Cancer Research Annual Meeting 2010 Genetics and Biology of Brain Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shakhova, O.
Right arrow Articles by Marino, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shakhova, O.
Right arrow Articles by Marino, S.
[Cancer Research 66, 5190-5200, May 15, 2006]
© 2006 American Association for Cancer Research

Cell, Tumor, and Stem Cell Biology

Lack of Rb and p53 Delays Cerebellar Development and Predisposes to Large Cell Anaplastic Medulloblastoma through Amplification of N-Myc and Ptch2

Olga Shakhova1, Carly Leung1, Erwin van Montfort2, Anton Berns2 and Silvia Marino1

1 Institute of Clinical Pathology, University Hospital, Zurich, Switzerland and 2 Division of Molecular Genetics and Centre of Biomedical Genetics, Netherlands Cancer Institute, Amsterdam, the Netherlands

Requests for reprints: Silvia Marino, Neuroscience Centre and Institute of Pathology, Barts and the London, Queen Mary University of London, 80 Newark Street, London E1 2ES, United Kingdom. Phone: 44-20-3246-0187; Fax: 44-20-3246-0216; E-mail: s.marino{at}qmul.ac.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Medulloblastomas are among the most common malignant brain tumors in childhood. They typically arise from neoplastic transformation of granule cell precursors in the cerebellum via deregulation of molecular pathways involved in normal cerebellar development. In a mouse model, we show here that impairment of the balance between proliferation and differentiation of granule cell precursors in the external granular layer of the developing cerebellum predisposes but is not sufficient to induce neoplastic transformation of these progenitor cells. Using array-based chromosomal comparative genomic hybridization, we show that genetic instability resulting from inactivation of the p53 pathway together with deregulation of proliferation induced by Rb loss eventually leads to neoplastic transformation of these cells by acquiring additional genetic mutations, mainly affecting N-Myc and Ptch2 genes. Moreover, we show that p53 loss influences molecular mechanisms that cannot be mimicked by the loss of either p19ARF, p21, or ATM. (Cancer Res 2006; 66(10): 5190-200)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Medulloblastomas are malignant embryonic tumors of the cerebellum and are among the most common malignancies in childhood. Modern multimodality treatment has considerably improved survival in recent years, but often it is still associated with long-term side effects.

One prevailing view is that medulloblastomas originate from actively proliferating granule cell progenitors located in the external granular layer (EGL) of the cerebellum. During early postnatal development, these progenitor cells give rise to mature granule cells through waves of proliferation, differentiation, and migration, events controlled by signaling pathways, such as Sonic Hedgehog (Shh), Wingless (WNT), and NOTCH. Deregulation of these signaling pathways is involved in medulloblastoma pathogenesis (for a recent review, see ref. 1). Indeed, mutations of genes of the Shh pathway, such as Ptch1 (2) and SUFU (3), and mutations of key players of the WNT pathway, such as APC, AXIN1 (4), and ß-catenin (5), as well as overexpression of NOTCH2 and HES1 (6) have been reported in hereditary and sporadic medulloblastoma.

The tumor suppressor Rb (retinoblastoma) regulates cell cycle progression, differentiation, and apoptosis (for recent review, see ref. 7). Absence of Rb function in the central nervous system causes cell cycle defects, such as inappropriate S-phase entry as well as apoptosis. We have recently shown that lack of Rb during cerebellar development causes prolonged proliferation, delayed differentiation, and migration of granule cell progenitors but is not sufficient to induce their neoplastic transformation (8). Only in combination with p53 loss do almost all mice develop medulloblastoma, in keeping with a model in which p53 and Rb act in synergy to promote tumor formation (9).

Additional clues to the predisposing role of p53 loss in medulloblastoma pathogenesis came from the dramatic increase of incidence and acceleration of disease in mice heterozygous for Ptch1 in a p53-null background (10) and in mice simultaneously deficient for poly(ADP-ribose) polymerase-1 and p53 (11) and for Lig4 (DNA ligase IV) and p53 (12). Interestingly, medulloblastoma arise also in mice lacking INK4c and p53, therefore suggesting that lack of this negative regulator of Rb function mimics the effects of Rb loss in the context of p53 deficiency (13).

Patients carrying germ-line mutations in the p53 gene (Li-Fraumeni syndrome) are predisposed to medulloblastoma, and alterations in the p53 pathway leading to overexpression of the protein have been implicated in up to 40% of cases of sporadic medulloblastoma (14, 15).

p53 coordinates the cellular response to stress, including DNA damage, hypoxia, and oncogenic stress, resulting in either cell cycle arrest, senescence, or apoptosis. Normal physiologic levels of p53 are low due to the negative feedback regulation by Mdm2, which binds to p53 and targets it for degradation through its E3 ubiquitin ligase activity. ATM and p19ARF are key upstream regulators of p53 that relay the response to DNA damage and oncogenic stress (for a review, see ref. 16). ATM is a major player in the cell cycle checkpoint machinery responsible for detecting DNA damage and induces growth arrest via phosphorylation leading to stabilization of the p53 protein. ATM knockout mice are highly sensitive to ionizing radiation and are tumor prone (17). Upon oncogenic stimuli, p19ARF binds to Mdm2 and inhibits its ubiquitin ligase activity, resulting in increased stability and accumulation of the p53 protein (18). Mice lacking p19ARF are also highly susceptible to spontaneous tumorigenesis (19). One of the main effectors of p53-dependent cell cycle arrest is p21, a cyclin-dependent kinase (CDK) inhibitor that can induce G1 arrest and block entry into the S-phase (20).

Here, we show that lack of Rb and p53 during granule cell development induces prolonged and ectopic proliferation as well as delayed differentiation, which is mainly counterbalanced by p53-independent apoptosis in the internal granular layer (IGL). Therefore, no overt cerebellar abnormalities are detected in adult double-mutant mice. However, small clusters of immature progenitors are retained on the outer surface of the cerebellum and are not undergoing apoptosis in the absence of p53. By means of array comparative genomic hybridization (CGH) to detect genomic alterations (21–23), we show that up-regulation of Shh pathway genes, such as N-Myc, Ptch2, and Gli2, are needed to induce progression of these lesions to medulloblastoma.

Moreover, lack of p19ARF, p21, or ATM cannot substitute for p53 loss in medulloblastoma formation in glial fibrillary acidic protein (GFAP)-Cre;RbLoxP/LoxP mice, therefore pointing at a unique role of p53 loss in our mouse model.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Generation and screening of compound mutant mice. RbLoxP/LoxP, p53LoxP/LoxP, RbLoxP/LoxPp53LoxP/LoxP, RbLoxP/LoxPp19ARF–/–, or RbLoxP/LoxPp21–/– mice were crossed with GFAP-Cre mice (line Tg2). The resulting heterozygous mice were intercrossed to obtain homozygous mutant. Genotyping was done according to published protocols (9, 19, 24).

