- Email: [email protected]
Somatic β-catenin mutation in gastric carcinoma – an infrequent event that is not specific for microsatellite instability
Cancer Letters 163 (2001) 125±130
www.elsevier.com/locate/canlet
Somatic b-catenin mutation in gastric carcinoma ± an infrequent event that is not speci®c for microsatellite instability J.H.M. Tong a, K.F. To a,*, E.K.W. Ng b, J.Y.W. Lau b, T.L. Lee a, K.W. Lo a, W.K. Leung c, N.L.S. Tang d, F.K.L. Chan c, J.J.Y. Sung c, S.C.S. Chung b a
Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, PR China b Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China c Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China d Department of Chemical Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China Received 2 September 2000; received in revised form 2 November 2000; accepted 4 November 2000
Abstract We screened 90 cases of gastric carcinoma (GCA) samples for b-catenin exon 3 mutation and assessed its possible relationship with microsatellite instability (MSI). Three mutations were detected in two samples, including a single mutation in an intestinal type and double mutations in a diffuse type GCA. One of the mutations found in the diffuse type GCA sample was a non-sense mutation at codon 68 (CAG ! TAG). This novel mutation was predicted to disrupt the binding of b-catenin to a-catenin and may be related to the diffuse type morphology. The other two mutations were missense mutations involved or related to the GSK-3b phosphorylation site, which have been reported previously. No MSI can be demonstrated in the two cases with b-catenin mutation. Our results suggested that b-catenin mutation was infrequent in GCA and appeared not speci®c for MSI. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Gastric carcinoma; b-Catenin; Microsatellite instability
1. Introduction b-Catenin was originally identi®ed as a cadherinbinding protein. In the adherens junction, b-catenin binds to the cytoplasmic domain of E-cadherin, providing a link to a-catenin and anchorage of the E-cadherin/catenin unit to actin cytoskeleton [1]. Complex formation of E-cadherin, b-catenin and acatenin is essential for the control and maintenance of normal intercellular adhesion. Therefore, cadherin* Corresponding author. Tel.: 1852-26322352; fax: 185226497286. E-mail address: [email protected] (K.F. To).
dependent cell±cell adhesion may be disrupted by abnormalities in one of the components of the complex. b-Catenin is also involved in growth signaling events, which is independent of the cadherin/catenin complex. It has been shown to be a downstream transcriptional activator of the Wnt signaling pathway, forming complexes with the DNA-binding proteins T-cell factor (Tcf) and lymphoid enhancerbinding factor (Lef-1) [2]. Constitutive activation of the b-catenin/Tcf complex caused by upregulation of b-catenin may be crucial in tumor development [3]. Free cytoplasmic b-catenin is targeted for degradation by the glycogen synthase kinase-3b (GSK-3b) and adenomatous polyposis coli (APC) proteins. Inac-
0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(00)00681-9
126
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
tivation of APC through mutation, deletion of the NH2 terminus, or mutation of one or more of the phosphorylation sites in exon 3 of the b-catenin gene can increase free b-catenin levels, thus enhancing its availability as a transcriptional activator. Activated b-catenin mutations have also been found to occur in several human cancers including colon cancer, melanomas and prostatic and endometrial cancer [3± 6]. Interestingly, b-catenin mutations were more common in colorectal tumors with microsatellite instability (MSI), which suggested that b-catenin mutations were relatively speci®c to the mismatch repair de®cient pathway of carcinogenesis in colorectal cancer [7±9]. Studies on the b-catenin mutation status in gastric cancer yielded discrepant results. One study revealed a lack of b-catenin mutation in 21 gastric carcinoma (GCA) samples [10]. However, another study from Korea showed that b-catenin mutations were detected in 26.9% (7/26 cases) of intestinal type, but not diffuse type GCA samples [11]. All of the mutations described so far were missense mutations involving exon 3 phosphorylation sites. The discrepancy of the results may be related to the relatively small sample size or the geographical±ethical difference. The possible relationship of b-catenin mutations and MSI has not been addressed. In the present study, we screened exon 3 of the b-catenin gene for mutation in a large cohort of 90 primary GCAs. We also examined MSI in these tumors and attempted to clarify the relationship between b-catenin mutation and MSI in gastric carcinogenesis.
