Allelotype of pediatric rhabdomyosarcoma - Nature

21 downloads 85 Views 312KB Size Report
An allelotype covering all autosomes was constructed for the embryonal form of childhood rhabdomyosarcoma. (ERMS) in order to identify regions ...
Oncogene (1997) 15, 1309 ± 1314  1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00

Allelotype of pediatric rhabdomyosarcoma Mike Visser1,2, Carin Sijmons2, Johannes Bras3, Robert J Arceci4, Mark Godfried2, Linda J Valentijn2, PA VouÃte1 and Frank Baas1,2 1

Department of Pediatric Oncology, 2Neurozintuigen Laboratory and 3Department of Pathology, Academic Medical Center, PO Box 22700, 1100 DE Amsterdam, The Netherlands; 4Department of Pediatrics, Division of Hematology and Oncology, Children's Hospital Medical Center and the Children's Hospital Research Foundation, 3333 Burnet Avenue, Cincinnati, Ohio 45229 ± 3039, USA

An allelotype covering all autosomes was constructed for the embryonal form of childhood rhabdomyosarcoma (ERMS) in order to identify regions encompassing tumorsuppressor genes (TSG) involved in ERMS. Thusfar most studies were focussed on chromosome 11p15.5, which frequently shows loss of heterozygozity (LOH) in embryonal tumors like RMS and Wilms' tumor (WT). In this study we show that, besides LOH of chromosome 11p15.5 (72%), LOH of chromosome 16q was present in 54% of the tumors analysed. Delineation of these two regions shows that the smallest region of overlap (SRO) for chromosome 11 was between D11S988 and D11S922. This region, estimated to be 7 cM and 3-5 Mb, is also the location of the putative Wilms' tumor WT2 TSG. It contains several genes including IGF2 and potential tumorsuppressor genes like H19 and p57kip2, which might contribute to the carcinogenesis of RMS. Analysis of chromosome 16q LOH de®ned the SRO between D16S752 and D16S413. LOH of chromosome 16 is also found in other tumors, including WT. Our data suggest that genes involved in the development of RMS and WT may not only be similar for chromosome 11 but also for chromosome 16. Keywords: RMS; Rhabdomyosarcoma; Alleotype; TSG, Tumor Suppressor Gene; LOH, loss of heterozygosity; SRD

Introduction Allelic loss at speci®c loci occurs in many types of tumors. These events are thought to represent the second hit in the Knudson hypothesis (Knudson, 1971) resulting in functional inactivation of the tumorsuppressor genes (TSGs). Since the allelic loss is not only restricted to the TSG, but usually encompasses large genomic regions, ¯anking polymorphic loci are frequently used to identify the genomic segment containing TSGs. A catalogue of the regions showing LOH in a tumor is called an allelotype. Rhabdomyosarcoma (RMS) accounts for 5 ± 10% of all solid tumors in children (Pappo et al., 1995; Gaiger et al., 1981). These tumors originate from the undi€erentiated mesenchymal cells resembling developing striated muscle. Three morphologic types are described: embryonal RMS (ERMS), alveolar RMS (ARMS) and pleomorphic RMS (Agamanolis et al., Correspondence: F Baas Received 3 February 1997; revised 27 May 1997; accepted 27 May 1997

