Many human medulloblastoma tumors overexpress repressor ... · medulloblastoma tumor specimens we examined overex-pressed REST/NRSF (6 strongly and 11 weakly). In contrast, adjacentnormalcerebellum

Many human medulloblastoma tumors overexpress repressorelement-1 silencing transcription (REST)/neuron-restrictivesilencer factor, which can be functionally countered byREST-VP16

Gregory N. Fuller,1 Xiaohua Su,2 Roger E. Price,3

Zvi R. Cohen,4 Frederick F. Lang,4

Raymond Sawaya,4 and Sadhan Majumder2,5

Brain Tumor Center and Departments of 1Pathology, 2MolecularGenetics, 3Imaging Physics, and 4Neurosurgery, University ofTexas M.D. Anderson Cancer Center, Houston, Texas; and5Program in Genes and Development, University of TexasGraduate School of Biomedical Sciences at Houston,Houston, Texas

AbstractMedulloblastoma, one of the most malignant pediatricbrain tumors, is believed to arise from the undifferenti-ated external granule-layer cells in the cerebellum. It is aheterogeneous cancer, and the mechanism of tumorigen-esis for the majority of types is unknown. Repressorelement-1 silencing transcription/neuron-restrictive si-lencer factor (REST/NRSF) is a transcriptional repressorthat can block transcription of a battery of neuronaldifferentiation genes by binding to a specific consensusDNA sequence present in their regulatory region.Previously, we found that some medulloblastoma celllines express REST/NRSF at high levels compared witheither neuronal progenitor cells or fully differentiatedneurons. However, it is not known if REST/NRSF isindeed overexpressed in human medulloblastoma tumorspecimens and, if so, in what frequency. Here, we did animmunohistochemical analysis of such tumor specimensusing an anti-REST antibody. We show that, among21 human medulloblastoma tumors, 17 expressed REST/NRSF (6 strongly and 11 weakly). In contrast, adjacentnormal cerebellum tissue sections and four of the tumorspecimens did not express REST/NRSF, indicating that

abnormal expression of REST/NRSF is observed in themajority of human medulloblastoma tumors. To deter-mine whether countering REST/NRSF activity blockstumorigenicity of medulloblastoma cells, especially inthe intracranial (i.c.) environment, we found that adeno-virus-mediated expression of REST-VP16, a recombinanttranscription factor that can compete with REST/NRSFand activate REST/NRSF target genes instead of repres-sing them, blocked the i.c. tumorigenic potential ofmedulloblastoma cells and inhibited growth of estab-lished tumors in nude mice. These results suggest thatREST/NRSF may serve as a therapeutic target formedulloblastoma and that forced expression of neuronaldifferentiation genes in medulloblastoma cells throughagents, such as REST-VP16, can interfere with theirtumorigenicity. [Mol Cancer Ther 2005;4(3):343–9]

IntroductionMedulloblastoma is believed to originate mostly from theundifferentiated neuroectodermal stem cells in the cerebel-lum (1–5). Although most medulloblastomas representundifferentiated phenotypes, others express markers rep-resenting various levels of differentiation in various celllineages, such as neuronal and glial cells, and undifferen-tiated medulloblastoma cells can be further differentiatedwhen exposed to various reagents (1–3, 6, 7). A fraction(f15%) of patients with sporadic medulloblastoma harborgenetic defects in the patched (PTCH) gene (8). In addition,some mutant ptc+/� mice produce medulloblastomas (9).The PTCH gene is an integral part of the hedgehog-hip-patched-smoothened pathway (10), and a mutation in anyof those genes may give rise to medulloblastoma (11).Mutations in the p53, h-catenin, and Rb pathways are alsoinvolved in this disease (3, 12–14). Other mechanismsinvolving JC virus and various other pathways regulatedby platelet-derived growth factor receptor A, insulin-likegrowth factor I receptor, or Trk receptor tyrosine kinase,etc., also have been correlated with medulloblastoma(4, 14–16). Perhaps these observations reflect the fact thatdifferent mechanisms under different genetic backgroundstrigger medulloblastoma tumorigenesis.

One challenge facing medulloblastoma research isunderstanding the mechanism or mechanisms that regulatemost of these tumors. Our previous work indicated thatseveral medulloblastoma cell lines, compared with eitherneuronal progenitor cells or fully differentiated neurons(17, 18), overexpress repressor element-1 silencing tran-scription/neuron-restrictive silencer factor (REST/NRSF).REST/NRSF is a global transcriptional repressor (19, 20)

Received 8/31/04; revised 1/13/05; accepted 1/17/05.

