- Email: [email protected]
Allelotype analysis of gallbladder carcinoma associated with anomalous junction of pancreaticobiliary duct
Cancer Letters 166 (2001) 135±141
www.elsevier.com/locate/canlet
Allelotype analysis of gallbladder carcinoma associated with anomalous junction of pancreaticobiliary duct Kentaro Nakayama a,b, Miki Konno a, Atsuko Kanzaki a, Takanori Morikawa c, Hitoshi Miyashita a, Toru Fujioka a, Takafumi Uchida a, Kohji Miyazaki b, Sonshin Takao d, Takashi Aikou d, Manabu Fukumoto a, Yuji Takebayashi a,* a
Department of Pathology, Institute of Development, Aging and Cancer, Tohoku University, Aobaku, Seiryomachi 4-1, Sendai, 9808575, Japan b Department of Obstetrics and Gynecology Shimane Medical University Enyacho, Izumo, 6930021, Japan c Department of First Surgery, Faculty of Medicine, Tohoku University, Aobaku, Seiryomachi 4-1, Sendai, 9808575, Japan d Department of First Surgery, Faculty of Medicine, Kagoshima University, Sakuragaoka 8-35-1, 890, Kagoshima, Japan Received 31 July 2000; received in revised form 12 October 2000; accepted 29 January 2001
Abstract Anomalous junction of pancreaticobiliary duct (AJPBD) patients has an increased risk of gallbladder and bile duct carcinomas. However, the relevance of carcinoma with AJPBD is not fully clari®ed. We performed analysis of loss of heterozygosity (LOH) at p53 locus and immunohistochemistry of p53 and K-ras gene mutation in ®ve cases of gallbladder carcinoma associated with AJPBD. LOH of p53 locus and overexpression of p53 were detected in two out of ®ve (40%) and ®ve out of ®ve (100%), respectively, in the present study. K-ras gene mutation at codon 12 and 13 was not detected (0%, 0/5). These results suggest that aberrations of p53 are involved in carcinogenesis of gallbladder carcinoma associated with AJPBD. Next, in order to ®nd the genetic events besides K-ras mutation and overexpression of mutant p53 in this disease, LOH analysis was performed using 72 microsatellite markers. High frequency of allelic loss (.50%) was found on 2p (81.8%), 4p (50%), 4q (50%), 8q (60%), 9q (50%), 10p (50%), 14p (60%), 14q (50%), 16p (60%), 19p (50%), 21p (50%) and Xp (66.6%). The highest deletion regions on chromosome 2p24 (3/3, 100%), 14q22 (3/4, 75%) and 21q22 (3/4, 75%) were found. The present study suggests that gallbladder carcinoma associated with AJPBD has high frequent allelic loss and has two new regions which may harbor putative tumor suppressor genes. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Anomalous junction of pancreaticobiliary duct; Carcinogensis; Gallbladder carcinoma; Loss of heterozygosity
1. Introduction Anomalous junction of pancreaticobiliary duct (AJPBD) was ®rst de®ned that AJPBD was a cause of congenital biliary dilatation by Babbit DP [1]. The junction of the pancreatic and common bile duct is * Corresponding author. Tel.: 1 81-22-717-8511; fax: 181-22717-8512. E-mail address: [email protected] (Y. Takebayashi).
present outside the duodenal wall where a sphincter system is absent. The frequency of AJPBD in Japan is more than that in western country [2±5]. AJPBD patients have an increased risk of gall bladder and bile duct carcinomas. This risk may approach a 20fold increase over the general population, if all sites are considered. Several investigations of an association between AJPBD and gall bladder carcinoma have been also reported in Japan. Kimura et al reported that gall bladder carcinoma with AJPBD was present in
0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00436-0
136
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
alteration of many tumor suppressor genes is the mutation of one allele and the loss of the other allele [11]. The loss of one allele can be detected as loss of heterozygosity (LOH) with informative markers in the tumor suppressor gene regions. Therefore, LOH assay has been widely used as an indirect approach in the search for a tumor suppressor gene [12]. The aim of this study was to search for a putative tumor suppressor gene in gallbladder carcinomas associated with AJPBD. We therefore performed the detection of Kras mutations, immunohistochemical analysis of p53 and allelotype studies of ®ve gallbladder carcinomas associated with AJPBD using 72 microsatellite markers.
24.6% among 65 cases with AJPBD, compared with 1.9% of those without AJPBD [2]. Yamauchi et al. [3] reported that patients with AJPBD gallbladder carcinoma were earlier than those without AJPBD. It was also suggested that the changes of bile juice induce chronic in¯ammation of biliary tract epithelium, leading to mucosal metaplasia and biliary tract malignancy. However, the mechanism of carcinogenesis of the gall bladder mucosa with AJPBD has not been demonstrated [2±5]. Previously, Hanada et al. showed that a sequence of hyperplastic changes with a corresponding increase in cellular kinetics and progression through dysplasia to carcinoma may be important in carcinogenesis in the gallbladders of patients with AJPBD by histopathological and immunohistochemical analysis using PCNA antibody [6]. Taken together with these ®ndings, AJPBD may be a risk and a model of carcinogenesis of gallbladder mucosa. A little has been known about the molecular mechanisms of development and/or progression of gallbladder carcinoma associated with AJPBD. Recently, Hanada et al. and Matsubara et al. demonstrated that K-ras mutations play an important role in cases of early gallbladder carcinoma with AJPBD by PCR-restriction-length polymorphism [7][8]. Furthermore, Hanada et al. also suggested that the speci®c Kras mutation in codon 12 (GGT to GAT) contribute to the early stage of carcinogenesis in the gallbladder with AJPBD [9]. Carcinogenesis of human cancer is due to the accumulation of genetic alterations both in oncogene and tumor suppressor gene [10]. Furthermore, the typical
2. Materials and methods 2.1. Patients and tumor samples Samples used in this study were collected from patients during the time of primary surgery at the Department of First Surgery, Faculty of Medicine, Kagoshima University, Kagoshima, Japan. Summary of clinicopathologic features of patients with gallbladder carcinoma associated with anomalous junction of panreaticobilliary duct (AJPBD) was shown in Table 1. All patients were female. The range of age was from 48 to 70. Histopathological diagnosis showed four well-differentiated and one moderately-differentiated adenocarcinoma. Clinical stages of gallbladder carcinoma patients according to TNM classi®cation were one case of stage II and four of stage III. None
Table 1 K-ras mutation, p53 LOH and p53 immunohistochemistry in gall bladder carcinoma associated with anomalous junction of pancreaticobiliary duct Patient no.
