The Neuronal Differentiation Factor Neurod1 Downregulates the Neuronal Repellent Factor Slit2 Expression and Promotes Cell Motil

Published OnlineFirst February 24, 2011; DOI: 10.1158/0008-5472.CAN-10-3524

Cancer Molecular and Cellular Pathobiology Research

The Neuronal Differentiation Factor NeuroD1 Downregulates the Neuronal Repellent Factor Slit2 Expression and Promotes Cell Motility and Tumor Formation of Neuroblastoma

Peng Huang1, Satoshi Kishida1, Dongliang Cao1, Yuko Murakami-Tonami1, Ping Mu1, Masato Nakaguro1, Naoshi Koide1, Ichiro Takeuchi2, Akira Onishi3, and Kenji Kadomatsu1

Abstract The basic helix-loop-helix transcription factor NeuroD1 has been implicated in the neurogenesis and early differentiation of pancreatic endocrine cells. However, its function in relation to cancer has been poorly examined. In this study, we found that NeuroD1 is involved in the tumorigenesis of neuroblastoma. NeuroD1 was strongly expressed in a hyperplastic region comprising neuroblasts in the celiac sympathetic ganglion of 2-week- old MYCN transgenic (Tg) mice and was consistently expressed in the subsequently generated neuroblastoma tissue. NeuroD1 knockdown by short hairpin RNA (shRNA) resulted in motility inhibition of the human neuroblastoma cell lines, and this effect was reversed by shRNA-resistant NeuroD1. The motility inhibition by NeuroD1 knockdown was associated with induction of Slit2 expression, and knockdown of Slit2 could restore cell motility. Consistent with this finding, shRNA-resistant NeuroD1 suppressed Slit2 expression. NeuroD1 directly bound to the first and second E-box of the Slit2 promoter region. Moreover, we found that the growth of tumor spheres, established from neuroblastoma cell lines in MYCN Tg mice, was suppressed by NeuroD1 suppression. The functions identified for NeuroD1 in cell motility and tumor sphere growth may suggest a link between NeuroD1 and the tumorigenesis of neuroblastoma. Indeed, tumor formation of tumor sphere–derived cells was significantly suppressed by NeuroD1 knockdown. These data are relevant to the clinical features of human neuroblastoma: high NeuroD1 expression was closely associated with poor prognosis. Our findings establish the critical role of the neuronal differentiation factor NeuroD1 in neuroblastoma as well as its functional relationship with the neuronal repellent factor Slit2. Cancer Res; 71(8); 2938–48. 2011 AACR.

Introduction NeuroD1-deficient mice with either transgenic mice expressing NeuroD1 in b cells or (mammalian atonal homolog) MATH- NeuroD1 is a basic helix-loop-helix (bHLH) transcription deficientmice,MiyataandcolleaguesaswellasSchwaband factor. NeuroD1-deficient mice die as a consequence of severe colleagues demonstrated that NeuroD1 deletion in the central hyperglycemia within 5 days of birth (1). This is attributed to nervous system leads to depletion of the cerebellar granule cells differentiation deficit and death of b cells in the pancreas during and loss of the dentate gyrus in the hippocampus (3, 4). Further- embryogenesis (1). In addition, NeuroD1 is involved in neurogen- more, NeuroD1 is involved in the differentiation of peripheral esis. It can convert embryonic epidermal cells into fully differ- neurons. Thus, NeuroD1-deficient mice exhibit neuronal cell loss entiated neurons in Xenopus and promotes premature cell-cycle in the inner ear (5, 6). In the retina of NeuroD1-deficient mice, exit and differentiation of neural precursor cells (2). By crossing amacrine cell differentiation is perturbed and the number of bipolar interneurons decreases (7). Furthermore, it has recently been reported that NeuroD1 plays an essential role in the main- Authors' Affiliations: 1Department of Biochemistry, Nagoya University Graduate School of Medicine; 2Department of Computer Science/Scien- tenance of neural precursor cells in adult neurogenesis: inducible tific and Engineering Simulation, Nagoya Institute of Technology, Nagoya; stem cell–specific deletion of NeuroD1 causes fewer newborn and 3Transgenic Animal Research Center, National Institute of Agrobio- neurons in the hippocampus and the olfactory bulb (8, 9). logical Sciences, Tsukuba, Ibaraki, Japan Neuroblastoma (NB) is the most common extracranial Note: Supplementary data for this article are available at Cancer Research pediatric solid tumor and is derived from the sympathetic Online (http://cancerres.aacrjournals.org/). neuron lineage of neural crest cells (10). A hierarchical expres- Corresponding Author: Kenji Kadomatsu, Department of Biochemistry, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, sion of regulatory molecules, such as Sox2, BMP, Mash1, Showa-ku, Nagoya 466-8550, Japan. Phone: 81-52-744-2059; Fax: 81- Phox2B, and Hand2, determines the fate of neural crest cells 52-744-2065; E-mail: [email protected] and their further differentiation to form sympathetic neurons doi: 10.1158/0008-5472.CAN-10-3524 (11). The molecular mechanism of the normal development of 2011 American Association for Cancer Research. sympathetic neurons may influence the biology of NB. For

