Consistent Deregulation of Gene Expression Between Human and Murine MLL Rearrangement Leukemias

Published OnlineFirst January 20, 2009; DOI: 10.1158/0008-5472.CAN-08-3381

Research Article

Consistent Deregulation of Gene Expression between Human and Murine MLL Rearrangement Leukemias

Zejuan Li,1 Roger T. Luo,1 Shuangli Mi,1 Miao Sun,1 Ping Chen,1 Jingyue Bao,2 Mary Beth Neilly,1 Nimanthi Jayathilaka,1 Deborah S. Johnson,3 Lili Wang,2 Catherine Lavau,4 Yanming Zhang,1 Charles Tseng,3 Xiuqing Zhang,2 Jian Wang,2 Jun Yu,2 Huanming Yang,2 San Ming Wang,5 Janet D. Rowley,1 Jianjun Chen,1 and Michael J. Thirman1

1Department of Medicine, University of Chicago, Chicago, Illinois; 2Beijing Genomics Institute, Chinese Academy of Sciences, Beijing, China; 3Department of Biological Sciences, Purdue University Calumet, Hammond, Indiana; 4CNRS UPR9051, IUH, Paris, France; and 5Center for Functional Genomics, Division of Medical Genetics, Department of Medicine, ENH Research Institute, Northwestern University, Evanston, Illinois

Abstract translocations associated with human acute leukemias (1, 2). MLL Important biological and pathologic properties are often is located on human chromosome 11 band q23 and on mouse conserved across species. Although several mouse leukemia chromosome 9. More than 50 different loci are rearranged models have been well established, the genes deregulated in in11q23 leukemias involving MLL in either acute myelogenous leukemia (AML) or acute lymphoblastic leukemia (ALL; ref. 3). both human and murine leukemia cells have not been studied systematically. We performed a serial analysis of gene MLL rearrangements are associated with a poor prognosis (4). expression in both human and murine MLL-ELL or MLL-ENL MLL-ELL and MLL-ENL that result from t(11;19)(q23;p13.1) and leukemia cells and identified 88 genes that seemed to be t(11;19)(q23;p13.3), respectively (1, 5), are two common examples of significantly deregulated in both types of leukemia cells, these rearrangements. These two fusions are frequently involved in including 57 genes not reported previously as being deregu- human AML, whereas MLL-ENL is also involved in human ALL. lated in MLL-associated leukemias. These changes were The translocations result in an in-frame fusion of the NH2 terminus validated by quantitative PCR. The most up-regulated genes of the MLL gene and the COOH terminus of each partner gene. include several HOX genes (e.g., HOX A5, HOXA9, and HOXA10) Retroviral-mediated gene transfer of MLL-ENL and MLL-ELL and MEIS1, which are the typical hallmark of MLL rearrange- transforms primary myeloid progenitor cells and cause acute myeloid leukemia in mice (6, 7). ment leukemia. The most down-regulated genes include LTF, LCN2, MMP9, S100A8, S100A9, PADI4, TGFBI,andCYBB. Gene expression profiles differ between distinct subtypes of Notably, the up-regulated genes are enriched in gene ontology leukemia and provide specific markers for clinical diagnosis. It is terms, such as gene expression and transcription, whereas the commonly observed that important biological/pathologic proper- down-regulated genes are enriched in signal transduction and ties are often conserved across species (8, 9). Model organisms have apoptosis. We showed that the CpG islands of the down- contributed substantially to our understanding of the etiology of regulated genes are hypermethylated. We also showed that human disease and the development of new treatment method- seven individual microRNAs (miRNA) from the mir-17-92 ologies (10). However, although genetically engineered mouse cluster, which are overexpressed in human MLL rearrange- leukemia models have been well established (6, 7, 11, 12), there are ment leukemias, are also consistently overexpressed in mouse few systematic studies to identify and study the genes that exhibit MLL rearrangement leukemia cells. Nineteen possible targets similar abnormal expression patterns in both human and murine of these miRNAs were identified, and two of them (i.e., APP leukemia cells. To perform an interspecies gene expression and RASSF2) were confirmed further by luciferase reporter comparative study in leukemia, we used the serial analysis of gene expression (SAGE) technique (13) to compare gene expression and mutagenesis assays. The identification and validation of consistent changes of gene expression in human and murine between MLL-ELL or MLL-ENL myeloid leukemia progenitor cells MLL rearrangement leukemias provide important insights and normal myeloid progenitor cells in both humans and mice. into the genetic base for MLL-associated leukemogenesis. Herein, we report the identification and validation of differentially [Cancer Res 2009;69(3):1109–16] expressed genes commonly present in both human and murine MLL-ELL or MLL-ENL leukemias. Introduction Chromosome translocations are among the most common Materials and Methods genetic abnormalities in human leukemia. The MLL (mixed lineage Patient samples. The patient samples were obtained at the time of leukemia) gene was identified as a common target of chromosomal diagnosis and with informed consent at University of Chicago and were stored in liquid nitrogen until use. SAGE assay and data analysis. Cell purification, total RNAisolation, cDNAsynthesis, and SAGEwere carried out according to our established Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). procedures (13–15). SAGE tag sequences were extracted with SAGE 2000 S.M. Wang, J.D. Rowley, J. Chen, and M.J. Thirman contributed equally to this work. software. Tag counts were converted to counts per 100,000, and the Requests for reprints: J. Chen, J.D. Rowley, or M.J. Thirman, 5841 South Maryland expression data were cross-linked to UniGene clusters by extracting the Avenue, MC 2115, Chicago, IL 60637. E-mail: [email protected], 3¶-most NlaIII SAGE tag for each transcript in each UniGene clusters. Only [email protected], or [email protected]. I2009 American Association for Cancer Research. tags that matched to a single gene cluster were taken into account. All doi:10.1158/0008-5472.CAN-08-3381 SAGE tags mapped to the same gene were then combined, and the sum of www.aacrjournals.org 1109 Cancer Res 2009; 69: (3).February 1, 2009

