Minichromosome Maintenance Protein MCM7 Is a Direct Target of the MYCN Transcription Factor in Neuroblastoma1

[CANCER RESEARCH 62, 1123–1128, February 15, 2002] Minichromosome Maintenance Protein MCM7 Is a Direct Target of the MYCN Transcription Factor in Neuroblastoma1

Jason M. Shohet, M. John Hicks, Sharon E. Plon, Susan M. Burlingame, Susan Stuart, Si-Yi Chen, Malcolm K. Brenner, and Jed G. Nuchtern2 Departments of Pediatrics [J. M. S., S. E. P., M. K. B.], Pathology [M. J. H.], Molecular and Human Genetics [S-Y. C.], and The DeBakey Department of Surgery [S. M. B., J. G. N.], Baylor College of Medicine, Houston, Texas 77030; The Center for Cell and Gene Therapy, Houston, Texas 77030 [J. M. S., S-Y. C., M. K. B., J. G. N.]; Texas Children’s Cancer Center, Texas Children’s Hospital, Houston, Texas 77030 [S. S.]; and Incyte Genomics, Palo Alto, California 94304 [S. S.]

ABSTRACT the E2F sites, the MCM7 promoter has an E-box binding site for the MYC oncoproteins (10). On the basis of initial observations demon- ϳ The MYCN oncogene is amplified in 25% of neuroblastoma tumors and strating differential expression of MCM7 between MYCN-amplified is the most significant negative prognostic factor. The direct transcriptional and nonamplified neuroblastoma tumor cell cultures (see below), we targets of MYCN in MYCN-amplified tumors have not been defined. Mi- croarray analysis of RNA from neuroblastoma primary cell cultures revealed pursued the hypothesis that MCM7 may be a direct target of MYCN. 10-fold higher MCM7 expression in MYCN-amplified versus nonamplified MYCN, a close homologue of MYCC, functions as a transactivator tumors. MCM7 is an essential component of DNA replication licensing factor, by binding to a core E-box promoter element CAC/TGTG after a hexameric protein complex that regulates DNA synthesis during the cell forming a heterodimer with MAX. Although MYCC, MYCN, and cycle, preventing rereplication and ensuring maintenance of DNA euploidy. MYCL can all bind to the canonical core element, promoter-flanking Additional experiments demonstrated markedly increased expression of sequences can markedly affect binding affinities of the different MYC MCM7 RNA and protein in MYCN-amplified neuroblastoma tumors and cell isoforms to individual target genes (11). In addition, other regulatory lines. Induction of MYCN in conditional cell lines results in increased expres- molecules that influence both the activity of MYC and the interaction sion of endogenous MCM7 mRNA and a 3-fold increase in protein levels. In of MYC with MAX contribute to the tissue specificity of MYCN and addition, luciferase activity from MCM7 promoter/luciferase gene reporter MYCC oncogene transactivation (12, 13). constructs was significantly increased under MYCN-induced conditions. Spe- ϳ cific electrophoretic mobility shifts of MCM7 promoter sequences are de- The MYCN proto-oncogene is amplified in 25% of neuroblasto- tected in extracts of MYCN-amplified cells. These findings demonstrate that mas. It is also frequently amplified in retinoblastomas, astrocytomas, in neuroblastoma, the MYCN oncogene directly activates genes required for gliomas, and small cell lung cancers. Amplification is the most DNA replication, and this may contribute to neoplastic transformation of reliable negative prognostic factor in neuroblastoma (14). Three-year these MYCN-amplified tumors. survival for patients with metastatic neuroblastoma decreases from ϳ40% to Ͻ10% with MYCN amplification (15). Despite intensive efforts, MYCN transcriptional targets responsible for the particularly INTRODUCTION malignant phenotype of these tumors have remained elusive (16). This report characterizes the direct interaction of MYCN with the The MCM3 molecules are a highly conserved group of DNA- MCM7 E-box, and consequent increased MCM7 mRNA and protein binding proteins with a vital function of “licensing” DNA synthesis expression in MYCN-amplified cell lines and tumors relative to their during the transition from G to S phase of the cell cycle (1). 1 nonamplified counterparts. We demonstrate consistent MYCN- Originally characterized in yeast, this group of proteins is required for dependent differential expression of MCM7 by direct immunohistochem- activation of autonomous replicating sequences and progression istry in primary tumor samples and by conditional expression of MYCN through the cell division cycle. In eukaryotes, MCM2 to MCM7 are in a neuroblastoma cell line. Transcriptional regulation of MCM7 by sequentially assembled into a heteromeric hexamer, the replication MYCN may have important implications for the pathogenesis, progres- licensing factor, that binds to DNA replication origins after the origin sion, and treatment of neuroblastoma and other MYCN amplified tumors. recognition complex has assembled at the end of G1 (2). (For detailed reviews of MCM structure and function see Refs. 3–5.) MCM-mediated regulation of DNA synthesis ensures that DNA replicates only MATERIALS AND METHODS once during each cell cycle and is essential for maintaining euploidy. Immunohistochemical studies in a variety of tissues demonstrate the Tissues and Cell Lines. The neuroblastoma primary tumor cultures (P4, increased expression of MCM2, 5, and 7 in solid tumors and prema- P46, P67, and P102) were established as described previously (17) and cul- lignant proliferative states (6–8). Their vital role in the maintenance tured in MEM with 10% FCS and antibiotics. The MYCN-inducible neuro- of chromosomal integrity in normal cells makes the MCM molecules blastoma cell line TET-21 derived from the SH-EP line was a kind gift from potential targets of the transforming effects of cellular oncogenes. Dr. Manfred Schwab. These cells were grown as described previously in 10% The promoter regions of MCM5, 6, and 7 each contain numerous serum/RPMI 1640 with penicillin and streptomycin (18). Tetracycline was used at 1 ␮g/␮l concentration. IMR-32 and SK-N-SH cell lines were obtained E2F transactivation sites, suggesting that the E2F transcription factor from the American Type Culture Collection and grown as above. The Texas may be primarily responsible for the coordinated increase in MCM Children’s Hospital Department of Pathology supplied neuroblastoma tumor mRNA noted at the G1-S boundary (9, 10). However, in addition to tissue for immunohistochemistry according to Institutional Review Board- approved protocols. This tissue was drawn from archived specimens that Received 6/4/01; accepted 12/17/01. remained after diagnostic studies and would have otherwise been discarded. The costs of publication of this article were defrayed in part by the payment of page Microarray Probing. Total RNA was extracted from exponentially growing charges. This article must therefore be hereby marked advertisement in accordance with cultures using the RNeasy Mini kit (Qiagen) according to the instructions of the 18 U.S.C. Section 1734 solely to indicate this fact. ϩ 1 Supported in part by USPHS Grant CA 78456 and an American College of Surgeons manufacturer. Poly(A) RNA was purified using the Oligotex mRNA Mini kit Faculty Research Fellowship award (to J. G. N.). (Qiagen). Each oligo-dT-selected mRNA (200 ng) was reverse transcribed with 2 To whom requests for reprints should be addressed, at Pediatric Surgery, 6621 Moloney murine leukemia virus reverse transcriptase enzyme (Promega). Random Fannin, CC 650.00, Houston, TX 77030. Phone: (832) 822-3135; Fax: (832) 825-3141. nonomers with fluorescent tags (CY3 or CY5) were used to prime each reaction 3 The abbreviations used are: MCM, minichromosome maintenance; INSS, Interna- tional Neuroblastoma Staging System; EMSA, electrophoretic mobility shift analysis; and were carried out with standard dideoxy nucleotides. After transcription, the DOTAP, N-[1-(2,3-dioleoyloxyl)propyl]-N,N,N-trimethylammoniummethyl sulfate. CY3 or CY5 samples were combined with prepared controls, purified over a 1123