Analysis of proliferation and apoptosis. Fifteen-day-old littermates were injected i.p. with 50 mg/kg body weight bromodeoxyuridine (BrdUrd) and killed 2 hours after injection. BrdUrd was immunohistochemically detected on sections of formalin-fixed and paraffin-embedded brains (see Histologic Analysis). Cells undergoing apoptosis were identified by terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) assay (Roche Diagnostics kit, Roche, Basel, Switzerland). Corresponding areas in the dorsal part of lobule VI or VII of the vermis of the cerebellum were chosen for counting BrdUrd-positive and TUNEL-positive cells. Ten high-power fields (x40 magnification) from at least six independent mice were counted.

Histologic analysis. Whole mouse brains were fixed for at least 12 hours in 4% buffered paraformaldehyde. Coronal slices were dehydrated through graded alcohols and embedded in paraffin. Sections of 4-µm nominal thickness were mounted on coated slides and routinely stained with H&E. For immunohistochemistry, sections were pretreated with microwave (20 minutes, 150 W) in citrate buffer (pH 6.0) for antigen retrieval followed by incubation with primary antibodies overnight at 4°C. The following antibodies were used: GFAP (rabbit polyclonal, 1:100, Dako, Glostrup, Denmark), TuJ1 recognizing neuronal class ßIII-tubulin (mouse monoclonal, 1:50, Abcam, Cambridge, United Kingdom), NeuN (monoclonal, 1:50, Chemicon, Temicula, CA), p27 (polyclonal, 1:400, Santa Cruz Biotechnology, Santa Cruz, CA), and BrdUrd (rat monoclonal, 1:200, Abcam). Secondary antibodies [Alexa 488 goat anti-mouse IgG, Alexa 647 donkey anti-mouse IgG, Alexa 546 goat anti-mouse IgG, and Alexa 546 goat anti-rabbit IgG (Molecular Probes, Eugene, OR)] were applied for 1 hour at room temperature and slides were mounted with Fluorescent Mounting Medium (Dako) to avoid bleaching. Images were captured with a Leica TCS SP confocal laser scanning microscope (Leica, Wetzlar, Germany).

Immunoblotting. For immunoblots, cerebella were homogenized and lysed in radioimmunoprecipitation assay buffer [1% NP40, 150 mmol/L NaCl, 50 mmol/L Tris-HCl (pH 8.0), 0.1% SDS, 0.5% deoxycholic acid, 2 mmol/L EDTA, and protease inhibitors (Complete Mini; Roche)] for 30 minutes at 4°C. Protein concentration was determined by BCA protein assay. Equal amounts of proteins (20 µg) were resolved on 10% SDS-polyacrylamide gel and incubated with primary antibodies: cdk2 (1:500, goat polyclonal antibodies, Santa Cruz Biotechnology), p27 and p21 (1:500, rabbit polyclonal antibodies, Santa Cruz Biotechnology), and {alpha}-tubulin (1:30,000, monoclonal antibodies, Sigma, Buchs, Switzerland). Detection was achieved with goat anti-mouse IgG peroxidase-conjugated secondary antibody (1:5,000, Zymed, South San Francisco, CA) and the enhanced chemiluminescence detection kit (Amersham, Amersham, United Kingdom).

RNA isolation and reverse transcription-PCR from paraffin sections. Six sections of 10-µm thickness were prepared from six human samples of large cell anaplastic medulloblastoma. The material was resuspended in 300 µL RNA extraction buffer [20 mmol/L Tris (pH 7.5), 20 mmol/L EDTA, 1% SDS], boiled at 95°C for 10 minutes, and centrifuged. Samples were digested with proteinase K for 48 hours at 55°C followed by homogenization with QIAshredder column (Qiagen, Hilden, Germany). RNA was extracted using Trizol reagent (Invitrogen, Basel, Switzerland) according to manufacturer's protocol. Reverse transcription-PCR (RT-PCR) was done using One-Step RT-PCR kit (Qiagen). Primers for gene amplification were as follows: Gli2 forward 5'-GCCATCAAGACCGAGAGCTC-3' and Gli2 reverse 5'-CGGCCCATGAGCAGGAATCC-3'; N-MYC forward 5'-GACCACAAGGCCCTCAGTAC-3' and N-MYC reverse 5'-GTGGATGGGAAGGCATCGTT-3'; Ptch2 forward 5'-CCATGGCTTCTCCCACAAATTC-3' and Ptch2 reverse 5'-TTGTAGCACTGTGCTGGCCTG-3'; and ß-actin forward 5'-AGCCTCGCCTTTGCCGA-3' and ß-actin reverse 5'-CTGGTGCCTGGGGCG-3'. PCR program used was as follows: 50°C for 30 minutes, 95°C for 15 minutes, and 40 cycles of 94°C for 1 minute, 57°C for 1 minute, and 72°C for 1 minute. For mouse tumors, total RNA was isolated from five samples using the RNeasy Mini kit (Qiagen). Total RNA (1 µg) was reverse transcribed using SuperScript III RT and oligo(dT) primers (Invitrogen). Primer sets were Ptch2 forward 5'-GCAGATCCATGCCTTCTCCTCCA-3' and Ptch2 reverse 5'-CAATGCCCACTGGTACTGCTCTGTC-3'; N-Myc forward 5'-CGAATTGGGCTACGGAGATGCT-3' and N-Myc reverse 5'-TTGTGCTGCTGATGGATGGG-3'; and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward 5'-TGGTCTACATGTTCCAGTATG-3' and GAPDH reverse 5'-TCCACCACCCTGTTGCTGTA-3'.

Array-based CGH. Genomic DNA was isolated using the QIAamp DNA Mini kit (Qiagen) and 5 µg DNA was digested using DpnII. Tumor and control DNA were labeled with Cy5-ULS or Cy3-ULS using the ULS labeling kit (Kreatech Biotechnology, Amsterdam, the Netherlands). Briefly, 2 µg digested DNA was mixed with 2 µL labeling solution, 1 µL Cy5-ULS or Cy3-ULS, and distilled water to a final volume of 20 µL. After a 30-minute incubation at 85°C, the mixture was cooled down to 4°C and labeled genomic DNA was purified using Kreapure columns (Kreatech Biotechnology) according to the manufacturer's instructions. Hybridization experiments were done on arrays of 2,803 murine bacterial artificial chromosome (BAC) clones spotted in duplicate as described previously (23). Hybridized slides were scanned using a Agilent G2565BA Scanner (Agilent, Palo Alto, CA) and quantified using Imagene software (Biodiscovery, Marina Del Rey, CA), and the intensity values were normalized per print block. All measurements for each BAC clone were combined, and weighed averages, errors, and confidence levels were calculated as described (23). Further data analysis was done using the Ensembl mouse genome server (v31, May 2005) based on National Center for Biotechnology Information m33 mouse assembly (freeze May 27, 2004, strain C57BL/6J).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Delayed proliferation to differentiation switch in EGL granule precursors lacking Rb and Rb/p53 during postnatal cerebellar development. Mice lacking Rb and p53 in EGL progenitors of the cerebellum develop tumors with morphologic and immunophenotypic features of human medulloblastoma with full penetrance (9). Morphologically, neoplastic cells are remarkably similar to progenitors during the phase of postnatal clonal expansion. The role of pocket protein family members (Rb, p107, and p130) in promoting the proliferation to differentiation switch of neural progenitor cells, such as retinal (25), telencephalic, and cerebellar (8) progenitors is well known. Therefore, we set out to analyze whether the lack of Rb in combination with p53 loss would result in developmental abnormalities, which may possibly evolve into early neoplastic lesions and ultimately into full-blown tumors.

