Loss of heterozygosity for loci on chromosome6q in human malignant melanoma

Item Type text; Thesis-Reproduction (electronic)

Authors Millikin, Dina Maria, 1953-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 11/04/2021 12:43:08

Link to Item http://hdl.handle.net/10150/277878

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may

be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate

the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by

sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company

300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600

Order Number 1343817

Loss of heterozygosity for loci on chromosome 6q in human malignant melanoma

Millikin, Dina Maria, M.S.

The University of Arizona, 1991

U M I 300 N. Zeeb Rd. Ann Arbor, MI 48106

NOTE TO USERS

THE ORIGINAL DOCUMENT RECEIVED BY U.M.I. CONTAINED PAGES WITH PHOTOGRAPHS WHICH MAY NOT REPRODUCE PROPERLY.

THIS REPRODUCTION IS THE BEST AVAILABLE COPY.

1

LOSS OF HETEROZYGOSITY FOR LOCI ON

CHROMOSOME 6q IN HUMAN MALIGNANT MELANOMA

by

Dina Maria Mi Hi kin

A Thesis Submitted to the Faculty of the

DEPARTMENT OF MICROBIOLOGY AND IMMUNOLOGY

In Partial Fulfillment of the Requirements For the Degree of

MASTER OF SCIENCE

WITH A MAJOR IN MICROBIOLOGY

In the Graduate College

THE UNIVERSITY OF ARIZONA

19 9 1

2

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgement the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: /<CUHOu hj.

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

y M. Trent, Ph.Dw WQP-

3

DEDICATION

I dedicate this thesis to my father, Alberto Torres E.,

whose love and dedication to scientific work has always been an

example to me.

4

ACKNOWLEDGEMENTS

I would like to give special thanks to Dr. Jeffrey M. Trent,

my research advisor, for his continous support and guidance. Also,

I thank the members of my committee Dr. Anne Cress and Dr. Tim

Bowden for sharing their time and knowledge.

I express my greatest appreciation to my husband, Robert,

and my children, David and Lauren, whose love and understanding made

this achievement possible for me.

5

TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS 6

LIST OF TABLES 7

ABSTRACT 8

INTRODUCTION 9

Cancer Genes 9 Tumor Suppressor Genes and Retinoblastoma 10 Evidence for Tumor Suppressor Genes 13 Loss of Genetic Sequences Associated with Human Malignant Melanoma 16

MATERIALS AND METHODS 17

Reagents 17 Polymorphic DNA Probes 18

Patient Samples 21 Isolation of Genomic DNA 21 Southern Transfer 22 Southern Prehybridization and Hybridization 23 Transformation of E.coli DH5a cells 24 Preparation of Plasmid DNA from Transformed Bacteria.... 24 Isolation of Insert DNA Probes by LMP Gel Electrophoresis 25 Radiolabeling of DNA Probes by the Random Primed Oligonucleotide Method 26 Removal of Repetitive Sequences from DNA Probes 26

RESULTS 27

Selection of Polymorphic DNA Probes 27 Detection of Allele Loss by Southern Blot Analysis 29 Normal and Tumor Genotypes of Twenty Melanoma Patients at Six Loci on Chromosome 6 41 LOH at Six Loci on Chromosome 6 41 RFLP Analysis for Other Chromosome Loci 44 Results Summary 46

DISCUSSION 49

REFERENCES 56

LIST OF ILLUSTRATIONS

6

Figure Page

1. Possible chromosomal mechanisms by which a first predisposing mutation can be revealed in retinoblastoma 12

2. Representative Southern blot showing allele loss with reduction to homozygosity" at the c-MYB locus 30

3. Representative Southern blot showing allele loss with "reduction to hemizygosity" at the c-MYB and S0D2 loci 32

4. Representative Southern blot showing allele loss at the CGA locus 34

5. Representative Southern blot showing allele loss at the D6S37 locus 35

6. Representative Southern blot showing allele loss at the S0D2 locus 37

7. Representative Southern blot showing allele loss at the ESR locus 39

8. Representative Southern blot showing no allele loss at the 6p locus D6S29 40

9. Schematic representation of LOH at six loci on chromosome 6 43

10. Representative Southern blot showing LOH at the lp locus H-RAS 45

7

LIST OF TABLES

Table Page

1. Polymorphic DNA Probes for Chromosome 6 19

2. Polymorphic DNA Probes for Other Chromosomes 20

3. Densitometric Analysis of cases with "reduction to hemizygosity" 33

4. List of Constitutional and Tumor Genotypes of Twenty Melanoma Patients at Six Loci on Chromosome 6 42

5. Frequency of LOH for Five Loci on Chromosome 6q in Malignant Melanoma 47

6. Frequency of LOH for Loci on Four Other Chromosomes in Malignant Melanoma 48

ABSTRACT

8

Restriction fragment length polymorphism (RFLP) analysis was

used as a molecular genetic approach to examine loci on chromosome

6q for loss of constitutional heterozygosity (LOH). Five

polymorphic DNA markers (that recognize RFLP's at specific

chromosome 6q loci), and one polymorphic DNA marker for 6p were used

to screen 20 autologous pairs of melanoma tumor DNA and normal blood

DNA to determine the tumor and constitutional genotypes of each

patient. LOH on chromosome 6q was identified at 21/53 informative

loci (40%). Five patients had allele losses consistent with the

loss of 6q (or of an entire copy of 6), and four other patients had

terminal deletions of 6q. The chromosomal region bearing the

highest frequency of 6q allele loss (60%) is defined by the marker

loci c-MYB and ESR (q22-23 and q24-27). Comparison was then made

with LOH observed at loci on four different chromosomes (lp, lip,

16q and 17p) with results indicating a loss in 5% of cases. These

results suggest that the loss of somatic sequences from chromosome

6q 1s a nonrandom and possibly biologically relevant event in human

malignant melanoma.

INTRODUCTION

9

Cancer Genes

Malignant transformation is a process that involves multiple

genetic alterations (Land, 1983). Significant progress has been

accomplished in this area during the past decade due to the

discovery of certain cellular genes that play important roles in

cancer development and progression. These genes can be grouped into

two main categories according to whether they contribute to

malignancy in a positive or negative fashion: 1) "oncogenes", and 2)

"tumor suppressor genes". The oncogenes have been very widely

studied, and are thought to act as positive effectors of cell growth

(Bishop, 1983; Varmus, 1984; Newmark, 1987). Several oncogenes have

been associated with specific cancers and are thought to derive from

preexisting cellular genes (proto-oncogenes) which are involved in

normal cellular growth, and which can be activated by mechanisms

such as gene amplification, chromosome translocation, point

mutation, or provirus insertion (Tabin et al., 1982; Cori, 1986;

Hayward et al., 1981; Manne et al., 1985; Schwab et al., 1984).

Tumor suppressor genes, on the other hand, are negative regulators

of cell growth and can therefore suppress cellular transformation

and/or tumorigenesis. As described below, it is thought that loss

or inactivation of both copies of a tumor suppressor gene, removes a

block to proliferation, leading to uncontrolled cell growth (Hansen

& Cavenee, 1987).

10

Tumor Suppressor Genes and Retinoblastoma

Tumor suppressor genes are part of a very important and complex

regulatory circuit whose function is to constrain or downregulate

cellular growth (for a review refer to: Geiser & Stanbridge, 1989;

Friend et al.f 1988; Klein, 1988; Weinberg, 1989). Their function

is opposite to the well studied oncogenes, and they have also been

refered to as "recessive oncogenes", "emerogenes", and "growth

suppressor genes" (Klein, 1985; Klein, 1987).

The modern concept of tumor suppressor genes stems from the work

of Alfred Knudson in retinoblastoma (Knudson, 1971). Retinoblastoma

is an embryonal childhood eye tumor that can be inherited as an

autosomal dominant trait with high penetrance. Knudson's theory

explained the genetic predisposition to this type of cancer by

suggesting that retinoblastoma (both inherited and sporadic forms)

arises as a consequence of two mutational events. In the case of

familial (inherited form) retinoblastoma, the first mutation is

inherited from a carrier parent in the germline, while the second

mutation occurs somatically in the fetal or infant retina. However,

in sporadic cases Knudson proposed that both mutations occur

somatically in the target retinal cells.

Cytogenetic analysis of retinoblastoma tumor cells revealed

frequent chromosomal interstitial deletions of band 13ql4 (Yunis &

Ramsey, 1978), which harbors a specific gene, the Rb gene.

Presently, we know that the Rb gene is the target site of mutations

and/or deletions which cause loss of Rb gene function leading to

retinoblastoma (Sparkes et al., 1983; Godbout et al.,1983; Cavenee

et al., 1983; Dryja et al., 1984). Because loss or inactivatlon of

both copies of the Rb gene is necessary for the establishment of the

tumorigenie phenotype, it is thought that the first predisposing

mutation is recessive and can only be unmasked by a second somatic

"hit"at the Rb locus. The possible chromosomal mechanisms by which

a recessive predisposing mutation (rb) can be revealed in

retinoblastoma are illustrated in figure 1 (taken from Hansen &

Cavenee, 1987). These include: mitotic chromosome nondisjunction,

mitotic nondisjunction with reduplication of the mutant rb allele,

deletion, mitotic recombination, and other events such as gene

conversion.

Thus, the Rb gene constitutes the prototype of a tumor

suppressor gene. Extensive recent work in the molecular mechanisms

of retinoblastoma has led to the isolation and cloffing of the Rb

gene (Friend et al., 1986; Lee et al., 1987; Fung et al., 1987).

Sequencing of the Rb cDNA revealed a phosphoprotein of 928 amino

acids, approximately 105 Kd in size, which is found in a variety of

tissues (Lee et al., 1987; Friend et al., 1987). Some of the

presently known biochemical properties of pl05-Rb include:

nonspecific binding to DNA (Lee et al, 1987), association with the

adenovirus E1A transforming protein (Whyte et al, 1988), association

with the large T antigen of SV40 (De Caprio et al, 1988), and

association with the E7 oncoprotein of human papilloma virus type 16

(Dyson et al, 1989).