Cancer Research their counts in a given sample represented the expression level/signal of ELL or MLL-ENL fusions and a leukemic mouse cell line with an 2 that gene in that sample (16, 17). m test was used to identify differentially MLL-ELL fusion (Table 1). SAGE tags (484,303) were identified m2 expressed genes, as Man and colleagues (18) showed that test has the from the nine samples (total SAGE tag counts of each sample best power and robustness in SAGE data analysis. varies from 40,000 to 100,000, with 53,811 tags per sample on Functional classification and annotation of the candidate genes. We average), yielding 103,899 unique tags in human and 60,993 in used the Database for Annotation, Visualization, and Integrated Discovery (DAVID; version 2008)6 for functional classification and annotation of mouse samples (see Table 1). candidate genes. DAVID provides a comprehensive set of functional We identified 88 candidate genes that seemed to be deregulated annotation tools for investigators to understand the biological meaning in both human and murine MLL-ELL and/or MLL-ENL leukemias. behind large list of genes (19). Agene cluster means a set of genes belonging Of them, 6 and 26 genes are up-regulated and down-regulated, to the same functional category, as defined by DAVID. respectively, in both types of leukemia; 3 and 13 genes are up- Quantitative real-time PCR assays of protein coding or microRNA regulated and down-regulated, respectively, in MLL-ELL leukemias genes and data analysis. We performed quantitative real-time PCR (qPCR) alone; 23 and 17 genes are up-regulated and down-regulated, to determine expression of protein coding genes with SYBR green dye using respectively, in MLL-ENL leukemias alone (see Fig. 1A and kits from Qiagene and microRNAs (miRNA) using TaqMan qPCR kits from Supplementary Table S1). All the significant genes have at least Applied Biosystems. PGK1 or glyceraldehyde-3-phosphate dehydrogenase 3-fold difference in expression (tag counts per 100,000) between was used as endogenous controls for protein coding genes, whereas U6 RNA m2 was used as an endogenous control for miRNAs. PCR reactions and data each leukemia sample and the normal control and with a test analyses were performed, as described previously (20, 21). P value of <0.05. As this criterion is stringent, it therefore limits the Methylation-specific PCR. The methylation status of the promoter total number of the candidate genes identified. region was determined by the methylation-specific PCR (MSP) method (22). Notably, the four most up-regulated genes (each has at least a The primers were designed to anneal specifically to methylated and 30-fold change in expression) in both types of human and mouse unmethylated CpG dinucleotides in promoter regions of genes using the leukemia samples are three homeobox (HOX) genes (i.e., HOXA5, Primer3 program. Genomic DNAwas isolated using QIAampDNAmini kit HOXA9, and HOXA10) and one HOX cofactor, MEIS1 (see Fig. 1A) (Qiagen). One microgram of DNAwas used for bisulfite modification using The eight most down-regulated genes (each has at least a 30-fold the CpGenome DNAmodification kit (Chemicon) according to the change in expression on average) in both types of human and manufacturer’s instructions. The bisulfite-converted DNAwas amplified in mouse leukemia samples include LTF, LCN2, MMP9, S100A8, a total volume of 20 AL using GeneAmp Gold buffer containing 4 mmol/L A S100A9, PADI4, TGFBI, and CYBB (see Supplementary Table S1). MgCl2,0.5 mol/L of each primer, 0.2 mmol/L deoxynucleotide triphos- phate, 5 Ag bovine serum albumin, and 1.25 units of AmpliTaq Gold DNA Note that the fold change of some up-regulated or down-regulated polymerase (Roche). Hot start PCR was performed for 30 cycles, which genes may be overestimated because of the limited number of consists of denaturation at 95jC for 30 s, annealing at 60jC for 30 s, and SAGE tags examined (40,000–100,000 tags per sample). extension at 72jC for 45 s, followed by a final 7-min extension for all primer Although 56 genes were identified to be significantly deregulated sets. The products were separated on 10% polyacrylamide gels. in only one type of leukemias, many of them exhibited a similar Cell culture and 5-aza-2¶-deoxycytidine (Decitabine) treatment. deregulation pattern in both types of leukemias (see Supplemen- j Human leukemia cell lines were grown at 37 C under 5% CO2 in RPMI tary Table S1), albeit they were significant only in one according to 1640 (Invitrogen) supplemented with 10% fetal bovine serum (Invitrogen), our stringent criterion. If more samples were analyzed and/or a less 1% penicillin/streptomycin, and 1% HEPES. The cells were plated at 2 106 stringent criterion was used, many of these genes would seem to be per flask in 4 mL of medium. After 24 h, cells were exposed to 5-aza-2¶- deoxycytidine (5-Aza-CdR; Fluka) at 1 Amol/L for 96 h. In parallel, untreated deregulated in both MLL-ELL and MLL-ENL leukemias. cells were used as a control. After 48 h of continuous exposure to 5-Aza- Thirty-one of the 88 genes identified by SAGE have been CdR, the medium was changed. Cells were harvested after an additional previously reported to be deregulated in MLL rearrangement 48 h of incubation. Then, total RNAand genomic DNAwere isolated from leukemia. Much effort has been expended previously to identify the treated and untreated cell lines for further qPCR and MSP assay, candidate genes that might be involved in MLL rearrangement respectively. leukemia using different methods, such as microarray and qPCR. Luciferase reporter and mutagenesis assay. MiR-17 expression To obtain a comprehensive list of candidate genes involved in plasmid or its control plasmid (i.e., MSCVpuro) was cotransfected into MLL rearrangement leukemia, we searched 35 relevant publica- HEK293T cells with a single report plasmid (pMIR-Report plasmid; Ambion) tions (see Supplementary Table S2A) and collected a total of 1,647 containing either wild-type or mutated 3¶ untranslated region (UTR) of genes that were reported to be significantly deregulated in MLL an individual predicted target gene. Luciferase was measured 42 h after transfection. The firefly luciferase activity was then normalized to h- rearrangement leukemia. The top two most frequently reported galactosidase activity. Experiments were repeated thrice, independently. genes are MEIS1 and HOXA9, having been reported as up-regulated genes by 16 and 15 different publications, respectively. Among the Results 1,647 genes, 370 were reported by at least two different studies. Of them, only 10 genes (all were reported by only two publications) Eighty-eight genes were identified by SAGE analysis to be have contradictory information regarding whether they are up- significantly abnormally expressed in both human and murine regulated or down-regulated in MLL rearrangement leukemia. MLL-ELL MLL-ENL and/or leukemia. We used a modified SAGE The remaining 360 genes were reported consistently by various technique (13) to compare gene expression profiles between MLL- different studies, with 213 up-regulated and 147 down-regulated in ELL or MLL-ENL myeloid leukemia progenitor cells and normal MLL rearrangement leukemia compared with other subtypes of myeloid progenitor cells (CD15+ or Gr-1+) in both humans and acute leukemias and/or normal controls. We found that 31 of the mice. We analyzed four patient samples (two with each fusion) and 88 genes (35%) identified from our SAGE analysis were reported by two retrovirally induced mouse leukemias containing either MLL- previous studies (see Supplementary Table S2B). Validation of 81 genes by a large-scale qPCR assay. To validate the deregulation patterns of the genes identified from our 6 http://david.abcc.ncifcrf.gov SAGE analysis and those from the previous studies, we performed

