bansal.molecular biology cns

8/9/2019 Bansal.molecular Biology CNS

1/23

Dear

Here are the electronic proofs of your article.

You can submit your corrections online or by fax. Together with your proof corrections

you must return the Copyright Transfer Statementto complete the proof process.

Print out theproof. (If you do not already have Acrobat Reader, just download it fromhttp://www.adobe.com.)

Check the metadata sheet to make sure that the header information, especially author

names and the corresponding affiliations, are correctly shown.

Check that the text is complete and that all figures, tables and their legends are

included. Also check the accuracy of special characters, equations, and electronic

supplementary material if applicable. If necessary refer to theEdited manuscript.

The publication of inaccurate data such as dosages and units can have serious

consequences. Please take particular care that all such details are correct.

Please do not make changes that involve only matters of style. We have generally

introduced forms that follow the journals style. Substantial changes in content, e.g.,

new results, corrected values, title and authorship are not allowed without the approvalof the responsible editor. In such a case, please contact the Editorial Office and return

his/her consent together with the proof.

For online submission please insert your corrections in the online correction form.

Always indicate the line number to which the correction refers.

For fax submission, please ensure that your corrections are clearly legible. Use a fine

black pen and write the correction in the margin, not too close to the edge of the page.

The cover sheets (including the Copyright Transfer Statementand the Offprint Order

Form) can either be scanned and sent electronically or sent by fax.

If we do not receive your corrections within 48 hours, we will send you a reminder.

Please note

Your article will be published Online First approximately one week after receipt of your

corrected proofs. This is the official first publication citable with the DOI. Further changes

are, therefore, not possible.

After online publication, subscribers (personal/institutional) to this journal will have access to

the complete article via the DOI using the URL:

http://dx.doi.org/

If you would like to know when your article has been published online, take advantage of our

free alert service. For registration and further information go to: http://www.springerlink.com.

Due to the electronic nature of the procedure, the manuscript and the original figures will onlybe returned to you on special request. When you return your corrections, please inform us if you

would like to have these documents returned.

The printed version will follow in a forthcoming issue.

James T. Rutka

http://dx.doi.org/10.1007/s00381-006-0187-3
http://dx.doi.org/10.1007/s00381-006-0187-3http://dx.doi.org/10.1007/s00381-006-0187-3http://dx.doi.org/10.1007/s00381-006-0187-3http://dx.doi.org/10.1007/s00381-006-0187-3


2/23

Fax to:

From:

Re:

Authors:

I. Permission to publish

Dear

I have checked the proofs of my article and

I have no corrections. The article is ready to be published without changes.

I have a few corrections. I am enclosing the following pages:

I have made many corrections. Enclosed is the complete article.

II. Offprint order Offprint order enclosed I do not wish to order offprints

Remarks:

Date / signature _________________________________________________________________________

III. Copyright Transfer Statement (sign only if not submitted previously)

The copyright to this article is transferred to Springer-Verlag (for U.S. government employees to the extent

transferable) effective if and when the article is accepted for publication. The author warrants that his/her

contribution is original and that he/she has full power to make this grant. The author signs for and accepts

responsibility for releasing this material on behalf of any and all co-authors. The copyright transfer covers the

exclusive right to reproduce and distribute the article, including reprints, translations, photographic reproductions,

microform, electronic form (offline, online) or any other reproductions of similar nature.

An author may self-archive an author-created version of his/her article on his/her own website and his/her

institutions repository, including his/her final version; however he/she may not use the publishers PDF version

which is posted on www.springerlink.com. Furthermore, the author may only post his/her version provided

acknowledgement is given to the original source of publication and a link is inserted to the published article on

Springers website. The link must be accompanied by the following text: The original publication is available at

www.springerlink.com.The author is requested to use the appropriate DOI for the article (go to the Linking Options in the article, then

to OpenURL and use the link with the DOI). Articles disseminated via www.springerlink.com are indexed,

abstracted and referenced by many abstracting and information services, bibliographic networks, subscription

agencies, library networks, and consortia.

After submission of this agreement signed by the corresponding author, changes of authorship or in the order of

the authors listed will not be accepted by Springer.

Date / Authors signature ___________________________________________________________________

Current concepts in the molecular genetics of pediatric brain tumors:implications for emerging therapies

Child's Nervous System DOI 10.1007/s00381-006-0187-3

Tamber Bansal Liang Mainprize Salhia Northcott Taylor Rutka

+1-703-5621873

Hazel Bacolod for Springer

Hazel Bacolod for Springer


3/23

Journal:

DOI:

Offprint Order Form

Please enter my order for:

Pages 1-4 1-4 5-8 5-8 9-12 9-12 13-16 13-16 17-20 17-20 21-24 21-24 25-28 25-28 29-32 29-32

Copies EUR USD EUR USD EUR USD EUR USD EUR USD EUR USD EUR USD EUR USD

50 2 50.00 2 75.00 3 00.00 330 .00 370 .00 405 .00 43 0.00 47 5.00 5 00.00 5 50.00 525 .00 575 .00 575 .00 63 0.00 61 0.00 67 0.00

100 3 00.00 3 30.00 3 65.00 405 .00 465 .00 510 .00 52 5.00 58 0.00 6 25.00 6 85.00 655 .00 720 .00 715 .00 78 5.00 76 5.00 84 0.00

200 400.00 440.00 525.00 575.00 645.00 710.00 740.00 815.00 860.00 945.00 925.00 1,015.00 1,005.00 1,105.00 1,105.00 1,190.00

300 500.00 550.00 680.00 750.00 825.00 910.00 955.00 1,050.00 1,095.00 1,205.00 1,190.00 1,310.00 1,295.00 1,425.00 1,425.00 1,530.00

400 610.00 670.00 855.00 940.00 1,025.00 1,130.00 1,195.00 1,315.00 1,360.00 1,495.00 1,485.00 1,635.00 1,615.00 1,775.00 1,775.00 1,915.00

500 720.00 790.00 1,025.00 1,130.00 1,225.00 1,350.00 1,430.00 1,575.00 1,625.00 1,780.00 1,780.00 1,960.00 1,930.00 2,125.00 2,090.00 2,300.00

Orders will only be processed if a credit card number has been provided. For German authors, payment by direct debit is also possible.

I wish to be charged in Euro USD

Please charge my credit card

Eurocard/Access/Mastercard

American ExpressVisa/Barclaycard/Americard

Number (incl. check digits):

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Valid until: _ _ / _ _

Date / Signature: ____________________________

Send receipt to: Ship offprints to:

To determine if your journal provides free offprints, pleasecheck the journals instructions to authors.

If you do not return this order form, we assume that you

do not wish to order offprints.

If you order offprints after the issue has gone to press, costs

are much higher. Therefore, we can supply offprints only inquantities of 300 or more after this time.

For orders involving more than 500 copies, please ask the

production editor for a quotation.

Prices include surface mail postage and handling.Customers in EU countries who are not registered for VATshould add VAT at the rate applicable in their country.

VAT registration number (EU countries only):

______________________________________________

For authors resident in Germany: payment by direct

debit:I authorize Springer to debit the amount owed from

my bank account at the due time.

Account no.: __________________________________

Bank code: ___________________________________

Bank: ________________________________________

Date / Signature: ______________________________

Child's Nervous System

10.1007/s00381-006-0187-3

James T. RutkaJames T. RutkaThe Division of Neurosurgery, Suite1504, The Hospital for SickChildren, 555 University Avenue,Toronto, Ontario, 518, Canada

The Division of Neurosurgery, Suite1504, The Hospital for SickChildren, 555 University Avenue,Toronto, Ontario, 518, Canada


4/23

Metadata of the article that will be visualized in OnlineFirst

1 Article Title Current concepts in the molecular genetics of pediatric braintumors: implications for emerging therapies

2 Journal Name Child's Nervous System3

CorrespondingAuthor

Family Name Rutka4 Particle5 Given Name James T.6 Suffix7 Organization The Hospital for Sick Children8 Division The Division of Neurosurgery, Suite 15049 Address 555 University Avenue, Toronto 518, Ontario, Canada

10 Organization The University of Toronto11 Division Division of Neurosurgery, The Hospital for Sick

Children12 Address Toronto , Ontario, Canada13 Organization The University of Toronto14 Division Arthur and Sonia Labatt Brain Tumor Research

Centre, The Hospital for Sick Children15 Address Toronto , Ontario, Canada16 e-mail [email protected]

Author

Family Name Tamber18 Particle19 Given Name Mandeep S.20 Suffix21 Organization The University of Toronto22 Division Division of Neurosurgery, The Hospital for Sick


Centre, The Hospital for Sick Children26 Address Toronto , Ontario, Canada27 e-mail28

Author

Family Name Bansal29 Particle30 Given Name Krishan31 Suffix32 Organization The University of Toronto33 Division Division of Neurosurgery, The Hospital for Sick

Children34 Address Toronto , Ontario, Canada

Page 1 of 3Testsite

8/16/2006file://D:\programs\Metadata\temp\hcn60187.htm


5/23

35 Organization The University of Toronto36 Division Arthur and Sonia Labatt Brain Tumor Research


Author

Family Name Liang40 Particle41 Given Name Muh-Lii42 Suffix43 Organization The University of Toronto44 Division Division of Neurosurgery, The Hospital for Sick



Author

Family Name Mainprize51 Particle52 Given Name Todd G.53 Suffix54 Organization The University of Toronto55 Division Division of Neurosurgery, The Hospital for Sick



Author

Family Name Salhia62 Particle63 Given Name Bodour64 Suffix65 Organization The University of Toronto66 Division Division of Neurosurgery, The Hospital for Sick


Centre, The Hospital for Sick Children70 Address Toronto , Ontario, Canada71

Page 2 of 3Testsite



6/23

e-mail72

Author

Family Name Northcott73 Particle74 Given Name Paul75 Suffix76 Organization The University of Toronto77 Division Division of Neurosurgery, The Hospital for Sick



Author

Family Name Taylor84 Particle85 Given Name Michael86 Suffix87 Organization The University of Toronto88 Division Division of Neurosurgery, The Hospital for Sick



ScheduleReceived 14 March 2006

95 Revised96 Accepted97 Abstract Background: The revolution in molecular biology that has taken place over the

past 2 decades has provided researchers with new and powerful tools fordetailed study of the molecular mechanisms giving rise to the spectrum ofpediatric brain tumors. Application of these tools has greatly advanced ourunderstanding of the molecular pathogenesis of these lesions.Review: After familiarizing readers with some promising new techniques in thefield of oncogenomics, this review will present the current state of knowledge as

it pertains to the molecular biology of pediatric brain neoplasms. Along the way,we hope to highlight specific instances where the detailed mechanisticknowledge acquired thus far may be exploited for therapeutic advantage.