Besides the Rb gene, other putative tumor suppressor genes

cloned to date include: the p53 gene from colon and lung carcinoma

Heritable

1 rb

1

2 +

12 Sporadic

a

o

1 2 + +

1 2

2 + 2

Constitutional

inherited ^ germinal mutation somatic mutation

nondisjunction nondisjunction

reduplication deletion recombination other events

Figure 1. Possible chromosomal mechanisms by which a first predisposing mutation can be revealed in retinoblastoma.

(Oren & Levine, 1983), the WT gene from Wilm's tumor (Call et al,

1990), and the DCC gene from colorectal carcinoma (Fearon et al,

1990). The normal biochemical and biological functions of these

gene products is under current investigation. The p53 protein,

which was identified by its ability to form stable complexes with

SV40 large T antigen and the Elb protein from adenovirus in virus-

transformed rodent cells (Lane & Crawford, 1979), is a nuclear

phosphoprotein which has recently been shown to have a

transcription-activating domain near the amino terminus (Fields &

Jang, 1990; Raycroft et al, 1990). The WT gene product also seems

to have a role 1n tissue-specific transcriptional activation (Call

et al, 1990), while the DCC gene product bears homology to proteins

implicated in cell adhesion (Fearon et al, 1990).

Clearly, understanding the biochemical properties of the

products of tumor suppressor genes will help explain the normal

role(s) of these genes in the cell and the way in which their loss

or inactivation contribute to cancer development.

Evidence for Tumor Suppressor Genes

Frequent nonrandom loss of specific chromosomal regions in a

specific tumor type (eg. retinoblastoma) has constituted the first

clue to the possible presence of a tumor suppressor gene. Nonrandom

chromosomal abnormalities such as deletions and translocations of

specific chromosome regions have been associated with a number of

tumor types (Sandberg, 1990).

Second, the area of somatic cell hybridization has provided

strong biological evidence for the involvement of growth suppressor

genes in malignant transformation and tumorigenesis (Sager, 1986;

Stanbridge, 1987; Geiser & Stanbridge, 1989). By fusing normal

(nontumorigenic) to malignant (tumorigenie) cells in culture,

researchers have been able to obtain stable hybrids that are

nontumorigenic like their normal parent cells. Suppression of

tumorigenicity in these hybrids indicates the presence of tumor

suppressor genes in the chromosome(s) derived from the normal

parent. Interestingly, the maintenance of the nontumorigenic

phenotype depends upon the stability of these hybrids, since loss of

certain normal parent-derived chromosomes results in reappearance of

tumorigenicity (Stanbridge et al., 1981). Presently, it is possible

to study suppression of tumorigenicity by specific human

chromosomes, through a relatively new method of microcell-mediated

chromosome transfer (Saxon et al., 1985; Saxon, et al., 1986), which

allows for the selective introduction of a specific human chromosome

Into the genome of a tumorigenie cancer cell.

Third, the above experimental evidence is strengthen by a

molecular genetic approach that identifies specific chromosome loci

that are frequently lost in human cancers. This approach uses

informative polymorphic DNA markers that recognize restriction

fragment length polymorphisms (RFLP's) for specific chromosome loci

in the DNA of cancer patients. Southern blot analysis of autologous

pairs of normal and tumor DNA digested with the appropriate

restriction enzyme and hybridized to these probes, may identify loss

of alleles in the tumor DNA. If an allele is lost, the loci defined

by these polymorphic DNA probes are said to have undergone a

"reduction to homozygosity or hemizygosity" (also referred to as

loss of constitutional heterozygosity LOH). Frequent LOH for

specific chromosome loci in different samples of a given tumor type

(e.g. retinoblastoma) represents strong evidence for the existence

of a tumor suppressor gene at those loci. Accordingly, the

polymorphic DNA markers can be used to define the region where a

putative tumor suppressor gene may reside.

Tumor-specific, nonrandom LOH has been observed in several human

cancers such as: retinoblastoma (Cavenee et al., 1983; Cavenee et

al., 1985; Benedict et al., 1987), Wilm's tumor (Koufos et al.,

1985; Dao et al., 1987; Mannens et al., 1988; Mannens et al., 1990),

breast carcinoma (Lundberg et al., 1987; Ali et al., 1987; Mackay et

al., 1988; Devi lee et al, 1989), small cell lung carcinoma (Brauch

et al., 1987; Naylor et al., 1987; Yokota et al., 1987; Harbour et

al., 1988), malignant astrocytoma (Fults et al., 1989; James et al.,

1989), neuroblastoma (Brodeur et al., 1988; Brodeur et al, 1989;

Fong et al., 1989; Suzuki et al, 1989), colorectal carcinoma (Fearon

et al., 1987; Okamoto et al., 1988; Vogelstein et al., 1988; Baker

et al., 1989; Vogelstein et al, 1989), meningioma (Dumanski et al.,

1987; Seizinger et al., 1987), and glioblastoma multiforme (Fujimoto

et al, 1989).

16

Loss of Genetic Sequences Associated with

Human Malignant Melanoma

Next to lung cancer in females, malignant melanoma is the tumor

with the highest increase in incidence in the U.S., annually

accounting for approximately 3% of all diagnosed cancers (Evans et

al., 1988; Thorn et al., 1989).

Cytogenetic analysis of melanoma tumor cells reveals numeric and

structural alterations of chromosomes 1, 2, 3, 6, and 7 (Trent et

al., 1989). However, the most common structural alteration involved

the long arm of chromosome 6 (6q), which was found to be deleted

(6q-) in about 40% of cases (Trent et al., 1983; Becher et al.,

1983; Balaban et al., 1984; Pedersen et al., 1986; Parmiter et al.,

1986; Trent et al., 1989).

Recently, direct support for the existence of a putative tumor

suppressor gene associated with malignant melanoma came from the

work of Trent et al. (1990). These investigators utilized the

technique of somatic cell hybridization by microcell transfer to

succesfully introduce a normal copy of chromosome 6 into two human

melanoma cell lines. Suppression of tumorigenicity was observed in

the hybrids, based on inability to form tumors in nude mice.

Furthermore, loss of the introduced normal chromosome 6 resulted in

reversion to the tumorigenie phenotype.

In order to augment the previously described cytogenetic and

biological evidence for the frequent loss of chromosome 6q in

malignant melanoma, I have used RFLP analysis to investigate the

frequency of loss of somatic sequences from chromosome 6q in a group

of 20 patients with histologically confirmed malignant melanoma.

Five informative DNA markers that detect RFLP's at specific

chromosome 6q loci, and one DNA marker for a 6p locus were used to

screen pairs of normal and tumor DNA from each patient for LOH by

Southern blot analysis. In addition, in order to confirm the

specificity of chromosome 6q allele loss, four polymorphic VNTR

(variable number of tandem repeats) probes for chromosomes lp, lip,

16q, and 17p were also used to identify LOH in the same group of

melanoma patients.

MATERIALS AND METHODS

Reagents

All chemical reagents for preparation of the Cell Lysis Buffer and

the proteinase K were purchased from Sigma Chemical Company (St.

Louis, MO). The phenol was nucleic acid grade and was obtained from

Bethesda Research Laboratories (Gaithesburg, MD). All restriction

endonucleases and restriction buffers were purchased from Boehringer

Mannheim (Indianapolis, IN). Frozen competent cells E. coli DH5a

were obtained from BRL, and the reagents for bacterial culture media

were obtained from Difco (Detroit, MI). The large fragment of

coli DNA polymerase I (Klenow) was purchased from Boehringer

Mannheim. The hexadeoxyribonucleotide primers (pdN6) were obtained

from Pharmacia LKB Biotechnology (Piscataway, NY). The dNTP's

(dCTP, dTTP & dGTP) were purchased from Sigma Chemical Corporation.

18

The [a-32P]dATP (3000 Ci/mmole, 10/iCi//il) was purchased from ICN

Biomedicals, Inc.(Costa Mesa, CA). The sodium dodecyl sulfate (SDS)

was from BRL, the Dextran sulfate was purchased from Pharmacia, and

the salmon sperm DNA was obtained from Sigma Chemical Company.

POLYMORPHIC DNA PROBES

The polymorphic human DNA probes used in this study were obtained

from the following sources: pHHH157 (D6S29) was purchased from ATCC

(American Type Culture Collection repository); pCGalpha (CGA) was

kindly provided by Dr. I. Boime (University of Washington, School of

Medicine); pJCZ30 (D6S37) was a gift from Dr. B. Vogelstein (Johns

Hopkins Oncology Center); pHMnS0D4 (S0D2), pHM2.6 (c-MYB), and p0R3

(ESR) were purchased from ATCC. The polymorphic (VNTR) probes pYNZ2

(D1S57), pEJ988 (H-ras), 79-2-23 (D16S7) and pYNZ22.1 (D17S30) were

kindly provided by Dr. B. Vogelstein (Johns Hopkins Oncology

Center). All the probes were checked by the size of the insert and

vector after agarose gel electrophoresis and by hybridization to

Southern blots of normal human DNA digested with the appropriate

restriction enzyme and used as controls. A complete description of

these DNA probes is provided in Tables 1 and 2. Most of this

information was taken from the report on Human Gene Mapping 9 (1987)

and 9.5 Proceedings (1988).

Table 1. Polymorphic DNA Probes for Chromosome 6.

Chromosome Region

Marker Probe Locus

Vector/Site Size Enzyme Alleles Freq. N

6p

6q12-q21

6q13-q21*

6q21

6q22-q23

6q24-q27

D6S29

CGA

D6S37

SOD2

c-MYB

ESR

pHHH157

pCGalpha

pJCZ30

pUC18/BamHI

pBR322/Pstl

pUC18/Hincll

5.5 Kb

pHM2.6 pKH47/EcoRI 2.6 Kb

pOR3 pBR322/EcoRI 1.3 Kb

BamHI

0.44 Kb EcoRI

3.0 Kb Hindlll

phMnSOD4 pBR327/EcoRI 0.82 Kb Taql

EcoRI

Pvull

A1 (13.0 Kb) A2 ( 5.5 Kb)

A1 (10.0 Kb) A2 ( 5.0 Kb)

0.50 0.50

0.36 0.64

216

110

VNTR 4 alleles Heterozygosity=.74 from 1.0-3.0 Kb Also detected with EcoRI, Taql, Mspl Pstl, Pvull and BamHI.

