Absence of β-Catenin Alteration in Hepatic Tumors Induced by p-Nitroanisole in Crj:BDF1 Mice

Abstract

In the present study, β-catenin localization in hepatocellular neoplasms and hepatoblastomas, induced by oral administration of p-Nitroanisole (pNA) in Crj:BDF1 for 2 years, was evaluated by immunohistochemistry along with genetic alterations in exon 2 of β-catenin by the polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) approach. Genomic DNA was isolated from paraffin sections of a total of 53 liver tumors. Immunohistochemical analysis revealed no abnormal accumulation of the β-catenin protein in any of the cases. No mutations (0/13), 20% silent mutations (3/15) and 8% silent plus 12% functional mutations (2 + 3/25), not in the multiple phosphorylation sites of β-catenin, were observed in hepatocellular adenomas, carcinomas and hepatoblastomas, respectively. The results indicate that β-catenin does not play an important role in development of hepatic tumors induced by pNA in Crj:BDF1 mice.

Introduction

β-catenin is an essential contributor to Wnt signaling which plays roles in intercellular adhesion, signal transduction and gene transcription (Cadigan and Nusse, 1997; Nelson and Nusse, 2004), acting in concert with other components of the pathway, such as adenomatous polyposis coli (APC) protein, glycogen synthase kinase 3β (GSK-3β) and Axin (Munemitsu et al., 1995; Rubinfeld et al., 1996; Aberle et al., 1997; Hart et al., 1998; Ikeda et al., 1998). There are multiple phosphorylation sites (codons 33, 37, 41 and 45) in exon 2 (corresponding to exon 3 in man (Nollet et al., 1996)) of theβ-catenin gene in the mouse that prevent the normal degradation and result in enhanced signaling (Kim et al., 2005). Immunohistochemical analysis of β-catenin has revealed accumulation of β-catenin in the cytoplasm and/or nucleus in tumors, instead of at the cell membrane, this occurring as a result of functional mutations in APC (Morin et al., 1997; Munemitsu et al., 1995; Rubinfeld et al., 1997) and Axin (Kikuchi, 2000) as well as β-catenin (Morin et al., 1997; Rubinfeld et al., 1997).

β-catenin mutations in human hepatocellular carcinomas (HCCs) are reported to be relatively frequent (Nhieu et al., 1999). In hepatoblastomas (HBs), the most common malignant hepatic neoplasms in children (Koesters and von Knebel Doeberitz, 2003), protein accumulation in the cytoplasm and/or nucleus has been reported to highly correlate with somatic mutations of β-catenin (Blaker et al., 1999; Jeng et al., 2000; Wei et al., 2000; Park et al., 2001; Koch et al., 2004; Udatsu et al., 2001). In the mouse, relationship between alterations in the β-catenin signaling pathway and chemically induced hepatocellular neoplasms and HBs has also been investigated, but the results have been conflicting. Chemically induced hepatocellular adenomas (HCAs) and HCCs in 2 studies had mutations in exon 2 of β-catenin (Hayashi et al., 2003; Huang et al., 2003), but were lacking in another investigation (Aydinlik et al., 2001). Concerning HBs, there have been many reports of high percentages of gene mutations and abnormal localization of β-catenin (de La Coste et al., 1998; Devereux et al., 1999; Anna et al., 2000, 2003; Hayashi et al., 2003), but we earlier found no change in N,N-Dimethylformamide-induced HBs in mice (Kakuni et al., 2004).

Recently we obtained interesting samples of hepatocellular neoplasms in a 104-week mouse study of the carcinogenicity of p-Nitroanisole (pNA), an intermediate used in the manufacture of p-Anisidine, a raw material for dyes and pharmaceuticals, directed by the Japan Bioassay Research Center (Japan Bioassay Research Center, 2005). Production of pNA in Western Europe was 5,800 tons in 1978 and 5,300 tons in 1983 (GDCh-Advisory Commitiee on Existing Chemicals of Enviromnental Relevance, 1993) and global production is now about 10,000 t/year, Japan importing about 400 tons in 2002 (Japan Bioassay Research Center, 2005; GDCh-Advisory Committee on Existing Chemicals of Enviromnental Relevance, 1993). In the Ames test pNA is mutagenic (Shimizu and Yano, 1986) and its LD₅₀ with oral administration is about 1.7 g/kg in mice and 2.6 g/kg in rats (GDCh-Advisory Commitiee on Existing Chemicals of Enviromnental Relevance, 1993).

The present study was conducted to assess whether β-catenin mutations and protein accumulation might be features of pNA-induced mouse hepatocellular neoplasms, with a combined immunohistochemical and polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) approach.