Age
Sex
Degree of differentiation a
Stage b
K-ras mutation codon 12/13
p53 LOH
p53 IHC c
1 2 3 4 5
55 62 70 62 48
F F F F F
Well Well Mode Well Well
III III III II III
2/2 2/2 2/2 2/2 2/2
1 2 1 2 2
1 1 1 1 1
a b c
Well/mode: well/moderately differentiated adenocarcinoma. Stage was determined by general rules for surgical and pathological studies on cancer of the biliary tract. Immunohistochemistry.
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
137
overnight at 588C in a digestion buffer (0.01 M NaCl, 0.5 M Tris±HCl, pH 8.0, 20 mM EDTA, 0.05% Tween20R, 0.1 mg/ml proteinase K). The samples were then heated to 958C for 10 min to inactivate proteinase K activity. After the digestion, DNA was extracted with phenol/chloroform treatment and ethanol precipitation. Normal DNA for archival specimens was similarly prepared from uninvolved areas of the same section. 2.3. Mutation analysis of K-ras gene
Fig. 1. Microdissection of gallbladder carcinoma with AJPBD. (A) Gallbladder carcinoma with AJPBD before microdissection ( £ 100, H&E). (B) Normal stromal component was removed by microdissection and the remaining carcinoma cells ( £ 100, H&E).
of the patients received chemotherapy or irradiation prior to surgery. After surgically removing tumors, tissues were ®xed by buffered formaline and embedded in paraf®n. 2.2. Microdissection and DNA extraction Before being processed for DNA extraction, all tumor samples were carefully examined histopathologically by use of sections stained with hematoxylin± eosin. These histopathological examinations revealed the presence of large amount of contaminating stroma. After deparaf®nization of 10 mm thickness sections stained with hematoxylin, non-neoplastic component were removed by LSM ver. 3.50 (Carl Zwiss, Germany), in order to minimize the number of nonneoplastic stromal cells (Fig. 1). The carcinoma cells were collected in Eppendorf tube and incubated
Analysis of K-ras mutation was performed as described before [13]. The exon 1 of the K-ras gene was ampli®ed by the nested PCR. The primers used for the ®rst-round PCR were 5 0 -TTATTTTATTATAAGGCCTGCTGAA-3 0 for upstream and 5 0 -TATGCATATTAAAACAAGATTTACCT-3 0 for downstream, and for the second-round PCR were 5 0 AAATGACTGAATATAAACTTGTGG-3 0 for upstream and 5 0 -CTCTATTGTTGGATCATATTCGTC-3 0 for downstream. After denaturation at 958C for 5 min, DNA ampli®cation was performed in 30 cycles consisting of denaturation at 958C for 30 s, primer annealing at 50 8C for 1 min, and extension for 1 min at 728C (Fig. 2A). The PCR product was directly sequenced using the ABI Prism Big Dye Terminator Sequencing kit and ABI 310 sequencers (Fig. 2B). 2.4. PCR ampli®cation and analysis of loss of heterozygosity Analysis of loss of heterozygosity (LOH) was performed by PCR ampli®cation using 72 microsatellite markers (Research Genetics, AL, information of primers can be obtained from Y. Takebayashi). Each primer was labeled with ¯uorescent dye. PCR reactions were performed in a total volume of 10 ml containing 10ng of DNA, dGTP, dATP, dTTP and dCTP at a concentration of 200 mM, 0.4 mM of each primer and 0.25 unit of platinum Taq polymerase (Gibco, BRL, Rockville, MD). The PCR condition consisted of 10 min at 948C followed by 45 cycles of 45 s at 948C, 1 min at the appropriate annealing temperature, and 1 min at 728C, followed by 728C for 10 min. After the denaturation of PCR products at 948C for 2 min, samples were subjected to electrophoresis using a Performance Optimized Polymer 4 in a 310 Genetic Analyzer (Applied Biosystems,
138
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
Foster, CA). LOH analysis was performed by Gene Scan version (2.1). When the different two parental alleles were obtained in normal DNA, the cases were regarded as informative cases. Cases with homogenous allele or non-PCR product detected in normal DNA were non-informative cases. Allelic intensity obtained from normal DNA was compared with those obtained from matched tumor samples derived from the same patients. Allelic loss was scored when band intensity of one allelic marker was signi®cantly decreased (more than 50% reduction) in tumor DNA as compared with that in normal DNA (Fig. 3). We performed the repeated experiments and con®rmed the results of LOH in several cases. 2.5. Immunohistochemical analysis of p53 Sections were deparaf®nized with xylene and dehydrated with 98% ethanol. For better immunohistochemical reactivity with p53 antibodies, deparaf®nized tissue sections were placed in 10 mmol/l citrate buffer, pH 6.0, and heated at 750W in a microwave oven, for 20 minutes. Endogenous peroxidase was blocked by immersing the slides in 0.3% hydrogen peroxide in absolute methanol for 30 min at room temperature. After washing three times for 5 min each with PBS, the sections were blocked by soaking for 30 min at room temperature in PBS containing 1% bovine serum albumin. The blocked sections were incubated overnight at 48C with 2.5 mg/ml of a monoclonal antibody against p53 protein (Clone
Fig. 2. Analysis of K-ras mutation in gallbladder carcinoma with AJPBD. (A) Nested-PCR. The PCR products of second-round PCR for K-ras were shown in 2% agarose gel. (B) Analysis of K-rassequence. The codon12 &13 of K-ras in case 1 was presented. (C) Immunohistochemistry of p53 using monoclonal antibody against p53 (DO-7). High expression of mutant p53 in case 1 ( £ 100).