Grant support: National Cancer Institute grants CA81255 and CA97124and grants from Katie’s Kids for the Cure and the Goodwin family.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

Note: G.N. Fuller, X. Su, and R.E. Price contributed equally to this work.Current address of Z.R. Cohen: Department of Neurosurgery, ShebaMedical Center, Tel Hashomer 52621, Israel.

Requests for reprints: Sadhan Majumder, Department of MolecularGenetics, University of Texas M.D. Anderson Cancer Center, 1515Holcombe Boulevard, Box 11, Houston, TX 77030.Phone: 713-792-8920; Fax: 713-792-6054.E-mail: [email protected]

Copyright C 2005 American Association for Cancer Research.

Molecular Cancer Therapeutics 343

Mol Cancer Ther 2005;4(3). March 2005

on August 7, 2020. © 2005 American Association for Cancer Research.mct.aacrjournals.org Downloaded from

http://mct.aacrjournals.org/

responsible for silencing the transcription of most neuronaldifferentiation genes by binding to a 23-bp consensus DNAsequence, the repressor element-1 binding site/neuronrestrictive silencer (RE1/NRSE), which is present in theregulatory regions of these genes. REST/NRSF is expressedin most, if not all, nonneuronal cells in vivo (19, 20), but it isnot expressed at high levels in differentiated neuronsduring embryogenesis. However, later studies showed thatit is expressed in certain mature neurons in adults (19, 20),suggesting that REST/NRSF may play a complex role thatin turn may depend on its cellular and physiologicenvironment. REST/NRSF-dependent promoter repressionrequires interaction with several cellular cofactors, includ-ing Co-REST, mSin3A, and histone deacetylase complex,and it requires histone deacetylase activity (19–21).

We constructed a recombinant transcription factor,REST-VP16, by replacing the repressor domains of REST/NRSF with the activation domain of the herpes simplexvirus protein VP16 (17, 18, 22, 23), which was designed tocounter the effects of REST/NRSF. Our work showed thatin transient transfection experiments, REST-VP16 operatesthrough RE1/NRSE, competes with endogenous REST/NRSF for DNA binding, activates plasmid-encoded neuro-nal promoters in medulloblastoma cells and other mam-malian cell types, and activates cellular REST/NRSF targetgenes even in the absence of factors otherwise requiredto activate such genes. Furthermore, high-efficiency ex-pression of REST-VP16 mediated by Ad.REST-VP16 inmedulloblastoma cells countered endogenous REST/NRSF-mediated repression of neuronal promoters, stimu-lated the expression of endogenous neuronal genes, andtriggered apoptosis through the activation of caspasecascades. In addition, i.t. injection of Ad.REST-VP16 inestablished s.c. medulloblastoma tumors in nude miceinhibited tumor growth. Using immunohistochemical anal-ysis, here we show that a majority of the 21 humanmedulloblastoma tumor specimens we examined overex-pressed REST/NRSF (6 strongly and 11 weakly). In contrast,adjacent normal cerebellum tissue sections were negative forREST/NRSF expression, indicating that overexpression ofREST/NRSF is a new and major marker for humanmedulloblastoma tumors. Furthermore, we also found thatAd.REST-VP16 blocked the intracranial (i.c.) tumorigenicpotential of medulloblastoma cell lines and inhibited growthof established tumors in nude mice, suggesting thatexpression of neuronal differentiation genes in medullo-blastoma cells can interfere with the tumorigenic potential ofthe cells. Based on these experiments, we suggest thatoverexpression of REST/NRSF in neuronal cells inhibits thetranscription of multiple neuronal differentiation genes,blocks these cells at a predifferentiation stage, and contrib-utes to medulloblastoma tumorigenesis.

