Loss of Heterozygosity in Human Ovarian Cancer on Chromosome 19q

GYNECOLOGIC ONCOLOGY ARTICLE NO.

66, 36–40 (1997)

GO974709

Loss of Heterozygosity in Human Ovarian Cancer on Chromosome 19q Annette Bicher,*,1 Kristen Ault,* Alec Kimmelman,* David Gershenson,† Eddie Reed,* and Bertrand Liang‡ *Medical Ovarian Cancer Section, Division of Clinical Sciences, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892; †Department of Gynecologic Oncology, M. D. Anderson Cancer Center, Houston, Texas 77030; and ‡Department of Neurology, Medicine (Medical Oncology), and Pathology, University of Colorado Health Sciences Center, Denver, Colorado Received December 3, 1996

the chromosome which contains the second allele [3]. By comparing polymorphic loci in normal and tumor DNA from the same patient, deletions can be mapped through the identification of sites of loss of heterozygosity (LOH). Using this strategy large areas of deletion can be identified. It has been implied that LOH of 35% or greater is more likely to represent potentially causative genetic events than a secondary phenomenon associated with generalized genomic instability [4]. Loss of heterozygosity studies in ovarian carcinomas have suggested the presence of tumor suppressor genes on chromosomes 5q, 6p, 6q, 9p, 11p, 13q, 17p, 17q, 18q, 19p, 22q, and Xp [5–8]. Very few studies have evaluated chromosome 19q in ovarian cancers. The DNA repair genes ERCC1, ERCC2, and LIG1 lie on 19q13.2-13.3. Prior studies have shown that abnormalities of mRNA expression of ERCC1 and ERCC2 may be characteristic of epithelial ovarian carcinoma [9] and brain tumors [10, 11]. While changes in mRNA expression patterns appear to be common [9, 10], genomic changes are infrequent in ovarian cancer [12]. Chromosome 19q13.2–13.4 has been found to be of interest in gliomas with 35–81% of these tumors showing loss of heterozygosity in this region of [13–18]. Extensive work searching for a tumor suppressor gene on this segment of chromosome 19 has been done in brain tumors. A 425-kb area of loss has been identified in gliomas centromeric to D19S112 on 19q13.3 [3]. The segment of chromosome 19q13.2–13.4 which we chose to analyze spans 18 cM by linkage analysis [19, 20]. An integrated metric physical map recently published by Ashworth and colleagues finds this area spanning 7 Mb [21, 22]. Given the preliminary results in gliomas and our interest in DNA repair genes, we selected six microsatellite polymorphic markers in this region to evaluate evidence of LOH in sporadic ovarian cancers.

Abnormalities in the function of oncogenes and tumor suppressor genes have been associated with many human malignancies. The recognition of sites of loss of heterozygosity (LOH) has led to the identification of such genes. We previously reported that abnormalities of mRNA expression of ERCC1 and ERCC2 may be characteristic of epithelial ovarian carcinoma and brain tumors. This led to an investigation of chromosome 19q13.2-q13.4 which contains these DNA repair genes. A 7-Mb region was analyzed using six microsatellite repeats. Loss of heterozygosity has been identified in 53% (8/15) of cases at marker D19S246 which lies in a 2-Mb segment between HRC and KLK1. The genetic material both centromeric and telomeric to the region of loss was conserved. This area is telomeric to three DNA repair genes where LIG1 is 1-Mb centromeric and ERCC1 and ERCC2 are 3.5- and 4.0-Mb centromeric, respectively. These findings represent the first report of a biologically significant rate of LOH on chromosome 19q13.2q13.4 in human ovarian carcinoma. q 1997 Academic Press

INTRODUCTION

Ovarian cancer is the leading cause of mortality from gynecologic malignancies in North America, with an estimated 14,800 deaths in 1996 [1]. The high mortality rate is due to the aggressive nature of the tumor and the fact that the majority of patients are diagnosed with advanced-stage disease. Unfortunately, no significant etiologic factors have been identified which allow for early diagnosis of sporadic disease [2]. Mutations within oncogenes and tumor suppressor genes have been implicated as possible etiologic factors in the development of several malignancies. Tumor suppressor genes are commonly involved in constraining the cell cycle or cellular growth. These genes act recessively, thus the loss or inactivation of both copies of the gene remove the normal controls of cellular proliferation and can lead to tumorigenesis. The inactivation of tumor suppressor genes typically occurs as a mutation of one allele and a loss of a portion of

