[CANCERRESEARCH57. 3126-3130,August1. 19971

Advances in Brief

Abrogation of the Rb/p16 Tumor-suppressive Pathway in Virtually AllPancreatic Carcinomas'

Mieke Schutte, Ralph H. Hruban, Joseph Geradts, Rob Maynard, Werner Hilgers, Sridhar K. Rabindran,Christopher A. Moskaluk, Stephan A. Hahn, Irmgard Schwarte-Waldhoff, Wolff Schmiegel, Stephen B. Baylin,Scott E. Kern,2 and James G. HermanDepartments of Oncolog',' (M. S., R. H. H., 5. B. B., S. E. K., J. G. H.J and Pathology (R. H. H., W. H., C. A. M., S. E. K.J, The Johns Hopkins Medical Institutions. Baltimore,Maryland 21205: Department of Patholog', and Lijboratorv Medicine, University of North Carolina School of Medicine. Chapel Hill, North Carolina 27599 (J. G., R. MI;Oncology/Immunology Research, Wyeth Ayerst Research, Pearl Riser, New York 10965 (5. K. RI; and Medical Clinic IMBL. Ruhr-University Bochum, Bochum, Germany(S. A. H., I. S. W., W. 5.1

Abstract

The Rb/p16 tumor-suppressive pathway is abrogated frequentlyin human tumors, either through inactivation of the Rb or

@ tumor-suppressor proteins, or through alterationor overexpression of the cyclin Dl or cyclin-dependent kinase 4 oncoproteins. We reported previously that the p16 gene was geneticallyinactivated in 82% of pancreatic carcinomas. Nearly half of theseinactivations were by intragenic mutation of p16, and the remainderwere by homozygous deletion of the gene. Here, we analyzed pancreaticcarcinomas for additional mechanisms by which the Rb/p16 pathwaymight be inactivated.

Transcriptional silencing of the p16 gene in association with methylation of its 5'-CpG island was examined by methylation-specific PCR in 18pancreatic carcinomas. Nine of these were known to harbor an intragenicmutation in p16, and nine had a wild-type p16 coding sequence Seven ofthe 18 tumors were hypermethylated, and all 7 were p16 wild-type

(P = 0.001). Complete silencing of transcription from methylated wildtype gene sequences was demonstrated. Immunohistochemical analysis

revealed normal expression levels of the Rb protein in all carcinomasstudied. None of the carcinomas had genomic amplification of the cycinDl or CDK4genes, and none had mutation of the p16-binding domain ofCDK4. An additional p16 mutation was identified.

In total, the Rb/p16 pathway was abrogated in 49 of the 50 carcinomas(98%) studIed, all through Inactivation of the pitS gene. Similar resultswere obtained in an independently analyzed series of 19 pancreatic carcinomas. These data demonstrate the central role of the Rb/p16 pathwayin the development of pancreatic carcinoma.

Introduction

Experimental models have suggested that both the Rb3 andp16INK4@@@@T5 I tumor-suppressor proteins function in a distinctcell cycle-regulatory pathway. In normal cells, progression throughthe G1 phase of the cell cycle requires functional inactivation of Rbthrough its phosphorylation by a complex of cyclin Dl and Cdk4 or

Received 5/6/97; accepted 6/12/97.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

@ This work was supported by the Specialized Programs of Research Excellence in

Gastrointestinal Cancer and Lung Cancer (NIH Grants CA62924 and CA58184, respectively). by the Deutsche Krebshilfe (Grant 10-1 137-Hal), and by the Ruhr-UniversitatBochum (Grant RUB-Med-FO-808004). S. E. K. is a McDonnell Foundation Scholar.S. B. B. and J. G. H. receive research funding and are entitled to sales royalties fromONCOR, which is developing products related to research described in this paper. Theterms of this arrangement have been reviewed and approved by The Johns HopkinsUniversity in accordance with its conflict of interest policies.

2 To whom requests for reprints should be addressed, at Department of Oncology, TheJohns Hopkins Medical Institutions, Baltimore, MD 21205-2196. Phone: (410) 614-3314;Fax: (410) 614-0671.