98 Keywordsseparated by ' - ' Brain neoplasms - Molecular biology - Pediatrics - Review - Therapeutics

99 Foot noteinformation

Page 3 of 3Testsite



7/23

UNCORREC

TED PR

OOF

1

2

3 INVITED PAPER

4 Current concepts in the molecular genetics of pediatric

5 brain tumors: implications for emerging therapies

6 Mandeep S. Tamber & Krishan Bansal & Muh-Lii Liang &

7 Todd G. Mainprize & Bodour Salhia & Paul Northcott &

8 Michael Taylor & James T. Rutka

9 Received: 14 March 2006

10 # Springer-Verlag 2006

13 Abstract14 Background The revolution in molecular biology that has

15 taken place over the past 2 decades has provided research-

16 ers with new and powerful tools for detailed study of the

17 molecular mechanisms giving rise to the spectrum of

18 pediatric brain tumors. Application of these tools has

19 greatly advanced our understanding of the molecular

20 pathogenesis of these lesions.

21 Review After familiarizing readers with some promising

22 new techniques in the field of oncogenomics, this review

23 will present the current state of knowledge as it pertains to

24 the molecular biology of pediatric brain neoplasms. Along

25 the way, we hope to highlight specific instances where the26 detailed mechanistic knowledge acquired thus far may be

27 exploited for therapeutic advantage.

28 Keywords Brain neoplasms . Molecular biology .

29 Pediatrics . Review . Therapeutics

3Introduction

3In the past 2 decades, the textbooks on the molecular

3 biology of pediatric brain tumors have been rewritten

3several times. This has been principally due to the

3tremendous advances that have been made in our under-

3standing of these tumors from a basic science standpoint,

3a dv an ce s which a re in large p art d ue to the n ew

3technologies that have come forward to query the genetic

3underpinnings of these lesions. Suffice it to say that we are

3now poised, greater than ever before, to utilize the

4information that has been garnered toward tangible

4improvements in the way children with brain tumors are4treated.

4In this review, we will summarize the salient molecular

4genetic changes that characterize pediatric astrocytomas,

4 both low- and high-grade lesions, ependymoma, medullo-

4 blastoma and primitive neuroectodermal tumors, atypical

4teratoid rhabdoid tumor (AT/RT), germ cell tumors (GCTs),

4and sarcomas. During the review of each tumor type, we

4will attempt to provide clues as to how detailed knowledge

5of the molecular pathogenesis of these lesions may be

5harnessed toward translational research opportunities

5designed to improve patient outcomes.

5New techniques in the field of oncogenomics

5In the past 30 years, advanced cancer cytogenetics have

5been applied to numerous cancer subtypes, including many

5of the pediatric brain tumors. For additional details

5regarding these techniques, the interested reader is referred

5to several excellent review articles on this topic [ 136, 159].

5Here we will describe array comparative genomic hybrid-

Childs Nerv Syst

DOI 10.1007/s00381-006-0187-3

M. S. Tamber: K. Bansal : M.-L. Liang : T. G. Mainprize :

B. Salhia: P. Northcott: M. Taylor: J. T. Rutka

Division of Neurosurgery, The Hospital for Sick Children,

The University of Toronto,

Toronto, Ontario, Canada

M. S. Tamber: K. Bansal : M.-L. Liang : T. G. Mainprize :

B. Salhia: P. Northcott: M. Taylor: J. T. Rutka

Arthur and Sonia Labatt Brain Tumor Research Centre,

The Hospital for Sick Children, The University of Toronto,

Toronto, Ontario, Canada

J. T. Rutka (*)

The Division of Neurosurgery, Suite 1504,

The Hospital for Sick Children,

555 University Avenue,

Toronto, Ontario 518, Canada

e-mail: [email protected]

JrnlID 381_ArtID 187_Proof# 1 - 15/08/2006


8/23

UNCORREC

TED PR

OOF

60 ization (CGH), single nucleotide polymorphism (SNP)

61 arrays, microarray analysis of gene expression, and exon

62 resequencing as new techniques which have been and will

63 be applied to the study of pediatric brain tumors (Table 1).

64 Array comparative genomic hybridization

65 Array comparative genomic hybridization is similar to66 CGH, but instead of hybridizing the labeled probes to

67 metaphase chromosomes, the probes are hybridized to

68 microarray slides in which thousands of fragments of

69 DNA that span the genome have been arrayed out [29,

70 70, 114, 115]. The array CGH chip can be spotted with

71 DNA composed of genomic clones, cDNA sets, or

72 oligonucleotide probes depending on the array design. A

73 major advantage of array CGH over traditional CGH is the

74 significantly higher resolution offered by the array plat-

75 form, allowing investigators to identify smaller areas of

76 aberration. For example, bacterial artificial chromosome

77 (BAC) arrays consisting of greater than 30,000 BAC clones78 representing the entire human genome are now widely used

79 in array CGH, permitting whole-genome copy number

80 analysis at resolutions of less than 1 Mb, compared with the

81 much lower resolution achieved using traditional CGH

82 (~5 Mb) [72, 83].

8The highest resolution for array CGH is now provided

8 by genome tiling arrays which interrogate entire genomes

8with evenly spaced oligonucleotide probes [8, 19]. The

8increased resolution offered by modern array CGH plat-

8forms will allow for the incredibly fine mapping of focal

8amplifications and deletions not possible using previous

8generation arrays. Weaknesses of array CGH include an

9inability to detect structural anomalies such as reciprocal9translocations and the inability to detect copy-number-

9neutral loss of heterozygosity (LOH) events.

9Array CGH technology has recently been applied by a

9number of groups studying DNA copy number changes in

9pediatric brain tumors [24, 68, 99, 135, 161, 164]. Using a

9combination of array CGH and oligonucleotide gene

9expression arrays (discussed below) in a study of ependy-

9moma, Taylor et al. recently reported that histologically

9similar anatomical subtypes of ependymoma could be

1distinguished based upon their distinct patterns of copy-

1number-driven gene expression [161]. These authors also

1used array CGH and expression microarrays to identify a1number of genes that are both amplified and overexpressed

1in ependymoma (i.e., oncogenes). The results of this study

1emphasize the power of array CGH in the identification of

1genomic aberrations in cancer and demonstrate its utility

1when combined with complementary technical platforms

1such as gene expression arrays.

t1.1 Table 1 Summary of molecular and cytogenetic techniques discussed in the text

Technique CGH SKY Array CGH SNP array Expression

array

Exon

resequencingt1.2

Primary

application

Mapping regions

of amplification

and deletion

Identification of

structural

aberrations

Copy number

analysis

Copy number

analysis; LOH

mapping

Gene

expression

profiling

Mutation

detectiont1.3

Platform type Fluorescence

microscopy

Fluorescence

microscopy

Microarray:

BAC,

oligonucleotide

Microarray:

oligonucleotide

Microarray:

oligonucleotide

Dideoxy

sequencingt1.4

Starting

material

Genomic DNA Metaphase

chromosomes from

cultured cells

Genomic DNA Genomic DNA Total RNA or

mRNA

Genomic DNAt1.5

Advantages Assay only

requires genomic

DNA

Capacity to detect

balanced

translocations

Good resolution;

whole-genome

screening

High

resolution;

whole-genome

screening

Whole

transcriptome

screening

possible

Highest

resolution;

no restriction

on choice of

genes

t1.6

Disadvantages Poor resolution;

unable to detect

balanced

translocations

Prone to erroneous

interpretations and

classification errors

LOH mapping

not possible

Data analysis

intensive

Data analysis

and

interpretation

issues

Limited to user-

defined genes

and gene

families;

expensive

t1.7

t1.8 Listed are some of the current platforms employed in cancer genetics to identify novel disease-causing genes, highlighting the type of platform

used, biological material required, as well as primary strengths and weaknesses of each technique. Detailed information for each assay type,

including relevant clinical examples, is described in the text.

CGH comparative genomic hybridization, SKY spectral karyotyping, SNP single nucleotide polymorphism, LOH loss of heterozygosity,

BAC bacterial artificial chromosome

Childs Nerv Syst



9/23

UNCORREC

TED PR

OOF

109 Single nucleotide polymorphism arrays

110 Single nucleotide polymorphisms represent the most com-

111 mon form of genetic variation in the human genome [2, 62,

112 137, 169]. SNPs are naturally occurring variations in DNA

113 sequence that occur regularly throughout the genome, with

114 an average frequency of one SNP for every 200 bases of

115 DNA. Recently, array-based technologies that use SNPs as116 a means of investigating both DNA copy number and SNP

117 genotype have become commercially available [67, 79, 95,

118 96]. Unlike array CGH which involves simultaneous

119 cohybridization of both a test sample and a reference

120 sample to the microarray, SNP arrays hybridize only a test

121 sample, with DNA copy number calculated by comparison

122 of signal intensity data obtained from a test sample to data

1from control samples hybridized to the same array type. In

1SNP array analysis, DNA from a test sample is first

1digested with a specific restriction enzyme (e.g., Hind or

1Xba) and then ligated to an adaptor which permits

1amplification of the digested fragments. Amplified DNA

1is subsequently fragmented and labeled before hybridiza-

1tion to a specific SNP array (e.g. 50K Hind or 50K Xba

1array) (Fig. 1).1The arrays consist of 25-mer oligonucleotide probes

1that are designed to genotype the test sample for either of

1two alleles of a given SNP (arbitrarily designated A and B

1alleles). Each SNP is represented by a set of perfect match

1A and perfect match B probes, as well as an equal number

1of mismatch probes for both alleles. For any given SNP,

1the possible genotypes for a test sample include AA, BB,

Fig. 1 Analysis of LOH and chromosomal copy number using SNP

arrays. Data obtained from SNP array experiments can be analyzed

visually to identify instances of LOH and copy number gains and

losses. a Inferred LOH view of a medulloblastoma cell line exhibiting

genetic loss on chromosome 17p. The middle panel depicts inferred

regions of LOH in blue and those retaining heterozygosity in yellow.