Cancer Res 2009; 69: (3).February 1, 2009 1110 www.aacrjournals.org

Genes Deregulated in Both Human and Murine MLL Leukemia

Table 1. SAGE libraries of human and mouse samples

SAGE library ID Fusion Clone(s)/description Total tags Unique tags

Hs_N None Human normal control* 105,006 38,871 Hs_ELL-1 MLL-ELL 46, XX, t(11;19)(q23;p13.1); 88% 42,450 15,385 Hs_ELL-2 MLL-ELL 46, XY, t(11;19)(q23;p13.1); 100% 48,704 17,856 Hs_ENL-1 MLL-ENL 46,XY, t(11;19)(q23;p13.3); 95% 41,151 12,801 Hs_ENL-2 MLL-ENL 46, XY, t(11;19)(q23;p13.3); 97% 49,807 18,986 c Mm_N None Mouse normal control 70,140 22,035 b Mm_ELL-1 MLL-ELL MLL-ELL retroviral mouse 44,741 13,032 Mm_ELL-2 MLL-ELL MLL-ELL retroviral mouse cell linex 41,819 10,526 Mm_ENL-1 MLL-ENL MLL-ENL retroviral mousek 40,485 15,400

NOTE: CD15+ and Gr-1+ myeloid progenitor cells were isolated from bone marrow samples for human and mouse samples, respectively. Atotal of 484,303 SAGE tags were obtained for the nine samples, with a total of 103,899 unique SAGE tags in the human and 60,993 in the mouse samples. Abbreviations: Hs, Homo sapiens (human); Mm, Mus musculus (mouse). *Human normal CD15+ myeloid progenitor cells (for details, see ref. 14). cMouse normal Gr-1+ myeloid progenitor cells (for details, see ref. 15). bMouse bone marrow cells with retrovirus-induced MLL-ELL fusion (for details, see ref. 7). xMouse cell line was derived from a leukemia mouse transduced with an MLL-ELL fusion. kMouse bone marrow cells with retrovirus-induced MLL-ENL fusion (for details, see ref. 6).

a large-scale qPCR validation for 81 genes in 20 samples. The 81 biological importance of these 88 candidate genes, we analyzed the genes (see Supplementary Table S3) include 43 identified from our functional categories of known genes in terms of gene ontology SAGE analysis and 38 reported only by others. Among the 43 (GO) using the DAVID tools (19). One gene (i.e., MAGT1; down- genes identified by SAGE, 19 were also reported by others. The regulated in MLL-ENL leukemia) is not in the DAVID gene 20 samples include 12 human and 8 mouse samples. The human database; therefore, only 87 genes were analyzed. We first samples include three MLL-ELL and three MLL-ENL leukemia performed Gene Functional Classification for the 87 genes and samples, in addition to six human normal control samples. The found that there were four gene clusters (only the clusters having at mouse samples include two retrovirally induced mouse leukemia least three members were considered). Gene cluster 1 contains samples containing either MLL-ELL or MLL-ENL fusion, two S100A8, S100A9, and S100A11 (enrichment score, 1.98); all of them leukemia cell lines with MLL-ELL fusions, and one leukemia cell were down-regulated in leukemia. Gene cluster 2 contains CTSS, line with an MLL-ENL fusion, in addition to three murine normal CTSH, and MAN2B1 (enrichment score, 1.18); the first two were control samples. down-regulated, whereas the last one was up-regulated in We found that 60 of the 81 tested genes (74%) seemed to be leukemia. Gene cluster 3 contains ARHGEF1, CENTB2, and SMAP1L significantly deregulated in leukemia samples (see Fig. 1B). Of (enrichment score, 0.86); all of them were down-regulated in these, 50 candidate genes were deregulated in human and mouse leukemia. Remarkably, gene cluster 4 contains eight genes, MLL-ELL leukemia (29 up-regulated and 21 down-regulated) and including HOXA5, HOXA9, HOXA10, MEIS1, MYB, SOX4, KLF4, and 48 were deregulated in human and mouse MLL-ENL leukemia MLL5 (enrichment score, 0.75); the first six were up-regulated (32 up-regulated and 16 down-regulated), whereas 55 genes seemed whereas the last two were down-regulated in leukemias. to be significantly deregulated (34 up-regulated and 21 down- We also conducted a gene enrichment and functional annotation regulated) when combining the MLL-ELL and MLL-ENL leukemia analysis on the 87 genes with the functional annotation tools. As samples together. As shown in Supplementary Table S3, 36 of the shown in Supplementary Table S4, among the 288 GO terms 43 SAGE candidate genes (84%) were confirmed to be significantly analyzed, our SAGE-detected candidate genes are specially deregulated in leukemia samples in a similar mode (i.e., up- enriched in 40 GO terms compared with the whole set of genes regulated or down-regulated) to SAGE data, whereas only 31 of the (30, 000) in the human genome, with at least 2-fold higher 57 genes (54%) reported previously were confirmed here in a enrichment (P < 0.05) in each GO term, such as cell differentiation, similar mode to previous report(s). Notably, MYB and SMARCA4 cell development, defense response, apoptosis, and signal trans- were detected by SAGE as being up-regulated but were reported by duction. Remarkably, we found that our candidate genes enriched others as being down-regulated in MLL rearrangement leukemia; in some GO terms are either preferentially up-regulated (e.g., those the opposite is true for PTPRC, S100A8, ANXA1, and NCF1. Our in gene expression) or down-regulated (e.g., those in apoptosis or qPCR result of these six genes is consistent with our SAGE analysis signal transduction) in our leukemia samples. rather than previous reports (see Supplementary Table S3). In Therefore, we further compared the percentage of up-regulated addition, four genes reported only by others, including PROM1, genes with that of down-regulated genes enriched in each GO term. ADCY9, LST1,andTCF4, were detected by qPCR as being We found that, in 125 GO terms, the percentage of enriched up- deregulated in MLL rearrangement leukemia but in a different regulated gene in the whole 32 up-regulated genes is at least 2-fold mode from previous reports. higher or lower than that of enriched down-regulated genes in the Functional classification and annotation of the candidate whole 55 down-regulated genes (see Supplementary Table S5). genes identified by SAGE. To gain further insight into the As exemplified in Fig. 2, up-regulated genes have a much higher www.aacrjournals.org 1111 Cancer Res 2009; 69: (3).February 1, 2009