3 The abbreviations used are: Rb and RBJ, retinoblastoma protein and gene; Cdk4 and

CDK4, cyclin-dependent kinase 4 protein and gene; LOH, loss of heterozygosity; MSP,methylation-specific PCR; RT-PCR, reverse transcription-PCR.

cyclin-dependent kinase 6. The p16 protein may act as a cyclindependent kinase inhibitor by binding competitively to Cdk4 andthereby preventing the interaction of Cdk4 with cyclin Dl (1). Theinactivation of either the RBJ or p16 gene would, therefore, abrogatean important pathway for inhibiting cell growth and thus promotetumorigenesis. In addition, the mutational abolition ofthe p16-bindingdomain of Cdk4 or the overexpression of either cycin Dl or Cdk4could negate the growth-inhibitory function of p16. The functionalconnections among Rb, p16, cyclin Dl, and Cdk4 imply that abrogation of the Rb/pl6 pathway through alteration of any of these fourproteins should result in similar effects upon tumor progression. Onewould also predict that subsequent additional alterations of othermembers of this pathway would not provide further selective growth

advantage to a neoplastic clone. Analyses of a variety of humancancers have indeed revealed a pattern of pathway inactivation in

which only one of the four members of the Rb/pl6 pathway isinactivated in a given tumor (Ref. 2 and references therein).

We reported previously that the pitS gene is genetically inactivatedin 82% of pancreatic carcinomas (3, 4). To further evaluate the fullextent of involvement of the Rb/p16 pathway in pancreatic tumorigenesis, we extended our investigation to examine additional mechanisms by which all four members of the Rb/p16 pathway could beinactivated.

Materials and Methods

Tumor Samples. Pancreatic adenocarcinoma specimens and normal duodenal tissues from the same patients were sampled from surgically resectedspecimens at The Johns Hopkins Hospital (5). At the time of surgery, normaland neoplastic tissues were fresh frozen or formalin fixed and paraffin embedded, and neoplastic tissues were also propagated as xenografts in nudemice, as has been described (6). Primary carcinomas were enriched for neo

plastic cell content by cryostat dissection, as described (5). The cell line PL45was obtained by in vitro culture of a pancreatic adenocarcinoma from thisseries. The pancreatic cancer cell lines AsPcl, BxPc3, Capanl, Capan2,CFPACI, Hs766T, MiaPaCa2, Pancl, and Su86.86 and the cervical cancer cell

line HeLa were obtained from American Type Culture Collection. TheColo357 cell line was obtained from European Collection of Animal CellCultures, Panc89 from T. Okabe (University of Tokyo, Tokyo, Japan),PancTul and PancTu2 from M. v. Billow (Johannes Gutenberg University,Mainz, Germany), and PaCa44 from M. Lohr (University ofRostock, Rostock,Germany). DNA was extracted using standard methods, as described (3). Allmutational analyses were performed without sample identifiers.

MSP. DNA methylationpatternsin the 5'-CpG island of pitS were determined by chemical modification of unmethylated, but not the methylated,cytosines to uracil and by subsequent PCR using primers specific for eithermethylateci or the modified unmethylated DNA (7). Briefly, DNA was dena

tured by NaOH and modified by sodium bisulfite. DNA samples were thenpurified using Wizard DNA purification resin (Promega), treated again withNaOH, precipitated with ethanol, and resuspended in water. PCR was per

3126

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

THE Rb/pl6 PAThWAY IN PANCREATIC CARCINOMA

CpG island, is an alternative mode for inactivating p16 in pancreaticcarcinoma. A subset of the xenografts and cell lines was thereforecoded and analyzed for pitS methylation patterns through MSP (7).Nine carcinomas were selected for analysis because they appeared toretain the wild-type pitS coding sequence, and nine samples wereselected because they were known to harbor an intragenic mutation inp16 (Table 1). Seven carcinomas (PXI7, PX29, PL45, PX93, PX1O4,CFPAC1, and Colo357) were found to harbor a methylated pitS, andupon breaking the code, all seven were found to have a wild-type pitScoding sequence (P = 0.001, Fisher's exact test; Fig. lA and Table I;data for CFPAC1 and Co1o357 not shown). None of the nine carcinomas in which p16 was inactivated by genetic changes were methylated. pI6 methylation patterns in 17 of the 18 xenografts and celllines were homogeneous; i.e. , the templates were either entirely methylated or entirely unmethylated. One xenograft (PX29) had strongMSP amplification of methylated templates with relatively weakamplification of unmethylated (or incompletely methylated) templates(Fig.1A).