MATERIALS AND METHODS

Patient Samples

1

To whom reprint requests should be addressed at Medical Ovarian Cancer Section, National Cancer Institute, Bldg. 10, Rm. 12N226, Bethesda, MD 20892. Fax: (301) 496-4572. 36

0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

Gyn 4709

/

6d1c$$$421

The proposed study was approved by the Institutional Review Board at M. D. Anderson Cancer Center (Houston,

06-12-97 18:30:01

goal

AP: GYN

37

LOSS OF HETEROZYGOSITY IN OVARIAN CANCER

TABLE 1 Clinical Data

LOH Studies

Patient

Age

Histology

Tumor grade

Stage

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

16 52 74 76 53 64 61 67 73 72 65 73 64 47 86 82 63 42 49 54

Mature teratoma Mixed epithelial MMMT Papillary serous Transitional cell Papillary serous Papillary serous Endometrioid Endometrioid Papillary serous Papillary serous Papillary serous Clear cell Papillary serous Mixed epithelial Mixed epithelial Mixed epithelial Papillary serous Mixed epithelial Mixed epithelial

Benign 3 3 3 3 3 3 3 2 3 2 3 3 3 3 3 3 LMP 3 3

N/A IV IV IV IIIc IIIc IIIc IIb IIIc IIIc IV IIIc Ic IIIc Ib IV IV Ib IV IIIc

Note. MMMT, malignant mixed mullerian tumor; LMP, low malignant potential.

TX). Patients undergoing initial exploratory laparotomy with possible cytoreductive surgery to rule out ovarian carcinoma were eligible. Exclusion criteria included a prior history of cancer (any type), prior chemotherapy, or prior radiation therapy. After consents were obtained, tumor tissue and blood samples were collected from patients at the time of surgery. An area of fresh tumor as designated by the histopathologist was flash frozen at 0807C. Lymphocytes were isolated from the venipuncture samples and similarly frozen at 0807C. The remainder of the sample was evaluated by a pathologist on staff at M. D. Anderson to obtain the histopathologic diagnosis. DNA was later extracted from both the tumor and lymphocytes using standard phenol–chloroform extraction [23]. All pathology was reviewed at the M. D. Anderson Cancer Center. Eighteen of the 20 patients had epithelial ovarian carcinomas. Eight of these had papillary serous carcinomas, 2 had endometrioid carcinomas, 1 had a clear cell carcinoma, 1 had a transitional cell carcinoma (TCC), and 6 had mixed epithelial carcinomas. Fifteen of the epithelial lesions were histopathologic grade 3, 2 were histopathologic grade 2, and 1 was a papillary serous tumor of low malignant potential (LMP). In the remaining cases, 1 was found to have a malignant mixed mullerian tumor (MMMT) of the ovary and the other a mature teratoma. Of the 19 patients with malignancies, 7 (37%) had stage IV disease, 7 (37%) had stage IIIc, 1 (5%) had stage IIIb, 1 (5%) had stage IIb, 1 (5%) had stage Ic, and 2 (11%) had stage Ib. The mean patient age in the study was 61.7 (median 64) (Table 1).