A1 (2.3 Kb) 0.40 30 A2 (2.0 Kb) 0.60 B1 (1.5 Kb) 0.27 31 B2 (1.2 Kb) 0.73

A1 (2.6 Kb) 0.48 98 A2 (1.5+1.1Kb) 0.52

A1 (1.5 Kb) 0.48 28 A2 (0.7 Kb) 0.52

* Localized for this study by hybridization to DNA from an available deletion hybrid mapping pannel.

Table 2. Polymorphic

Chromosome Marker Probe Vector/Site Region Locus

1p D1S57 pYNZ2 pBR322/Hindlll

11 p H-RAS pEJ988 Bluescript/?

16q D16S7 79-2-23 pSP65/BamHI

17p D17S30 pYNZ22.1 pBR322/BamHI

DNA Probes for Other Chromosomes.

Size Enzyme Alleles Frequency

3.0 Kb Mspl VNTR with 8 alleles Het.= .65 from 1.0-3.0 Kb

Also detected with Taql, Hindlll, Pvull, Bglll, EcoRI BamHI, Pstl & Rsal

? Mspl VNTR with >3 alleles Het.= .85 from 1.0-5.0 Kb Also detected with Taql

1.45 Kb Taql VNTR with >10 alleles Het.= .80 Also detected with Mspl, Hindlll, Pvull, EcoRIS Rsal

1.70 Kb Mspl VNTR with 10 alleles Het.= .86 Rsal from 1.5-3.0 Kb

Also detected with Taql

21

Methods

PATIENT SAMPLES

Twenty histologically confirmed malignant melanoma metastatic tumor

biopsies were obtained from a total of twenty patients immediately

after surgery. The fresh solid tissues were stored at -70°C until

DNA was extracted. Peripheral blood samples (15 ml) were obtained

from each melanoma patient and used to prepare DNA from the

lymphocyte fraction. Peripheral blood lymphocytes (PBL) were

isolated by using a Histopaque 1077 kit obtained from Sigma Chemical

Co. (St. Louis, MO) following the procedure provided by the

manufacturer.

ISOLATION OF GENOMIC DNA

The frozen melanoma tumor specimens were removed from -80°C storage,

immersed in liquid Nitrogen and pulverized. The powdered tissue was

lysed in 5 ml. of Lysis Buffer (20mM Tris-HCl pH8.0; 10 mM EDTA;

0.5% SDS; 100 mM NaCl) and 100 /<g/ml of proteinase-K, and incubated

in a water bath at 55'C overnight. The lymphocytes (PBL) were

washed twice with Phosphate Buffer Saline (PBS) and the final cell

pellet was lysed in 1-2 ml of lysis buffer and 50 /jg/ml of

proteinase-K under the above conditions. The next day the lysates

were extracted by mixing an equal volume of buffer saturated phenol,

mixed and centrifuging at 4,000 RPM for 10 min at room temperature

using a table top Beckman centrifuge Model TJ-6 (Beckman

Instruments, Palo Alto CA). The top aqueous phase was then

transferred to a clean centrifuge tube and extracted with an equal

volume of chloroform:isoamyl alcohol (24:1), mixed and centrifuged

at 1,500 RPM for 5 min. at room temperature. The top aqueous phase

was again transfered to a clean centrifuge tube and precipitated

with a double volume of 100% cold ethanol (or an equal volume of

cold isopropanol) and incubated at -20°C for 1 hr. The precipitated

DNA was then pelleted at 9,000 RPM for 10 min using a Beckman

midspeed centrifuge, model J2-21M. The DNA pellet was rinsed with

70% cold ethanol and centrifuged at 9,000 RPM for 10 min in the same

centrifuge. The pellet was then air dried and resuspended in TE

buffer (10 mM Tris-HCl; 1 mM EDTA pH 8.0). The resuspended DNA was

treated with RNase (10 /jg/ml) and re-extracted with phenol and

chloroform and ethanol precipitated. The final DNA pellet is rinsed

with 70% ethanol, air dried and resuspended in TE buffer.

SOUTHERN TRANSFER

Five micrograms of genomic DNA was digested with the appropriate

restriction endonuclease (EcoRI, PvuII, TaqI or BamHI) according to

the manufacturers recommended conditions. The DNA was tested for

complete digestion by running 300-400 fiq of digested genomic DNA

through a 0.9% agarose mini gel at 70V for 1-2 hrs, using a Mini-

Horizontal Agarose Submarine Electrophoresis apparatus Model HE 33B.

The mini gel is then stained with Ethidium Bromide (0.5 /ig/ml) and

the DNA fragments are visualized under U.V. light, using a

Transilluminator apparatus model UVT-400M by IBI (International

Biotechnologies, Inc., New Haven, CT). Photographs of the gels were

23

taken using a Quickshooter Photosystem model QSP-46400 by IBI (New

Haven, CT). Next, the completely digested DNA samples were loaded

on a 0.9% agarose gel (11.0 X 14.0 cm) in DNA loading buffer (0.04%

bromophenol blue, 2.5% ficoll) and electrophoresed overnight at 22V

in 1XTBE running buffer (0.09 M Tris-Borate; 0.002 M EDTA pH8.2),

using a Horizontal Gel Electrophoresis apparatus model H5 (BRL,

Gaithesburg, MD), and a model 250 Power Supply (BRL, Gaithesburg,

MD). Following electrophoresis, the gel is stained in Ethidium

Bromide (0.5 /xg/ml) and photographed under short wave UV light. The

DNA 1n the gel is denatured in Transfer Solution (0.4 M NaOH; 0.5 M

NaCl) for 20-30 min prior to transfer on to Zeta-probe membrane

(Bio-Rad Laboratories, Richmond, CA). The transfer is allowed to

proceed for 24 hrs, after which the blot is neutralized in

Neutralization Solution (0.5 M Tris-HCl; 1.0 M NaCl pH 7) for 30 min

and allowed to air dry. The dry southern blots were baked in a

vaccum oven at 80°C for 1.5-2 hrs.

SOUTHERN PREHYBRIDIZATION AND HYBRIDIZATION

The southern blots were prehybridized at 65*C for 5 hrs to overnight

in a plastic bag with Prehybridization/Hybridization solution

containing 4X SET (0.6 M NaCl; 8 mM EDTA; 120 mM Tris-HCl pH 7.4),

0.1% NaPPi, 1.0% SDS, 10% Dextran sulfate, 100 /ig/ml denatured

salmon sperm DNA. After prehybridization, the radioactively labeled

DNA probe (denatured by boiling for 5-10 min) was added to the bag

and hybridization was allowed to proceed for 24 hrs at 65°C.

Next day, the blots were washed several times at 56°C in 1% SDS,

0.IX SSC and exposed to X-ray film for autoradiography. The

hybridized southern blots can be stripped by incubating the blots in

Wash solution at 90°C for 20 min with gentle shaking. After testing

for complete removal of the radiolabeled probe by autoradiography,

the stripped blots can be used in subsequent hybridizations to other

DNA probes. Multiple blot stripping did not significantly reduce

the hybridization signal.

TRANSFORMATION OF E.coli DH5a CELLS

The transformation of competent E.coli DH5o was performed according

to the manufacturer's instructions (BRL, Gaithesburg, MD). To 20 /*1

of competent cells (kept chilled on ice), add 1 /jl of diluted probe

DNA (about 10 ng), shake gently and place on ice for 30 min. Heat

pulse 40 sec at 42*C, and place on ice. Add 80 /*1 of S.O.C. media

(2% Bactotryptone; 0.5% Yeast extract; 10 mM NaCl; 2.5 mM KC1; 20mM

MgCl-MgS04, 20mM Glucose, pH 7.0) and incubate for 1 hr at 37°C,

shaking at 225 rpm. Plate cells in LB agar (1% Bactotryptone; 0.5%

Yeast extract; 170 mM NaCl; 1.5% Bactoagar; pH 7.0) in the presence

of the appropriate antibiotic.

PREPARATION OF PLASMID DNA FROM TRANSFORMED BACTERIA

This procedure is as described by Birnboim & Doly (1979). It

involves alkaline lysis of the bacteria cells. A 100 ml liquid

culture (LB broth) is grown at 37°C in the appropriate antibiotic.

The cells are pelleted by centrifugation at 4000xg for 10 min using

a Beckman midspeed centrifuge model J2-21M (Beckman Instruments,

Palo Alto, CA). The bacterial pellets are vigorously resuspended in

4 ml of Solution I (50mM glucose; 25mM Tris-HCl pH 8; lOmM EDTA) and

2 mg/ml of Lysozyme (Sigma Chemical Company, St. Louis, MO). After

a 30 min incubation on ice, the cells are lysed in 8 ml of 0.2M

NaOH; 0.5% SDS, mixed and allow to stand on ice for 10 min.

Differential precipitation of E. coli chromosomal DNA is carried on

in the presence of 6 ml of 5M potassium acetate (pH 4.8). After a

5-10 min incubation on ice, the suspension is centrifuged at 13,000

RPM for 20 min at 4*C using the same Beckman J2-21M model

centrifuge. The bacterial DNA becomes entangled in the cell debris

and forms a tight pellet. The supernatant is saved and used in a

typical DNA extraction procedure as described earlier to isolate

plasmid DNA.

ISOLATION OF INSERT DNA PROBES

BY LMP GEL ELECTROPHORESIS

Purified plasmid DNA was digested with the appropriate restriction

enzyme, in order to excise the cloned insert DNA. Typically 1-5 pg

of plasmid DNA were digested for 1-2 hrs under conditions

recommended by the enzyme supplier. A small aliquot of the reaction

mixture, containing 100-500 ng of total DNA, was loaded on to a 1%

LMP agarose minigel alongside a ladder of XDNA/Hindlll fragments

(1.0-2.0 /ig/lane), and electrophoresed at 60V for about 2 hrs in IX

OLB buffer (40mM Tris-acetate; 5mM sodium acetate; ImM EDTA; pH

8.0). The gel is then stained with ethidium bromide and

26

photographed under UV light. The desired band is excised from the

gel and solubilized in sterile water (0.3ml water/mg gel), boiled

for 7 min and stored at -20°C.