To confirm these data, methylation patterns were determined in theprimary carcinomas and control normal tissues that were availablefrom four of the seven carcinomas having pitS methylation. MSPanalysis revealed strong amplification of methylated templates, withsimilarly strong amplification of unmethylated templates, in the primary tumors corresponding to cell line PL45 and xenograft PX1O4.The primary tumor for xenograft PX17 had a weak MSP amplificationof the methylated template, although it was nearly as strong as theamplification of the unmethylated template. The primary tumor for

PX93, however, had only a very weak MSP amplification of themethylated template, whereas the amplification of the unmethylatedtemplate was strong. A similarly weak amplification suggested somedegree of methylation in the normal tissue corresponding to xenograftPX93 but not in the other normal tissues (Fig. IB).

Expression of pitS was analyzed in detail in the pancreatic carcinoma cell line PL45. PL45 had been obtained from the same patientgroup as were the xenografts, and unlike the xenografts, RNA andprotein analyses of this cell line were not complicated by the presenceof nonneoplastic murine cells in the supportive stroma. Analysis ofpitS RNA transcripts by RT-PCR and of p1(1 protein expression by

Western blotting both revealed no pitS gene expression (Fig. 2). Incontrast, the neighboring and highly homologous INK4 family member p15 was found to be transcribed in this cell line (Fig. 2A).Repeated sequencing of the pitS gene of the PL45 culture used forthese assays again did not reveal any alterations, supporting thehypothesis that pitS expression in this tumor was silenced in association with methylation of its promoter region CpG island. Of note,retention of heterozygosity at the pitS locus in PL45 (Table 1) impliedthat both p16 alleles were methylated, a phenomenon that has alsobeen observed in other tumor types (13, 14).

Fourteen of45 informative pancreatic carcinomas exhibited LOH atchromosome 13q. Rb protein expression was analyzed in 46 availablecarcinomas by immunohistochemical analysis using the anti-Rb antibody 3C8. All carcinomas were found to express the Rb protein toimmunohistochemically detectable levels (Table I). Twenty-six carcinomas were also analyzed by Southern blotting for genomic amplification of the cyclin Dl and CDK4 genes. None of the tumors had agreater than 3-fold increase in the intensity of the hybridization signalfor either gene (Table 1). The nine pitS wild-type carcinomas wereanalyzed for genetic alterations in the p16-binding domain of CDK4,for which an arginine-to-cysteine missense mutation at codon 24 hasbeen reported to critically impair the interaction between the p16 andCdk4 proteins (8, 9, 15). No genetic alterations in CDK4 wereidentified (Table 1).

The two carcinomas (PX66 and PX67) from this series that had no

formed using primers designed near the transcriptional start site of p16 andwere specific for modified DNA containing either methylated or unmethylatedcytosines. Primer sequences were as described (7).

PCR and Sequencing. Genomic sequences of exon 3 of p16 and of thep16-binding domain of CDK4 (including codon 24; Refs. 8 and 9) wereanalyzed for alterations, as has been described (10). Briefly, tumor DNA wasamplified by PCR, treated with exonuclease I and shrimp alkaline phosphatase(United States Biochemical Corp.) and directly sequenced by cycle sequencing(Sequitherm). PCR and sequencing primer sequences were (5â€â€”@3')AGAAC1TrATCCATAAGTA'ITrC, GTGACTGATGATcTAAG1TFC,

and CGGTAGGGACGGCAAGAGAGG forpl6 and ATCTCGATATGAGCCAGTGG, CCTCCTCCATTGGGGACTC,and CTCCTCCATFGGGGACTCTCACAC for cDK4. Sequence data from exons 1 and 2 of pitS were

available (3, 4) and were confirmed by independent amplification and sequencing of a subset of samples.

RT-PCR. Oligodeoxythymidylic acid cDNA was generated from totalRNA using SuperScript II reverse transcriptase according the manufacturer'srecommendations (Life Technologies, Inc.). p16 or p15 transcripts were amplified by PCR using Pfu polymerase according the manufacturer's recommendations (Stratagene) and with the addition of dimethylsulfoxide (5%).Primer sequences were (5' —@3') AGCAGCATGGAGCCUCGGCTGACTand AGAGCCTCTCTGGUC'ITFCAATCG for p16 and CTGCGGAATGCGCGAGGAGAACAAG and TAAAGTCGTFGTGGGCGGCTGGGGAfor p15 and were designed to encompass the entire coding sequences of thegenes.