Materials andMethodsHistologic and Immunohistochemical Assays for

MedulloblastomaSamplesFor histologic examination, surgically excised human

brain tumor tissue and adjacent normal tissue samples fixed

in 10% buffered formalin and embedded in paraffin wereobtained from the Brain Tumor Center tissue bank at M.D.Anderson Cancer Center, stained with H&E, and examinedunder light microscope as previously described (17). Forimmunohistochemical assays, the brain sections werestained with an anti-REST antibody (a gift of Gail Mandel,State University of New York, Stony Brook, NY) essentiallyas previously described (17). In brief, the routinely pro-cessed, formalin-fixed, paraffin-embedded tissue was sec-tioned at a thickness of 4 to 5 Am. The tissue sections weredeparaffinized, incubated in 0.3% hydrogen peroxide inmethanol for 10 minutes at room temperature, washed, andtreated with 1% acetic acid in PBS for 5 minutes. Nonspecificbackground binding was blocked with normal goat serum.The slides were incubated with primary anti-REST antibodyovernight at 4jC. The secondary polyclonal antibody(biotinylated goat anti-rabbit antibody, diluted 1:100, Onco-gene Science, Cambridge, MA) was applied for 1 hour at37jC. Detection was accomplished after incubation withavidin alkaline phosphatase (DAKO Corporation, Carpin-teria, CA), diluted 1:200, for 1 hour at 37jC, followed byapplication of an alkaline phosphatase substrate (LifeTechnologies, Gaithersburg, MD) and incubation in thedark for 20 to 30 minutes. The staining intensity was gradedas negative (�), weakly positive (+), or strongly positive(++), and the number of tumor cells that were positivelystained was expressed as a percentage of the total number oftumor cells, estimated on the basis of 15 to 20 representativefields from each section.Cell Culture,Viruses, and Mouse Brain Immunohisto-

chemistryD283 (or Daoy) medulloblastoma cells were purchased

from American Type Culture Collection (Rockville, MD) andgrown as previously described (17). Replication-incompetent adenoviral vectors encoding the vector DNA(Ad), green fluorescent protein cDNA (Ad.GFP), or theREST-VP16 cDNA (Ad.REST-VP16) under the control ofcytomegalovirus promoter/enhancer elements were gen-erated and used to infect the Daoy cells as previouslydescribed (17). Deparaffinization and immunohistochemi-cal staining were done on a Ventana BenchMark XTAutostainer (Ventana, Tucson, AZ). Initially, antigenretrieval was incorporated using citrate buffer (pH 6.0) ina rice steamer for 45 minutes at 92jC. After rinsing in PBS,the sections were incubated with the primary antibody,synaptophysin (Abcam, Cambridge, MA; monoclonalmouse clone sy38), at the dilution of 1:50. The avidin-biotin-peroxidase conjugation method was used by thedetection kit, LSAB (DAKO). The immunoreaction wasvisualized using 3-amino-9-ethylcarbazole as the chromo-gen. Sections then were counterstained with Mayer’shematoxylin for visualization of nuclear detail.

Intracranial Inoculation of Cells into Mice and AssayofTumor Formation

These experiments were done following M.D. AndersonInstitutional Animal Care and Use guidelines. Daoy cellsinfected with Ad or Ad.REST-VP16 (2 � 105 cells present in5 AL PBS) were inoculated in a group of six mice for each

REST/NRSF in Medulloblastoma344




experimental condition using an implantable guide-screwsystem we described elsewhere (24). This method, whichuses a guide screw, a stylet, a modified Hamilton syringewith a 26-gauge needle, and an infusion pump, allowsdelivery of cells to a very specific location in the brain.Several days before the inoculation, the guide screws wereplaced in the dorsal cranium of mice. The animals wereanesthetized by i.p. injection of 1.2% avertin solution madein sterile PBS per gram of body weight (100% avertinsolution: 10 g tribromoethylalcohol + 10 mL tertiary amylalcohol) at a dosage of 15 to 17 AL per gram of body weight.The guide screw entry site was marked at a point 2.5 mmlateral and 1 mm anterior to the bregma, which is locateddirectly above the caudate nucleus. A small, hand-controlled twist drill was used to make the hole, and aspecially devised screwdriver was used to thread the screwinto the hole and secure it. After a recovery period of 5 daysafter the guide screws were placed, the cells wereinoculated. On the day of implantation, the cells wereharvested and resuspended in Ca2+-free and Mg2+-freePBS. The cells were kept on ice until implantation. Micewith the guide screws in place were reanesthetized asdescribed above. The tumor cell suspension was drawninto a Hamilton syringe, which was fitted with a cuff tocontrol the depth of injection. The needle of the Hamiltonsyringe was slowly lowered into the center of the guidescrew until the cuff rested on the screw surface. Up to 10mice attached to 10 syringes were positioned in a row in aHarvard Apparatus (Holliston, MA) infusion pump thatprovided continuously controlled slow injection of the cellsinto the caudate nucleus at a rate of 1 AL/min. After theentire volume of the cell suspension was injected, theneedles were manually removed and with a fine forceps,the stylets were positioned in the screw holes to close thesystem and prevent the tumor cells from leaking afterinoculation. The mice were sacrificed after 4 weeks by CO2