2938 Cancer Res; 71(8) April 15, 2011

NeuroD1 in Neuroblastoma Tumorigenesis

example, the homeodomain DNA-binding transcription factor (Open Biosystems), and human Slit2 (Open Biosystems) were Phox2B is mutated in familial and sporadic forms of NB (12). used. The shRNA sequences are listed in Supplementary Phox2B mutations predispose to NB by promoting the pro- Table SII. Together with the lentivirus packaging plasmids liferation and dedifferentiation of cells in the sympathoadre- psPAX2 (Addgene plasmid 12260) and pMD2.G (Addgene nergic lineage (13). Phox2B induction in a NB cell line plasmid 12259), all plasmids were cotransfected into HEK293T downregulates expression of the homeobox protein Msx1, cells to produce infectious virus particles. The titer was whereas Msx1 upregulates expression of Notch3 and Hey1 evaluated by using a QuickTiter Lentivirus Titer kit (Cell and downregulates NeuroD1 expression (14). The bHLH tran- Biolabs). Approximately 2.5 109/mL virus particles were scription factor Hand2, the proto-oncogene MYCN, and tyr- obtained. NB cells were infected in the presence of 8 mg/ osine hydroxylase, which is essential for catecholamine mL polybrene (Sigma). For the infection of tumor sphere cells, metabolism, are expressed in proliferating precursors of sym- virus particles were concentrated by PEG-it Virus Precipita- pathetic neurons (15). MYCN gene amplification is often tion Solution (System Biosciences) and suspended in a sphere observed in NB, and it is the strongest prognostic factor of growth medium without polybrene. NB (16). Furthermore, the bHLH transcription factor Mash1- deficient mice lose sympathetic neurons (17). Mash1 is Plasmids expressed in multipotent neuron progenitors immediately A human NeuroD1 cDNA clone was amplified from human after expression of neural stem cells during postnatal hippo- HEK293 cell cDNA with attB-flanked primers. For shRNA- campus granule neurogenesis, whereas NeuroD1 is expressed resistant expression constructs, the shRNA-targeting site on in more differentiated progenitors (18). Mash1 is expressed the cDNA was mutated at 6 nucleotides without affecting the mostly in poorly differentiated neuroendocrine carcinoma, amino acid sequence. Thereafter, following the instructions of whereas NeuroD1 is expressed in well-differentiated neuroen- Invitrogen Gateway Technology and using the ClonaseII Kit, docrine carcinoma (19). the mutated NeuroD1 cDNA was inserted into the gate- Weiss and colleagues generated MYCN transgenic (Tg) way destination vector CSII-CMV-RfA-IRES2-Venus (RIKEN mice, in which MYCN expression is targeted to the sympa- BioResource Center, Ibaraki, Japan), which is a lentivirus- thetic neuron lineage by the rat tyrosine hydroxylase (20). based cDNA over-expression vector. CSII-CMV-Venus These mice develop NB in the celiac sympathetic ganglion, (RIKEN) was used as the negative control. which has histology indistinguishable from that of human NB (20). In particular, hemizygous mice develop tumors with a Histochemistry and Western blot analysis stepwise increase of the frequency of MYCN amplification (21). Tissues were fixed in 4% paraformaldehyde, dehydrated, Therefore, this model mimics human NB development. More- and embedded in paraffin according to the standard proce- over, this model demonstrates a marked increase of neuro- dure. Then, 5-mm sections were stained with hematoxylin and blast hyperplasia at approximately 2 weeks after birth that is eosin stain or incubated with anti-NeuroD1 polyclonal anti- hypothesized to represent a precancerous lesion (21). body (Cell Signaling Technology), which was diluted at a In spite of the accumulating evidence showing the critical concentration of 1:50 in 3% normal goat serum for 2 hours roles of NeuroD1 in neurogenesis, its involvement in cancer at room temperature, and then incubated with biotin goat development has not been fully explored. As NB tumorigenesis anti-rabbit secondary antibody (1:50; BD Pharmingen) for is closely related to the normal development of the sympa- 30 minutes at room temperature. For Western blot analysis, thetic neuron lineage, we hypothesized that NeuroD1 could be anti-NeuroD1 (1:500; Santa Cruz Biotechnology), anti-Slit2 an important regulator of NB. To examine this hypothesis, we (1:500; Millipore), and anti-b-actin (1:10,000; Sigma) antibodies used MYCN hemizygous mice and found that NeuroD1 plays a were used. critical role in NB tumorigenesis. ChIP assay Materials and Methods SH-SY5Y cells were cross-linked by 1% formaldehyde, and the reaction was quenched with 0.125 mol/L glycine. Cells RT-PCR were harvested by using PBS containing phenylmethylsulfo- Thermal conditions for PCR included 28–35 cycles of dena- nylfluoride (PMSF) followed by homogenization with hypo- turing at 94C for 30 seconds, annealing at 55–56C for 30 tonic buffer. Then, DNA was digested to 150 to 900-bp seconds, and elongation at 72C for 30 seconds. The primers fragments with micrococcal nuclease (New England Biolabs). used and the expected product sizes are listed in Supplemen- The pellet obtained was resuspended in 300 mL of lysis buffer tary Table SI. Detection of NeuroD1 and Slit2 was performed containing propidium iodide and sonicated. The lysates were by using the TaqMan Gene Expression Assays kit (Applied centrifuged, and the supernatants obtained were diluted with Biosystems) according to the manufacturer's instructions. The dilution buffer. Furthermore, these lysates were incubated assay ID for NeuroD1 is Hs00159598_m1 and that for Slit2 is with control Immunoglobulin G1 (IgG1) antibody (R&D Sys- Hs00191193_m1. tems) or anti-NeuroD1 (Santa Cruz Biotechnology) accompanied by rotation for 4 hours at 4C this was followed by the RNA interference addition of 15 mL of protein G Sepharose (GE Healthcare) and Nontargeting short hairpin RNA (shRNA; Sigma) and spe- rotation for 1 hour at 4C. The precipitants obtained were cific shRNA for human NeuroD1 (Sigma), mouse NeuroD1 washed, eluted with elution buffer, and subjected to PCR. The

www.aacrjournals.org Cancer Res; 71(8) April 15, 2011 2939

Huang et al.

chromatin immunoprecipitation (ChIP) primer sequences are into 1-month-old 129/SVJ WT mice. Eighteen days later, the listed in Supplementary Table SI. tumors were dissected and weighed.