2. Materials and methods 2.1. Patients and samples A total of 90 pairs of GCA and the corresponding normal tissue were obtained from patients at the Prince of Wales Hospital, Shatin, Hong Kong from 1984 to 1990. No patient had a positive family history. The mean age of the patients at the time of diagnosis was 58.3 years (range 29±82 years). The male to female ratio was 1.8:1. The formalin-®xed, paraf®nembedded tissues were retrieved and the hematoxylin and eosin stained sections were reviewed by a pathologist to con®rm the diagnosis. For each tumor speci-
men, ten pieces of 7 mm sections were microdissected manually for DNA extraction. Resected lymph node free of neoplastic invasion from the same patient was used as a normal control. DNA extraction was performed using a High Pure PCR Template Preparation Kit (Boehringer Mannheim) according to the manufacturer's protocol. 2.2. Immunohistochemistry Formalin-®xed, paraf®n-embedded tissue blocks of 61 gastric tumors were retrieved. An immunohistochemical study was carried out with the avidin/biotin complex immunoperoxidase technique using an antib-catenin mouse monoclonal IgG1 antibody raised against the 23 kDa carboxyl-terminal fragment of b-catenin corresponding to amino acid residues 571±781 (Transduction Laboratories, Lexington, KY). Paraf®n sections (4 mm) were cut and af®xed to 3aminopropyl triethoxysilane (APES, Sigma)-coated slides and air-dried overnight at 378C. The sections were then dewaxed in xylene, rehydrated through graded ethanol, treated with 0.3% hydrogen peroxide for 20 min to block the endogenous peroxidase activity prior to immersion in 0.01 M citrate buffer (pH 6.0) and microwaved for 15 min. After washing the sections in 0.01 M (pH 7.6) Tris-buffered saline (TBS), the sections were incubated with normal rabbit serum (DAKO) diluted at 1:20 for 10 min to block non-speci®c antibody binding. Sections were then incubated with primary antibody diluted at 1:1000 for 2 h at room temperature. After three separate washes with TBS, the sections were incubated with biotinylated rabbit anti-mouse (dilution 1:200, DAKO) for 45 min at room temperature followed by incubation with avidin/biotinylated peroxidase complex (dilution 1:100, DAKO) for 45 min at room temperature. The signal was developed using 3,3-diaminobenzidine tetrahydrochloride and counterstained with hematoxylin. 2.3. Microsatellite analysis DNA derived from microdissected tumor and normal tissue was analyzed using a panel of nine microsatellite markers, namely BAT25, BAT26, BAT40, BATRII, D1S158, D2S123, D5S346, D5S421 and D8S199. The primer sequences, anneal-
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
ing temperatures, heterozygosity frequency and estimated length of PCR products from each locus were obtained from the Genome Data Base (GDB). The samples were ampli®ed using 32P-labeled primers and processed according to the standard protocol [12]. MSI was demonstrated by the presence of new fragments of variable size in tumor DNA. The tumors were classi®ed as MSI-H if .30% of loci displayed MSI, MSI-L if ,30% of loci displayed MSI, or MSS if no locus displayed MSI. 2.4. PCR and DNA sequencing for b -catenin DNA sequences of the third coding exon of the b-catenin gene were ampli®ed using the forward 5 0 -primer GCTGATTTGATGGAGTTGGA and the reverse 3 0 -primer GCTACTTGTTCTTGAGTGAA. These ampli®ed a 227 bp fragment of exon 3 of the b-catenin gene encompassing the sequence for GSK3b phosphorylation. The PCR reaction was performed using the GeneAmp PCR system 9600 (Perkin± Elmer). Each PCR reaction was generally performed under standard conditions in a 25 ml reaction containing 50 ng of DNA derived from microdissected tumor tissue, 0.2 mM of each primer, 2 mM MgCl2, 0.2 mM of each dNTP, 0.4 units of AmpliTaq Gold (Perkin± Elmer) and 2.5 ml of 10 £ PCR buffer. The PCR conditions consisted of one cycle at 958C for 10 min, 40 cycles at 958C for 30 s, 508C for 1 min and 728C for 1 min, and a ®nal extension cycle at 728C for 5 min. The PCR products were puri®ed using a High Pure PCR Puri®cation Kit (Roche Diagnostics) and then sequenced using the dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (Perkin± Elmer). The sequencing reactions were run on an Applied Biosystems 310 Genetic Analyzer. The data were collected and analyzed using Applied Biosystems sequencing analysis software. Mutations were veri®ed by repeat ampli®cation and sequencing in both directions.
127
(54.4%) GCAs showed MSI at one or more loci. Forty-one of 90 (45.6%) gastric cancers were classi®ed as MSS, 31 (34.4%) cases were MSI-L and 18 (20%) cases were MSI-H. Aberrant cellular localization of b-catenin was investigated by immunohistochemistry in 61 tumors for which the paraf®n sections were available. Nineteen cases (19/61, 31.1%) showed immunoreactivity in nuclei and/or cytoplasm. In these cases, aberrant immunoreactivity was present in 10±50% of tumor cells. b-Catenin gene mutations were found in two (2.2%) GCAs tested. One of the patients (case 19) with diffuse type adenocarcinoma had double mutations. One of the mutations was a missense mutation at codon 33 (TCT ! TTT), changing the serine to
3. Results The GCAs were classi®ed into three histopathological types: intestinal, diffuse and mixed. In the 90 GCAs analyzed, 48 (53.3%) cases were intestinal type, 34 (37.8%) cases were diffuse type and 8 (8.9%) cases were mixed type. Overall, 49 of 90
Fig. 1. Electropherograms for the regions comprising codons 32 and 33 of b-catenin. (A) Wild-type sequences derived from a control sample. (B) Case 19, mutation at codon 33 (TCT ! TTT). (C) Case 102, mutation at codon 32 (GAC ! AAC). The affected nucleotides are indicated by an arrow.
