

1.0 3 10

4

17.4

damaging

tolerated

yes

1.2 3 10

3

3.1 3 10

3

4

3.2 3 10

5

Gly337Ser (G337S)

50467774

G

A

missense

10

3.3

benign

tolerated

yes

Met347Val (M347V)

50467804

A

G

missense

1

0

0

1.1

benign

tolerated

yes

Tyr348Cys (Y348C)

50467808

A

G

missense

2

0

0

18.3

damaging

deleterious

yes

Ala365Val (A365V)

50467859

C

T

missense

1

0

0

17.2

damaging

tolerated

yes

Cys394* (C394*)

50467947

C

A

nonsense

1

0

0

35.0

NA

tolerated

yes

Leu411Phe (L411P)

50467996

C

T

missense

1

0

0

19.7

damaging

tolerated

no

Pro420Gln (P420Q)

50468024

C

A

missense

1

0

0

13.2

benign

tolerated

no

Arg423Cys (R423C)

50468032

C

T

missense

1

0

0

10.1

damaging

deleterious

yes

His432Gln (H423Q)

50468061

C

G

missense

1

0

0

0.7

benign

tolerated

no

Ala434Gly (A434G)

50468066

C

G

missense

1

0

0

16.4

probably damaging

deleterious

yes

Leu449Phe (L449F)

50468110

C

T

missense

1

0

0

0.1

benign

tolerated

yes

Met459Val (M459V)

50468140

A

G

missense

1

0

0

0.2

benign

tolerated

yes

Met476Thr (M476T)

50468192

T

C

missense

1

0

0

22.7

damaging

tolerated

yes

Met518Lys (M518K)

50468318

T

A

missense

1

1.1 3 10

0

15.6

probably damaging

deleterious

no

5

ExAC, Exome Aggregation Consortium; TCGA, The Cancer Genome Atlas; gnomAD, Genome Aggregation Database; CADD, Combined Annotation-Dependent Depletion; Polyphen2, Polymorphism Phenotyping v2; SIFT, Sorting Intolerant From Tolerant; NA, not available. See also Figure S2.

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

4 Cancer Cell 33, 1–12, May 14, 2018

Table 1. Germline IKZF1 Variants in Pediatric ALL Cases

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

N-terminal DNA-binding domain and within the C-terminal dimerization domain, whereas previously identified germline IKZF1 variants in immunodeficient patients are restricted to the N-terminal zinc fingers (Hoshino et al., 2016; Kuehn et al., 2016) (Figures 1C and S1B). We also compared the occurrence of rare and damaging IKZF1 coding variants in ALL cases (n = 4,963) and non-ALL controls (i.e., the ExAC cohort excluded TCGA data). Using a Combined Annotation-Dependent Depletion (CADD) score (Kircher et al., 2014) of greater than 15 as a threshold for deleteriousness, and focusing on variants with allele frequency <0.01%, we observed a significant over-representation of rare and predicted damaging IKZF1 variants in ALL (0.34% versus 0.13%, hazard ratio 2.7, p < 0.001). This enrichment remained significant with varying cutoffs of CADD score and allele frequency (Figure S2), suggesting that loss-offunction IKZF1 variants are related to ALL disease risk. Again, using the gnomAD non-ALL controls, we observed similar patterns of enrichment for damaging IKZF1 variants (Table S2). Genomic Profiling of Germline IKZF1 Variant Cases Germline variants in other ALL predisposition genes exhibit association with somatic genomic features, such as dicentric chromosome 9 and germline PAX5 mutations, low hypodiploidy in ALL with germline TP53 mutations, and high hyperdiploidy with germline ETV6 mutations. We thus used whole-exome and RNA sequencing, and SNP array analysis, for all available patient leukemia samples to identify potential collaborating genetic events that were somatically acquired in our cohort of patients harboring germline IKZF1 variants. We observed a high proportion of high hyperdiploidy (8/21, 38%), and other known recurring ALL-associated genomic rearrangements, including ETV6-RUNX1 (n = 5), KMT2A (MLL) rearrangement (3 cases), and rearrangement of CRLF2 and MEF2D-BCL9 (Tables S2 and S3). Two cases displayed broad loss of 9p, affecting the region harboring the CDKN2A and PAX5 loci; these genes were also affected by focal deletions in one case each and missense mutations in PAX5 were observed in 5 of 27 cases sequenced (Tables S2 and S3). Mutations in KRAS (5/27) and NRAS (2/27) were also prevalent, including six activating mutations affecting glycine 13 (G13D/V; Tables S2 and S3). Additional, less frequently mutated genes included TP53, CREBBP, JAK2, MTOR, and FLT3 (Tables S2 and S3); the full list is available in Table S3. Interestingly, none of patients with damaging IKZF1 variants carried other known rare ALL predisposition variants in PAX5, ETV6, or TP53, and these IKZF1 variants were not associated with known common ALL risk variants (e.g., ARID5B, GATA3, and PIP4K2A; Table S4), suggesting independent effects of these variants on ALL pathogenesis. Functional Effects of IKZF1 Variants Using a mouse model of BCR-ABL1+ B-ALL, we previously showed that somatic IKZF1 alterations observed in B-ALL patients, including the isoform IK6 that lacks exons 4–7 and the N-terminal zinc-finger domains, Ikzf1 haploinsufficiency, and DNA-binding domain point mutants, all result in aberrant leukemia cell-cell and cell-stromal adhesion, extravascular invasion in the bone marrow niche, and decreased sensitivity to antileukemia drug treatment in vitro and in vivo (Churchman et al., 2015). Herein we performed a similar series of experiments to evaluate

how the germline variants identified in this study affect IKAROS activity and leukemia cell phenotypes. These included the ability of the range of IKZF1 alleles to transcriptionally repress target genes, bind DNA, dimerize, localize to the nucleus, promote aberrant leukemic cell aggregation, and influence drug sensitivity of leukemic cells. Using a reporter assay in HEK293T cells, all variants were tested for their ability to transcriptionally repress the promoter of MCL1, a known IKAROS target (Iacobucci et al., 2012). Variants affecting the DNA-binding domain (R162P and H163Y) and truncating mutations (D186fs, M306*, and C394*) were unable to repress MCL1 promoter activity, whereas all other variants retained repressive function (Figure 2A and data not shown). These five variants were then tested for dominant-negative effects using the same reporter system by co-transfection with wild-type IKZF1. Variants within the DNA-binding domain (R162P and H163Y) displayed dominant-negative effects with loss of repressive ability despite concomitant wild-type IKZF1 expression. In contrast, restoration of transcriptional repression was observed when wild-type IKZF1 was expressed with truncating variants (D186fs, M306*, and C394*; Figure 2B). To determine whether these five variants lose the ability to repress the MCL1 promoter due to impaired DNA binding, we used electrophoretic mobility-shift assays whereby nuclear extracts from transfected HEK293T cells were incubated with DNA probes containing an IKAROS consensus-binding sequence (IK-bs4) (Cobb et al., 2000; Kuehn et al., 2016) and assessed for binding. Of five variants that failed to repress MCL1, those residing within the N-terminal zinc fingers (R162P, H163Y, and D186fs) were unable to directly bind DNA, whereas truncating variants M306* and C394* retained DNA-binding affinity (Figure 2C), consistent with the notion that both DNA binding and dimerization are required for efficient transcriptional regulation (Koipally et al., 2002). To examine effects of these germline variants on IKAROS dimerization, we ectopically expressed FLAG-tagged wild-type IKAROS and hemagglutinin (HA)-tagged variant IKAROS in HEK293T cells and then performed co-immunoprecipitation. Immunoblotting showed that D186fs, M306*, and C394* failed to dimerize with wild-type IKAROS, whereas the DNA-binding domain and C-terminal missense variants retained the ability to dimerize with the wild-type protein (Figure 2D). Thus, with the exception of DNA-binding domain and truncating variants, the majority of the newly identified germline variants do not directly affect transcriptional activation of known IKAROS target genes or dimerization. To assess the function of these germline IKZF1 variants in a cellular context that is more biologically relevant to ALL, we introduced each variant into primary mouse Arf / BCR-ABL1+ pre-B cells by retroviral transduction and evaluated whether they displayed aberrant phenotypes known to be associated with IKZF1 alterations. First we performed IKAROS immunostaining and confocal microscopy to visualize subcellular localization, which is typically observed as punctate foci at pericentromeric DNA within the nucleus (Cobb et al., 2000). Mislocalization to the cytoplasm was observed for the DNA-binding domain variants (R162P and H163Y), truncating variants D186fs and M306*, dimerization domain variant M476T, as well as variants residing outside of the zinc-finger domains, including N-terminal variants (M31V, V52L, V53M, D81N, and S105L) and variants Cancer Cell 33, 1–12, May 14, 2018 5

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 2. Effects of IKZF1 Variants on Transcriptional Repression, Dimerization, DNA Binding, and Nuclear Localization (A) Relative MCL1 promoter activity in luciferase reporter assays using HEK293T cells ectopically expressing IKAROS variants compared with wildtype (WT), empty vector (EV), and dominantnegative IK6 as controls. (B) Relative MCL1 promoter activity after wild-type IKAROS is co-expressed 1:1 with the subset of IKAROS variants that lose the ability to transcriptionally repress MCL1. (C) Electrophoretic shift assays performed using nuclear extracts of HEK293T cells ectopically expressing wild-type or variant IKAROS. Extracts were allowed to bind biotin-labeled probes containing a consensus IKAROS binding site (IKbs4); as controls, excess unlabeled probe and IKAROS antibody were used as binding competitors. Arrow indicates full-length IKAROS-containing complexes and asterisks indicate truncated size of nonsense variants. (D) Co-immunoprecipitation (IP) of FLAG-tagged wild-type and HA-tagged wild-type or variant IKAROS in HEK293T cells to demonstrate the ability of each variant to dimerize. Data in (A) and (B) are presented as mean ± SD of three technical replicates; Student’s t test compared with wild-type, ***p < 0.0005, **p < 0.005, *p < 0.05. See also Figure S3.