Cancer Research

Figure 1. Expression profiles of candidate genes. A, expression profiles of the 88 significant genes from nine human and mouse samples, as detected by SAGE. The genes have at least 3-fold difference in expression value (tag number per 100,000 of total SAGE tag in each library after normalization) between each leukemia sample and the normal control sample (human CD15+; mouse:Gr-1+) and with a m2 test P value of <0.05. See Supplementary Table S1 for details. B, expression profiles of the 60 significant genes from 20 human and mouse samples, as detected by qPCR. Because the number of mouse samples is too small, we combined human and mouse quantitative real-time PCR data together for significance analysis with SAM. Data are presented as DCT. Unsupervised, average linkage hierarchical clustering was performed with Pearson correlation as distance. All the significant genes have a q value of <0.05, with the overall FDR of <5%. Expression data were mean centered. _ELL, MLL-ELL; _ENL, MLL-ENL; _cl, cell line; _N, normal control; m or Mm., mouse; h or Hs., human; hm, human and mouse. In the columns of qPCR, ‘‘–’’ means no significant change. percentage of enrichment in GO terms, such as gene expression, response to stress, signal transduction, and cell communication transcription, and regulation of transcription compared with (see Supplementary Table s6 for the detailed gene lists). For down-regulated genes; opposite is true in GO terms, such as example, HOXA5, HOXA9, HOXA10, MEIS1, MYB, SMARCA2, apoptosis, cell death, extracellular space, immune response, SMARCA4, RPUSD1, CDK6, RPL7A, RRAGC,andSOX4 are

Cancer Res 2009; 69: (3).February 1, 2009 1112 www.aacrjournals.org

Genes Deregulated in Both Human and Murine MLL Leukemia

Figure 2. Example of 23 GO terms in which enriched up-regulated genes have an over 2-fold difference of enrichment percentage compared with enriched down-regulated genes. The percentages of the up-regulated and down-regulated genes relative to their total number (i.e., 32 and 55, respectively) in each GO term are shown in the plot for a direct comparison.

up-regulated genes that are involved in gene expression, whereas genes have putative miRNAregulation sites in their 3 ¶ UTRs MCL1, LITAF, STK17B, FCER1G, TNFRSF1B, PTPRC, MMP9, ANXA1, (see Supplementary Table S7). and CYFIP2 are down-regulated genes that are involved in cell A miRNA cluster is abnormally overexpressed in both death and apoptosis (see Supplementary Tables S5 and S6). human and murine leukemia samples with MLL-ELL or Down-regulation of some candidate genes is likely associated MLL-ENL fusion. We recently performed a large-scale, genome- with DNA methylation. As shown in Fig. 1A, the number of down- wide miRNAexpression profiling assay and observed that all the regulated genes is much greater than that of up-regulated genes seven individual miRNAs from a unique polycistronic miRNA (56 versus 32). The epigenetic silencing of tumor suppressor genes cluster, namely mir-17-92, residing in the C13orf25 gene locus at is a common event during carcinogenesis and often involves 13q31, including miR-17-5p, miR-17-3p, miR-18a, miR-19a, miR- aberrant DNAmethylation. To investigate whether DNAmethyla- 20a, miR-19b, and miR-92, were significantly up-regulated in MLL tion is involved in the down-regulation of candidate genes in the rearrangement leukemias relative to normal controls or other MLL rearrangement leukemias, we performed MSP assay (22) to leukemias (20). The mir-17-92 is a well-known oncogene (29). To detect the status of CpG islands of three significantly down- validate whether this miRNAcluster is also abnormally overex- regulated genes, including LTF, G0S2, and TGFBI. As shown in pressed in murine leukemia cells with MLL-ELL or MLL-ENL Fig. 3A, the CpG islands of all three tested genes are hyper- fusion, we performed a qPCR to evaluate the expression levels of methylated in three human ALL cell lines bearing MLL-ENL fusion all seven individual miRNAs in 33 human and 7 mouse samples. (i.e., KOPN1, KOCL33, and KOCL51). After the treatment of 5-Aza- The 33 human samples include 18 MLL rearrangement and CdR on these three cell lines, the CpG islands of the three genes 9 non–MLL rearrangement leukemia samples and 6 normal were partially demethylated (see Fig. 3B), and the expression of the controls. The seven mouse samples include two MLL-ELL and two three genes was dramatically up-regulated/restored with a >10-fold MLL-ENL leukemia samples and three Gr-1+ normal control change (Fig. 3C). Indeed, a similar up-regulation of these three samples. As shown in Fig. 4A, as in human samples, all seven genes was also observed in five other MLL rearrangement leukemia individual miRNAs are overexpressed in mouse leukemia cells, cells lines, including both ALL (i.e., KOCL45, KOCL69, and SEM) although the level is not as high as in corresponding human and AML (KOCL48 and MV4-11) cell lines, after the demethylating leukemias. treatment (see Fig. 3C). Therefore, the down-regulation of these Candidate targets of the miRNAs in the mir-17-92 cluster. It three candidate genes in MLL rearrangement leukemia cells is is believed that the central role of miRNAs is to regulate the likely associated with DNAhypermethylation. expression of their target genes (23–25); thus, it is important to Putative miRNA regulators of the 88 candidate genes identify their targets. Because miRNAs usually function as negative detected by SAGE. One of the most exciting findings in life regulators, it is expected that they will exhibit an inverse sciences in recent years is the discovery of an abundant class of correlation with the expression of their targets. Recent findings small (f22 nucleotides) noncoding RNAs, namely miRNAs. Recent indicate that animal miRNAs can not only repress protein synthesis studies showed that miRNAs play important regulatory roles in but also induce mRNAdegradation of a large portion of targets diverse biological processes and may function as oncogenes or (30, 31). To identify targets of the miRNAs in the mir-17-92 cluster, tumor suppressors (23–25). Using the four currently available we searched for candidate genes that were significantly down- major prediction programs, including TargetScanS (26), Miranda regulated in MLL rearrangement leukemias as being reported by at (27), PicTar (28), and MAMI,7 we found that all the 88 candidate least two different studies (including all the previous 35 studies and our SAGE analysis) and/or having been confirmed by our qPCR assay. We obtained a collection of 167 such candidate genes and 7 http://mami.med.harvard.edu found that 19 of them were predicted as putative targets of the www.aacrjournals.org 1113 Cancer Res 2009; 69: (3).February 1, 2009