inhalation and their brains fixed with formalin andembedded in paraffin; 4 to 5 Am brain sections wereexamined as described above for the human medulloblas-toma brain specimens. For i.t. adenovirus injection experi-ments, i.c. implanted medulloblastoma cells were allowedto grow for 2 weeks and then each mouse was injectedthrice at the interval of 3 days with 2 � 105 plaque-formingunits of Ad.GFP or Ad.REST-VP16 present in 3 AL volumeusing the guide-screw system. The mice brains were thenexamined by magnetic resonance imaging after an addi-tional 4 weeks, sacrificed, and their brains were analyzedby histologic methods as described above. The diameter ofeach tumor was determined by measuring the average ofthe largest and the smallest diameters and this value wasused to calculate the average volume of each tumor.Because we have small number of independent samples,the P value was calculated by the Mann-Whitney Testusing SPSS statistical package (SPSS Inc., Chicago, IL).

Magnetic Resonance ImagingThe magnetic resonance images were acquired on a

clinical 1.5 T GE Signa (GE Medical Systems, Waukesha,WI) in the Small Animal Cancer Imaging Research Facility

at University of Texas M.D. Anderson Cancer Center(Houston, TX). Mice were anesthetized for imaging using1.5% to 5% isoflurane inhalation anesthesia. The i.c. screwswere removed from each of the mice before imaging. Theimages of the brains of the mice were acquired using a 3 cmquadrature volume resonator. Axial T1-, T2-, and postcon-trast T1-weighted images using gadopentetate dimeglu-mine (Magnevist-Berlex Laboratories, Wayne, NJ) as acontrast agent were acquired of the brain. The axial headimages covered a 4.0 cm square field-of-view. Data wasacquired at 256 � 192 data matrices, resulting in anisotropic in-plane resolution of f150 � 200 Am. All sliceswere 1.5 mm thick and separated by 0.5 mm gaps. The T1-weighted images were acquired with a spin-echo pulsesequence at TE/TR = 14/450 ms, whereas the T2-weightedimages were acquired with fast spin-echo pulse sequence atTE/TR = 90/5,000 ms with an echo train length of 12.

In situ Apoptosis Detection AssayApoptosis in tumor sections was determined by using

the TACS in situ apoptosis detection kit (R&DSystems, Minneapolis, MN) essentially according to themanufacturer’s instructions. Formalin-fixed, paraffin-embedded mouse brain sections were deparaffinized andthen digested with proteinase K for permeabilization. Theapoptotic DNA fragments were labeled with biotinylatednucleotides using terminal deoxynucleotidyl transferaseand the biotinylated nucleotides were then detected bystreptavidin-horseradish peroxidase conjugate and diami-nobenzidine.

ResultsHuman MedulloblastomaTumors Overexpress REST/

NRSFTo determine whether human medulloblastoma

tumors overexpressed REST/NRSF, we examined 21surgically excised, formalin-fixed, paraffin-embeddedtissue specimens organized in a tissue array, first by H&Estaining (Fig. 1A and B) and then by immunohisto-chemicalanalysis with an anti-REST antibody (Fig. 1C;Table 1). All specimens were obtained from the M.D.Anderson Brain Tumor Center tissue bank and werepreviously diagnosed by histologic examination as medul-loblastoma. Six normal cerebellar tissue samples adjacentto tumors, which were histopathologically distinct from thetumor tissue, were included in the array as negativecontrols. Medulloblastoma patient age ranged from 2 to34 years (Table 1). The morphology shown in Fig. 1A and Bshows that normal cerebellar cortex tissue was comprisedof granular-cell neurons with monotonously uniform,round nuclei and interspersed pale neuropil islands(glomeruli), whereas the medulloblastoma displayeddisheveled, densely packed tumor cells with irregular,hyperchromatic, pleomorphic nuclei, as expected for thesetumors (25). The immunohistochemical analysis (Fig. 1C)done using anti-REST/NRSF antibodies revealed a strongpositive signal with HeLa cells (HeLa cytospin), whichexpress REST/NRSF, and they served as a positive control(17). These cells were spun down, fixed with formalin, and





paraffin-embedded and processed the same way as thebrain tissue. As described above, normal cerebellar tissueadjacent to the tumor was used as a negative control.Whereas no immunoreactivity was observed in normalcerebellar tissue, REST/NRSF positivity was observed in 17of 21 medulloblastoma tumor samples, with strong signalsin 6 and weak signals in 11. The four specimens that did notshow immunoreactivity came from patients ages 2, 8, 12,and 25 years, indicating that immunopositivity is notuniversally present in younger patients. These resultsindicated that REST/NRSF is a major determinant forhuman medulloblastoma tumors.