To this end, we generated GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice which express Cre recombinase not only in astrocytes but also, importantly, in EGL precursor cells during their clonal expansion (9). Cerebella of mice lacking one or both genes were isolated at postnatal time points critical for granule cell development (postnatal days 1, 8, 15, and 20). No histopathologic abnormalities were observed in the cerebella of mice lacking Rb and/or p53 at postnatal days 1 and 8 (data not shown; Fig. 1A ). No significant increase in the proliferation rate of compound mutant EGL progenitors were observed based on expression analysis of the cell cycle–associated protein cdk2 (Fig. 1C) and BrdU incorporation (Fig. 1B, left). Moreover, no significant increase in apoptotic activity was observed as evaluated by TUNEL staining (Fig. 1B, right). Comparable proliferation and apoptotic rates were found in the IGL of compound mutant cerebella and control littermates (Fig. 1B, left and right).


Figure 1
View larger version (46K):
[in this window]
[in a new window]
  		Figure 1. Lack of developmental abnormalities in the cerebellum of GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice at postnatal day 8. A, H&E (x20) staining of coronal sections of control (RbLoxP/LoxP;p53LoxP/LoxP) and mutant (GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP) cerebella. PC, Purkinje cell layer; fIGL, forming IGL. B, comparable proliferation (BrdU incorporation) and apoptosis (TUNEL) of granule cells and granule cell progenitors in the IGL and EGL of GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice and control littermates. n = 6 mice per genotype. C, expression analysis of cdk2 confirms the lack of proliferation abnormalities (Western blot). 1, GFAP-Cre;p53LoxP/LoxP; 2, GFAP-Cre;RbLoxP/LoxP; 3, GFAP-Cre;p53LoxP/LoxP;RbLoxP/LoxP; 4, p53LoxP/LoxP;RbLoxP/LoxP.

 
The most striking morphologic abnormalities were observed in GFAP-Cre;RbLoxP/LoxP and GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice at postnatal day 15. H&E-stained coronal sections of the cerebellum revealed a reduced thickness of the IGL compared with wild-type littermates (Fig. 2A ). Conversely, a strikingly thicker EGL (six to seven cell layers in the mutant compared with one to two cell layers in the control) containing granule cell progenitors was noted in both single-mutant and double-mutant mice (Fig. 2A). GFAP-Cre;RbLoxP/LoxP;p53LoxP/+ mice showed a similar phenotype (data not shown). Instead, cerebella of GFAP-Cre;p53LoxP/LoxP and GFAP-Cre;RbLoxP/+;p53LoxP/LoxP mice were indistinguishable from their wild-type littermates on morphologic examination.


Figure 2
View larger version (70K):
[in this window]
[in a new window]
  		Figure 2. Granule cell abnormalities in the cerebella of GFAP-Cre;RbLoxP/LoxP and GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice at postnatal day15. A, H&E and immunohistochemical stainings for p27 and BrdU on coronal cerebellar sections of control and mutant mice (x20). Note the six to seven cell layer thick EGL in the mutant mice compared with the one to two cell layer thick EGL of the controls. B, analysis of proliferation and apoptosis of granule cells and granule cell progenitors in the IGL and EGL of mice lacking either Rb or p53 or both in cells expressing Cre under control of the GFAP promoter. BrdU-positive cells in the EGL: *, P = 0.005 (t test). BrdU-positive cells in the IGL: *, P = 0.005 (t test). TUNEL-positive cells in the EGL: *, P = 0.05 (t test). TUNEL-positive cells in the IGL: *, P = 0.05 (t test). n = 6 mice per genotype. C, Western blot analysis of expression of cdk2 and p27 confirms the proliferation abnormalities observed. 1, GFAP-Cre;p53LoxP/LoxP; 2, GFAP-Cre;RbLoxP/LoxP; 3, GFAP-Cre;p53LoxP/LoxPRbLoxP/+; 4, GFAP-Cre;p53LoxP/+RbLoxP/LoxP; 5, GFAP-Cre;p53LoxP/LoxP;RbLoxP/LoxP; 6, p53LoxP/LoxP;RbLoxP/LoxP; 7, medulloblastoma sample.

 
The mitotically active and the postmitotic fractions of EGL cells lacking Rb or Rb and p53 were identified by immunohistochemical detection of incorporated BrdU and p27 expression, respectively (Fig. 2A). The proliferation rate was assessed by enumeration of BrdU-positive cells (Fig. 2B). We found significantly higher percentage of BrdU-positive cells in Rb-deficient cells versus control and in cells lacking Rb and p53 compared with those lacking only Rb and controls (Fig. 2B). Increased expression of cdk2 and decreased expression of p27, as assessed by Western blot analysis, confirmed the prolonged proliferation of the mutant progenitor cells (Fig. 2C). Moreover, immunohistochemical analysis of differentiation markers revealed expression of ßIII-tubulin, a neuronal lineage marker known to be expressed in EGL granule cell progenitors at the beginning of their differentiation (reviewed in ref. 26) and lack of expression of NeuN, therefore showing a delayed differentiation in double-mutant EGL progenitors (Fig. 3, middle and bottom ). Indeed, in control littermates of this time point, all granule cells are postmitotic, fully differentiated, and almost completely migrated to the IGL.


Figure 3
View larger version (103K):
[in this window]
[in a new window]
  		Figure 3. Delayed proliferation to differentiation switch in granule cell progenitors in GFAP-Cre;RbLoxP/LoxP and GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice at postnatal day 15. Expression of differentiation markers of BrdU-positive granule cells in the IGL. Double staining for BrdU (green) and GFAP (red) indicates that BrdU-positive cells in the IGL of mutant mice are not all GFAP-positive glial cells or GFAP-positive stem cells. They can be identified as immature granule cells because they are NeuN negative [double staining for NeuN (red) and BrdU (green) overview, 20x; detail, 300µ 4x] and ßIII-tubulin positive [double staining for ßIII-tubulin (red) and BrdU (green) overview, 20x; detail, 300µ 4x].

 
The IGL of wild-type mice is completely formed at postnatal day 15 and contains fully differentiated granule cells, Golgi cells, astrocytes, and few neural stem cells. Only scattered BrdU-positive cells were observed, which could represent either astrocytes or possibly also neural stem cells when GFAP-positive (double labeling for BrdU and GFAP; Fig. 3, top row) or fully differentiated granule cells migrated from the EGL when NeuN positive (double labeling for BrdU and NeuN; Fig. 3, middle row). However, we observed a significant increase of BrdU-positive cells in the IGL of Rb and Rb/p53 double-mutant mice (Figs. 2B and 3). Some of these cells were astrocytes as assessed by morphology and GFAP positivity, but some were neither astrocytes (negative for GFAP) nor fully differentiated granule cells (negative for NeuN). They expressed ßIII-tubulin and could therefore be identified as immature granule cells (Fig. 3, bottom row).

p53-independent apoptotic cell death of immature granule cells ectopically proliferating in the IGL and lack of overt cerebellar abnormalities in adult GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice. Continued proliferation of EGL progenitors lacking Rb induced apoptosis as assessed by TUNEL assay done on p15 cerebella and by enumeration of positive cells (Fig. 2B, right). The apoptotic mechanism involved was partially p53 dependent, as the apoptotic rate was slightly but significantly lower in Rb- and p53-deficient EGL cells compared with cells lacking only Rb (Fig. 2B, right). Conversely, ectopically proliferating granule cells in the IGL showed massive apoptosis mainly through p53-independent mechanism as judged from the dramatic increase in the apoptotic rate observed in the double-mutant compared with the single-mutant Rb (Fig. 2B, right).