within putative SIN3 and HDAC2 interaction domains (R423C, A434G, and L449F) (Figures 3A and 3B). Marked mislocalization of the D186fs variant was confirmed in immortalized B cells isolated from index family members that carry the variant allele (Figure 3C). Furthermore, we confirmed the localization of all variants in HEK293T cells by immunofluorescence (Figure S3A). To confirm that our observations were not artifacts of retrovirally overexpressing each variant, we created CRISPR/Cas9-engineered Arf / BCR-ABL1+ pre-B cell lines with endogenous expression of M31V, R162P, or A434G (A430G in mouse) from the Ikzf1 locus and confirmed subcellular mislocalization of each of these variants (Figure S3B). Protein expression levels of retrovirally expressed variants in Arf / BCR-ABL1+ pre-B cells were assessed to examine potential associations between level of expression and observed phenotypes for each variant. While expression levels were variable, there was not a direct correlation with the severity of the phenotypes observed (Figures S3C and S3D). In particular, D186fs and M306* exhibited very low level expression on western blot, yet manifested highly aberrant phenotypes. To determine whether their protein stability was affected, we expressed D186fs, M306*, and C394* compared with wild-type IKAROS and IK6 in HEK293T cells 6 Cancer Cell 33, 1–12, May 14, 2018

that were treated with cycloheximide, and observed that D186fs and M306* underwent rapid protein degradation (Figure S3E). Several somatic IKZF1 alterations have been previously shown in a mouse model of BCR-ABL1+ ALL to result in increased expression of genes involved in cellcell and cell-stroma adhesion, leading to the aggregation of leukemic cells in vitro, and to adherence and extravascular invasion of leukemic cells in the bone marrow microenvironment in vivo (Churchman et al., 2015, 2016). To examine the effects of germline IKZF1 variants on adhesion in vitro, we measured the number and size of cell aggregations that were formed in suspension cultures as a result of the expression of each variant in Arf / BCR-ABL1+ pre-B leukemia cells. Ectopic expression of the P18T, P420Q, H432Q, or M518K variants resulted in cell death, suggesting that these retain wild-type IKAROS function, as we have previously shown that overexpression of IKAROS is not tolerated in pre-B cells (Churchman et al., 2015). This observation was consistent with the ability of these variants to repress the MCL1 promoter in vitro and localize normally to the nucleus of HEK293T cells (Figure S3A). The remaining 24 variants had a broad range of effects on adhesion and cell viability with no clear correlation with level of overexpression (Figures S3C and S3D), as follows. Variants affecting DNA binding and/or dimerization (R162P, H163Y, D186fs, M306*, and C394*) resulted in aggregation comparable with that of the IK6 deletion isoform that lacks DNA-binding activity. Several variants located outside of the DNA-binding or

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 3. Subcellular Mislocalization of IKAROS Variants (A) Subcellular localization of endogenous wild-type (WT) and ectopically expressed variant IKAROS (green) in primary mouse Arf / BCR-ABL1+ pre-B cells detected by immunofluorescence using an antibody to the IKAROS N terminus capable of detecting all variant proteins and an AlexaFluor 488-conjugated secondary antibody. Nuclei (blue) are observed by DAPI staining. (B) Quantification of the amount of IKAROS immunostaining found in the cytoplasm of Arf / BCR-ABL1+ pre-B cells ectopically expressing each N-terminal variant. n R 25 cells; Student’s t test compared with EV, ***p < 0.0005, *p < 0.05. Short horizontal lines of each corresponding color indicate the mean for each variant. The dotted horizontal line indicates the maximum percentage of cytoplasmic IKAROS staining observed in empty vector (EV) cells. All variants were tested but those that retained IKAROS activity were not evaluable as their ectopic expression resulted in cell death. (C) Immunostaining of IKAROS in B cells isolated from individuals III-1 and III-2 in the index family shown in Figure 1 compared with ND340 cells isolated from a normal bone marrow donor harboring a wild-type IKZF1. Scale bars, 20 mm (A) and 10 mm (C). See also Figure S3.

dimerization zinc-finger domains (V53M, M31V, S105L, T333A, M347V, Y348C, A365V, R423C, A434G, L449F, and M459V) also resulted in pronounced aggregation (Figures 4A, 4B, and S4A). Cellular aggregations were also observed in Arf / BCR-ABL1+ pre-B leukemia cells with endogenous expression of M31V, R162P, and A430G engineered by CRISPR/Cas9 genome editing (Figure S4A). Increased cell-cell adhesion was further confirmed by irregular leukemic cell morphology and abnormal cell-stromal associations in the bone marrow microenvironment in vivo (Figures 4C and S4B). Notably, Arf / BCR-ABL1+ pre-B leukemia cells overexpressing adhesionrelated variants also showed significant upregulation of one or more of the adhesion molecules THY1 (also known as CD90), ITGA5 (integrin a5), or SELL (L-selectin), known to be transcriptionally repressed by wild-type IKAROS (Churchman et al., 2015) (Figures 4D and S4C).

IKZF1 Variants Affect Antileukemia Drug Sensitivity of Leukemic Cells We next examined the effects of the germline variant IKZF1 alleles on drug responsiveness in Arf / BCR-ABL1+ pre-B cells in vitro and in vivo. The tyrosine kinase inhibitor dasatinib and dexamethasone potently induce apoptosis in Arf / BCRABL1+ pre-B cells in vitro (LC50 = 0.05 nM and 2.7 nM, respectively) and expression of the dominant-negative IK6 isoform significantly increased the concentration of drug required to induce 50% cell killing (LC50) (Figures 5A and 5B). We selected ten germline IKZF1 variants found to be most damaging in other functional assays (M31V, R162P, H163Y, D186fs, M306*, M347V, C394*, R423C, A434G, and L449F) and evaluated their effects on sensitivity to dasatinib and dexamethasone in vitro. In these studies, the N-terminal variant M31V and the C-terminal variants A434G and L449F, which lie within the putative Cancer Cell 33, 1–12, May 14, 2018 7

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 4. Effects of IKZF1 Variants on Cell Adhesion (A) Representative in vitro suspension cultures of primary mouse Arf / BCR-ABL1+ pre-B cells transduced with empty vector (EV) or expressing the dominant-negative IK6 or IKAROS variants. (B) Quantification of the number and size of cell-cell aggregate clusters in suspension cultures of Arf / BCR-ABL1+ pre-B cells expressing IKAROS variants compared with empty vector (EV) and IK6. (C) Ex vivo confocal imaging of the undisturbed bone marrow niche in calvaria of mice 3 days after inoculation with Arf / BCR-ABL1+ pre-B cells expressing IKAROS variants. (D) Flow-cytometric analyses of Arf / BCRABL1+ pre-B cells co-expressing IKAROS variants for adhesion molecules, THY1, ITGA5, and L-selectin (SELL); Data represent the mean ± SD for three technical replicates; colored dashed lines indicate mean percentage of EV cells. Student’s t test compared with EV, ***p < 0.0005. Scale bars, 20 mm. See also Figure S4.

SIN3-binding domain proximal to the dimerization domain, demonstrated the most striking reduction in sensitivity to dasatinib and dexamethasone. Variants R162P, M306*, and R423C showed reduced sensitivity to dasatinib comparable with IK6, whereas H163Y and D186fs were less sensitive to dexamethasone (Figures 5A and 5B; Table S5). In vivo efficacy studies revealed that mice inoculated with Arf / BCR-ABL1+ pre-B cells expressing the H163Y, A433G, or L449F variants showed a significantly shortened survival in response to dasatinib treatment compared with the empty vector control group despite all untreated mice succumbing to disease with no change in latency (Figure 5C). These findings are consistent with our previous reports that IKZF1-altered leukemias, including those expressing IK6, are less responsive to dasatinib in vivo (Churchman et al., 2015, 2016). While the H163Y-expressing cells were sensitive to dasatinib in vitro, these cells were more resistant to dasatinib than wild-type cells in vivo, highlighting the importance of perturbed interactions of tumor cells with the tumor microenvironment in determining overall response to therapy in vivo. Thus, we have identified 22 IKZF1 variants that adversely affected gene function based on at least one assay in vitro, and were considered as functionally ‘‘damaging’’ with the remaining six variants classified as ‘‘benign’’ (Figure 6; Tables 1 and S5). Comparing patient characteristics among patients with damaging, benign variants, or wild-type IKZF1 in COG and SJCRH ALL cohorts with data available (n = 4,698), we did not observe any significant difference in demographic and ALL presenting features by germline IKZF1 status (Table S1). DISCUSSION Multiple studies have shown that somatic alterations in IKZF1 are common in ALL and enriched in high-risk forms of leukemia that are driven by genetic alterations activating kinase signaling pathways, such as Ph+ and Ph-like ALL (Den Boer et al., 2009; 8 Cancer Cell 33, 1–12, May 14, 2018