Wound-healing assay and Boyden chamber assay Cell culture A total of 2 105 cells were plated in a 6-well plate, and, SH-SY5Y cells were obtained from the American Type 24 hours later, infected with virus. For the wound-healing Culture Collection. IMR32 cells were procured from RIKEN. assay, another 48 hours later, a scratch on the confluent We monitored the morphology of these cells by microscopic monolayers of cells was made using a plastic pipette tip. examination and confirmed that their morphology was the The cell-migration area was calculated by using Metamorph same as that of the original cells. In addition to cell growth and 6.1 software (Molecular Devices), and it was represented as the morphology, we also confirmed that IMR32 cells strongly change in length (mm). For the Boyden chamber assay, cells expressed MYCN whereas SH-SY5Y cells did not. No further were harvested 48 hours after virus infection and counted. validation was performed during these experiments. Stable Transwell plates (pore size: 8 mm; Corning) were used. The NeuroD1-knockdown SH-SY5Y cells were maintained in undersurface of the chamber was coated with 10 mg/mL DMEM with 10% FBS and 1 mg/mL puromycin (InvivoGen). collagen I (Upstate Biotechnology). Virus-infected cells were seeded in the upper chamber (4 104 per well) with 1% Data analysis bovine serum albumin. Thereafter, 10% FBS was added to Results are expressed as the mean SD. Statistical sig- the lower chamber. Cells were allowed to migrate for 6 hours. nificance was evaluated by Student's t test. Nonmigrated cells in the upper chamber were removed by using a swab. Migrated cells attached to the lower side of Results chamber were stained with 40,6-diamidino-2-phenylindole and counted in 4 randomly selected fields. Then, 4 fields at 100 NeuroD1 is expressed in the hyperplastic region of times the magnitude per sample were selected. The experi- sympathetic ganglions ment was performed twice, with each sample in triplicate. We examined the NeuroD1 expression profile during the tumorigenesis of NB in MYCN Tg mice. As previously reported Animals (21), we found that neuroblast hyperplasia occurred in the The MYCN Tg mice (20) were maintained in our animal celiac sympathetic ganglions approximately 2 weeks after facility, where they were housed under a controlled environ- birth in MYCN Tg mice but not in WT mice [Fig. 1A (a and ment and provided with standard nourishment and water. b)]. There were several loci of neuroblast hyperplasia in a Tumor tissues from hemizygous MYCN Tg mice were dis- ganglion [Fig. 1A (b)]. Because the incidence of hyperplasia sected, minced, and then inoculated subcutaneously into 1- correlates with the incidence of tumor formation in MYCN Tg month-old 129/SVJ wild-type (WT) mice (the mouse strain of mice, neuroblast hyperplasia is considered to be the precan- MYCN Tg mice). After approximately 15 such serial passages, a cerous status of NB (21). We found that ganglion cells of WT more invasive tumor was obtained. This constituted our mice, at 2 weeks after birth, showed only low-level expression allograft model in the present study. This study was approved of NeuroD1 [Fig. 1A (c)]. Surprisingly, NeuroD1 was strongly by the Animal Care and Use Committee of Nagoya University expressed in the hyperplastic loci in MYCN Tg mice [Fig. 1A Graduate School of Medicine, Nagoya, Japan. (d), arrowheads], whereas only low-level expression of Neu- roD1 was detected in the surrounding ganglion cells [Fig. 1A Tumor sphere culture and allograft tumorigenesis (d)]. The expression was strong in the majority of hyperplastic A 0.5-cm3 tumor tissue was dissected and washed several cells but not necessarily in all cells [Fig. 1A (d)]. We confirmed times with Hanks’ balanced salt solution (HBSS). After thor- the presence of NeuroD1 mRNA expression in the hyperplastic ough mincing, cells were digested with 0.25% trypsin (5 mL) at ganglions of MYCN Tg mice by using RT-PCR, but the expres- 37C for 10 minutes. The same volume of trypsin inhibitor sion in WT ganglion was below the detection threshold (Sigma) was used to stop the digestion. After centrifuging and (Fig. 1B). Furthermore, we found that there was continued washing with HBSS, the tissue pellet was mixed with 5 mL NeuroD1 expression in advanced cancerous tissues of NB HBSS and allowed to react for 10 minutes. Thereafter, the derived from MYCN Tg mice (Fig. 1C). Interestingly, NeuroD1 supernatant was carefully transferred to a new tube and expression was not homogenous, but rather a portion of the centrifuged. The cell pellet was suspended in a growth med- total cancer cells strongly expressed it (Fig. 1C). ium for stem cells [Dulbecco's modified Eagle Medium Hyperplasia in WT mice usually regresses approximately (DMEM):F12K 1:1, epidermal growth factor 10 ng/mL, basic 2 weeks after birth (21). Therefore, we subsequently ques- fibroblast growth factor 15 ng/mL) and seeded onto a 10-cm tioned whether hyperplasia in WT mice at an early age could petri dish. The medium was changed 24 hours later to remove also express NeuroD1. We found neuroblast hyperplastic dead cells. Approximately 2 days later, tumor spheres were regions in the celiac sympathetic ganglions at day 0 after recognized. These tumor spheres were digested with trypsin birth in both WT and MYCN Tg mice, as previously reported and broken down into single cells and were passed through [Fig. 1D (a and b); ref. 21]. While there was low-level expression fresh culture every 3–4 days. For the allograft tumorigenesis of NeuroD1 in immature ganglion cells in both genotypes experiment, 1 104 shRNA-treated cells mixed with 30% [Fig. 1D (a and b, white arrowheads], NeuroD1 expression was Matrigel (BD Biosciences) were subcutaneously inoculated barely detectable in neuroblast hyperplastic regions at this age