To address the question of what is the consequence of this disarray of the proliferation/differentiation switch of granule cell progenitors with consequent apoptosis, we set out to analyze the cerebella of adult GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice. Surprisingly, no overt histologic abnormalities were detected in Rb or Rb/p53 mutant cerebella at postnatal day 30 (Fig. 4A ). Morphology and immunophenotyping using differentiation markers, such as NeuN, NF200, and parvalbumin, showed terminal differentiation of neuronal subpopulations. We conclude that delay of the proliferation/differentiation switch of granule cell progenitors observed during early postnatal development did not result in permanent cerebellar defects. To exclude that this finding was due to complete depletion of mutant cells due to massive apoptosis in a context of partial penetrance of Cre expression and therefore recombination, we set out to analyze the recombination status of granule cells in cerebella of adult GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice. We show that at postnatal day 30 both recombined and unrecombined alleles are detectable by PCR analysis of DNA extracted from the whole cerebellum in equal ratio (Fig. 4B). Because the granule cells located in the IGL are by far the most numerous cell population of the cerebellum, these findings exclude the complete loss of cells lacking Rb and Rb/p53 in the adult cerebellum.


Figure 4
View larger version (61K):
[in this window]
[in a new window]
  		Figure 4. No histologic abnormalities in cerebella of GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice compared with control littermates at postnatal day 30. A, top row, control; bottom row, GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP. H&E staining (x1.2; left) and immunostainings for NeuN (x20; middle left), NF200 (x20; middle right), and parvalbumin (x20; right). B, PCR analysis of recombination for p53 and Rb constructs shows presence of recombined cells in adult cerebellum of GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP. Top, 1 to 6, DNA isolated from cerebellum of GFAP-Cre;p53LoxP/LoxP mice; 7 and 8, DNA isolated from tails of GFAP-Cre;p53LoxP/+ and GFAP-Cre;RbLoxP/+. 584 bp, floxed p53 allele (primers 10F and 10R); 612 bp, recombined p53 allele (primers 1F and 10R). Bottom, 1 to 6, DNA isolated from cerebellum of GFAP-Cre;RbLoxP/LoxP mice; 7 and 8, DNA isolated from tails of GFAP-Cre;RbLoxP/LoxP. 260 bp, recombined Rb allele (primers Rb212 and Rb18); 283 bp, floxed allele (primers Rb19E and Rb18).

 
Amplification of N-Myc and Ptc2 in medulloblastoma arising in GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice. Serial histologic sections of cerebella of GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP mice at age 1 to 2 months showed clusters of immature granule cells, possibly representing early neoplastic lesions at the outer surface of the cerebellum in the majority of animals. These ectopic cells share many features with tumor cells of fully developed medulloblastoma but are probably in an early phase of neoplastic transformation because it takes 1 to 2 additional months to give rise to overt tumors. To address the question of whether lack of Rb and p53 predispose to additional genetic changes, which are needed for the development of tumors, we performed CGH using BAC arrays. We used arrays comprising 2,803 mouse BAC clones at 1-Mb intervals; experiments were done in dye-swapped pairs of two-color hybridizations and data were processed as described (23). DNA extracted from eight medulloblastoma was compared with DNA obtained from the cerebellum of a non-Cre-expressing control mouse. Tumors showed copy alterations for various chromosomes as well as small regions of amplification (Fig. 5C ).


Figure 5
View larger version (69K):
[in this window]
[in a new window]
  		Figure 5. Amplification of Shh pathway players in medulloblastoma arising in GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP. Histologic analysis (H&E; x20) showing remarkable similarities between the murine tumors (A) and the large cell anaplastic variant of human medulloblastoma (B). Note the prominent nucleoli, the molding of the nuclei, and the high proliferation and apoptosis. C, left, genome-wide DNA copy number profile for tumor #8m, gain of part of chromosome 12, and loss of (part of) chromosomes 5, 15, 17, and 19 are visible. Ratios (log2) are plotted against BAC clones ordered by position on the genome. Red and green spots, significant gains or losses, respectively. Right, chromosome 12 CGH profile for tumor #8, clearly visible is the amplification of part of this chromosome, including N-Myc. D, RT-PCR analysis shows overexpression of Ptch2 and N-Myc in the same tumor samples showing amplification in the array CGH. t1 to t5, murine medulloblastoma samples; p8, postnatal day 8; A, adult cerebellum. E, RT-PCR analysis shows overexpression of Gli2, Ptch2, and N-Myc in two, three, and four human large cell anaplastic medulloblastoma. 1 to 6, human medulloblastoma samples; C, cerebellum of a 4-month-old child. A, adult cerebellum. Controls without RNA and without RT are presented in the last two lanes.

 
Besides several infrequent chromosomal changes in a range of chromosomes, we found frequent deletion of one copy of chromosome 13 and one copy of chromosome 7 (in 75% of the tumors) as well as loss of one copy of chromosome 19 (in 50% of the tumors). Further, amplification of a region of chromosome 12 was observed in 6 of 8 tumors, and amplification of a small region of chromosome 4 and 1 was detected in 4 of 8 and 1 of 8 tumors, respectively (Table 1 ). The amplified region on chromosome 12 contained the coding region of N-Myc. The smallest region of overlap between amplicons harboring N-Myc included only six other (putative) transcripts that have not been previously associated with tumorigenesis making N-Myc the prime candidate. The amplicon on chromosome 1 harboring Gli2 was 3 Mb, making it likely that Gli2 is the critical gene in this region. The minimal region of overlap between the amplicons found on chromosome 4 and encompassing Ptch2 was ~2 Mb, supporting the view that Ptch2 is the responsible gene in this region (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of recurrent copy number changes in chromosomal DNA regions of eight GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP medulloblastomas

 
To further confirm that the above-mentioned genes were indeed involved in tumorigenesis in our mouse model, we set out to analyze their expression level in RNA extracted from the same tumor samples. Here, we show overexpression of N-Myc and Ptch2 in the murine tumors showing amplification in the CGH analysis (Fig. 5D).

Strikingly, we find overexpression of N-Myc, Ptch2, and Gli2 also in human samples of large cell anaplastic medulloblastoma. We analyzed RNA extracted from paraffin blocks of six cases retrieved from our archives by RT-PCR and showed overexpression of N-Myc, Ptch2, and Gli2 in four, three, and two cases, respectively (Fig. 5E).