Iacobucci et al., 2009; Lana et al., 2015; Martinelli et al., 2009; Mullighan et al., 2008, 2009; Roberts et al., 2014, 2017; Van der Veer et al., 2014). These alterations include deletions or sequence mutations that result in either loss of function and/or dominant-negative effects, particularly by disrupting binding to DNA targets and mislocalizing IKAROS from sites of DNA binding at heterochromatin to a broad distribution in the nucleus and/or cytoplasm. Such alterations result in impaired maturation of leukemia cells and acquisition of a hematopoietic stem celllike phenotype, as well as in deregulation of cellular adhesion within the bone marrow niche, leading to aberrant stromal adhesion of leukemic cells. In this study, we have identified coding germline IKZF1 variation as a risk factor for genetic predisposition to ALL and show that the majority of variants have deleterious effects on IKAROS functions. Moreover, these IKZF1 variants demonstrate differences in the type and magnitude of effects. The germline IKZF1 variants identified in this study are distributed across the entire length of the IKAROS protein, which is in stark contrast to those identified in familial common variable immune deficiency, which are exclusively found within the DNAbinding zinc fingers (Kuehn et al., 2016) and suggest severe perturbation of protein function when this domain is affected. While many of the variants identified in our study are located outside of the known DNA-binding or dimerization domains of IKAROS, remarkably, 22 out of the 28 unique variants identified (79%) were functionally damaging in at least one of our functional assays in HEK293T cells or primary mouse BCR-ABL1+ B-ALL cells. Strikingly, five variants located outside of the zinc-finger domains (M31V, M347V, R423C, A434G, and L449F) were among the most deleterious, alongside two variants affecting the N-terminal DNA-binding domain (R162P and H163Y) and three truncating frameshift and nonsense variants (D186fs, M306*, and C394*). Combining the predictions of three in silico tools used to categorize variants as deleterious or tolerated (CADD, Polyphen2, and SIFT), eight of these ten most damaging

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 5. Effects of IKZF1 Variants on Sensitivity to Drugs (A and B) Cell viability of Arf / BCR-ABL1+ pre-B cells transduced with empty vector (EV) or expressing IKAROS variants at increasing concentrations of dasatinib (A) or dexamethasone (B) after 72 hr of drug exposure in vitro. Data represent mean ± SD for three biological replicates performed in triplicate. (C) Kaplan-Meier survival curve of mice inoculated by tail vein with Arf / BCRABL1+ pre-B cells expressing either empty vector or IKAROS variants and dosed daily starting 7 days post inoculation with either vehicle or 10 mg/kg dasatinib by mouth; n = 5 mice per group; Mantel-Cox log rank compared with EV, **p < 0.005, *p < 0.05. See also Table S5.

variants were accurately predicted by at least one method, whereas L449F and M348V were predicted to be benign; overall only 65% of variants were accurately predicted when compared

with our functional results. Several variants that were highly deleterious in cellular localization, adhesion, and drug responsiveness assays did not fall within the DNA-binding or dimerization domains and had no effect detected by more widely used assays such as transcriptional repression and DNA binding. This result suggests that these variants perturb IKAROS functions through misassembly of complexes such as the SIN3 histone deacetylase, Mi-2/NuRD, and PRC2 complexes. However, in contrast to other regions of IKZF1, the putative C-terminal SIN3-interacting domain that harbors many of the germline variants shows limited conservation between mouse and human proteins, and we observed minimal interaction between human IKAROS and SIN3A (data not shown). Thus, the variants in this region are likely to perturb or confer interaction with other transcriptional regulatory proteins, and future studies directly examining the role of these regions in transcriptional complex assembly and gene regulation will be of interest. Gene mutations that predispose to familial leukemia, such as PAX5, TP53, and ETV6, impair function of the encoded protein. Our findings add to the growing body of evidence of inherited predisposition to ALL by showing that germline IKZF1 variants associated with leukemia predisposition also directly influence responsiveness of leukemia cells to chemotherapy. We have previously shown that somatic IKZF1 alterations result in resistance to kinase inhibitor therapy, in part by the induction of aberrant adhesion and bone marrow microenvironmental mislocalization, and that this effect may be reversed by therapeutic approaches that induce expression of wild-type IKZF1, such as retinoid receptor agonists (Churchman et al., 2015), or agents that inhibit adhesion molecule signaling pathways, such as focal adhesion kinase inhibitors (Churchman et al., 2016). The current findings extend these data by showing that several of the germline variants also reduce responsiveness to both conventional chemotherapeutic agents as well as kinase inhibitors, accompanied by aberrant adhesion. Thus, we have shown that germline variants in IKZF1 predispose to ALL and can also influence the response of leukemia cells to therapeutic agents. These findings also broaden the paradigm of germline variants in lymphoid transcription factors influencing the risk of developing ALL, as previously described for ETV6 (Duployez et al., 2018; Moriyama et al., 2015; Noetzli et al., 2015; Topka et al., 2015; Zhang et al., 2015). As is the case for ETV6, in which a highly deleterious variant was first identified in a family with multiple affected members, in our study an index family revealed a frameshift variant in IKZF1 that prematurely truncates the IKAROS protein and removes the C terminus. The current study demonstrates the power of sequencing large cohorts of sporadic ALL cases for similar variants. Familial ALL is uncommon, but it is unclear whether relatives of the majority of patients with germline IKZF1 variants from the sporadic cohort are also at increased risk of ALL, as clinical records for the majority of such cases were not accessible; however, none of the six children treated on SJCRH ALL protocols had a family history of ALL or other hematological malignancies. However, this study demonstrates how analysis of families can inform our understanding of leukemia predisposition by identifying germline genetic variants that may play a more general role in susceptibility to the disease, and possibly treatment outcome. Cancer Cell 33, 1–12, May 14, 2018 9

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Figure 6. Summary of Variant Effects The position of each variant within the protein is shown along with the consequences noted for each assay, denoted as ‘‘+’’ if an effect was observed, and ‘‘ ‘‘ if the variant had no effect; blanks denote that the variant was not included in the assay. Putative protein interaction domains that have been previously described for mouse IKAROS are noted below the protein structure. See also Table S5.

STAR+METHODS

ACKNOWLEDGMENTS

Detailed methods are provided in the online version of this paper and include the following:

We thank the Genome Sequencing Facility, Hartwell Center for Bioinformatics and Biotechnology, and the flow cytometry and cell sorting core facility of St. Jude Children’s Research Hospital. This work was supported in part by the American Lebanese Syrian Associated Charities of St. Jude Children’s Research Hospital; European Commission FP7, ERA-NET on Translational Cancer Research, TRANSCALL, Health-F2-2010-260791 (to M.S. and G.K.); Parents Association for Children with Cancer, Tuebingen (to R.H.); a St. Baldrick’s Foundation Scholar award and Robert J. Arceci award (to C.G.M.); NIH grant P50 GM115279 to (to C.G.M., J.J.Y., M.V.R., and W.E.E.); National Cancer Institute grants P30 CA021765 (St. Jude Cancer Center support grant), U01CA176063 (to. J.J.Y.), R35 CA197695-01A1 (to C.G.M.); and grants to the Children’s Oncology Group: CA98543, CA114766, CA98413, CA180886, and CA180899.

d d d

d

d

d

KEY RESOURCES TABLE CONTACT FOR REAGENT AND RESOURCE SHARING EXPERIMENTAL MODEL AND SUBJECT DETAILS B Human Subjects B Animals B Primary Mouse Pre-B Cell Culture B Cell Lines METHOD DETAILS B IKZF1 Sequencing B Genomic Analysis of Patient Samples B Expression Vectors and Retroviral Production B Co-immunoprecipitation B Electrophoretic Mobility Shift Assays (EMSA) B Immunofluorescence B Luciferase Reporter Assay B Calvarial Imaging B Flow Cytometric Analysis of Adhesion Proteins B Cell Viability Assays B CRISPR/Cas9 Genome Editing QUANTIFICATION AND STATISTICAL ANALYSIS B Quantification of Cell-Cell Adhesion In Vitro B Statistical Analyses DATA AND SOFTWARE AVAILABILITY

SUPPLEMENTAL INFORMATION Supplemental Information includes four figures and six tables and can be found with this article online at https://doi.org/10.1016/j.ccell.2018.03.021.

10 Cancer Cell 33, 1–12, May 14, 2018

AUTHOR CONTRIBUTIONS J.J.Y. and C.G.M. initiated and led the project; M.L.C., M.Q., G.t.K., W.E.E., M.V.R., S.P.H., M.L.L, J.J.Y., and C.G.M. designed experiments; M.L.C., R.Z., P.T., R.B., K.V., I.M.M., and H.Y. performed functional experiments, J.L.P. acquired and assisted in analyses of microscopy images; S.M.P.-M. and S.N.P. designed and provided CRISPR/Cas9 reagents, G.t.K., H.Z., T.L., M.D., E.R., E.L., P.L.M., W.P.B., N.W., E.M., R.F., and M.S. contributed to the data gathering; J.J.Y., M.Q.,C.-H.P., S.P.H., R.H., K.E.N., C.G.M., and M.L.C. interpreted the data; M.Q., W.Y., Z.G., C.Q., and G.W. analyzed genomic data; M.L.C., C.G.M., M.Q., and J.J.Y. wrote the manuscript. All of the authors reviewed and commented on the manuscript. DECLARATION OF INTERESTS C.G.M. received research support for the study of FAK inhibitors in IKZF1mutated leukemia from Verastem. Received: June 21, 2017 Revised: February 8, 2018 Accepted: March 19, 2018 Published: April 19, 2018

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

REFERENCES

Hunger, S.P., and Mullighan, C.G. (2015). Acute lymphoblastic leukemia in children. N. Engl. J. Med. 373, 1541–1552.

Adzhubei, I., Jordan, D.M., and Sunyaev, S.R. (2013). Predicting functional effect of human missense mutations using PolyPhen-2. Curr. Protoc. Hum. Genet. Chapter 7, Unit7.20.

Iacobucci, I., Iraci, N., Messina, M., Lonetti, A., Chiaretti, S., Valli, E., Ferrari, A., Papayannidis, C., Paoloni, F., Vitale, A., et al. (2012). IKAROS deletions dictate a unique gene expression signature in patients with adult B-cell acute lymphoblastic leukemia. PLoS One 7, e40934.