Southern Blotting. Southern blots were prepared as described (5). Briefly,10 @tgof genomic DNA were digested with BgIII restriction endonuclease andseparated by electrophoresis in a 1.2% agarose gel. After transfer to Zeta-ProbeGT membranes(Bio-Rad), the DNA was evaluatedsequentiallyfor hybridization to a cyclin DI probe (dbEST clone ID no. 110022; Genbank accessionno. T85217) and to a CDK4 probe (dbEST clone ID no. 345984; Genbankaccession no. W72136).

Western Blotting. Protein extracts were prepared as described (I I), exceptthat SDS was omitted. Total protein (1.5—2mg) was reacted with either theanti-pl6 antibody SC-468 according the manufacturer's recommendations

(Santa Cruz Biotechnology) or with normal rabbit serum as a control (LifeTechnologies, Inc.). Immune complexes were collected using protein A agarose beads (Life Technologies, Inc.), and proteins were released throughboiling in SDS-containing buffer. Following SDSIPAGE electrophoresis in a4—20%acrylamide gradient gel, Western blots were prepared and incubatedwith anti-p16 primary antibody and antirabbit IgG secondary antibody conjugated to horseradish peroxidase (Amersham Corp.). Reactions were visualizedthrough enhanced chemiluminescence (Amersham, ECI).

Immunohistochemistry. Rb protein expression was determined using immunohistochemical assays, as described (12). Briefly, 5-nm sections weremounted and baked for 20 mm at 60°Con Probe-on Plus slides (FisherScientific). Sections were reacted with the anti-Rb antibody 3C8 (QED, SanDiego, CA) at 2 pg/mi for 2 h following an antigen retrieval step in hot citratebuffer. As a control, adjacent sections were treated with isotype-matchedantibodies at equivalent conditions. Secondary reagents were obtained from theVector Elite avidin-biotin complex kit (Vector, Burlingham, CA), and diami

nobenzidine with hematoxylin counterstain was used for color development.Staining patterns were scored using previously published criteria (12).

Results

We previously reported the results of a genetic analysis of the@ tumor suppressor gene in a panel of 39 pancre

atic carcinoma xenografts and 11 pancreatic cancer cell lines (Table 1;Refs. 3 and 4). All tumors but one (PL45) exhibited LOH at the p16locus. Seventeen of the 50 pancreatic carcinomas (34%) had anintragenic mutation of the second allele of the pitS gene, and 24 of the50 (48%) had a homozygous deletion involving pitS. In the presentstudy, we analyzed the nine tumors that had retained a wild-typecoding sequence ofpl6 for sequence alterations in exon 3 ofpl6. No

mutations were identified in these nine cases (Table 1).Next, we examined whether transcriptional silencing of the pitS

gene, in association with hypermethylation of the promoter region3127

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table I The Rb/p16 tumor-suppressive pathway in pancreatic carcinoma

p16—exon 3sequencedp'6 methylatione

Cyclin DIamplification5CDK4 amplification5CDK4 sequence@l

THE Rb/p16 PATHWAY IN PANCREATIC CARCINOMA

p/6—exonPancreatic

carcinomas―9pallelicI + 2

sequencec13qallelic

@bRbprotein1PXI3LOHMutatedRetExpressedPx

I6LOHMutatedUnmethylatedRetExpressed(—)(—)PX2OLOHMutatedRetExpressedPX2

ILOHMutatedRetExpressedPX24LOHMutatedRetExpressed(—)(—)PX27LOHMutatedLOHExpressed(—)(—)PX3OLOHMutatedUnmethylatedRetExpressed(—)(—)PX6ILOHMutatedRetExpressed(—)(—)PX6SLOHMutatedUnmethylatedRetExpressed(—)(—)PX67LOHMutatedWild-typeUnmethylatedRetExpressed(—)(—)Wild-typePX68LOHMutatedUnmethylatedRetExpressedPX92LOHMutatedUnmethylatedRetExpressedPXlOlLOHMutatedLOHExpressedPx

105LOHMutatedUnmethylatedNIExpressedPXI22LOHMutatedNIExpressedAsPc

ILOHMutatedUnmethylatedLOHExpressed(—)(—)Capan2LOHMutatedUnmethylatedRetExpressed(—)(—)Hs766TLOHMutatedUnmethylatedRetExpressed(—)(—)Px