Cancer Research

Discussion The power of comparative genomic analysis relies on the assumption that important biological properties are often conserved across species (9). Cross-species sequence comparison has been widely used to infer gene function, but it is becoming apparent that sequence similarity is not always proportional to functional similarity (32, 33). To determine the function of a gene precisely, therefore, we need to investigate not only its sequence characteristics but also its expression characteristics (9). The expression pattern of a gene can thus serve as a sensitive indicator of its function. In the present work, we conducted a genome-wide gene expression profiling assay on both human and mouse leukemia cells bearing MLL fusions using the SAGE method. We obtained 484,303 total SAGE tags for the nine samples and a total of 103,899 unique SAGE tags in the human and 60,993 in the mouse samples. We identified 88 genes that seemed to be significantly deregulated (32 up-regulated and 56 down-regulated) in both human and murine MLL-ELL and/or MLL-ENL leukemia (see Fig. 1A). In a comparison with previous studies of human MLL rearrangement leukemias that used different methods, such as cDNAmicroarray, oligonucleotide microarray, and qPCR, we found that 31 of the 88 genes identified by SAGE have been reported previously but 57 are previously undescribed candidate genes that seem to be deregulated in MLL rearrangement leukemias. We further performed a large-scale qPCR assay to validate the deregulation of 81 candidate genes, including 43 genes identified from our SAGE analysis and 38 genes reported only by the previous studies in 20 samples, including 12 human and 8 mouse samples. Eighty-four percent (36 of 43) of the candidate genes detected by SAGE were confirmed, as were 54% (31 of 57) of the genes identified by previous studies. Alower confirmation/validation rate of the genes identified by previous studies relative to genes identified by our SAGE analysis is not surprising, because most of the previous studies compared MLL rearrangement leukemias with other acute leukemias, not with normal controls, and our qPCR Figure 3. DNA methylation analysis of three down-regulated candidate genes (i.e., LTF, TGFBI, and G0S2)inMLL rearrangement leukemias. A, MSP focused on MLL-ELL and MLL-ENL samples that might not be assay of CpG islands on the promoter regions of LTF, TGFBI, and G0S2. included in a large number of previous studies. The fact that only Lanes U, unmethylated products; lanes M, methylated products. MNC-1 and 31 of the 88 genes detected by SAGE were reported previously and MNC-2 are two normal controls (bone marrow mononuclear cells). Pos., positive control; Neg., negative control. B, the CpG islands of LTF, TGFBI, and G0S2 84% of the tested SAGE candidate genes were confirmed by qPCR were partially demethylated after the treatment with 5-Aza-CdR. KOPN1 indicates that, although MLL rearrangement leukemias have been is the only example shown here. C, effects of 5-Aza-CdR on the expression of LTF, TGFBI, and G0S2, as detected by qPCR. KOPN1, KOCL33, and intensively studied previously, there are still many potentially KOCL51 are ALL cell lines bearing MLL-ENL fusions; KOCL45, KOCL69, and important genes that have not been identified; thus, our study, SEM are ALL cell lines bearing MLL-AF4 fusions; KOCL48 and MV4-11 are using a different strategy/method, could provide additional AML cell lines bearing MLL-AF4 fusions. important information to extend our understanding of the complex genetic alterations in MLL fusion-induced leukemogenesis. miRNAs in the mir-17-92 cluster by at least two of the four Remarkably, HOXA5, HOXA9, HOXA10, and the HOX cofactor prediction programs described above (see Supplementary MEIS1 are the four most up-regulated genes observed in both types Table S8). Of these, three genes (i.e., APP, RASSF2, and SH3BP5) (i.e., MLL-ELL and MLL-ENL) of human and murine leukemia have already been confirmed by our qPCR assay to be significantly samples (Fig. 1A), which were further confirmed by our large-scale down-regulated in both MLL-ELL and MLL-ENL samples (see qPCR assay (Fig. 1B). HOX genes and MEIS1 are the best studied Fig. 1B). downstream targets of MLL and MLL fusion proteins, and their We further performed a luciferase reporter and mutagenesis aberrant overexpression has been consistently recognized as the assay to validate APP and RASSF2. Asignificantly negative effect hallmark of MLL rearrangement leukemias (34–41). Interestingly, in (P < 0.01; paired t test) on luciferase activity was observed in the a gene functional classification analysis using the DAVID tools (19), presence of miR-17 on 3¶ UTR of APP or RASSF2, and such we found that HOXA5, HOXA9, HOXA10, and MEIS1 were clustered repression was lost when the predicted miRNAbinding site in the with four other candidate genes, including MYB, SOX4, KLF4, and 3¶ UTR of APP or RASSF2 (in the reporter plasmid) was mutated MLL5; the first six were up-regulated whereas the last two were (Fig. 4B), indicating that both APP and RASSF2 are direct targets down-regulated in MLL-associated leukemias. Notably, all of these of miR-17. eight genes are regulators of gene expression and/or cell cycle; the