Adenovirus-Mediated Expression of REST-VP16Blocks the IntracranialTumorigenic Potential ofMedul-loblastoma Cells in NudeMice

Several medulloblastoma cell lines (D283, Daoy, andD341) overexpress REST/NRSF, and they do not expressneuronal differentiation genes (17). These cells also form i.c.tumors in nude mice (6). Previously, we showed that therecombinant molecule REST-VP16, which can competewith REST/NRSF for binding to RE1/NRSE and canactivate neuronal differentiation genes (REST/NRSF targetgenes) instead of repressing them, could also blocktumorigenic potential of various medulloblastoma cells ats.c. locations in mice (17). Because the i.c. environment maybe critical for medulloblastoma tumorigenesis, we wantedto determine whether REST-VP16 could also block thetumorigenic potential of medulloblastoma cells in ortho-topic mouse models. For this purpose, we inoculated D283(or Daoy cells) cells infected with either Ad or Ad.REST-VP16 into the brain of six 8-week-old nude mice using theimplantable guide-screw system described in Materials andMethods. We sacrificed the mice after 5 weeks andexamined their paraffin-embedded brain sections by H&Estaining (Fig. 2). Whereas no tumor was visible in any of theAd.REST-VP16 – inoculated mice tumors, large tumorsappeared in five of six Ad-inoculated mice. Similar results

were also obtained with Daoy medulloblastoma cells.These experiments suggest that countering REST/NRSFfunction and the forced expression of neuronal differentia-tion genes in medulloblastoma cells can interfere with tumo-rigenic potential of the cells in orthotopic mouse models.

Adenovirus-Mediated Delivery of REST-VP16 InhibitsGrowth of Medulloblastoma Cells at Intracranial Loca-tions in NudeMice

To examine whether the i.t. injection of adenoviralvectors carrying REST-VP16 would inhibit the growth of

Table 1. Status of REST/NRSF immunoreactivity in humanmedulloblastoma patient samples

Patient no. Age REST

1 2 �2 9 +3 34 +4 8 �5 4 +6 15 ++7 22 ++8 20 +9 2 +

10 26 +11 11 +12 2 +13 2 ++14 12 �15 25 �16 26 ++17 20 ++18 10 +19 3 +20 7 +21 11 ++

NOTE: �, negative; +, weakly positive; ++, strongly positive.

Figure 1. Histologic analysis of normalhuman brain tissue sections (A) and medullo-blastoma tumor specimens (B). Paraffin-embedded tissue sections were processed,stained with H&E, and analyzed as describedin Materials and Methods. Primary medullo-blastoma specimens overexpress REST/NRSF(C). Immunohistochemical analysis was doneusing paraffin-embedded HeLa cytospin prep-aration, normal cerebellum tissue, and medul-loblastoma primary tumor specimens withpreimmune serum and an anti-REST antibody.HeLa cells have high levels of REST/NRSF andserved here as a positive control. Magnifiedversions of medulloblastoma specimens in thepresence of the preimmune serum and anti-REST antibody are also shown to exhibitnuclear localization of REST/NRSF.





preformed medulloblastoma i.c. tumors, uninfected D283cells (or Daoy cells) were injected i.c. in nude mice asdescribed above. After 2 weeks, one group of mice wasinjected with Ad.GFP and the other with Ad.REST-VP16.Three injections were carried out with an interval of 3days each, for a total of three injections. At the end of anadditional 4-week period, all mice were analyzed bymagnetic resonance imaging, sacrificed, and their paraf-fin-embedded brain sections were analyzed by H&E to

detect the tumor volume (Fig. 3). The Ad-REST-VP16-injected mice showed smaller tumors with a volume of2.10 mm3 (SE = 0.54) compared with mice injected withAd.GFP that showed an average tumor volume of13.57 mm3 (SE = 3.07). The P value for comparing bothsets of data is 0.002 using Mann-Whitney test. Again,similar results were obtained with Daoy cells. It isinteresting that in the tumorigenicity experiment describedin the previous section, in which cells were directly infectedwith adenovirus in vitro and then inoculated into nudemice, tumors grew in none of the mice receiving Ad.REST-VP16 – infected cells. In contrast, the i.t. injection ofAd.REST-VP16 caused inhibition of tumor growth ratherthan regression of the tumor. This suggests that i.t. injectiondelivers the virus only to a limited number of cells in thetumor, presumably causing them to undergo apoptosis(17). However, the spread of the virus among neighboringtumor cells was restricted because of the fact that these arereplication-defective viruses. Thus, tumor cells that did notreceive the virus would grow at a normal rate for thetumor.