128
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
Fig. 2. Electropherograms for the b-catenin sequences of case 19. (A) Wild-type sequences derived from normal control. (B) A nonsense mutation at codon 68 (CAG ! TAG) (arrowed).
phenylalanine (S33F) (Fig. 1), and the second mutation was a non-sense mutation at codon 68 (CAG ! TAG), leading to a premature translational stop (Fig. 2). The other patient (case 102) with intestinal type adenocarcinoma had a mutation at codon 32 (GAC ! AAC) (Fig. 1), changing a highly conserved amino acid, aspartic acid, to asparagine (D32N). Mutations were determined to be somatic because the normal DNA derived from the lymph node of the same patient contained a wild-type b-catenin DNA sequence in both cases. Both tumors showed no MSI in any of the loci tested. 4. Discussion b-Catenin is a multifunctional protein involved in both intercellular adhesion and signal transduction. In the APC-b-catenin -Tcf pathway, free cytoplasmic bcatenin behaves as an oncoprotein, which acquires oncogenic activity when it is mutated or when it is upregulated by inactivation of APC [13]. Somatic mutations in the b-catenin gene were found in colorectal and some other tumors [4±7]. Mutations described in primary tumors were involved or related
to the GSK-3b phosphorylation consensus motif in exon 3 of the b-catenin gene. These active mutations eliminate a GSK-3b phosphorylation site from b-catenin, resulting in a protein that is no longer regulated by APC, but continues to function as a transactivator when complexed with Tcf [13]. In our study, mutations in exon 3 of the b-catenin gene in GCA were rare. Of the 90 cases of primary GCA, only three mutations were found in two (2.2%) cases (case 19 contained two mutations). A mutation found in case 19 involved codon 33 (S33F), one of the important serine and threonine sites for GSK-3b phosphorylation. This mutation has been reported in various tumors, including endometroid carcinoma [6], prostatic cancer [5] and pilomatricoma [14]. The other case (case 102) contained a missense mutation at codon 32 (D32N). Codon 32 was not known to be phosphorylated. However, mutations at this site were frequently found in human tumors and druginduced rat colon tumors [5±7,11,15]. The mutation altered the amino acid residues immediately adjacent to the serine residue at codon 33. Substitution of a small amino acid (Asp32) for a charged residue (Asn32) may probably affect the phosphorylation by altering the recognition sequences or tertiary protein structure. It has been proposed that this residue is important for b-catenin ubiquitination and proteasome-dependent degradation [16]. Active mutations of b-catenin would lead to aberrant cytoplasmic and/or nuclear accumulation. However, such aberrant expressions were identi®ed in 19 out of 61 (31.1%) cases in this series. The presence of aberrant expressions without b-catenin active mutation suggested that disruptions in other candidate genes in b-catenin destruction machinery are likely involved in gastric carcinogenesis. The tumor suppressor protein APC, for example, is a critical component in b-catenin regulation. It has been reported that mutant APC proteins promote b-catenin accumulation within cytoplasm [3]. Many proteins that interact with APC in regulation b-catenin have been identi®ed, including GSK-3b and axin. Upregulation or nuclear translocation of b-catenin could hypothetically be achieved through alterations in these genes that function in this pathway. Interestingly, double mutations were detected in case 19, a diffuse type adenocarcinoma. In addition to the active mutation found in codon 33, there was a
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
non-sense mutation at codon 68, resulting in a premature translational stop. This mutation has not been reported before. By immunohistochemistry, overall downregulation of b-catenin was observed in this case with about 10% of the tumor cells exhibiting nuclear staining. b-Catenin plays a role in cell±cell adhesion by linking the cytoplasmic domain of Ecadherin to a-catenin, which anchors the adhesion complex to cytoskeleton [1]. It has been proposed that the a-catenin binding site might be located between codons 90 and 132 in the NH2 terminus of b-catenin [11]. Cells lacking an a-catenin binding site within b-catenin are unable to form stable adheren junctions despite normal E-cadherin and b-catenin expression. For example, there was a 321 bp inframe deletion in the NH2 terminus of b-catenin in two cell lines (HSC-39 and HSC-40) established from signet ring cell carcinoma of the stomach. This truncated b-catenin molecule failed to link E-cadherin and a-catenin and resulted in a non-functional Ecadherin complex [17]. The second mutation detected in case 19 results in a premature translational stop at codon 68. The resulting truncated proteins are predicted to be non-functional in the linkage between E-cadherin and a-catenin, as they lack the a-catenin binding domain at the NH2 terminus. The altered Ecadherin/catenin complex may contribute to the loose cell±cell adhesion pattern shown in diffuse type adenocarcinoma. Further studies on the E-cadherin/ catenin system are required to con®rm our hypothesis. Genetic instability, which is recognized by the presence of replication error (RER) detected by microsatellite analysis, is characteristic of colorectal carcinomas in hereditary non-polyposis colorectal cancer families but is also found in sporadic colorectal and gastric cancer. We identi®ed MSI-H in 20% of GCA samples, which is in keeping with the previous reports of 15±38% [18±20]. Previous studies have found a higher frequency of b-catenin mutation in MSI-H colorectal cancers and thereby suggested that b-catenin mutation was relatively speci®c to MSI [7,8]. However, in our study, both gastric tumors with b-catenin mutations were classi®ed as microsatellite stable. A similar ®nding has also been reported in endometrial carcinoma, in which the b-catenin mutation status was independent of MSI [7]. The discrepancy in the mutation pro®les may re¯ect the different molecular pathways