Cancer Res 2009; 69: (3).February 1, 2009 1114 www.aacrjournals.org

Genes Deregulated in Both Human and Murine MLL Leukemia first six genes (41–44) are positive regulators, whereas KLF4 (45) group proteins, which act as negative regulators of transcription and MLL5 (46) are negative regulators. Thus, their up-regulation or (47). Thus, it is important to conduct systematic studies in the down-regulation would promote gene expression and cell prolif- future to determine whether down-regulation of many genes in eration, which in turn would contribute to the development of MLL-associated leukemias is related to the deregulation of BMI1 leukemia. and/or other transcription repressor(s). LTF, LCN2, MMP9, S100A8, S100A9, PADI4, TGFBI, and CYBB are We found that all of the 88 candidate genes have putative miRNA the eight most down-regulated genes (each has at least 30-fold regulators (see Supplementary Table S7). Interestingly, 19 putative change in expression) in both types of human and murine leukemia targets of mir-17-92 seemed to be down-regulated in MLL samples (Fig. 1A) and were also confirmed by our qPCR assay rearrangement leukemias, as reported previously or detected by (Fig. 1B). Interestingly, five (i.e., LTF, MMP9, S100A8, S100A9, and our SAGE assay; the down-regulation of APP, RASSF2, and SH3BP5 TGFBI) of these eight genes are enriched in a GO term, namely and the up-regulation of the mir-17-92 cluster have been confirmed extracellular space, and the other two genes (i.e., TNFAIP2 and LYZ) by our qPCR. We have confirmed both APP and RASSF2 as direct enriched in this term are also down-regulated in MLL-associated targets of miR-17 through luciferase reporter and mutagenesis leukemias. In addition, LTF, S100A8, S100A9, and CYBB are involved assay (see Fig. 4B). Interestingly, RASSF2 encodes a protein that in both response to stimulus and defense response, along with contains a Ras association domain and is a novel tumor-suppressor TGFBI involved in the former (see Supplementary Tables S5 and S6). gene, which has been observed to regulate Ras signaling and play a As shown in Fig. 2, up-regulated genes have a much higher pivotal role in the early stages of colorectal tumorigenesis (48). percentage of enrichment in GO terms related to gene expression Therefore, miRNAs may play an important role in the development and transcription, whereas down-regulated genes are more of MLL rearrangement leukemia. In addition, recent studies also enriched in GO terms related to apoptosis, signal transduction, showed that some miRNAs, e.g., the miR-290 cluster, may control and response. Thus, the up-regulation of genes responsible for gene de novo DNAmethylation through regulating Rbl2-dependent expression and transcription but the down-regulation of genes Dnmt expression (49, 50). Thus, besides the direct regulation of the responsible for apoptosis, signal transduction, and response can expression of target genes, some miRNAs may also contribute to promote cell proliferation and inhibit apoptosis and, thereby, the DNAmethylation of many genes and, thereby, regulate contribute to the development of leukemia. expression of these genes indirectly. Therefore, a combined study It is well known that MLL fusion proteins can promote of miRNAand mRNAexpression profiles would provide a more expression of many target genes, such as HOX genes and MEIS1 complete understanding of the complexity of molecular networks (3, 4). However, the mechanism underlying the down-regulation of in diseases, such as MLL rearrangement leukemia. a large number of genes in MLL rearrangement leukemias is In all, our study showed that the deregulation patterns of many unclear. In the present study, we showed that down-regulation of protein-coding (e.g., HOX genes and MEIS1) and noncoding many genes in the leukemias might be associated with DNA (e.g., individual miRNAs within the mir-17-92 cluster) genes observed methylation (see Fig. 3). Nonetheless, the genetic or epigenetic in human leukemia cells are also conserved in mouse leukemia cells, changes that caused the DNAmethylation remain to be explored. highlighting the importance of these genes in leukemogenesis. Interestingly, we found that BMI1 was consistently overexpressed The identification and validation of consistent changes of gene in human and murine MLL-associated leukemia cells relative to the expression in human and murine MLL rearrangement leukemias normal controls (see Fig. 1B). BMI1 is a member of polycomb contribute important information leading to the understanding

Figure 4. Expression profiles and targets of the miRNAs in the mir-17-92 cluster. A, expression profiles of the seven miRNAs in 40 human and mouse samples, as detected by TaqMan qPCR. All the significant genes have a q value of <0.05, with an overall FDR of <0.05. _cl, cell line; _N, normal control; m, mouse. B, regulation of APP and RASSF2 by miR-17 was confirmed by luciferase reporter and mutagenesis assay. HEK293T cells were cotransfected with an expression construct for miR-17 or an empty vector (control; i.e., MSCVpuro); the luciferase reporter construct containing the wild-type or mutated 3¶ UTR of APP or RASSF2, along with a h-galactosidase reporter control vector to monitor transfection efficiency. Luciferase was measured 42 h after transfection. The firefly luciferase activity was normalized to h-galactosidase activity. The normalized luciferase activities represent the firefly/h-galactosidase ratios normalized to the control sample. Error bars present SD obtained from three independent experiments. *, P < 0.01 (paired t test).