AID

Gyn 4709

/

6d1c$$$422

06-12-97 18:30:01

Polymerase chain reaction (PCR) was performed on tumor and lymphocyte DNA pairs with each of the oligonucleotide primers; D19S178, APOC2, D19S112, HRC, D19S246, KLK (Research Genetics, Inc.). PCR conditions have been previously reported [18] and were further optimized for each primer. The reaction mixture consisted of 40 ng genomic DNA, 50 pmol of each primer (forward and reverse), 1.25 mM dNTPs, 1 unit AmpliTaq polymerase (Perkin-ElmerCetus, Norwalk, CT), 11 amplification buffer ({supplied} 10 mM Tris, 50 mM KCl, 1.5 mM MgCl2 , 0.1% Triton X-100, 0.01% gelatin, pH 8.3, without magnesium). The magnesium concentrations ([Mg]) varied depending on the primer and were as follows: D19S178 [Mg] 1 mM, APOC2 [Mg] 2 mM, D19S112 and HRC [Mg] of 2.5 mM, D19S246 and KLK required [Mg] 1.5 and 2.0 mM, respectively. PCR was performed on a Programmable Thermal Controller (MJ Research) in 25(l volumes. The cycling conditions also varied depending upon the primer used. D19S112, D19S246, KLK underwent 35 cycles at 947C for 30 sec, 557C for 75 sec, and 727C for 15 sec with a 6-min final elongation step at 727C. HRC required an annealing temperature of 687C with the remainder of the cycling conditions as above. D19S178 required ramping annealing temperature 25–377C for 20 cycles followed by the above conditions for 30 cycles using an annealing temperature of 557C. APOC2 also required a ramping annealing temperature of 15–377C for 10 cycles followed by the above conditions with the annealing temperature at 527C. The segments of DNA produced varied from 110 to 202 bp in length depending on the primer used. The PCR products were analyzed by electrophoresis. PCR products were denatured at 997C for 5 min and separated on 6% denaturing polyacrylamide gels (Gibco BRL, Gaithersburg, MD) at 60 W, for 2–3 hr with normal and tumor samples in adjacent wells. All gels were run with a 10- to 500-bp DNA ladder (Gibco BRL) to allow for identification of the isolated DNA segment. The gels were developed using the silver staining technique (Promega, Madison, WI). Once the developed gel dried, a permanent record was created by exposing Silver Sequence automatic processor compatible film (Promega) for 5–8 min. Absent signal in a tumor sample where signal was present in the respective normal sample was interpreted as loss of heterozygosity. The results were interpreted based on the consensus of three reviewers (A.B., K.A., B.L.). RESULTS

The region of 19q13.2 – 19q13.4 was tested using 6 microsatellite markers which span a 7-Mb segment. The order of the markers and distances between them and the DNA repair genes are shown in Fig. 1. These were determined by Ashworth et al. using pronuclear fluorescence in situ hybridization (FISH) [22]. A total of 120 loci were

goal

AP: GYN

38

BICHER ET AL.

DISCUSSION

FIG. 1. Metric physical map of chromosome 19q13.2-q13.4. Distances between markers are in megabases. Adapted from Ashworth et al. [21].

tested with 94 (78%) found to be informative. Eight tumors showed no areas of loss. However, of these, tumors 10 and 13 were uninformative in 4 of the 6 loci tested. LOH was found in 3/15 (20%) of informative tumors at D19S112, 1/16 (6%) at HRC, 8/15 (53%) at D19S246, and 3/19 (19%) at KLK. There was no evidence of LOH at D19S178 (0/16) or APOC2 (0/13). Loss of heterozygosity at D19S246 was seen in 3/8 papillary serous carcinomas, 1/2 endometrioid carcinomas, 1/1 TCC, 0/1 clear cell carcinomas, 1/6 mixed epithelial carcinomas, 1/1 MMMT, and 1/1 mature teratomas (Table 2). The mean age of patients with tumor showing LOH at D19S246 was 58.6 years (median 64). The mean age of those patients with retention of genetic material at D19S246 was 63.6 years (median 64). Six of the 8 (75%) tumors with LOH at D19S246 had grade 3 lesions, while 5/7 (71%) of those without LOH at this marker had grade 3 tumors. There was no significant difference between patient age, histopathologic subtype or tumor grade, and LOH at the D19S246 marker. None of the three stage I tumors showed LOH at any of the loci tested, although tumor 13 is difficult to interpret because it was uninformative in 4/6 loci. Figure 2 reveals a deletion map of the region. Of note is the retention of genomic material both centromeric (HRC) and telomeric (KLK) to the region of loss in 5 of the 7 cases with LOH at D19S246. Case 4 revealed only telomeric loss, while case 3 was uninformative at HRC (Fig. 3).