2940 Cancer Res; 71(8) April 15, 2011 Cancer Research

NeuroD1 in Neuroblastoma Tumorigenesis

Figure 1. Expression of NeuroD1 in the mouse hyperplastic ganglion and tumor. A, the celiac sympathetic ganglions from 2-week-old WT (a) and MYCN hemizygote (b) mice. Normal ganglion cells show large nuclei (arrowheads in a). Neuroblast hyperplasia appears as clusters of basophilic, small round neuroblasts (arrowheads in b). c and d, immunochemical staining of NeuroD1 in 2-week-old WT (c) and MYCN hemizygote mice (d). Arrowheads in d indicate hyperplastic regions. B, RT-PCR for NeuroD1 expression in ganglions. C, immunochemical staining of NeuroD1 in tumor. Cells strongly expressing NeuroD1 are indicated by arrowheads. D, immunochemical staining for NeuroD1 at day 0 in WT (a) and MYCN hemizygote (b) mice. Black arrowheads indicate hyperplastic regions, and white ones indicate immature ganglion cells. Scale bar, 100 mm. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

[Fig. 1D (a and b), black arrowheads]. These results indicate in NB cells. In the MYCN amplified cell line IMR32, NeuroD1 that the hyperplasia at 2 weeks in MYCN Tg mice is different expression was significantly suppressed by 2 shRNAs acting from that at day 0 in both WT and MYCN Tg mice. Thus, against human NeuroD1 that were transiently expressed via a NeuroD1 expression is switched on in hyperplastic regions in lentivirus expression system (Fig. 2A). Although there was no MYCN Tg mice during the first 2 weeks after birth. difference in cell growth between the control shRNA and NeuroD1 shRNA-treated cells, we found that NeuroD1 shRNAs NeuroD1 is involved in cell motility inhibited cell motility in a scratch assay (Fig. 2A). Furthermore, The in vivo expression profile suggests a possible involve- we found that SH-SY5Y cells, a human NB cell line without ment of NeuroD1 in NB tumorigenesis or development. To MYCN amplification, showed a high level of NeuroD1 expression examine this hypothesis, we investigated the role of NeuroD1 (Fig. 2B; Supplementary Fig. S1). In addition, NeuroD1 shRNA

www.aacrjournals.org Cancer Res; 71(8) April 15, 2011 2941

Huang et al.

effectively suppressed NeuroD1 expression (Fig. 2B) and inhibited cell motility in a scratch assay [Fig. 2C (a and b)]. In support of these data, we report that a Boyden chamber cell migration assay showed NeuroD1 shRNA-mediated inhibition [Fig. 2C (c)]. Next, we established cells in which NeuroD1 shRNA or control shRNA was stably expressed, such that we could perform rescue experiments and analyze the underlying molecular mechanism. The established NeuroD1-knockdown cells showed a significantly suppressed expression of NeuroD1 (Fig. 3A). The NeuroD1 shRNA-treated cells became round, which was strikingly different from that of the shape of parent cells (Fig. 3B). However, data obtained with regard to cell growth and cell-cycle analyses showed that cell growth was comparable between the control shRNA cells and NeuroD1 shRNA-treated cells (Supplementary Fig. S2). As expected, a scratch assay revealed that NeuroD1 shRNA-treated cells showed less motility (Fig. 3C). This inhibition was observed up to 72 hours after scratch (Supplementary Fig. S3). This phenomenon was attributed to the knockdown of NeuroD1 expression, because the shRNA-resistant NeuroD1 expression reversed the NeuroD1 shRNA-induced suppression of cell motility (Fig. 3D).

NeuroD1 downregulates Slit2 expression What is the underlying mechanism? Coincidentally, we found that Slit2 mRNA expression was induced by knockdown of NeuroD1 expression in both IMR32 cells and SH-SY5Y cells (Fig. 4A). Consistent with this finding, SH-SY5Y cells stably expressing NeuroD1 shRNA were found to express the Slit2 protein at a much higher level than cells expressing the control shRNA (Fig. 4B). It is of importance that shRNA-resistant NeuroD1 expression in the NeuroD1 shRNA-treated cells significantly suppressed Slit2 protein expression (Fig. 4B). It is known that NeuroD1 is a transcription factor that binds to the E-box in the regulatory region of target genes (22). Therefore, in the next step, we explored whether NeuroD1 was recruited to the regulatory region of the human Slit2 gene. There are 6 E-box stretches surrounding the Slit2 gene (Fig. 4C). A ChIP assay revealed that NeuroD1 directly bound to the first and second E-box but not to the other E-boxes (Fig. 4D). NeuroD1 knockdown led to the inhibition of cell motility and an increase in Slit2 expression and therefore Slit2 might possibly be involved in cell motility. To test this hypothesis, we knocked down Slit2 expression in the cells expressing Neu- roD1 shRNA. shRNAs functioning against human Slit2 efficiently suppressed Slit2 expression in NeuroD1 shRNA-treated cells (Fig. 5A). These double-knockdown cells restored cell motility (Fig. 5B and C), and this supports the hypothesis that Figure 2. Transient knockdown of NeuroD1 leads to motility inhibition NeuroD1 promotes cell motility by suppressing Slit2 expres- of IMR32 cells and SH-SY5Y cells. A, a quantitative RT-PCR showed sion. When we knocked down only Slit2 expression in SH-SY5Y significant decrease in NeuroD1 expression by 2 lentivirus-based shRNAs against NeuroD1 in IMR32 cells (left). Directional cell migration cells, cell motility was enhanced (Supplementary Fig. S4). of IMR32 cells was evaluated by wound-healing assay (right). B, RT-PCR and quantitative PCR results showing efficient knockdown NeruoD1 knockdown suppresses tumor sphere growth of NeuroD1 by the 2 shRNAs in SH-SY5Y cells. C, a and b, directional and tumor growth cell migration of SH-SY5Y cells was evaluated by wound-healing assay. c, migration of SH-SY5Y cells was evaluated by the Boyden chamber assay. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's To further examine the functional relevance of NeuroD1 to t test). GAPDH, glyceraldehyde 3-phosphate tumorigenesis, we knocked down NeuroD1 expression in dehydrogenase. tumor spheres established from NB tumor tissue derived from MYCN Tg mice. We examined 2 shRNAs against mouse

