Notch-Mediated Expansion of Cord Blood Progenitors

Letters to the Editor 1948 patient number places some limitation on interpretation, there is a REFERENCES suggestion that this treatment could still be considered for patients 1 Auner HW, Szydlo R, Rone A, Chaidos A, Giles C, Kanfer E et al. Salvage autologous who may have not previously been referred for auto-HCT but who stem cell transplantation for multiple myeloma relapsing or progressing after up- now have relapsed disease. front autologous transplantation. Leuk Lymphoma 2013; 54: 2200–2204. Our adaptive study design ultimately determined that the 2 Shah N, Ahmed F, Bashir Q, Qureshi S, Dinh Y, Rondon G et al. Durable remission highest doses of lenalidomide (75 and 100 mg) were equivalent to with salvage second autotransplants in patients with multiple myeloma. Cancer 118 – achieve a balance between tolerability and efficacy. Owing to 2012; : 3549 3555. relatively small patient numbers, a longer follow-up time may 3 Lemieux E, Hulin C, Caillot D, Tardy S, Dorvaux V, Michel J et al. Autologous stem cell transplantation: an effective salvage therapy in multiple myeloma. Biol Blood further differentiate the true clinical effect of the dose escalation Marrow Transplant 2013; 19:445–449. and the impact of post-auto-HCT maintenance therapy, which 4 Cook G, Williams C, Brown JM, Cairns DA, Cavenagh J, Snowden JA et al. High- 63% of patients received. Unfortunately (although not surpris- dose chemotherapy plus autologous stem-cell transplantation as consolidation ingly), poor-risk cytogenetics independently predicted for poorer therapy in patients with relapsed multiple myeloma after previous autologous OS, reminding us of the continued need for innovative treatment stem-cell transplantation (NCRI myeloma × relapse (Intensive trial)): a randomised, options for these patients. open-label, phase 3 trial. Lancet Oncol 2014; 15: 874–885. In conclusion, combination of high-dose melphalan with high- 5 Kumar SK, Lacy MQ, Hayman SR, Stewart K, Buadi FK, Allred J et al. Lenalidomide, cyclophosphamide and dexamethasone (CRd) for newly diagnosed multiple dose lenalidomide appears to be a well tolerated and safe 86 – preparative regimen in the setting of salvage auto-HCT. As novel myeloma: results from a phase 2 trial. Am J Hematol 2011; :640 645. 6 van de Donk NW, Wittebol S, Minnema MC, Lokhorst HM. Lenalidomide (Revlimid) agents, maintenance regimens and immunotherapies emerge, it is combined with continuous oral cyclophosphamide (endoxan) and prednisone conceivable that they too may be combined with this auto-HCT (REP) is effective in lenalidomide/dexamethasone-refractory myeloma. Br J platform to offer the relapsed patient yet another chance at a Haematol 2010; 148: 335–337. durable response. 7 Durie BG, Harousseau JL, Miguel JS, Blade J, Barlogie B, Anderson K et al. International uniform response criteria for multiple myeloma. Leukemia 2006; 20:1467–1473. 8 Thall PF, Cook JD. Dose-finding based on efficacy-toxicity trade-offs. Biometrics CONFLICT OF INTEREST 2004; 60:684–693. χ2 Drs N Shah, Bashir, JJ Shah, Hosing and Orlowski receive research funding from 9 Fisher R. On the interpretation of from contingency tables, and the 85 – Celgene. Drs N Shah, JJ Shah and Orlowski have served on Celgene Advisory Boards. calculation of P. J R Stat Soc 1922; :87 94. 10 Freeman GH, Halton JH. Note on exact treatment of contingency, goodness of fit fi 38 – 1 2 2 1 3 1 4 and other problems of signi cance. Biometrika 1951; :141 149. N Shah , PF Thall , PS Fox , Q Bashir , JJ Shah , S Parmar , P Lin , 11 Gelman ACJ, Stern HS, Rubin DB. Bayesian Data Analysis, 2nd edn. Chapman & 1 1 1 1 1 P Kebriaei , Y Nieto , UR Popat , CM Hosing , A Cornelison , Hall/CRC Press: New York, NY, USA, 2004. 1 3 1 1 EJ Shpall , RZ Orlowski , RE Champlin and MH Qazilbash 12 Ibrahim JG, Chen M-H, Sinha D. Bayesian Survival Analysis.Springer:NewYork,NY,2001. 1Department of Stem Cell Transplantation and Cellular Therapy, The 13 Doo NW, Thompson PA, Prince HM, Seymour JF, Ritchie D, Stokes K et al. University of Texas MD Anderson Cancer Center, Houston, TX, USA; Bortezomib with high dose melphalan conditioning for autologous transplant is 2Department of Biostatistics, The University of Texas MD Anderson safe and effective in patients with heavily pretreated and high risk multiple 54 – Cancer Center, Houston, TX, USA; myeloma. Leuk Lymphoma 2013; : 1465 1472. 3Department of Lymphoma and Myeloma, The University of Texas 14 Orlowski RZ. Novel agents for multiple myeloma to overcome resistance in phase III clinical trials. Semin Oncol 2013; 40:634–651. MD Anderson Cancer Center, Houston, TX, USA and 4 15 Kumar SK, Lee JH, Lahuerta JJ, Morgan G, Richardson PG, Crowley J et al. Risk of Department of Hematopatholoy, The University of Texas MD progression and survival in multiple myeloma relapsing after therapy with IMiDs Anderson Cancer Center, Houston, TX, USA and bortezomib: a multicenter international myeloma working group study. E-mail: [email protected] Leukemia 2012; 26:149–157.