I9LOHHDRetExpressedPX23LOHHDRetExpressedPX28LOHHDLOHExpressed(—)(—)Px55LOHHDLOHExpressedPX56LOHHDRetExpressed(—)(—)PX64LOHHDLOHExpressedPX72LOHHDNIExpressed(—)(—)PX74LOHHDLOHExpressed(—)(—)PX75LOHHDRetNDPX76LOHHDRetExpressedPX86LOHHDRetExpressed(—)(—)PX88LOHHDRetExpressedPX9OLOHHDRetND(-)(-)PX9ILOHHDNINDPX94LOHHORetExpressedPXIO2LOHHDRetExpressed(—)(—)PXIO7LOHHDRetExpressedPx

117LOHHDNIExpressedPxI20LOHHORetExpressedBxPc3LOHHDLOHExpressedCapanlLOHHDLOHND(-)(-)Miapaca2LOHHDRetExpressed(—)(—)Panc

ILOHHDLOHExpressed(-)(-)Su86.86LOHHDLOHExpressed(—)(—)Px

I7LOHWild-typeWild-typeMethylatedRetExpressedWild-typePX29LOHWild-typeWild-typeMethylatedRetExpressedWild-typePL45RetWild-typeWild-typeMethylatedLOHExpressed(—)(—)Wild-typePX93LOHWild-typeWild-typeMethylatedRetExpressed(—)(—)Wild-typePXIO4LOHWild-typeWild-typeMethylatedRetExpressedWild-typeCFPAC

ILOHWild-typeWild-typeMethylatedLOHExpressed(—)(—)Wild-typeColo3SlLOHWild-typeWild-typeMethylatedRetExpressed(—)(—)Wild-typePX66LOHWild-typeWild-typeUnmethylatedLOHExpressed(—)(—)Wild-type

a Pancreaticcarcinomaspecimenswerepropagatedasxenografts(PX series)orasa cellline(PL45).Othercelllineswerefromcommercialsources(see“MaterialsandMethods―).b Allelic loss was determined using microsatellite markers IFNA and D9S171 for 9p (near the pitS locus) and markers D13S260 and D13S267 for I3q (centromeric to RB)). Ret,

retention of heterozygosity; NI, noninformative for both markers.C Mutation data previously published (3, 4), obtained by sequencing analysis of the genomic sequences of exons 1 and 2 ofp)6. Upon reanalysis, xenograft PX55 from this series

was found to harbor a homozygous deletion of p16. PX67 had a nonsense mutation at codon 50 of p)6, and cell line Colo357 had a wild-type p16 gene sequence. HD, homozygousdeletion involving p16.

d Mutation analysis of the genomic sequence of exon 3 of pitS, detennined in tumors having a wild-type sequence at exons I and 2.e MSP analysis for methylation of the 5'-CpG island of p16, determined in all nine p16 wild-type tumors and nine selected p16-mutated tumors.

@Immunohistochemical analysis for nuclear Rb protein expression of cancer cells in tissue or in cell pellets. ND, not determined due to inadequate sample.SSouthernblot analysisfor cyclinDI andCDK4amplification,determinedin a subsetof thetumors.(—),copynumberchangeof lessthanlimit of assaysensitivity(about3-fold).h Sequence analysis of the p16-binding domain of CDK4 (including codon 24; see “Materialsand Methods―), determined in the nine p16 wild-type tumors.

evidence of abrogation of the Rb/pl6 pathway were reanalyzed forgenetic alterations in the pitS coding sequence. Xenograft PX66again had a wild-type pitS gene sequence, but PX67 had a nonsensemutation at codon 50 of pitS (CGA —*TGA). Review of previoussequence analysis data of PX67 revealed that the mutation wasevident but had been overlooked. Re-evaluation of the histology ofthe tumors confirmed the diagnosis of pancreatic adenocarcinomafor both cases. It remains possible that our analyses failed to detectan existing alteration within one of the four Rb/pl6 pathwaymembers in case PX66.

In addition to results of this tumor series, p16 methylation

patterns were determined in selected samples from another seriesof 19 pancreatic carcinoma xenografts and cell lines (16). Thesecarcinomas had been analyzed independently for genetic alterations of the pitS gene. Five of the 19 carcinomas (26%) had anintragenic mutation ofpi6, and 10 of the 19 (53%) had a homozygous deletion involving p16 ( I 6). Four cell lines in this series hada wild-type pitS coding sequence (Panc89, PancTul, PancTu2, andPaCa44) and lacked pitS transcripts and p16 protein expression(Ref. 16 and reanalysis of Panc89). MSP analysis of these four celllines revealed that all four tumors had pitS methylation (data notshown).