TE-30 spin column (Clontech), and concentrated. UniGEM V microarrays (Incyte presence of tetracycline (MYCN noninduced) using DOTAP (gift from Dr. Genomics) were hybridized with two different fluorescent-labeled probes (CY3 Nancy Templeton). Endotoxin-free plasmid preps of each construct were and CY5). The probes were simultaneously applied to the microarray for compet- prepared and mixed with DOTAP to produce lipid/DNA complexes. For each itive hybridization. After scanning and image analysis, a ratio of the two fluores- experiment, the wells of a six-well plate with 70% confluent TET-21 cells were cent intensities provided a quantitative measurement of relative gene expression transfected with DOTAP/plasmid complexes prepared as described (20). To levels. Only genes that showed significant expression in at least one channel were ensure equivalent transfection conditions between MYCN-induced and non- considered for calculation of differential expression. No differential expression induced cells, sufficient DOTAP/plasmid complexes were prepared to transfect ratio was calculated for any element of which the signal did not measure Ն2.5 the entire plate (ϳ10 ␮g DNA/well) and were added to each well under times the background signal in at least one channel. For our analyses, we used a identical conditions and incubated with serum-free medium for 3 h. Standard background subtraction level of 200 relative fluorescent units. Identity of spotted medium containing tetracycline and serum was then added, and the cells were cDNA was verified by PCR. A failure in PCR testing resulted in exclusion of the grown for 36 h. Medium was exchanged, and tetracycline-free medium was spot measurement. A total of 6800 gene products (cDNAs) passed these quality added to half the wells to induce MYCN expression. Cells were lysed in control criteria. Of these, 370 showed Ͼ3.5-fold differential expression for one of reporter lysis buffer and luciferase activity measured after 10 h of induction per the tumor cell cultures. protocol (Promega luciferase assay kit). In pilot experiments cotransfection cDNA Northerns. cDNA Northerns were used in this study to maximize with a ␤-galactoside plasmid was used to confirm that transfection efficiency detection sensitivity for MCM7 and other differentially expressed messages. This was uniform between wells and that reporter gene activity was proportional to method has been demonstrated to accurately reflect expression levels and to total protein in the cell lysates (␤-galactoside activity did not change on correlate well with standard Northern blots (19). Total RNA from the induced and removal of tetracycline; data not shown). Subsequently, luciferase activity was noninduced TET-21 cell line was isolated with the Qiagen RNeasy kit and normalized to total protein in each well (luciferase activity/␮g protein). reverse-transcribed using Superscript-II reverse transcriptase (Life Technologies, Immunohistochemistry. MCM7 immunohistochemistry was performed Inc.) with the CDS and Smart-II primer oligonucleotides per the Clontech SMART on formalin-fixed deparaffinized tumor sections with a mouse monoclonal cDNA synthesis kit. The resulting cDNA was amplified for 15 cycles. The number anti-MCM7 antibody at 1:200 dilution (Santa Cruz Biotechnology). Prepared of amplification cycles was optimized with electrophoretic analysis of the cDNA sections were washed in Tris-EDTA, and endogenous peroxidase was blocked on 1.4% agarose gels to prevent over-cycling per protocol. Next, 1.5 ␮gof by standard protocols. Biotinylated primary antibody was applied for 25 min MYCN-induced and noninduced cDNA was electrophoresed and electroblotted at room temperature. Sections were incubated with horseradish peroxidase/ onto nylon membranes. Probes for MYCN and MCM7 were generated from PCR streptavidin (BioGenex) for 10 min at room temperature per standard methods. amplified fragments (MYCN:5Ј-CCTGCCCGCCGAGCTCG-3Ј and reverse 5Ј- Counterstaining was done with hematoxylin and samples mounted with aque- CTCGCTGGACTGAGCCCA-3Ј, MCM7:5Ј-AGCAGAACATACAGCTAC- ous mounting medium (Dako Faramount). Negative (withholding primary CTG-3Ј and reverse 5Ј-CCCTTGTCTCCTAGAAGAGAG-3Ј) and either random antibody) and positive controls (antibody specific for neuron-specific enolase) hexamer labeled with [␣-32P]CTP or labeled with alkaline phosphatase (Amer- were performed with each tissue specimen. No significant staining occurred sham AlkphosDirect). Probes were hybridized at 42°C in Ultrahyb (Ambion) without primary antibody. All of the cells with neuronal/neuroblast-like mor- hybridization buffer or alkaline phosphatase hybridization buffer (Alkphos Direct phology stained positive with the neuron-specific enolase antibody. kit, Amersham) overnight and washed. Probe signals were detected using a Molecular Dynamics Phosphoimager. Expression levels were normalized against RESULTS the signal from ␤-actin. Electrophoretic Mobility Shift Assays. Nuclear extracts were prepared as MCM7 Expression Correlates with MYCN Amplification in follows from logarithmically growing cell cultures. Cell pellets were solubi- Neuroblastoma Tumors. We performed a cDNA microarray-based lized in lysis buffer without NaCl [20 mM HEPES (pH 7.4), 3 mM MgCl2, 0.2 analysis comparing mRNA expression between tumor explants from mM EGTA, 10% glycerol, and protease inhibitors (EDTA free Complete; two INSS stage 4 neuroblastoma patients, P4 and P67. In the MYCN- Roche)] on ice then centrifuged 15 min at 4°C. The resulting pellet was washed amplified (P4) cells, MCM7 message was increased 10.4-fold over the twice in lysis buffer and resuspended in lysis buffer plus 400 mM NaCl. Protein level in the nonamplified (P67) tumor. Table 1 lists some of the genes content was measured by the Bradford method (Bio-Rad) and salt concentration adjusted to 150 mM for binding assays. Oligonucleotide EMSA probes associated with MYCN overexpression. were generated by annealing commercially prepared sense and complement Differential Expression Is the Ratio of the Greater to the Lesser sequences followed by end-labeling with [␥-32P]ATP and T4 polynucleotide Signal after Correction for Background. The observed relationship kinase. Segments (34-bp) of the MCM7 promoter containing the wild-type and between MYCN and MCM7 expression was additionally investigated by mutated E-box sequences with the following sequences were used (sense immunohistochemistry on tissue sections from 15 MYCN-amplified and strand with E-box region underlined, mutant bases italicized): wild-type ϭ 5Ј- 14 nonamplified INSS stage III and stage IV neuroblastoma tumors (as CGC AGT GGC GCC GGT CAC GTG GGG GGC GAC GTT T-3Ј, mutant well as 4 mature ganglioneuromas, which were entirely negative for ϭ Ј Ј 1 5 -CGC AGT GGC GCC GGT CATATC GGG GGC GAC GTT T-3 , MCM7; not shown; Fig. 1; Table 2). An anti-MCM7 monoclonal anti- ϭ Ј and mutant 2 5 -CGC AGT GGC GCC GGT CATATG GGG GGC GAC body stained the majority of tumor cells positive with a well-defined GTT T-3Ј. Probe (1 ␮l; 105 cpm) was mixed with 5 ␮l of binding buffer (50 nuclear staining pattern in the MYCN-amplified samples. By contrast, ␮g/ml random hexamer p(dN)6, 2.5 mg/ml BSA, 4% Ficoll 400, and 10% only scattered foci of positive staining were demonstrated in the MYCN glycerol) and incubated with 10 ␮l of lysate (7 ␮g) after dilution to 150 mM NaCl and addition of 2 ␮g poly(dI⅐dc), for 30 min at 4°C. Competitors and nonamplified samples, although the overall proliferative fractions of the antibodies were preincubated with the nuclear lysate with poly(dI⅐dc) for 30 tumor samples were very similar (see Fig. 1 legend). These foci may min before the addition of probe. Reactions were loaded onto 6% native represent cells with increased MYCN expression, although this study acrylamide gels and electrophoresed on prerun gels for1hat250Vat4°C. does not address this hypothesis. No MYCN-amplified tumors had Ͻϩϩ Antibodies (sources) were as follows: polyclonal anti-MYCN (C-19; Santa staining, whereas no nonamplified tumor had Ͼϩϩ staining. H&E Cruz Biotechnology Research), monoclonal anti-MYCN (AB-1; Oncogene staining and neuron-specific enolase staining demonstrated no gross Research), anti-MAX (C-124; Santa Cruz Biotechnology Research), and anti- histological differences between the amplified and nonamplified speci- IgG FC control (005098; Jackson ImmunoResearch). mens tested (data not shown). Luciferase Reporter Assays. The luciferase reporter constructs used in MCM7 Is Rapidly Up-Regulated by Inducible MYCN Expres- this report were a gracious gift from Dr. Tohru Kiyono. The full 0.5 Kb MCM7 promoter contains several E2F binding sites and one E-box. In luciferase sion. To eliminate a coincidental correlation between MYCN and reporter assays, a promoter fragment (A8) containing the E-box and the 3Ј E2F MCM7 expression, this phenomenon was additionally analyzed using transactivation site accounted for more than half of the activity of the full- a MYCN nonamplified neuroblastoma cell line, SH-EP, stably trans- length promoter (10). This construct and a control plasmid containing mutated fected with the MYCN gene under the control of the rTet-inducible E-box and E2F sites (A8M) were transfected into TET-21 cells grown in the expression system (MYCN expression is induced on removal of tet- 1124