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

Notch-mediated expansion of cord blood progenitors: maintenance of transcriptional and epigenetic ﬁdelity Leukemia (2015) 29, 1948–1951; doi:10.1038/leu.2015.61 provide more rapid short-term myeloid reconstitution when co- infused with a non-manipulated CB unit(s). However, infusion of ex vivo generated HSPCs for long-term hematopoietic reconstitu- Recent advances have been made towards successful generation tion remains a goal in the context of stem cell transplantation and of human hematopoietic stem and progenitor cells (HSPCs) gene therapy, thus the development of methods to assess the derived from embryonic, induced pluripotent or hematopoietic safety of these cultured products are essential. stem cells. However, whether these stem cell populations, which Our group and others have demonstrated successful expansion are generated after ex vivo manipulation, are suitable for clinical of CB HSPCs for both pre-clinical and clinical applications. application depends on the faithful preservation of their Standardized methods for assessing transcriptional quality control transcriptional, epigenetic and functional properties with respect of pluripotent cell lines have been developed; however, there has to their primary cell counterparts. been no standard approach to assessing long-term safety of To overcome the signiﬁcant delay in neutrophil recovery cultured progenitor cells infused in humans.1 Here we investi- following cord blood (CB) transplantation (CBT), we developed gated the transcriptional and epigenetic states of CB HSPC methods for the ex vivo expansion of CD34+ CB-derived HSPC by generated after ex vivo expansion on Delta1. We found that the culture with the Notch-ligand Delta1. When infused into patients transcriptomes and methylomes of the expanded HSPC overall undergoing a myeloablative CBT, these cells have been shown to recapitulate those of their primary cell counterparts.

Accepted article preview online 6 March 2015; advance online publication, 27 March 2015