3128

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

I@@@ U.@ H

THE Rb/pl6 PAThWAY IN PANCREATIC CARCINOMA

Fig. I. MSP analysis of the p16 gene in pancreaticcarcinoma. A, pancreatic carcinoma xenografts and celllines. B, cryostat-dissected primary pancreatic carcinomasand constitutional normal duodenal tissues. Identifiersabove each set of lanes correspond to the similarly numbered xenografts and cell lines (A; Table 1). N, normaltissue; 1', primary tumor; U and M, MSP analysis specificfor unmethylated and methylated 5'-CpG islands of p16,respectively.

B

Discussion

In the present study, an extensive analysis of all four members ofthe Rb/p16 tumor-suppressive pathway revealed the abrogation of thispathway in virtually all pancreatic carcinomas (98%). Although thispathway can be abrogated in other tumor types by inactivation of anyof several members of the pathway (15, 17—23),in pancreatic carcinoma, the Rb/p16 pathway appears to be almost exclusively mactivated by alteration of the p16 protein and, only rarely, the Rb protein(5, 24, 25).

The pitS gene was inactivated through intragenic mutation, homozygous deletion, and methylation-associated transcriptional silencing of pitS. The association of silencing of gene expression withmethylation of the 5'-CpG island of the gene has been established forgenes of the X chromosome in females and for nonexpressed allelesof selected imprinted genes. This methylation change has also beendemonstrated for tumor suppressor genes, and it appears to constitutea nonmutational alternative to gene inactivation (26). Importantly, ourresults clearly indicate that hypermethylation occurred exclusively in

the pancreatic carcinomas in which p16 was not genetically altered,confirming the specificity of this mechanism of gene inactivation.

Analysis of the primary carcinomas from which the xenografts andcell lines were obtained confirmed these results. Interestingly, theprimary tumor for xenograft PX93 showed an unusually weak MSPamplification of the methylated template but a relatively strong amplification of the unmethylated template (even when taking the proportion of neoplastic cells into consideration). This weak amplification could represent a low, and somewhat variable, background levelof methylation, as suggested by a similarly weak amplification of thenormal duodenal tissue from this same patient. Alternatively, only aportion of the primary tumor for PX93 might have been methylatedand selectively propagated in the xenograft. Another alternative is thatthe weak MSP amplification in the primary carcinoma might representsubpopulations of tumor cells with variable degrees of p16 methylation. Although we do not understand this apparent heterogeneity inpitS methylation, methylation-associated silencing of p16 may be agradual process, during which a subset of cytosine residues within

Ap15

c@ to_l ‘@@-

G) .J .@.:i: o@

PL45

CD CO‘- -@‘ ‘-

0.@ 0.

Fig. 2. Analysis of transcriptional silencing of p16 in the p16-methylated pancreatic carcinoma cell line PL45. The HeLa cell lineserved as a positive control. Ticksat the right of each image indicatemarker positions. A. RT-PCR of pitS and p15 transcripts; (—),atemplate-negative RT-PCR control. B, Westem blot for p16 protein,performed on immunoprecipitated proteins. Lanes marked p16 and(—), immunoprecipitation using anti-pl6 antibody and normal rabbitserum, respectively. The lower band represents the p16 protein. Thehigher molecular weight band, present in all lanes, represents immunoglobulin heavy chain from the immunoprecipitation.

3129

A I— t'J,-‘ Lf) @O ,— ,@.@

11)0@ u 0') @D@ t@ 0 m@ 1%. LI)w @- N@W @. C@@J1@O- @.m a @D@-

UN UN UN UN UN UN UN UNUN UMUM UN UN UN

17 45 93 104

N 1° N 1° N 1° N 1°

UMUMUMUMUMUMUMUM

BHeLap16

(@ LI)_j ‘@J.a) _JI O@@

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

THE Rb/p16 PAThWAY IN PANCREATIC CARCINOMA

11. Rabindran, S. K., Wisniewski, J., Li, L., Li, G. C., and Wu, C. Interaction betweenheat shock factor and hsp7O is insufficient to suppress induction of DNA bindingactivity in vivo. Mol. Cell Biol., 14: 6552—6560,1994.

12. Geradts, J., Hu, S-X., Lincoln, C. E., Benedict, W. F., and Xu, H-J. Aberrant RB geneexpression in routinely processed archival tumor tissues determined by three differentanti-RB antibodies. lot. J. Cancer, 58: 161—167,1994.