Fig. 3. Microsatellite analysis of gallbladder carcinoma with AJPBD. LOH was detected at marker D2S1400 in gallbladder carcinoma with AJPBD.
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
DO-7, DAKO, Denmark). After 12 h the slides were incubated for 30 min at room temperature with biotinylated anti-mouse IgG diluted 100-fold with PBS, washed three times for 15 min in PBS, and incubated for 50 min with avidin±biotin±peroxidase complex diluted with PBS. After three washes for 15 min with PBS, the sections were incubated for 7 min with 0.5 mg/ml diaminobenzidine and 0.03% (v/v) H2O2 in PBS and ®nally counterstained with hematoxylin prior to mounting. 2.6. Evaluation of the Stained Sections. For p53 analysis, we examined at least 1000 carcinoma cells to determine whether the cells were positive for p53 at high power (£400) after screening for the areas with the highest intensity of staining at lower power (£100). When more than 10% of the carcinoma cells were stained, the specimen was de®ned as positive (Fig. 2C) according to the previous report [7]. Immunohistochemical evaluations were performed by two investigators (K.N. and Y.T.) independently without prior information of the clinicopathologic features. 3. Results and discussion Recently, various oncogenes and tumor suppressor genes have been investigated. p53 and K-ras play an important role in carcinogenesis in a lot of tumor [13], including gallbladder carcinoma [14±16]. As for p53 overexpression and mutation, the incidence of them
139
were 67 and 50%, respectively, in the cases of gallbladder carcinoma with AJPBD by immunohistochemical and PCR-SSCP [7]. In our study, loss of heterozygosity (LOH) at p53 locus and overexpression of p53 by immunohistochemistry were determined in two of ®ve (40%) and ®ve of ®ve (100%) in gallbladder carcinoma associated with AJPBD (Table 1). These results suggested that two carcinomas with LOH at p53 locus and overexpression of p53 were compatible with two-hit theory by Kunudson [11]. The mutations of p53 appeared to be took place in the other three carcinomas without LOH at p53 locus. There might be mutation(s) of p53 which could not be speculated by the microsatellite marker used in this study. Concerning K-ras mutation at codon 12, Hanada et al. has demonstrated that the incidence of K-ras mutation were 15% (3/20) in hyperplasia, 60% (6/10) in stage I carcinoma, and 100% (5/5) in stage II±IV carcinoma in the cases with AJPBD. The incidence of K-rasmutation were 0% (0/4) in normal gallbladder, 0% (0/4) in hyperplasia, 17% (1/6) in adenoma, 7% (1/16) in stage I carcinoma and 38% (3/8) in stage II±IV carcinoma. These suggested that K-ras mutation in codon 12 contribute more to the early stage of carcinogenesis in the gallbladder carcinoma with AJPBD than to that without AJPBD [9]. Unexpectedly, K-ras mutation at codon 12 and 13 was not detected in the present study (0%, 5/5) (Table 1, Fig. 2B), although we obtained genomic DNA from paraf®n embedded sections using microdissection (Fig. 1). Previous study also revealed the incidence of K-ras mutation was 0% using a method of direct sequence [14]. Taken together with these results, Kras mutation and overexpression of p53 are partly involved in carcinogenesis of gallbladder with AJPBD. Next, in order to ®nd another genetic events besides Table 2 High frequent loci of loss of heterozygocity in gall bladder carcinoma associated with anomalous junction of pancreaticobiliary duct
Fig. 4. Allelogram of gallbladder carcinoma with AJPBD using 72 markers for genome wide scanning of LOH. LOH frequency of chromosome arm is de®ned as percentage of total LOH divided by number of informative cases on each chromosome arm.
Locus
Chromosomal location
Frequency of LOH LOH/ informative cases (%)
D25 1400 D145599 D21S1400
2p 24.1 14q12 21q22.2
3/3 (100) 3/4 (75) 3/4 (75)
140
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
aberration of p53 in the carcinogenesis of gallbladder with AJPBD, we performed genome wide scanning of LOH study (Fig. 4; Table 2). Forty (93%) of 43 chromosome arms revealed LOH in at least one patient. LOH could not be analyzed at chromosome 15p, 18p and 21p since of running failure and homogenous allele. Six chromosomal arms (7q, 9p, 11p, 12p, 13p and 22p) did not show any LOH. Every ®ve patients with gallbladder carcinoma with AJPBD showed LOH and the frequency of LOH in each patient were 56.4% (44/78), 38.6% (27/70), 32.8% (18/58), 30.1% (24/78) and 27.8% (10/36) from case 1±5. Allelotyping was performed to determine which chromosomes carry a tumor suppressor gene for gallbladder carcinomas with AJPBD. High frequency of allelic loss (.50%) was found on 2p (81.8%), 4p (50%), 4q (50%), 8q (60%), 9q (50%), 10p (50%), 14p (60%), 14q (50%), 16p (60%), 19p (50%), 21p (50%) and Xp (66.6%). Other chromosome arms had LOH frequency of less than 50% (Fig. 3). The highest deletion regions on chromosome 2p24 (3/3, 100%), 14q22 (3/4, 75%) and 21q22 (3/4, 75%) were found. The LOH on chromosome 21q22 has been observed in oral cancer [17]. As for the deletion regions on chromosome 14q22, LOH of this has not been also reported. Interestingly, this region includes a CDK-associated phosphatase, KAP