Vol. 6, 239-243, April !997 Cancer Epidemiology, Biomarkers & Prevention 239

3 The abbreviations used are: NNC. N-nitroso compound: PNET. primitive neu-

roectodermal tumor; NMU, N-nitroso-N-methylurea.

ras Oncogene Mutations in Childhood Brain Tumors’

Terese H. Maltzman, Beth A. Mueller, Jesara Schroeder,Joe C. Rutledge, Kathy Patterson, Susan Preston-Martin,and Elaine M. Faustman2

Division of Gastroenterology, University of Colorado Health Sciences Center,

Denver, Colorado, 80262 [T. H. Ml; Fred Hutchinson Cancer Research

Center, Seattle, Washington 98006 [B. A. Ml; Departments of Epidemiology

[B. A. M.l and Environmental Health [J. S., E. M. F.l and Center onHuman Development and Disability [E. M. Fl. University of Washington,

Seattle, Washington 98195; Department of Laboratory Medicine. Children’sHospital Medical Center, Seattle, Washington 98105 [J. C. R.. K. P.1; and

Department of Preventive Medicine, University of Southern California,

Los Angeles, California 90007 [S. P-M.l

Abstract

Although N-nitroso compounds (NNC) are ubiquitous in thehuman environment and are known neurocarcinogens inanimal models, results of epidemiological studies have notyet convincingly associated NNCs with brain tumoroccurrence in humans. Animal studies have suggested thatspecific codons (12, 13, and 61) in the ms family aremutable by exposure to NNCs. The purpose of this studywas to measure the presence of mutations in the ras familyof oncogenes in tissue from childhood brain (CB) tumors asa preliminary step toward investigating their potential useas biomarkers of chemical exposure. DNA was extractedfrom paraffin-embedded formalin-fixed CB tumors fromtissues resected during neurosurgical operations. Using thePCR, designed RFLP-screening methods, and sequencing,we attempted to screen brain tumors from 46 children forthe presence of H, K, and N-ms mutations at codons 12, 13,and 61. Screening for oncogene mutations using PCR,RFLP methods, and DNA sequencing was successfullycompleted for a high proportion of the available specimens.Astrocytoma specimens from three children for whomscreening with PCR was successfully completed were foundto contain CAA -� GAA point mutations in K-ms at codon61. None of the specimens contained mutations at any of theother locations. These results, although preliminary, providea potential clue for future mechanistic studies of CB tumors.The possible roles of NNCs in inducing this mutation, or ofthis mutation as an early or late event in tumor progression,however, remain unclear.

Introduction

ras oncogene mutations of varying types have been detected inapproximately 20-25% of all human tumors (1-4), with spe-

Received 7/12/96; revised 1/6/97; accepted 1/6/97.

The costs of publication of this article were defrayed in pars by the payment of

page charges. This article must therefore be hereby marked advertisement in

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

I Supported by Grants ROl-NIHES-03157, l-RO1CA-47082 Nd, and T32-

ES07262 and a Royalty Research Fund grant from the University of Washington.

2 To whom requests for reprints should be addressed, at Department of Environ-

mental Health, 4225 Roosevelt Way NE, Suite 100, University of Washington,

Seattle, WA 98195.

cific mutation patterns more commonly associated with tumorsof different types and at different locations (1, 5). NNCs3 have

been shown to induce tumors in several animal models (6, 7)and ras activation has been recognized as a potential mutationalevent in many of these tumors (8-11). This animal evidence,

coupled with evidence of specificity of ras mutations by tumortype, suggests that NNCs may be associated with tumorigenesis

in humans, and that ras oncogene mutations may have potentialas biomarkers of NNC exposure in humans.

Despite the convincing animal evidence of an NNC-brain

tumor association via transplacental exposure in young rodents(7, 9, 1 2), the causes of brain tumors in humans remain largely

unknown. If ras oncogene mutations similar to those observedin NNC-induced tumors in rodents are detected in childhoodbrain tumors, this may provide evidence for an NNC-childhoodbrain tumor association via a similar pathway.

Materials and Methods

Subjects. Paraffin-embedded formalin-fixed brain tissue speci-mens were obtained from 46 children who had undergone neuro-surgical operations for brain tumors at Children’s Hospital Mod-ical Center in Seattle, WA. All pathology specimens werereviewed by a neuropathologist and were classified by histology.