Adzhubei, I.A., Schmidt, S., Peshkin, L., Ramensky, V.E., Gerasimova, A., Bork, P., Kondrashov, A.S., and Sunyaev, S.R. (2010). A method and server for predicting damaging missense mutations. Nat. Methods 7, 248–249. Appelmann, I., Rillahan, C.D., de Stanchina, E., Carbonetti, G., Chen, C., Lowe, S.W., and Sherr, C.J. (2015). Janus kinase inhibition by ruxolitinib extends dasatinib- and dexamethasone-induced remissions in a mouse model of Ph+ ALL. Blood 125, 1444–1451. €schendorf, F., Gombert, M., Husemann, P., Ginzel, S., Izraeli, S., Auer, F., Ru Harit, M., Weintraub, M., Weinstein, O.Y., Lerer, I., et al. (2013). Inherited susceptibility to pre B-ALL caused by germline transmission of PAX5 c.547G>A. Leukemia 28, 1136–1138. Boulos, N., Mulder, H.L., Calabrese, C.R., Morrison, J.B., Rehg, J.E., Relling, M.V., Sherr, C.J., and Williams, R.T. (2011). Chemotherapeutic agents circumvent emergence of dasatinib-resistant BCR-ABL kinase mutations in a precise mouse model of Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 117, 3585–3595. Bowman, W.P., Larsen, E.L., Devidas, M., Linda, S.B., Blach, L., Carroll, A.J., Carroll, W.L., Pullen, D.J., Shuster, J., Willman, C.L., et al. (2011). Augmented therapy improves outcome for pediatric high risk acute lymphocytic leukemia: results of Children’s Oncology Group trial P9906. Pediatr. Blood Cancer 57, 569–577.

Iacobucci, I., and Mullighan, C.G. (2017). Genetic basis of acute lymphoblastic leukemia. J. Clin. Oncol. 35, 975–983. Iacobucci, I., Storlazzi, C.T., Cilloni, D., Lonetti, A., Ottaviani, E., Soverini, S., Astolfi, A., Chiaretti, S., Vitale, A., Messa, F., et al. (2009). Identification and molecular characterization of recurrent genomic deletions on 7p12 in the IKZF1 gene in a large cohort of BCR-ABL1-positive acute lymphoblastic leukemia patients: on behalf of Gruppo Italiano Malattie Ematologiche dell’Adulto Acute Leukemia Working Party (GIMEMA AL WP). Blood 114, 2159–2167. Joshi, I., Yoshida, T., Jena, N., Qi, X., Zhang, J., Van Etten, R.A., and Georgopoulos, K. (2014). Loss of Ikaros DNA-binding function confers integrin-dependent survival on pre-B cells and progression to acute lymphoblastic leukemia. Nat. Immunol. 15, 294–304. Kamijo, T., Zindy, F., Roussel, M.F., Quelle, D.E., Downing, J.R., Ashmun, R.A., Grosveld, G., and Sherr, C.J. (1997). Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell 91, 649–659. Karol, S.E., Larsen, E., Cheng, C., Cao, X., Yang, W., Ramsey, L.B., Fernandez, C.A., McCorkle, J.R., Paugh, S.W., Autry, R.J., et al. (2017). Genetics of ancestry-specific risk for relapse in acute lymphoblastic leukemia. Leukemia 31, 1325–1332.

Chen, X., Gupta, P., Wang, J., Nakitandwe, J., Roberts, K., Dalton, J.D., Parker, M., Patel, S., Holmfeldt, L., Payne, D., et al. (2015). CONSERTING: integrating copy-number analysis with structural-variation detection. Nat. Methods 12, 527–530.

Kim, J., Sif, S., Jones, B., Jackson, A., Koipally, J., Heller, E., Winandy, S., Viel, A., Sawyer, A., Ikeda, T., et al. (1999). Ikaros DNA-binding proteins direct formation of chromatin remodeling complexes in lymphocytes. Immunity 10, 345–355.

Churchman, M.L., Low, J., Qu, C., Paietta, E.M., Kasper, L.H., Chang, Y., Payne-Turner, D., Althoff, M.J., Song, G., Chen, S.C., et al. (2015). Efficacy of retinoids in IKZF1-mutated BCR-ABL1 acute lymphoblastic leukemia. Cancer Cell 28, 343–356.

Kircher, M., Witten, D.M., Jain, P., O’Roak, B.J., Cooper, G.M., and Shendure, J. (2014). A general framework for estimating the relative pathogenicity of human genetic variants. Nat. Genet. 46, 310–315.

Churchman, M.L., Evans, K., Richmond, J., Robbins, A., Jones, L., Shapiro, I.M., Pachter, J.A., Weaver, D.T., Houghton, P.J., Smith, M.A., et al. (2016). Synergism of FAK and tyrosine kinase inhibition in Ph+ B-ALL. JCI Insight 1, https://doi.org/10.1172/jci.insight.86082. Cobb, B.S., Morales-Alcelay, S., Kleiger, G., Brown, K.E., Fisher, A.G., and Smale, S.T. (2000). Targeting of Ikaros to pericentromeric heterochromatin by direct DNA binding. Genes Dev. 14, 2146–2160. Den Boer, M.L., van Slegtenhorst, M., De Menezes, R.X., Cheok, M.H., BuijsGladdines, J.G., Peters, S.T., Van Zutven, L.J., Beverloo, H.B., Van der Spek, P.J., Escherich, G., et al. (2009). A subtype of childhood acute lymphoblastic leukaemia with poor treatment outcome: a genome-wide classification study. Lancet Oncol. 10, 125–134. Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21. Duployez, N., Abou Chahla, W., Lejeune, S., Marceau-Renaut, A., Letizia, G., Boyer, T., Geffroy, S., Peyrouze, P., Grardel, N., Nelken, B., et al. (2018). Detection of a new heterozygous germline ETV6 mutation in a case with hyperdiploid acute lymphoblastic leukemia. Eur. J. Haematol. 100, 104–107.

Koipally, J., and Georgopoulos, K. (2002). A molecular dissection of the repression circuitry of Ikaros. J. Biol. Chem. 277, 27697–27705. Koipally, J., Heller, E.J., Seavitt, J.R., and Georgopoulos, K. (2002). Unconventional potentiation of gene expression by Ikaros. J. Biol. Chem. 277, 13007–13015. Kuehn, H.S., Boisson, B., Cunningham-Rundles, C., Reichenbach, J., Stray-Pedersen, A., Gelfand, E.W., Maffucci, P., Pierce, K.R., Abbott, J.K., Voelkerding, K.V., et al. (2016). Loss of B Cells in patients with heterozygous mutations in IKAROS. N. Engl. J. Med. 374, 1032–1043. Kumar, P., Henikoff, S., and Ng, P.C. (2009). Predicting the effects of coding non-synonymous variants on protein function using the SIFT algorithm. Nat. Protoc. 4, 1073–1081. Lana, T., de Lorenzo, P., Bresolin, S., Bronzini, I., den Boer, M.L., Cave´, H., kova´, E., Stanulla, M., Zaliova, M., Harrison, C.J., et al. (2015). Fron Refinement of IKZF1 status in pediatric Philadelphia-positive acute lymphoblastic leukemia. Leukemia 29, 2107–2110.

Georgopoulos, K. (2017). The making of a lymphocyte: the choice among disparate cell fates and the IKAROS enigma. Genes Dev. 31, 439–450.

Larsen, E.C., Devidas, M., Chen, S., Salzer, W.L., Raetz, E.A., Loh, M.L., Mattano, L.A., Jr., Cole, C., Eicher, A., Haugan, M., et al. (2016). Dexamethasone and high-dose methotrexate improve outcome for children and young adults with high-risk B-acute lymphoblastic leukemia: a report from Children’s Oncology Group Study AALL0232. J. Clin. Oncol. 34, 2380–2388.

Hoshino, A., Okada, S., Yoshida, K., Nishida, N., Okuno, Y., Ueno, H., Yamashita, M., Okano, T., Tsumura, M., Nishimura, S., et al. (2016). Abnormal hematopoiesis and autoimmunity in human subjects with germline IKZF1 mutations. J. Allergy Clin. Immunol. 140, 223–231.

Lek, M., Karczewski, K.J., Minikel, E.V., Samocha, K.E., Banks, E., Fennell, T., O’Donnell-Luria, A.H., Ware, J.S., Hill, A.J., Cummings, B.B., et al. (2016). Analysis of protein-coding genetic variation in 60,706 humans. Nature 536, 285–291.

Hu, Y., Zhang, Z., Kashiwagi, M., Yoshida, T., Joshi, I., Jena, N., Somasundaram, R., Emmanuel, A.O., Sigvardsson, M., Fitamant, J., et al. (2016). Superenhancer reprogramming drives a B-cell-epithelial transition and high-risk leukemia. Genes Dev. 30, 1971–1990.

Li, H., and Durbin, R. (2009a). Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760. Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., and Durbin, R.; 1000 Genome Project Data Processing

Cancer Cell 33, 1–12, May 14, 2018 11

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Subgroup (2009b). The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079.

childhood acute lymphoblastic leukemia without cranial irradiation. N. Engl. J. Med. 360, 2730–2741.

Liu, Y., Easton, J., Shao, Y., Maciaszek, J., Wang, Z., Wilkinson, M.R., McCastlain, K., Edmonson, M., Pounds, S.B., Shi, L., et al. (2017). The genomic landscape of pediatric and young adult T-lineage acute lymphoblastic leukemia. Nat. Genet. 49, 1211–1218.

Pui, C.H., Relling, M.V., Sandlund, J.T., Downing, J.R., Campana, D., and Evans, W.E. (2006). Total Therapy study XV for newly diagnosed childhood acute lymphoblastic leukemia: study design and preliminary results. Ann. Hematol. 85 (Suppl 1 ), 88–91.

Martinelli, G., Iacobucci, I., Storlazzi, C.T., Vignetti, M., Paoloni, F., Cilloni, D., Soverini, S., Vitale, A., Chiaretti, S., Cimino, G., et al. (2009). IKZF1 (Ikaros) deletions in BCR-ABL1-positive acute lymphoblastic leukemia are associated with short disease-free survival and high rate of cumulative incidence of relapse: a GIMEMA AL WP report. J. Clin. Oncol. 27, 5202–5207.