The significance curve shown on the right demonstrates a high

probability of LOH on 17p for this sample. b Copy number view of a

medulloblastoma cell line with high-level genomic amplifications on

chromosome 8q24. Copy number is displayed in the middle panel

using a user-defined scale, with gains and losses represented by darker

and lighter color intensities, respectively. The right panel clearly

identifies the 8q24 amplifications in this sample as sharp peaks in copy

number, with values dramatically exceeding the threshold value of 2 at

these loci

Childs Nerv Syst



10/23

UNCORREC

TED PR

OOF

138 or AB, with genotype determined by the pattern of perfect

139 match and mismatch signal intensities obtained for that

140 SNP. Accuracy of SNP genotyping relies on the availabil-

141 ity of a highly purified DNA sample; analysis of tumor

142 tissue may therefore be problematic due to the presence of

143 contaminating normal cells. Modern array platforms

144 interrogate from 100,000 to 500,000 SNPs, resulting in

145 high-resolution genome coverage with median SNP146 spacing of ~8.5 and ~2.5 kb, respectively. This resolution

147 is more than 1,000 times better than classic CGH. As a

148 result of the high resolution and sensitivity associated

149 with SNP arrays, it is possible to identify microamplifi-

150 cations and microdeletions present in cancer samples, as

151 well as larger aberrations including aneuploidy. A major

152 advantage of SNP array technology over other platforms

153 used in cancer genetics is the simultaneous acquisition of

154 both copy number and genotype information. Whereas

155 copy number data are critical for mapping regions of

156 amplification and homozygous deletion in tumor samples,

157 genotyping information can identify loci which have158 undergone loss of heterozygosity without a decrease in

159 copy number (copy-number-neutral LOH). Although

160 SNP-array-based approaches have not yet been reported

161 for pediatric brain tumors, they have proven to be a

162 valuable tool when applied to studies of other tumor types

163 [49, 65, 140, 162, 176, 179, 180].

164 Microarray analysis of gene expression

165 In recent years, microarrays designed to study global gene

166 expression profiles have become an indispensable tool for

167 cancer researchers [16, 127, 142]. Significant improve-

168 ments with respect to array design, in combination with a

169 more comprehensive knowledge of the human transcrip-

170 tome, has resulted in the evolution of arrays which allow

171 for evaluation of the expression of every known human

172 gene [47, 60, 87, 88, 141]. In addition, new-generation

173 high-resolution arrays have recently been developed,

174 including platforms which interrogate every annotated

175 human exon. Now, genome tiling arrays capable of

176 measuring gene expression every 25 base pairs across

177 entire genome have been developed in order to measure

178 expression from cryptic transcriptional units [21, 75, 78].

179 This technology has found expression of countless tran-

180 scripts that previously lacked annotation, many from so-

181 called gene deserts.

182 Conventional arrays for gene expression analysis consist

183 of 25-mer oligonucleotide probes that are complementary in

184 sequence to their target, with multiple perfect match and

185 mismatch probes existing for each interrogated target and

186 multiple probe sets present for each gene represented on the

187 array. Total RNA from a tumor sample is first reverse-

188 transcribed using a T7-oligo(dT) promoter primer to

1generate first-strand cDNA. First-strand cDNA is then

1converted to double-stranded cDNA which serves as a

1template for the synthesis of complementary RNA (cRNA)

1in an in vitro transcription (IVT) reaction. IVT is carried out

1by T7 RNA polymerase in the presence of biotin-labeled

1ribonucleotides. The biotinylated cRNA is subsequently

1purified, fragmented, and hybridized to an expression array.

1Hybridized arrays are then stained, washed, and scanned to1determine the signal intensity for each probe on the array,

1with the signal intensity being proportional to the amount

1of bound target. Data obtained from such a gene expression

2array can then be used to identify the gene expression

2 profile characteristic of a particular tumor [100]. Gene

2expression arrays have proven useful in the identification of

2aberrantly expressed genes and gene families in pediatric

2 brain tumors, with some of these candidates potentially

2representing important prognostic markers and/or serving

2as targets for future therapeutics [44, 51, 92, 119].

2Exon resequencing

2Advances in DNA sequencing technologies, along with the

2availability of comprehensive genomic databases, have

2recently led to a number of large-scale resequencing efforts

2aimed at identifying novel mutations in groups or families

2of preselected candidate genes [7, 27, 28, 122, 145, 146,

2170, 171]. The resequencing approach involves choosing a

2set of genes for analysis (e.g., kinases) and performing

2 polymerase chain reaction (PCR) amplification of all

2coding exons including the intronexon splice site junctions

2for that set of genes, with genomic DNA isolated from

2 patient tumor samples or cell lines serving as a template.

2The PCR is carried out using a high-fidelity polymerase to

2avoid introduction of de novo mutations during the

2amplification reaction. PCR products are purified and then

2subjected to bidirectional dideoxy sequencing to identify

2putative mutations. Ideally, patient-matched normal DNA is

2analyzed using the same procedure and then compared with

2results from sequencing of the tumor DNA to allow for

2distinction between possible mutations and natural poly-

2morphisms. This strategy for mutation identification has

2 been successfully applied in studies of lung, colorectal,

2 breast, and brain tumors [7, 27, 28, 122, 145, 146, 170,

2171]. For example, a resequencing project analyzing the

2receptor tyrosine kinase family of genes in human

2glioblastomas found mutations of the fibroblast growth

2receptor-1 gene and the platelet-derived growth factor

2receptor (PDGFR) gene [122]. To date, exon resequenc-

2ing has been most extensively applied to the tyrosine kinase

2and associated phosphatase gene families [7, 27, 28, 122,

2145, 146, 171]; however, there are no limitations on the

2types or families of genes that can be interrogated using this

2strategy.

Childs Nerv Syst



11/23

UNCORREC

TED PR

OOF

240 Molecular genetics of specific pediatric brain tumors

241 High-grade astrocytoma

242 High-grade astrocytomas in children are relatively rare

243 lesions when compared with adults. Whereas the prognosis

244 of children with anaplastic astrocytoma or glioblastoma

245 multiforme may be better than that of adults with a246 comparable histological lesion, the majority of children

247 will still succumb from progressive disease within 25 years

248 after diagnosis [152]. What has been very interesting, and

249 as yet incompletely characterized, is the fact that the genetic

250 alterations that accompany the childhood high-grade astro-

251 cytic neoplasms are distinct from those that occur in adults.

252 The genetic alterations that are the hallmark of adult

253 anaplastic astrocytomasloss of p53, PTEN, p14ARF, and

254 amplification of the epidermal growth factor receptor

255 (EGFR) III mutantare relatively rare in pediatric anaplas-

256 tic astrocytomas (see below). In light of this and other

257 evidence, it appears as if the development of pediatric high-258 grade astrocytomas may follow pathways distinct from the

259 well-established primary and secondary paradigm of adult

260 glioblastomas.

261 Common chromosomal losses in pediatric high-grade

262 gliomas include those of chromosome 16p, 17p, 19p, 19q,

263 and 22 [129]. Within the spectrum of high-grade gliomas,

264 distinct cytogenetic changes are observed in pediatric

265 anaplastic astrocytomas, which are typified by gains of

266 chromosome 5q and losses of chromosomes 6q, 9q, 12q,

267 and 22q, and in pediatric glioblastoma, which is character-

268 ized by gains in chromosomes 1q and 16p and losses of

269 chromosomes 8q and 17p [129]. The finding of 1q

270 amplification is noteworthy because it may be a marker of

271 poor survival.

272 Whereas microsatellite instability appears to be generally

273 absent in the setting of adult high-grade astrocytomas,

274 nearly 25% of pediatric malignant astrocytomas may

275 display a microsatellite instability phenotype as a manifes-

276 tation of mutations in various DNA mismatch repair genes

277 [1, 22].

278 In addition to being characterized by their respective

279 patterns of structural chromosomal abnormalities, pediatric

280 high-grade astrocytomas may be distinguished from their

281 adult counterparts on the basis of differential expression of

282 oncogenes and tumor suppressor genes that are involved in

283 signal transduction pathways critical to the process of

284 gliomagenesis. The pertinent pathways for the adult tumors

285 include the growth factor/growth factor receptor/PI3-kinase/

286 Akt/PTEN pathway, the p53/MDM2/p14ARF pathway, and

287 the pRB/cyclinD1/cyclin-dependent kinase (CDK) 4/p16

288 pathway. The interested reader is referred to a review by

289 Ichimura et al. for a recent synopsis of the interplay of these

290 various pathways in the molecular pathogenesis of astro-

2cytic tumors [69]. In the remainder of this section, each

2pathway will be examined in turn in order to characterize

2the patterns of abnormality present within and to highlight

2the often dichotomous nature of the importance of an

2individual pathway to the genesis of pediatric vs adult high-

2grade gliomas.

2Although amplification of the EGFR gene is observed in

2up to 40% of adult glioblastomas and 15% of adult2anaplastic astrocytomas, it is not a common finding in

3 pediatric high-grade astrocytomas [22, 121, 148]. Similarly,

3whereas de novo adult glioblastomas have a high frequency

3of mutations in the PTEN tumor suppressor gene, pediatric

3malignant gliomas rarely contain such mutations. Although

3LOH at 10q2325 may be present in as many as 80% of

3informative cases, homozygous deletions of the PTEN gene

3are seen in only 8% of pediatric high-grade astrocytomas

3[22]. Where mutations of PTEN are observed, they may

3herald a poor prognosis for children with high-grade

3astrocytomas.