129
involved in the carcinogenesis of colonic cancer and gastric cancer. In summary, b-catenin mutations appeared infrequent in GCA, and only a small subset of tumors showed b-catenin mutations. The detected mutations affect highly conserved serine phosphorylation sites involved in b-catenin degradation and hence could lead to b-catenin accumulation and transduction of oncogenic signals. In addition to the active mutations found in the GSK-3b phosphorylation consensus region, a coexisting non-sense mutation was identi®ed in one of the diffuse type GCAs. The resulting truncated b-catenin that lacked an a-catenin binding site may produce a non-functional E-cadherin/catenin complex, which may contribute to the impaired cell adhesion in diffuse type tumor. On the other hand, both gastric tumors with b-catenin mutations in our study were the MSS type. In contrast to the published data on colorectal tumors, there is no evidence that bcatenin mutation is speci®c to MSI in gastric cancer. References [1] H. Aberle, S. Butz, J. Stappert, H. Weissig, R. Kemler, H. Hoschuetzky, Assembly of the cadherin-catenin complex in vitro with recombinant proteins, J. Cell Sci. 107 (Pt 12) (1994) 3655±3663. [2] A.I. Barth, I.S. Nathke, W.J. Nelson, Cadherins, catenins and APC protein: interplay between cytoskeletal complexes and signaling pathways, Curr. Opin. Cell Biol. 9 (5) (1997) 683± 690. [3] P.J. Morin, A.B. Sparks, V. Korinek, N. Barker, H. Clevers, B. Vogelstein, K.W. Kinzler, Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC (see comments), Science 275 (5307) (1997) 1787±1790. [4] B. Rubinfeld, P. Robbins, M. El-Gamil, I. Albert, E. Por®ri, P. Polakis, Stabilization of beta-catenin by genetic defects in melanoma cell lines (see comments), Science 275 (5307) (1997) 1790±1792. [5] H.J. Voeller, C.I. Truica, E.P. Gelmann, Beta-catenin mutations in human prostate cancer, Cancer Res. 58 (12) (1998) 2520±2523. [6] T. Fukuchi, M. Sakamoto, H. Tsuda, K. Maruyama, S. Nozawa, S. Hirohashi, Beta-catenin mutation in carcinoma of the uterine endometrium, Cancer Res. 58 (16) (1998) 3526±3528. [7] L. Mirabelli-Primdahl, R. Gryfe, H. Kim, A. Millar, C. Luceri, D. Dale, E. Holowaty, B. Bapat, S. Gallinger, M. Redston, Beta-catenin mutations are speci®c for colorectal carcinomas with microsatellite instability but occur in endometrial carcinomas irrespective of mutator pathway, Cancer Res. 59 (14) (1999) 3346±3351.
130
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
[8] M.N. Kitaeva, L. Grogan, J.P. Williams, E. Dimond, K. Nakahara, P. Housner, J.W. Denobile, P.W. Soballe, I.R. Kirsch, Mutations in beta-catenin are uncommon in colorectal cancer occurring in occasional replication error-positive tumors, Cancer Res. 57 (20) (1997) 4478±4481. [9] O. Muller, I. Nimmrich, U. Finke, W. Friedl, I. Hoffmann, A beta-catenin mutation in a sporadic colorectal tumor of the RER phenotype and absence of beta-catenin germline mutations in FAP patients, Genes Chromosomes Cancer 22 (1) (1998) 37±41. [10] S. Candidus, P. Bischoff, K.F. Becker, H. Ho¯er, No evidence for mutations in the alpha- and beta-catenin genes in human gastric and breast carcinomas, Cancer Res. 56 (1) (1996) 49± 52. [11] W.S. Park, R.R. Oh, J.Y. Park, S.H. Lee, M.S. Shin, Y.S. Kim, S.Y. Kim, H.K. Lee, P.J. Kim, S.T. Oh, N.J. Yoo, J.Y. Lee, Frequent somatic mutations of the beta-catenin gene in intestinal-type gastric cancer, Cancer Res. 59 (17) (1999) 4257± 4260. [12] L. Mao, D.J. Lee, M.S. Tockman, Y.S. Erozan, F. Askin, D. Sidransky, Microsatellite alterations as clonal markers for the detection of human cancer, Proc. Natl. Acad. Sci. USA 91 (21) (1994) 9871±9875. [13] Y. Nakamura, Cleaning up on beta-catenin (news), Nat. Med. 3 (5) (1997) 499±500.
[14] E.F. Chan, U. Gat, J.M. McNiff, E. Fuchs, A common human skin tumour is caused by activating mutations in beta-catenin, Nat. Genet. 21 (4) (1999) 410±413. [15] R.H. Dashwood, M. Suzui, H. Nakagama, T. Sugimura, M. Nagao, High frequency of beta-catenin (ctnnb1) mutations in the colon tumors induced by two heterocyclic amines in the F344 rat, Cancer Res. 58 (6) (1998) 1127±1129. [16] H. Aberle, A. Bauer, J. Stappert, A. Kispert, R. Kemler, Betacatenin is a target for the ubiquitin-proteasome pathway, EMBO J. 16 (13) (1997) 3797±3804. [17] J. Kawanishi, J. Kato, K. Sasaki, S. Fujii, N. Watanabe, Y. Niitsu, Loss of E-cadherin-dependent cell-cell adhesion due to mutation of the beta-catenin gene in a human cancer cell line, HSC-39, Mol. Cell. Biol. 15 (3) (1995) 1175±1181. [18] H.J. Han, A. Yanagisawa, Y. Kato, J.G. Park, Y. Nakamura, Genetic instability in pancreatic cancer and poorly differentiated type of gastric cancer, Cancer Res. 53 (21) (1993) 5087± 5089. [19] M.G. Rhyu, W.S. Park, S.J. Meltzer, Microsatellite instability occurs frequently in human gastric carcinoma, Oncogene 9 (1) (1994) 29±32. [20] J.M. Chong, M. Fukayama, Y. Hayashi, T. Takizawa, M. Koike, M. Konishi, R. Kikuchi-Yanoshita, M. Miyaki, Microsatellite instability in the progression of gastric carcinoma, Cancer Res. 54 (17) (1994) 4595±4597.