A majority of these children (34, or 74%) were also involved in a

large, multicenter case-control study of risk factors for childhoodbrain tumors diagnosed between 1984 and 1990; for these, demo-graphic information was also available. For 12 of the children,however, no demographic details were available because their

specimens were accessed from a general pool of tissue blockscomprised of excess specimens obtained routinely during opera-tions and made available for research without access to patient-identifying information. Tumors were classified based on the neu-

ropathology report as astroglial (ICD-9 9380-9384, 9400-9421,and 9424-9442), PNET (lCD 9362, 9470-9473, and 9500), or

other. Access to specimens and laboratory procedures for thisproject was approved by the Institutional Review Boards at theFred Hutchinson Cancer Research Center and Children’s HospitalMedical Center (Seattle, WA).

DNA Extraction. Paraffin-embedded formalin-fixed brain tis-sue specimens were assayed for mutations in ras genes. In the

laboratory, l0-40-�tm-thick sections of the formalin-fixed par-affin tissue blocks were cut and stored in sterile microfuge

tubes for DNA extraction. Extraction of DNA involved a mod-ification of a method described previously (13). Briefly, sam-pies were deparaffinized using one extraction with xylene,followed by one rinse with 100% ethanol and then dried downin a 55#{176}Cheat block. Tissues were then resuspended in a buffer

containing 7.5 mM Tris-HCI (pH 8.5), 37.5 mr�i KC1, I msiMgCl2, 0.5% Tween 20, and 20 �g/ml protemnase K and incu-bated in a 55#{176}Cheat block for at least 3 h. Samples containing

on June 20, 2021. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

ASCA TG BUCA TG

I

I

� I� I

�

240 ras Mutations in Childhood Brain Tumors

Fig. !. N-n,s. codon 61. Due to lack of an

appropriate primer design/restriction site for

codon 61 of N-ras, PCR amplification of a

portion of the gene surrounding codon 61

and direct sequencing were used for the de-

termination of mutations at this site. A, con-

trol DNA without mutation at codon 61(lymph). B, control DNA with mutation at

codon 61 (HL-60). Arrow, A to T mutation.

more tissue were incubated overnight. Samples were then

boiled for 10 mm, centrifuged, and then used for PCR or storedat 4#{176}C.Some samples were extracted according to a modified

protocol (14) including a DNA purification step ofan extractionwith phenol:chloroform:isoamyl alcohol (25:24: 1) and one of

chloroform following the digestion step above.

RFLP and Sequencing to Detect ras Mutations. Analysis ofsamples for ras mutations were performed using a modifiedscreening method for mutations in the ras genes ( 15). This

method used PCR and designed primers that created RFLP todistinguish between mutated and nonmutated ras genes. The

analysis was designed to look at mutations at the 12th, 13th, and61st codons of H-, K-, and N-ras. For this study, 1-10 �l of thesample extractions were used to amplify products. Generally,the PCR conditions were initial denaturation at 94#{176}Cfor 2 mm

followed by 30-40 cycles of denaturation (94#{176}Cfor 1 mm),annealing (at the specific temperature for the primer set, 1 mm),and extension (72#{176}Cfor I mm). After PCR, the products were

then subjected to the appropriate restriction enzyme digestionand run on 2-4% agarose gels or 8% polyacrylamide gels toobserve cutting patterns. This assay was designed so that ifthere were mutations, the restriction enzyme would not cut thePCR product. Samples that were not cut completely by therestriction enzyme (indicating mutation in one allele) were gel

purified and sequenced for confirmation.The primers originally designed for RFLP to detect mu-

tations at codon 61 of N-ras did not amplify efficiently becausethere was a mismatch at the 3 ‘ end of the sense primer. Becauseof this problem, we chose to directly sequence the PCR product.A new sense primer was designed which did not contain any

mismatches (5’-gat tct tac aga aaa caa gtg-3’). Using this primerand the previous original antisense primer, PCR amplificationwas performed at an annealing temperature of 54#{176}C.The am-

plified product was purified by electrophoresis on a 2% agarose

gel, excised, and centrifuged in a 0.2 /LM Spin-X column(Costar). PCR cycle sequencing was accomplished using thedideoxy termination method by end labeling one of the primers(with [y-32P]ATP) using the fmol Sequencing System (Pro-

mega). An example of this sequencing is given in Fig. I.Screening for H-ras-61 required a heminest strategy be-