RADIOLABELING OF DNA PROBES

BY THE RANDOM PRIMED OLIGONUCLEOTIDE METHOD

The LMP-isolated DNA probes or plasmid DNA were first denatured by

boiling for 3-5 min and placed at 37'C for a brief time until

initiating the labeling reaction. Thirty to sixty nanograms of DNA

were radioactively labeled with 50/*Ci of d-ATP[a-32P] (3000

Ci/mmole) by the random primed oligonucleotide method (Feinberg &

Vogelstein, 1983) to a specific radioactivity >108 cpm//ig at room

temperature for more than 5 hrs. The radiolabeled probe is

denatured by boiling for 5-10 min prior to hybridization to blotted

genomic DNA.

REMOVAL OF REPETITIVE SEQUENCES

FROM DNA PROBES

When human DNA probes are derived from genomic libraries, it is not

uncommon to isolate fragments containing highly repetitive sequences

which interfere with the identification of bands of single copy DNA.

In order to circumvent the complicated process of subcloning these

fragments, Litt and White (1985) described a method of pre-

reassociation of these human probes with a large excess of sonicated

human placental DNA. The probe 79-2-23 was treated in this manner

with positive results, by incubating the radiolabeled probe (50/* 1

reaction) with 200 /xg of sonicated human placental DNA (Sigma

Chemical Company, St. Louis, MO) and 5X SSC. This is boiled for 5

min, cooled in ice for 1 min and incubated at 68*C for 100 min

(CotlOO) prior to adding to prehybridized filters.

RESULTS

Selection of Polymorphic DNA Probes

The human DNA segments used as probes in this study are

polymorphic and can detect restriction fragments of different

lengths (alleles) after digestion with DNA sequence-specific

restriction endonucleases. Variation in allele sizes can be due to

loss or formation of a new restriction enzyme cleavage site

("restriction site polymorphism"), resulting in two possible alleles

identified (A1 and A2): the homozygous individual will have one of

these fragment lengths(Al/Al or A2/A2), whereas the heterozygous

individual will have both fragments(A1/A2). In addition to

restriction site polymorphisms, insertion or deletion of blocks of

moderately repetitive DNA can also alter the size of a restriction

fragment. This latter type of polymorphism is termed a VNTR

(variable number of tandem repeats) polymorphism and is highly

variable within a population (and can therefore identify multiple

alleles of different lengths). DNA probes that identify VNTR

polymorphisms are thus highly informative in their capacity to

identify heterozygosity among individuals and represent a very

useful tool in gene mapping studies (Nakamura et al., 1987).

Selection of DNA probes for this study was made by considering

the following:

1) High PIC value. The PIC value (polymorphism information content)

is used in linkage studies as an indication of how informative a

marker locus is in determining linkage to an index locus (Botstein

et al., 1980). Marker loci with PIC >.5 are considered highly

informative, if .5 > PIC >.25 then the marker 1s reasonably

informative, and if PIC <.25 then the marker would be siightly

informative. The polymorphic probes used in this study had PIC

values of .32 to .38, corresponding to reasonably informative marker

loci.

2) Common Restriction enzyme polymorphism. Due to the limited

amount of patient DNA, it was necessary to select probes that shared

specificity for a certain restriction enzyme site polymorphism.

However, it is important to mention that when using polymorphic VNTR

probes, this limitation is eased by the fact that VNTR alleles can

be revealed by several restriction enzymes (e.g. marker D6S37

detects several VNTR alleles with either Hindlll, EcoRI, TagI, MsPI,

PstI, PvuII or BamHI digests).

3) Chromosomal Localization. The 6g marker loci CGA (6gl2-21),

D6S37 (ql3-21), S0D2 (6q21), c-MYB (6q22-23), and ESR (6q24-27) span

the entire long arm of chromosome 6; localization of the marker

D6S37 was performed for this study by hybridization to genomic DNA

from an available somatic cell hybrid deletion mapping panel for 6q

(data not shown). The 6p marker D6S29 was selected as a control for

6p loss and has not been assigned a specific chromosome band on 6p.

Table 1 (Methods section) comprises information on each one of these

DNA markers. The VNTR markers for chromosomes lp, lip and 17p were

selected based on the reported presence of growth suppressor genes

on these chromosome regions, while the 16q marker was chosen

arbitrarily. Complete information on these VNTR markers is included

in table 2 (Methods section).

Detection of Allele Loss by

Southern Blot Analysis

Two kinds of allele losses were detected in southern blots of

autologous combinations of tumor (T) and normal blood (N) DNA from

melanoma patients. The first is loss of an allele from a

constitutionally heterozygous individual, resulting in reduction to

homozygosity or LOH in the tumor DNA. Figure 2 is an autoradiogram

of a southern blot of EcoRI digested tumor and normal DNA, probed

with pHM2.6 (c-MYB) (Dozier et al., 1986). This probe detects two

alleles: Al (2.6 Kb) and A2 (1.5 Kb + 1.1 Kb). Patient SH is

constitutionally homozygous for the A2 allele, while patient MN is

homozygous for the Al allele. Patient ZO, however, is

constitutionally heterozygous (A1/A2) and therefore "informative" by

heterozygosity. This patient has lost the A2 allele in his tumor,

resulting in loss of constitutional heterozygosity (LOH) at the c-

MYB locus (6q22-23). The faint bands detected at the site of the

lost A2 allele are due to the presence of trace amounts of normal

DNA derived from normal cells infiltrating the tumor specimen.

30

C-MYB (pHM 2.6)

EcoRI

Figure 2. Allele loss with "reduction to homozygosity" or loss of heterozygosity (LOH). Southern blot of EcoRI digested tumor (T) and normal (N) DNA (5/ig/lane) from melanoma patients SH, MN and ZO, probed with c-MYB (pHM2.6). Patient SH is homozygous for the A2 allele (1.5 Kb + 1.1 Kb); MN is homozygous for the A1 allele (2.6 Kb); ZO is constitutionally heterozygous (informative) and his tumor underwent a "reduction to homozygosity" or LOH.

31

A second kind of allele loss was detected in patients that were

constitutionally homozygous but had lost one allele in their tumor

DNA, resulting in reduction to hemizygosity. A 50% reduction in the

intensity of the hybridization signal in the tumor cells is

indicative of this event, and can be confirmed by scanning

densitometry. Figure 3 shows an example of reduction to

hemizygosity in patient KY at the c-MYB locus (panel A) and at the

S0D2 locus (panel B). Southern blots that had been hybridized to

these 6q probes were stripped and rehybridized to other probes used

as controls for DNA loading. In this way, three other cases of

reduction to hemizygosity were identified. Table 3 contains the

results of densitometric analysis and calculated T/N ratios for

patients GZ, HY and KY at the c-MYB locus and for patient KY at the

S0D2 locus. The probes used for controls in each case are specified

in the same table.

Figure 4 is a southern blot of normal and tumor DNA from

patients CL, MG and ZO, digested with EcoRI and probed with a 440 bp

5' PstI fragment of pCGa (Boothby et al., 1981) which identifies two

alleles of 10 Kb (Al) and 5.0 Kb (A2) in size. All three patients

were constitutionally heterozygous at the CGA locus (6ql2-21), with

ZO showing loss of the 10 Kb allele (Al) in the tumor. The faint

signal observed in the autoradiogram is most likely due to normal

DNA contaminating the tumor sample.

L0H at the D6S37 locus (6ql3-21) is represented in figure 5.

EcoRI digested DNA from patient tumor and blood was southern blotted

32

EcoRI

Figure 3. Allele loss with "reduction to hemizygosity". Southern blots of tumor (T) and normal (N) DNA (5/ig/lane) from patient KY. Panel A: EcoRI digests probed with top: c-MYB (pHM2.6) shows constitutional homozygosity of KY at this locus and allele loss or "reduction to hemizygosity" in his tumor; bottom: same blot reprobed with D6S3 for control. Panel B: TaqI digests probed with top: S0D2 (phMnS0D4) showing reduction to hemi zygosity in the tumor of KY; bottom: reprobed with H-RAS (pEJ988) for loading control.

Table 3. Densitometry of cases with reduction to

DNA samples hemizvaositv.

33 from melanoma

6q marker control 6q/control

GZ Tumor c-MYB 0.75

D6S40 0.66 1.14

Normal-. 1.53 0.65 2.4

HY Tumor c-MYB 0.96

D6S3 0.80 1.2

Normal 1.39 0.67 2.07

KY Tumor c-MYB 0.69

D6S3 1.18 0.58

Normal 1.64 1.61 1.02

KY Tumor SQD2 0.76

H-RAS 1.0 0.76

Normal 1.21 1.08 1.12

T/N

0.48

0.58

0.57

0.68

To determine if the reduction in the intensity of the hybridization signal in the tumor of these constitutionaly homozygous patients was due to allelic loss, scanning densitometry was performed.The data for the 6q marker was normalized to other probes.T/N ratios reflect the loss of approximately 50% of the signal intensity or loss of one allele in these tumors.

34

CGA(pCG alpha) EcoRI

Figure 4. Southern blot of tumor (T) and normal (N) DNA (5/jg/lane) from three melanoma patients, digested with EcoRI and probed with the chromosome 6-specific probe CGA (pCGalpha). LOH or loss of the 10 Kb allele (Al) can be observed in the tumor of ZO, while patients CL and MG maintained heterozygosity in their tumor.

35

SH N N

ZO N

A1-A2-A3

3.6 3.4

D6S37 (p JCZ30) EcoRI

Figure 5. Southern blot of tumor (T) and normal (N) DNA (5/*g/lane) from three melanoma patients, digested with EcoRI and probed with the VNTR probe D6S37 (pJCZ30). Three different allele fragments approximately 3.0-4.0 Kb in size are identified with this probe. LOH at this locus is observed in the tumor of patients SH and ZO, while patient MN retained heterozygosity in his tumor.

and hybridized to the polymorphic VNTR probe pJCZ30 (Nakamura et

al., 1988). Three different allele fragments 3.0-4.0 Kb in size are

identified by this probe, due to the nature of the polymorphism.