1986). Approximately 80% of all new diagnosed cases of RMS are ERMS (60%) or ARMS (20%). More than 90% of the ARMS are characterized by a translocation t(2;13)(q35;q14) or t(1;13)(q36;q14), which results in a chimaeric transcript encoding a fusion protein consisting of the intact PAX3 or PAX7 DNA binding domain and the distal half of the fork head domain of the FKHR gene (Barr et al., 1993, 1995; Galili et al., 1993; Davis et al., 1994). ERMS is frequently characterized by LOH of chromosome 11p15.5 and these LOH studies de®ne the putative TSG distal to D11S988 locus (Besnard-GueÂrin et al., 1996). Several genes on 11p15.5 are subjected to imprinting which complicates the LOH studies. Early studies by Scrable et al. (1989a) show a preferential retention of the paternal chromosome 11 in ERMS suggesting a role of epigenetic modi®cations in the process of tumor development. H19, p57kip2 and IGF2 are located distal to D11S988 and might contribute to the malignant transformation. H19 and p57kip2 are potential tumor suppressor genes and IGF2 stimulates cell growth and lowers the incidence of apoptosis (Hao et al., 1993; Matsouka et al., 1995; Christofori et al., 1994). All three genes are imprinted (Zemel et al., 1992; Hatada et al., 1996a; Matsuoka et al., 1996) and paternal isodisomy with maintenance of imprinting status would theoretically result in loss of expression of H19 and p57kip2 and expression of both IGF2 alleles. Somatic cell hybridization studies were used to determine the locus for the suppressor of proliferation on 11p. Koi et al. (1993) showed that the suppressor of proliferation was located centromeric to D11S724 in the RMS derived RD cell line. Both H19 and IGF2 are located outside this region, however they may still contribute to tumor development because only the proliferative capacity of the tumor derived cell lines were tested. In case of a contribution of multiple genes at the same loci to oncogenic development, this analysis might result in the underestimation of the role of genes surrounding a suppressor of growth. The paternal imprinted gene p57kip2 is currently the only candidate TSG located in this region (Matsuoka et al., 1996; Reid et al., 1996a; Hatada et al., 1996a) and in one study reduced p57kip2 expression was found in three out of seven WT tested (Hatada et al., 1996a). However, no mutations have been identi®ed in the p57kip2 gene in a wide variety of tumors tested such as soft tissue sarcomas, WT and carcinomas of breast, bladder and liver (Orlow et al., 1996; Tokina et al., 1996). Moreover, in somatic cell hybrid analysis of the WT G401 cell line no correlation of p57kip2 mRNA

Allelotype in RMS M Visser et al

1310

with suppression was found (Reid et al., 1996a). Thus far, all studies on ERMS were focused on 11p15. Here we present a study in which we constructed a genome wide allelotype for ERMS in order to identify other chromosome regions harboring putative TSGs. In this study we identify a new common region of LOH on chromosome 16q in RMS and in combination with the data from Koi et al. (1993) narrow the region for the tumor suppressor on chromosome 11 between D11S988 and D11S724. Results Thirty-three primary RMS were analysed. Of the 33 primary tumors, 27 tumors were histologically typed as ERMS and six as ARMS. Eight relapses and three metastases of 10 primary ERMS and one relapse and six metastases of three primary ARMS were included in this study. In addition 10 human RMS xenografts were analysed. Six were histologically typed as ERMS and four as ARMS. Table 1 shows the 74 DNA markers tested, the majority being microsatellites. For each autosomal chromosome arm one or more markers were used for the identi®cation of LOH. The frequency of LOH per marker is shown. A compilation of the LOH pattern detected with all markers used for the primary ERMS, is shown in Figure 1. The most obvious alterations are the frequent LOH on chromosome 11p, 11q and 16q with a total LOH of 72%, 55% and 54% respectively. The ERMS demonstrating LOH of chromosome 11q also demonstrated LOH of chromosome 11p suggesting whole chromosome loss or isodisomic state. Twenty percent of the RMS, demonstrated LOH on chromosome 11p without LOH of chromosome 11q suggesting 11p (interstitial) deletion or mitotic recombination event. There was only one ERMS sample demonstrating LOH of chromosome 11q without LOH of chromosome 11p tested with four informative markers. Four out of six ARMS also demonstrated LOH of chromosome 11p. The parane embedded ARMS samples could not be tested for the translocations t(2;13)(q35;q14) or t(1;13)(q36;q14) because no RNA or karyotype was available. The two informative ARMS xenografts with a translocation t(2;13)(q35:q14) demonstrated no LOH of chromosome 11p15.5 (Table 2).