Table 1 Comparison of mRNA hybridization signal intensity for selected genes between protein is first detectable (by Western blot) in the TET-21 line 3 h MYCN-amplified (P4) and non-amplified (P67) neuroblastoma tumor cell after tetracycline (18). Thus, up-regulation of MCM7 transcription is primary cultures an early response to increased MYCN levels. MYCN levels are Differential Gene name P4 signal P67 signal expression maximal in the TET-21 cell line 6 h after removal of tetracycline (Ref. 18 and data not shown). Furthermore, Western blots indicated that the MYCN (Incyte PD:2811651) 7514 228 41.9 Laminin (Incyte PD:1851696) 223 4373 17.6 level of MCM7 protein increased ϳ3-fold on induction of MYCN Interferon-inducible protein 9–27 209 3891 16.9 (Fig. 2C). These studies in the MYCN-inducible cell line support a (Incyte PD:2902903) V-myb AMV oncogene homolog-like 4992 417 14.1 causal relation between MYCN and MCM7 expression in neuro- 2 (Incyte PD:494905) blastoma. Ferritin, light chain (Incyte 1324 18665 13.2 MYCN Activates the MCM7 Promoter. Transient transfection of a PD:2868138) Integrin beta 1 (Incyte PD:417451) 971 13166 12.2 reporter construct consisting of the 0.1-Kb proximal portion of the MCM7 Inhibin, beta A (Incyte PD:2242648) 403 5308 11.9 promoter fused to luciferase (10) into the MYCN-inducible cell line was Ferritin, heavy chain (Incyte PD: 2330 29888 11.8 27775) Homeo box A9 (Incyte PD:459651) 297 3507 10.7 MCM7 (Incyte PD:986752) 7338 850 10.4 Table 2 Summary of MCM7 immunohistochemistry MHC, class II, DR beta 5 (Incyte 349 3786 10.2 Neuroblastoma tumors from patients with known MYCN amplification status and stage PD:2105963) were analyzed for MCM7 expression by immunohistochemistry. MYCN status was Cell division cycle 2,G1 to S and G2 1656 246 8.0 determined as part of tumor biology studies performed at diagnosis as part of a Pediatric to M (Incyte PD:1525795) Oncology Group protocol. Grading is as follows: Ͻ1% ϭ trace, 1–25% ϭϩ,25– ␤ Transforming growth factor 1039 214 5.4 50% ϭϩϩ,50–75% ϭ ϩϩϩ, Ͼ75% ϭ ϩϩϩϩ. (Incyte PD:2375329) CDC21 homolog, MCM4 (Incyte 1510 409 3.9 MYCN-amplified tumors MYCN nonamplified tumors PD:103669) Tumor MCM7 INSS stage Tumor MCM7 INSS stage S1 ϩϩϩϩ 4 S16 ϩ 4 S2 ϩϩϩ 4 S17 ϩ 3 racycline). Northern and Western blots confirmed the tetracycline- S3 ϩϩϩϩ 4 S18 ϩϩ 4 dependent expression of exogenous MYCN described previously (data S4 ϩϩϩϩ 3 S19 ϩϩ 4 S5 ϩϩϩϩ 4 S20 ϩ 3 not shown; Ref. 18). Phenotypic changes observed on MYCN induc- S6 ϩϩϩϩ 3 S21 ϩ 4 tion in this cell line, TET-21, have been well characterized and S7 ϩϩϩ 4 S22 ϩ 3 ϩϩϩϩ ϩϩ include cell cycle alterations, increased proliferation rate, and in- S8 3 S23 3 S9 ϩϩϩϩ 4 S24 ϩϩ 4 creased metastatic potential (18). Fig. 2A demonstrates increased S10 ϩϩϩ 3 S25 ϩ 4 expression of MCM7 corresponding to increased MYCN expression in S11 ϩϩϩϩ 4 S26 ϩϩ 4 S12 ϩϩϩ 4 S27 ϩ 3 the MYCN-induced cells. Kinetic analysis (Fig. 2B) shows the signal S13 ϩϩϩϩ 4 S28 ϩϩ 3 for MCM7 increasing within 2–3 h after the initiation of MYCN S14 ϩϩϩϩ 4 S29 ϩ 4 induction at a point when MYCN levels are not yet maximal. MYCN S15 ϩϩϩ 4