We conclude that lack of Rb and p53 in granule cell progenitors predispose to medulloblastoma through acquisition of additional mutations. At least some of these mutation are identical to those seen in the human counterpart, lending support to the notion that similar pathways are altered in both systems.

p21 and p19ARF are dispensable for p53-mediated tumor suppression in medulloblastoma pathogenesis in vivo. The CDK inhibitor p21, a key downstream target of p53, is a critical cdk2 regulator essential for proliferation control of Rb-deficient cells (24). Although Rb-deficient mouse embryo fibroblasts do not show a transformed phenotype, the additional deficiency of p21 leads to anchorage-independent growth and reduced serum dependence. In addition, p21 deficiency in mice heterozygous for Rb accelerates tumorigenesis. Because expression of p21 has been reported during early postnatal cerebellar development in both EGL and IGL (27), we set out to test whether the lack of p21 could predispose to medulloblastoma development. Therefore, we generated GFAP-Cre;RbLoxP/LoxP mice in a p21-null background (24). Double-mutant mice were kept under observation for up to age 6 months, well beyond the age at which GFAP-Cre;RbLoxP/LoxP;p53LoxP/LoxP develop medulloblastoma. Cohorts of 4 to 13 mice were monitored for survival and tumor formation (Fig. 6D ). No tumors were detected on histologic serial and level sections.


Figure 6
View larger version (15K):
[in this window]
[in a new window]
  		Figure 6. Survival curves of compound mutant mice. A, Rb/p53. Neither mice lacking Rb alone nor p53 in the granule cell progenitors develop medulloblastoma; however, when both tumor suppressor genes are inactivated, the incidence of medulloblastoma reaches 70% at 150 days. B, Rb/ATM. The lack of ATM gene induces aggressive thymic lymphomas at age ~3 months. Lack of Rb does not change the life span of these mice. No medulloblastoma were observed. C, Rb/p19ARF. Double-mutant mice develop highly aggressive pituitary carcinomas but no medulloblastoma. D, Rb/p21. Deletion of Rb and p21 genes does not result in medulloblastoma formation. The number of animals for each group is indicated in the corresponding survival curve.

 
p19ARF is a tumor suppressor gene that mediates p53 response to oncogenic stimuli. Lack of Rb or its inactivation by DNA tumor viruses (e.g., E1a) leads to up-regulation of p19ARF expression (28) and to negative regulation of p53 via interaction with Mdm2. It results in impairment of cell cycle arrest and p53-dependent apoptosis leading to loss of growth control and to rapid tumor formation. To test whether p53-mediated tumor suppression in this model is dependent on p19ARF, we generated GFAP-Cre;RbLoxP/LoxP;p19ARF–/– mice. Cohorts of 10 to 13 mice were monitored for survival and tumor formation. No tumors were detected in mice up to age 5 months (Fig. 6C), suggesting that lack of p53-dependent apoptosis is most likely not responsible for transformation of Rb–/– EGL cells in vivo. These results are in keeping with data showing that loss of p53 but not of p19ARF accelerates medulloblastoma in mice heterozygous for ptch1 (10). Interestingly, we observed a high incidence of aggressive skull base tumors that could be identified as pituitary carcinomas in Rb/p19ARF compound mutant mice, in keeping with previous reports (29). The occurrence of pituitary carcinomas in Rb+/– mice (30) and IRBP-Cre;RbLoxP/LoxP mice (31) has been reported. However, the markedly increased incidence of this tumor type in a p19ARF–/– background compared with a p53–/– background is striking and further supports the notion that p19ARF exerts also p53-independent functions.

Lack of ATM is not sufficient to induce neoplastic transformation of Rb deficient EGL progenitor cells. p53-mediated response to DNA damage depends on ATM protein kinase. Mutations in ATM cause ataxia telangiectasia, a rare autosomal recessive disease characterized by cerebellar degeneration with ataxia, immunodeficiency, sterility, high sensitivity to ionizing radiation, and genomic instability, predisposing to cancer. The most common neoplasms are thymic lymphomas and, at lesser frequency, medulloblastomas and astrocytomas (32). A mouse model with targeted deletion of ATM has provided important clues to the understanding of the role of ATM in cell cycle control. In addition, deregulated E2F1 activity on Rb inactivation has been recently shown to increase ATM levels and p53 phosphorylation (33), the latter inhibiting Mdm2-p53 interaction, thus contributing to p53 stabilization and activation.

It has become increasingly evident that ATM plays a dual role depending on the mitotic activity of the cells. In actively proliferating cells, such as thymic cells or embryonic fibroblasts, lack of ATM leads to impairment of G1 arrest on DNA damage and therefore causes unleashed proliferation and possibly neoplastic transformation (34). In postmitotic cells, such as Purkinje cells, lack of ATM causes abortive cell cycle entry and cell death (35). ATM expression has been shown high in granule cell progenitors during their late embryonic and early postnatal development and to be turned down in fully differentiated postmitotic granule neurons (36).

We reasoned that p53-dependent DNA damage repair could have been implicated in tumor suppression triggered by the cell cycle dysfunction due to Rb inactivation in granule cell progenitors. If this were the case, lack of ATM would allow neoplastic transformation of Rb-deficient cerebellar progenitors as did the lack of p53. To test this hypothesis, we generated mice lacking both Rb and ATM in granule cell progenitors (GFAP-Cre;RbLoxP/LoxP;ATM–/– mice). Because disruption of ATM results in thymic lymphomas (37), GFAP-Cre;RbLoxP/LoxP;ATM–/– double-mutant mice were sacrificed at age 4 months and a complete histologic work-up of these cerebella was done. No medulloblastoma and no early neoplastic lesions on the EGL were observed in these mice (Fig. 6B). Moreover, we set out to determine whether ATM deficiency had an effect on proliferation of Rb-deficient EGL precursor cells and analyzed the cerebella of GFAP-Cre;RbLoxP/LoxP;ATM–/– mice at postnatal day 15. No morphologic abnormalities were detected, and both proliferation and apoptotic rate, as determined by BrdU incorporation and TUNEL labeling, respectively, were comparable with that of GFAP-Cre;RbLoxP/LoxP mice (data not shown).