13. Herman, J. G., Merlo, A., Mao, L., Lapidus, R. G., Issa, J-P. J., Davidson, N. E.,Sidransky, D., and Baylin, S. B. Inactivation of the CDKN2JpI6.IMTSIgene isfrequently associated with aberrant DNA methylation in all common human cancers.Cancer Res., 55: 4525—4530,1995.

14. Merlo, A., Herman, J. G., Mao, L., Lee, D. J., Gabrielson, E., Burger, P. C., Baylin,S. B., Sidransky, D. 5' CpG island methylation is associated with transcriptionalsilencing of the tumour suppressor p16/CDKN2/MTSI in human cancers. Nat. Med.,1: 686—692, 1995.

15. Bartkova, J., Lukas, J., Guldberg, P., Alsner, J., Kirkin, A. F., Zeuthen, J., and Bartek,J. The p16-cydin D/Cdk4-pRb pathway as a functional unit frequently altered inmelanoma pathogenesis. Cancer Res., 56: 5475—5483,1996.

16. Naumann, M., Savitskaia, N., Eilert, C., Schramm, A., Kalthoff, H., and Schmiegel,W. Frequent codeletion of p16/MTSI and p15/MTS2 and genetic alterations inp16(MTSJ in pancreatic tumors. Gastroenterology, 110: 1215—1224,1996.

17. Jiang,W., Zhang,Y-J.,Kahn,S. M., Hollstein,M. C., Santella,R. M., Lu, S-H.,Harris, C. C., Montesano, R., and Weinstein, I. B. Altered expression ofthe cyclin Dland retinoblastoma genes in human esophageal cancer. Proc. Nail. Acad. Sci. USA,90: 9026—9030, 1993.

18. Ouerson, G. A., Kratzke, R. A., Coxon, A., Whan Kim, Y., and Kaye, F. J. Absenceof pl6INK4protein is restricted to the subset of lung cancer lines that retains wildtypeRB.Oncogene,9:3375—3378,1994.

19. He, J., Allen, J. R., Collins, V. P., Allalunis-Turner, M. J., Godbout, R., Day, R. S.,m, andJames,C. D. CDK4amplificationis an alternativemechanismof pitSgenehomozygous deletion in glioma cell lines. Cancer Res., 54: 5804—5807,1994.

20. Schmidt,E. E., Ichimura,K., Reifenberger,G., and Collins,V. P. CDKN2(p16/MTSJ) gene deletion or CDK4 amplification occurs in the majority of glioblastomas.Cancer Res., 54: 6321—6324,1994.

21. Kratzke, R. A., Greatens, T. M., Rubins, J. B., Maddaus, M. A., Niewoehner, D. E.,Niehaus, G. A., and Geradts, J. Rb and @16tr'@@.expression in resected non-smalllung tumors. Cancer Res., 56: 3415-3420, 1996.

22. Kinoshita, I., Dosaka-Akita, H., Mishina, T., Akie, K., Nishi, M., Hiroumi, H.,Hommura, F., and Kawakami, Y. Altered pl6@―and retinoblastoma protein statusin non-small cell lung cancer: potential synergistic effect with altered p53 protein onproliferative activity. Cancer Res., 56: 5557-5562, 1996.

23. Ichimura, K., Schmidt, E. E., Goike, H. M., and Collins, V. P. Human glioblastomaswith no alterations of the CDKN2A(p16ht@@*@4A,/pff5l) and CDK4 genes have frequentmutations of the retinoblastoma gene. Oncogene, 13: 1065—1072,1996.

24. Barton,C. M.,McKie,A. B.,Hogg,A., Bia,B.,Elia,G., Phillips,S. M.A., Ding,S-F., and Lemoine, N. R. Abnormalities of the RB) and DCC tumor suppressorgenes: uncommon in human pancreatic adenocarcinoma. Mol. Carcinog., 13:61—69,1995.

25. Huang, L., Lang, D., Geradts, J., Obara, T., Klein-Szanto, A. J. P., Lynch, H. T., andRugged, B. A. Molecular and immunochemical analyses of RB1 and cyclin Dl inhuman ductal pancreatic carcinomas and cell lines. Mol. Carcinog., 15: 85—95,1996.

26. Baylin, S. B., Herman, J. G., Graff, J. R., Vertino, P. M., and Issa, J-P. Alterations inDNA methylation: a fundamental aspect of neoplasia. Adv. Cancer Res., in press,1997.