To delineate the LOH region on chromosome 11p we tested 13 additional markers on chromosome 11p15.5. The smallest region of overlap (SRO) that was lost in this study is between the markers D11S988 and D11S922 (Figure 2: tumor 17). The candidate region as de®ned in the study of Koi et al. (1993) is also depicted. Table 1 Chromosome

Locus

Location

LOH/ informative patients

1p

FGR D1S322 D1S175 D1S158 D1S102 MYCN D2S123 HGBC D3S11 GLUT2 D4S174 GABRB1 D4S171 D5S208 D5S107 IL-9 F13A1 D6S89 D6S251 D7S472 D7S466 D8S201 D8S166 D9S104 D9S51 D9S66 D10S89 D10S109 HRAS1 TH D11S875 D11S554 D11S922 D11S873 TYRSIN D11S35 D11S874 D12S62 D12S60 D13S71 D14S43 D14S51 CYP19 HBAP1 D16S749 D16S540 D16S541 D16S752 D16S539 D16S413 D16S305 D17S513 HGF D18S59 MFD80 MBP D19S177 APOC2 D20S59 D20S120 D21S13E D21S156 CYP2D

1p36.2-p36.1 1p31 1p21-q12 1q32-q41 1q42-q43 2p24 2p15-p16 2q34-q35 3p21-p14 3q26-q26.3 4p14 4p13-p12 4q33-q35 5p15-3-15.1 5q11.2-q13.3 5q22.3-q31.3 6p25-p24 6p24-p23 6q13-21.1 7p21-22 7q32 8p23 8q11-q12 9p21 9q3 9q34.1-q34.3 10pter-p11.2 10q11.2-qter 11p15.5 11p15.5 11p15.5-15.4 11p12-11.2 11p15.5 11q13-q23 11q14-q21 11q22 11q23-qter 12pter-p12 12qter 13q31-q33 14q24.3 14q31.1-qter 15q21.1 16p13.3 16p 16q12.1 16q12.1 16q23 16q23.1-qter 16q24.3 16q24.3 17p13 17q22-q24 18pter 18p11.32-11.31 18q22-qter 19p13.3 19q12-q13.2 20p12 20q12-q13 21q11.2 21q22.3 22q11

0/11 1/17 0/12 0/14 1/8 0/12 2/14 3/16 2/21 0/9 1/8 4/24 2/16 1/24 1/6 3/22 5/22 4/13 3/27 0/18 0/20 2/16 0/21 2/14 0/12 4/26 0/13 1/21 6/10 20/24 12/22 13/24 4/13 1/3 1/2 3/7 13/21 1/20 6/28 0/15 0/8 5/20 3/19 1/17 4/15 2/12 3/13 4/17 6/13 1/17 5/17 2/14 1/17 6/25 5/17 0/12 1/20 1/21 0/15 0/20 2/12 0/29 0/12

1q 2p 2q 3p 3q 4p 4q 5p 5q 6p 6q 7p 7q 8p 8q 9p 9q 10p 10q 11p

11q

12p 12q 13q 14q 15q 16p 16q

Delineation of the TSG locus on 11p and 16q Chromosome 11 Ten human RMS xenografts were analysed in more detail for LOH of chromosome 11p (six ERMS, four ARMS). Four out of ®ve informative histologically typed ERMS xenograft showed LOH of 11p, while the two informative ARMS xenografts demonstrated no LOH of chromosome 11p. The karyotype analysis of the xenografts showed that LOH was either due to chromosome isodisomy or accompanied by the presence of multiple copies of chromosome 11 (the ERMS showing no LOH has two translocations involving chromosome 11), Table 2.

Markers

17p 17q 18p 18q 19p 19q 20p 20q 21q 22q

Markers tested on the tumors of the patients and the xenografts. All markers are derived from published sources (GDB)

Allelotype in RMS M Visser et al

Chromosome 16 Two markers on chromosome 16p and six markers on chromosome 16q were also tested on the parane embedded tumor samples and the xenografts. Of the ®ve ARMS, two demonstrated LOH and of the 22 ERMS, 11 demonstrated LOH for chromosome 16q markers. Marker D16S539 shows LOH in six out of 13 informative samples. The overall LOH of chromosome 16q is 54%. In the xenografts only one of seven informative tumors showed LOH for D16S539. The overall LOH of chromosome 16q in the xenografts is 1/ 10 (Table 2). Although this di€erence in the frequency of chromosome 16q LOH between the tumors and the xenografts seems large, it is not statistically signi®cant (Fisher exact test, two sided, P40.05). Figure 3 shows all the tumors demonstrating LOH of chromosome 16q. Assuming a single locus on 16q the smallest region of overlap is between the markers D16S752 and