2942 Cancer Res; 71(8) April 15, 2011 Cancer Research

NeuroD1 in Neuroblastoma Tumorigenesis

Figure 3. Stable knockdown of NeuroD1 leads to motility inhibition. A, the whole-cell extract of SH-SY5Y stably expressing control shRNA (conshRNA) or NeuroD1 shRNA-2 was subjected to Western blot analysis. B, phase-contrast photographs showing the shape change induced by NeuroD1 knockdown. Insets show higher magnification. C, wound-healing assay of SH- SY5Y cells stably expressing conshRNA or NeuroD1 shRNA-2. D, shRNA-resistant human NeuroD1 restores the motility of NeuroD1 shRNA SH-SY5Y cells. Cells indicated as 1 and 2 in D were treated as follows: 1, NeuroD1 shRNA-2 and CMV- Venus plasmid; 2, NeuroD1 shRNA-2 and hNeuroD1-Venus plasmid. Scale bar, 50 mm. * P < 0.05, *** P < 0.001 (Student's t-test).

NeuroD1. They efficiently suppressed NeuroD1 expression in and secondary sphere-formation ability was impaired after tumor spheres after infection via a lentiviral system (Fig. 6A). NeuroD1 knockdown when evaluated by sphere number Control shRNA-expressing tumor spheres showed a tightly (Fig. 6C). In addition, NeuroD1 knockdown spheres were packed shape and thus the interface between cells was smaller (Fig. 6C). nearly invisible (Fig. 6A). In contrast, cell–cell contact in The NeuroD1 functions, that is, cell motility and tumor the NeuroD1-knockdown tumor spheres appeared loose sphere growth, revealed during this study further support the (Fig. 6A). Furthermore, NeuroD1 shRNAs strongly suppressed idea that NeuroD1 might be involved in NB tumor develop- the growth of tumor spheres (Fig. 6B). Effects of NeuroD1 on ment. To examine this idea, we injected 1 104 cells from sphere self-renewal ability are shown in Figure 6C. Primary tumor spheres, which had been infected with lentivirus

www.aacrjournals.org Cancer Res; 71(8) April 15, 2011 2943

Huang et al.

Figure 4. Reverse expression of NeuroD1 and Slit2. A, the RT-PCR results indicate the inverse relationship between NeuroD1 and Slit2 expression. a, IMR32 cells; b, SH-SY5Y cells. B, Western blot analysis for Slit2 and NeuroD1 expression in NeuroD1- knockdown cells. A lentivirus- based vector expressing shRNA- resistant human NeuroD1 cDNA, but not control vector, suppressed Slit2 expression in NeuroD1- knockdown cells. C, E-boxes in the promoter region of the Slit2 gene. There are 6 E-boxes (1–6) in the promoter region. ChIP 1–3, the primer location for the ChIP assay. D, for primers for the E-boxes, the DNA fragment was coprecipitated with anti-NeuroD1 antibody. GAPDH, glyceraldehyde 3- phosphate dehydrogenase. * P < 0.05, ** P < 0.01 (Student's t-test).

expressing either control shRNA or NeuroD1 shRNA, into 129/ closely associated with poor prognosis of human NB in both SVJ WT mice. Because control shRNA-treated mice died from of the probes tested. their tumors at approximately 4 weeks, the deadline was set at Furthermore, we stratified the samples by MYCN amplifica- 3 weeks in this experiment. We found that tumor sphere cells tion and age because these are important prognosis factors for expressing NeuroD1 shRNA showed significantly less tumori- NB. With regard to the relationship between NeuroD1 expres- genic ability (Fig. 6D). sion and MYCN status on survival probability of patients, we could not draw a conclusion, probably because of the limited High NeuroD1 expression is closely associated with number of patients with amplified MYCN (data not shown). On poor prognosis of human NB the contrary, when we stratified the samples by age, we Finally, we considered whether the data obtained were obtained a significant effect of the NeuroD1 expression on relevant to the clinical features of human NB. To examine this, survival of patients older than 1 year with one probe we utilized the data of the AMC cohort study (23) and (206282_at; Fig. 7C). For another probe (1556057_s_at), we analyzed the overall and relapse-free survival probabilities observed the same tendency as noted for the first probe, with regard to the expression levels of NeuroD1 (2 probes). As although larger and more balanced samples are needed to shown in Figure 7A and B, high NeuroD1 expression was obtain firm statistical results: the P values were 0.041, in