3The majority of adult secondary glioblastomas demon-3strate mutations of the p53 tumor suppressor gene located

3on chromosome 17p. A relatively small subset of high-

3grade gliomas, mostly occurring in older children, may also

3harbor frequent p53 mutations. In this population, there

3appears to be a nearly 2:1 frequency of p53 gene mutations

3in high-grade brainstem astrocytomas as opposed to non-

3brainstem astrocytomas [22, 90, 117].

3Homozygous deletions of the CDKN2A/pl4ARF locus at

39p21, which encodes both the p16 and p14ARFproteins, are

3found in 10% of pediatric malignant astrocytomas [103].

3Overexpression of the MDM2 oncogene is seen in 67% of

3 pediatric malignant astrocytomas, although there is no

3concomitant amplification of the MDM2 gene at a genomic

3level [148]. Overall, inactivation of the p53/MDM2/p14ARF

3 pathway, either by mutation of p53, overexpression of

3MDM2, or deletion of p14ARF, is seen in more than 95% of

3 pediatric malignant astrocytomas, a situation which is

3similar to the observed frequency of p53/MDM2/p14ARF

3 pathway inactivation in the setting of adult malignant

3astrocytomas [148]. However, as outlined earlier, mutations

3of p53 are seldom seen in malignant gliomas of children

3younger than 3 years, once again suggesting that malignant

3gliomas in very young children may follow a distinct

3molecular pathway as compared with older children and

3adults [118].

3Sure et al. were able to demonstrate loss of expression of

3p16 in 11 of 18 pediatric glioblastomas [149]. However, the

3 pRb/cyclin D1/CDK4/p16 pathway appears to be inacti-

3vated in only about 25% of pediatric malignant astrocyto-

3mas, as opposed to inactivation in more than 80% of the

3corresponding adult tumors [148].

3To date, there have been no published reports of cDNA

3microarray analyses on pediatric high-grade gliomas.

Childs Nerv Syst



12/23

UNCORREC

TED PR

OOF

344 Low-grade astrocytoma

345 Whereas tremendous advances have been made in the

346 understanding of the molecular pathogenesis of pediatric

347 high-grade astrocytomas, there have been comparatively

348 few studies focusing on the low-grade and pilocytic

349 astrocytomas of childhood.

350 Consistent with the fact that they are low-grade lesions,351 cytogenetic studies have revealed a normal karyotype in the

352 majority of pilocytic astrocytomas examined, and where

353 abnormal profiles have been observed, no consistent

354 karyotype abnormalities have been identified [138, 178].

355 Pediatric pilocytic astrocytomas show fewer chromosomal

356 changes than adult tumors; where observed, they are

357 usually gains of only a single chromosome, such as

358 chromosome 7 or 8, as has been demonstrated in

359 approximately one third of pediatric pilocytic astrocytomas

360 using fluorescent in situ hybridization (FISH) analysis

361 [173].

362 Individuals with neurofibromatosis type I (NF1) have an363 increased propensity to develop pilocytic astrocytomas,

364 especially of the optic/hypothalamic region. As sporadic

365 pilocytic astrocytomas occasionally show LOH on chro-

366 mosome 17q (the location of the NF1 tumor suppressor

367 gene), one would predict that mutations of the NF1 gene

368 with consequent loss of expression would be found in

369 sporadic pilocytic astrocytomas. In fact, this appears not to

370 be the case. In actuality, the expression of the NF1 tumor

371 suppressor gene in sporadic pilocytic astrocytomas is often

372 upregulated, perhaps as a reactive response to excessive

373 cellular proliferation [116, 175]. This is in contrast to

374 pilocytic astrocytomas arising in NF1 patients, where loss

375 of NF1 expression is an obligatory finding [54]. The precise

376 origin of the differing contribution of the NF1 gene product

377 in the setting of NF1 vs non-NF1 pilocytic astrocytomas

378 remains to be elucidated.

379 Even with the use of various sensitive p53 mutation

380 detection assays, it appears that p53 mutations are absent or

381 at most infrequently present in the setting of pediatric

382 pilocytic astrocytomas [23, 71, 174]. Such results are in

383 concordance with the low frequency of allelic loss detected

384 at the p53 locus on chromosome 17p [23, 71, 174]. Taken

385 together, these findings indicate that abnormalities of p53

386 do not contribute in any significant way to the genesis of

387 pilocytic astrocytomas.

388 Most cases of pediatric pilocytic astrocytoma show

389 immunopositivity for p16 and CDK4, indicating that

390 abnormalities in the pRb/cyclinD1/CDK4/p16 pathway

391 likely do not play an important role in the evolution of

392 these tumors [23]. PTEN mutations are also likely not a

393 significant contributor to the pathogenesis of these lesions

394 [23, 37], as mutations in this tumor suppressor gene tend to

395 be reserved for higher-grade tumors [123].

3Ependymoma

3Chromosomal 22 defects are frequently found in ependy-

3moma. Mutation of the NF-2 gene product on chromosome

322 has been documented to predispose to the formation of

4various tumor types, including ependymoma, especially in

4 patients with NF-2 [41]. However, the vast majority of

4sporadic (non-NF-2) cases lack mutations in the NF-2 gene.4The most plausible explanation for these findings is the

4existence of another oncogene or tumor suppressor gene on

4chromosome 22 that is more commonly involved in the

4genesis of sporadic ependymoma than the NF-2 gene

4product [166].

4Pediatric intracranial ependymomas appear to have a

4chromosomal signature distinct from their adult counter-

4 parts. Pediatric tumors, which often occur in the

4infratentorial compartment, display balanced CGH pro-

4files in 3050% of cases, in comparison to only 10% of

4adult tumors [38, 63]. This finding suggests that pediatric

4intracranial ependymomas may progress along substantially4different pathways than those giving rise to adult supra-

4tentorial or spinal ependymomas [161].

4Where chromosomal aberrations are observed in pediat-

4ric ependymomas, monosomy 17 appears to be one of the

4most common lesions, with an approximately 50% fre-

4quency [166]. The p53 tumor suppressor gene resides on

417p, and individual case reports have identified some

4instances of p53 mutation in ependymoma patients,

4including a family who had concurrent mutation in

4chromosome 22 [104].

4Gain of chromosome 1q is also a frequent finding in

4pediatric ependymoma [38]. A putative region of interest on

4chromosome 1q may be the region between 1q22 and 31

4[82]. Several studies have made an association between the

4 presence of 1q gain and a poor clinical course, suggesting

4that the presence of one or more genes located on 1q may

4 be responsible for tumor progression and/or response to

4therapy [38, 63].

4Loss of a region on 6q has been documented in a number

4of case reports; however, no known oncogene or tumor

4suppressor gene has yet been implicated at this locus [82,

4125, 131]. Other frequent chromosomal alterations include

4losses of chromosomes 1p, 6, 9q, 11, 13q, 16p, 16q, 19q,

420p, and 20q, and gains of 1q, 2, 5, 9, 12, 15, 18, 20q, and

4X [63, 163].

4Microarray experiments designed to document the gene

4expression profile of ependymomas have been performed.

4A recent examination of a panel of 12,627 genes identified

4a subset of 112 genes as being abnormally expressed in

4 pediatric ependymoma when compared with normal brain

4controls [147]. Genes with increased expression included

4the oncogene WNT5A, the p53 homologue p63, and

4several cell cycle, cell adhesion, and cell proliferation

Childs Nerv Syst



13/23

UNCORREC

TED PR

OOF

448 genes. Underexpressed genes included the NF2 interacting

449 gene SCHIP-1 and the adenomatous polyposis coli (APC)-

450 associated gene EB1 among others. These genes represent

451 candidate genes for further study; further validation work is

452 necessary in order to clarify their precise contribution to

453 ependymoma tumorigenesis.

454 Our understanding of the origins of ependymoma

455 increased dramatically with the recent publication of Taylor456 et al. in which histologically identical, but genetically

457 distinct, ependymomas showed patterns of gene expression

458 that recapitulate those of radial glia cells in corresponding

459 regions of the central nervous system [161]. In this study,

460 supratentorial, infratentorial, and intraspinal ependymomas

461 demonstrated distinct genetic signatures and were shown to

462 arise from restricted populations of radial glia stem cells.

463 For the supratentorial tumors, CDK4 and several Notch

464 signaling pathway genes were overexpressed; for the

465 infratentorial tumors, IFG-1 and several HOX homeobox

466 genes were overexpressed; and for the spinal tumors, the ID

467 genes and the aquaporins were overexpressed. The impli-468 cations of this study are that ependymomas should be

469 treated with therapies that target the cell signal pathways

470 that maintain subsets of ependymoma stem cells rather than

471 the histological or clinical forms of the disease [161].

472 Atypical teratoid rhabdoid tumor

473 Molecular studies have been able to distinguish CNS AT/

474 RT from the other primitive pediatric brain tumors,

475 establishing it as a unique pathological entity [133, 134].

476 AT/RT frequently demonstrates deletion of the long arm of

477 chromosome 22q11.2, and further molecular studies have

478 led to the identification of the INI1/hSNF5 tumor suppres-

479 sor gene at this location [1013]. A somatic mutation in this

480 gene predisposes children to develop AT/RT [1013]. The

481 hSNF5 protein is the smallest member of a highly

482 conserved family of proteins that function in chromatin

483 remodeling via the nucleosome. By winding and un-

484 winding DNA, this complex changes the configuration of

485 genomic DNA, thus allowing or denying transcription

486 factors access to the DNA, consequently changing patterns

487 of gene expression.

488 Some children with AT/RT are born with heterozygous

489 germ line mutations of the hSNF5 gene, suggesting that

490 these children were predisposed to develop AT/RT [155]; in

491 most cases, however, these germ line mutations arise de

492 novo.