AID

Gyn 4709

/

6d1c$$$422

06-12-97 18:30:01

The results of this study showed loss of heterozygosity in 53% of the ovarian tumors on chromosome 19q at the polymorphic locus D19S246. This marker lies in a region spanning 2 Mb of chromosome 19q13.2–q13.4 which is flanked by HRC and KLK. The majority of the tumors showing LOH in this area were grade 3 epithelial ovarian carcinomas with one tumor being an MMMT and the other being a mature teratoma. No stage I lesions showed LOH on chromosome 19q, although the stage IIb epithelial cancer did reveal LOH at D19S246. Several studies of gliomas have revealed a potential tumor suppressor gene on chromosome 19q13.2–13.4 with LOH reported from 31 to 81% of the cases [13–18]. Yong et al. [3] have recently published a loss of heterozygosity study in gliomas using six microsatellite polymorphisms between APOC2 and HRC. Their results suggest the presence of a tumor suppressor gene at 19q13.3 telomeric to D19S219 and centromeric to D19S112 which narrows the segment to 425 kb. This region is centromeric to D19S246 and does not include ERCC1, ERCC 2, and LIG1 (DNA repair/metabolism genes). Several studies in epithelial ovarian cancer have found up to 38% LOH on chromosome 19p in the region of the insulin receptor gene (INSR) [7, 24, 25]; however, very few studies have investigated the long arm of chromosome 19. Thompson et al. [12] identified cytogenetic abnormalities on chromosome 19 in 26/237 (11%) of epithelial ovarian cancer cases on both arms of chromosome 19 using FISH analysis. In 4 patient samples and 7 cell lines they noted that 65% of chromosome 19 structural abnormalities contained 19q13.1– q13.2 sequences. Overexpression of the AKT2 putative oncogene and ERCC2 repair gene was evaluated using Southern blot analysis in the ovarian cancer cell lines and normal ovary. The expression of AKT2 transcripts was increased relative to normal ovary in OVCAR-3, CoLo-316, and SKOV-3 cell lines which have chromosome 19 abnormalities. A2780 and 367 cell lines do not have chromosome 19 abnormalities and were not found to have increased expression of AKT2. There was no increase in ERCC2 gene copies or transcripts in any of the cell lines. Their findings suggest

TABLE 2 LOH on Chromosome 19q13.2-q13.4: Summary of Informative Cases Genomic site D19S178 APOC2 D19S112 HRC D19S246 KLK

goal

AP: GYN

LOH (%) 0/16 0/13 3/15 1/16 8/15 3/19

(0) (0) (20) (6) (53) (19)

LOSS OF HETEROZYGOSITY IN OVARIAN CANCER

FIG. 2. alleles.

Deletion map of chromosome 19q13.2-q13.4. Note increased loss of heterozygosity at marker D19S246 with retention of HRC and KLK

overrepresentation of chromosomal material on chromosome 19q in a region centromeric to the area of loss identified in our study. They suggest a possible role of AKT2 or another oncogene in ovarian cancer [12]. Amfo et al. [24] detected LOH in 9/26 (35%) of sporadic ovarian cancers using RFLP analysis at D19S119. This polymorphic marker lies just telomeric to ERCC1 on chromosome 19q13.3. The deletion at 19q was primarily associated with stage IV disease [24]. Our findings were not limited to stage IV disease, although the area of loss identified herein

FIG. 3. Example of loss of heterozygosity evaluation using microsatellite markers. Normal peripheral blood lymphocyte DNA (N) and tumor DNA (T) from patients 4, 5, 11, and 20 underwent PCR with the noted primers. The products were separated on 6% polyacrylamide gels to determine allelic imbalance and demonstrate loss of heterozygosity at marker D19S246.