Pui, C.H., Sandlund, J.T., Pei, D., Campana, D., Rivera, G.K., Ribeiro, R.C., Rubnitz, J.E., Razzouk, B.I., Howard, S.C., Hudson, M.M., et al. (2004). Improved outcome for children with acute lymphoblastic leukemia: results of total therapy study XIIIB at St Jude Children’s Research Hospital. Blood 104, 2690–2696.

McKenna, A., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K., Kernytsky, A., Garimella, K., Altshuler, D., Gabriel, S., Daly, M., and DePristo, M.A. (2010). The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303. Merkenschlager, M. (2010). Ikaros in immune receptor signaling, lymphocyte differentiation, and function. FEBS Lett. 584, 4910–4914. Moriyama, T., Metzger, M.L., Wu, G., Nishii, R., Qian, M., Devidas, M., Yang, W., Cheng, C., Cao, X., Quinn, E., et al. (2015). Germline genetic variation in ETV6 and risk of childhood acute lymphoblastic leukaemia: a systematic genetic study. Lancet Oncol. 16, 1659–1666. Mullighan, C.G., Miller, C.B., Radtke, I., Phillips, L.A., Dalton, J., Ma, J., White, D., Hughes, T.P., Le Beau, M.M., Pui, C.H., et al. (2008). BCR-ABL1 lymphoblastic leukaemia is characterized by the deletion of Ikaros. Nature 453, 110–114. Mullighan, C.G., Su, X., Zhang, J., Radtke, I., Phillips, L.A., Miller, C.B., Ma, J., Liu, W., Cheng, C., Schulman, B.A., et al. (2009). Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. N. Engl. J. Med. 360, 470–480. Nicorici, D., Satalan, M., Edgren, H., Kangaspeska, S., Murumagi, A., Kallioniemi, O., Virtanen, S., and Kilkku, O. (2014). FusionCatcher-a tool for finding somatic fusion genes in paired-end RNA-sequencing data. bioRxiv. https://doi.org/10.1101/011650. Noetzli, L., Lo, R.W., Lee-Sherick, A.B., Callaghan, M., Noris, P., Savoia, A., Rajpurkar, M., Jones, K., Gowan, K., Balduini, C.L., et al. (2015). Germline mutations in ETV6 are associated with thrombocytopenia, red cell macrocytosis and predisposition to lymphoblastic leukemia. Nat. Genet. 47, 535–538. O’Neill, D.W., Schoetz, S.S., Lopez, R.A., Castle, M., Rabinowitz, L., Shor, E., Krawchuk, D., Goll, M.G., Renz, M., Seelig, H.-P., et al. (2000). An ikaros-containing chromatin-remodeling complex in adult-type erythroid cells. Mol. Cell. Biol. 20, 7572–7582.

Roberts, K.G., Gu, Z., Payne-Turner, D., McCastlain, K., Harvey, R.C., Chen, I.M., Pei, D., Iacobucci, I., Valentine, M., Pounds, S.B., et al. (2017). High frequency and poor outcome of Philadelphia chromosome-like acute lymphoblastic leukemia in adults. J. Clin. Oncol. 35, 394–401. Roberts, K.G., Li, Y., Payne-Turner, D., Harvey, R.C., Yang, Y.L., Pei, D., McCastlain, K., Ding, L., Lu, C., Song, G., et al. (2014). Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N. Engl. J. Med. 371, 1005–1015. Robinson, J.T., Thorvaldsdottir, H., Winckler, W., Guttman, M., Lander, E.S., Getz, G., and Mesirov, J.P. (2011). Integrative genomics viewer. Nat. Biotechnol. 29, 24–26. Shah, S., Schrader, K.A., Waanders, E., Timms, A.E., Vijai, J., Miething, C., Wechsler, J., Yang, J., Hayes, J., Klein, R.J., et al. (2013). A recurrent germline PAX5 mutation confers susceptibility to pre-B cell acute lymphoblastic leukemia. Nat. Genet. 45, 1226–1231. Topka, S., Vijai, J., Walsh, M.F., Jacobs, L., Maria, A., Villano, D., Gaddam, P., Wu, G., McGee, R.B., Quinn, E., et al. (2015). Germline ETV6 mutations confer susceptibility to acute lymphoblastic leukemia and thrombocytopenia. PLoS Genet. 11, e1005262. Trevino, L.R., Yang, W., French, D., Hunger, S.P., Carroll, W.L., Devidas, M., Willman, C., Neale, G., Downing, J., Raimondi, S.C., et al. (2009). Germline genomic variants associated with childhood acute lymphoblastic leukemia. Nat. Genet. 41, 1001–1005. Van der Auwera, G.A., Carneiro, M.O., Hartl, C., Poplin, R., Del Angel, G., LevyMoonshine, A., Jordan, T., Shakir, K., Roazen, D., Thibault, J., et al. (2013). From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline. Curr. Protoc. Bioinformatics 43, 1–33.

Oravecz, A., Apostolov, A., Polak, K., Jost, B., Le Gras, S., Chan, S., and Kastner, P. (2015). Ikaros mediates gene silencing in T cells through Polycomb repressive complex 2. Nat. Commun. 6, 8823.

Van der Veer, A., Zaliova, M., Mottadelli, F., De Lorenzo, P., Te Kronnie, G., Harrison, C.J., Cave, H., Trka, J., Saha, V., Schrappe, M., et al. (2014). IKZF1 status as a prognostic feature in BCR-ABL1-positive childhood ALL. Blood 123, 1691–1698.

Perez-Andreu, V., Roberts, K.G., Harvey, R.C., Yang, W., Cheng, C., Pei, D., Xu, H., Gastier-Foster, J., E, S., Lim, J.Y., et al. (2013). Inherited GATA3 variants are associated with Ph-like childhood acute lymphoblastic leukemia and risk of relapse. Nat. Genet. 45, 1494–1498.

Wang, K., Li, M., and Hakonarson, H. (2010). ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 38, e164.

Perez-Andreu, V., Roberts, K.G., Xu, H., Smith, C., Zhang, H., Yang, W., Harvey, R.C., Payne-Turner, D., Devidas, M., Cheng, I.M., et al. (2015). A genome-wide association study of susceptibility to acute lymphoblastic leukemia in adolescents and young adults. Blood 125, 680–686. Powell, B.C., Jiang, L., Muzny, D.M., Trevino, L.R., Dreyer, Z.E., Strong, L.C., Wheeler, D.A., Gibbs, R.A., and Plon, S.E. (2013). Identification of TP53 as an acute lymphocytic leukemia susceptibility gene through exome sequencing. Pediatr. Blood Cancer 60, E1–E3. Pruitt, K.D., Brown, G.R., Hiatt, S.M., Thibaud-Nissen, F., Astashyn, A., Ermolaeva, O., Farrell, C.M., Hart, J., Landrum, M.J., McGarvey, K.M., et al. (2014). RefSeq: an update on mammalian reference sequences. Nucleic Acids Res. 42, D756–D763. Pui, C.H., Campana, D., Pei, D., Bowman, W.P., Sandlund, J.T., Kaste, S.C., Ribeiro, R.C., Rubnitz, J.E., Raimondi, S.C., Onciu, M., et al. (2009). Treating

12 Cancer Cell 33, 1–12, May 14, 2018

Whitlock, C.A., and Witte, O.N. (1987). Long-term culture of murine bone marrow precursors of B lymphocytes. Methods Enzymol. 150, 275–286. Williams, R.T., Roussel, M.F., and Sherr, C.J. (2006). Arf gene loss enhances oncogenicity and limits imatinib response in mouse models of Bcr-Ablinduced acute lymphoblastic leukemia. Proc. Natl. Acad. Sci. USA 103, 6688–6693. Yang, J.J., Cheng, C., Devidas, M., Cao, X., Campana, D., Yang, W., Fan, Y., Neale, G., Cox, N., Scheet, P., et al. (2012). Genome-wide association study identifies germline polymorphisms associated with relapse of childhood acute lymphoblastic leukemia. Blood 120, 4197–4204. Zhang, M.Y., Churpek, J.E., Keel, S.B., Walsh, T., Lee, M.K., Loeb, K.R., Gulsuner, S., Pritchard, C.C., Sanchez-Bonilla, M., Delrow, J.J., et al. (2015). Germline ETV6 mutations in familial thrombocytopenia and hematologic malignancy. Nat. Genet. 47, 180–185.