From this blot, two different constitutional genotypes are

identified in these three patients: A2/A3 for patients SH and Z0,

and A1/A3 for patient MN. Loss of the larger (A2) allele is

apparent in the tumor of SH, whereas the shorter allele (A3) is

deleted from the tumor of Z0; no LOH was observed for tumor MN.

The DNA probe used to analyze the S0D2 locus (6q21) for LOH was

an 800 bp insert from phMnS0D4, which encodes the 222 amino acid

MnSOD precursor, 91 bp of 5' and 21 bp of 3' untranslated regions

(Xiang, et al., 1987). A representative southern blot of TaqI

digested DNA from six melanoma patients (AN, CO, CL, CR, GZ and KY)

probed with this insert is shown in figure 6. The probe detects two

constant bands of 3.8 Kb and 3.2 Kb, and two different two allele

systems: A1-A2 and B1-B2. For simplicity, only the A1-A2 system

will be considered here. Allele loss at this locus was detected in

two of the patients of figure 6. Patient CO lost the 2.0 Kb (A2)

allele in 50% of the cells from his tumor specimen; this finding was

confirmed by densitometric analysis using the hybridization signal

for the lip probe pEJ988 as a control for DNA loading (not shown).

The second case with allele loss identified in figure 6 is tumor KY

which suffered a reduction to hemizygosity for the 2.0 Kb (A2)

allele; this case was also confirmed by densitometry and has

previously been discussed.

37

AN co T N T N T N T

M KX. N T N T N

V

A1 A2

«0P

i

W)> -2 .3 2.0

B1

B2

m- m

S0D2(phMnS0D4) TaqI

1 .5

1 2

Figure 6. Southern blot of tumor (T) and normal (N) DNA (5/*g/lane) from six melanoma patients: AN, CO, CL, CR, GZ and KY. The DNA was digested with TaqI and probed with the 6q probe S0D2 (phMnS0D4). The tumor of patient CO lost the 2.0 Kb allele (A2) in 50% of the cells (densitometry not shown); the tumor of patient KY lost one of his constitutional 2.0 Kb (A2) alleles.

38

Figure 7 is an autoradiogram of a representative southern blot

showing LOH at the ESR locus (6q24-27). Five micrograms of tumor

and blood DNA from patients AN, CO, CL, CR, GZ and KY were digested

with PvuII, southern blotted and probed with pOR3 which is a 1.3 Kb

insert of the human estrogen receptor c-DNA (Castagnoli et al.,

1987). This probe identifies five invariant bands of 13 Kb, 5 Kb,

3.3 Kb, 2.8 Kb and 1.0 Kb, and two polymorphic alleles Al and A2 of

size 1.5 Kb and 0.7 Kb respectively. Loss of the 1.5 Kb (Al) allele

can be observed in the tumor of patient KY.

Lastly, loss of 6p alleles was investigated in the tumor DNA of

ten patients. The paired tumor and blood DNA samples were digested

with BamHI, blotted and hybridized to the 6p probe pHHH157 (Hoff et

al., 1988) from the D6S29 marker locus (6p). Two bands

corresponding to polymorphic allelic fragments of 13 Kb (Al) and 5.5

Kb (A2) are identified by this probe and can be observed in the

autoradiogram of figure 8. No allele losses are observed in any of

the tumors from figure 8 or in the other five tumors analyzed at

this 6p locus (not shown). Analysis of the remaining ten melanomas

was not possible due to insufficient genetic material from normal

blood.

39

CO CL CR GZ KY

i I » T N T N T N T N T N

3* #

V;y£

## m ̂ w • i§ *•

' i i i 5 8rt' 1

'fr, *•-*

#* fc#

«• -1.5

-mm iyb: 4* qpHpP- .»S*-' '

m -0.7

ESR(pOR3)

Pvun

Figure 7. Southern blot of tumor (T) and normal (N) DNA (5/ig/lane) from six melanoma patients, digested with PvuII and probed with the 6q probe ESR (pOR3). Patients AN, CR and GZ are homozygous for the 0.7 Kb (A2) allele; CO and CL are constitutionally heterozygous (A1/A2) and retained heterozygosity in their tumor; the tumor of patient KY, however, shows LOH at this locus with loss of the 1.5 Kb (Al) allele.

40

D6S29 (pHHH157)

Bam HI

Figure 8. Southern blot of tumor (T) and normal (N) DNA (5/ig/lane) from melanoma patients KT, PS, RR, SO and ZO. The DNA was digested with BamHI and probed with the 6p probe D6S29 (pHHH157). Patient KT is constitutionally homozygous for the 13 Kb (Al) allele, while PS and ZO are homozygous for the 5.5 Kb (A2) allele. Patients RR and SO are constitutionally heterozygous and their tumor retained heterozygosity at this 6p locus.

41

Normal and Tumor Genotypes of Twenty Melanoma Patients

at Six Loci on Chromosome 6

From the RFLP data, it was possible to determine the normal

(constitutional) and tumor genotypes of each patient at each

chromosome 6 loci analyzed. This information is shown in table 4.

A total of 53 informative phenotypes were identified on 6q; 49 were

informative by heterozygosity and 4 were informative by

densitometry. Comparison of the normal and tumor genotypes of each

autologous set of DNA's revealed that 17 evidenced LOH and 4

underwent reduction to hemizygosity in their tumor DNA (enclosed in

boxes). This translates into a 40% (21/53) frequency of 6q allele

loss in this group of melanoma patients. In contrast, no LOH was

identified at the 6p locus defined by the marker D6S29.

LOH at Six Loci on Chromosome 6

An ideogram of chromosome 6, showing the patterns of allele loss

among the 20 melanoma patients was constructed and is presented in

figure 9. The physical location of each marker locus is represented

by a solid bar. It can be observed that for each 6q locus analyzed,

at least 3 informative patients had loss of allelic sequences, while

no losses were observed at the 6p locus. Furthermore, from the 17

cases with more than one informative locus on 6q, five (HY, HN, KY,

RZ, and ZO) had allele losses that were consistent with the loss of

either an entire copy of chromosome 6, or of the entire long arm.

For instance, patient KY was constitutionally heterozygous

(informative by heterozygosity) at the 6p locus, and maintained

42

Table 4. List of Constitutional and Tumor Genotypes of Twenty Melanoma Patients at Six Loci on Chromosome 6.

Locus CGA D6S37 S0D2

Prob*

c-MYB ESR 08S29

pcGa1pha pJCZ30 pHMnS0D4

Localization

pHM2.8 pOR3 pHHHlS7

8ql2-21 0ql3-21 6q21 6q22-23 6q24-27 6p

Restriction •nzvm*

Patient EcoRI EcoRI TaqI EcoRI PvuII BamHI

AN - N AN - T

CO - N

CO - T

CL - N

CL - T

CR - N

CR - T

GZ - N

GZ - T

HY - N HY - T

HN - N

HN - T

KY - N

KY - T

KT - N KT - T

MN - N

MN - T

MG - N

MG - T

PS - N

PS - T

RR - N

RR - T

RZ - N

RZ - T SN - N

SN - T

SH - N

SH - T

SO - N

SO - T

VM - N

VM - T

VA - N

VA - T

Z0 - N

Z0 - T

A1/A2 A1/A2 A1/A2 A1/A2 A1/A2

A1/A2

A1/A2

Al

A1/A2

A1/A2

A1/A2

A1/A2

A1/A2 A1/A2

A1/A2

A1/A2

Al A1

A1/A2

A1

A1/A2

A1/A2

A1

A1

A1/A2

A1/A2

A1/A2

A2 A2 A2 A2 A3 A3

A2/A3

A2/A3

A1/A3

A1/A3 A2/A3

A3

A1/A3

A3 Al A1

A1/A3

A1/A3

A2/A3

A2/A3

A2/A3 A2/A3

Al

Al

A2

A2

A2/A3 A2/A3

A2/A3

A3 A2

A2

A2

A2

A2

Al Al

A1/A2

AJ A1/A2

A1/A2

A1/A2

A1/A2

A1/A2

A1/A2

A1/A2

Al A2 |

lA2 1 A2

A2

A2

A2

A1/A2

A1/A2

A1/A2

A1/A2

A1/A2 A1/A2

Al Al

Al

Al

A1/A2 A1/A2 A1/A2

A2/A3

A2 A1/A2

A2

A1/A2

IA2

A1/A2 A1/A2 A2 A2

A1/A2

A1/A2

A2

A2 |M—1

-! ,A1 I Al i

A1/A2

Al

;ai ; I_A1 _ J A2 A2

Al

Al

A2

A2 A1/A2

A1/A2

A2

A2.. A1/A2

A2 A2

A2

A2

A2

A1/A2 A1/A2

A2

A2

A2 A2

A2 A2

A1/A2

Al

A1/A2

A1/A2 A2

A2

A2 A2

A1/A2 A2

A1/A2

A1/A2

A2

A2

A1/A2

Al

A1/A2

A2

A1/A2

Al

A2

A2 A1/A2

A1/A2

A1/A2

A1/A2

Al

Al

Al Al Al Al

Al

Al

A1/A2

A1/A2

A1/A2

A1/A2

Al

Al

A2

A2

A1/A2

A1/A2

A1/A2

A1/A2

A2

A2

LOSS OF CONSTITUTIONAL HETEROZYGOSITY AT SIX LOCI ON CHROMOSOME 6

D6S29

6

— CGA

•D6S37

-SOD2

-c-MYB

—ESR

AN CO CL CR GZ HY HN KY KT MNMG PS RR RZ SN SH SO VM VA ZO

ooo@ •o o® • o

= no information • = informative - no loss O = non-informative # = informative - loss

• • 9 • © • • • o • ® o # • O o o 9 9 • • o 9 • • o o • • o o o • O • # A igffl 9 • • o o JtBb ® o o # # • IP o O • • • • o o o @ o • o o • o o • O • 9 O O • • o • • o # @ o

Figure 9. Schematic representation of somatic allele loss at six loci on chromosome 6 in twenty metastatic melanoma tumors. The chromosomal localization of each marker locus is indicated with vertical solid lines to the right of the idiogram. Gross chromosomal deletions are observed in tumors from patients HY, HN, KY, RZ, and ZO. Terminal deletions of 6q are observed in patients CO, GZ, KT, and RR.

heterozygosity in the tumor DNA, indicating loss of only the long

arm of 6. Cytogenetic analysis of patient KY's melanoma showed that

this tumor had indeed lost the 6q arm

(-6q) (data not shown). When the tumor from patient HN was

karyotyped, deletion of one entire copy of 6 was identified (-6)

(not shown). Chromosome analysis of tumors from the other three

patients HY, RZ, and ZO was not possible, due to poor chromosome

morphology.