AID

39

Gyn 4709

/

6d1c$$$423

06-12-97 18:30:01

lies telomeric to those areas previously identified in sporadic ovarian cancer and that described in brain tumors [3]. In addition, the DNA repair genes ERCC1, ERCC2, and LIG 1 are not involved in the area of loss seen. A well-described oncogene, rRAS, is found within the region of loss [22]; however, the significance of this remains unclear. The results of this study are limited by the small patient volume. The majority of the specimens were grade 3 lesions at an advanced stage; thus, analysis of incidence of LOH relative to stage and grade is limited. In addition, the specimens were not microdissected; therefore, the results may underestimate the actual incidence of loss of heterozygosity due to contamination by normal cells which could be integrated amongst the tumor cells. This study shows loss at locus D19S246 on chromosome 19q with retention of genomic material 200 kb telomeric (KLK) and 1.8 Mb centromeric (HRC) in 53% of the ovarian cancer cases studied. Further evaluation of the 2-Mb segment is being undertaken using five additional microsatellite polymorphic markers. This may indicate the presence of a tumor suppressor gene in this region. Should a tumor suppressor gene more directly related to sporadic ovarian cancer be localized, it could increase our understanding of the molecular pathogenesis of this disease. In turn, this may lead to advancements in our diagnostic and therapeutic capabilities in ovarian cancer.

goal

AP: GYN

40

BICHER ET AL.

ACKNOWLEDGMENTS We gratefully acknowledge Adel K. El-Naggar, M.D., Ph.D., Department of Pathology, and Donna L. Warner, R.N., Department of Gynecology, of the M.D. Anderson Cancer Center, Houston, Texas, for their assistance in specimen collection and patient data. We also gratefully acknowledge Linda Ashworth, Ph.D., Lawrence Livermore National Laboratory, Livermore, California, for providing information regarding the physical map of chromosome 19.

13.

14.

15.

REFERENCES 1. Parker SL, Tong T, Bolden S, Wingo PA: Cancer statistics, 1996. CA Cancer J Clin 65:5–27, 1995

16.

2. Ozols RF, Rubin SC, Dembo AJ, Robboy SJ: Epithelial Ovarian Cancer, in Hoskins WJ, Perez CA, Young RC (eds): Principles and Practice of Gynecologic Oncology. Philadelphia, PA, Lippincott, 1992, pp 731– 782

17.

3. Yong WH, Chou D, Ueki K, Harsh GR, vonDeimling A, Gusella JF, Mohrenweiser HW, Lewis DN: Chromosome 19q deletions in human gliomas overlap telomeric to D19S219 and may target a 425 kb region centromeric to D19S112. J Neuropathol Exp Neurol 54:622–626, 1995

18.

4. Cliby W, Ritland S, Hartmann L, Dodson M, Halling KC, Keeney G, Podratz KC, Jenkins RB: Human epithelial ovarian cancer allelotype. Cancer Res 53:2393–2398, 1993

19.

5. Shelling AN, Cooke IE, Ganesan TS: The genetic analysis of ovarian cancer. Br J Cancer 72:521–527, 1995

20.

6. Osborne RJ, Leech V: Polymerase chain reaction allelotyping of human ovarian cancer. Br J Cancer 69:429–439, 1994 7. Jenkins RB, Bartelt D Jr, Stalboerger P, Persons D, Dahl RJ, Podartz K, Keeney G, Hartmann L: Cytogenetic studies of epithelial ovarian carcinoma. Cancer Genet Cytogenet 71:76–86, 1993 8. Gallion HH, Powell DE, Morrow JK, Pieretti M, Case E, Turker MS, DePriest PD, Hunter JE, van Nagell JR Jr: Molecular genetic changes in human epithelial ovarian malignancies. Gynecol Oncol 47:137–142, 1992 9. Dabholkar M, Bostick-Bruton F, Reed E: ERCC1 and ERCC2 expression in fresh human tumor tissues. J Natl Cancer Inst 84:1512–1517, 1992

21.

22.

23.

10. Dabholkar MD, Berger MS, Vionnet JA, Egwuagu C, Silber JR, Yu JJ, Reed E: Malignant and nonmalignant brain tissues differ in their messenger RNA expression patterns for ERCC1 and ERCC2. Cancer Res 55:1–6, 1995

24.

11. Liang BC, Ross DA, Reed E: Genomic copy number changes of DNA repair genes ERCC1 and ERCC2 in human gliomas. J Neuro-Oncol 26:17–23, 1995

25.