gene which was shown to dephosphorylate Thr160 in human Cdk2. The binding of cyclin A to Cdk2 inhibited the dephosphorylation of Thr160 by KAP. Moreover, the dephosphorylation of Thr160 by KAP prevented Cdk2 kinase activity upon subsequent association with cyclin A [18]. With this ®nding, deletion of KAP may lead the abnormality of cell cycle and the carcinogenesis of gallbladder carcinoma with AJPBD. The frequent LOH on 2p24 is the most interesting ®nding in this study. This has not been found previously in any malignancies and the candidate genes are not known. In summary, comprehensive allelotype study of gallbladder carcinoma with AJPBD was completed using microsatellite markers. This study suggests that another genetic alterations may play an important role in carcinogenesis in gallbladder carcinoma with AJPBD besides p53 and K-ras. At least two new sites of frequent LOH, which may harbor putative tumor suppressor genes were identi®ed, one is 2p24 and the other is 14q22. Further studies of ®ne mapping to localize the area of loss on these chromosomes are
required to identify tumor suppressor genes in gallbladder carcinoma associated with AJPBD. References [1] D.P. Babbitt, Congenital choledochal cyst: New etiological concept based on anomalous relationships of the common bile duct and pancreatic bulb, Ann. Radiol. 12 (1969) 231±240. [2] K. Kimura, M. Ohto, H. Saisho, T. Unozawa, Y. Tsuchiya, M. Morita, M. Ebara, S. Matsutani, K. Okuda, Association of gallbladder carcinoma and anomalous pancreaticobiliary ductal union, Gastroenterology 89 (1985) 1258±1265. [3] S. Yamauchi, A. Koda, S. Matsumoto, M. Tanaka, F. Nakayama, Anomalous junction of pancreaticobiliary duct without congenital choledochal cyst: a possible risk factor for gallbladder cancer, Am. J. Gastroenterol. 82 (1987) 20±24. [4] T. Ohta, T. Nagakawa, K. Ueno, K. Maeda, N. Ueda, M. Kayahara, T. Akiyama, M. Kanno, I. Konishi, R. Izumi, Clinical experience of biliary tract carcinoma associated with anomalous union of the pancreaticobiliary ductal system, Jpn. J. Surg. 20 (1990) 36±43. [5] M. Yamamoto, S. Nakajo, E. Tahara, M. Ito, K. Taniyama, F. Shimamoto, N. Miyoshi, Y. Hayashi, H. Akiyama, S. Nakai, Mucosal changes of the gallbladder in anomalous union with the pancreatico-biliary duct system, Pathol. Res. Pract. 187 (1991) 241±246. [6] K. Hanada, M. Itoh, K. Fujii, A. Tsuchida, M. Hirata, S. Ishimaru, T. Iwao, N. Eguchi, G. Kajiyama, Pathology and cellular kinetics of gallbladder with an anomalous junction of the pancreaticobiliary duct, Am. J. Gastroenterol. 91 (1996) 1007±1011. [7] K. Hanada, M. Itoh, K. Fujii, A. Tsuchida, H. Ooishi, G. Kajiyama, K-ras and p53 mutations in stage I gallbladder carcinoma with an anomalous junction of the pancreaticobiliary duct, Cancer 77 (1996) 452±458. [8] T. Matsubara, Y. Sakurai, Y. Sasayama, H. Hori, M. Ochiai, T. Funabiki, K. Matsumoto, I. Hirono, K-ras point mutations in cancerous and non-cancerous biliary epithelium in patients with pancreaticobiliary maljunction, Cancer 77 (1996) 1752± 1757. [9] K. Hanada, A. Tsuchida, T. Iwao, N. Eguchi, T. Sasaki, K. Morinaka, K. Matsubara, Y. Kawasaki, S. Yamamoto, G. Kajiyama, Gene mutations of K-ras in gallbladder mucosae and gallbladder carcinoma with an anomalous junction of the pancreaticobiliary duct, Am. J. Gastroenterol. 94 (1999) 1638±1642. [10] B. Vogelstein, E.R. Fearon, S.R. Hamilton, S.E. Kern, A.C. Preisinger, M. Leppert, Y. Nakamura, R. White, A.M. Smits, J.L. Bos, Genetic alterations during colorectal-tumor development, N. Engl. J. Med. 319 (1988) 525±532. [11] A.G. Kunudson, Antioncogenes and human cancer, Proc. Natl. Acad. Sci. USA 1 (1993) 10914±10921. [12] B. Vogelstein, E.R. Fearon, S.E. Kern, S.R. Hamilton, A.C. Preisinger,, Y. Nakamura, R. White, Allelotype of colorectal carcinomas, Science 244 (1989) 207±211. [13] M. Mandai, I. Konishi, H. Kuroda, T. Komatsu, S. Yamamoto,
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141 K. Nanbu, K. Matsushita, M. Fukumoto, H. Yamabe, T. Mori, Heterogeneous distribution of K-ras-mutated epithelia in mucinous ovarian tumors with special reference to histopathology, Hum. Pathol. 29 (1998) 34±40. [14] M. Tada, M. Omata, M. Ohto, Analysis of ras gene mutations in human hepatic malignant tumors by polymerase chain reaction and direct sequencing, Cancer Res. 50 (1990) 1121±1124. [15] M. Watanabe, M. Asaka, J. Tanaka, M. Kurosawa, M. Kasai, T. Miyazaki, Point mutation of K-ras gene codon 12 in biliary tract tumors, Gastroenterology 107 (1994) 1147±1153. [16] I.I. Wistuba, K. Sugio, J. Hung, Y. Kishimoto, A.K. Virmani,
141
I. Roa, J. Albores-Saavedra, A.F. Gazdar, Allele-speci®c mutations involved in the pathogenesis of endemic gallbladder carcinoma in Chile, Cancer Res. 55 (1995) 2511±2515. [17] N. Yamamoto, K. Uzawa, T. Miya, T. Watanabe, H. Yokoe, T. Shibahara, H. Noma, H. Tanzawa, Frequent allelic loss/imbalance on the long arm of chromosome 21 in oral cancer: evidence for three discrete tumor suppressor gene loci, Oncol. Rep. 6 (1999) 1223±1227. [18] R.Y. Poon, T. Hunter, Dephosphorylation of Cdk2 Thr160 by the cyclin-dependent kinase- interacting phosphatase KAP in the absence of cyclin, Science 270 (1995) 90±93.