Terminal deletions of 6q were identified in four cases. The

tumor DNA of patient CO retained heterozygosity at the CGA locus,

but showed LOH at the S0D2 and ESR loci and was uninformative

(homozygous) for D6S37 and c-MYB loci, consistent with a gross

chromosomal deletion involving at least half of the long arm of 6,

with a breakpoint distal to CGA. Patients GZ and RR, on the other

hand, had a much smaller terminal deletion of 6q, possibly involving

regions q22-ter and q24-ter respectively. The size of the deletion

in the tumor of KT however, could not be assessed, as it had a

homozygous (uninformative) genotype for three 6q loci separating the

informative locus with no loss (CGA) and the informative locus with

loss (ESR).

RFLP Analysis for Other Chromosome Loci

For the purpose of determining the specificity of allele losses

at chromosome 6q loci in malignant melanoma, four other highly

informative VNTR probes localized to chromosomes lp, lip, 16q and

17p were used to probe Taq I digested normal and tumor DNA for LOH.

Figure 10 is a representative autoradiogram of a southern blot

45

SH AN CO £L T N T N T N T N

' ;v.V"V. ••

A1

A2

A3

4 . 5

3 . 4

2 . 6

H-RAS(pEJ988) TaqI

r,

Figure 10. Representative Southern blot of tumor (T) and normal (N) DNA (5/*g/lane) from four melanoma patients: SH, AN, CO, and CL. The DNA was digested with TaqI and probed with the VNTR probe pEJ988 for the H-RAS locus on chromosome lip. Three possible allelic fragments are identified: 4.5 Kb (Al), 3.4 Kb (A2), and 2.6 Kb (A3). Patient SH shows LOH at this locus with loss of the 3.4 Kb (A2) allele. Patients AN and CO retained heterozygosity in their tumor, and patient CL was homozygous for the 2.6 Kb (A3) allele.

hybridized with the probe pEJ988 from the H-RAS marker locus on

chromosome lip. Three possible alleles are apparent in the four

cases shown: 4.5 Kb (Al), 3.4 Kb (A2) and 2.6 Kb (A3). LOH for the

A2 allele occurred in the tumor of SH. This case was the only one

with allele loss at the lip locus out of 10 informative heterozygous

cases (10%) in a total of 17 cases analyzed. The 17p locus (D17S30)

also had just one case of LOH out of 7 informative cases (14%) in a

total of 17 cases studied. No LOH cases were observed at the lp

(D1S57) or the 16q (D16S7) loci (table 6).

Results Summary

The LOH data for 6q is summarized in table 5. A total of 95

loci were analyzed for loss of somatic allelic sequences. Twenty

one loci with loss were identified out of 53 informative loci,

representing a 40% frequency of loss. It also can be observed from

this table that the frequency of allele loss from 6q loci increases

towards the terminal region of the long arm, with region 6q22-27

bearing the highest frequency (60%) of allele losses. This region

is defined by the markers c-MYB (q22-23) and ESR (q24-27).

In contrast to the frequent loss of 6q sequences, no

significantly frequent allele losses were detected when other

chromosome loci were analyzed. Table 6 summarizes these results,

and indicates that out of 65 total loci analyzed, 39 were

informative with only 2 cases of LOH identified, representing an

overall frequency of somatic loss of only 5% for loci on chromosomes

47

Table 5. Frequency of LOH for Five Loci on Chromosome 6q in Malignant Melanoma.

Marker Locus

Chromosome Region

Number of loci analyzed

Number of informative loci

Number of loci with LOH

LOH percent of informatives

CGA 6q12-21 1 5 1 3 3 23 %

D6S37 6q13-21 1 9 1 0 4 40 %

SOD2 6q21 1 7 1 2 4 33 %

C-MYB 6q22-23 20 1 0 6 60 %

ESR 6q24-27 1 4 8 4 50 %

Totals 95 53 21 40 %

48

Table 6. Frequency of LOH for Loci on Four Other Chromosomes in Malignant Melanoma.

Probe Chromosome Reqion

Number of loci analvzed

Number of informative loci

Number of loci with LOH

LOH percent of

informatives

pYNZ2 1P 1 7 12 0 0 %

pEJ988 11p 1 7 1 0 1 10 %

79-2-23 16q 1 4 1 0 0 0%

pYNZ22.1 17p 1 7 7 1 14 %

Totals 6 5 3 9 2 5%

49

lp, lip, 16q and 17p.

DISCUSSION

As mentioned previously, the frequent loss of somatic alleles

may be indicative of somatic mutations which are important in the

establishment and/or progression of a given malignancy. These

allele losses result in a reduction to homozygosity (or

hemizygosity) at the targeted loci in the tumor DNA of cancer

patients. Molecular genetic analysis of restriction fragment length

polymorphisms (RFLP's), identifying loss of constitutional

heterozygosity or reduction to homozygosity/hemizygosity have led to

the chromosomal localization of putative tumor suppressor genes,

whose normal function is thought to be as negative regulators of

cell growth.

In the present study, loss of allelic sequences from chromosome

6q in a group of 20 malignant melanoma patients has been identified

by RFLP analysis. The polymorphic DNA markers CGA(6ql2-21),

D6S37(6ql3-21), S0D2(6q21), c-MYB(6q22-23), and ESR(6q24-27)

spanning the long arm of 6, and the polymorphic marker D6S29

localized to 6p were used to detect loss of constitutional

heterozygosity at these loci. The results indicate that loss of

alleles occurred in 21 out of 53 informative cases, representing a

total frequency of loss of 40% in this group of melanoma patients.

Furthermore, it was observed that the frequency of loss at each 6q

locus tends to increase towards distal 6q, with the highest number

of losses mapping to chromosome region 6q22-27, defined by the

markers c-MYB(6q22-23) and ESR(6q24-27). No allele losses were

observed at the 6p locus in 4 informative cases.

In order to confirm the specificity of loss of 6q allelic

sequences, other chromosome loci were also analyzed by RFLP. LOH

for loci on chromosomes lp, lip, 16q, and 17p was detected in only 2

of 39 informative cases, with a frequency of loss of 5% for these

chromosome loci. Here, it is important to consider that the DNA

probes pYNZ2 (lp), pEJ988 (lip), 79-2-23 (16q) and pYNZ22.1 (17p)

were all highly informative polymorphic VNTR probes, identifying

multiple alleles, and therefore, a large number of heterozygous

individuals. Yet, LOH was observed only in 1/10 and 1/7 informative

cases at the lip and 17p loci respectively.

The molecular mechanisms by which chromosome 6q sequences are

lost in a large subset of melanoma tumors appear to involve gross

chromosomal deletions such as loss of an entire copy of 6 (-6), loss

of the long arm of 6 (-6q), or terminal deletions of 6q (6q-). An

ideogram of 6, illustrating the loss of 6q alleles from tumor DNA

(figure 9), indicates that from the group of 20 melanoma patients

studied here, 5 patients (25%) had allele losses throughout 6q,

consistent with the loss of one homologue of 6 or of the entire long

arm, while 4 patients (20%) had terminal deletions of 6q. This data

may indicate that mitotic chromosomal nondisjunction and terminal

deletions of the long arm are the most frequent mechanisms for the

loss of somatic sequences from chromosome 6 in melanoma tumors.

Interstitial deletions of 6q were not identified in this group of

51

patients. However, it is important to consider that these may very

well occur in a subset of melanomas, but their identification

requires the use of a larger number of polymorphic probes localized

close to the deleted target gene. Isolation of new polymorphic DNA

sequences from chromosome 6 that can be mapped to a specific band

and linked to other marker loci, is currently underway. These

probes will help provide a means to identify the putative tumor

suppressor gene on 6q.

The RFLP data obtained in this study is consistent with and

confirms multiple previous cytogenetic findings involving nonrandom

loss of 6q in malignant melanoma (Becher et al., 1983; Trent et al.,

1983; Pathak et al., 1983; Trent et al., 1989). In addition, this

study represents the first confirmation by molecular genetic means

of chromosome 6-specific somatic mutations in malignant melanoma.

Previous molecular studies by Dracopoli et al (1985) reported

frequent but nonspecific loss of polymorphic restriction fragments

in cell lines established from cutaneous malignant melanomas. Their

study analyzed loci from chromosomes 1,3,5,7,10,13,15,17,18 and 20,

reporting a frequency of allele loss ranging from 8% at the

chromosome 18 locus D18S1, to 67% at the D10S1 locus. However,

chromosome 6q marker loci were not analyzed in their study. More

recent reports by Nordenskjold et al (1990) that included a group of

11 melanoma metastasis from different patients, also indicated a

random pattern of allele loss when one marker locus from each human

chromosome were screened for LOH. The DNA marker for chromosome 6

(D6S10) utilized by these investigators has been localized to the

short arm of 6 (6p), which has not been frequently implicated in

chromosomal deletions by cytogenetics; on the contrary, 6p is

frequently found duplicated as iso(6p), or translocated [t(6p;?)]

(Pathak, et al., 1983; Becher, et al., 1983). Furthermore, Pathak

et al (1983) suggest that the i(6p) and t(6p;?) are the result of

complete loss of 6q, while terminal deletions of the long arm of 6

give the 6q- or t(6q-;?) (Trent et al, 1989).