Thus, in our mouse model, lack of ATM is dispensable for neoplastic transformation of Rb-deficient EGL cells.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have shown here that Rb inactivation in actively proliferating granule cell progenitors leads to aberrant proliferation as well as delayed differentiation and migration. Continued proliferation of EGL progenitors lacking Rb induced apoptosis partially through a p53-dependent mechanism, as the apoptotic rate was slightly lower in Rb- and p53-deficient EGL cells compared with cells lacking only Rb. Conversely, ectopically proliferating granule cells in the IGL showed massive apoptosis mainly through p53-independent mechanism as judged from the dramatic increase in the apoptotic rate observed in the double-mutant compared with the single-mutant Rb. Our data support the notion that lack of Rb delays the switch between proliferation and differentiation in granule cell progenitors and leads to ectopic proliferation of immature granule cells as shown previously (8). Moreover, the data presented here provide new evidence that lack of Rb in EGL progenitors is sufficient to hamper their proliferation/differentiation switch and complement previous experiments showing no secondary granule cell abnormalities in mice lacking Rb in Purkinje cells (8). Rb function in EGL progenitor cells is cell autonomous and the observed abnormalities do not result from aberrant signaling by Rb-deficient Purkinje cells. Lack of p53 significantly aggravates the phenotype in both compartments. Interestingly, although phenotypically indistinguishable, immature and mitotically active granule cell progenitors undergo apoptosis through different mechanisms depending on their location. Indeed, small clusters of progenitor cells are retained on the outer surface of the cerebellum until adulthood. This implies a crucial role of p53-mediated apoptosis in this compartment and is indicative for microenvironmental factors that command proliferating cells into apoptosis when they have reached the IGL. Therefore, we do not observe overt cerebellar abnormalities in adult mutant mice. These findings point to several possible scenarios. (a) It is conceivable that only a subset of granule cell progenitors express Cre recombinase during early postnatal development, leaving the other population with intact alleles and normal levels of Rb or Rb/p53, respectively. In that case, the increase in apoptotic rate may only affect abnormally and ectopically proliferating granule cells lacking Rb or Rb/p53 and would then be responsible for the depletion of the mutant cells. The rescue of the cerebellar defect would then be due to compensation through nonrecombined wild-type granule cells. (b) An alternative possibility suggests that all granule cells of the IGL carry recombined Rb or Rb/p53, leaving only non-GFAP-expressing cells, such as Purkinje cells, Golgi cells, and oligodendrocytes, unrecombined. Additional mutations have to be assumed that either force mutant cells into apoptosis or allow them to escape from apoptosis, explaining the heterogeneity of the response and the lack of overt abnormalities in adult mutant mice.

A striking observation in the developing cerebellum was the retention of small clusters of Rb- and p53-deficient immature granule cell progenitors on the outer cerebellar surface. They were still present during adulthood long before medulloblastoma developed but were morphologically quite similar to medulloblastoma and therefore suggestive of being early precursor lesions. We also considered the possibility that some tumors might have developed from neoplastic transformation of the recently described stem cell fraction present in the IGL (38). However, because the majority of proliferating cells in the IGL of our mice are GFAP negative and early neoplastic lesions have never been observed thus far in the IGL, we consider this possibility less likely. However, we cannot formally exclude this possibility, because it is at the moment unclear whether these stem cells are targeted in the GFAP-Cre mice.

It is conceivable that p53 loss in aberrantly proliferating Rb-deficient preneoplastic lesions located in the EGL may promote accumulation of other genetic mutations to induce the malignant phenotype. In keeping, Ptch1+/– mice show similar early lesions before they develop medulloblastoma. These lesion show a specific expression profile characterized by down-regulation of genes involved in migration and differentiation, clearly different from normal granule cell progenitors and from overt tumors (39). Interestingly, this model also shows an increased tumor incidence in a p53-null background.

However, the underlying molecular mechanism of p53 function in this context remained unclear. Does p53 loss contribute to medulloblastoma pathogenesis by disrupting a specific p53-driven tumor suppressor pathway or rather by destabilizing the genome?

Here, we set out to dissect the function of p53 in this context. ARF is generally believed to transmit oncogenic stress signals to p53 (40) and previous studies on B-cell lymphomas have shown similar effects of a p53+/– and a p19ARF+/– background in accelerating c-myc-induced lymphomagenesis in mice (41). Moreover, mutually exclusive mutations of p53 and p19ARF were detected in a significant number of medulloblastoma of the large cell anaplastic variant (14), which morphologically most closely resembles the tumors arising in our mouse model. Here, we present evidence that ARF-mediated p53-dependent apoptosis does not suppress tumorigenesis in our system. This is in agreement with a previous report showing an increased incidence of medulloblastoma in p53-deficient but not in p19ARF-deficient Ptch1+/– mice (10). The medulloblastoma model of Zindy et al., where lack of p53 but not p19ARF induces medulloblastoma in mice lacking the Rb-negative regulator p18INK4c (13) also supports this notion.

Another mechanism of tumor suppression by p53 implicates the CDK inhibitor p21 as a crucial player in regulating normal and neoplastic cell growth. This model is based on the critical function of p21 in proliferation control of Rb-deficient mouse embryo fibroblasts and on its role in limiting tumor growth in Rb+/– mice, because tumorigenesis in p21–/–;Rb+/– mice is accelerated (24). The data presented here indicate that p21 plays no role in preventing neoplastic transformation of Rb-deficient granule cell progenitors, because no medulloblastoma were observed in double-mutant mice. Alternatively, activation of p53 through DNA damage might have been the critical mechanism of tumor suppression triggered by cell cycle dysfunction induced by lack of Rb. If this would have been the case, lack of Rb and ATM, the main regulator of p53 response to DNA damage, should have induced medulloblastoma in our system. As we did not observe tumors or early neoplastic lesions in Rb/ATM double-mutant cerebella, ATM does not have any effect on p53-mediated tumor suppression in these mice.

Therefore, p53 prevents medulloblastoma development in Rb-deficient EGL cells by ATM-, p19ARF-, and p21-independent mechanisms. The most likely mechanism by which p53 loss confers susceptibility to medulloblastoma development entails chromosomal instability. The extensive chromosomal imbalances in chromosome copy number seen in our array CGH analysis support this notion.

The large cell anaplastic variant of medulloblastoma is a recently described histologic subtype of medulloblastoma associated with a particularly aggressive clinical course (42, 43). These neoplasms have been recently shown to be characterized by multiple genomic alterations, the most common being amplification of N-Myc and loss of chromosome 17 (44). The experimental medulloblastomas arising in our mouse model are indistinguishable from the human large cell anaplastic variant on morphologic grounds (Fig. 5A and B). Strikingly, array CGH analysis showed amplification of N-Myc and RT-PCR analysis showed overexpression of this gene in the majority of tumors analyzed, thus providing not only phenotypic but also genotypic similarity of our mouse model to its human counterpart.

Mutations leading to overexpression of the human Ptch2 gene have been described in basal cell carcinoma (45) but not in medulloblastoma thus far. However, the overexpression of this gene has been reported in microarray-based expression analysis of medulloblastomas arising in various mouse models (46, 47). This finding was interpreted as a sign of activation of the Shh pathway in these tumors because Ptch2 has sequence and possibly functional similarity to Ptch1 and it is a downstream target of Shh (48). We show here for the first time amplification of Ptch2 at the DNA level in a significant fraction of medulloblastoma arising in our mouse model. Moreover, we show overexpression of Ptch2 in three of six human anaplastic medulloblastoma analyzed. These results might open a new perspective on the function of Ptch2 in medulloblastoma pathogenesis possibly through a dominant-negative mechanism. Alternatively, overexpression of Ptch2 (in two cases of three concomitant with overexpression of Gli2) in the human neoplasm might be interpreted as evidence of activation of the Shh pathway, therefore supporting the notion that activation of this pathway is not restricted to the desmoplastic variant of medulloblastoma. It will be interesting to test human medulloblastoma samples, especially those of the large cell anaplastic variant for amplification of Ptch2 at the DNA level.

Interestingly, we observed loss of one copy of chromosome 13 in 75% of the tumors analyzed, which implies loss of one allele of Ptch1 that is located on chromosome 13 in the mouse. However, we did not observe loss of the remaining allele of Ptch1 in these tumors on CGH array analysis, although we cannot exclude that small deletions or mutations have occurred. Our data therefore suggest that complete loss of Ptch1 is not necessary for neoplastic transformation in our mouse model, in keeping with published data (49, 50).