To determine whether the delivery of REST-VP16 inpreformed medulloblastoma tumors indeed caused expres-sion of terminal neuronal differentiation genes andapoptosis, we subjected the tumor sections to immunohis-tochemical analysis with antisynaptophysin antibody(Fig. 4) and an in situ apoptosis detection assay thatmeasures apoptotic cells by detecting DNA fragmentation(Fig. 5). As shown, Ad.REST-VP16, and not Ad.GFP,injected tumors showed expression of sypatophysin anddistinct signs of apoptosis.

DiscussionOur observations described here indicated that REST/NRSF overexpression is involved in a majority of the

Figure 2. REST-VP16 blocks the tumorigenic potential of medulloblas-toma cells at i.c. locations in nude mice. Medulloblastoma cell lines (D283or Daoy) infected with Ad or Ad.REST-VP16 were inoculated i.c. into nudemice using a guide-screw system, and their brains were examinedhistologically after 4 wks, as described in Materials and Methods. Theneedle mark produced during the inoculation of cells can be seen in theREST-VP16 specimens.

Figure 3. Adenovirus-mediated delivery of REST-VP16 inhibits growth of medulloblastoma cells at i.c. locations in nude mice. Medulloblastoma cell lines(D283 or Daoy) were inoculated i.c. into nude mice using a guide-screw system and after a 2-wk period, the implanted cells were injected with Ad.GFP orAd.REST-VP16 and their brains were examined by magnetic resonance imaging (left ; T1, T1 + Gd , and T2) and by histologic methods (right ; H&E) asdescribed in Materials and Methods. The example shows the largest tumors observed for each group (arrow ). Spherical dark magnetic resonance imagingsusceptibility artifacts due to the presence of contaminating metal from the guide screws in the shape of a circle are present in some of the images.





human medulloblastoma tumors. REST/NRSF has severaldifferentially spliced isoforms (26, 27). One such isoform,REST4, functions as a dominant-negative regulator byinterfering with REST/NRSF activity (27, 28). Because theantibody used in this study was raised against the full-length REST/NRSF protein, it is not clear whether themedulloblastoma samples express the full-length REST/NRSF or any of its isoforms or both. However, wepreviously found that several human medulloblastomacell lines expressed the repressor activity of REST/NRSFand subsequently blocked its multiple target neuronaldifferentiation genes. Furthermore, countering REST/NRSF activity in those cells through REST-VP16 causedactivation of the target genes. These results suggestedthat the human medulloblastoma cells expressed therepressor activity of REST/NRSF and most likely not theREST4 isoform, which would have inhibited REST/NRSFactivity and activated neuronal differentiation genes.Similarly, the data shown in Fig. 4 also indicated thatthe i.c. injected tumor cells did not express the neuronaldifferentiation genes, but they did so only when infectedwith REST-VP16. However, experiments done with anti-bodies specific for defined domains of the REST/NRSFprotein would be needed to definitively answer thisquestion.

Here, we also found that expression of REST-VP16 inhuman medulloblastoma cells caused apoptosis andblocked i.c. tumorigenic potential of the cells in mice.Previously, we found that expression of REST-VP16 inhuman medulloblastoma cells did not convert them toneurons but rather caused apoptosis (17). The exactmechanism for this process is not known. One possibilityis that although REST-VP16 does activate neuronal differ-entiation genes, it does not block the expression of genesencoding stem cell properties and the simultaneousexpression of genes regulating both differentiation and

undifferentiation properties caused apoptosis. The secondpossibility is that the medulloblastoma cells, like most othercancer cells, harbor genetic instability and might not havecontained all the cellular machinery required to proceedthrough normal neuronal differentiation pathways. Thisseems to be a reasonable possibility because REST-VP16was found to convert neural stem cells and myoblasts to aphysiologically active neuronal phenotype (22, 23). It willbe interesting to determine the effect of REST-VP16expression in medulloblastoma tumors that do not over-express REST/NRSF. If the latter tumors are blocked inneuronal differentiation by some other mechanisms andthey do not contain the functional machinery for normaldifferentiation, those tumors might behave very similar tothe tumors studied here. In contrast, if those tumors are

Figure 4. Adenovirus-mediated delivery of REST-VP16 causes expres-sion of synaptophysin in medulloblastoma cells at i.c. locations in nudemice. Preformed medulloblastoma tumors injected with either Ad.GFP orAd.REST-VP16, as described in Fig. 3, were sectioned and subjected toimmunohistochemical analysis using antisynaptophysin antibody.