D16S539, 16q23-16qter. This region is estimated to be 25 ± 30 cM. Chromosome stability and fractional allelic loss (FAL) From the LOH study it is clear that some RMS frequently demonstrated LOH whereas others did not. Analysis of metastases of 10 ERMS shows that there

Allelotype of RMS

A

Figure 1 Chromosomes are depicted and divided into p arms (white bars) and q arms (black bars). For each chromosome arm the percentage of observed LOH in all tumors is shown. For chromosome 3q, 7p, 8q, 10p, 13q, 18q, 20p, 20q, 21q and 22q no LOH was observed in the tumors tested

E

E

A

A

E

Xenograft

Karotype

Chromosome 16q

Chromosome 11p15.5

1 ± 2 ERMS

46xy

LOH

LOH

1 ± 9 ERMS

55xxq+, +2, +2, +11, +13, +1 Marker 52xx, +2, der (2), del (2), t (x; 11) (q28; q13), 711, t (8; 11) (q22; q13), +13, +5 Markers 55xyy, +2, +11, +11 53xy, +2, +11 59xy, rcp (1; 11) (q21; q12), 2q+, +2q7, +13, +16p, +4 Markers t (2; 13) 54xy, t (2; 13) (q35; q14), +2q+, +5 Markers 46xx, t (2; 13) (q35; q14), 713, +1 Marker 88xxx, t (2; 13) (q35; q14)

nLOH

ni

ni

nLOH

nLOH nLOH nLOH

LOH LOH LOH

nLOH nLOH

nLOH nLOH

nLOH

ni

nLOH

ni

2 ± 3 ERMS 3 ± 1 ERMS 3 ± 5 ERMS 2 ± 8 ARMS 1 ± 10 ARMS 2 ± 7 ARMS 3 ± 6 ARMS

E

E

E

Figure 2 LOH analysis on chromosome 11p: Schematic genetic presentation of chromosome 11p15.5 with the markers tested on seven tumors from patients and three xenografts. The candidate region de®ned by Koi et al. (1993) is also depicted between the markers D11S724 and D11S719. Only informative markers are shown. Black circles: LOH; white circles: no LOH. A: alveolar RMS, E: embryonal RMS. The SRO is de®ned by tumor nr. 17

Table 2 Partial karytypes of ten xenografts, six ERMS and four ARMS, to show involvement of chromosome 2, 11, 13, 16

1 ± 12 ERMS

A

All ARMS tumors contain the translocation t (2;13), Xenograft 1-2 was the only tumor with LOH of chromosome 16q.

1311

Allelotype in RMS M Visser et al

1312

Figure 3 LOH analysis on chromosome 16: Schematic representation of chromosome 16 with the markers used in this study. The scale in centyRays (cR) is from Stanford Human Genome Center (SHGC) radiation hybrid map and represents the distance from the p arm telomere. The markers D16S540 and D16S541 are mapped by the SHGC. The other markers are mapped by the Whitehead Institute for Genome Research and Comparative Human Linkage Center (CHLC), and their locations on the SHGC map deduced on the basis of ¯anking markers present on both maps. Thirteen tumors are tested. Only the informative markers are shown: black circles: LOH; white circles: no LOH. The SRO is between the markers D16S539 and D16S752. Tumors 34 and 35 do not contribute to the SRO, but do not exclude the SRO

were only minor di€erences in LOH pattern between the primary tumor and its subsequent relapse or metastases (data not shown). This suggests that allelic loss in RMS is rather stable. For each tumor the FAL was calculated, which is the fraction of the total number of informative chromosome arms that shows LOH. The six ARMS demonstrated a mean FAL of 0.06, the ERMS had a mean FAL of 0.142. The di€erence in the FAL between the ARMS and the ERMS was signi®cant (P50.05 by Student t-test). On the basis of FAL, the ERMS group was divided in two subgroups one with a FAL higher than the mean and one with FAL lower than the mean. There was no correlation of LOH of chromosome 11p or 16q with any of the two FAL groups. Discussion The allelotype of ERMS demonstrated a high frequency of LOH on chromosome 11p, 11q and 16q. Chromosome 11p, which is already being investigated in RMS for a long time demonstrated LOH in 72%. Scrable et al. (1989b) demonstrated LOH of chromosome 11p15.5 in ERMS whereas this was not the case in ARMS (Scrable et al., 1989a). Four out of six informative histologically classi®ed ARMS demonstrated LOH of chromosome 11p15.5 while the two informative ARMS xenografts demonstrated no LOH of chromosome 11p15.5. Thus, LOH of chromosome 11p15.5 is not speci®c for ERMS.