Fig. 1. MCM7 immunohistochemistry in neuroblastoma tumors. A and C, two representative MYCN-amplified tumors. B and D, two representative MYCN nonamplified tumors. ϩ Overall, tumor histology and architecture were similar as indicated by H&E and neuron specific enolase staining (data not shown). Proliferative fractions (percentage S G2-M phase) of these four tumors, A–D, were all markedly elevated (24.0, 26.0, 31.7, and 14.1, respectively), and staining for the proliferation marker, Ki-67 antigen, was equivalent among all samples. This staining pattern was consistently found for all 29 tumor specimens analyzed. 1125

E-box within the MCM7 promoter sequence. Nuclear lysates from several MYCN-amplified and nonamplified neuroblastoma cell lines and tumor cell primary cultures were used in EMSA experiments with a 34-bp end-labeled probe containing the E-box site and flanking regions from the MCM7 promoter (see “Materials and Methods”). Fig. 4A demonstrates a specific complex of protein bound to oligonucleotide probe retarded relative to free probe on a native acrylamide gel. Nuclear lysates from MYCN-amplified cell lines and tumors (IMR-32 and P46) consistently showed much higher intensity bands than their nonamplified counterparts (SK-N-SH shown for example; Fig. 4B) likely reflecting increased MYCN/MAX heterodimer content of MYCN-amplified tumor lines. Competition experiments using cold probes with wild-type and mutated E-box sequences were performed to test the sequence specificity of the EMSA. Fig. 4, A and C show pronounced inhibition of complex formation with wild-type sequence as the cold competitor. This inhibition is abrogated when bases within the E-box motif of the competitor oligonucleotide are altered (CACGTG to CATATC or CATATG). Further- more, anti-MAX and anti-MYCN antibodies specifically interfere with protein/DNA complex formation as shown in Fig. 4D. This effect is seen whether the antibodies are added before or after probe and lysate are combined. There was no such effect with the irrelevant polyclonal anti- immunoglobulin G FC antibody. No super-shift of the protein/DNA complex could be detected with the antibodies used suggesting that the antibodies interfere with formation of the MYCN/MAX/promoter complex and/or do not bind to the intact protein/DNA adduct. Taken together these results indicate that in neuroblastoma tumor samples, primary tumor cell cultures, and cell lines, high MCM7 expression correlates with overexpression of MYCN. EMSA studies showing specific interaction of MYCN with the MCM7 promoter E-box element, combined with the kinetics of MCM7 gene activation closely after that of MYCN, implicate MCM7 as a definite transactivation target for the MYCN transcription factor. Thus, direct binding of MYCN to the transactivation site in the MCM7 promoter is partially responsible for the increased expression of MCM7 observed in MYCN-amplified tumor

Fig. 2. MCM7 expression correlates with MYCN induction. A, cDNA Northern blot from mRNA of MYCN-conditional neuroblastoma cell line TET-21. Blot was probed sequentially for MCM7 and then MYCN. B, cDNA Northern blot demonstrating the time course of MCM7 expression on induction of MYCN in the TET-21 cell line. Tetracycline removed at time 0 h. Values are normalized to the actin content of the same samples. C, Western blot of MCM7 in MYCN conditional cells. Cell lysate (15 ␮g) from MYCN- induced and noninduced TET-21 cells was subjected to electrophoresis and electroblotting then probed sequentially with antibodies to MCM7 and actin. Quantification of the ratio of MCM7:actin was performed by PhosphorImager. MYCN induction resulted in an ϳ3-fold increase in MCM7 when normalized to actin. used to study the effect of MYCN on MCM7 promoter activity. Fig. 3 presents the data from five independent experiments demonstrating significant increases in promoter activity on MYCN induction (removal of tetracycline) with a mean increase of 64% (P Ͻ 0.01). A control construct with a mutated E-box and proximal E2F sites (A8M) consistently gave Ͻ5% of the activity seen from the native E-box containing vector and showed no significant difference on MYCN induction. This increase in MCM7 promoter activity in these transient transfections is less than that demonstrated for the chromosomal promoter in tumors (10-fold). Previ- ous studies have noted similar results when comparing episomal transcription from transiently transfected reporter constructs to expression of integrated chromosomal genes (21–23). These transient transfection assays additionally support the direct interaction of MYCN with the MCM7 Fig. 3. MCM7 promoter activity in the MYCN-inducible Tet-21 cell line. The MYCN- promoter. inducible Tet-21 cell line was transfected with the MCM7 promoter/luciferase reporter MYCN Protein Binds Specifically to the MCM7 Promoter. The construct illustrated above. Induction of MYCN led to a significant increase in luciferase activity (from 56.2 ϩ/Ϫ 16 to 92 ϩ/Ϫ 17; P Ͻ 0.005). Data summarized from five human MCM7 promoter consists of a DNA sequence 5Ј to the coding separate experiments run in triplicate and evaluated with two-tailed Student’s t test and region that contains several transcription factor-binding sites including Wilcoxon signed rank test. Control vector is promoter construct with a mutated E-box that does not bind MYCN. Transfections were done in the presence of tetracycline as described three E2F sites, three GC boxes, and an E-box (10). EMSAs were (see “Materials and Methods”); cells were replated with and without tetracycline and performed to determine whether MYCN is capable of binding to the luciferase activity measured 36 h after transfection; bars, ϮSD. 1126