www.aacrjournals.org 1115 Cancer Res 2009; 69: (3).February 1, 2009

Cancer Research of the complex functional pathways underlying MLL-associated Grant support: NIH grant CA127277 (J. Chen), Cancer Research Foundation Young Investigator Award (J. Chen), G. Harold and Leila Y. Mathers Charitable Foundation leukemia, which in turn may lead to more effective therapy for this (J. Chen), Leukemia and Lymphoma Society Translational Research Grant (J.D. Rowley), type of leukemia. Spastic Paralysis Foundation of Illinois, Eastern Iowa Branch of Kiwanis International (J.D. Rowley), NIH grants RO1CA104449 and P01CA105049 (M.J. Thirman), Leukemia and Lymphoma Society Scholar Award (M.J. Thirman), Leukemia and Lymphoma Disclosure of Potential Conflicts of Interest Society Specialized Center of Research (M.J. Thirman), family of Robert A. Chapski (M.J. Thirman), and NIH grant HG002600 (S.M. Wang). No potential conflicts of interest were disclosed. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance Acknowledgments with 18 U.S.C. Section 1734 solely to indicate this fact. We thank Drs. Y. Sato and K. Sugita for kindly providing KOPN1, KOCL33, KOCL45, Received 8/29/2008; revised 10/22/2008; accepted 11/11/2008. KOCL48, KOCL51, and KOCL69 cell lines.