The search for a melanoma gene that may behave as a recessive

cancer gene like Rb in retinoblastoma, has led some investigators to

consider the locus for cutaneous melanoma-dysplastic nevus (CMM-DS)

as a possible candidate. The reasons for this are several fold: 1)

approximately 8-12 percent of cutaneous melanomas are familial and

occur in association with a precursor lession or dysplastic nevi

(Greene & Fraumeni, 1979); 2) hereditary cutaneous malignant

melanoma and dysplastic nevi exhibit autosomal dominant mode of

inheritance with high penetrance (Greene, et al., 1983; Bale, et

al., 1986); 3) chromosome lp is frequently deleted in melanoma

tumors (Balaban et al., 1984); 4)somatic cell hybrid studies (Stoler

& Bouck, 1985) of normal human fibroblasts fused to carcinogen-

transformed baby hamster kidney cells (BHK), showed that all

transformation-suppressed hybrids retained human chromosome 1 and

that loss of this chromosome resulted in re-expression of the

tranformed phenotype; lastly, 5) the CMM-DS locus has been mapped by

multipoint linkage analysis to chromosome band lp36, between the

marker D1S47 and the pronatrodilatin gene (PND) (Bale et al., 1989).

However, recent experimental evidence indicates that loss of

53

sequences from chromosome lp may not represent a germinal event in a

pathway of melanoma tumorigenesis. Specifically, molecular genetic

analysis by RFLP's was performed to detect loss of alleles from 24

loci on chromosome lp in melanoma tumors, melanoma cell lines,

multiple metastasis from a single patient, and from a family with

familial cutaneous melanoma-dysplastic nevus (Dracopoli et al.,

1989). The results revealed that the loss of alleles from lp is a

late event in malignant melanoma, possibly associated with tumor

progression rather than with establishment of malignancy. In

addition, cytogenetic data obtained by Herlyn et al (1985) on the

chromosome changes of melanocytes at various stages of tumor

progression, support the idea that structural abnormalities of

chromosome 1 occur late in tumor development and are associated with

metastatic potential of the cells. Of further interest, the present

study has tested 17/20 metastatic melanoma tumors for LOH at the lp

locus D1S57 which maps to chromosome band lp36.1, and found no loss

of polymorphic alleles in 12 informative cases, adding to the

evidence against involvement of lp allele losses in melanoma tumor

establishment.

LOH for distal lp loci has also been observed in neuroblastoma

and multiple endocrine neoplasia (MEN2) which like melanoma, are

embryogenically derived from neural crest cells. Fong et al. (1989)

detected LOH for lp loci in 28% of neuroblastomas in their study,

most of them showing terminal deletions; furthermore, the frequency

of LOH was found to be associated with N-myc gene amplification,

which in turn has been implicated in advanced tumor stage (Brodeur

et al., 1984). In respect to MEN2, Mathew et al. (1987) reported

that LOH for lp occurred in 50% of tumors, but analysis of the

parental origin of the deleted allele in two families with

hereditary MEN2 showed that in one case it was derived from the

affected parent, indicating that the deleted lp does not reflect the

site of the original mutation, and therefore does not fit the "two

hit" model of carcinogenesis. Recently, the MEN2A gene has been

mapped by linkage studies to chromosome 10, close to the retinol-

binding protein gene (Simpson et al., 1987). Thus, it appears that

the frequent deletion of lp in melanoma constitutes a genetic

alteration associated with tumor progression as opposed to tumor

initiation.

Whether the deletion of 6q constitutes a key mutational event

that leads to loss or inactivation of a putative melanoma tumor

suppressor gene, is the subject of further investigation. The

complexity of this project is due in part to the presence of

chromosomal alterations in addition to those on 6q, so that

identification of the "initial" genetic alterations becomes a very

difficult task. In addition, tumor progression is known to be

associated with an accumulation of genetic changes as a result of

increased genetic instability that may result in loss or

Inactivation of multiple tumor suppressors genes. Thus, tumor

suppressor genes involved in initiation of certain types of tumors

may be involved in progression of others. Alternatively, it is

possible that certain tumor suppressor genes may have a common role

in different types of cancers. For example, the Rb gene (13ql4)

55

which serves as a prototype for tumor suppressor genes, has been

shown to be the target for "two hit" mutations that lead to the

establishment of retinoblastoma, but allele losses at 13ql4 have

also been observed in breast carcinoma (Ali et al., 1987; Lee et

al., 1988), ductal breast carcinoma (Lundberg et al., 1987),

osteosarcoma (Hansen, et al., 1985; Friend, et al., 1986), and small

cell lung carcinoma (Yokota et al., 1987; Harbour et al., 1988),

suggesting a general role for Rb in different tissues.

The results of this study indicate that the loss of somatic

alleles from chromosome 6q in human malignant melanoma is a frequent

and nonrandom event. Together with cytogenetic and biological data,

these studies strongly suggest the existence of a putative tumor

suppressor gene localized to 6q. It is tempting to speculate that

while loss of lp material seems to be a characteristic of advanced

melanomas, loss of 6q may represent an earlier genetic event of

significant importance in melanoma tumorigenesis.

Finally, LOH studies have proven very effective in mapping

putative tumor suppressor genes to a specific chromosome region. I

believe this study will provide the basis for future research that

will lead to the identification and cloning of a tumor suppressor

gene (or genes) from chromosome 6q, implicated in the development of

human malignant melanoma.

REFERENCES

56

Ali, I.U., Lidereau, R., Theillet, C.,& Callakan, R. (1987) Science 238, 185-188.

Baker, S.J., Fearon, E.R., Nigro, J.M., Hamilton, S.R., Preisinger, A.C., Jessup, J.M., Tuinen, P., Ledbetter, D.H., Barker, D.F., Nakamura, Y., White, R.& Vogelstein, B. (1989) Science 244, 217-221.

Balaban, G., Herlyn, M.t Guerry, D., et al. (1984) Cancer Genet.Cytogenet. 11, 429-439.

Bale, S.J., Chakravarti, A., Green, M.H. (1986) Am.J.Hum.Genet. 38, 188-196.

Bale, S.J., Dracopoli, N.C., Tucker, M.A., Clark, W.H., Fraser, M.C., Stanger, B.Z., Green, P., Donis-Keller, H., Housman, D.E. & Green, M. (1989) N.Engl.J.Med. 320, 1367-1372.

Becher, R., Gibas, Z.& Sandberg, A.A. (1983) Cancer Genet.Cytogenet. 9, 173-175.

Benedict, W.F., Srivatsan, E.S., Mark, C., et al. (1987) Cancer Res. 47, 4189.

Bishop, J.M. (1983) Ann. Rev. Biochem. 52, 301.

Bishop, J.M. (1987) Science 235, 305.

Brauch, H., Johnson, B., Hovis, J., et al. (1987) N.Engl.J.Med. 317, 1109-1113.

Boothby, M., Ruddon, R.W., Anderson, C., McWilliams, D. and Boime, I. (1981) J.Biol.Chem. 256,(10) 5121-5127.

Botstein, D., White,R., Skolnick, M., Davis, R. (1980) Am. J. Hum. Genet. 32, 314-331.

Brodeur, G.M., Seeger, R.C., Scwab, M., Varmus, H.E., Bishop, J.M. (1984) Science 224, 1121-1124.

Brodeur, G.M., Fong, C.T., Michon, M., et al. (1988) Adv.Neuroblastoma Res. 2, 3-15.

Brodeur, G.M., Fong, C.T. (1989) Cancer Genet.Cytogenet. 41, 153-174.

Call, K.M., Glaser, T., Ito, C.Y., Buckler, A.J., Pelletier, J., et al. (1990) Cell 60, 509-520.

57

Castagnoli, A., Maestri, I., Bernardi, F. and Del Senno, L. (1987) Nucl.Acid.Res. 15(2), 866.

Cavenee, W.K., Dryja, T.P., Phillips, R.A., et al.(1983) Nature 305, 779-784.

Cavenee, W.K., Hansen, M.F., Nordenskjold, M.,et al. (1985) Science 228, 501.

Cori, S. (1986) Adv. Cancer Res. 47, 189.

Dao, D.D., Schroeder, W.T., Chao, L-Y., et al. (1987) Am.J.Hum.Genet. 41, 202.

De Caprio, J.A., Ludlow, J.W., Figge, J., Shew, J.-Y., Huan, C.M., Lee, W.H., Marsilio, E., Paucha, E., & Livingston, D.M. (1988) Cell 54, 275-283.

Devilee, P., van den Broek, M., Kuipers-Dijkshoorn, N., et al. (1989) Genomics 5, 554-560.

Dozier, C., Walbaum, S., Leprince, D., & Stehelin, D. (1986) Nucl.Acid.Res. 14(4), 1928.

Dracopoli, N.C., Houghton, A.N.& Old, L.J. (1985) Proc.Natl.Acad.Sci.USA. 82, 1470-1474.

Dracopoli, N.C., Harnett, P., Bale, S.J., Stanger, B.Z., Tucker, M.A., Housman, D.E. & Kefford, R.F. (1989) Proc.Natl .Acad.Sci.USA. 86, 4614-4618.

Dryja, T.P., Cavenee, W., White, R., Rapaport, J.M., Petersen, R., Albert, D.M., & Bruns, G.A.P. (1984) N.Engl.J.Med. 310, 550-553.

Dumanski, J.P., Carlbom, E., Collins, V.P. & Nordenskjold, M. (1987) Proc.Natl.Acad.Sci.USA. 84, 9275-9279.

Dyson, N., Howley, P.M., Munger, K., & Harlow, E. (1989) Science 243, 934-937.

Evans, R.D., Kopf, A.W., Lew, R.A., Rigel, D.S., Bart, B.R., Friedman, R.J.& Rivers, J.K. (1988) J. Dermatol. Surg. Oncol. 14, 393-408.

Fields, S., & Jang, S.K. (1990) Science 249, 1046-1049.

Fearon, E.R., Hamilton, S.R.& Vogelstein, B. (1987) Science 238, 193-196.

Fearon, E.R., Cho, K.R., Nigro, J.M., Kern, S.E., Simon, S.R., et al. (1990) Science 247, 49-56.

58

Feinberg, A.P. & Vogelstein, B. (1983) Anal. Biochem. 132, 6-13.

Fong, C., Dracopoli, N.C., White, P.S., Merrill, P.T., Griffith, R.C., Housman, D.E., Brodeur, G.M. (1989) ProctNatl.Acad.Sci.USA. 86, 3753-3757.