3130

each CpG island become progressively methylated. Such gradual methylation of CpG islands perhaps results in gradually decreasing levels ofp16 transcription. If so, the concentration-dependent competitive bindingof p16 to Cdk4, and thus its growth-inhibitory function, would then alsodiminish gradually. With continued growth, this heterogeneous population of tumorigenic clones would thus evolve toward a more uniformpopulation of tumor cells having more extensive methylation of the 5'CpG island of p16. Indeed, the finding of pitS methylation exclusively

within the group of p16 wild-type carcinomas strongly suggests thatmethylation silencing of p16 imposed a selective growth advantage forthe tumor cells. The comparable frequencies of p16 methylation foundhere in two series of pancreatic carcinomas (14% and 21%, respectively)underscores the widespread importance of this mechanism of pitS mactivation in the development of carcinomas of the pancreas.

References1. Sherr, C. J. Cancer cell cycles. Science (Washington DC), 274: 1672—1677,1996.2. Hall, M., and Peters, 0. Genetic alterations of cycins, cyclin-dependent kinases, and

cdk inhibitors in human cancer. Adv. Cancer Res., 68: 67—108,1996.3. Caldas, C., Hahn, S. A., da Costa, L. T., Redston, M. S., Schuue, M., Seymour, A. B.,

Weinstein, C. L., Hruban, R. H., Yeo, C. J., and Kern, S. E. Frequent somaticmutations and homozygous deletions of the p16 (MTSI) gene in pancreatic adenocarcinoma. Nat. Genet., 8: 27—31,1994.

4. Rozenblum, E., Schutte, M., Goggins, M., Hahn, S. A., Panzer, S., Zuhurak, M.,Goodman, S. N., Sohn, T. A., Hruban, R. H., Yeo, C. J., and Kem, S. E. Tumorsuppressive pathways in pancreatic carcinoma. Cancer Res., 57: 1731—1734,1997.

5. Seymour, A. B., Hruban, R. H., Redston, M. S., Caldas, C., Powell, S. M., Kinzler,K. W., Yeo, C. H., and Kern, S. E. Allelotype of pancreatic adenocarcinoma. CancerRes., 54: 2761—2764,1994.

6. Hahn, S. A., Seymour, A. B., Hoque, A. T. M. S., Schutte, M., tin Costa, L. T.,Redston, M. S., Caldas, C., Weinstein, C. L., Fischer, A., Yea, C. J., Hniban, R. H.,and Kern, S. E. Allelotype of pancreatic adenocarcinoma using xenograft enrichment.Cancer Res. 55: 4670—4675, 1995.

7. Herman, J. G., Graff, J. R., Myohanen, S., Nelkin, B. D., and Baylin, S. B. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc.NatI. Aced. Sci. USA, 93: 9821—9826,1996.

8. WOlfel, 1., Hauer, M., Schneider, J., Serrano, M., Wölfel,C., Klehmann-Hieb, E., DcPlaen, E., Hankeln, T., Meyer zum Bllschenfelde, K-H., and Beach, D. A pl6@K4a@insensitive CDK4 mutant targeted by cytolytic T lymphocytes in a human melanoma.Science (Washington DC), 269: 1281—1284,1995.

9. Zuo, L., Weger, J., Yang, Q., Goldstein, A. M., Tucker, M. A., Walker, G. J.,Hayward, N., and Dracopoli, N. C. Germline mutations in the pl6tr@@4@@bindingdomain of CDK4 in familial melanoma. Nat. Genet., 12: 97—99,1996.

10. Schutte, M.. Hruban, R. H., Hedrick, L., Cho, K. R., Molnar Nadasky, G., Weinstein,C. L., Bova, G. S., Isaacs, W. B., Cairns, P., Nawroz, H., Sidransky, D., Casero, R. A.,Jr., Meltzcr, P. S., Hahn, S. A., and Kem, S. E. DPC4 gene in various tumor types.Cancer Res. 56: 2527—2530,1996.

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from

1997;57:3126-3130. Cancer Res   Mieke Schutte, Ralph H. Hruban, Joseph Geradts, et al.   Virtually All Pancreatic CarcinomasAbrogation of the Rb/p16 Tumor-suppressive Pathway in

  Updated version

  http://cancerres.aacrjournals.org/content/57/15/3126

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://cancerres.aacrjournals.org/content/57/15/3126To request permission to re-use all or part of this article, use this link

Research. on October 26, 2020. © 1997 American Association for Cancercancerres.aacrjournals.org Downloaded from