www.elsevier.com/locate/canlet
Allelotype analysis of gallbladder carcinoma associated with anomalous junction of pancreaticobiliary duct Kentaro Nakayama a,b, Miki Konno a, Atsuko Kanzaki a, Takanori Morikawa c, Hitoshi Miyashita a, Toru Fujioka a, Takafumi Uchida a, Kohji Miyazaki b, Sonshin Takao d, Takashi Aikou d, Manabu Fukumoto a, Yuji Takebayashi a,* a
Department of Pathology, Institute of Development, Aging and Cancer, Tohoku University, Aobaku, Seiryomachi 4-1, Sendai, 9808575, Japan b Department of Obstetrics and Gynecology Shimane Medical University Enyacho, Izumo, 6930021, Japan c Department of First Surgery, Faculty of Medicine, Tohoku University, Aobaku, Seiryomachi 4-1, Sendai, 9808575, Japan d Department of First Surgery, Faculty of Medicine, Kagoshima University, Sakuragaoka 8-35-1, 890, Kagoshima, Japan Received 31 July 2000; received in revised form 12 October 2000; accepted 29 January 2001
Abstract Anomalous junction of pancreaticobiliary duct (AJPBD) patients has an increased risk of gallbladder and bile duct carcinomas. However, the relevance of carcinoma with AJPBD is not fully clari®ed. We performed analysis of loss of heterozygosity (LOH) at p53 locus and immunohistochemistry of p53 and K-ras gene mutation in ®ve cases of gallbladder carcinoma associated with AJPBD. LOH of p53 locus and overexpression of p53 were detected in two out of ®ve (40%) and ®ve out of ®ve (100%), respectively, in the present study. K-ras gene mutation at codon 12 and 13 was not detected (0%, 0/5). These results suggest that aberrations of p53 are involved in carcinogenesis of gallbladder carcinoma associated with AJPBD. Next, in order to ®nd the genetic events besides K-ras mutation and overexpression of mutant p53 in this disease, LOH analysis was performed using 72 microsatellite markers. High frequency of allelic loss (.50%) was found on 2p (81.8%), 4p (50%), 4q (50%), 8q (60%), 9q (50%), 10p (50%), 14p (60%), 14q (50%), 16p (60%), 19p (50%), 21p (50%) and Xp (66.6%). The highest deletion regions on chromosome 2p24 (3/3, 100%), 14q22 (3/4, 75%) and 21q22 (3/4, 75%) were found. The present study suggests that gallbladder carcinoma associated with AJPBD has high frequent allelic loss and has two new regions which may harbor putative tumor suppressor genes. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Anomalous junction of pancreaticobiliary duct; Carcinogensis; Gallbladder carcinoma; Loss of heterozygosity
1. Introduction Anomalous junction of pancreaticobiliary duct (AJPBD) was ®rst de®ned that AJPBD was a cause of congenital biliary dilatation by Babbit DP [1]. The junction of the pancreatic and common bile duct is * Corresponding author. Tel.: 1 81-22-717-8511; fax: 181-22717-8512. E-mail address: [email protected] (Y. Takebayashi).
present outside the duodenal wall where a sphincter system is absent. The frequency of AJPBD in Japan is more than that in western country [2±5]. AJPBD patients have an increased risk of gall bladder and bile duct carcinomas. This risk may approach a 20fold increase over the general population, if all sites are considered. Several investigations of an association between AJPBD and gall bladder carcinoma have been also reported in Japan. Kimura et al reported that gall bladder carcinoma with AJPBD was present in
0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00436-0
136
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
alteration of many tumor suppressor genes is the mutation of one allele and the loss of the other allele [11]. The loss of one allele can be detected as loss of heterozygosity (LOH) with informative markers in the tumor suppressor gene regions. Therefore, LOH assay has been widely used as an indirect approach in the search for a tumor suppressor gene [12]. The aim of this study was to search for a putative tumor suppressor gene in gallbladder carcinomas associated with AJPBD. We therefore performed the detection of Kras mutations, immunohistochemical analysis of p53 and allelotype studies of ®ve gallbladder carcinomas associated with AJPBD using 72 microsatellite markers.
24.6% among 65 cases with AJPBD, compared with 1.9% of those without AJPBD [2]. Yamauchi et al. [3] reported that patients with AJPBD gallbladder carcinoma were earlier than those without AJPBD. It was also suggested that the changes of bile juice induce chronic in¯ammation of biliary tract epithelium, leading to mucosal metaplasia and biliary tract malignancy. However, the mechanism of carcinogenesis of the gall bladder mucosa with AJPBD has not been demonstrated [2±5]. Previously, Hanada et al. showed that a sequence of hyperplastic changes with a corresponding increase in cellular kinetics and progression through dysplasia to carcinoma may be important in carcinogenesis in the gallbladders of patients with AJPBD by histopathological and immunohistochemical analysis using PCNA antibody [6]. Taken together with these ®ndings, AJPBD may be a risk and a model of carcinogenesis of gallbladder mucosa. A little has been known about the molecular mechanisms of development and/or progression of gallbladder carcinoma associated with AJPBD. Recently, Hanada et al. and Matsubara et al. demonstrated that K-ras mutations play an important role in cases of early gallbladder carcinoma with AJPBD by PCR-restriction-length polymorphism [7][8]. Furthermore, Hanada et al. also suggested that the speci®c Kras mutation in codon 12 (GGT to GAT) contribute to the early stage of carcinogenesis in the gallbladder with AJPBD [9]. Carcinogenesis of human cancer is due to the accumulation of genetic alterations both in oncogene and tumor suppressor gene [10]. Furthermore, the typical
2. Materials and methods 2.1. Patients and tumor samples Samples used in this study were collected from patients during the time of primary surgery at the Department of First Surgery, Faculty of Medicine, Kagoshima University, Kagoshima, Japan. Summary of clinicopathologic features of patients with gallbladder carcinoma associated with anomalous junction of panreaticobilliary duct (AJPBD) was shown in Table 1. All patients were female. The range of age was from 48 to 70. Histopathological diagnosis showed four well-differentiated and one moderately-differentiated adenocarcinoma. Clinical stages of gallbladder carcinoma patients according to TNM classi®cation were one case of stage II and four of stage III. None
Table 1 K-ras mutation, p53 LOH and p53 immunohistochemistry in gall bladder carcinoma associated with anomalous junction of pancreaticobiliary duct Patient no.