12. Thompson FH, Nelson MA, Trent JM, Guan X, Liu Y, Yang J, Emerson J, Adair L, Wymer J, Balfour C, Massey K, Weinstein R, Alberts DS,

AID

Gyn 4709

/

6d1c$$$423

06-12-97 18:30:01

Teatle R: Amplification of 19q13.1–q13.2 sequences in ovarian cancer. Cancer Genet Cytogenet 87:55–62, 1996 Reifenberger J, Reifenberger G, Liu L, James CD, Wechsler W, Collins VP: Molecular genetic analysis of oligodendroglial tumors shows preferential allelic deletions on 19q and 1p. Am J Pathol 145:1175–1190, 1994 Bello MJ, Leone PE, Nebreda P, Kusak ME, DeCampos JM, Vaquero J, Sarasa JL, Pestana A, Rey JA: Molecular abnormalities of chromosome 19 in malignant gliomas: preferntial involvement of the 19q13.2– 19q13.4 region. Int J Oncol 6:655–658, 1995 Bello MJ, Leone PE, Vaquero J, De Campos JM, Kusak ME, Sarasa JL, Pestana A, Rey JA: Allelic loss at 1p and 19q frequently occurs in association and may represent early oncogenic events in oligodendroglial tumors. Int J Cancer 64:207–210, 1995 Ritland SR, Ganju V, Jenkins RB: Region-specific loss of heterozygosity on chromosome 19 is related to the morphologic type of human glioma. Genes Chromosom Cancer 12:277–282, 1995 vonDeimling A, Nagel J, Bender B, Lenartz D, Schramm J, Louis DN, Wiestler OD: Deletion mapping of chromosome 19 in human gliomas. Int J Cancer 57:676–680, 1994 Rubio MP, Correa DM, Ueki K, Mohrenweiser HW, Gusella JF, von Deimling A, Louis DN: The putative glioma tumor suppressor gene on chromosome 19q maps between APOC2 and HRC. Cancer Res 54:4760–4763, 1994 Buetow KH, Weber JL, Ludwigsen S, Scherpbier-Heddema T, Duyk GM, Sheffield VC, Wang Z, Murray JC: Integrated human genomewide maps constructed using the CEPH reference panel. Nature Genet 6:391–393, 1994 Weber JL, Wang Z, Hansen K, Stephenson M, Kappel C, Salzman S, Wilkie PJ, Keats B, Dracopoli NC, Brandriff BF, Olsen AS: Evidence for human meiotic recombination interference obtained through construction of a short tandem repeat-polymorphism linkage map of chromosome 19. Am J Hum Genet 53:1079–1095, 1993 Gordon LA, Bergmann A, Christensen M, Danganan L, Lee DA, Ashworth LK, Nelson DO, Olsen AS, Mohrenwieser HW, Carrano AV, Brandriff BF: A 30-MB metric flourescence in situ hybridization map of human chromosome 19q. Genomics 30:187–192, 1995 Ashworth LK, Batzer MA, Brandriff B, Branscomb E, deJong P, Garcia E, Garnes JA, Gordon LA, Lamerdin JE, Lennon G, Mohrenweiser H, Olsen AS, Slezak T, Carrano AV: An integrated metric physical map of human chromosome 19. Nature Genet 1:422–427, 1995 Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A Laboratory Manual, 2nd ed. Plainview, NY, Cold Spring Harbor Laboratory Press, 1989, pp E3–E7 Amfo K, Neyns B, Teugels E, Lissens W, Bourgain C, De Sutter P, Vandamme B, Vamos E, De Greve J: Frequent deletion of chromosome 19 and a rare rearrangement of 19p13.3 involving the insulin receptor gene in human ovarian cancer. Oncogene 11:351–358, 1995 Kiechle-Schwarz M, Bauknecht T, Karck U, Kommoss F, DuBois A, Pfleiderer A: Recurrent cytogenetic aberration and loss of constitutional heterozygosity in ovarian carcinomas. Gynecol Oncol 55:198–205, 1994

goal

AP: GYN