Other tumors such as breast cancer, Wilms' tumor, hepatocellular carcinoom and bladder cancer also frequently show LOH of chromosome 11p15 (Winqvist et al., 1993; Reeve et al., 1989; Fujimori et al., 1991; Knowles et al., 1994). This suggests either the presence of multiple TSGs or a single pleomorphic TSG on 11p15.5. Chromosome 11p15.5 might also harbor growth stimulating genes which might be a prerequisite for tumorigenesis. A gene located in this region is the maternally imprinted gene IGF2. Paternal isodisomy or maternal loss of imprint theoretically results in an increase of IGF2 expression and subsequently in a stronger signal for proliferation and a lower incidence of apoptosis (Christofori et al., 1994). The SRO in this study was between the markers D11S988 and D11S922, which encompasses about 7 cM or 3 ± 5 Mb and includes IGF2 and H19. Combining our data with the transfection studies in the RMS RD cell line by Koi et al. (1993) the region where the TSG for RMS should be located is between the markers D11S988 and D11S724, excluding H19 and IGF2 as suppressor of proliferation. Fusion of the G401 Wilms' tumor cell line with radiation reduced t(X;11) chromosomes suggested that a suppressor lies telomeric to D11S601 (Reid et al., 1996b). However, growth suppression was only obtained by transfer of a Dt(X;11) chromosome from one of three subclones that contain the same breakpoint, suggesting that the suppressor is inactivated by either mutations or epigenetic events like methylation. If one assumes that WT2 is the same TSG as in RMS, the region of interest is even smaller between the marker D11S601 and D11S724. This region is estimated to be 500 kb (Reid et al., 1996b). However, the fact that at least two copies of chromosome 11 are still present in the ERMS xenografts, suggests a selection for multiple copies of this chromosome which would result in a higher expression of IGF2. The putative TSG p57kip2 is located in the region between the markers D11S601 and D11S724 (Reid et al., 1996b). p57kip2 is a cyclin dependent kinase-cyclin complex inhibitor that causes G1 arrest and is subjected to paternal imprinting. Paternal isodisomy of chromosome 11 is a phenomenon frequently found in studies of both RMS (Scrable et al., 1989a) and Wilms' tumor (Mannens et al., 1988) and should result in a low expression of p57kip2 and subsequent a decrease of inhibition of cdkcyclin complexes and progression through the cell cycle. In di€erent tumors including RMS no gross rearrangements of p57kip2 or mutations in the coding region have been found (Orlow et al., 1996; Tokino et al., 1996). Recently in two out of nine BeckwithWiedemann syndrome (BWS) patients, who have a high susceptibility for WT and RMS, mutations were found in p57kip2 (Hatada et al., 1996b). However, in an earlier study ®ve breakpoints in BWS patients have been mapped on chromosome 11p15.5, together with the breakpoint of a rhabdoid tumor cell line (Hoovers et al., 1995; Newsham et al., 1994). These breakpoints do not a€ect the structure of the p57kip2 gene. Our analysis of p57kip2 in RMS also does not show mutations in p57kip2 (LJV et al., manuscript in preparation), but this still does not exclude p57kip2 as a TSG in RMS. It is conceivable that cis-acting elements are disrupted and therefore silencing of genes could be induced.