Leukemia (2015) 1939 – 1958 © 2015 Macmillan Publishers Limited Letters to the Editor 1949 CD34+ CB HSPCs were cultured under conditions identical to cells. To determine whether BEX2 downregulation was the result those in our ongoing clinical trials.2 We validated in vitro and of altered expression of transcription factors known to regulate in vivo growth characteristics of cells used for transcriptional and BEX2, we looked at the expression levels of SOX2, β-catenin, p65 methylation studies to ensure their similarity with our previously and ERBB2 and found no difference between primary and published results. Mean CD34+ fold expansion was 191 ± 62-fold cultured cells. as compared with 184 ± 35 previously. In vivo repopulation in We compared genes upregulated in the cultured cells with an sublethally irradiated (275 rad) NOD-SCID IL-2Rγ-null mice (NSG) established list of proto-oncogenes (www.uniprot.org/uniprot, approved for use by the Fred Hutchinson Cancer Research Center search terms ‘proto-oncogene’, ‘human’) as upregulation of these Institutional Animal Care and Use Committee was at least as genes might be expected to result in unintended cell proliferation. robust as our previously published results. Among the list of 232 proto-oncogenes, 11 were upregulated in For transcriptional analysis, CD34+ CB HSPCs were isolated from cultured cells (Figure 1c). Although many of these genes have six individual CB units within 24 h of collection. Freshly isolated been implicated in hematologic malignancies, they also have CD34+ CB HSPCs were then frozen for subsequent RNA extraction, important roles in normal hematopoiesis, and thus may be whereas remaining cells were placed in culture for 14 days as upregulated in this setting. Because the CB HSPCs were cultured in previously described.2 Upon cell harvest at day 14 of culture, the presence of Delta1, we also determined whether there was CD34+ cells were selected for RNA extraction and analysis on the persistent altered expression of Notch downstream target genes Illumina HT12v3 platform (Illumina Inc, San Diego, CA, USA). All as some of these have known roles in cell cycle entry and data were analyzed using the Bioconductor framework.3 Lumi proliferation. We found no difference in the expression of known package4 was used for reading and normalizing the raw data Notch targets SKP2, HES1, HEY1, CCND1 or BCL2. As shown in output by GenomeStudio software (Illumina Inc.). Limma package3 Figure 1c, MYC expression was upregulated following culture on was used for identifying differentially expressed genes between Delta1 ligand; however, Notch is one of many regulators of MYC freshly isolated and cultured CD34+ CB HSPCs. Significant genes expression. were defined as those with false discovery rate (FDR) o1% and Whereas transcriptional activity represents cellular function, we 4twofold change between non-manipulated and ex vivo expected that DNA methylation states would offer a more detailed expanded cells. picture of the fidelity of the expanded populations with respect to Of 24 912 genes analyzed, 668 genes were upregulated in the cell identity and thus safety for clinical use. We assessed the ex vivo expanded groups and 616 genes were downregulated as epigenetic landscape of the cultured cells through genome-wide compared with non-manipulated control HSPCs from the same CB DNA methylation studies. CD34+-selected CB HSPCs were cultured units. Using DAVID Bioinformatics Resources v6.7, a web-based for 14 days and again CD34 selected for comparison with primary functional annotation tool for data analysis of microarray data cells. DNA methylation analysis was performed by high- (http://david.abcc.ncifcrf.gov/home.jsp), we performed gene throughput reduced representation bisulfite sequencing profiling ontology (GO) analysis for biological processes (GOTERM_BP_FAT) of CD34-selected primary and cultured cells.10 The original data of these differentially regulated genes.5,6 Differentially expressed set contained 24 981 genes. M-value statistics were used to genes were used as the inputs and ‘GOTERM_BP_FAT’ was used as measure the methylation levels (M-value = log2((methy+1)/ the GO search term to identify terms related to biological (unmethy+1))). Genes with M-values of 0 across all samples processes. FDR o1% was used to identify GO terms that were were filtered out leaving 17 346 genes for analysis. Analysis was significantly different between fresh and cultured CD34+ cells. performed on primary and cultured cell populations from six Most GO terms identified were those related to cell culture individual CB units using Limma package3 with an FDR threshold conditions. Only four significant GO terms were identified in the of 1%, as previously described. Overall, DNA methylation downregulated gene set and these were associated with non- patterns were similar between primary and cultured cells replicative cellular state including nucleosome assembly, chroma- (Figure 2a) with only 21 genes identified with increased tin assembly and protein–DNA complex assembly. Significant GO methylation following culture, 14 of which have known function. terms with upregulated genes were overlapping and related to This is in contrast to many cultured cell lines that show cells in a replicative state as expected for cells in culture. These GO hypermethylation including at TSGs.11–13 None of these genes terms represented more general cell cycle phase (for example, overlappedwithtumorsuppressororproto-oncogeneseval- mitosis, M phase and regulation of cell cycle) or specific steps of uated in our transcriptional analysis. GO terms associated with cell division (for example, organelle fission, spindle organization these genes include cytoskeleton, translation, proteolysis and and chromosome segregation; Figure 1a). Notably, absent were cell activation (Figure 2b). GO terms associated with differentiation, lineage commitment or To better determine whether small numbers of cells were cellular proliferation. generated that had functionally significant alterations able to lead Given concern for acquisition of genomic alterations in cultured to clonal dominance, we transplanted freshly isolated or cultured cells and the risk these pose for dysregulation of cell growth, we CB HSPCs into NSG mice to determine whether in vivo growth compared up- and downregulated genes with the known gene selected for and subsequently allowed detection of clones of groups for tumor suppressor genes (TSGs) and proto-oncogenes. cultured cells with altered DNA methylation. CD34+ CB HSPCs We compared our list of downregulated genes with those were sorted from mice following 3–22 weeks of in vivo growth and on TSG, a TSG database, (http://bioinfo.mc.vanderbilt.edu/TSGene/ global DNA methylation analysis was performed. Using the Limma browser.cgi),7 as loss of function of these genes would suggest risk package, we compared the methylomes of transplanted primary of unregulated cell proliferation. Of the 718 human TSGs in the CD34+ cells and transplanted cultured CD34+ cells and their database, 38 overlapped with downregulated genes in cultured progeny and found no differential methylation at an FDR level of cells (Figure 1b). Looking more specifically at TSGs described in 1%. This included the 21 genes identified when comparing fresh leukemia and acute myeloid leukemia (40), only BEX2 overlapped and cultured cells further suggesting that this differential with our downregulated gene list. Although better described as a methylation in vitro did not provide in vivo growth advantage. TSG in malignant gliomas and breast cancer, hypomethylation and We additionally compared the methylomes of transplanted increased expression of BEX2 have been described in leukemia cultured CD34+ cells with primary CD34+ cells to determine with mixed lineage leukemia gene rearrangements.8,9 Down- whether any differentially methylated genes overlapped with regulation of BEX2 in the cultured cells is of unclear significance. those identified when fresh and cultured cells were compared. However, our subsequent methylation studies showed no Two genes were identified with differential methylation (DRG2 difference in BEX2 methylation between primary and cultured and SERPINA1D) but these did not overlap with those shown in