493 Heterozygous mSNF5 +/ knockout mice develop

494 tumors resembling AT/RT, supporting the role of hSNF5 as

495 a tumor suppressor gene [130]. Although most AT/RTs

496 show evidence of some genetic derangement at the hSNF5

497 locus, mutational analysis of the hSNF5 gene in a series of

498 primitive neuroectodermal tumors/medulloblastomas dis-

4covered mutations in only 4 of 52 tumors [155]. Of those

5four, two were reclassified as AT/RT upon reexamination of

5the pathology, but there was insufficient clinical material to

5establish an accurate diagnosis in the other two cases. This

5suggests that tumors which are histologically diagnosed as

5 primitive neuroectodermal tumors/medulloblastomas but

5which also harbor hSNF5 mutations are most likely AT/

5RT. Although mutation/deletion of hSNF5 is not currently5sufficient for a diagnosis of AT/RT, it appears to be related

5to the clinical outcome; consequently, determination of the

5status of the AT/RT gene by DNA sequence analysis, FISH,

5or immunohistochemistry is rapidly becoming an integral

5 part of the neuropathological diagnosis of primitive

5pediatric malignant brain tumors (Fig. 2).

5Medulloblastoma

5Recent gene expression studies have shown that medullo-

5 blastoma is molecularly distinct from the supratentorial

5 primitive neuroectodermal tumors (sPNETs) [119]. The5Sonic Hedgehog (SHH), Wingless(WNT/WG), and

5receptor tyrosine kinase I family ERBB pathway are

5emerging as central developmental signaling pathway

5systems in the formation of medulloblastoma [17, 110].

5The specific contributions of these pathways to the genesis

5of medulloblastoma will be discussed in turn.

5Medulloblastoma has been reported in the setting of a

5number of different hereditary cancer syndromes, including

5nevoid basal cell carcinoma syndrome (NBCCS), Turcot

5syndrome, LiFraumeni syndrome and RubensteinTaybi

5syndrome [156]. In the mid-1990s, it was discovered that

5NBCCS arises from germ line mutations of PTCH1, which

Fig. 2 Pattern of INI-1 staining in AT/RTs of childhood. Immuno-

histochemical staining with an anti-INI-1 antibody fails to stain the

nuclei of tumor cells but does stain the nuclei of normal or reactive

tissues in the vicinity

Childs Nerv Syst



14/23

UNCORREC

TED PR

OOF

529 encodes the receptor for the Hedgehog (Hh) family of

530 signaling proteins. PTCH1 is a 12-pass transmembrane

531 receptor protein that represses activity of the Hh signaling

532 pathway in its unbound state by suppressing the activity of

533 the seven-pass transmembrane protein Smoothened(S-

534 MOH) [56, 74, 94]. Upon ligand binding, PTCH1s tonic

535 inhibition of SMOH is lifted, leading to the release of GLI

536 transcription factor from a tetra-protein complex which also537 includes suppressor of fused (SUFU), fused (FU), and

538 costal-2 (COS2). Upon its release, GLI is activated and

539 translocated to the nucleus where it binds to promoters in a

540 sequence-specific fashion to increase the transcription of

541 various growth-promoting genes such as cyclin D [94].

542 Mutations in the various molecules comprising this signal-

543 ing complex, including SHH, PTCH1, SMOH, and SUFU,

544 have been shown to occur in both familial and sporadic

545 tumors [43, 113, 126, 158]. Another recent report demon-

546 strated a crucial role for polycomb group gene Bmi1 in

547 clonal expansion of granule cell precursors in vivo and

548 linked overexpression of BMI1 and patched (PTCH), a549 finding that is suggestive of an alternative or additive

550 mechanism of SHH pathway activation in the pathogenesis

551 of medulloblastomas [84]. A novel putative tumor suppres-

552 sor, human REN(KCTD11), incidentally maps to 17p13.2,

553 a chromosomal region which is frequently lost in medul-

554 loblastoma [45]. The REN tumor suppressor inhibits

555 medulloblastoma growth by impairing both Gli2-dependent

556 gene transcription and SHH-enhanced expression of the

557 target Gli1 mRNA, thereby decreasing the expression of

558 downstream genes [34, 35].

559 Cyclopamine, a plant-derived teratogen that binds to

560 SMOH and interrupts its ability to activate downstream

561 signaling, may have some important therapeutic implica-

562 tions. Mice harboring human medulloblastoma explants

563 showed tumor regression upon drug administration. Less

564 toxic derivatives of cyclopamine are being used in

565 preclinical trials of medulloblastoma. Although this com-

566 pound will probably only be effective for patients with

567 NBCCS or those whose tumors harbor PTCH and SMOH

568 mutations, it carries great potential as a targeted molecular

569 therapy in such instances [9, 151].

570 Germ line mutations of APC in patients with Turcot

571 syndrome have been reported to increase the risk of

572 developing medulloblastomas [58]. The inactivation of the

573 APC gene leads to aberrant signaling in the WNT pathway,

574 resulting in the inability of the multiprotein complex

575 containing APC, axin, and glycogen synthase kinase-3

576 (GSK-3) to phosphorylate and cause degradation of -

577 catenin. This leads to an overabundance of -catenin and

578 its translocation to the nucleus where it activates the TCF/

579 LEF transcriptional complex. The -catenin/TCF dimer

580 activates the transcription of other growth-regulating genes,

581 some of which are known oncogenes (e.g., c-MYC and

5cyclin D) [42, 101, 150]. Mutations in WNT pathway

5members occur in about 15% of sporadic medulloblastoma

5[25, 66, 77, 181]. A recent report also demonstrated that

5mutations in SUFU can cause a decreased efficiency of-

5catenin nuclear export, thereby also resulting in increased

5-catenin/TCF signaling [160].

5The ERBB or epidermal growth factor family of receptor

5tyrosine kinases plays a crucial role in regulating cellular5 proliferation, apoptosis, migration, and differentiation.

5Activation of these receptors through ligand binding,

5dimerization, and autophosphorylation culminate in down-

5stream signaling through mitogen-activated protein kinase

5(MAPK), AKT, and STAT [107]. Interestingly, one of the

5four members of the ERBB family, ERBB2, has been

5shown to be highly expressed in medulloblastoma [52]. A

5high level of ERBB2 and ERBB4 coexpression signifies an

5increased risk of metastases and is associated with poor

5prognosis in cases of medulloblastoma [48, 50]. Microarray

6analysis of medulloblastoma cell lines demonstrate that

6S100A4, a prometastatic gene known to be linked to breast6and bladder cancers, is a direct target of ERBB2 signaling

6in medulloblastoma cells via the AKT/PI3K pathway [61].

6Compounds such as OSI-774 (erlotinib, also known as

6gefitinif) inhibit ERBB2 signaling in human medulloblas-

6toma cells and may have therapeutic potential [61].

6The PDGFR and downstream activation of the RAS/

6MAPK signaling pathway (including MAP2K1, MAP2K2,

6and MAPK1/3) have also been uncovered as potential

6mediators of medulloblastoma metastasis [33, 51, 92].

6The mRNA expression level of TrkC, a neurotrophin

6signaling receptor, was identified as a positive prognostic

6factor for progression-free and overall survival in medul-

6loblastoma [53, 124]. The MYC protooncogenes (MYCN

6and MYCC) are amplified and/or overexpressed in a

6subset of large cell medulloblastomas and also correlate

6with poor clinical outcome [39, 40]. However, other

6studies of TrkC, MYCC, and MYCN mRNA expression

6or immunoreactivity did not demonstrate a significant

6 prognostic effect [48, 80].

6Loss of chromosome 17p, often through the formation of

6an isochromosome 17q i(17)(q10), is the most common

6chromosome aberration in childhood medulloblastoma,

6occurring in about 25% to 35% of cases. Either isolated

617p deletion or i(17)(q10) has been reported as a significant

6negative prognostic factor [50, 109]. Recent cytogenetic

6analysis using matrix CGH suggested that overexpression

6of CDK6 correlates with a worse prognosis [99].

6Most of the tumor suppressor genes shown to play a role

6in the formation of medulloblastoma have been identified

6through mutations; however, other epigenetic phenomena

6may lead to their decreased expression. Aberrant methyla-

6tion of CpG islands located in promoter regions represents

6one of the major mechanisms for silencing of cancer-related

Childs Nerv Syst



15/23

UNCORREC

TED PR

OOF

635 genes in tumor cells. Extensive hypermethylation of the

636 RASSF1A gene (ras-association domain family protein 1,

637 isoform A), an identified tumor suppressor gene located at

638 chromosome 3p21.3, occurs in about 80% of primary

639 medulloblastomas [86, 91]. Furthermore, complete methyl-

640 ation of the putative tumor suppressor HIC-1 (hyper-

641 methylated in cancer) and the apoptosis effector molecule

642 caspsase-8 has been found in a subset of medulloblastoma643 [132, 168, 182].

644 Germ cell tumors

645 Although the exact molecular and cytogenetic aberrations

646 in GCTs are still not well defined, the most consistent

647 cytogenetic abnormality observed in such tumors is

648 isochromosome 12p (i12p) [26, 31, 89]. Some tumors

649 without i12p have overrepresentation of 12p by other

650 mechanisms, such as duplication of the entire chromosome.

651 The exact role of genes implicated in 12p overrepresenta-652 tion in GCTs remains uncertain, but evidence suggests that

653 the genes located on this chromosome, such as cyclin D2,

654 may play a role in facilitating entry into S phase of the cell

655 cycle. The rather frequent finding of i12p is also highly

656 indicative of the existence of novel putative oncogenes on

657 12p which may be involved in the pathogenesis of GCTs.

658 Other chromosomal aberrations have also been observed

659 in pineal GCTs. In a study of 15 pineal region GCTs,

660 Rickert et al. reported that the most common chromosomal

661 imbalances in pineal germinomas were losses of 13q and

662 18q, and loss of chromosomes 4 and 5 [128]. Interestingly,

663 these authors infrequently found gain of 12p. A summary

664 of other reported genetic aberrations in these tumors can be

665 found in a recent review by Taylor et al. [157].