Allelotype in RMS M Visser et al

Chromosome 16q demonstrated LOH in 54% of the tested ERMS between the markers D16S752 and D16S539 on chromosome band 16q23-16qter. Chromosome 16q is also frequently a€ected in Wilms' tumors, 20% (Maw et al., 1992), hepatocellular carcinoma, 30% (Tsuda et al., 1990), prostate, 30% (Carter et al., 1990) and breast cancers, 52% (Tsuda et al., 1994). Recently, Weber-Hall et al. (1996) demonstrated 16q LOH in 40% of the tested ERMS by comparative genomic hybridization which con®rms our allelotype data. These observations make it likely that chromosome 16q harbors a TSG which plays an important role in RMS and other tumors. Several studies in Wilms' tumor describe di€erent regions on chromosome 16q which harbors the WT3 and the region of interest until now is between the chromosome bands 16q12.1-16q24.1 (Maw et al., 1992; Grundy et al., 1994; Austruy et al., 1995; Newsham et al., 1995). There might be more than one TSG located on chromosome 16q and this might explain the large region. Many other chromosomes are lost at a low frequency in RMS but, in view of the high FAL in ERMS, this might be random. The losses of chromosome 6p and 18p with a total LOH of 28% and 32% respectively also might not be random and warrant further analysis. The histological subgroups showed a di€erence with respect to their FAL. ARMS had a low FAL (0.06) while the ERMS had a high FAL (0.142). In some tumor types high FAL may be associated with a more advanced disease (Vogelstein et al., 1989), but in view of the poor prognosis of ARMS and the low FAL, this does not hold for the whole group of RMS. When we divided the ERMS based on their FAL in two groups (high/low FAL), no clinical parameters correlated with the FAL (data not shown). Recently, we reported a correlation between FAL and microsatellite instability. In the ERMS group, high FAL was correlated with a RER+phenotype (Visser et al., 1996). The FAL remained stable during the course of the disease, only minor di€erences were found in the FAL between the primary tumor and its subsequent metastases and relapses. This suggests that LOH of a critical region is an early and stable event in the carcinogenesis of ERMS and not essential for the development of metastases.

10 mm sections were prepared and the last section was reevaluated to test if there were no large changes in the tissue architecture. Isolation of DNA Three 10 mm sections were treated two times with xylene to dissolve the paran and washed with 100% ethanol. Sections were dried and incubated in 200 ml 10 mM Tris (pH 8.3), 1 mM EDTA 0.5% Tween 20 and 10 mg ml71 of proteinase K and incubated overnight at 378C. The DNA obtained was phenol/chloroform puri®ed according to standard methods. For the HE stained bone marrow smears, cells were scraped of the slides and DNA was extracted as described above. PCR ampli®cation Primers for the ampli®cation of the microsatellites were obtained from published sources and GDB (Table 1). Standard PCR (Saiki et al., 1988) was performed in a volume of 10 ml with 10 ± 80 ng DNA, 50 ng of each primer, 200 mM of each dNTP, 50 mM KCL, 10 mM Tris.Cl (pH 8.3), 0.8 U of Taq polymerase (Gibco BRL). The MgCl2 concentration was adjusted for each primer pair to get optimal ampli®cation and ranged from 0.5 ± 2.0 mM. PCR was performed in a multi well thermocycler (MJ Research, Waltham, MA) for 35 cycles of 1 min each at 948C, 558C and 728C. For detection of the PCR ampli®ed products, either 1 mCi of [a-32P]dATP (Amersham Life Sciences, Bucks, UK) was added to the PCR reaction or one primer was FITC end-labeled. The PCR products were denaturated and separated on a denaturing 6% polyacrylamide gel and exposed to an X-ray ®lm or in the case the FITC labeled primer, the PCR products were detected on an automated sequencer. Xenografts Human RMS xenografts were passaged two or three times through nude mice and a karyotype was constructed for each tumor. Tumor DNA was isolated by standard methods. Normal DNA derived from patients leucocytes was available. Criteria for LOH LOH was scored if the ratio of the two ampli®ed allels from tumor DNA di€ers at least twofold with the ratio of the two ampli®ed alleles in normal DNA.

Materials and methods Microdissection of normal and tumor tissue Tumor tissue was obtained from formalin-®xed, paran embedded tissue. If no constitutional tissue was available in the form of peripheral blood, normal tissue was obtained from the paran blocks or from bone marrow smears. Histological sections from the paran blocks were hematoxylin and eosin (HE) stained and microdissection was performed to separate normal and tumor tissue. Ten

Acknowledgements Thanks to Darryl Hake and Lisa Adams for assistance at RMS xenografts and Lilianne Wijnaendts for collecting tissue samples and help with the statistical analysis and Prof Dr A Westerveld, Dr Mannens and Drs M Alders for their comments on this manuscript. This work was supported by the Dutch Cancer Society, by the Foundation of Pediatric Cancer Research and by the Phi Beta Psi Sorority.