Table I Demographic and disease characteristics of 46 children with brain

tissue specimens assayed for the presence of ras oncogene mutations

Characteristics No. %

Age at diagnosis” (yr)

Cancer Epidemiology, Biomarkers & Prevention 24/

Table 2 Results of r u.s oncogene mutation ass ays on 56 brain tissue speci mens from children wit h brain tumors

Codon

ras Family

N- K-

12

H-

13 61

Total

12 13 61 12 13 61

No. of screens attempted

Inadequate quantity DNA

Unable to amplify

Negative

Positive

56

0

I

55

0

56

0

5

51

0

56

3

0

53

0

56

0

2

54

0

56

0

0

56

0

56

0

0

53

3

56

0

3

53

0

56

0

0

56

0

56

0

4

52

0

5t�4

3

IS

483

3

Results

General Characteristics of Subjects. We attempted to assaya total of 56 paraffin-embedded brain tissue specimens from 46

children for the presence of ras family oncogene mutations.Nearly one-half (47. 1%) of the children for whom demographic

information was available were less than 5 years old; 38.2%were between the ages of 5 and 9 years (Table 1). The majority(58.8%) were male. Over one-half of the specimens were fromastrocytomas and glial tumors (54.3%); approximately one-

quarter (26.1%) were PNETs.

Results of Assay for ras Mutations. A total of 56 specimenswere obtained; 37 of the 46 children had a single tissue block

available for assay, 8 children had two specimen blocks, and 1child had three specimen blocks. Five of the 56 specimens

contained nontumor brain tissue resected during the neurosur-gical operation.

Although we attempted to screen all specimens from tissue

blocks for K-, N-, and H-ras mutations at codons 12, 13, and 61

(a total of 9 oncogene screens which included 1 1 separate assaysfor each specimen), not all specimens contained a sufficient quasi-tity or adequate quality of DNA for successful PCR amplification,resulting in a slightly lower number of completed assays for each

ras mutation screen (Table 2). Of the 504 total ras mutationscreening procedures attempted, 3 (or less than 1%) were aborted

because an inadequate quantity of DNA was available for assay.We were successfully able to complete 486 (97%) of the remain-

ing 501 screens; PCR amplification did not occur for 15.Mutations in the K-ras oncogene at codon 61 were present

in three specimens. These were observed in astrocytoma tumortissues resected in first operations for primary tumors of three

children, all female, ages 4, 5, and 12 years. Specimens fromthe 5- and 12-year olds were both from low-grade fibrillaryastrocytomas located in the cerebellum and brainstem, respec-

tively. Specimens from the 4-year-old were from a pilocyticastrocytoma in the ventricular region. The two younger children

each had two astrocytoma tumor specimens screened, only oneof which contained the K-ras mutation. Testing of these sam-

pies was repeated to confirm these results. None of thesechildren had nontumor brain tissue available for assay. Theresult of the RFLP demonstrating this mutation in one of thesespecimens is shown in Fig. 2. Subsequent sequencing analysis(Fig. 3) indicated that all three mutations were identical andconsisted of CAA -� GAA transversions. No N- or H-ras orother K-ras mutations were detected at any of the three codons

examined (codon 12, 13, or 61) in any of the other specimens.

An example of the RFLP assay is presented in Fig. 4.

Discussion

To complete the screening for mutations in codons 1 2, 13, and

61 of the H-, K-, and H-ras oncogenes, at least 10 PCRamplifications per specimen were necessary. Some samples

Fig. 2. K-ras. codon 61 . Four percent agarose gel demonstrating the use ofRFLP. Lanes 2, 4, 6. and 8, amplified DNA incubated with the restriction enzyme

Bell; Lwtes I, 3, 5. 7. and 9, amplified DNA alone. Lane 1. molecular weight

ladder; Lanes 2 and 3, control DNA without mutation at codon 61 (lymph); Lam’s

4 and 5, sample 30; Lanes 6 and 7. sample 3 1 : and Lanes 8 and 9. sample 34. Note

that Lane 4 has two bands at about half the intensity of Lane 2 or 6. indicating a

mutation in one of the alleles of K-ras-61.

would not amplify specific PCR products, possibly due to thepresence of badly degraded DNA; however, we were able to

complete our battery of assays for 97% of the specimens.