2944 Cancer Res; 71(8) April 15, 2011 Cancer Research

NeuroD1 in Neuroblastoma Tumorigenesis

knockdown led to suppression of NB cell motility and NB tumor sphere growth. Consequently, NeuroD1 knockdown suppressed in vivo tumor growth derived from tumor sphere cells. Consistent with these data, high expression of NeuroD1 was closely associated with poor prognosis of NB patients. Taken together, our data suggest that NeuroD1 plays a critical role in NB tumorigenesis. Our results concur with the report by Revet and colleagues that inhibition of cell proliferation and anchorage-indepen- dent growth of a NB cell line by the Phox2B downstream target Msx1 is accompanied by downregulation of NeuroD1 (14). However, they found no significant difference in NeuroD1 expressions in NB and ganglioneuroma (differentiated benign tumor); this finding does not necessarily fit in with our results. Therefore, further studies are needed for understanding the roles of NeuroD1, particularly in terms of neural differentiation and NB tumorigenesis. In this context, it may be important to discuss the stemness of the neural progenitors and NB cells. NeuroD1 is expressed in proliferating neuronal progenitor cells not only during embryogenesis but also during adult neurogenesis. Its loss leads to a reduction in the number of differentiated neurons, such as cerebellar granule cells, the dentate gyrus cells, inner ear sensory neurons, amacrine cells, and newborn neurons in the adult hippocampus and olfactory bulb (4–10). In adult neurogenesis, the Wnt pathway activates b-catenin, which accumulates in the nucleus where it forms an activating complex with T-cell factor/lymphoid enhancer factor (TCF/LEF). The complex binds to the Sox/LEF element located in the promoter region of the NeuroD1 gene and activates the transcription of NeuroD1. Thereafter, the neurogenesis process begins in the neural progenitor cells (9, 10). NeuroD1 may preserve the stemness of the neural progenitor cells, that is, it may promote self-renewal and differentiation. Furthermore, our findings showing its involvement in NB tumorigenesis support this idea. Interestingly, NeuroD1 was heterogeneously expressed, that is, it was not expressed in all the cells, and it was expressed in hyperplastic regions of 2- week-old MYCN Tg mice and NB tumors of MYCN Tg mice. Figure 5. Slit2 knockdown restores the motility of NeuroD1-knockdown Furthermore, NeuroD1 knockdown led to growth suppression cells. Cells were treated with the following plasmids: 1, conshRNA in vivo plasmid; 2, conshRNA and NeuroD1 shRNA-2 plasmid; 3, NeruoD1 of tumor spheres and tumors derived from tumor shRNA-2 and Slit2 shRNA-1 plasmid; 4, NeuroD1 shRNA-2 and Slit2 sphere cells. Considering that tumor-initiating cells are shRNA-2 plasmid. A, RT-PCR showing that NeuroD1-knockdown enriched in tumor spheres (24), it is conceivable that NeuroD1 upregulates Slit2 (lane 2); this Slit2 expression is sufficiently suppressed by is involved in the stemness property of NB cells. In line with Slit2 shRNA (lanes 3 and 4). B and C, wound-healing assay data show that NeuroD1 knockdown inhibits cell motility whereas double knockdown of this idea, NeuroD1 expression was induced in hyperplatic NeuroD1 and Slit2 restores the motility. ***, P < 0.001 (Student's t test). regions of MYCN Tg mice during the first 2 weeks after birth, GAPDH, glyceraldehyde 3-phosphate dehydrogenase. thus suggesting that dynamic molecular events occur in these regions and lead to NB development. Slit proteins (Slit1–Slit3) are secreted as extracellular overall survival probability, and 0.063, in relapse-free survival matrix-associated glycoproteins and function as ligands for probability (Fig. 7D). the repulsive guidance receptor Roundabout (Robo1–Robo4) family. Slit2 is known to bind with the Robo 1 receptor and Discussion regulates the migration of neurons during development (25). In the present study, we found that NeuroD1 downregulated In this study, we have demonstrated that NeuroD1 is highly Slit2 expression and promoted cell motility whereas NeuroD1 expressed in the neuroblast hyperplastic region, which is knockdown upregulated Slit2 expression and inhibited cell hypothesized to be a precancerous lesion of NB, suggesting motility. This is the first study to illustrate an inverse relation- that NeuroD1 is involved in NB tumorigenesis. Evidence in ship between NeuroD1 and Slit2. Significantly, Slit2 knock- support of this hypothesis includes the fact that NeuroD1 down in NeuroD1-knockdown cells restored cell motility.

www.aacrjournals.org Cancer Res; 71(8) April 15, 2011 2945

Huang et al.