References Agamanolis DP, Dasu S and Krill CE. (1986). Hum. Pathol., 17, 778 ± 779. Austruy E, Candon S, Henry I, Gyapay G, Tournade M, Mannens M, Callen D, Junien C and Jeanpierre C. (1995). Genes, Chrom. & Cancer, 14, 285 ± 294.

Barr FG, Galili N, Holick J, Biegel JA, Rovera G and Emanuel BS. (1993). Nature Genet., 3, 113 ± 117. Barr FG, Chatten J, D'Cruz CM, Wilson AE, Nauta LE, Nycum LM, Biegel JA and Womer RB. (1995). JAMA, 27, 553 ± 557.

1313

Allelotype in RMS M Visser et al

1314

Besnard-GueÂrin C, Newsham I, Winqvist R and Cavenee WK. (1996). Hum. Genet., 97, 163 ± 170. Carter BS, Ewing CM, Ward S, Treiger BF, Aalders TW, Schalken JA, Epstein JI and Isaacs WB. (1990). Proc. Natl. Acad. Sci. USA, 87, 8751 ± 8755. Christofori G, Naik P and Hanahan D. (1994). Nature, 369, 414 ± 418. Davis RJ, D'Cruz CM, Lovell MA, Biegel JA and Barr FG. (1994). Cancer Res., 54, 2869 ± 2872. Fujimori M, Tokino T, Hino O, Kitagawa T, Imamura T, Okamoto E, Mitsunobu M, Ishikawa T, Nakagama H, Harada H, Yagura M, Matsubara K and Nakamura Y. (1991). Cancer Res., 51, 89 ± 93. Gaiger AM, Soule EH and Newtom WA. (1981). Natl. Cancer Inst. Monogr., 56, 19 ± 27. Galili N, Davis RJ, Fredericks WJ, Mukhopadhyay S, Rauscher FJ, Emanual BS, Rovera G and Barr FG. (1993). Nature Genet., 5, 230 ± 235. Grundy PE, Telzerow PE, Breslow N, Moksness J, Hu€ V and Paterson C. (1994). Cancer Res., 54, 2331 ± 2333. Hao Y, Crenshaw T, Moulton T, Newcomb E and Tycko B. (1993). Nature, 365, 764 ± 767. Hatada I, Inazawa J, Abe T, Nakayama M, Kaneko Y, Jinno Y, Niikawa N, Ohashi H, Fukushima Y, Lida K, Yutani C, Takahashi S, Chiba Y, Ohishi S and Mukai T. (1996a). Hum. Mol. Genet., 5, 783 ± 788. Hatada I, Ohashi H, Fukushima Y, Kaneko Y, Inoue M, Komoto Y, Okado A, Ohishi S, Nabetani A, Morisaki H, Nakayama M, Niikawa N and Mukai T. (1996b). Nature Genet., 14, 171 ± 173. Hoovers JMN, Kalikin LM, Johnson LA, Alders M, Redeker B, Law DJ, Bliek J, Steenman M, Benedict M, Wiegant J, Lengauger C, Taillon-Miller P, Schlessinger D, Edwards MC, Elledge SJ, Ivens A, Westerveld A, Little P, Mannens M and Feinberg AP. (1995). Proc. Natl. Acad. Sci. USA. 92, 12456 ± 12460. Knowles MA, Elder PA, Williamson M, Cairns JP, Shaw ME and Law MG. (1994). Cancer Res., 54, 531 ± 538. Knudson AG. (1971). Proc. Natl. Acad. Sci. USA, 68, 820 ± 823. Koi M, Johnson LA, Kalikin LM, Little PFR, Nakamura Y and Feinberg AP. (1993). Science (Washington DC), 260, 361 ± 364. Mannens M, Slater RM, Heyting C, Bliek J, Kraker de J, Coad N, Pagter-Holthuizen de P and Pearson PL. (1988). Hum. Genet., 81, 41 ± 48. Matsuoka S, Edwards MC, Bai C, Parker S, Zhang P, Baldini A, Harper JW and Elledge SJ. (1995). Genes & Development, 9, 650 ± 662. Matsuoka S, Thompson JS, Edwards MC, Barletta JM, Grundy P, Kalikin LM, Harper JW, Elledge S and Feinberg AP. (1996). Proc. Natl. Acad. Sci. USA, 93, 3026 ± 3030.