In conclusion, we provide a thorough characterization of the molecular mechanisms leading to medulloblastoma formation in mice with a deregulated proliferation to differentiation switch of granule cell progenitors. The neoplasms arising are indistinguishable from the human medulloblastoma of large cell anaplastic type on morphologic grounds. Here, we show that the genetic events involved in the pathogenesis of these experimental murine neoplams recapitulate the human situation. The availability of genetically defined mouse models of human cancers will have a major effect on the future development of more targeted therapeutic strategies for this highly aggressive neoplasm of childhood.


   Acknowledgments
 
Grant support: Swiss Cancer League and University of Zürich (S. Marino), Sassella-Stiftung (S. Marino and O. Shakhova), and Swiss National Science Foundation (S. Marino and C. Leung).

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 with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank H. Steinmann for excellent technical support for the histology and Dr. Simona Frigerio for helpful comments on RNA isolation from paraffin-embedded tissue.

Received 10/ 3/05. Revised 2/17/06. Accepted 2/27/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Marino S. Medulloblastoma: developmental mechanisms out of control. Trends Mol Med 2005;11:17–22.[CrossRef][Medline]
  2. Dong J, Gailani MR, Pomeroy SL, et al. Identification of PATCHED mutations in medulloblastomas by direct sequencing. Hum Mutat 2000;16:89–90.[Medline]
  3. Taylor MD, Liu L, Raffel C, et al. Mutations in SUFU predispose to medulloblastoma. Nat Genet 2002;31:306–10.[CrossRef][Medline]
  4. Baeza N, Masuoka J, Kleihues P, et al. AXIN1 mutations but not deletions in cerebellar medulloblastomas. Oncogene 2003;22:632–6.[CrossRef][Medline]
  5. Eberhart CG, Tihan T, Burger PC. Nuclear localization and mutation of ß-catenin in medulloblastomas. J Neuropathol Exp Neurol 2000;59:333–7.[Medline]
  6. Fan X, Mikolaenko I, Elhassan I, et al. Notch1 and notch2 have opposite effects on embryonal brain tumor growth. Cancer Res 2004;64:7787–93.[Abstract/Free Full Text]
  7. Lipinski MM, Jacks T. The retinoblastoma gene family in differentiation and development. Oncogene 1999;18:7873–82.[CrossRef][Medline]
  8. Marino S, Hoogervoorst D, Brandner S, et al. Rb and p107 are required for normal cerebellar development and granule cell survival but not for Purkinje cell persistence. Development 2003;130:3359–68.[Abstract/Free Full Text]
  9. Marino S, Vooijs M, van Der Gulden H, et al. Induction of medulloblastomas in p53-null mutant mice by somatic inactivation of Rb in the external granular layer cells of the cerebellum. Genes Dev 2000;14:994–1004.[Abstract/Free Full Text]
  10. Wetmore C, Eberhart DE, Curran T. Loss of p53 but not ARF accelerates medulloblastoma in mice heterozygous for patched. Cancer Res 2001;61:513–6.[Abstract/Free Full Text]
  11. Tong WM, Ohgaki H, Huang H, et al. Null mutation of DNA strand break-binding molecule poly(ADP-ribose) polymerase causes medulloblastomas in p53(–/–) mice. Am J Pathol 2003;162:343–52.[Abstract/Free Full Text]
  12. Lee Y, McKinnon PJ. DNA ligase IV suppresses medulloblastoma formation. Cancer Res 2002;62:6395–9.[Abstract/Free Full Text]
  13. Zindy F, Nilsson LM, Nguyen L, et al. Hemangiosarcomas, medulloblastomas, and other tumors in Ink4c/p53-null mice. Cancer Res 2003;63:5420–7.[Abstract/Free Full Text]
  14. Frank AJ, Hernan R, Hollander A, et al. The TP53-ARF tumor suppressor pathway is frequently disrupted in large/cell anaplastic medulloblastoma. Brain Res Mol Brain Res 2004;121:137–40.[Medline]
  15. Giordana MT, Duo D, Gasverde S, et al. MDM2 overexpression is associated with short survival in adults with medulloblastoma. Neuro-oncol 2002;4:115–22.[Abstract]
  16. Sherr CJ. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 2000;60:3689–95.[Abstract/Free Full Text]
  17. Westphal CH, Hoyes KP, Canman CE, et al. Loss of atm radiosensitizes multiple p53 null tissues. Cancer Res 1998;58:5637–9.[Abstract/Free Full Text]
  18. Honda R, Yasuda H. Association of p19(ARF) with Mdm2 inhibits ubiquitin ligase activity of Mdm2 for tumor suppressor p53. EMBO J 1999;18:22–7.[CrossRef][Medline]
  19. Kamijo T, Zindy F, Roussel MF, et al. Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell 1997;91:649–59.[CrossRef][Medline]
  20. Chen X, Bargonetti J, Prives C. p53, through p21 (WAF1/CIP1), induces cyclin D1 synthesis. Cancer Res 1995;55:4257–63.[Abstract/Free Full Text]
  21. Pinkel D, Segraves R, Sudar D, et al. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 1998;20:207–11.[CrossRef][Medline]
  22. Hodgson G, Hager JH, Volik S, et al. Genome scanning with array CGH delineates regional alterations in mouse islet carcinomas. Nat Genet 2001;29:459–64.[CrossRef][Medline]
  23. Chung YJ, Jonkers J, Kitson H, et al. A whole-genome mouse BAC microarray with 1-Mb resolution for analysis of DNA copy number changes by array comparative genomic hybridization. Genome Res 2004;14:188–96.[Abstract/Free Full Text]
  24. Brugarolas J, Bronson RT, Jacks T. p21 is a critical CDK2 regulator essential for proliferation control in Rb-deficient cells. J Cell Biol 1998;141:503–14.[Abstract/Free Full Text]
  25. Chen D, Livne-bar I, Vanderluit JL, et al. Cell-specific effects of RB or RB/p107 loss on retinal development implicate an intrinsically death-resistant cell-of-origin in retinoblastoma. Cancer Cell 2004;5:539–51.[CrossRef][Medline]
  26. Wang VY, Zoghbi HY. Genetic regulation of cerebellar development. Nat Rev Neurosci 2001;2:484–91.[CrossRef][Medline]
  27. Shambaugh GE III, Haines GK III, Koch A, et al. Immunohistochemical examination of the INK4 and Cip inhibitors in the rat neonatal cerebellum: cellular localization and the impact of protein calorie malnutrition. Brain Res 2000;855:11–22.[CrossRef][Medline]
  28. de Stanchina E, McCurrach ME, Zindy F, et al. E1A signaling to p53 involves the p19(ARF) tumor suppressor. Genes Dev 1998;12:2434–42.[Abstract/Free Full Text]
  29. Tsai KY, MacPherson D, Rubinson DA, et al. ARF mutation accelerates pituitary tumor development in Rb+/– mice. Proc Natl Acad Sci U S A 2002;99:16865–70.[Abstract/Free Full Text]
  30. Nikitin AY, Juarez-Perez MI, Li S, et al. RB-mediated suppression of spontaneous multiple neuroendocrine neoplasia and lung metastases in Rb+/– mice. Proc Natl Acad Sci U S A 1999;96:3916–21.[Abstract/Free Full Text]
  31. Vooijs M, te Riele H, van der Valk M, et al. Tumor formation in mice with somatic inactivation of the retinoblastoma gene in interphotoreceptor retinol binding protein-expressing cells. Oncogene 2002;21:4635–45.[CrossRef][Medline]
  32. Liberzon E, Avigad S, Cohen IJ, et al. ATM gene mutations are not involved in medulloblastoma in children. Cancer Genet Cytogenet 2003;146:167–9.[Medline]
  33. Berkovich E, Ginsberg, D. ATM is a target for positive regulation by E2F-1. Oncogene 2003;22:161–7.[CrossRef][Medline]
  34. Kastan MB, Zhan Q, el-Deiry WS, et al. A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 1992;71:587–97.[CrossRef][Medline]
  35. Yang XW, Zhong R, Heintz N. Granule cell specification in the developing mouse brain as defined by expression of the zinc finger transcription factor RU49. Development 1996;122:555–66.[Abstract]
  36. Oka A, Takashima S. Expression of the ataxia-telangiectasia gene (ATM) product in human cerebellar neurons during development. Neurosci Lett 1998;252:195–8.[CrossRef][Medline]
  37. Xu Y, Baltimore D. Dual roles of ATM in the cellular response to radiation and in cell growth control. Genes Dev 1996;10:2401–10.[Abstract/Free Full Text]
  38. Lee A, Kessler JD, Read TA, et al. Isolation of neural stem cells from the postnatal cerebellum. Nat Neurosci 2005;8:723–9.[CrossRef][Medline]
  39. Oliver TG, Read TA, Kessler JD, et al. Loss of patched and disruption of granule cell development in a pre-neoplastic stage of medulloblastoma. Development 2005;132:2425–39.[Abstract/Free Full Text]
  40. Sherr CJ. Tumor surveillance via the ARF-p53 pathway. Genes Dev 1998;12:2984–91.[Free Full Text]
  41. Hsu B, Marin MC, el-Naggar AK, et al. Evidence that c-myc mediated apoptosis does not require wild-type p53 during lymphomagenesis. Oncogene 1995;11:175–9.[Medline]
  42. Giangaspero F, Rigobello L, Badiali M, et al. Large-cell medulloblastomas. A distinct variant with highly aggressive behavior. Am J Surg Pathol 1992;16:687–93.[Medline]
  43. Eberhart CG, Burger PC. Anaplasia and grading in medulloblastomas. Brain Pathol 2003;13:376–85.[Medline]
  44. Reardon DA, Michalkiewicz E, Boyett JM, et al. Extensive genomic abnormalities in childhood medulloblastoma by comparative genomic hybridization. Cancer Res 1997;57:4042–7.[Abstract/Free Full Text]
  45. Zaphiropoulos PG, Unden AB, Rahnama F, et al. PTCH2, a novel human patched gene, undergoing alternative splicing and up-regulated in basal cell carcinomas. Cancer Res 1999;59:787–92.[Abstract/Free Full Text]
  46. Lee Y, Miller HL, Jensen P, et al. A molecular fingerprint for medulloblastoma. Cancer Res 2003;63:5428–37.[Abstract/Free Full Text]
  47. Hallahan AR, Pritchard JI, Hansen S, et al. The SmoA1 mouse model reveals that notch signaling is critical for the growth and survival of sonic hedgehog-induced medulloblastomas. Cancer Res 2004;64:7794–800.[Abstract/Free Full Text]
  48. Ingram WJ, Wicking CA, Grimmond SM, et al. Novel genes regulated by Sonic Hedgehog in pluripotent mesenchymal cells. Oncogene 2002;21:8196–205.[CrossRef][Medline]
  49. Wetmore C, Eberhart DE, Curran T. The normal patched allele is expressed in medulloblastomas from mice with heterozygous germ-line mutation of patched. Cancer Res 2000;60:2239–46.[Abstract/Free Full Text]
  50. Zurawel RH, Allen C, Chiappa S, et al. Analysis of PTCH/SMO/SHH pathway genes in medulloblastoma. Genes Chromosomes Cancer 2000;27:44–51.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
B. Bhatia, P. A. Northcott, D. Hambardzumyan, B. Govindarajan, D. J. Brat, J. L. Arbiser, E. C. Holland, M. D. Taylor, and A. M. Kenney
Tuberous Sclerosis Complex Suppression in Cerebellar Development and Medulloblastoma: Separate Regulation of Mammalian Target of Rapamycin Activity and p27Kip1 Localization
Cancer Res., September 15, 2009; 69(18): 7224 - 7234.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
P.-O. Frappart, Y. Lee, H. R. Russell, N. Chalhoub, Y.-D. Wang, K. E. Orii, J. Zhao, N. Kondo, S. J. Baker, and P. J. McKinnon
Recurrent genomic alterations characterize medulloblastoma arising from DNA double-strand break repair deficiency
PNAS, February 10, 2009; 106(6): 1880 - 1885.
[Abstract] [Full Text] [PDF]