Age
Sex
Degree of differentiation a
Stage b
K-ras mutation codon 12/13
p53 LOH
p53 IHC c
1 2 3 4 5
55 62 70 62 48
F F F F F
Well Well Mode Well Well
III III III II III
2/2 2/2 2/2 2/2 2/2
1 2 1 2 2
1 1 1 1 1
a b c
Well/mode: well/moderately differentiated adenocarcinoma. Stage was determined by general rules for surgical and pathological studies on cancer of the biliary tract. Immunohistochemistry.
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
137
overnight at 588C in a digestion buffer (0.01 M NaCl, 0.5 M Tris±HCl, pH 8.0, 20 mM EDTA, 0.05% Tween20R, 0.1 mg/ml proteinase K). The samples were then heated to 958C for 10 min to inactivate proteinase K activity. After the digestion, DNA was extracted with phenol/chloroform treatment and ethanol precipitation. Normal DNA for archival specimens was similarly prepared from uninvolved areas of the same section. 2.3. Mutation analysis of K-ras gene
Fig. 1. Microdissection of gallbladder carcinoma with AJPBD. (A) Gallbladder carcinoma with AJPBD before microdissection ( £ 100, H&E). (B) Normal stromal component was removed by microdissection and the remaining carcinoma cells ( £ 100, H&E).
of the patients received chemotherapy or irradiation prior to surgery. After surgically removing tumors, tissues were ®xed by buffered formaline and embedded in paraf®n. 2.2. Microdissection and DNA extraction Before being processed for DNA extraction, all tumor samples were carefully examined histopathologically by use of sections stained with hematoxylin± eosin. These histopathological examinations revealed the presence of large amount of contaminating stroma. After deparaf®nization of 10 mm thickness sections stained with hematoxylin, non-neoplastic component were removed by LSM ver. 3.50 (Carl Zwiss, Germany), in order to minimize the number of nonneoplastic stromal cells (Fig. 1). The carcinoma cells were collected in Eppendorf tube and incubated
Analysis of K-ras mutation was performed as described before [13]. The exon 1 of the K-ras gene was ampli®ed by the nested PCR. The primers used for the ®rst-round PCR were 5 0 -TTATTTTATTATAAGGCCTGCTGAA-3 0 for upstream and 5 0 -TATGCATATTAAAACAAGATTTACCT-3 0 for downstream, and for the second-round PCR were 5 0 AAATGACTGAATATAAACTTGTGG-3 0 for upstream and 5 0 -CTCTATTGTTGGATCATATTCGTC-3 0 for downstream. After denaturation at 958C for 5 min, DNA ampli®cation was performed in 30 cycles consisting of denaturation at 958C for 30 s, primer annealing at 50 8C for 1 min, and extension for 1 min at 728C (Fig. 2A). The PCR product was directly sequenced using the ABI Prism Big Dye Terminator Sequencing kit and ABI 310 sequencers (Fig. 2B). 2.4. PCR ampli®cation and analysis of loss of heterozygosity Analysis of loss of heterozygosity (LOH) was performed by PCR ampli®cation using 72 microsatellite markers (Research Genetics, AL, information of primers can be obtained from Y. Takebayashi). Each primer was labeled with ¯uorescent dye. PCR reactions were performed in a total volume of 10 ml containing 10ng of DNA, dGTP, dATP, dTTP and dCTP at a concentration of 200 mM, 0.4 mM of each primer and 0.25 unit of platinum Taq polymerase (Gibco, BRL, Rockville, MD). The PCR condition consisted of 10 min at 948C followed by 45 cycles of 45 s at 948C, 1 min at the appropriate annealing temperature, and 1 min at 728C, followed by 728C for 10 min. After the denaturation of PCR products at 948C for 2 min, samples were subjected to electrophoresis using a Performance Optimized Polymer 4 in a 310 Genetic Analyzer (Applied Biosystems,
138
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
Foster, CA). LOH analysis was performed by Gene Scan version (2.1). When the different two parental alleles were obtained in normal DNA, the cases were regarded as informative cases. Cases with homogenous allele or non-PCR product detected in normal DNA were non-informative cases. Allelic intensity obtained from normal DNA was compared with those obtained from matched tumor samples derived from the same patients. Allelic loss was scored when band intensity of one allelic marker was signi®cantly decreased (more than 50% reduction) in tumor DNA as compared with that in normal DNA (Fig. 3). We performed the repeated experiments and con®rmed the results of LOH in several cases. 2.5. Immunohistochemical analysis of p53 Sections were deparaf®nized with xylene and dehydrated with 98% ethanol. For better immunohistochemical reactivity with p53 antibodies, deparaf®nized tissue sections were placed in 10 mmol/l citrate buffer, pH 6.0, and heated at 750W in a microwave oven, for 20 minutes. Endogenous peroxidase was blocked by immersing the slides in 0.3% hydrogen peroxide in absolute methanol for 30 min at room temperature. After washing three times for 5 min each with PBS, the sections were blocked by soaking for 30 min at room temperature in PBS containing 1% bovine serum albumin. The blocked sections were incubated overnight at 48C with 2.5 mg/ml of a monoclonal antibody against p53 protein (Clone
Fig. 2. Analysis of K-ras mutation in gallbladder carcinoma with AJPBD. (A) Nested-PCR. The PCR products of second-round PCR for K-ras were shown in 2% agarose gel. (B) Analysis of K-rassequence. The codon12 &13 of K-ras in case 1 was presented. (C) Immunohistochemistry of p53 using monoclonal antibody against p53 (DO-7). High expression of mutant p53 in case 1 ( £ 100).