Fig. 4. Electrophoretic mobility shifts demonstrate specific interaction of MYCN with the MCM7 promoter. Electrophoretic mobility shift assays were performed on nuclear lysates of various tumors and cell lines illustrat- ing a clear specific complex, which was inhibited on addition of excess cold probe, was specific for wild-type E-box sequence, and was inhibited by anti-MYCN and anti-MAX antibodies. A, nuclear lysates from MYCN- amplified cell lines (IMR-32) and a primary tumor culture (P46) were used to assay for gel shifts. All assays were performed with the labeled 34-bp MCM7 promoter fragment containing the E-box element and flanking regions as probe. Whereas several bands were detected, the arrow indicates the band specific for MYCN/MAX binding to the MCM7 promoter fragment as shown by its complete competition with cold probe (50 ϫ molar concentration) and inhibition with anti-MYCN antibody. B, gel shifts performed with equal quantities (4 ␮g) of nuclear lysates from MYCN-amplified and nonamplified tumors demonstrate markedly increased binding in cells with high MYCN. SK-N-SH is a non-MYCN-amplified tumor cell line and is shown in comparison to IMR-32 and P46. C, left panel illustrates EMSA complexes formed from cell line nuclear lysates of P46 neuroblastoma tumor cell culture in the presence of unlabeled wild-type E-box probe. Relative molar concentrations are as indicated. Right panels demonstrate lack of specific competition with mutated E-box sequences (CATATC or CATATG) at the equivalent molar concentration. Open arrow indicates non-E box-specific competition of high molecular weight adducts. D, inhibition of complex formation in the presence of anti-MYC or anti-MAX antibodies. Anti-MYCN and anti-MAX antibodies markedly inhibit the formation of the complex.

cells. These findings have important implications for the pathogenesis (28). Although MYCN has been well characterized as a transcriptional and biology of neuroblastoma as discussed below. regulator with clear transforming ability, the transcriptional target(s) that are responsible for this activity are not understood (16, 28). Recently, DISCUSSION Boon et al. (29) reported significantly enhanced expression of ribosomal proteins, as well as proteins involved in ribosome biogenesis and protein Neuroblastoma is clinically and biologically heterogeneous. MYCN- synthesis in MYCN-amplified neuroblastoma tumors and a MYCN-con- amplified tumors are particularly aggressive and respond poorly to ther- ditional cell line. The full characterization of the downstream genes apy. Whereas no mechanism linking gene amplification to this distinct activated (or repressed) by MYCN amplification should help elucidate the clinical phenotype has been established, the bulk of evidence, both with mechanism underlying the clinical phenotype of MYC-overexpressing MYCN as well as MYCC, supports a model in which increased expression tumors in general and may suggest novel therapeutic approaches. of the oncogene, either by amplification or deregulated transcription, We have demonstrated that the binding of the MYCN/MAX het- directly increases tumor cell growth rate and metastatic potential (16). A direct correlation exists between MYCN expression levels and growth erodimer to its promoter E-box activates MCM7. This result is consistent rate, and motility and cell cycle alterations in cell culture models (18, 24, with the increased expression of MCM7 protein in a MYCN-inducible 25). In vitro studies have also demonstrated that MYCN affects adhesion, neuroblastoma cell line, as well as in MYCN-amplified tumor tissue invasiveness, response to growth factors, and autocrine factor production samples and tumor cell primary cultures. Whereas previous reports have (26, 27). Additional evidence of the primary transforming ability of demonstrated an association between cell proliferation and expression of MYCN comes from the demonstration that tissue specific overexpression MCM molecules (6, 7), several observations suggest that the MYCN- of MYCN leads to a neuroblastoma-like malignancy in transgenic mice dependent up-regulation of MCM7 in neuroblastoma tumor cells is at 1127