www.elsevier.com/locate/canlet
Somatic b-catenin mutation in gastric carcinoma ± an infrequent event that is not speci®c for microsatellite instability J.H.M. Tong a, K.F. To a,*, E.K.W. Ng b, J.Y.W. Lau b, T.L. Lee a, K.W. Lo a, W.K. Leung c, N.L.S. Tang d, F.K.L. Chan c, J.J.Y. Sung c, S.C.S. Chung b a
Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, PR China b Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China c Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China d Department of Chemical Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, PR China Received 2 September 2000; received in revised form 2 November 2000; accepted 4 November 2000
Abstract We screened 90 cases of gastric carcinoma (GCA) samples for b-catenin exon 3 mutation and assessed its possible relationship with microsatellite instability (MSI). Three mutations were detected in two samples, including a single mutation in an intestinal type and double mutations in a diffuse type GCA. One of the mutations found in the diffuse type GCA sample was a non-sense mutation at codon 68 (CAG ! TAG). This novel mutation was predicted to disrupt the binding of b-catenin to a-catenin and may be related to the diffuse type morphology. The other two mutations were missense mutations involved or related to the GSK-3b phosphorylation site, which have been reported previously. No MSI can be demonstrated in the two cases with b-catenin mutation. Our results suggested that b-catenin mutation was infrequent in GCA and appeared not speci®c for MSI. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Gastric carcinoma; b-Catenin; Microsatellite instability
1. Introduction b-Catenin was originally identi®ed as a cadherinbinding protein. In the adherens junction, b-catenin binds to the cytoplasmic domain of E-cadherin, providing a link to a-catenin and anchorage of the E-cadherin/catenin unit to actin cytoskeleton [1]. Complex formation of E-cadherin, b-catenin and acatenin is essential for the control and maintenance of normal intercellular adhesion. Therefore, cadherin* Corresponding author. Tel.: 1852-26322352; fax: 185226497286. E-mail address: [email protected] (K.F. To).
dependent cell±cell adhesion may be disrupted by abnormalities in one of the components of the complex. b-Catenin is also involved in growth signaling events, which is independent of the cadherin/catenin complex. It has been shown to be a downstream transcriptional activator of the Wnt signaling pathway, forming complexes with the DNA-binding proteins T-cell factor (Tcf) and lymphoid enhancerbinding factor (Lef-1) [2]. Constitutive activation of the b-catenin/Tcf complex caused by upregulation of b-catenin may be crucial in tumor development [3]. Free cytoplasmic b-catenin is targeted for degradation by the glycogen synthase kinase-3b (GSK-3b) and adenomatous polyposis coli (APC) proteins. Inac-
0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(00)00681-9
126
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
tivation of APC through mutation, deletion of the NH2 terminus, or mutation of one or more of the phosphorylation sites in exon 3 of the b-catenin gene can increase free b-catenin levels, thus enhancing its availability as a transcriptional activator. Activated b-catenin mutations have also been found to occur in several human cancers including colon cancer, melanomas and prostatic and endometrial cancer [3± 6]. Interestingly, b-catenin mutations were more common in colorectal tumors with microsatellite instability (MSI), which suggested that b-catenin mutations were relatively speci®c to the mismatch repair de®cient pathway of carcinogenesis in colorectal cancer [7±9]. Studies on the b-catenin mutation status in gastric cancer yielded discrepant results. One study revealed a lack of b-catenin mutation in 21 gastric carcinoma (GCA) samples [10]. However, another study from Korea showed that b-catenin mutations were detected in 26.9% (7/26 cases) of intestinal type, but not diffuse type GCA samples [11]. All of the mutations described so far were missense mutations involving exon 3 phosphorylation sites. The discrepancy of the results may be related to the relatively small sample size or the geographical±ethical difference. The possible relationship of b-catenin mutations and MSI has not been addressed. In the present study, we screened exon 3 of the b-catenin gene for mutation in a large cohort of 90 primary GCAs. We also examined MSI in these tumors and attempted to clarify the relationship between b-catenin mutation and MSI in gastric carcinogenesis.