Fig. 3. Microsatellite analysis of gallbladder carcinoma with AJPBD. LOH was detected at marker D2S1400 in gallbladder carcinoma with AJPBD.
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
DO-7, DAKO, Denmark). After 12 h the slides were incubated for 30 min at room temperature with biotinylated anti-mouse IgG diluted 100-fold with PBS, washed three times for 15 min in PBS, and incubated for 50 min with avidin±biotin±peroxidase complex diluted with PBS. After three washes for 15 min with PBS, the sections were incubated for 7 min with 0.5 mg/ml diaminobenzidine and 0.03% (v/v) H2O2 in PBS and ®nally counterstained with hematoxylin prior to mounting. 2.6. Evaluation of the Stained Sections. For p53 analysis, we examined at least 1000 carcinoma cells to determine whether the cells were positive for p53 at high power (£400) after screening for the areas with the highest intensity of staining at lower power (£100). When more than 10% of the carcinoma cells were stained, the specimen was de®ned as positive (Fig. 2C) according to the previous report [7]. Immunohistochemical evaluations were performed by two investigators (K.N. and Y.T.) independently without prior information of the clinicopathologic features. 3. Results and discussion Recently, various oncogenes and tumor suppressor genes have been investigated. p53 and K-ras play an important role in carcinogenesis in a lot of tumor [13], including gallbladder carcinoma [14±16]. As for p53 overexpression and mutation, the incidence of them
139
were 67 and 50%, respectively, in the cases of gallbladder carcinoma with AJPBD by immunohistochemical and PCR-SSCP [7]. In our study, loss of heterozygosity (LOH) at p53 locus and overexpression of p53 by immunohistochemistry were determined in two of ®ve (40%) and ®ve of ®ve (100%) in gallbladder carcinoma associated with AJPBD (Table 1). These results suggested that two carcinomas with LOH at p53 locus and overexpression of p53 were compatible with two-hit theory by Kunudson [11]. The mutations of p53 appeared to be took place in the other three carcinomas without LOH at p53 locus. There might be mutation(s) of p53 which could not be speculated by the microsatellite marker used in this study. Concerning K-ras mutation at codon 12, Hanada et al. has demonstrated that the incidence of K-ras mutation were 15% (3/20) in hyperplasia, 60% (6/10) in stage I carcinoma, and 100% (5/5) in stage II±IV carcinoma in the cases with AJPBD. The incidence of K-rasmutation were 0% (0/4) in normal gallbladder, 0% (0/4) in hyperplasia, 17% (1/6) in adenoma, 7% (1/16) in stage I carcinoma and 38% (3/8) in stage II±IV carcinoma. These suggested that K-ras mutation in codon 12 contribute more to the early stage of carcinogenesis in the gallbladder carcinoma with AJPBD than to that without AJPBD [9]. Unexpectedly, K-ras mutation at codon 12 and 13 was not detected in the present study (0%, 5/5) (Table 1, Fig. 2B), although we obtained genomic DNA from paraf®n embedded sections using microdissection (Fig. 1). Previous study also revealed the incidence of K-ras mutation was 0% using a method of direct sequence [14]. Taken together with these results, Kras mutation and overexpression of p53 are partly involved in carcinogenesis of gallbladder with AJPBD. Next, in order to ®nd another genetic events besides Table 2 High frequent loci of loss of heterozygocity in gall bladder carcinoma associated with anomalous junction of pancreaticobiliary duct
Fig. 4. Allelogram of gallbladder carcinoma with AJPBD using 72 markers for genome wide scanning of LOH. LOH frequency of chromosome arm is de®ned as percentage of total LOH divided by number of informative cases on each chromosome arm.
Locus
Chromosomal location
Frequency of LOH LOH/ informative cases (%)
D25 1400 D145599 D21S1400
2p 24.1 14q12 21q22.2
3/3 (100) 3/4 (75) 3/4 (75)
140
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141
aberration of p53 in the carcinogenesis of gallbladder with AJPBD, we performed genome wide scanning of LOH study (Fig. 4; Table 2). Forty (93%) of 43 chromosome arms revealed LOH in at least one patient. LOH could not be analyzed at chromosome 15p, 18p and 21p since of running failure and homogenous allele. Six chromosomal arms (7q, 9p, 11p, 12p, 13p and 22p) did not show any LOH. Every ®ve patients with gallbladder carcinoma with AJPBD showed LOH and the frequency of LOH in each patient were 56.4% (44/78), 38.6% (27/70), 32.8% (18/58), 30.1% (24/78) and 27.8% (10/36) from case 1±5. Allelotyping was performed to determine which chromosomes carry a tumor suppressor gene for gallbladder carcinomas with AJPBD. High frequency of allelic loss (.50%) was found on 2p (81.8%), 4p (50%), 4q (50%), 8q (60%), 9q (50%), 10p (50%), 14p (60%), 14q (50%), 16p (60%), 19p (50%), 21p (50%) and Xp (66.6%). Other chromosome arms had LOH frequency of less than 50% (Fig. 3). The highest deletion regions on chromosome 2p24 (3/3, 100%), 14q22 (3/4, 75%) and 21q22 (3/4, 75%) were found. The LOH on chromosome 21q22 has been observed in oral cancer [17]. As for the deletion regions on chromosome 14q22, LOH of this has not been also reported. Interestingly, this region includes a CDK-associated phosphatase, KAP