24912 genes compared primary v. cultured cells

Down- Up- regulated regulated FDR < 1% Tumor 38 Genes Proto 11 Genes LogFC ≥1 Suppressor Genes (616) Oncogenes (668) (718) (232)

668 genes up- 616 genes down- GENE LogFC GENE LogFC regulated regulated BEX2 -3.15745 AURKA 3.339183 GOTERM_BP_FAT TP53INP1 -2.86841 PIM1 2.264683 (FDR< 1%) BTG2 -2.73918 CSF1R 2.165061 RARRES3 -2.70021 MYC 2.077349 SNF1LK -2.12784 UP-REGULATED (42) DOWN-REGULATED (4) HCK 1.738054 TNFAIP3 -2.09496 M phase Nucleosome assembly PTTG1 1.702821 YPEL3 -1.90184 Nuclear Division Chromatin assembly STIL 1.542976 TXNIP -1.66932 Mitosis Protein-DNA complex BRI3BP 1.523564 MEG3 -1.66932 M phase of mitotic cycle Nucleosome organization FOS 1.33147 FOXO1 -1.57346 Cell cycle phase MYCN 1.299442 ETS1 -1.56665 Organelle fission RAB8A 1.104032 RASSF1 -1.51952 Mitotic cell cycle FOXO3 -1.50933 Cell cycle process CDKN2D -1.48259 Cell Cycle MN1 -1.4615 Cell division MAL -1.44081 Chromosome segregation CYGB -1.43286 DNA metabolic process KLF6 -1.41318 Mitotic sister chromatid separation HOPX -1.39124 DNA replication PPP3CC -1.33092 Sister chromatid segregation SOD2 -1.32811 Microtubule-based process PDCD4 -1.32682 Regulation of cell cycle SPTBN1 -1.28715 Microtubule cytoskeleton HBP1 -1.26385 organization PANX2 -1.25607 Spindle organization SMAD3 -1.25166 Regulation of mitotic cycle JUP -1.24726 DNA repair STAT5A -1.22794 Regulation of cell cycle LOX -1.22041 Cell cycle checkpoint XAF1 -1.18263 Sterol biosynthesis process BNIP3L -1.16532 Response to DNA damage EPHB4 -1.13012 Chromosome organization PNN -1.12302 Cellular response to stress TCTA -1.11545 Meiotic cell cycle HECA -1.11409 DNA-dependent DNA replication PTPRD -1.10937 Cholesterol biosynthetic process EIF1 -1.1022 DNA packaging ZFP36 -1.07601 Mitotic spindle organization M phase meiotic cycle Meiosis Chromosome condensation Macromolecular complex subunit organization Cellular macromolecular complex subunit organization Mitotic cell cycle checkpoint Mitotic chromosome condensation Regulation of mitosis Regulation of nuclear division Cellular macromolecular complex assembly Figure 1. Transcriptional analysis of expanded CB HSPCs. (a) Gene ontology (GO) analysis of differentially expressed genes between primary and expanded CB HSPCs. Differentially expressed genes were deﬁned as those with FDR o1% and LogFC ⩾ 1. GO analysis was performed using DAVID Bioinformatics Resources v6.7 (http://david.abcc.ncifcrf.gov/home.jsp), input term GOTERM_BP_FAT. Signiﬁcantly enriched GO terms are shown. (b, c) Differentially expressed genes were compared with established tumor suppressor (b) and oncogene (c) lists. Overlapping genes and enrichment are shown.