2. Materials and methods 2.1. Patients and samples A total of 90 pairs of GCA and the corresponding normal tissue were obtained from patients at the Prince of Wales Hospital, Shatin, Hong Kong from 1984 to 1990. No patient had a positive family history. The mean age of the patients at the time of diagnosis was 58.3 years (range 29±82 years). The male to female ratio was 1.8:1. The formalin-®xed, paraf®nembedded tissues were retrieved and the hematoxylin and eosin stained sections were reviewed by a pathologist to con®rm the diagnosis. For each tumor speci-
men, ten pieces of 7 mm sections were microdissected manually for DNA extraction. Resected lymph node free of neoplastic invasion from the same patient was used as a normal control. DNA extraction was performed using a High Pure PCR Template Preparation Kit (Boehringer Mannheim) according to the manufacturer's protocol. 2.2. Immunohistochemistry Formalin-®xed, paraf®n-embedded tissue blocks of 61 gastric tumors were retrieved. An immunohistochemical study was carried out with the avidin/biotin complex immunoperoxidase technique using an antib-catenin mouse monoclonal IgG1 antibody raised against the 23 kDa carboxyl-terminal fragment of b-catenin corresponding to amino acid residues 571±781 (Transduction Laboratories, Lexington, KY). Paraf®n sections (4 mm) were cut and af®xed to 3aminopropyl triethoxysilane (APES, Sigma)-coated slides and air-dried overnight at 378C. The sections were then dewaxed in xylene, rehydrated through graded ethanol, treated with 0.3% hydrogen peroxide for 20 min to block the endogenous peroxidase activity prior to immersion in 0.01 M citrate buffer (pH 6.0) and microwaved for 15 min. After washing the sections in 0.01 M (pH 7.6) Tris-buffered saline (TBS), the sections were incubated with normal rabbit serum (DAKO) diluted at 1:20 for 10 min to block non-speci®c antibody binding. Sections were then incubated with primary antibody diluted at 1:1000 for 2 h at room temperature. After three separate washes with TBS, the sections were incubated with biotinylated rabbit anti-mouse (dilution 1:200, DAKO) for 45 min at room temperature followed by incubation with avidin/biotinylated peroxidase complex (dilution 1:100, DAKO) for 45 min at room temperature. The signal was developed using 3,3-diaminobenzidine tetrahydrochloride and counterstained with hematoxylin. 2.3. Microsatellite analysis DNA derived from microdissected tumor and normal tissue was analyzed using a panel of nine microsatellite markers, namely BAT25, BAT26, BAT40, BATRII, D1S158, D2S123, D5S346, D5S421 and D8S199. The primer sequences, anneal-
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
ing temperatures, heterozygosity frequency and estimated length of PCR products from each locus were obtained from the Genome Data Base (GDB). The samples were ampli®ed using 32P-labeled primers and processed according to the standard protocol [12]. MSI was demonstrated by the presence of new fragments of variable size in tumor DNA. The tumors were classi®ed as MSI-H if .30% of loci displayed MSI, MSI-L if ,30% of loci displayed MSI, or MSS if no locus displayed MSI. 2.4. PCR and DNA sequencing for b -catenin DNA sequences of the third coding exon of the b-catenin gene were ampli®ed using the forward 5 0 -primer GCTGATTTGATGGAGTTGGA and the reverse 3 0 -primer GCTACTTGTTCTTGAGTGAA. These ampli®ed a 227 bp fragment of exon 3 of the b-catenin gene encompassing the sequence for GSK3b phosphorylation. The PCR reaction was performed using the GeneAmp PCR system 9600 (Perkin± Elmer). Each PCR reaction was generally performed under standard conditions in a 25 ml reaction containing 50 ng of DNA derived from microdissected tumor tissue, 0.2 mM of each primer, 2 mM MgCl2, 0.2 mM of each dNTP, 0.4 units of AmpliTaq Gold (Perkin± Elmer) and 2.5 ml of 10 £ PCR buffer. The PCR conditions consisted of one cycle at 958C for 10 min, 40 cycles at 958C for 30 s, 508C for 1 min and 728C for 1 min, and a ®nal extension cycle at 728C for 5 min. The PCR products were puri®ed using a High Pure PCR Puri®cation Kit (Roche Diagnostics) and then sequenced using the dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (Perkin± Elmer). The sequencing reactions were run on an Applied Biosystems 310 Genetic Analyzer. The data were collected and analyzed using Applied Biosystems sequencing analysis software. Mutations were veri®ed by repeat ampli®cation and sequencing in both directions.
127
(54.4%) GCAs showed MSI at one or more loci. Forty-one of 90 (45.6%) gastric cancers were classi®ed as MSS, 31 (34.4%) cases were MSI-L and 18 (20%) cases were MSI-H. Aberrant cellular localization of b-catenin was investigated by immunohistochemistry in 61 tumors for which the paraf®n sections were available. Nineteen cases (19/61, 31.1%) showed immunoreactivity in nuclei and/or cytoplasm. In these cases, aberrant immunoreactivity was present in 10±50% of tumor cells. b-Catenin gene mutations were found in two (2.2%) GCAs tested. One of the patients (case 19) with diffuse type adenocarcinoma had double mutations. One of the mutations was a missense mutation at codon 33 (TCT ! TTT), changing the serine to
3. Results The GCAs were classi®ed into three histopathological types: intestinal, diffuse and mixed. In the 90 GCAs analyzed, 48 (53.3%) cases were intestinal type, 34 (37.8%) cases were diffuse type and 8 (8.9%) cases were mixed type. Overall, 49 of 90
Fig. 1. Electropherograms for the regions comprising codons 32 and 33 of b-catenin. (A) Wild-type sequences derived from a control sample. (B) Case 19, mutation at codon 33 (TCT ! TTT). (C) Case 102, mutation at codon 32 (GAC ! AAC). The affected nucleotides are indicated by an arrow.
128
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
Fig. 2. Electropherograms for the b-catenin sequences of case 19. (A) Wild-type sequences derived from normal control. (B) A nonsense mutation at codon 68 (CAG ! TAG) (arrowed).