Maw MA, Grundy PE, Millow LJ, Eccles MR, Dunn RS, Smith PJ, Feinberg AP, Law DJ, Paterson MC, Telzerow PE, Callen DF, Thompson AD, Richards RI and Reeve AE. (1992). Cancer Res., 52, 3094 ± 3098. Newsham I, Daub D, Besnard-GueÂrin C and Cavenee W. (1994). Genomics, 19, 433 ± 440. Newsham I, Kindler-Rohrborn A, Daub D and Cavenee W. (1995). Genes, Chrom. & Cancer, 12, 1 ± 7. Orlow I, Iavarone A, Crider-Miller SJ, Bonilla F, Latres E, Lee M-H, Gerald WL, Massaque J, Weissman BE and Cordon-Cardo C. (1996). Cancer Res., 56, 1219 ± 1221. Pappo AS, Shapiro DN, Crist WM and Maurer HM. (1995). J. of Clinical Oncology, 13, 2123 ± 2139. Reeve AE, Sih SA, Raizis AM and Feinberg AP. (1989). Mol. Cell. Biol., 9, 1799 ± 1803. Reid LH, Crider-Miller SJ, West A, Lee M, Massague J and Weissman BE. (1996a). Cancer Res., 56, 1214 ± 1218. Reid LH, West A, Gioeli DG, Phillips KK, Kelleher KF, Araujo D, Stanbridge EJ, Dowdy SF, Gerhard DS and Weissman BE. (1996b). Hum. Mol. Genet., 5, 239 ± 247. Saiki RK, Gelfand DH, Sto€el S, Scharf SJ, Higuchi R, Horn GT, Mullis KB and Erlich HA. (1988). Science (Washington DC), 239, 487 ± 489. Scrable H, Cavenee W, Ghavimi F, Lovell M, Morgan K and Sapienza C. (1989a). Proc. Natl. Acad. Sci. USA, 86, 7480 ± 7484. Scrable H, Witte D, Shimada H, Seemayer T, Wang-Wuu S, Soukup S, Koufos A, Houghton P, Lamkin B and Cavenee W. (1989b). Genes, Chrom. & Cancer, 1, 23 ± 25. Tokino T, Urano T, Furuhata T, Matsushima M, Miyatsu T, Sasaki S and Nakamura Y. (1996). Hum. Genetics, 97, 625 ± 631. Tsuda H, Zhang W, Shimosota Y, Yokota J, Terada M, Sugimura T, Miyamura T and Hirohashi S. (1990). Proc. Natl. Acad. Sci. USA, 87, 6791 ± 6794. Tsuda H, Callen DF, Fukutomi T, Nakamura Y and Hirohashi S. (1994). Cancer Res., 54, 513 ± 517. Weber-Hall S, Anderson J, McManus A, Abe S, Nojima T, Pinkerton R, Pritchard-Jones K and Shipley J. (1996). Cancer Res., 56, 3220 ± 3224. Visser M, Bras J, Sijmons C, Devilee P, Wijnaendts LCD, van der Linden JC, VouÃte PA and Baas F. (1996). Proc. Natl. Acad. Sci. USA, 93, 9172 ± 9176. Vogelstein B, Fearo ER, Kern SE, Hamilton SR, Preisinger AC, Nakamara Y and White R. (1989). Science (Washington), 244, 207 ± 210. Winqvist R, Mannermaa A, Alavaikko M, Blanco G, Taskinen PJ, Kiviniemi H, Newsham I and Cavenee W. (1993). Cancer Res., 53, 4486 ± 4488. Zemel S, Bartolomei MS and Tilghman SM. (1992). Nature Genet., 2, 61 ± 65.