Mutations at K-ras-61 were detected in specimens from threechildren; all contained the same mutation, a C to G transversionin the first nucleotide of codon 61, substituting glutamic acid

for glutamine.The ras family of oncogenes appears to play a role for human

tumors at several sites ( 1 ), and ras mutations have been shown toresult from NNC exposure in laboratory animals (8, 9). Mutatedras proteins have lost the ability to be turned off and thus stimulategrowth or differentiation autonomously. Point mutations identified

to date that lead to ras gene activation appear to be specific andinclude those screened at codons 12, 13 and 61 ( 16). ras oncogenemutations have been detected in varying proportions of humantumors screened, ranging from 10 to 90% depending on tissue type(2). Relatively high prevalences of ras mutation have been ob-

served in adenocarcinomas of the pancreas, colon, and lung (3, 4,17). Of the three ras family oncogenes, activated K-ras has been

detected most often in human tumors, and it has been suggestedthat this is due to exposure to specific environmental agents ( I).

The pattern of ras mutations observed in several chemically in-

duced rodent tumors indicates that certain chemicals may exhibitspecificity for mutation occurrence at specific sites on the gene( I 0, 1 1, I 8, 19), raising the potential for use of ras mutations asbiomarkers for selected chemical exposures.

Given the variety of oncogene characteristics examinedand the wide range of patient ages included in previous studies

on June 20, 2021. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

con 1 2

,�

�

-

U

p � *l� -

242 ms Mutations in Childhood Brain Tumors

of brain tumors, it is difficult to characterize the role of the rasoncogenes in childhood brain tumors. Overexpression of H- orN-ras was reported in 0, 43, and 71% of low-, intermediate-,and high-grade astrocytomas (20), and increased expression ofN-ras was measured in all tumors in a series of five humanglioblastomas assayed (2 1 ). However, amplification or rear-rangement of N-ras was not observed in a separate series of 65astrocytic tumors (22). These previous studies were not, how-

ever, restricted to childhood tumors. Previous reports suggest

that the spectrum of chromosomal abnormalities (23) and on-cogene amplification (24) differs between brain tumors ofadults and children, although neither of these focused specifi-

cally on ras oncogenes. To our knowledge, our report repre-sents the first systematic characterization of mutations in theH-, K-, and N-ras oncogenes to date.

Our results suggest that K-ras-61 mutations may be present

in a measurable proportion of childhood brain tumors. Three of the46 children in our series, or nearly 7%, had tumor specimenscontaining K-ras mutations: 3 of 25 (12%) of those with a.stroglial

tumors. These levels appear to be lower than the proportions of rasmutations reported for tumors from other sites. In an earlier study,no ras mutations were detected in any of 18 neuroblastomas,

sarcomas of neural origin affecting mostly infants and children upto age 10 years (25). More recently, a mutation in K-ras at codon12 (G to A transition) was found in a PNET: however, this was

thought to be a result of therapeutic radiation to the central nervoussystem (26). It would be of interest to compare the levels and

specific type of mutation we observed with those measured inother series of childhood brain tumors.

It should be noted that two of the three children whosespecimens contained mutations each had two tumor specimensscreened, of which only one was positive for K-ras-6 1 muta-

tions. Although it is possible that the lack of a mutation ob-served in the second specimen of these two children might be

artifactual, perhaps the result of DNA degradation, we excludedfrom our results specimens for which PCR amplification was

unsuccessful, reducing this likelihood. Furthermore, specimenscontaining mutations were rescreened to verify our results, andthus our positive findings are unlikely to be due to screening

error. For these two children with two tumor specimensscreened, this suggests that not all of the cells within these

tumors contained this mutation, a finding that may be indicativeof heterogeneity present within these tumors. Other possible cx-planations are that this mutation arose a.s a late event in tumor

progression, or that the collection techniques for removing braintissues also removed some adjacent nontumor tissues. The K-ras-61 mutation that we did see is, however, compatible with anhypothesis of a C to G change occurring due to NNC-induced

Fig. 3. K-ras. codon 61. Sequencing analysis of two of themutated samples and wild-type control DNA. Changed Se-

quence )G to C) can be seen at the (Arrow: in sample 2. two

hands of equal intensity are seen in the G and C lanes. whereas

sample I has one band at C and the control DNA has a single

hand in the G lane.