666 Although most GCTs are sporadic, a few genetic

667 syndromes do predispose individuals to their development.

668 Pineal germinomas or teratomas in patients with Klinefelter

669 syndrome have been reported [120]. That patients with

670 Klinefelter syndrome generally have increased risk of

671 developing malignancy is in keeping with the idea of the

672 existence of a putative oncogene on the X chromosome.

673 Patients with trisomy 21 have an increased risk of a number

674 of cancers including leukemia and gonadal and extragona-

675 dal GCTs [139].

676 Ewing sarcoma

677 Primary Ewing sarcoma (ES)/peripheral primitive neuro-

678 ectodermal tumors (pPNET) of the central nervous system

679 constitute a clinically important, albeit rare, subset of

680 pediatric soft tissue tumors. They typically arise extracra-

681 nially or paraspinally and often result in secondary invasion

682 of critical neural structures [18].

6In recent years, an examination of the genetic alterations

6 present within the spectrum of pediatric soft tissue tumors

6has demonstrated a fairly specific (although not absolute)

6association between specific nonrandom reciprocal chro-

6mosomal translocations and individual soft tissue tumor

6types [20]. Ongoing study of these fusion products has

6yielded profound insights into the biology of these tumors

6and may hold great promise for novel diagnostic and6therapeutic applications.

6Ewing sarcoma/pPNET-associated translocations charac-

6teristically involve the EWS gene on 22q12 and various

6members of the ETS family of protooncogenes (Fig. 3).

6Approximately 85% of ES/pPNET tumors harbor the

6translocation t(11;22)(q24;q12); in this subset of tumors,

6the translocation partner for EWS is the FLI1 gene product

6found on 11q24 [32]. In nearly 10% of cases, the

6translocation partner is ERG [t(21;22)(q22;q12)] [144],

7and in rare cases, EWS may be fused to the ETS domains

7of ETV-1 [t(7;22)(p22;q12)] [73], E1AF [t(17;22)(q12;

7q12)] [76], or FEV [t(2;22)(q33;q12)] [111].7The EWS gene product is a member of a growing family

7of highly conserved RNA-binding proteins [105]. Although

7the exact biological function of wild-type EWS and its

7homologues remains largely unknown, a growing body of

7evidence suggests that they are involved in mRNA

7transcription. EWS has been shown to form an adaptor

7ternary complex with RNA polymerase II and other

7heterogeneous RNA-binding proteins [112]; these findings

7are highly suggestive of an important role of EWS in basic

7transcriptional regulation.

7ETS-domain-containing proteins are DNA-binding tran-

7scription factors that are implicated in the control of cellular

7 proliferation. ETS family members appear to cooperate

7with other nuclear proteins to help establish promoter

7specificity, modulate transcriptional regulation, and facili-

7tate linkage to various signal transduction pathways,

7including RAS [36, 64, 143, 172].

7The ES/pPNET-associated translocations result in chi-

7meric proteins containing the N-terminal domain of EWS

7fused to the site-specific nucleic acid binding domain of the

7ETS transcription factor translocation partner (Fig. 3). This

7structure suggests that the chimeric protein is directed at the

7 promoter region of specific genes recognized by the

7translocated DNA-binding domain of the ETS member [5,

7106]. However, the ultimate targets of the EWS-ETS

7chimeric protein may not be solely dependent on the ETS

7domain; rather, other proteinprotein interactions unique to

7the chimeric molecule may be at play [98]. The actual

7target genes contributing to tumorigenesis are not known,

7but analysis of mRNAs differentially expressed in cell lines

7stably transfected with the EWS-FLI1 fusion has produced

7some interesting candidates, including manic fringe, c-myc,

7cyclin D1, mE2-C, MMP-1, and TGF-RIII [4, 6, 14].

Childs Nerv Syst



16/23

UNCORREC

TED PR

OOF

736 The fairly consistent presence of these recombinant gene

737 products in ES/pPNET suggests that they play a critical role

738 in the underlying biology of these tumors (Fig. 3). Trans-739 fection of EWS-FLI1 or EWS-ERG can transform mouse

740 NIH-3T3 cells if both the EWS and ETS domains are

741 functionally intact [98]. EWS-FLI1 antisense RNA trans-

742 fected into ES/pPNET cells results in marked growth

743 inhibition, suggesting that the EWS-ETS gene rearrange-

744 ment may also be necessary for maintaining the malignant

745 phenotype of ES/pPNET cell lines [153]. Furthermore,

746 expression of EWS-ETS fusion constructs may contribute

747 to tumorigenesis via inhibition of apoptosis; not surprisingly,

748 antisense inhibition of EWS-ETS fusion genes may enhance

749 susceptibility to chemotherapy-induced apoptosis [177].

750 The gene fusions characteristic of ES/pPNET exhibit an751 underlying molecular heterogeneity. There are two under-

752 lying sources of variability: the specific ETS fusion partner

753 and the breakpoint location within the genes. A better

754 outcome for patients with localized tumors expressing the

755 most common chimeric transcript (type I: EWS exon 7

756 fused to FLI1 exon 6) compared with the next most

757 common fusion types (type II: EWS exon 7 fused to FLI1

758 exon 5 or type III: EWS exon 10 fused to FLI1 exon 6) has

759 been reported, raising the possibility that heterogeneity in

7chimeric transcripts may reliably define clinically distinct

7risk groups [30, 85]. Preliminary work suggests that the

7better outcome associated with type I fusion transcripts may7be related to the weaker transcriptional activation properties

7of the type I transcript [30].

7Other less common numerical and structural chromo-

7somal aberrations may also be found in ES/pPNET [3, 15,

797, 102, 108, 154]. The most common numerical abnor-

7malities are +1q, +8, +12, and +20, and 1p, 16q, and

719q. More complex, multichromosome translocations,

7such as t(11;14;22)(q24;q11;q12) and t(10;11;22)(p11.2;

7q24;q12), may also occur and generally portend a poor

7prognosis.

7Changes in known tumor suppressor genes may be

7observed in some cases. Homozygous deletion of7CDKN2A (p14ARF) on 9p21 has been described in

7approximately 30% of cases [81]. Although mutations in

7p53 have been described in up to 50% of ES/pPNET cell

7lines [57], this appears to be a rare event in primary tumors

7[167]. Amplification of the MDM2 gene, an inactivator of

7 p53, is also rare in ES/pPNET, consistent with the

7hypothesis that p53 and regulators of its activity may not

7 play a dominant role in the pathogenesis of ES/pPNET

7[93].

Fig. 3 Schematic representation

of the two commonest fusion

genes in Ewing sarcoma and the

proposed effects of the onco-

genic chimeric protein on cellu-

lar proliferation and

tumorigenicity

Childs Nerv Syst



17/23

UNCORREC

TED PR

OOF

784 Conclusions

785 As is evident from this review, our understanding of the

786 fundamental mechanisms of brain tumorigenesis in children

787 has increased markedly over the past 2 decades. The pace

788 of subsequent advances will only quicken, owing to such

789 monumental technical feats as the sequencing of the human

790 genome. Concomitant developments in the field of bio-791 informatics will be equally important as they will allow us

792 to make sense of the volumes of data that will undoubtedly

793 emerge from the ongoing interrogation of the fundamental

794 molecular processes at work in the genesis of pediatric

795 brain tumors. The challenge for the next 2 decades will be

796 to consolidate our understanding of the molecular patho-

797 genesis of childhood brain tumors and to apply this

798 sophisticated knowledge toward the development of clini-

799 cally useful treatment strategies.

800 References

801 1. Alonso M, Hamelin R, Kim M, Porwancher K, Sung T, Parhar P,

802 Miller DC, Newcomb EW (2001) Microsatellite instability

803 occurs in distinct subtypes of pediatric but not adult central

804 nervous system tumors. Cancer Res 61:21242128

805 2. Altshuler D, Brooks LD, Chakravarti A, Collins FS, Daly MJ,

806 Donnelly P, International HapMap Consortium (2005) A

807 haplotype map of the human genome. Nature 437:12991320

808 3. Armengol G, Tarkkanen M, Virolainen M et al (1997) Recurrent

809 gains of 1q, 8 and 12 in the Ewing family of tumors by

810 comparative genomic hybridization. Br J Cancer 75:14031409

811 4. Baer C, Nees M, Breit S et al (2004) Profiling and functional

812 annotation of mRNA gene expression in pediatric rhabdomyo-

813 sarcoma and Ewings sarcoma. Int J Cancer 110:687694

814 5. Bailly RA, Bosselut R, Zucman J et al (1994) DNA-binding and

815 transcriptional activation properties of the EWS-FLI-1 fusion

816 protein resulting from the t(11;22) translocation in Ewing

817 sarcoma. Mol Cell Biol 14:32303241

818 6. Bandres E, Malumbres R, Escalada A et al (2005) Gene

819 expression profile of Ewing sarcoma cell lines differing in their

820 EWS-FLI1 fusion type. J Pediatr Hematol Oncol 27:537542

821 7. Bardelli A, Parsons DW, Silliman N, Ptak J, Szabo S, Saha S,

822 Markowitz S, Willson JK, Parmigiani G, Kinzler KW, Vogelstein

823 B, Velculescu VE (2003) Mutational analysis of the tyrosine

824 kinome in colorectal cancers. Science 300:949

825 8. Barrett MT, Scheffer A, Ben-Dor A, Sampas N, Lipson D,

826 Kincaid R, Tsang P, Curry B, Baird K, Meltzer PS, Yakhini Z,

827 Bruhn L, Laderman S (2004) Comparative genomic hybridiza-

828tion using oligonucleotide microarrays and total genomic DNA.829 Proc Natl Acad Sci USA 101:1776517770

830 9. Berman DM, Karhadkar SS, Hallahan AR, Pritchard JI,

831 Eberhart CG, Watkins DN et al (2002) Medulloblastoma

832 growth inhibition by hedgehog pathway blockade. Science

833 297:15591561

834 10. Biegel JA, Rorke LB, Packer RJ et al (1990) Monosomy 22 in

835 rhabdoid or atypical tumors of the brain. J Neurosurg 73:710 714

836 11. Biegel JA, Zhou JY, Rorke LB et al (1999) Germ-line and

837 acquired mutations of INI1 in atypical teratoid and rhabdoid

838 tumors. Cancer Res 59:7479

839 12. Biegel JA, Fogelgren B, Zhou JY, James CD, Janss AJ, Allen JC,

840 Zagzag D, Raffel C, Rorke LB (2000) Mutations of the INI1

8rhabdoid tumor suppressor gene in medulloblastomas and

8primitive neuroectodermal tumors of the central nervous system.