Materials and Methods

Tumor Samples

Samples of mouse liver tumors, induced in a 104 week mouse carcinogenicity study of pNA, were obtained from Japan Bioassay Research Center as paraffin-embedded tissue. Briefly, the study employed a total of four-hundred Crj:BDF1 mice (Males: 200, Females: 200), purchased from Charles River Japan (Kanagawa, Japan) and divided into 4 weight-matched groups, each consisting of 50 mice of both sexes. Three different dose levels of 5000, 10000 and 20000 ppm pNA in γ-irradiation-sterilized CRF-1 powdered diet (Oriental Yeast Co., Tokyo, Japan) were administered at libitum for 104 weeks. Livers were then removed from the euthanized mice, fixed with 10% neutral-buffered formalin and routinely processed for paraffin embedding. Histological diagnoses were determined with 4-μm-thick H&E stained sections (Japan Bioassay Research Center, 2005). We selected typical non-necrotic and non-autolytic neoplasms from scheduled or moribund sacrifice male animals in pNA treatment groups for the present analysis. The numbers of HCAs, HCCs and HBs were 13, 15, and 25, respectively. Mice were cared in accordance with Guide for the Care and Use of Laboratory Animals, and the study was approved by the ethics committee of the Japan Bioassay Research Center.

Immunohistochemistry

The localization of β-catenin protein was detected with an anti- mouse β-catenin antibody (Clone: 14, Transduction laboratories Inc., Lexington, KY, USA), and a VECTASTAIN elite ABC kit (Vector Laboratories, Inc., Burlingame, CA, USA). Normal liver tissue, adjacent to lesions, was employed as the positive control for β-catenin immunohistochemical staining.

DNA Preparation

DNA for PCR-SSCP was prepared from paraffin-embedded sections using a microdissection approach, as described previously (Yamamoto et al., 1995). Briefly, serial sections adjacent to those used for immunohistochemical study were prepared at a thickness of 5–10 μm, deparaffinized and placed in DNase free water. Using a fine needle, appropriate amounts of neoplasmic lesions were dissected out under a microscope and collected in 20–50 μL of protein lysis buffer containing 0.1 mg/mL proteinase K. Samples were initially digested at 37°C over night, and then underwent phenol/chloroform DNA extraction, and ethanol precipitation. The resulting DNA pellets were diluted with distilled water for PCR.

PCR-SSCP

PCR-SSCP analysis was carried out on PCR products of β-catenin exon 2 (corresponds to exon 3 in the human gene (Nollet et al., 1996)), which contains the GSK-3βphosphorylation sites (de La Coste et al., 1998). The sequences of the intronic PCR primers flanking the borders of exon 2 were: BCAT-1F, 5′-TACAGGTAGCATTTTCAGTTCAC-3′; Ctnnb m-R1; 5′-CACTCAGGGAAGGAGCTGTG-3′and Ctnnb m-F2; 5′-CCACCACCACAGCTCCTT-3′; BCAT-2R: 5′-TAGCTTCCAAACACAAATGC-3′ (de La Coste et al., 1998). PCR-SSCP was performed using non-radioisotopic SSCP analysis as described previously (Takahashi-Fujii et al., 1994). The primers were designed with a 5′FITC label. Hot start PCR was carried out in a 20 μL of reaction volume using AmpliTaq Gold (Perkin-Elmer Cetus Instruments, Norwalk, CT, USA) under the following conditions: initial preheating at 94°C for 9 min to achieve enzymatic activity followed by 40 cycles of denaturing (94°C) for 1 min, annealing (58°C) for 1 min and extension (72°C) for 1 min using Gene Amp PCR system 9700 (Perkin-Elmer Cetus Instruments). Five microliters of each PCR product were mixed with 20 μL of stop solution (95% formamide, 20 mM EDTA), denatured for 3 min at 95 0176C, immediately put on ice for 10 min then loaded (3 μL/lane) onto 0.5×MDE gels, with or without glycerol. Gels were run at 8 W for 16 h (with glycerol) or at 4 W for 16 h (without glycerol) at room temperature. After electrophoresis, gels were visualized using an FMBIO II Multi-View fluorescent image analyzer (Takara SHUZO CO., Shiga, Japan). The patterns of single-strand DNA bands were compared with those from normal control mouse DNA (from liver) to check for the presence of any mobility-shifted bands. Small areas of gels corresponding to the position of mobility shift bands were cut out, immersed in 400 μL distilled H₂O and shaken overnight to elute the DNA. Fifteen microliters of each extracted DNA solution were amplified by PCR using the no-labeled same sequence primers used in PCR-SSCP.