Fig. 4. K-ras. codon 12. Amplified DNA that has been digested with Bail and

run on a 4’7c agarose gel. Lanes 1. 3. 5. 7. and 9 are those with PCR product that

has been incubated with the restriction enzyme BanI. Lane.s 2. 4. 6. and 8 havethe same PCR product without the restriction enzyme. Lane I. molecular weightladder: Lanes 2 and 3. control DNA without mutation at codon 12 (lymph): Lanes

4 and 5. control DNA with a mutation at codon I 2 (Calu- I ): Lanes 6 and 7. sample

I 03- 10: and La,tes 8 and 9. sample I 03- I I.

mutational changes in C, possibly as a result of 3-alkyl or 02-alkyl

DNA lesions. It would be interesting to know whether or notsimilar mutations were also present in the nontumor brain of these

three children; unfortunately, such specimens were not available.Mutations were not observed in any of the five specimens con-taming nontumor brain tissue from other children. Future studies

examining the presence of mutations in brain specimens of youngneurosurgery patients without tumor, possibly from epilepsy pa-

tients, would help to clarify further the role of these mutations inchildhood brain tumors. Investigations ofthe potential relationship

between exposure to NNCs and the occurrence of K-ras-61 mu-tations would help us understand whether these mutations mayhave resulted from environmental exposures.

Although the causes of K-ras-61 and other ras mutations

in humans have not been identified, it has been suggested thatthey may be due to exposure to specific agents in the environ-ment such as NNCs (27). For example, the majority of ras

mutations observed in colon cancer are G to A transitions at thesecond G of a GG pair. a mutation characteristic of exposure to

alkylating agents such as NNC (2). Most K-ras mutations inlung cancer are G to T transversions, a finding consistent withmutation patterns associated with mutagenic components in

NNC-containing tobacco smoke (2). However, the levels ofthese mutations occurring in nontumor tissues is less frequently

on June 20, 2021. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

Cancer Epidemiology, Biomarkers & Prevention 243

considered and of importance as well. One frequently observedH-ras- 1 2 mutation in NMU- induced rat mammary tumors

(aGGA to GAA mutation) was also measured in the mammaryepithelium of 70% of the untreated rats (28). This level was notsignificantly altered following NMU exposure, leading the au-thors to conclude that it was an independent effect of NMU ontumor formation. An epigenetic mechanism for NMU has been

proposed whereby NMU-induced alterations in promoter con-figuration irreversibly deregulated H-ras- 1 expression and

causing mammary carcinogenesis (29). A similar scenario iscertainly possible for the K-ras-61 mutations we observed inbrain tumors; unfortunately, we can gain no further insight fromthe five nontumor brain specimens we assayed with negativeresults. It will be of interest to expand our work to a larger

number of children for whom both tumor and normal tissues areavailable and for whom epidemiological data exist for potential

dietary and environmental exposures to NNCs.Application of this RFLP assay that was developed to screen

for ras mutations in tumor cell lines appeared to work for archivedformalin-fixed paraffin-embedded tissue sections despite somelimitations. Although the majority of the RFLPs were conducted

successfully with our specimens, two procedural modificationswere necessary. The first modification involved screening muta-tions at N-ras-61; the primers originally designed did not amplify

efficiently due to a mismatch at the 3 ‘ end of the sense primer.

Although the original procedure might have amplified DNA fromtumor cell lines, it would not do so with formalin-fixed paraffin-

embedded tissue sections, likely because of a somewhat degradedquality of DNA. Because of this, we directly sequenced the PCRproduct. The second modification was necessary to conduct

screening for H-ras-61 . Originally, a heminested procedure wasattempted wherein a larger PCR fragment was first amplifiedfollowed by a second amplification using this larger fragment as atemplate, a procedure that resulted in inefficient amplification.

This was dealt with by eliminating the heminest step and directlyamplifying the PCR product at a lower annealing temperature,55#{176}Cinstead of 60#{176}C.Presumably, this second modification wasnecessary because of possible degradation of the formalin-fixed

paraffin-embedded tissue sections.Despite the fact that in some instances as many as 10 years

had elapsed since the subjects’ neurosurgical operations, the ma-

jority (486/504, or 96%) of the screens performed using formalin-fixed paraffm-embedded tumor sections in this study yielded aPCR product. This was likely due to the fact that the size of thePCR products were relatively small, less than 300 bp. Because theDNA is often partly degraded from tumor necrosis and cross-

linked from formalin fixation, it may be difficult to successfullyamplify larger products using DNA from formalin-fixed paraffin-embedded tissue sections. For detection of ras oncogene mutationsin archived childhood brain tumor specimens, however, this tech-

nique appears to be feasible and could be used in future large-scale

epidemiological studies.