phenylalanine (S33F) (Fig. 1), and the second mutation was a non-sense mutation at codon 68 (CAG ! TAG), leading to a premature translational stop (Fig. 2). The other patient (case 102) with intestinal type adenocarcinoma had a mutation at codon 32 (GAC ! AAC) (Fig. 1), changing a highly conserved amino acid, aspartic acid, to asparagine (D32N). Mutations were determined to be somatic because the normal DNA derived from the lymph node of the same patient contained a wild-type b-catenin DNA sequence in both cases. Both tumors showed no MSI in any of the loci tested. 4. Discussion b-Catenin is a multifunctional protein involved in both intercellular adhesion and signal transduction. In the APC-b-catenin -Tcf pathway, free cytoplasmic bcatenin behaves as an oncoprotein, which acquires oncogenic activity when it is mutated or when it is upregulated by inactivation of APC [13]. Somatic mutations in the b-catenin gene were found in colorectal and some other tumors [4±7]. Mutations described in primary tumors were involved or related
to the GSK-3b phosphorylation consensus motif in exon 3 of the b-catenin gene. These active mutations eliminate a GSK-3b phosphorylation site from b-catenin, resulting in a protein that is no longer regulated by APC, but continues to function as a transactivator when complexed with Tcf [13]. In our study, mutations in exon 3 of the b-catenin gene in GCA were rare. Of the 90 cases of primary GCA, only three mutations were found in two (2.2%) cases (case 19 contained two mutations). A mutation found in case 19 involved codon 33 (S33F), one of the important serine and threonine sites for GSK-3b phosphorylation. This mutation has been reported in various tumors, including endometroid carcinoma [6], prostatic cancer [5] and pilomatricoma [14]. The other case (case 102) contained a missense mutation at codon 32 (D32N). Codon 32 was not known to be phosphorylated. However, mutations at this site were frequently found in human tumors and druginduced rat colon tumors [5±7,11,15]. The mutation altered the amino acid residues immediately adjacent to the serine residue at codon 33. Substitution of a small amino acid (Asp32) for a charged residue (Asn32) may probably affect the phosphorylation by altering the recognition sequences or tertiary protein structure. It has been proposed that this residue is important for b-catenin ubiquitination and proteasome-dependent degradation [16]. Active mutations of b-catenin would lead to aberrant cytoplasmic and/or nuclear accumulation. However, such aberrant expressions were identi®ed in 19 out of 61 (31.1%) cases in this series. The presence of aberrant expressions without b-catenin active mutation suggested that disruptions in other candidate genes in b-catenin destruction machinery are likely involved in gastric carcinogenesis. The tumor suppressor protein APC, for example, is a critical component in b-catenin regulation. It has been reported that mutant APC proteins promote b-catenin accumulation within cytoplasm [3]. Many proteins that interact with APC in regulation b-catenin have been identi®ed, including GSK-3b and axin. Upregulation or nuclear translocation of b-catenin could hypothetically be achieved through alterations in these genes that function in this pathway. Interestingly, double mutations were detected in case 19, a diffuse type adenocarcinoma. In addition to the active mutation found in codon 33, there was a
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
non-sense mutation at codon 68, resulting in a premature translational stop. This mutation has not been reported before. By immunohistochemistry, overall downregulation of b-catenin was observed in this case with about 10% of the tumor cells exhibiting nuclear staining. b-Catenin plays a role in cell±cell adhesion by linking the cytoplasmic domain of Ecadherin to a-catenin, which anchors the adhesion complex to cytoskeleton [1]. It has been proposed that the a-catenin binding site might be located between codons 90 and 132 in the NH2 terminus of b-catenin [11]. Cells lacking an a-catenin binding site within b-catenin are unable to form stable adheren junctions despite normal E-cadherin and b-catenin expression. For example, there was a 321 bp inframe deletion in the NH2 terminus of b-catenin in two cell lines (HSC-39 and HSC-40) established from signet ring cell carcinoma of the stomach. This truncated b-catenin molecule failed to link E-cadherin and a-catenin and resulted in a non-functional Ecadherin complex [17]. The second mutation detected in case 19 results in a premature translational stop at codon 68. The resulting truncated proteins are predicted to be non-functional in the linkage between E-cadherin and a-catenin, as they lack the a-catenin binding domain at the NH2 terminus. The altered Ecadherin/catenin complex may contribute to the loose cell±cell adhesion pattern shown in diffuse type adenocarcinoma. Further studies on the E-cadherin/ catenin system are required to con®rm our hypothesis. Genetic instability, which is recognized by the presence of replication error (RER) detected by microsatellite analysis, is characteristic of colorectal carcinomas in hereditary non-polyposis colorectal cancer families but is also found in sporadic colorectal and gastric cancer. We identi®ed MSI-H in 20% of GCA samples, which is in keeping with the previous reports of 15±38% [18±20]. Previous studies have found a higher frequency of b-catenin mutation in MSI-H colorectal cancers and thereby suggested that b-catenin mutation was relatively speci®c to MSI [7,8]. However, in our study, both gastric tumors with b-catenin mutations were classi®ed as microsatellite stable. A similar ®nding has also been reported in endometrial carcinoma, in which the b-catenin mutation status was independent of MSI [7]. The discrepancy in the mutation pro®les may re¯ect the different molecular pathways