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

STAR+METHODS KEY RESOURCES TABLE

REAGENT or RESOURCE

SOURCE

IDENTIFIER

FITC-conjugated mouse anti-human CD90

BD

Cat# 555595; RRID: AB_395969

PE-conjugated mouse anti-human CD73

BD

Cat# 550257; RRID: AB_393561

APC-conjugated mouse anti-human CD105

BD

Cat# 562408; RRID: AB_11154045

APC-H7-conjugated mouse anti-human CD45

BD

Cat# 560178; RRID: AB_1645479

PE-conjuated mouse anti-human CD14

BD

Cat# 561707; RRID: AB_10924593

PerCP-Cy5.5-conjugated mouse anti-human CD34

BD

Cat# 347203; RRID: AB_400266

PE-conjugated mouse anti-human CD19

BD

Cat# 349209; RRID: AB_400407

APC-conjugated mouse anti-human HLA-DR

BD

Cat# 560896; RRID: AB_10563218

Rabbit anti-IKAROS (H-100)

Santa Cruz

Cat# sc-13039; RRID: AB_2124699

Rabbit anti-HA

Abcam

Cat# ab9110; RRID: AB_307019

Mouse anti-FLAG (clone M2)

Sigma

Cat# F1804; RRID: AB_262044

Human bone marrow aspirates

COG, SJCRH, and Children’s €bingen, University Hospital, Tu Germany

AALL0232, AALL0331, P9904, P9905, P9906, Total Therapy XIIIA, XIIIB, XV, and Berlin-Frankfurt-Munster (BFM) 99

Human peripheral blood samples

COG, SJCRH, and Children’s €bingen, University Hospital, Tu Germany

AALL0232, AALL0331, P9904, P9905, P9906, Total Therapy XIIIA, XIIIB, XV, and BFM99

Epstein-Barr virus (EBV)-transformed lymphocytes

Children’s University Hospital, €bingen, Germany Tu

BFM99

Patient SNP array, whole exome sequencing, and RNA-sequencing data

This paper

EGAS00001002838

Patient SNP array data

This paper

phs001350.v1.p1

Patient SNP array data

This paper

phs000638.v1.p1

Antibodies

Biological Samples

Deposited Data

Experimental Models: Cell Lines HEK293T cells

ATCC

CRL-3216

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IK6-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1M31V-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1V52L-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1V53M-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1R69H-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1D81N-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCV-IKZF1S105L-ires-RFP pre-B cells

This paper

N/A (Continued on next page)

Cancer Cell 33, 1–12.e1–e8, May 14, 2018 e1

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1R162P-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1H163Y-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1D186fs-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1D252N-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1S258P-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1M306*-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1T333A-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1G337S-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1M347V-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1Y348C-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1A365V-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1C394*-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1L411F-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1R423C-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1A434G-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1L449F-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1M459V-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 + MSCVIKZF1M476T-ires-RFP pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 CRISPRIKZF1M31V/M31Vpre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 CRISPRIKZF1R162P/R162P pre-B cells

This paper

N/A

Arf-/- MSCV-BCR-ABL1-ires-Luc2 CRISPRIKZF1A430G/A430G pre-B cells

This paper

N/A

Arf-/- mice (C57Bl/6 background)

Kamijo et al., 1997

N/A

C57Bl/6 mice

Jackson Laboratory

Cat# 000664; RRID: IMSR_JAX000664

C1934 forward primer 5’ ttcccctccccggtt gtagatttca 3’

This paper

N/A

C1935 reverse primer 5’gccttggagagcag cagcaggttc 3’

This paper

N/A

C1936 forward primer 5’gagaacctgct gctgctctccaagg 3’

This paper

N/A

Experimental Models: Organisms/Strains

Oligonucleotides

(Continued on next page)

e2 Cancer Cell 33, 1–12.e1–e8, May 14, 2018

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

C1937 reverse primer 5’gagaaggcggca gtccttgtgctt 3’

This paper

N/A

C1866 forward primer 5’actggctcca cccagtacct 3’

This paper

N/A

C1867 reverse primer 5’cccatcctgc tgatctttgt 3’

This paper

N/A

IK-bs4 3’ biotin-labeled forward: 5’gtcatga cagggaatacacattcccaaaagc 3’

Cobb et al., 2000

N/A

IK-bs4 5’ biotin-labeled reverse: 5’gtcagc ttttgggaatgtgtattccctgtca 3’

Cobb et al., 2000

N/A

MSCV-BCR-ABL1-ires-Luc2

Boulos et al., 2011

N/A

MSCV-IKZF1-ires-RFP

Mullighan et al., 2008

N/A

MSCV-IK6-ires-RFP

Mullighan et al., 2008

N/A

MSCV-IKZF1P18T-ires-RFP

This paper

N/A

MSCV-IKZF1M31V-ires-RFP

This paper

N/A

MSCV-IKZF1V52L-ires-RFP

This paper

N/A

MSCV-IKZF1V53M-ires-RFP

This paper

N/A

MSCV-IKZF1R69H-ires-RFP

This paper

N/A

MSCV-IKZF1D81N-ires-RFP

This paper

N/A

MSCV-IKZF1S105L-ires-RFP

This paper

N/A

MSCV-IKZF1R162P-ires-RFP

This paper

N/A N/A

Recombinant DNA

H163Y

-ires-RFP

This paper

MSCV-IKZF1D186fs-ires-RFP

This paper

N/A

MSCV-IKZF1D252N-ires-RFP

This paper

N/A

MSCV-IKZF1

MSCV-IKZF1S258P-ires-RFP

This paper

N/A

MSCV-IKZF1M306*-ires-RFP

This paper

N/A

MSCV-IKZF1T333A-ires-RFP

This paper

N/A

MSCV-IKZF1G337S-ires-RFP

This paper

N/A

MSCV-IKZF1M347V-ires-RFP

This paper

N/A

MSCV-IKZF1Y348C-ires-RFP

This paper

N/A

MSCV-IKZF1A365V-ires-RFP

This paper

N/A

MSCV-IKZF1C394*-ires-RFP

This paper

N/A

L411F

-ires-RFP

This paper

N/A

MSCV-IKZF1P420Q-ires-RFP

This paper

N/A

MSCV-IKZF1

MSCV-IKZF1R423C-ires-RFP

This paper

N/A

MSCV-IKZF1H432Q-ires-RFP

This paper

N/A

MSCV-IKZF1A434G-ires-RFP

This paper

N/A

MSCV-IKZF1L449F-ires-RFP

This paper

N/A

MSCV-IKZF1M459V-ires-RFP

This paper

N/A

MSCV-IKZF1M476T-ires-RFP

This paper

N/A

MSCV-IKZF1M518K-ires-RFP

This paper

N/A

pcDNA3.1-FLAG-IKZF1

This paper

N/A

pcDNA3.1-FLAG-IK6

This paper

N/A

pcDNA3.1-HA-IKZF1

This paper

N/A

pcDNA3.1-HA-IK6

This paper

N/A

pcDNA3.1-HA-IKZF1P18T

This paper

N/A

pcDNA3.1-HA-IKZF1M31V

This paper

N/A

pcDNA3.1-HA-IKZF1V52L

This paper

N/A

pcDNA3.1-HA-IKZF1V53M

This paper

N/A (Continued on next page)

Cancer Cell 33, 1–12.e1–e8, May 14, 2018 e3

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Continued REAGENT or RESOURCE

SOURCE

IDENTIFIER

pcDNA3.1-HA-IKZF1R69H

This paper

N/A

pcDNA3.1-HA-IKZF1D81N

This paper

N/A

pcDNA3.1-HA-IKZF1S105L

This paper

N/A

pcDNA3.1-HA-IKZF1R162P

This paper

N/A

pcDNA3.1-HA-IKZF1H163Y

This paper

N/A

pcDNA3.1-HA-IKZF1D186fs

This paper

N/A

pcDNA3.1-HA-IKZF1D252N

This paper

N/A

pcDNA3.1-HA-IKZF1S258P

This paper

N/A

pcDNA3.1-HA-IKZF1M306*

This paper

N/A

pcDNA3.1-HA-IKZF1T333A

This paper

N/A

pcDNA3.1-HA-IKZF1G337S

This paper

N/A

pcDNA3.1-HA-IKZF1M347V

This paper

N/A

pcDNA3.1-HA-IKZF1Y348C

This paper

N/A

A365V

This paper

N/A

pcDNA3.1-HA-IKZF1C394*

This paper

N/A

pcDNA3.1-HA-IKZF1L411F

This paper

N/A

pcDNA3.1-HA-IKZF1P420Q

This paper

N/A

pcDNA3.1-HA-IKZF1R423C

This paper

N/A

pcDNA3.1-HA-IKZF1H432Q

This paper

N/A

pcDNA3.1-HA-IKZF1A434G

This paper

N/A

pcDNA3.1-HA-IKZF1L449F

This paper

N/A

pcDNA3.1-HA-IKZF1M459V

This paper

N/A

pcDNA3.1-HA-IKZF1M476T

This paper

N/A

pcDNA3.1-HA-IKZF1M4518K

This paper

N/A

pGL3-MCL1

This paper

N/A

pRL-TK

Promega

Cat# E2241

STRUCTURE (version 2.3.4)

Pritchard Lab, Stanford University

https://web.stanford.edu/group/pritchardlab/ structure_software/release_versions/ v2.3.4/html/structure.html

CONSERTING

Chen et al., 2015

http://www.stjuderesearch.org/site/ lab/zhang

Prism (version 7.03)

Graphpad

www.graphpad.com

GATK

Broad Institute

https://gatkforums.broadinstitute.org/gatk

STAR

Dobin et al., 2013

https://github.com/alexdobin/STAR/releases

Samtools

Li et al., 2009b

https://github.com/samtools

CICERO

Roberts et al., 2014

N/A

FusionCatcher

Nicorici et al., 2014

https://github.com/ndaniel/fusioncatcher

pcDNA3.1-HA-IKZF1

Software and Algorithms

CONTACT FOR REAGENT AND RESOURCE SHARING Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Charles Mullighan ([email protected]). An MTA will be required for all requested materials. EXPERIMENTAL MODEL AND SUBJECT DETAILS Human Subjects €bingen, Germany, and treated with comThe proband of the index family with ALL was identified at Children’s University Hospital, Tu bination chemotherapy on a Berlin-Frankfurt-Munster protocol (BFM 99). The recurrence cohort of ALL cases was comprised of 4,963 children with newly-diagnosed ALL (n=4,857 B-ALL and 106 T-ALL) enrolled on the Children’s Oncology Group (COG) AALL0232, AALL0331, P9904, P9905, P9906 protocols, and St Jude Children’s Research Hospital (SJCRH) Total Therapy XIIIA, XIIIB e4 Cancer Cell 33, 1–12.e1–e8, May 14, 2018