Figure 2a, providing further support that Notch signaling does genes shown to be up- or downregulated in our transcriptional not induce aberrant epigenetic changes. Overall, these data analysis to see whether altered methylation could contribute to demonstrate that culture of CB HSPCs on Delta ligand does not differential gene expression and found no difference in result in signiﬁcant alterations in DNA methylation of the methylation at these genes. Looking speciﬁcally at methylation cultured cells. Finally, we compared the methylation levels of status of the differentially expressed TSG and proto-oncogenes

10 Gene GOTERM_BP_FAT KRTAP6-1 Cytoskeleton, organelle KDR Angiogenesis, regulation of cell motion 5 TAGAP1 RhoGTPase, T-cell activation RHOH RAS GTPase, cell activation MRPL49 Translation DPP9 Proteolysis 0 KATNAL2 Cytoskeleton, microtubule MAMDC4 Protein localization/transport TGS1 RNA splicing/processing M value -5 PRAM1 Regulation of leukocyte function FAU Translation/elongation TBX19 Regulation of transcription -10 CYYR1 Intrinsic to membrane ITGB2 Integrin beta-2 component

-15 Fresh Cultured Figure 2. Methylation analysis of expanded CB HSPCs. (a) Violin plot depicting methylation density (M-value) of fresh (red) and cultured (blue) CB HSPCs. (b) Gene ontology (GO) analysis of hypermethylated genes in cultured CB HSPCs as compared with fresh cells. Differentially methylated genes were deﬁned as those with adjusted P-value (FDR) o0.01. GO analysis was performed using DAVID Bioinformatics Resources v6.7 (http://david.abcc.ncifcrf.gov/home.jsp), input term GOTERM_BP_FAT.