129
involved in the carcinogenesis of colonic cancer and gastric cancer. In summary, b-catenin mutations appeared infrequent in GCA, and only a small subset of tumors showed b-catenin mutations. The detected mutations affect highly conserved serine phosphorylation sites involved in b-catenin degradation and hence could lead to b-catenin accumulation and transduction of oncogenic signals. In addition to the active mutations found in the GSK-3b phosphorylation consensus region, a coexisting non-sense mutation was identi®ed in one of the diffuse type GCAs. The resulting truncated b-catenin that lacked an a-catenin binding site may produce a non-functional E-cadherin/catenin complex, which may contribute to the impaired cell adhesion in diffuse type tumor. On the other hand, both gastric tumors with b-catenin mutations in our study were the MSS type. In contrast to the published data on colorectal tumors, there is no evidence that bcatenin mutation is speci®c to MSI in gastric cancer. References [1] H. Aberle, S. Butz, J. Stappert, H. Weissig, R. Kemler, H. Hoschuetzky, Assembly of the cadherin-catenin complex in vitro with recombinant proteins, J. Cell Sci. 107 (Pt 12) (1994) 3655±3663. [2] A.I. Barth, I.S. Nathke, W.J. Nelson, Cadherins, catenins and APC protein: interplay between cytoskeletal complexes and signaling pathways, Curr. Opin. Cell Biol. 9 (5) (1997) 683± 690. [3] P.J. Morin, A.B. Sparks, V. Korinek, N. Barker, H. Clevers, B. Vogelstein, K.W. Kinzler, Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC (see comments), Science 275 (5307) (1997) 1787±1790. [4] B. Rubinfeld, P. Robbins, M. El-Gamil, I. Albert, E. Por®ri, P. Polakis, Stabilization of beta-catenin by genetic defects in melanoma cell lines (see comments), Science 275 (5307) (1997) 1790±1792. [5] H.J. Voeller, C.I. Truica, E.P. Gelmann, Beta-catenin mutations in human prostate cancer, Cancer Res. 58 (12) (1998) 2520±2523. [6] T. Fukuchi, M. Sakamoto, H. Tsuda, K. Maruyama, S. Nozawa, S. Hirohashi, Beta-catenin mutation in carcinoma of the uterine endometrium, Cancer Res. 58 (16) (1998) 3526±3528. [7] L. Mirabelli-Primdahl, R. Gryfe, H. Kim, A. Millar, C. Luceri, D. Dale, E. Holowaty, B. Bapat, S. Gallinger, M. Redston, Beta-catenin mutations are speci®c for colorectal carcinomas with microsatellite instability but occur in endometrial carcinomas irrespective of mutator pathway, Cancer Res. 59 (14) (1999) 3346±3351.
130
J.H.M. Tong et al. / Cancer Letters 163 (2001) 125±130
[8] M.N. Kitaeva, L. Grogan, J.P. Williams, E. Dimond, K. Nakahara, P. Housner, J.W. Denobile, P.W. Soballe, I.R. Kirsch, Mutations in beta-catenin are uncommon in colorectal cancer occurring in occasional replication error-positive tumors, Cancer Res. 57 (20) (1997) 4478±4481. [9] O. Muller, I. Nimmrich, U. Finke, W. Friedl, I. Hoffmann, A beta-catenin mutation in a sporadic colorectal tumor of the RER phenotype and absence of beta-catenin germline mutations in FAP patients, Genes Chromosomes Cancer 22 (1) (1998) 37±41. [10] S. Candidus, P. Bischoff, K.F. Becker, H. Ho¯er, No evidence for mutations in the alpha- and beta-catenin genes in human gastric and breast carcinomas, Cancer Res. 56 (1) (1996) 49± 52. [11] W.S. Park, R.R. Oh, J.Y. Park, S.H. Lee, M.S. Shin, Y.S. Kim, S.Y. Kim, H.K. Lee, P.J. Kim, S.T. Oh, N.J. Yoo, J.Y. Lee, Frequent somatic mutations of the beta-catenin gene in intestinal-type gastric cancer, Cancer Res. 59 (17) (1999) 4257± 4260. [12] L. Mao, D.J. Lee, M.S. Tockman, Y.S. Erozan, F. Askin, D. Sidransky, Microsatellite alterations as clonal markers for the detection of human cancer, Proc. Natl. Acad. Sci. USA 91 (21) (1994) 9871±9875. [13] Y. Nakamura, Cleaning up on beta-catenin (news), Nat. Med. 3 (5) (1997) 499±500.
[14] E.F. Chan, U. Gat, J.M. McNiff, E. Fuchs, A common human skin tumour is caused by activating mutations in beta-catenin, Nat. Genet. 21 (4) (1999) 410±413. [15] R.H. Dashwood, M. Suzui, H. Nakagama, T. Sugimura, M. Nagao, High frequency of beta-catenin (ctnnb1) mutations in the colon tumors induced by two heterocyclic amines in the F344 rat, Cancer Res. 58 (6) (1998) 1127±1129. [16] H. Aberle, A. Bauer, J. Stappert, A. Kispert, R. Kemler, Betacatenin is a target for the ubiquitin-proteasome pathway, EMBO J. 16 (13) (1997) 3797±3804. [17] J. Kawanishi, J. Kato, K. Sasaki, S. Fujii, N. Watanabe, Y. Niitsu, Loss of E-cadherin-dependent cell-cell adhesion due to mutation of the beta-catenin gene in a human cancer cell line, HSC-39, Mol. Cell. Biol. 15 (3) (1995) 1175±1181. [18] H.J. Han, A. Yanagisawa, Y. Kato, J.G. Park, Y. Nakamura, Genetic instability in pancreatic cancer and poorly differentiated type of gastric cancer, Cancer Res. 53 (21) (1993) 5087± 5089. [19] M.G. Rhyu, W.S. Park, S.J. Meltzer, Microsatellite instability occurs frequently in human gastric carcinoma, Oncogene 9 (1) (1994) 29±32. [20] J.M. Chong, M. Fukayama, Y. Hayashi, T. Takizawa, M. Koike, M. Konishi, R. Kikuchi-Yanoshita, M. Miyaki, Microsatellite instability in the progression of gastric carcinoma, Cancer Res. 54 (17) (1994) 4595±4597.