References 17. Chen J, Sun M, Hurst LD, Carmichael GG, Rowley JD. of function in myeloid malignancies. Blood 2004;103: Genome-wide analysis of coordinate expression and 229–35. 1. Tkachuk DC, Kohler S, Cleary ML. Involvement of a evolution of human cis-encoded sense-antisense tran- 36. Ross ME, Mahfouz R, Onciu M, et al. Gene expression homologue of Drosophila trithorax by 11q23 chromo- scripts. Trends Genet 2005;21:326–9. profiling of pediatric acute myelogenous leukemia. somal translocations in acute leukemias. Cell 1992;71: 18. Man MZ, Wang X, Wang Y. POWER_SAGE: compar- Blood 2004;104:3679–87. 691–700. ing statistical tests for SAGE experiments. Bioinfor- 37. Krivtsov AV, Twomey D, Feng Z, et al. Transforma- 2. Thirman MJ, Gill HJ, Burnett RC, et al. Rearrangement matics 2000;16:953–9. tion from committed progenitor to leukaemia stem cell of the MLL gene in acute lymphoblastic and acute 19. Huang DW, Sherman BT, Tan Q, et al. DAVID initiated by MLL-AF9. Nature 2006;442:818–22. myeloid leukemias with 11q23 chromosomal trans- Bioinformatics Resources: expanded annotation data- 38. Kohlmann A, Schoch C, Dugas M, et al. New locations. N Engl J Med 1993;329:909–14. base and novel algorithms to better extract biology from insights into MLL gene rearranged acute leukemias 3. Popovic R, Zeleznik-Le NJ. MLL: how complex does it large gene lists. Nucleic Acids Res 2007;35:W169–75. using gene expression profiling: shared pathways, get? J Cell Biochem 2005;95:234–42. 20. Li Z, Lu J, Sun M, et al. Distinct microRNAexpression lineage commitment, and partner genes. Leukemia 4. Ayton PM, Cleary ML. Molecular mechanisms of profiles in acute myeloid leukemia with common 2005;19:953–64. leukemogenesis mediated by MLL fusion proteins. translocations. Proc Natl Acad Sci U S A 2008;105: 39. Ferrando AA, Armstrong SA, Neuberg DS, et al. Gene Oncogene 2001;20:5695–707. 15535–40. expression signatures in MLL-rearranged T-lineage and 5. Thirman MJ, Levitan DA, Kobayashi H, Simon MC, 21. Mi S, Lu J, Sun M, et al. MicroRNAexpression B-precursor acute leukemias: dominance of HOX Rowley JD. Cloning of ELL, a gene that fuses to MLL in a signatures accurately discriminate acute lymphoblastic dysregulation. Blood 2003;102:262–8. t(11;19)(q23; p13.1) in acute myeloid leukemia. Proc Natl leukemia from acute myeloid leukemia. Proc Natl Acad 40. Chiaretti S, Li X, Gentleman R, et al. Gene expression Acad Sci U S A 1994;91:12110–4. Sci U S A2007;104:19971–76. profiles of B-lineage adult acute lymphocytic leukemia 6. Lavau C, Szilvassy SJ, Slany R, Cleary ML. Immortal- 22. Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin reveal genetic patterns that identify lineage derivation ization and leukemic transformation of a myelomono- SB. Methylation-specific PCR: a novel PCR assay for and distinct mechanisms of transformation. Clin Cancer cytic precursor by retrovirally transduced HRX-ENL. methylation status of CpG islands. Proc Natl Acad Sci Res 2005;11:7209–19. EMBO J 1997;16:4226–37. U S A1996;93:9821–6. 41. Drabkin HA, Parsy C, Ferguson K, et al. Quantitative 7. Lavau C, Luo RT, Du C, Thirman MJ. Retrovirus- 23. He L, Hannon GJ. MicroRNAs: small RNAs with a big HOX expression in chromosomally defined subsets of mediated gene transfer of MLL-ELL transforms primary role in gene regulation. Nat Rev Genet 2004;5:522–31. acute myelogenous leukemia. Leukemia 2002;16:186–95. myeloid progenitors and causes acute myeloid leuke- 24. Esquela-Kerscher A, Slack FJ. Oncomirs—microRNAs 42. Kawagoe H, Humphries RK, Blair A, Sutherland HJ, mias in mice. Proc Natl Acad Sci U S A 2000;97:10984–9. with a role in cancer. Nat Rev Cancer 2006;6:259–69. Hogge DE. Expression of HOX genes, HOX cofactors, 8. Hu Y, Sun H, Drake J, et al. From mice to humans: 25. Wu W, Sun M, Zou GM, Chen J. MicroRNAand and MLL in phenotypically and functionally defined identification of commonly deregulated genes in cancer: current status and prospective. Int J Cancer subpopulations of leukemic and normal human hema- mammary cancer via comparative SAGE studies. Cancer 2007;120:953–60. topoietic cells. Leukemia 1999;13:687–98. Res 2004;64:7748–55. 26. Lewis BP, Burge CB, Bartel DP. Conserved seed 43. Liu P, Ramachandran S, Ali Seyed M, et al. Sex- 9. Zhou XJ, Gibson G. Cross-species comparison of genome- pairing, often flanked by adenosines, indicates that determining region Y box 4 is a transforming oncogene wide expression patterns. Genome Biol 2004;5:232. thousands of human genes are microRNAtargets. Cell in human prostate cancer cells. Cancer Res 2006;66: 10. Huang H, Winter EE, Wang H, et al. Evolutionary 2005;120:15–20. 4011–9. conservation and selection of human disease gene 27. Krek A, Grun D, Poy MN, et al. Combinatorial 44. Lang G, White JR, Argent-Katwala MJ, Allinson CG, orthologs in the rat and mouse genomes. Genome Biol microRNAtarget predictions. Nat Genet 2005;37:495–500. Weston K. Myb proteins regulate the expression of 2004;5:R47. 28. John B, Enright AJ, Aravin A, Tuschl T, Sander C, diverse target genes. Oncogene 2005;24:1375–84. 11. Corral J, Lavenir I, Impey H, et al. An Mll-AF9 fusion Marks DS. Human MicroRNAtargets. PLoS Biol 2004;2: 45. Shields JM, Christy RJ, Yang VW. Identification and gene made by homologous recombination causes acute e363. characterization of a gene encoding a gut-enriched leukemia in chimeric mice: a method to create fusion 29. He L, Thomson JM, Hemann MT, et al. AmicroRNA Kruppel-like factor expressed during growth arrest. oncogenes. Cell 1996;85:853–61. polycistron as a potential human oncogene. Nature J Biol Chem 1996;271:20009–17. 12. Lavau C, Du C, Thirman M, Zeleznik-Le N. Chroma- 2005;435:828–33. 46. Deng LW, Chiu I, Strominger JL. MLL 5 protein forms tin-related properties of CBP fused to MLL generate a 30. Bagga S, Bracht J, Hunter S, et al. Regulation by let-7 intranuclear foci, and overexpression inhibits cell cycle myelodysplastic-like syndrome that evolves into mye- and lin-4 miRNAs results in target mRNA degradation. progression. Proc Natl Acad Sci U S A 2004;101:757–62. loid leukemia. EMBO J 2000;19:4655–64. Cell 2005;122:553–63. 47. Rice KL, Hormaeche I, Licht JD. Epigenetic regulation 13. Lee S, Chen J, Zhou G, Wang SM. Generation of high- 31. Nilsen TW. Mechanisms of microRNA-mediated gene of normal and malignant hematopoiesis. Oncogene quantity and quality tag/ditag cDNAs for SAGE analysis. regulation in animal cells. Trends Genet 2007;23:243–9. 2007;26:6697–714. Biotechniques 2001;31:348–50, 52–4. 32. Gerlt JA, Babbitt PC. Can sequence determine 48. Akino K, Toyota M, Suzuki H, et al. The Ras effector 14. Lee S, Zhou G, Clark T, Chen J, Rowley JD, Wang SM. function? Genome Biol 2000;1:REVIEWS0005. RASSF2 is a novel tumor-suppressor gene in human The pattern of gene expression in human CD15+ 33. Wilson CA, Kreychman J, Gerstein M. Assessing colorectal cancer. Gastroenterology 2005;129:156–69. myeloid progenitor cells. Proc Natl Acad Sci U S A annotation transfer for genomics: quantifying the 49. Sinkkonen L, Hugenschmidt T, Berninger P, et al. 2001;98:3340–5. relations between protein sequence, structure and MicroRNAs control de novo DNAmethylation through 15. Chen J, Rowley DA, Clark T, et al. The pattern of gene function through traditional and probabilistic scores. regulation of transcriptional repressors in mouse expression in mouse Gr-1(+) myeloid progenitor cells. J Mol Biol 2000;297:233–49. embryonic stem cells. Nat Struct Mol Biol 2008;15: Genomics 2001;77:149–62. 34. Rozovskaia T, Feinstein E, Mor O, et al. Upregulation 259–67. 16. Sun M, Hurst LD, Carmichael GG, Chen J. Evidence of Meis1 and HoxA9 in acute lymphocytic leukemias 50. Benetti R, Gonzalo S, Jaco I, et al. Amammalian for variation in abundance of antisense transcripts with the t(4:11) abnormality. Oncogene 2001;20:874–8. microRNAcluster controls DNAmethylation and between multicellular animals but no relationship 35. Poppe B, Vandesompele J, Schoch C, et al. Expression telomere recombination via Rbl2-dependent regulation between antisense transcriptionand organismic com- analyses identify MLL as a prominent target of 11q23 of DNAmethyltransferases. Nat Struct Mol Biol 2008;15: plexity. Genome Res 2006;16:922–33. amplification and support an etiologic role for MLL gain 268–79.

Cancer Res 2009; 69: (3).February 1, 2009 1116 www.aacrjournals.org

Consistent Deregulation of Gene Expression between Human and Murine MLL Rearrangement Leukemias

Zejuan Li, Roger T. Luo, Shuangli Mi, et al.

Cancer Res 2009;69:1109-1116. Published OnlineFirst January 20, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-08-3381

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2009/01/19/0008-5472.CAN-08-3381.DC1

Cited articles This article cites 50 articles, 17 of which you can access for free at: http://cancerres.aacrjournals.org/content/69/3/1109.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/69/3/1109.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/69/3/1109. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.