Figure 5. Adenovirus-mediated delivery of REST-VP16 causes apopto-sis of medulloblastoma cells at i.c. locations in nude mice. Preformedmedulloblastoma tumors injected with either Ad.GFP or Ad.REST-VP16,as described in Fig. 3, were sectioned and subjected to an in situapoptosis detection assay. The brown nuclear signal indicates apoptosis.

Figure 6. A role of REST/NRSF in medulloblastoma. REST-VP16 cancounter the function of REST/NRSF in human medulloblastoma cells andblock the tumorigenic potential of the cells (see text for details).





blocked in neuronal differentiation but also contain thefunctional differentiation machinery, they would likelyundergo terminal differentiation as was observed with theneural stem cells (23).

On the basis of our observations, we suggest a model(Fig. 6) in which, when normal neuronal stem cells thatmay express REST/NRSF undergo differentiation, theexpression of REST/NRSF is blocked and the cellsdifferentiate into mature neurons in response to environ-mental cues leading to cascades of activators that aretranscribed in a stage-specific manner. However, underabnormal conditions, the expression of REST/NRSF iseither maintained or reinitiated in response to abnormalenvironmental cues along the differentiation pathway,repressing the transcription of most terminal differentiationgenes (REST/NRSF target genes), capturing these still-dividing cells at a stage before full differentiation, andeventually initiating medulloblastoma tumor formation.Our hypothesis is not inconsistent with the possibilitythat REST/NRSF cooperates with other abnormallyexpressed regulators and that, after tumors are initiatedby REST/NRSF, tumor maintenance and progression arethen regulated by other genes. Furthermore, REST/NRSFmay have yet-unknown functions in this process that giverise to additional mechanisms.

Acknowledgments

We thank Belinda Rivera for her assistance with the anesthesia andpositioning of the animals during the magnetic resonance procedures,Murlidhar Tekchandani for acquiring the magnetic resonance images, JoyGumin for help with the terminal deoxynucleotidyl transferase–mediatednick end labeling assay, Lynda Corley for help with the immunohisto-chemical assays, and Bunmi Owolabi for help with the statistical analysisof the data.

References

1. Eberhart CG, Burger PC. Anaplasia and grading in medulloblastomas.Brain Pathol 2003;13:376–85.

2. Pomeroy SL, Sturla L-M. Molecular biology of medulloblastomatherapy. Peditr Neurosurg 2003;39:299–304.

3. Marino S, Vooijs M, van Der Gulden H, Jonkers J, Berns A. Induction ofmedulloblastomas in p53-null mutant mice by somatic inactivation of Rbin the external granular layer cells of the cerebellum. Genes Dev2000;14:994–1004.

4. MacDonald TJ, Brown KM, LaFleur B, et al. Expression profiling ofmedulloblastoma: PDGFRA and the RAS/MAPK pathway as therapeutictargets for metastatic disease. Nat Genet 2001;29:143–52.

5. Pomeroy SL, Tamayo P, Gaasenbeek M, et al. Prediction of centralnervous system embryonal tumour outcome based on gene expression.Nature 2002;415:436–42.

6. Li XN, Parikh S, Shu Q, et al. Phenylbutyrate and phenylacetate inducedifferentiation and inhibit proliferation in human medulloblastoma cells.Clin Cancer Res 2004;10:1150–9.

7. Wang Q, Li H, Wang X-W, Wu D-C, Chen X-Y, Liu J. Resveratrolpromotes differentiation and induces Fas-independent apoptosis of humanmedulloblastoma cells. Neurosci Lett 2003;351:83–6.

8. Xie J, Johnson RL, Zhang X, et al. Mutations of the PATCHED gene inseveral types of sporadic extracutaneous tumors. Cancer Res 1997;57:2369–72.

9. Goodrich LV, Milenkovic L, Higgins KM, Scott MP. Altered neural cellfates and medulloblastoma in mouse patched mutants. Science 1997;277:1109–13.