Figure 6. NeruoD1 knockdown suppresses tumor sphere growth and tumor growth. A, tumor sphere cells derived from NB tumors of MYCN Tg mice were infected with lentivirus expressing shRNA (photographs were taken 4 days after virus infection). RT-PCR shows efficient knockdown of NeuroD1expression by shRNAs. B, tumor sphere cells were completely dissociated into single cells by trypsin treatment and were infected with lentivirus expressing NeuroD1 shRNA or control shRNA. Cells were then cultured for 4 days (images indicated by an "a"). Cells were dissociated again and were further cultured for 4 days (images indicated by a "b") to form tumor spheres so that the ability of sphere formation and sphere growth could be evaluated. C, self-renewal ability of tumor spheres. An equal number (1 104) of sphere cells were cultured for 4 days, and the sphere number was counted for each generation. Furthermore, sphere size is shown for the first generation. The results for the second generation were similar (data not shown). D, an equal number (1 104) of control or NeuroD1 shRNA-infected sphere cells were subcutaneously injected into 129/SVJ WT mice (n ¼ 5 for conshRNA and shRNA-1 treated group, n ¼ 2 for shRNA-2 treated group). The tumors generated were removed and weighed 18 days after inoculation. Scale bar, 200 mm. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (Student's t test). GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Therefore, Slit2 is, at least partly, involved in the NeuroD1- with cell migration and growth in tumor development (26–29). mediated regulation of cell motility. Cell motility is closely However, we did not observe a significant relationship linked to tumor invasion and metastasis. Indeed, in our between Slit2 expression and NB patient survival (data not allograft NB model, we can find metastasized small tumors shown). This may suggest that Slit2 does not entirely account in the ipsilateral axillary lymph nodes usually at approximately for the role of NeuroD1 in NB development, although it is an 4 weeks after sphere cell inoculation (data not shown). Evi- important player downstream from NeuroD1. dence in support of our findings includes the fact that Slit2 is In summary, we found an inverse relationship between considered a candidate tumor suppressor gene; it is frequently NeuroD1 and Slit2 expression. NeuroD1 promotes cell motility inactivated in various cancers due to hypermethylation of its and growth of tumor spheres as well as in vivo tumor growth. promoter region and allelic loss. Slit2 has a close relationship NeuroD1-mediated cell motility is at least in part due to Slit2

2946 Cancer Res; 71(8) April 15, 2011 Cancer Research

NeuroD1 in Neuroblastoma Tumorigenesis

Figure 7. Kaplan–Meier analysis of the prognosis of NB patients in relation to NeruoD1 expression. A and B, Kaplan–Meier analysis of the AMC cohort (n ¼ 88). Two probes (206282_at and 1556057_s_at) are available for NeuroD1 (A and B, respectively). C and D, effects of NeuroD1 on the survival of NB patients older than 1 year. www.aacrjournals.org Cancer Res; 71(8) April 15, 2011 2947

Huang et al.

suppression. Our findings highlight the close relation of early CSII-CMV-RfA-IRES2-Venus and control vector CSII-CMV-Venus. We also thank Tomoko Masuoka and Misako Tanase for their technical assistance. neurogenesis and NB tumorigenesis and provide candidate molecular targets for the development of NB therapy. Grant Support

Disclosure of Potential Conflicts of Interest This work was supported in part by a Grant-in-Aid for Cancer Research (20- 13) from the Ministry of Health, Labour and Welfare, Japan, to K. Kadomatsu, by No potential conflicts of interest were disclosed. Grants-in-Aid (22790311 to S. Kishida; 21790317 to Y. Murakami-Tonami) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), Japan, by a Grant-in-Aid from the Ministry of Agriculture, Forestry and Fisheries, Acknowledgments Japan, to K. Kadomatsu and A. Onishi, and by funds from the Global COE program of MEXT, Japan, to Nagoya University. The costs of publication of this article were defrayed in part by the payment We thank Motoshi Suzuki (Nagoya University) for instruction us in FACS advertisement analysis, Didier Trono from the Swiss Federal Institute of Technology (EPFL, of page charges. This article must therefore be hereby marked in Switzerland) for providing the Lentivirus Packaging and Production plasmids accordance with 18 U.S.C. Section 1734 solely to indicate this fact. psPAX2 and pMD2.G, and Hiroyuki Miyoshi from the Subteam for Manipulation of Cell Fate, RIKEN BioResource Center (Japan), and Atsushi Miyawaki from Received September 29, 2010; revised February 1, 2011; accepted February 7, RIKEN Brain Science Institute (Japan) for providing the cDNA expression vector 2011; published OnlineFirst February 24, 2011.