Home page
 Genes Dev.Home page
J. D. Kessler, H. Hasegawa, S. N. Brun, B. A. Emmenegger, Z.-J. Yang, J. W. Dutton, F. Wang, and R. J. Wechsler-Reya
N-myc alters the fate of preneoplastic cells in a mouse model of medulloblastoma
Genes & Dev., January 15, 2009; 23(2): 157 - 170.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. Cattelani, R. Defferrari, S. Marsilio, R. Bussolari, O. Candini, F. Corradini, G. Ferrari-Amorotti, C. Guerzoni, L. Pecorari, C. Menin, et al.
Impact of a Single Nucleotide Polymorphism in the MDM2 Gene on Neuroblastoma Development and Aggressiveness: Results of a Pilot Study on 239 Patients
Clin. Cancer Res., June 1, 2008; 14(11): 3248 - 3253.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
A. B. Martinez-Cruz, M. Santos, M. F. Lara, C. Segrelles, S. Ruiz, M. Moral, C. Lorz, R. Garcia-Escudero, and J. M. Paramio
Spontaneous Squamous Cell Carcinoma Induced by the Somatic Inactivation of Retinoblastoma and Trp53 Tumor Suppressors
Cancer Res., February 1, 2008; 68(3): 683 - 692.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
K. Sasai, J. T. Romer, H. Kimura, D. E. Eberhart, D. S. Rice, and T. Curran
Medulloblastomas Derived from Cxcr6 Mutant Mice Respond to Treatment with a Smoothened Inhibitor
Cancer Res., April 15, 2007; 67(8): 3871 - 3877.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
F. Zindy, T. Uziel, O. Ayrault, C. Calabrese, M. Valentine, J. E. Rehg, R. J. Gilbertson, C. J. Sherr, and M. F. Roussel
Genetic Alterations in Mouse Medulloblastomas and Generation of Tumors De novo from Primary Cerebellar Granule Neuron Precursors
Cancer Res., March 15, 2007; 67(6): 2676 - 2684.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Shakhova, O.
Right arrow Articles by Marino, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Shakhova, O.
Right arrow Articles by Marino, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online