References

1 . Bos, J, L. ras oncogenes in human cancer: a review. Cancer Res., 49:4682-4689, 1989.

2. Bos, J. L. The ras gene family and human carcinogenesis. Mutat. Res.. 195:

255-271, 1988.

3. Almoguera, C., Shibata, D., Forrester, K, Martin, J., Arnheim, N., and

Perucho, M. Most human carcinomas of the exocrine pancreas contain mutant

c-K-ras genes. Cell, 53: 549-554, 1988.

4. Forrester, K., Almoguera, C., Han, K., Grizzle, W. E., and Perucho, M.Detection of high incidence of K-ras oncogenes during human colon tumorigen-

esis. Nature (Lond.), 327: 298-303, 1987.

5. Garte, S. J., Hood, A. 1., Hochwalt, A. E., D’Eustachio, P. D., Snyder, C. A.,

Segal, A., and Albert, R. E. Carcinogen specificity in the activation of transform-

ing genes by direct-acting alkylating agents. Carcinogenesis (Lond.), 6: 1709-

1712, 1985.

6. Anderson, L. M., Higiwara, A., Kovatch, R. M., Rehm, S., and Rice, J. M.

Transplacental initiation of liver lung, neurogenic, and connective tissue tumors

by N-nitroso compounds in mice. Fundam. AppI. Toxicol., 12: 604-620, 1989.

7. Bilzer, T., Reifenberger, G., and Wechsler, W. Chemical induction of brain

tumors in rats by nitrosoureas: molecular biology and neuropathology. Neuro-

toxicol. Teratol., 11: 551-556, 1989.

8. Ohgaki, H., Kleihues, P., and Hard, G. C. Ki-ras mutations in spontaneous and

chemically induced renal tumors of the rat. Mol. Carcinog., 4: 455-459, 1991.

9. Sukumar, S., and Barbacid, M. Specific patterns of oncogene activation in

transplacentally induced tumors. Proc. NatI. Acad. Sci. USA, 87: 718 -722, 1990.

10. Sukumar, S., Notario, V., Martin-Zanaca. D., and Barbacid, M. Induction of

mammary carcinogens in rats by nitroso-methylurea involves malignant activation of

H-ras-l locus by single point mutations. Nature (Land.), 306: 658-661. 1983.

I I . Fujita. J., Ohuchi, N., Ito, N., Reynolds, S. H., Yoshida, 0., Nakayama, H.,

and Kitamura, Y. Activation of H-ras oncogene in rat bladder tumors induced by

N-butyl-N-(4-hydroxybutyl)nitrosamine. J. Natl. Cancer Res., 80: 37-42, 1988.

12. Rice, J. M., Rehm, S., Donovan, P. J., and Perantoni, A. 0. Comparative

transplacental carcinogenesis by directly acting and metabolism-dependent alky-

lating agents in rodents and nonhuman primates. In: N. P. Napalkov, J. M. Rice,

L. Tomatis, and H. Yamasaki (eds.), Perinatal and Multigeneration Carcinogen-

esis. Lyon, France: IARC Publications, 1989.

13. Optimization of PCRs. PCR Protocols: A Guide to Methods and Applica-

tions, Chap. 19, pp. 153-158. San Diego: Academic Press, 1990.

14. Kosel, S. Use of neuropathological tissue for molecular genetic studies:

parameters affecting DNA extraction and polymerase chain reaction. Acts Neu-

ropathol., 88: 19-25, 1994.

15. Mitsudomi, T., Viallet, J., Mushine, R., Ilona, L.. Minna, J. D., and Gazdar,

A. F. Mutations of ras genes distinguish a subset of non-small-cell lung cancer

cell lines from small-cell lung cancer cell lines. Oncogene, 6: 1353-1362, 1991.