8Clin Cancer Res 6:27592763

813. Biegel JA, Tan L, Zhang F, Wainwright L, Russo P, Rorke LB

8(2002) Alterations of the hSNF5/INI1 gene in central nervous

8system atypical teratoid/rhabdoid tumors and renal and extra-

8renal rhabdoid tumors. Clin Cancer Res 8:34613467

814. Braun BS, Frieden R, Lessnick SL et al (1995) Identification

8of target genes for the Ewings sarcoma EWS/Fli fusion

8 protein by representational difference analysis. Mol Cell Biol

815:46234630

815. Brisset S, Schleiermacher G, Peter M et al (2001) CGH analysis

8of secondary genetic changes in Ewing tumors: correlation with

8metastatic disease in a series of 43 cases. Cancer Genet

8Cytogenet 130:5761

816. Bucca G, Carruba G, Saetta A, Muti P, Castagnetta L, Smith CP

8(2004) Gene expression profiling of human cancers. Ann N Y

8Acad Sci 1028:2837

817. Buhren J, Christoph AH, Buslei R, Albrecht S, Wiestler OD,

8Pietsch T (2000) Expression of the neurotrophin receptor

8 p75NTR in medulloblastomas is correlated with distinct histo-

8logical and clinical features: evidence for a medulloblastoma

8subtype derived from the external granule cell layer. J Neuro-

8pathol Exp Neurol 59:229240

818. Carlotti CG Jr, Drake JM, Hladky JP, Teshima I, Becker LE,

8Rutka JT (1999) Primary Ewings sarcoma of the skull in

8children. Utility of molecular diagnostics, surgery and adjuvant

8therapies. Pediatr Neurosurg 31:307315

819. Carvalho B, Ouwerkerk E, Meijer GA, Ylstra B (2004) High

8resolution microarray comparative genomic hybridisation analy-

8sis using spotted oligonucleotides. J Clin Pathol 57:644646

820. Chang CC, Shidham VB (2003) Molecular genetics of pediatric

8soft tissue tumors. J Mol Diagn 5:143154

821. Cheng J, Kapranov P, Drenkow J, Dike S, Brubaker S, Patel S,

8Long J, Stern D, Tammana H, Helt G, Sementchenko V,

8Piccolboni A, Bekiranov S, Bailey DK, Ganesh M, Ghosh S,

8Bell I, Gerhard DS, Gingeras TR (2005) Transcriptional maps of

810 human chromosomes at 5-nucleotide resolution. Science

8308:11491154

822. Cheng Y, Ng HK, Zhang SF, Ding M, Pang JC, Zheng J, Poon

8WS (1999) Genetic alterations in pediatric high-grade astrocyto-

8mas. Hum Pathol 30:12841290

823. Cheng Y, Pang JC, Ng HK, Ding M, Zhang SF, Zheng J, Liu

8DG, Poon WS (2000) Pilocytic astrocytomas do not show most

8of the genetic changes commonly seen in diffuse astrocytomas.

8Histopathology 37:437444

824. Cowell JK, Matsui S, Wang YD, LaDuca J, Conroy J, McQuaid

8D, Nowak NJ (2004) Application of bacterial artificial chromo-

8some array-based comparative genomic hybridization and spec-

8tral karyotyping to the analysis of glioblastoma multiforme.

8Cancer Genet Cytogenet 151:3651

825. Dahmen RP, Koch A, Denkhaus D et al (2001) Deletions of

8AXIN1, a component of the WNT/wingless pathway, in sporadic

8medulloblastomas. Cancer Res 61:7039

7043 826. Dal Cin P, Dei Tos AP, Qi H, Giannini C, Furlanetto A, Longatti

8PL, Marynen P, Van den Berghe H (1998) Immature teratoma of

8the pineal gland with isochromosome 12p. Acta Neuropathol

8(Berl) 95:107110

827. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S,

9Teague J, Woffendin H, Garnett MJ, Bottomley W, Davis N,

9Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J,

9Kosmidou V, Menzies A, Mould C, Parker A, Stevens C, Watt S,

9Hooper S, Wilson R, Jayatilake H, Gusterson BA, Cooper C,

9Shipley J, Hargrave D, Pritchard-Jones K, Maitland N, Chene-

9vix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A,

9Flanagan A, Nicholson A, Ho JW, Leung SY, Yuen ST, Weber

Childs Nerv Syst



18/23

UNCORREC

TED PR

OOF

907 BL, Seigler HF, Darrow TL, Paterson H, Marais R, Marshall CJ,

908 Wooster R, Stratton MR, Futreal PA (2002) Mutations of the

909 BRAF gene in human cancer. Nature 417:949954

910 28. Davies H, Hunter C, Smith R, Stephens P, Greenman C, Bignell

911 G, Teague J, Butler A, Edkins S, Stevens C, Parker A, OMeara S,

912 Avis T, Barthorpe S, Brackenbury L, Buck G, Clements J, Cole J,

913 Dicks E, Edwards K, Forbes S, Gorton M, Gray K, Halliday K,

914 Harrison R, Hills K, Hinton J, Jones D, Kosmidou V, Laman R,

915 Lugg R, Menzies A, Perry J, Petty R, Raine K, Shepherd R, Small

916 A, Solomon H, Stephens Y, Tofts C, Varian J, Webb A, West S,917 Widaa S, Yates A, Brasseur F, Cooper CS, Flanagan AM, Green

918 A, Knowles M, Leung SY, Looijenga LH, Malkowicz B, Pierotti

919 MA, Teh BT, Yuen ST, Lakhani SR, Easton DF, Weber BL,

920 Goldstraw P, Nicholson AG, Wooster R, Stratton MR, Futreal PA

921 (2005) Somatic mutations of the protein kinase gene family in

922 human lung cancer. Cancer Res 65:75917595

923 29. Davies JJ, Wilson IM, Lam WL (2005) Array CGH technologies

924 and their applications to cancer genomes. Chromosome Res

925 13:237248

926 30. de Alava E, Panizo A, Antonescu CR, Huvos AG, Pardo-Mindan

927 FJ, Barr FG, Ladanyi M (2000) Association of EWS-FLI1 type 1

928 fusion with lower proliferative rate in Ewings sarcoma. Am J

929 Pathol 156:849855

930 31. de Bruin TW, Slater RM, Defferrari R, Geurts van Kessel A,

931 Suijkerbuijk RF, Jansen G, de Jong B, Oosterhuis JW (1994)932 Isochromosome 12p-positive pineal germ cell tumor. Cancer Res

933 54:15421544

934 32. Delattre O, Zucman J, Plougastel B et al (1992) Gene fusion with

935 an ETS DNA-binding domain caused by chromosome translo-

936 cation in human tumors. Nature 359:162165

937 33. Del Valle L, Enam S, Lassak A, Wang JY, Croul S et al (2002)

938 Insulin-like growth factor I receptor activity in human medullo-

939 blastomas. Clin Cancer Res 8:18221830

940 34. De Smaele E, Di Marcotullio L, Ferretti E, Screpanti I, Alesse E,

941 Gulino A (2004) Chromosome 17p deletion in human medullo-

942 blastoma: a missing checkpoint in the Hedgehog pathway. Cell

943 Cycle 3:12631266

944 35. Di Marcotullio L, Ferretti E, De Smaele E, Argenti B, Mincione

945 C et al (2004) REN(KCTD11) is a suppressor of Hedgehog

946 signaling and is deleted in human medulloblastoma. Proc Natl

947 Acad Sci USA 101:1083310838

948 36. Dittmer J, Nordheim A (1998) ETS transcription factors and

949 human disease. Biochim Biophys Acta 1377:F1F11

950 37. Duerr EM, Rollbrocker B, Hayashi Y, Peters N, Meyer-Puttlitz

951 B, Louis DN, Schramm J, Wiestler OD, Parsons R, Eng C, von

952 Deimling A (1998) PTEN mutations in gliomas and glioneuronal

953 tumors. Oncogene 16:22592264

954 38. Dyer S, Prebble E, Davison V, Davies P, Ramani P, Ellison D,

955 Grundy R (2002) Genomic imbalances in pediatric intracranial

956 ependymomas define clinically relevant groups. Am J Pathol

957 161:21332141

958 39. Eberhart CG, Kepner JL, Goldthwaite PT et al (2002) Histo-

959 pathologic grading of medulloblastomas: a Pediatric Oncology

960 Group study. Cancer 94:552560

961 40. Eberhart CG, Kratz J, Wang Y, Summers K, Stearns D et al

962 (2004) Histopathological and molecular prognostic markers in

963 medulloblastoma: c-myc, n-myc, TrkC, and anaplasia. J Neuro-

964 pathol Exp Neurol 63:441449

965 41. Ebert C, von Haken M, Meyer-Puttlitz B, Wiestler OD,

966 Reifenberger G, Pietsch T, von Deimling A (1999) Molecular

967 genetic analysis of ependymal tumors. NF2 mutations and

968 chromosome 22q loss occur preferentially in intramedullary

969 spinal ependymomas. Am J Pathol 155:627632

970 42. Elberhart CG, Tihan T, Burger PC (2000) Nuclear localization

971 and mutation of beta-catenin in medulloblastoma. J Neuropathol

972 Exp Neurol 59:333337

943. Erez A, Ilan T, Amariglio N et al (2002) GLI3 is not mutated

9commonly in sporadic medulloblastomas. Cancer 95:2831

944. Fernandez-Teijeiro A, Betensky RA, Sturla LM, Kim JY,

9Tamayo P, Pomeroy SL (2004) Combining gene expression

9 profiles and clinical parameters for risk stratification in medul-

9loblastomas. J Clin Oncol 22:994998

945. Ferretti E, Smaele ED, Marcotullio LD, Screpanti I, Gulino A

9(2005) Hedgehog checkpoints in medulloblastoma: the chromo-

9some 17p deletion paradigm. Trends Mol Med 11:537545

946. Fink KL, Rushing EJ, Schold SC Jr, Nisen PD (1996)9Infrequency of p53 gene mutations in ependymomas. J Neuro-

9oncol 27:111115

947. Freeman WM, Robertson DJ, Vrana KE (2000) Fundamentals of

9DNA hybridization arrays for gene expression analysis. Bio-

9techniques 29:10421046, 10481055

948. Gajjar A, Hernan R, Kocak M, Fuller C, Lee Y et al (2004)

9Clinical, histopathological, and molecular markers of prognosis:

9toward a new disease risk stratification system for medulloblas-

9toma. J Clin Oncol 22:984993

949. Garraway LA, Widlund HR, Rubin MA, Getz G, Berger AJ,

9Ramaswamy S, Beroukhim R, Milner DA, Granter SR, Du J,

9Lee C, Wagner SN, Li C, Golub TR, Rimm DL, Meyerson ML,

9Fisher DE, Sellers WR (2005) Integrative genomic analyses

9identify MITF as a lineage survival oncogene amplified in

9malignant melanoma. Nature 436:117122950. Gilbertson R, Wickramasinghe C, Hernan R, Balaji V, Hunt D et

9al (2001) Clinical and molecular stratification of disease risk in

1medulloblastoma. Br J Cancer 85:705712

151. Gilbertson RJ, Clifford SC (2003) PDGFRB is overexpressed in

1metastatic medulloblastoma. Nat Genet 35:197198

152. Gilbertson RJ, Clifford SC, MacMeekin W, Meekin W, Wright C

1et al (1998) Expression of the ErbB-neuregulin signaling network

1during human cerebellar development: implications for the

1biology of medulloblastoma. Cancer Res 58:39323941

153. Grotzer MA, Janss AJ, Fung KM, Biegel JA, Sutton LN, Rorke

1LB (2000) TrkC expression predicts good clinical outcome in

1primitive neuroectodermal brain tumors. J Clin Oncol 18:1027

11035

154. Gutmann DH, Donahoe J, Brown T, James CD, Perry A (2000)

1Loss of neurofibromatosis 1 (NF1) gene expression in NF1-

1associated pilocytic astrocytomas. Neuropathol Appl Neurobiol

126:361367

155. Gyure KA (2005) Newly defined central nervous system

1neoplasms. Am J Clin Pathol 123(Suppl):S3S12

156. Hahn H, Wicking C, Zaphiropoulous PG, Gailani MR, Shanley S

1et al (1996) Mutations of the human homolog of Drosophila

1 patched in the nevoid basal cell carcinoma syndrome. Cell

185:841851

157. Hamelin R, Zucman J, Melot T et al (1994) p53 mutations in

1human tumors with chimeric EWS/FLI-1 genes. Int J Cancer

157:336340

158. Hamilton SR, Liu B, Parsons RE, Papadopoulos N, Jen J, Powell

1SM et al (1995) The molecular basis of Turcots syndrome. N

1Engl J Med 332:839847

159. Harkness W, Hayward R, Darling J, Thomas D (2001)

1Identification of extensive genomic loss and gain by comparative

1genomic hybridisation in malignant astrocytoma in children and

1young adults. Genes Chromosomes Cancer 31:1522

160. Harrington CA, Rosenow C, Retief J (2000) Monitoring gene

1expression using DNA microarrays. Curr Opin Microbiol

13:285291

161. Hernan R, Fasheh R, Calabrese C, Frank AJ, Maclean KH et al

1(2003) ERBB2 up-regulates S100A4 and several other prometa-

1static genes in medulloblastoma. Cancer Res 63:140148

162. Hinds DA, Stuve LL, Nilsen GB, Halperin E, Eskin E, Ballinger

1DG, Frazer KA, Cox DR (2005) Whole-genome patterns of

Childs Nerv Syst



19/23

UNCORREC

TED PR

OOF

1039 common DNA variation in three human populations. Science

1040 307:10721079

1041 63. Hirose Y, Aldape K, Bollen A, James CD, Brat D, Lamborn K,

1042 Berger M, Feuerstein BG (2001) Chromosomal abnormalities

1043 subdivide ependymal tumors into clinically relevant groups. Am

1044 J Pathol 158:11371143

1045 64. Hromas R, Klemsz M (1994) The ETS oncogene family in

1046 development, proliferation and neoplasia. Int J Hematol 59:257

1047 265

1048 65. Hu N, Wang C, Hu Y, Yang HH, Giffen C, Tang ZZ, Han XY,1049 Goldstein AM, Emmert-Buck MR, Buetow KH, Taylor PR, Lee

1050 MP (2005) Genome-wide association study in esophageal cancer

1051 using GeneChip mapping 10 K array. Cancer Res 65:25422546

1052 66. Huang H, Mahler-Araujo BM, Sankila A, Chimelli L, Yonekawa

1053 Y, Kleihues P (2000) APC mutations in sporadic medulloblas-

1054 tomas. Am J Pathol 156:433437

1055 67. Huang J, Wei W, Zhang J, Liu G, Bignell GR, Stratton MR,

1056 Futreal PA, Wooster R, Jones KW, Shapero MH (2004) Whole

1057 genome DNA copy number changes identified by high density

1058 oligonucleotide arrays. Hum Genomics 1:287299

1059 68. Hui AB, Takano H, Lo KW, Kuo WL, Lam CN, Tong CY,

1060 Chang Q, Gray JW, Ng HK (2005) Identification of a novel

1061 homozygous deletion region at 6q23.1 in medulloblastomas

1062 using high-resolution array comparative genomic hybridization

1063 analysis. Clin Cancer Res 11:470747161064 69. Ichimura K, Ohgaki H, Kleihues P, Collins VP (2004) Molecular

1065 pathogenesis of astrocytic tumors. J Neurooncol 70:137160

1066 70. Inazawa J, Inoue J, Imoto I (2004) Comparative genomic

1067 hybridization (CGH)-arrays pave the way for identification of

1068 novel cancer-related genes. Cancer Sci 95:559563

1069 71. Ishii N, Sawamura Y, Tada M, Daub DM, Janzer RC, Meagher-

1070 Villemure M, de Tribolet N, Van Meir EG (1998) Absence of

1071 p53 gene mutations in a tumor panel representative of pilocytic

1072 astrocytoma diversity using a p53 functional assay. Int J Cancer

1073 76:797800

1074 72. Ishkanian AS, Malloff CA, Watson SK, DeLeeuw RJ, Chi B,

1075 Coe BP, Snijders A, Albertson DG, Pinkel D, Marra MA, Ling

1076 V, MacAulay C, Lam WL (2004) A tiling resolution DNA

1077 microarray with complete coverage of the human genome. Nat

1078 Genet 36:299303

1079 73. Jeon IS, Davis JN, Braun BS et al (1995) A variant Ewings

1080 sarcoma translocation (7;22) fuses the EWS gene to the ETS

1081 gene ETV1. Oncogene 10:12291234

1082 74. Johnson RL et al (1996) Human homolog of patched, a candidate

1083 gene for the basal cell nevus syndrome. Science 272:16681671

1084 75. Kampa D, Cheng J, Kapranov P, Yamanaka M, Brubaker S,

1085 Cawley S, Drenkow J, Piccolboni A, Bekiranov S, Helt G,

1086 Tammana H, Gingeras TR (2004) Novel RNAs identified from

1087 an in-depth analysis of the transcriptome of human chromosomes

1088 21 and 22. Genome Res 14:331342

1089 76. Kaneko Y, Yoshida K, Handa M et al (1996) Fusion of an ETS-

1090 family gene, E1AF, to EWS by t(17;22)(q12;q12) chromosome

1091 translocation in an undifferentiated sarcoma of infancy. Genes

1092 Chromosomes Cancer 15:115121

1093 77. Kang DE, Soriano S, Xia X, Eberhart CG, De Strooper B et al

1094 (2002) Presenilin couples the paired phosphorylation of beta-

1095 catenin independent of axin: implications for beta-catenin

1096 activation in tumorigenesis. Cell 110:751762

1097 78. Kapranov P, Cawley SE, Drenkow J, Bekiranov S, Strausberg

1098 RL, Fodor SP, Gingeras TR (2002) Large-scale transcriptional

1099 activity in chromosomes 21 and 22. Science 296:916919

1100 79. Kennedy GC, Matsuzaki H, Dong S, Liu WM, Huang J, Liu G,

1101 Su X, Cao M, Chen W, Zhang J, Liu W, Yang G, Di X, Ryder T,

1102 He Z, Surti U, Phillips MS, Boyce-Jacino MT, Fodor SP, Jones

1103 KW (2003) Large-scale genotyping of complex DNA. Nat

1104 Biotechnol 21:12331237

180. Korshunov A, Savostikova M, Ozerov S (2002) Immunohisto-

1chemical markers for prognosis of average-risk pediatric medul-

1loblastomas. The effect of apoptotic index, TrkC, and c-myc

1expression. J Neurooncol 58:271279

181. Kovar H, Jug G, Aryee DN et al (1997) Among genes involved

1in the RB dependent cell cycle regulatory cascade, the p16 tumor

1suppressor gene is frequently lost in the Ewing family of tumors.

1Oncogene 15:22252232

182. Kramer DL, Parmiter AH, Rorke LB, Sutton LN, Biegel JA

1(1998) Molecular cytogenetic studies of pediatric ependymomas.1J Neurooncol 37:2533

183. Krzywinski M, Bo

bansal.molecular biology cns

Documents