gene which was shown to dephosphorylate Thr160 in human Cdk2. The binding of cyclin A to Cdk2 inhibited the dephosphorylation of Thr160 by KAP. Moreover, the dephosphorylation of Thr160 by KAP prevented Cdk2 kinase activity upon subsequent association with cyclin A [18]. With this ®nding, deletion of KAP may lead the abnormality of cell cycle and the carcinogenesis of gallbladder carcinoma with AJPBD. The frequent LOH on 2p24 is the most interesting ®nding in this study. This has not been found previously in any malignancies and the candidate genes are not known. In summary, comprehensive allelotype study of gallbladder carcinoma with AJPBD was completed using microsatellite markers. This study suggests that another genetic alterations may play an important role in carcinogenesis in gallbladder carcinoma with AJPBD besides p53 and K-ras. At least two new sites of frequent LOH, which may harbor putative tumor suppressor genes were identi®ed, one is 2p24 and the other is 14q22. Further studies of ®ne mapping to localize the area of loss on these chromosomes are
required to identify tumor suppressor genes in gallbladder carcinoma associated with AJPBD. References [1] D.P. Babbitt, Congenital choledochal cyst: New etiological concept based on anomalous relationships of the common bile duct and pancreatic bulb, Ann. Radiol. 12 (1969) 231±240. [2] K. Kimura, M. Ohto, H. Saisho, T. Unozawa, Y. Tsuchiya, M. Morita, M. Ebara, S. Matsutani, K. Okuda, Association of gallbladder carcinoma and anomalous pancreaticobiliary ductal union, Gastroenterology 89 (1985) 1258±1265. [3] S. Yamauchi, A. Koda, S. Matsumoto, M. Tanaka, F. Nakayama, Anomalous junction of pancreaticobiliary duct without congenital choledochal cyst: a possible risk factor for gallbladder cancer, Am. J. Gastroenterol. 82 (1987) 20±24. [4] T. Ohta, T. Nagakawa, K. Ueno, K. Maeda, N. Ueda, M. Kayahara, T. Akiyama, M. Kanno, I. Konishi, R. Izumi, Clinical experience of biliary tract carcinoma associated with anomalous union of the pancreaticobiliary ductal system, Jpn. J. Surg. 20 (1990) 36±43. [5] M. Yamamoto, S. Nakajo, E. Tahara, M. Ito, K. Taniyama, F. Shimamoto, N. Miyoshi, Y. Hayashi, H. Akiyama, S. Nakai, Mucosal changes of the gallbladder in anomalous union with the pancreatico-biliary duct system, Pathol. Res. Pract. 187 (1991) 241±246. [6] K. Hanada, M. Itoh, K. Fujii, A. Tsuchida, M. Hirata, S. Ishimaru, T. Iwao, N. Eguchi, G. Kajiyama, Pathology and cellular kinetics of gallbladder with an anomalous junction of the pancreaticobiliary duct, Am. J. Gastroenterol. 91 (1996) 1007±1011. [7] K. Hanada, M. Itoh, K. Fujii, A. Tsuchida, H. Ooishi, G. Kajiyama, K-ras and p53 mutations in stage I gallbladder carcinoma with an anomalous junction of the pancreaticobiliary duct, Cancer 77 (1996) 452±458. [8] T. Matsubara, Y. Sakurai, Y. Sasayama, H. Hori, M. Ochiai, T. Funabiki, K. Matsumoto, I. Hirono, K-ras point mutations in cancerous and non-cancerous biliary epithelium in patients with pancreaticobiliary maljunction, Cancer 77 (1996) 1752± 1757. [9] K. Hanada, A. Tsuchida, T. Iwao, N. Eguchi, T. Sasaki, K. Morinaka, K. Matsubara, Y. Kawasaki, S. Yamamoto, G. Kajiyama, Gene mutations of K-ras in gallbladder mucosae and gallbladder carcinoma with an anomalous junction of the pancreaticobiliary duct, Am. J. Gastroenterol. 94 (1999) 1638±1642. [10] B. Vogelstein, E.R. Fearon, S.R. Hamilton, S.E. Kern, A.C. Preisinger, M. Leppert, Y. Nakamura, R. White, A.M. Smits, J.L. Bos, Genetic alterations during colorectal-tumor development, N. Engl. J. Med. 319 (1988) 525±532. [11] A.G. Kunudson, Antioncogenes and human cancer, Proc. Natl. Acad. Sci. USA 1 (1993) 10914±10921. [12] B. Vogelstein, E.R. Fearon, S.E. Kern, S.R. Hamilton, A.C. Preisinger,, Y. Nakamura, R. White, Allelotype of colorectal carcinomas, Science 244 (1989) 207±211. [13] M. Mandai, I. Konishi, H. Kuroda, T. Komatsu, S. Yamamoto,
K. Nakayama et al. / Cancer Letters 166 (2001) 135±141 K. Nanbu, K. Matsushita, M. Fukumoto, H. Yamabe, T. Mori, Heterogeneous distribution of K-ras-mutated epithelia in mucinous ovarian tumors with special reference to histopathology, Hum. Pathol. 29 (1998) 34±40. [14] M. Tada, M. Omata, M. Ohto, Analysis of ras gene mutations in human hepatic malignant tumors by polymerase chain reaction and direct sequencing, Cancer Res. 50 (1990) 1121±1124. [15] M. Watanabe, M. Asaka, J. Tanaka, M. Kurosawa, M. Kasai, T. Miyazaki, Point mutation of K-ras gene codon 12 in biliary tract tumors, Gastroenterology 107 (1994) 1147±1153. [16] I.I. Wistuba, K. Sugio, J. Hung, Y. Kishimoto, A.K. Virmani,
141
I. Roa, J. Albores-Saavedra, A.F. Gazdar, Allele-speci®c mutations involved in the pathogenesis of endemic gallbladder carcinoma in Chile, Cancer Res. 55 (1995) 2511±2515. [17] N. Yamamoto, K. Uzawa, T. Miya, T. Watanabe, H. Yokoe, T. Shibahara, H. Noma, H. Tanzawa, Frequent allelic loss/imbalance on the long arm of chromosome 21 in oral cancer: evidence for three discrete tumor suppressor gene loci, Oncol. Rep. 6 (1999) 1223±1227. [18] R.Y. Poon, T. Hunter, Dephosphorylation of Cdk2 Thr160 by the cyclin-dependent kinase- interacting phosphatase KAP in the absence of cyclin, Science 270 (1995) 90±93.