References 1. Naya FJ, Huang H-P, Qiu Y, Mutoh H, DeMayo FJ, Leiter AB, et al. 16. Brodeur GM. Neuroblastoma: biological insights into a clinical Diabetes, defective pancreatic morphogenesis, and abnormal enter- enigma. Nat Rev Cancer 2003;3:203–16. oendocrine differentiation in BETA2/NeuroD-deficient mice. Genes 17. Guillemot F, Lo L-C, Johnson JE, Auerbach A, Anderson DJ, Joyner Dev 1997;11:2323–34. AL. Mammalian achaete-scute homolog 1 is required for the early 2. Lee JE, Hollenberg SM, Snider L, Turner DL, Lipnick N, Weintraub H. development of olfactory and autonomic neurons. Cell 1993;75:463– Conversion of Xenopus ectoderm into neurons by NeuroD, a basic 76. helix-loop-helix protein. Science 1995;268:836–44. 18. Roybon L, Hjalt T, Stott S, Guillemot F, Li J-Y, Brundin P. Neurogenin2 3. Miyata T, Maeda T, Lee JE. NeuroD is required for differentiation of the directs granule neuroblast production and amplification while Neu- granule cells in the cerebellum and hippocampus. Genes Dev roD1 specifies neuronal fate during hippocampal neurogenesis. PLoS 1999;13:1647–52. One 2009;4:e4779. 4. Schwab MH, Bartholomae A, Heimrich B, Feldmeyer D, Druffel- 19. Shida T, Furuya M, Kishimoto T, Nikaido T, Tanizawa T, Koda K, et al. Augustin S, Goebbels S, et al. Neuronal basic helix-loop-helix proteins The expression of NeuroD and mASH1 in the gastroenteropancreatic (NEX and BETA2/Neuro D) regulate terminal granule cell differentiation neuroendocrine tumors. Mod Pathol 2008;21:1363–70. in the hippocampus. J Neurosci 2000;20:3714–24. 20. Weiss WA, Aldape K, Mohapatra G, Feuerstein BG, Bishop JM. 5. Kim WY, Fritzsch B, Serls A, Bakel LA, Huang EJ, Reichardt LF, et al. Targeted expression of MYCN causes neuroblastoma in transgenic NeuroD-null mice are deaf due to a severe loss of the inner ear sensory mice. EMBO J 1997;16:2985–95. neurons during development. Development 2001;128:417–26. 21. Hansford LM, Thomas WD, Keating JM, Burkhart CA, Peaston AE, 6. Liu M, Pereira FA, Price SD, Chu M-j, Shope C, Himes D, et al. Norris MD, et al. Mechanisms of embryonal tumor initiation: distinct Essential role of BETA2/NeuroD1 in development of the vestibular roles for MycN expression and MYCN amplification. Proc Natl Acad and auditory systems. Genes Dev 2000;14:2839–54. Sci U S A 2004;101:12664–9. 7. Morrow EM, Furukawa T, Lee JE, Cepko CL. NeuroD regulates 22. Naya FJ, Stellrecht CM, Tsai MJ. Tissue-specific regulation of the multiple functions in the developing neural retina in rodent. Develop- insulin gene by a novel basic helix-loop-helix transcription factor. ment 1999;126:23–36. Genes Dev 1995;9:1009–19. 8. Kuwabara T, Hsieh J, Muotri A, Yeo G, Warashina M, Lie DC, et al. 23. Huang S, Laoukili J, Epping MT, Koster J, Holzel€ M, Westerman BA, Wnt-mediated activation of NeuroD1 and retro-elements during adult et al. ZNF423 is critically required for retinoic acid-induced differen- neurogenesis. Nat Neurosci 2009;12:1097–105. tiation and is a marker of neuroblastoma outcome. Cancer Cell 9. Gao Z, Ure K, Ables JL, Lagace DC, Nave K-A, Goebbels S, et al. 2009;15:328–40. Neurod1 is essential for the survival and maturation of adult-born 24. Hansford LM, McKee AE, Zhang L, George RE, Gerstle JT, Thorner neurons. Nat Neurosci 2009;12:1090–2. PS, et al. Neuroblastoma cells isolated from bone marrow metastases 10. Nakagawara A, Ohira M. Comprehensive genomics linking between contain a naturally enriched tumor-initiating cell. Cancer Res neural development and cancer: neuroblastoma as a model. Cancer 2007;67:11234–43. Lett 2004;204:213–24. 25. Wong K, Ren X-R, Huang Y-Z, Xie Y, Liu G, Saito H, et al. Signal 11. Howard MJ. Mechanisms and perspectives on differentiation of transduction in neuronal migration: roles of GTPase activating pro- autonomic neurons. Dev Biol 2005;277:271–86. teins and the small GTPase Cdc42 in the Slit-Robo pathway. Cell 12. Raabe EH, Laudenslager M, Winter C, Wasserman N, Cole K, LaQua- 2001;107:209–21. glia M, et al. Prevalence and functional consequence of PHOX2B 26. Prasad A, Paruchuri V, Preet A, Latif F, Ganju RK. Slit-2 induces a mutations in neuroblastoma. Oncogene 2007;27:469–76. tumor-suppressive effect by regulating b-catenin in breast cancer 13. Reiff T, Tsarovina K, Majdazari A, Schmidt M, del Pino I, Rohrer H. cells. J Biol Chem 2008;283:26624–33. Neuroblastoma Phox2b variants stimulate proliferation and dediffer- 27. Tseng R-C, Lee S-H, Hsu H-S, Chen B-H, Tsai W-C, Tzao C, et al. Slit2 entiation of immature sympathetic neurons. J Neurosci 2010;30:905– attenuation during lung cancer progression deregulates {beta}-cate- 15. nin and e-cadherin and associates with poor prognosis. Cancer Res 14. Revet I, Huizenga G, Chan A, Koster J, Volckmann R, van Sluis P, et al. 2010;70:543–51. The MSX1 homeobox transcription factor is a downstream target of 28. Mertsch S, Schmitz N, Jeibmann A, Geng J-G, Paulus W, Senner V. PHOX2B and activates the Delta-Notch pathway in neuroblastoma. Slit2 involvement in glioma cell migration is mediated by Robo1 Exp Cell Res 2008;314:707–19. receptor. J Neurooncol 2008;87:1–7. 15. Jogi€ A, Øra I, Nilsson H, Poellinger L, Axelson H, Pahlman S. Hypoxia- 29. Kim HK, Zhang H, Li H, Wu T-T, Swisher S, He D, et al. Slit2 inhibits induced dedifferentiation in neuroblastoma cells. Cancer Lett growth and metastasis of fibrosarcoma and squamous cell carci- 2003;197:145–50. noma. Neoplasia 2008;10:1411–20.

2948 Cancer Res; 71(8) April 15, 2011 Cancer Research

The Neuronal Differentiation Factor NeuroD1 Downregulates the Neuronal Repellent Factor Slit2 Expression and Promotes Cell Motility and Tumor Formation of Neuroblastoma

Peng Huang, Satoshi Kishida, Dongliang Cao, et al.

Cancer Res 2011;71:2938-2948. Published OnlineFirst February 24, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-10-3524

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2011/02/24/0008-5472.CAN-10-3524.DC1

Cited articles This article cites 29 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/71/8/2938.full#ref-list-1

Citing articles This article has been cited by 8 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/71/8/2938.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/71/8/2938. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.