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2002 American Association for Cancer Research. MYCN TRANSACTIVATION OF MCM7 IN NEUROBLASTOMA least partially independent of other MYCN-induced proliferative 10. Suzuki, S., Adachi, A., Hiraiwa, A., Ohashi, M., Ishibashi, M., and Kiyono, T. changes. First, MCM7 is rapidly up-regulated on induction of MYCN Cloning and characterization of human MCM7 promoter. Gene (Amst.), 216: 85–91, 1998. expression in the TET-21 cell line, with a time course more consistent 11. Prochownik, E. V., and VanAntwerp, M. E. Differential patterns of DNA binding by with direct transactivation rather than as a secondary effect of increased myc and max proteins. Proc. Natl. Acad. Sci. USA, 90: 960–964, 1993. 12. O’Hagan, R. C., Schreiber-Agus, N., Chen, K., David, G., Engelman, J. A., Schwab, proliferation. Second, the proliferative index and G1-S fraction of cells R., Alland, L., Thomson, C., Ronning, D. R., Sacchettini, J. C., Meltzer, P., and derived from the MYCN-amplified and nonamplified tumor specimens DePinho, R. A. Gene-target recognition among members of the myc superfamily and illustrated in Fig. 1 are very similar (see legend, Fig. 1), implying that this implications for oncogenesis. Nat. Genet., 24: 113–119, 2000. 13. Hurlin, P. J., Queva, C., Koskinen, P. J., Steingrimsson, E., Ayer, D. E., Copeland, broad range of MCM7 expression is largely independent of proliferation N. G., Jenkins, N. A., and Eisenman, R. N. Mad3 and Mad4: novel Max-interacting rate. Third, a previous study by Ohtani et al. (9) demonstrates that serum transcriptional repressors that suppress c-myc dependent transformation and are stimulation or forced exogenous E2F expression activates the coordinate expressed during neural and epidermal differentiation. EMBO J., 14: 5646–5659, 1995. expression of the MCM proteins at the G1-S boundary in rat embryonic 14. Bordow, S. B., Norris, M. D., Haber, P. S., Marshall, G. M., and Haber, M. Prognostic fibroblasts. Human MCM5, 6 and 7, as well as mouse Mcm3, have significance of MYCN oncogene expression in childhood neuroblastoma. J. Clin. Oncol., 16: 3286–3294, 1998. functional E2F binding sites in their promoters suggesting that this 15. Matthay, K. K., Villablanca, J. G., Seeger, R. C., Stram, D. O., Harris, R. E., Ramsay, transcription factor is partly responsible for the coordinate MCM protein N. K., Swift, P., Shimada, H., Black, C. T., Brodeur, G. M., Gerbing, R. B., and response to progression through the cell cycle (9, 10). Thus MYCN Reynolds, C. P. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. Chil- binding to the unique E-box of MCM7 likely increases its expression dren’s Cancer Group. N. Engl. J. Med., 341: 1165–1173, 1999. independent of the other MCM molecules. 16. Nesbit, C. E., Tersak, J. M., and Prochownik, E. V. MYC oncogenes and human The concept that MCM7 may play an important regulatory role in neoplastic disease. Oncogene, 18: 3004–3016, 1999. 17. Sarkar, A. K., and Nuchtern, J. G. Lysis of MYCN-amplified neuroblastoma cells by tumor cell biology arises from several recent findings of novel MYCN peptide-specific cytotoxic T lymphocytes. Cancer Res., 60: 1908–1913, MCM7/protein interactions. Yeast two-hybrid experiments have dem- 2000. 18. Lutz, W., Stohr, M., Schurmann, J., Wenzel, A., Lohr, A., and Schwab, M. Condi- onstrated MCM7 binding to RB1-related protein p107, resulting in tional expression of N-myc in human neuroblastoma cells increases expression of down-regulation of DNA replication when MCM7 is limiting (30) to ␣-prothymosin and ornithine decarboxylase and accelerates progression into S-phase the HPV E6 oncoprotein (31) and to FLH-2 (a heart-specific LIM-zinc early after mitogenic stimulation of quiescent cells. Oncogene, 13: 803–812, 1996. 19. Franz, O., Bruchhaus, I., and Roeder, T. Verification of differential gene transcription finger protein also expressed in lung cancer and chronic myelogenous using virtual northern blotting. Nucleic Acids Res., 27: e3, 1999. leukemia cell lines; Ref. 32). MCM7 interacts with MAT-1, a subunit 20. Templeton, N. S., and Lasic, D. D. New directions in liposome gene delivery. Mol. of the cyclin-dependent kinase activating kinase that promotes the Biotechnol., 11: 175–180, 1999. 21. Cook, J. L., Zhang, Z., Alam, J., and Re, R. N. Effects of chromosomal integration stability and activation of the CDK7-cyclin H complex (33). It is also site upon p53 interactions with DNA consensus sequence homologies. Oncogene, 18: a component of a DNA helicase/ATPase (Refs. 34, 35; composed of 2373–2379, 1999. 22. Stein, G. S., van Wijnen, A. J., Stein, J. L., and Lian, J. B. Interrelationships of MCM4, 6, and 7) and associates with the RNA polymerase II holoen- transcriptional machinery with nuclear architecture. Crit. Rev. Eukaryot. Gene Expr., zyme complex (36). The biochemical and functional properties of 9: 183–190, 1999. MCM7 suggest a regulatory role for this protein in cell division cycle 23. Stein, G. S., Montecino, M., van Wijnen, A. J., Stein, J. L., and Lian, J. B. Nuclear structure-gene expression interrelationships: implications for aberrant gene expres- and RNA synthesis. As such, its transcriptional activation by the sion in cancer. Cancer Res., 60: 2067–2076, 2000. MYCN oncoprotein is remarkable. In concert with other MYCN- 24. Fotsis, T., Breit, S., Lutz, W., Rossler, J., Hatzi, E., Schwab, M., and Schweigerer, L. mediated effects, transcriptional activation of MCM7 could be a Down-regulation of endothelial cell growth inhibitors by enhanced MYCN oncogene expression in human neuroblastoma cells. Eur. J. Biochem., 263: 757–764, 1999. significant component of MYCN-induced tumorigenesis. 