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

and XV studies (Table S1) (Appelmann et al., 2015; Bowman et al., 2011; Larsen et al., 2016; Pui et al., 2004, 2006, 2009). This study was approved by institutional review boards at SJCRH and COG affiliated institutions and informed consent was obtained from parents, guardians, or patients, and assent from the patient, as appropriate. Family histories were not available for patients on the COG studies, and thus ALL cases were considered to be sporadic. Age and gender information for participants of this study can be found in Tables S1 and S2. Germline DNA was extracted from bone marrow samples or peripheral blood from children with ALL during remission. Most of the ALL cases had been previously genotyped with genome-wide single nucleotide polymorphism (SNP) arrays and genetic ancestry (European, African, Native American, and Asian) was estimated with STRUCTURE (version 2.3.4) on the basis of genotypes at 30,000 randomly selected SNPs with HapMap samples and indigenous Native Americans as ancestral populations (Boulos et al., 2011; Karol et al., 2017; Moriyama et al., 2015; Yang et al., 2012). Animals Mice were housed in an American Association of Laboratory Animal Care (AALAC)-accredited facility and all experiments were approved and in compliance with the St. Jude Children’s Research Hospital Institutional Animal Care and Use Committee (IACUC)-approved protocol (584-100506) in accordance with NIH guidelines. Female C57Bl/6 Arf-/- mice (Kamijo et al., 1997) were used as bone marrow donors and female wild-type C57Bl/6 mice (Jackson Laboratory, Bar Harbor, ME) were used as recipient mice for transplant experiments. For all pre-B cell transplantation experiments, 200,000 BCR-ABL1-expressing pre-B cells were transplanted by tail vein injection into sublethally (5.5 Gy) irradiated 8-10 week old female C57Bl/6 recipients. For in vivo imaging, three mice were injected per group. For in vivo treatment studies, five mice were randomly allocated to each study arm. Dasatinib (LC Laboratories, Woburn, MA) was administered once-daily by oral gavage at 10 mg/kg in 80 mM citric acid (pH 3.1) and continued until animals became moribund. Animals were monitored daily and sacrificed when moribund or upon clinically manifest central nervous system involvement. Primary Mouse Pre-B Cell Culture MSCV vectors expressing human BCR-ABL1 and luciferase were used for transduction of whole bone marrow from 8 week old female C57Bl/6 Arf-/- mice (Kamijo et al., 1997) and generation of in vitro pre-B cell cultures without stromal support or addition of cytokines (Whitlock and Witte, 1987; Williams et al., 2006). Upon establishment of BCR-ABL1-transformed cultures after 7-8 days, Arf-/- BCR-ABL1+ pre-B cells were transduced with MSCV retroviral supernatants expressing MSCV-IKZF1variant-iresRFP or emptyvector. Cells were RFP-sorted prior to subsequent culture and/or transplantation into mice. All pre-B cell cultures were grown at 37 Cwith 8% CO2 and maintained in Roswell Park Memorial Institute medium (RPMI 1640) supplemented with 10% fetal calf serum(Sigma, penicillin, streptomycin, glutamine and 55 mM b-mercaptoethanol. Cell Lines HEK293T cells (ATCC, Manassas, VA) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal calf serum, penicillin, streptomycin, and glutamine. METHOD DETAILS IKZF1 Sequencing In the index family, IKZF1 exons were amplified from genomic DNA (germline and matched ALL whenever applicable) by PCR and variants were identified by Sanger sequencing. For the deceased individual II-2 of this kindred (Figure 1B), DNA was isolated from archived unstained slides smeared with bone marrow aspirates containing <5% blasts from remission (germline) samples or 70% blasts from relapse samples. Fluorescence in situ hybridization was attempted using archived slides containing 73% blasts from a second relapse sample, however the assay failed and was inconclusive. In the sporadic ALL cohort for targeted IKZF1 sequencing, Illumina dual-indexed libraries were created from the germline DNA of 4,963 children with ALL, and pooled in sets of 96 before hybridization with customized Roche NimbleGene SeqCap EZ probes (Roche, Roche NimbleGen, Madison, WI, USA) to capture the IKZF1 genomic region. Quantitative PCR was used to define the appropriate capture product titer necessary to efficiently populate an Illumina HiSeq 2000 flowcell for paired-end 2x100 bp sequencing. Coverage of at least 20-fold depth was achieved across the targeted IKZF1 locus for 99.2% of samples. Sequence reads in FASTQ format were mapped and aligned using the Burrows-Wheeler Aligner (BWA) (Li and Durbin, 2009a), and genetic variants were called using the GATK pipeline (version 3.1) (McKenna et al., 2010), as previously described, and annotated using the ANNOVAR program (Wang et al., 2010) with the annotation databases including RefSeq (Pruitt et al., 2014), CADD (Kircher et al., 2014), Polyphen2 (Adzhubei et al., 2010, 2013) and SIFT (Kumar et al., 2009). Non-coding, and synonymous coding variants were excluded from further consideration for this study. All remaining variants were evaluated for their prevalence in general populations (minor allele frequency <0.2% in the Exome Aggregation Consortium [ExAC] cohort excluded TCGA) (Lek et al., 2016) and manually reviewed to ensure variant-containing sequence reads comprised at least 30% of sequence coverage at each locus. All the IKZF1 non-silent variants were manually reviewed in the Integrative Genomics Viewer (IGV) (Robinson et al., 2011). The minimum alternative allele frequency was > 30% in the final variant list. Finally, we also estimated germline copy number variation from genome-wide SNP array data for the vast majority of the ALL cases studied here but did not observe any focal amplifications or deletions affecting IKZF1 (not shown).

Cancer Cell 33, 1–12.e1–e8, May 14, 2018 e5

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

Four frozen remission bone marrow aspirate samples were available for Sanger sequencing of genomic DNA isolated from mesenchymal stromal cells from patients ALL_05, ALL_06, ALL_21, ALL_36. These samples were thawed and cultured on culture dish in Minimum Essential Medium Eagle (Sigma M4526) supplemented with 10% FBS and Penicillin-Streptomycin-Glutamine (Gibco 10378016). Culture dish was washed with PBS and supplemented with fresh media after 24 hr of culture. Medium was changed every 3 days. When the dish reached confluence, trypsin was added to the dish and cells were replated into larger dish. After 20 days of culture, all cells reached confluence and were collected. Rabbit immunoglobulin was added to each sample for blocking. Cells were stained with the following antibodies; FITC mouse anti-human CD90 (BD 555595), PE Mouse Anti-Human CD73 (BD 550257), APC Mouse Anti-Human CD105 (BD 562408), and APC-H7 Mouse Anti-Human CD45 (BD 560178). Cells positive for CD105, CD73, CD90, and negative for CD45 were sorted for DNA extraction. Before sorting, immunophenotype of cells were characterized with the following additional antibodies; PE Mouse Anti-Human CD14 (BD561707), PerCP-Cy5.5 Mouse Anti-Human CD34 (BD347203), PE Mouse Anti-Human CD19 (BD 349209), and APC Mouse Anti-Human HLA-DR (BD 560896). DNA was extracted using QIAamp DNA mini kit (QIAGEN 51306). IKZF1 exons were amplified from genomic DNA by PCR and variants were confirmed by Sanger sequencing (primer pairs for detecting variants: C1934 + C1935 for G337S; C1936 + C1937 for P420Q; and C1866 + C1867 for V53M, see Key Resources Table for sequences). To determine the germline structural variation at IKZF1 locus, we performed a focal screening of germline CNV results identified CONSERTING method (Chen et al., 2015). Specifically, the focal gain or loss segments at IKZF1 with absolute coverage log ratio (tumor against normal) greater than 0.15 were retained. We required at least two unique pairs of discordant reads to support the copy number deletion. Tumor in normal contamination was further evaluated using the other somatic mutations or somatic copy number changes for filtering. Genomic Analysis of Patient Samples Whole exome sequencing, RNA sequencing and SNP array analysis were performed for tumors with available material using approaches as described in Liu et al., 2017 and Roberts et al., 2017. Whole exome sequencing data was analyzed using Genome Analysis ToolKit (GATK, v3.7) (Van der Auwera et al., 2013) (https://gatkforums.broadinstitute.org/gatk). RNA-sequencing data was analyzed using STAR (Dobin et al., 2013; https://github.com/alexdobin/STAR/releases), Samtools (Li et al., 2009b; https://github. com/samtools/), FusionCatcher (Nicorici et al., 2014; https://github.com/ndaniel/fusioncatcher), and CICERO (Roberts et al., 2014). Expression Vectors and Retroviral Production cDNAs encoding BCR-ABL1 were cloned into mouse stem cell virus-internal ribosome entry site-luciferase, MSCV-IRES-Luc2 (Boulos et al., 2011). cDNAs encoding wild-type dominant-negative IKAROS (IK6) were PCR amplified from human leukemic cell RNA (Mullighan et al., 2008) and cloned into MSCV-IRES-RFP. IKZF1 point mutant vectors were generated from MSCV-IKZF1-IRESRFP using the QuikChange II XL Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA). These vectors were packaged into replication-incompetent, ecotropic retroviral particles by transient transfection of HEK293T cells with a triple plasmid system (MSCV-IKZF1 variant-IRES-RFP vector, pMD old gag pol and pCAG4-Eco). Co-immunoprecipitation Human wild-type IKZF1 and IK6 cDNA with an N-terminus HA tag were cloned into pcDNA3.1 expression vector using HiFi DNA Assembly Master Mix kit (NEB). Site directed mutagenesis was performed on the wild-type construct to obtain all IKZF1 variants for the following assay. HEK293T cells (1.25 x 106) were co-transfected with 2.5 mg pCDNA3.1-N-FLAG-IKZF1 (wild-type) and 2.5 mg pCDNA3.1-N-HA-IKZF1 (wild-type or indicated variants). Cultured 24 h after transfection, cells are lysed in lysis buffer (20 mM Tris-Hcl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP-40, 1% Sodium dexycholate) supplemented with proteosome inhibitors. Cell lysates were incubated with ANTI-FLAG M2 Magnetic Beads (Sigma) or Mouse IgG Magnetic Bead Conjugate for 4 hr, washed and re-suspended in the peptide buffer. Protein samples were separated on SDS-PAGE gels and immunoprecipitated with anti-HA (Abcam, 9110). Membrane was stripped and blotted with anti-FLAG (Sigma, F1804). Electrophoretic Mobility Shift Assays (EMSA) EMSA was performed using the LightShift Chemiluminescent EMSA Kit (Thermo Scientific, Waltham, MA) according to the manufacturer’s protocol with the following specifications. Nuclear extracts were prepared by transfecting HEK293T cells with 6 mg of each IKZF1 variant plasmid using FuGENE-HD (Promega, Madison, WI) and harvesting cells after 48 hr. Cells were washed with PBS and incubated in 5X packed cell volume (PCV) of hypotonic lysis buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, DTT and protease inhibitor cocktail) for 15 min on ice. Cells were centrifuged for 5 min at 420 g and resuspended in 2X PCV hypotonic lysis buffer, sheared with a 27 g needle, and centrifuged for 20 min at 10,000 g. The supernatant was collected as the cytoplasmic fraction. The pellet was incubated in extraction buffer (20 mM HEPES, pH 7.9, 1.5 mM MgCl2, 0.42 M NaCl, 0.2 mM EDTA, 25% glycerol, DTT and protease inhibitor cocktail) for 30 min on ice. Sample was centrifuged for 5 min at 20,000 g and the supernatant was collected as the nuclear fraction. Western blot was used with anti-IKAROS (H-100, Santa Cruz Biotechnology, Dallas, TX) to confirm the nuclear and cytoplasmic location of wild-type IKAROS and its variants. EMSA probes were derived from the regulatory regions of IK-bs4 (3’ biotin-labeled forward: 5’-gtcaTGACAGGGAATACACATTCCCAAAAGC; 5’ biotin-labeled reverse: 5’-gtcaGCTTTTGG GAATGTGTATTCCCTGTCA) (Cobb et al., 2000; Kuehn et al., 2016) and were synthesized by Integrated DNA Technologies (Coralville, IA). Unlabeled probes of the same sequence were synthesized to use in competition assays. Probes were annealed by e6 Cancer Cell 33, 1–12.e1–e8, May 14, 2018