Friend, S.H., Bernards, R., Rogelj, S., Weinberg, R.A., Rapaport, J.M., Alberts, D.M. & Dryja, T.P. (1986) Nature 323, 643-646.

Friend, S.H. et al. (1987) Proc.Natl.Acad.Sci.USA. 84, 9059-9063.

Friend, S.H., Dryja, T.P.& Weinberg, R.A. (1988) N.Engl.J.Med. 318, 618-622.

Fujimoto, N., Fults, D.W., Thomas, G.A., Nakamura, Y. et al. (1989) Genomics 4, 210-214.

Fults, D., Tippets, R.H., Thomas, G.A., Nakamura, Y.& White, R. (1989) Cancer Res. 49, 6572-6577.

Fung, Y-KT, Murphree, A.L., T'Ang, A. et al. (1987) Science 236, 1657.

Geiser, A.G., Stanbridge, E.J. (1989) Crit.Rev.in Oncogenesis 1, 261-276.

Godbout, R., Dryja, T.P., Squire, J., Gallie, B.L. & Phillips, R.A. (1983) Nature 304, 451-453.

Greene, M.H., Fraumeni, J.F.Jr.. In: Human Malignant Melanoma. New York: Grune & Stratton 1979: 139-166.

Greene, M., Goldin, L., Clark, W., Lovrien, E., et al. (1983) Proc.Natl.Acad.Sci.USA. 80, 6071-6075.

Hansen, M.F., Koufos, A., Gallie, B.L., Phyllips, R.A., Fodstad, 0., Brooger, A., Gedde-Dahl, T. & Cavenee, W.K. (1985) Proc.Natl.Acad.Sci.USA. 82, 6216.

Hansen, M.F., Cavenee, W.K. (1987) Cancer Res.47, 5518-5527.

Harbour, J.W., Lai, S.L., Whang-Peng, J., et al. (1988) Science 241, 353-357.

Herlyn, M., Thurin, J., Balaban, G., Bennicelli, J.L., Herlyn, D., Elder, D.E., Bondi, E., Guerry, D., Nowell, P., Clark, W.H. & Koprowski, H. (1985) Cancer Res. 45, 5670-5676.

Hoff, M., Nakamura, Y., Holm, T., Gillilan, S., 0'Connell, P., Leppert, M., Lathrop, G.M., Lalouel, J.-M. and White, R. (1988) Nucl.Acid.Res. 16(11), 5217.

59

Hayward, W.S., Neel, B.G.& Astrin, S.M. (1981) Nature 290, 475.

Human Gene Mapping 9 (1987) Cytogenet. Cell Genet. 46, (Nos.1-4).

Human Gene Mapping 9.5 (1988) Cytogenet. Cell Genet. 49, (Nos 1-3).

James, C.D., Carlbom, E., Nordenskjold, M.( Collins, V.P. & Cavenee, W.K. (1989) Proc.Natl.Acad.Sci.USA. 86, 2858-2862.

Klein, G. (1985) Nature 315, 190-195

Klein, G. (1987) Science 238, 1539-1544.

Klein, G. (1988) Acta Oncologica 27, 427-437.

Knudson, A.G. (1985) Jr.Cancer Res. 45, 1437-1443.

Knudson, A.G.(1971) Proc.Natl.Acad.Sci.USA. 68, 820-823.

Koufos, A., Hansen, M.F., Copeland, N.G., et al. (1985) Nature 316, 330.

Land, H., Parada, L.F., Weinberg, R.A. (1983) Science 222, 771-778.

Lane, D.P., & Crawford, L.V. (1979) Nature 278, 261-263.

Lee, W.H., Bookstein, R., Hong, F., Young, L.J., Shew, J.Y. & Lee, E.YHP. (1987) Science 235, 1394-1399.

Lee, W. et al. (1987) Nature 329, 642-645.

Lee, E.Y.-H.P., To, H., Shew, J.-Y., Bookstein, R., Scully, P. & Lee, W.-H. (1988) Science 241, 218.

Litt, M. and White, R. (1985) Proc.Natl.Acad.Sci.USA. 82, 6206.

Lundberg, C., Skoog, L., Cavenee, W.K., et al.(1987) Proc.Natl.Acad.Sci.USA. 84, 2372-2376.

Mackay, J., Elder, P.A., Porteous, D.J., et al. (1988) Br.J.Cancer 58, 710-714.

Mackay, J., Elder, P.A., Steel, C.M. & Forrest, A.P.M. (1988) Lancet 2, 1384-1385.

Maniatis, T., Fritsch, E.F. & Sambrook, J. (1989) Cold spring Harbor Laboratory Press.

Manne, V., Bekesi, E. & Kung, H.F. (1985) Proc. Natl. Acad. Sci. USA. 82, 376-380.

60

Mannens, M.t Devilee, P., Bliek, J., et al. (1990) Cancer Res. 50, 3279-3283.

Mannens, M., Slater, R.M., Heyting, C., et al. (1988) Human Genet. 81, 41-48.

Mathew, C.G.P., Smith, B.A., Thorpe, K., Wong, Z., Royle, N.J., Jeffreys, A.J. & Ponder, B.A.J. (1987) Nature 328, 524-526.

Nakamura, Y., Leppert, M., O'Connell, P., Wolff, R., Holm, T., Culver, M., Martin, C., Fujimoto, E., Hoff, M., Kumlin, E. & White, R. (1987) Science 235, 1616-1622.

Nakamura, Y., Cowen, J., Martin, C., Leppert, M., O'Connell, P., Lathrop, G.M., Lalouel, J.-M. and White, R. (1988) Nucl.Acid.Res. 16(10), 4743.

Naylor, S.L., Johnson, B.E., Minna, D.& Sakaguchi, A.Y. (1987) Nature (Lond.) 329, 451-454.

Newmark, P. (1987) Nature 327, 101.

Nordenskjold, M., Werelius, B. & Ringborg, U. (1990) Anticancer Res. 10, 883-886.

Okamoto, M., Sasaki, M., Sugio, K. et al. (1988) Nature 331, 273-277.

Oren, M. & Levine, A.J. (1983) Proc. Natl. Acad. Sci, USA. 80, 56.

Parmiter, A.H., Balaban, G., Herlyn, M., Clark, W.H., Nowel, P.C. (1986) Cancer Res. 46, 1526

Pathak, S., Drwinga, H.L., & Hsu, T.C. (1983) Cytogenet. Cell Genet. 36, 573-579.

Pedersen, M.I., Benett, J.W., Wang, N. (1986) Cancer Genet.Cytogenet. 20, 11.

Raycroft, L., Wu, H., & Lozano, G. (1990) Science 249, 1049-1051.

Sager, R. (1986) Cancer Res. 46, 1573.

Sandberg, A.A. In: The Chromosomes in Human Cancer and Leukemia. Elsevier Science Publishing Co. The Netherlands. 1990.

Saxon, P.J., Srivatson, E.S., Leipzig, G.V., Sameshima, J.H., & Stanbridge, E.J. (1985) Mol.Cel1.Biol. 5, 140.

Saxon, P.J., Srivatson, E.S., & Stanbridge, E.J. (1986) EMB0 J. 5, 3461.

61

Schwab, M., Ellison, J., Busch, M., Rosenau, W., Varmus, H.E. (1984) Proc. Natl. Acad. Sci. USA. 81, 4940-4944.

Seizinger, B.R., De la Monte, S., Atkins, L., Gusella, J.F.& Martuza, R.L. (1987) Proc.Natl.Acad.Sci.USA. 84, 5419-5423.

Simpson, N.E., Kidd, K.K., Goodfellow, P.J., et al. (1987) Nature 328, 528-530.

Southern, E.M. (1975) J. Mol. Biol. 98, 503-517.

Sparkes, R.S., Murphree, A.L., Ligua, R.W., Sparkes, M.C., Field, L.L., Funderburk, S.J. & Benedict, W.F. (1983) Science 219, 971.

Stanbridge, E.J., Flandermeyer, R.R., Daniels, D.W. & Nelson-Rees, W.A. (1981) Somatic Cell Mol.Genet. 7, 699.

Stanbridge, E.J. (1987) Adv. Viral Oncol. 6, 83.

Stoler, A. & Bouck, N. (1985) Proc.Natl.Acad.Sci.USA. 82, 570-574.

Suzuki, T., Yokota, J., Mugishima, H., et al. (1989) Cancer Res. 49, 1095-1098.

Tabin, C.J., Bradley, S.M., Bargman, C.I. & Weinberg, R.A. (1982) Nature 300, 143.

Thorn, M., Adami, H-O., Bergstrom, R., Ringborg, V.& Kruscmo, U.B. (1989) JNCL 81, 611-617.

Trent, J.M., Rosenfeld, S.B., Meyskens, F.L. (1983) Cancer Genet.Cytogenet. 9, 177-180.

Trent, J.M., Stanbridge, E.J., Mc Bride, H.L., Meese, E.U., Casey, G., Araujo, D.E., Witowski, C.M.& Nagle, R.B. (1990) Science 247, 568-571.

Trent, J.M., Thompson, F.H., Meyskens, F.L. (1989) Cancer Res. 49, 420.

Varmus, H.E. (1982) Science 216, 812.

Vogelstein, B., Fearon, E.R., Hamilton, S.R. et al. (1988) N.Engl.J.Med. 9, 525-532.

Vogelstein, B., Fearon, E.R., Kern, S.E., Hamilton, S.R., Preisinger, A.C., Nakamura, Y. & White, R. (1989) Science 244, 207-211.

Weinberg, R. (1989) Biochemistry 28(21) , 8263-8269.

Whyte, P., Buchkovich, K.J., Horowitz, J.M., Friend, S.H., Raybuck, M., Weinberg, R.& Harlow, E. (1988) Nature 334, 124-129.

X1ang, K., Cox, N.J., Hallewell, R.A. and Bell, G.I. (1987) Nucl.Acid.Res. 15(18), 7654.

Yokota, J., Wada, M., Shimosato, Y., et al. (1987) Proc.Natl.Acad.Sci.USA. 84, 9252-9256.

Yunis, J.J. & Ramsay, N. (1978) Am.J.Dis.Child. 132, 161-163.