25. Schweigerer, L., Breit, S., Wenzel, A., Tsunamoto, K., Ludwig, R., and Schwab, M. Augmented MYCN expression advances the malignant phenotype of human neuroblastoma cells: evidence for induction of autocrine growth factor activity. Cancer ACKNOWLEDGMENTS Res., 50: 4411–4416, 1990. 26. Chambery, D., Mohseni-Zadeh, S., de Galle, B., and Babajko, S. N-myc regulation of We thank Tohru Kiyono (Aichi Cancer Center, Nagoya, Japan) for the gifts type I insulin-like growth factor receptor in a human neuroblastoma cell line. Cancer of the MCM7 promoter/reporter constructs and full length MCM7 cDNA and Res., 59: 2898–2902, 1999. 27. Schweigerer, L., Scheurich, P., and Fotsis, T. Enhanced MYCN oncogene expression Manfred Schwab for the Tet-21 cell line. We also thank Jeffery Rosen (Baylor in human neuroblastoma cells results in increased susceptibility to growth inhibition College of Medicine, Houston, Texas) and Philip Hastings (Baylor College of by TNF ␣. Biochem. Biophys. Res. Commun., 170: 1301–1307, 1990. Medicine, Houston, Texas) for critical review and discussion of the manu- 28. Weiss, W. A., Aldape, K., Mohapatra, G., Feuerstein, B. G., and Bishop, J. M. script. Targeted expression of MYCN causes neuroblastoma in transgenic mice. EMBO J., 16: 2985–2995, 1997. 29. Boon, K., Caron, H. N., van Asperen, R., Valentijn, L., Hermus, M-C., van Sluis, P., REFERENCES Roobeek, I., Weis, I., Voute, P. A., Schwab, M., and Versteeg, R. N-myc enhances the expression of a large set of genes functioning in ribosome biogenesis and protein 1. Takisawa, H., Mimura, S., and Kubota, Y. Eukaryotic DNA replication: from pre- synthesis. EMBO J., 20: 1383–1393, 2001. replication complex to initiation complex. Curr. Opin. Cell Biol., 12: 690–696, 2000. 30. Sterner, J. M., Dew-Knight, S., Musahl, C., Kornbluth, S., and Horowitz, J. M. 2. Maiorano, D., Lemaitre, J. M., and Mechali, M. Stepwise regulated chromatin Negative regulation of DNA replication by the retinoblastoma protein is mediated by assembly of MCM2–7 proteins. J. Biol. Chem., 275: 8426–8431, 2000. its association with MCM7. Mol. Cell. Biol., 18: 2748–2757, 1998. 3. Kearsey, S. E., and Labib, K. MCM proteins: evolution, properties, and role in DNA 31. Kukimoto, I., Aihara, S., Yoshiike, K., and Kanda, T. Human papillomavirus onco- replication. Biochim. Biophys. Acta, 1398: 113–136, 1998. protein E6 binds to the C-terminal region of human minichromosome maintenance 7 4. Tye, B. K. MCM proteins in DNA replication. Annu. Rev. Biochem., 68: 649–686, protein. Biochem. Biophys. Res. Commun., 249: 258–262, 1998. 1999. 32. Chan, K. K., Tsui, S. K., Ngai, S. M., Lee, S. M., Kotaka, M., Waye, M. M., Lee, 5. Thommes, P., and Blow, J. J. The DNA replication licensing system. Cancer Surv., C. Y., and Fung, K. P. Protein-protein interaction of FHL2, a LIM domain protein 29: 75–90, 1997. preferentially expressed in human heart, with hCDC47. J. Cell. Biochem., 76: 6. Todorov, I. T., Werness, B. A., Wang, H. Q., Buddharaju, L. N., Todorova, P. D., 499–508, 2000. Slocum, H. K., Brooks, J. S., and Huberman, J. A. HsMCM2/BM28: a novel 33. Wang, Y., Xu, F., and Hall, F. L. The MAT1 cyclin-dependent kinase-activating proliferation marker for human tumors and normal tissues. Lab. Investig., 78: 73–78, kinase (CAK) assembly/targeting factor interacts physically with the MCM7 DNA 1998. licensing factor. FEBS Lett., 484: 17–21, 2000. 7. Freeman, A., Morris, L. S., Mills, A. D., Stoeber, K., Laskey, R. A., Williams, G. H., 34. Ishimi, Y., Komamura-Kohno, Y., You, Z., Omori, A., and Kitagawa, M. Inhibition and Coleman, N. Minichromosome maintenance proteins as biological markers of of Mcm4,6,7 helicase activity by phosphorylation with cyclin A/Cdk2. J. Biol. Chem., dysplasia and malignancy. Clin. Cancer Res., 5: 2121–2132, 1999. 275: 16235–16241, 2000. 8. Hiraiwa, A., Fujita, M., Adachi, A., Ono, H., Nagasaka, T., Matsumoto, Y., Ohashi, 35. Labib, K., Tercero, J. A., and Diffley, J. F. Uninterrupted MCM2–7 function M., Tomita, Y., and Ishibashi, M. Specific distribution patterns of hCDC47 expres- required for DNA replication fork progression. Science (Wash. DC), 288: 1643– sion in cutaneous diseases. J. Cutan. Pathol., 25: 285–290, 1998. 1647, 2000. 9. Ohtani, K., Iwanaga, R., Nakamura, M., Ikeda, M., Yabuta, N., Tsuruga, H., and 36. Yankulov, K., Todorov, I., Romanowski, P., Licatalosi, D., Cilli, K., McCracken, S., Nojima, H. Cell growth-regulated expression of mammalian MCM5 and MCM6 genes Laskey, R., and Bentley, D. L. MCM proteins are associated with RNA polymerase mediated by the transcription factor E2F. Oncogene, 18: 2299–2309, 1999. II holoenzyme. Mol. Cell. Biol., 19: 6154–6163, 1999. 1128

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2002 American Association for Cancer Research. Minichromosome Maintenance Protein MCM7 Is a Direct Target of the MYCN Transcription Factor in Neuroblastoma

Jason M. Shohet, M. John Hicks, Sharon E. Plon, et al.

Cancer Res 2002;62:1123-1128.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/62/4/1123

Cited articles This article cites 36 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/62/4/1123.full#ref-list-1

Citing articles This article has been cited by 16 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/62/4/1123.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/62/4/1123. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.