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

combining forward and reverse oligonucleotides at a concentration of 10 mM in TE buffer, heating to 95 C for 5 min and cooling to room temp over 1.5 hr. EMSA reaction conditions were as follows: 1X binding buffer, 2.5% glycerol, 5 mM MgCl2, 50 ng/mL Poly (dI-dC), 0.05% NP-40, 10 mg HEK293T nuclear extract, and 25 nM biotin-labeled probe in a total volume of 20 mL. Unlabeled probes at various concentrations or an IKAROS antibody (Santa Cruz Biotechnology, Santa Cruz, CA) were included in reactions to confirm specific binding of the probes to IKAROS. Reactions were incubated at room temperature for 20 min prior to loading onto a 6% DNA retardation gel (Thermo Scientific, Waltham, MA) and running at 100 V in 0.5X TBE for approximately 1 hr. Gels were transferred to a Biodyne B Nylon Membrane (Thermo Scientific, Waltham, MA) at 30 V for 1 hr in 0.5X TBE. Transferred DNA was crosslinked for 10 min on a transilluminator with a 312 nm bulb according to the LightShift Chemiluminescent EMSA Kit protocol. Detection of Biotin-labeled DNA was carried out according to the kit protocol: the membrane was incubated in blocking buffer for 15 min at room temperature with shaking, followed by incubation with Streptavidin-HRP Conjugate in blocking buffer for 15 min. Membrane was washed 4X with 1X wash buffer for 5 min each, incubated in substrate equilibration buffer for 5 min, and finally incubated in substrate working solution before being imaged on a ChemiDoc Touch Imaging System (Bio-Rad, Hercules, CA). Immunofluorescence Cytospins of mouse pre-B cells and cryopreserved Epstein-Barr virus (EBV)-transformed lymphocytes from members of the index family, were fixed with 4% paraformaldehyde and washed with phosphate-buffered saline (PBS), followed by a 30 min incubation in a blocking/permeabilization solution of 10% normal goat serum (NGS)/0.1% Triton-X 100/PBS, and then incubated for 1 hr with primary anti-IKAROS antibody (H-100, Santa Cruz Biotechnology, Santa Cruz, CA) diluted in 3% NGS/0.1% Triton-X 100/ PBS. Slides were washed three times in PBS, and then incubated for 45 min in 3% NGS/0.1% Triton-X 100/PBS containing a secondary antibody conjugated to Ig-Alexa Fluor 488 (Invitrogen, Carlsbad, CA). Slides were washed three times in PBS and mounted with Vectashield (Vector Labs, Burlingame, CA) containing 4’-6-diamidino-2-phenylindol (DAPI). All steps were carried out at room temperature. Images were captured using a Nikon C2 confocal fluorescence microscope and analyzed for nuclear versus cytoplasmic localization using NIS Elements software. Luciferase Reporter Assay Human wild-type IKZF1 and IK6 cDNA with an N-terminus FLAG tag were cloned into pCDNA3.1 expression vector using HiFi DNA Assembly Master Mix kit (NEB). Site directed mutagenesis was performed to generate plasmids expressing IKZF1 variants (QuikChange II Site-Directed Mutagenesis Kit, Agilent Technologies, Santa Clara, CA). The promoter of MCL1 (hg19 Chr1:150551973-150553274) was cloned into pGL3 luciferase vector (Promega). For luciferase assays, 2.5x104 HEK293T cells cultured in 96-well plates were transiently transfected with 50 ng of pGL3-MCL1 plasmid, 50 ng of pcDNA IKZF1 plasmid (wildtype, IK6 or IKZF1 sequence variants) and 5 ng of pRL-TK (Renilla luciferase) using Lipofectamine 2000 (Thermo Scientific, Waltham, MA). For dominant negative luciferase assays, HEK293T cells were co-transfected with 10 ng of wild-type IKZF1 expression vector with the varying amounts (5 ng to 40 ng) of variant vector for IK6, R162P, H163Y, D186fs, M306*, or C394* (with empty vector so that total expression vector was 50 ng), 50 ng of pGL3-MCL1 reporter construct and 5 ng pRL-TK Renilla luciferase. Firefly luciferase activity was measured 24 hr post-transfection using the Dual Luciferase Assay (Promega). The results were normalized to Renilla luciferase activity. These experiments were repeated at least 3 times, and each sample was assayed in triplicate. Calvarial Imaging Calvaria were harvested at 72 hr after transplant, placed in PBS, and immediately imaged using a 25x 1.05 NA XLPlan water immersion objective on an Olympus Fluoview FVMPE-RS microscope equipped with a SpectraPhysics InSight DS+ Dual-OL laser. Excitation of RFP was at 1,040 nm, which simultaneously generates a second harmonic signal from the collagen in the bone at 520 nm. Flow Cytometric Analysis of Adhesion Proteins SELL-eFluor605, THY1.2-PerCP5.5, and ITGA5-AlexaFluor 647 conjugated antibodies (BD Biosciences, Franklin Lakes, NJ) were used for flow cytometric analysis of mouse in vitro overexpression. Cellular fluorescence data were collected for three technical replicates on an LSR II flow cytometer (BD Biosciences, Franklin Lakes, NJ) using DIVA software (BD Biosciences, Franklin Lakes, NJ), and analyzed with FlowJo vX.0.6 (Tree Star, Inc., Ashland, OR). Cell Viability Assays Arf-/- BCR-ABL1+ pre-B cells were plated 20,000/well in 96 well plates and treated with dasatinib or dexamethasone dissolved in DMSO for 72 hr at the indicated concentrations. Cells were incubated for 4 hr with resazurin and read on a Synergy HT (Biotek, Winooski, VT). Three technical replicates were used for each of three biological replicate experiments. CRISPR/Cas9 Genome Editing Modified Arf-/- BCR-ABL1+ pre-B cells were generated using CRISPR-Cas9 technology. Briefly, 600,000 Arf-/- BCR-ABL1+ pre-B cells were transiently co-transfected with 200 ng of gRNA expression plasmid, 500 ng Cas9 expression plasmid, 200 ng of pMaxGFP, and 5 mg of single-stranded donor oligonucleotide (ssODN) via nucleofection (Lonza, 4D-NucleofectorTM X-unit) using solution SG, program CM-147 in small cuvettes according to the manufacturers recommended protocol. Cells were single cell sorted by FACS to Cancer Cell 33, 1–12.e1–e8, May 14, 2018 e7

Please cite this article in press as: Churchman et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell (2018), https://doi.org/10.1016/j.ccell.2018.03.021

enrich for GFP+ (transfected) cells, clonally selected and verified for the desired targeted modification via targeted deep sequencing. At least three biological replicate were identified and assessed in relevant assays. Editing construct sequences are listed in Table S6. QUANTIFICATION AND STATISTICAL ANALYSIS Quantification of Cell-Cell Adhesion In Vitro Cells were seeded 2x105 per well and allowed to grow for 72 hr before being imaged at 10X magnification using an Eclipse TS100 light microscope (Nikon, Melville, NY). Images were analyzed using NIS Elements (Nikon, Melville, NY) to quantitate the number of aggregrates and measure the area of each aggregate. Statistical Analyses Analyses of functional data were performed using GraphPad Prism Version 6.0 (GraphPad, La Jolla, CA). All data are presented as mean ± SD. Significance was determined using Student’s t test, ANOVA, or Mantel-Cox log rank test as appropriate. A p value of less than 0.05 was considered significant. *p % 0.05, **p %0.005, *** p % 0.0005. DATA AND SOFTWARE AVAILABILITY The accessions numbers for the patient sequencing data reported in this paper are Database of Genotypes and Phenotypes (dbGAP): phs001350.v1.p1 and phs000638.v1.p1; and European Genome-phenome Archive: EGAS00001002838.

e8 Cancer Cell 33, 1–12.e1–e8, May 14, 2018