16. Barbacid, M. ras genes. Annu. Rev. Biochem., 56: 779-827, 1987.

17. Mitsudomi, T., Steinberg, S. M., Oie, H. K., Mulshine, J. L., Phelps, R.,

Viallet, J., Pass, H., Minna, J. D., and Bazdar, A. F. ras gene mutations in

non-small cell lung cancers are associated with shortened survival irrespective of

treatment intent. Cancer Res., 52: 4999-5002, 1991.

18. lopol. M. D. DNA repair. oncogenes and carcinogenesis. Carcinogenesis

(Land.), 9: 691-696, 1988.

19. Sukumar, S., Perantoni, A., Reed. C., Rice, J. M., and Wenk, M. L. Activated

K-ras and N-ms oncogenes in primary renal mesenchymal tumors induced in P344

rats by methyl(methosymethyl)nitrosamnine. Mol. Cell. Biol., 6: 2716-2720, 1986.

20. Orian, J. M., Vasilopoulos, K., Yoshida, S., Kaye, A. H., Chow, C. W, and

Gonzales, M. F. Overexpression of multiple oncogenes related to histological

grade of astrocytic glioma. Br. J. Cancer, 66: 106-1 12, 1992.

21. Gerosa, M. A.. Talarico, D., Fognani, C.. Raimondi, E.. Colombatti, M.,

Tridente, G.. Dc Carli, L., and Della Valle, G. Overexpression of N-ras oncogeneand epidermal growth factor receptor gene in human glioblastomas. J. NatI.

Cancer Inst., 81: 63-67, 1989.

22. Lang, F. F., Miller, D. C., Koslow, M., and Newcomb, E. W. Pathways

leading to glioblastoma multiforme: a molecular analysis of genetic alterations in

65 astrocytic tumors. J. Neurosurg., 81: 427-436, 1994.

23. Griffin, C. A., Hawkins, A. L., Packer, R. J., Rorke. L. B., and Emanuel, B.

S. Chromosome abnormalities in pediatric brain tumors. Cancer Res., 48: 175-

180, 1988.

24. Watson, J. C., Saylors, R. L., Zeltzer, P. Friedman, H. S., Bigner, S. H., Burger,P. C., Bigner, D. D., Look, A. T., Douglass, E. C., and Brodeur, GM. Oncogene

amplification in pediatric brain tumors. Cancer Res., 50: 2987-2990, 1990.

25. Ballas, K., Lyons, J., Janssen, J. W. G., and Bartram, C. R. Incidence of rasgene mutations in neuroblastoma. Eur. J. Pediatr.. 147: 313-314, 1988.

26. Brustle, 0., Ohgaki, H., Schmitt, H. P., Walter, G. F., Ostertag, H., andKleihues, P. Primitive neuroectodermal tumors after prophylactic central nervous

system irradiation in children. Cancer (Phila.), 69: 2385-2392, 1992.

27. Preston-Martin, S., Yu, M. C., Benton, B., and Henderson, B. E. N-nitroso

compounds and childhood brain tumors. Cancer Res., 42: 5240-5245, 1982.

28. Cha, R. S., Thilly, W. G., and Zarbl, H. N-nitroso-N-methylurea-induced rat

mammary tumors arise from cells with preexisting oncogenic Hrasl gene muta-tions. Proc. Natl. Acad. Sci. USA. 91: 3749-3753. 1994.

29. Zhuang, J., Houle, B., Mikheev, A. M., Cha, R. S., and Zarbl, H. Alterations

in H-rasl promoter conformation during N-nitroso-N-methylurea-induced mam-

mary carcinogenesis and pregnancy. Cancer Res., 56: 4927-4935, 1996.

on June 20, 2021. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

1997;6:239-243. Cancer Epidemiol Biomarkers Prev T H Maltzman, B A Mueller, J Schroeder, et al. Ras oncogene mutations in childhood brain tumors.

Updated version

http://cebp.aacrjournals.org/content/6/4/239

Access the most recent version of this article at:

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

Subscriptions

Reprints and

.pubs@aacr.orgDepartment at

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

Permissions

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

.http://cebp.aacrjournals.org/content/6/4/239To request permission to re-use all or part of this article, use this link

on June 20, 2021. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/content/6/4/239http://cebp.aacrjournals.org/cgi/alertsmailto:pubs@aacr.orghttp://cebp.aacrjournals.org/content/6/4/239http://cebp.aacrjournals.org/