10. McMahon AP. More surprises in the hedgehog signaling pathway. Cell2000;100:185–8.

11. Wechsler-Reya R, Scott MP. The developmental biology of braintumors. Annu Rev Neurosci 2001;24:385–428.

12. Wetmore C, Eberhart DE, Curran T. Loss of p53 but not ARFaccelerates medulloblastoma in mice heterozygous for patched. CancerRes 2001;61:513–6.

13. Krynska B, Del Valle L, Gordon J, Otte J, Croul S, Khalili K.Identification of a novel p53 mutation in JCV-induced mouse medullo-blastoma. Virology 2000;274:65–74.

14. Lee Y, Miller HL, Jensen P, et al. A molecular fingerprint formedulloblastoma. Cancer Res 2003;63:5428–37.

15. Wang JY, Del Valle L, Gordon J, et al. Activation of the IGF-IR systemcontributes to malignant growth of human and mouse medulloblastomas.Oncogene 2001;20:3857–68.

16. Nakagawara A. Trk receptor tyrosine kinases: a bridge betweencancer and neural development. Cancer Lett 2001;169:107–14.

17. Lawinger P, Venugopal R, Guo ZS, et al. Expression of REST/NRSF isa critical phenotype of medulloblastoma cells. Nat Med 2000;6:826–31.

18. Immaneni A, Lawinger P, Zhao Z, Rastelli L, Majumder S. REST-VP16activates expression of multiple neural differentiation genes in NT2 cells.Nucleic Acids Res 2000;28:3403–10.

19. Shimojo M, Hersh LB. Regulation of the cholinergic gene locus by therepressor element-1 silencing transcription factor/neuron restrictivesilencer factor (REST/NRSF). Life Sci 2004;74:2213–25.

20. Lunyak VV, Prefontaine GG, Rosenfeld MG. REST and peace for theneuronal-specific transcriptional program. Ann N Y Acad Sci 2004;1014:110–20.

21. Ballas N, Battaglioli E, Atouf F, et al. Regulation of neuronal traits bya novel transcriptional complex. Neuron 2001;31:353–65.

22. Watanabe Y, Kameoka S, Gopalakrishnan V, et al. Conversion ofmyoblasts to physiologically active neuronal phenotype. Genes Dev2004;18:889–900.

23. Su X, Kameoka S, Lentz S, Majumder S. Activation of REST/NRSFtarget genes in neural stem cells is sufficient to cause neuronaldifferentiation. Mol Cell Biol 2004;24:8018–25.

24. Lal S, Lacroix M, Tofilon P, Fuller GN, Sawaya R, Lang FF. Animplantable guide-screw system for brain tumor studies in small animals.J Neurosurg 2002;92:326–33.

25. Fuller GN. Central nervous system tumors in pediatric neoplasia:morphology and biology. Philadelphia: Lippincott-Raven Publishers;1996.

26. Palm K, Belluardo N, Metsis M, Timmusk T. Neuronal expression ofzinc finger transcription factor REST/NRSF/XBR gene. J Neurosci1998;18:1280–96.

27. Shimojo MA, Paquette AJ, Anderson DJ, Hersh LB. Protein kinase Aregulated cholinergic gene expression on PC12 cells: REST4 silences thesilencing activity of neuron-restrictive silencer factor/REST. Mol Cell Biol1999;19:6788–95.

28. Shimojo M, Hersh LB. REST/NRSF-interacting LIM domain protein,a putative nuclear translocation receptor. Mol Cell Biol 2003;23:9025–31.





2005;4:343-349. Mol Cancer Ther Gregory N. Fuller, Xiaohua Su, Roger E. Price, et al. functionally countered by REST-VP16(REST)/neuron-restrictive silencer factor, which can berepressor element-1 silencing transcription Many human medulloblastoma tumors overexpress

Updated version

http://mct.aacrjournals.org/content/4/3/343

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/4/3/343.full#ref-list-1

This article cites 25 articles, 10 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/4/3/343.full#related-urls

This article has been cited by 8 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://mct.aacrjournals.org/content/4/3/343To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/content/4/3/343.full#ref-list-1

http://mct.aacrjournals.org/content/4/3/343.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Many human medulloblastoma tumors overexpress repressor ... · medulloblastoma tumor specimens we examined overex-pressed REST/NRSF (6 strongly and 11 weakly). In contrast, adjacentnormalcerebellum