from our transcriptional analysis, we again found no significant 6Department of Pathology, Massachusetts General Hospital and differences in methylation. Harvard Medical School, Boston, MA, USA Taken together, our studies indicate that, in contrast to E-mail: [email protected] transcriptional differences between primary and cultured cells including at some TSG and proto-oncogenes reflecting differences in cellular proliferation, DNA methylation showed strong fidelity REFERENCES between the primary and cultured cells. Methods for assessing the 1 Bock C, Kiskinis E, Verstappen G, Gu H, Boulting G, Smith ZD et al. Reference Maps safety of cultured progenitor cell products for clinical intent will of human ES and iPS cell variation enable high-throughput characterization of become increasingly important as these products become more pluripotent cell lines. Cell 2011; 144: 439–452. widely utilized and intended for sustained, long-term engraftment. 2 Delaney C, Heimfeld S, Brashem-Stein C, Voorhies H, Manger RL, Bernstein ID. This will be particularly true for ex vivo derived cells, either from Notch-mediated expansion of human cord blood progenitor cells capable of rapid myeloid reconstitution. Nat Med 2010; 16: 232–236. induced pluripotent stem cells (iPSCs) or embryonic stem cells 3 Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S et al. (ESCs), or by direct conversion to hematopoietic stem cells. These Bioconductor: open software development for computational biology and data provide a baseline for future critical assessment of expanded bioinformatics. Genome Biol 2004; 5: R80. populations of HSPCs including iPSC- and ESC-derived HSPCs. 4 Du P, Kibbe WA, Lin SM. Lumi: a pipeline for processing Illumina microarray. Bioinformatics 2008; 24: 1547–1548. 5 Huang DW, Sherman BT, Lempicki RA. Systematic and integrative analysis of large CONFLICT OF INTEREST gene lists using DAVID Bioinformatics Resources. Nat Protoc 2009; 4:44–57. The Fred Hutchinson Cancer Research Center holds a patent on ‘methods for 6 Huang DW, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths immortalizing cells’ that covers the use of Notch ligand for expansion of toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37 – hematopoietic stem cells. IDB is an inventor on this patent. The remaining authors 2009; :1 13. declare no conflict of interest. 7 Zhao M, Sun J, Zhao Z. TSGene: a web resource for tumor suppressor genes. Nucleic Acids Res 2013; 41: D970–D976. 8 Rohrs S, Dirks WG, Meyer C, Marschalek R, Scherr M, Slany R et al. Hypomethy- ACKNOWLEDGEMENTS lation and expression of BEX2, IGSF4, TIMP3 indicative of MLL translocations in acute myeloid leukemia. Mol Cancer 2009; 8: 86. This work was supported by National Heart, Lung and Blood Institute grant 9 Naderi A, Liu J, Hughes-Davies L. BEX2 has a functional interplay with c-Jun/JNK U01HL100395 and National Institutes of Health Ruth L. Kirschstein National Research and p65/RelA in breast cancer. Mol Cancer 2010; 9:111. Service Award T32CA009351 (AD) and K12CA076930 (AD), as well as Hyundai Hope on 10 Meissner A, Gnirke A, Bell GW, Ramsahoye B, Lander ES, Jaenisch R. Reduced Wheels 2013 Hope Grant (AD). CD is a Damon Runyon Clinical Investigator. representation bisulfite sequencing for comparative high-resolution DNA methylation analysis. Nucleic Acids Res 2005; 33: 5868–5877. A Dahlberg1,2, S Woo3, C Delaney1,2, P Boyle4, A Gnirke4, C Bock4, 11 Meissner A, Mikkelsen TS, Gu H, Wernig M, Hanna J, Sivachenko A et al. BE Bernstein4,5,6, A Meissner4, R Gottardo3 and ID Bernstein1,2 Genome-scale DNA methylation maps of pluripotent and differentiated cells. 1Pediatric Oncology, Clinical Research Division, Fred Hutchinson Nature 2008; 454:766–770. 12 Doi A, Park I-H, Wen B, Murakami P, Aryee MJ, Irizarry R et al. Differential Cancer Research Center, Seattle, WA, USA; fi 2Department of Pediatrics, University of Washington, Seattle, methylation of tissue- and cancer-speci c CpG island shores distinguishes human induced pluripotent stem cells, embryonic stem cells and fibroblasts. Nat Genet Washington, USA; 41 – 3 2009; :1350 1353. PS Statistics, Vaccine and Infectious Disease Division, Fred 13 Laurent L, Ulitsky I, Slavin I, Tran H, Schork A, Morey R et al. Dynamic changes Hutchinson Cancer Research Center, Seattle, WA, USA; in the copy number of pluripotency and cell proliferation genes in human ESCs 4 Broad Institute of MIT and Harvard, Cambridge, MA, USA; and iPSCs during reprogramming and time in culture. Cell Stem Cell 2011; 8: 5Howard Hughes Medical Institute, Chevy Chase, MD, USA and 106–118.