Direct DNA Sequencing

DNA sequencing was carried out for both strands using a Big Dye Terminator cycle sequencing Ready Reaction kit (ABI PRISM, Applied Biosystems, CA, USA) on a Genetic Analyzer (ABI PRISM 3100, Applied Biosystems, CA, USA).

Results

Histopathology and Immunohistochemistory for β-Catenin in Mouse Liver

Histological sections from liver of mice were examined. Figure 1 shows histological features of an HCA, an HCC and an HB induced by pNA (Fig. A, C, E). HBs in this study were characterized by small round to elongated cells with scant basophilic cytoplasm (Figure 1).

The immunohistochemical analysis was used to identify the localization of β-catenin protein. Immunohistochemical analysis revealed membrane staining of β-catenin in normal liver cells, HCAs and HCCs (Figure 1B, D). In HBs, β-catenin staining was also detected in membranes, but its intensity was weak (Figure 1F). No abnormal cytoplasmic or nuclear staining was exhibited by any of the neoplasms.

PCR-SSCP

Immunohistochemical analysis raised the question whether β-catenin gene mutations occurred in mice liver tumors, as there were no abnormal accumulations of the β-catenin protein in any of the cases. To determine β-catenin gene mutations occurred, DNA was extracted from mice liver tumors and analyzed by PCR-SSCP. Typical results of PCR-SSCP followed by direct DNA sequencing was shown in Table 1 and Figure 2. In HCAs, no β-catenin mutations were detected. In HCCs, the mutation frequency was 20% (3/15), all the mutations being silent and present at the third base of codons 16, 35, or 70. The mutation frequency in HBs was 20% (5/25), comprising 8% silent (2/25) and 12% functional mutations (3/25) (Table 1, Figure 2). The silent mutations were C to T and A to G transitions in codons 46 and 76, respectively. Mutations with amino acid change were observed in codons 49, 51, 67 and 72. One mutation was a transversion and the others were transitions. None occurred in or adjacent to the four known regulatory phosphorylation sites (codons 33, 37, 41, and 45).

Discussion

In the present immunohistochemical study, we demonstrated no abnormal localization of β-catenin protein in three types of liver neoplasms, HCAs, HCCs and HBs, induced by pNA in Crj:BDF1 males. Furthermore, PCR-SSCP analysis of β-catenin exon 2 revealed no mutations observed in any of the multiple phosphorylation sites of the gene. It has been reported that functional abnormalities of both β-catenin and other Wnt signal components, which interfere with the degradation of β-catenin, result in its accumulation in the cell cytoplasm and translocation into the nucleus (Morin et al., 1997; Rubinfeld et al., 1997). Therefore, the absence of abnormal accumulation of β-catenin in cytoplasm and/or nucleus indicated functional normality of the Wnt signal pathway in our neoplasms. Furthermore, the lack of genetic abnormalities in the multiple-phosphorylation sites of β-catenin exon 2 supports our immunohistochemical results in indicating that β-catenin does not play an important role in hepatic tumors, particularly HBs, induced by pNA in the Crj:BDF1 mouse.

There are many reports of high percentages of HBs with gene mutation and abnormal localization of β-catenin in mice (de La Coste et al., 1998; Devereux et al., 1999; Anna et al., 2000, 2003; Hayashi et al., 2003), in contrast to our previous (Kakuni et al., 2004) and present studies. The root causes are unclear, but it is likely that specificity regarding inducing compounds and strains of mice is involved. pNA is an example of a genotoxic carcinogen (Shimizu and Yano, 1986), generally considered to interact with DNA in their target organ cells, resulting in DNA damage and finally causing genetic alterations that are considered to be irreversible (Butterworth, 1990). Even if chemicals are genotoxic, however, some studies have indicated that the involvement of β-catenin in tumor formation varies with the chemical. For example, the mutation frequencies in exon 2 of the β-catenin gene in HCAs and HCCs induced by diethanolamine (Hayashi et al., 2003), 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (Huang et al., 2003) and diethylnitrosamine (Aydinlik et al., 2001) in mice were 32, 12.5, and 0%, respectively. All these three compounds are genotoxic carcinogens. Therefore, the absence of β-catenin gene mutations in present study provides further evidence that β-catenin gene mutations are not essential for liver tumor formation in the mouse. Several reports revealed that liver tumors might evolve through ras-dependent or ras-independent pathways (de La Coste et al., 1998; Devereux et al., 1999; Aydinlik et al., 2001; Hayashi et al., 2003). These reports indicated the H-ras mutations to be few or absent when β-catenin gene mutations were frequent and vice versa. Furthermore, the pathway appears to depend on the chemical. To elucidate the genetic mechanisms involved in the pNA case, analysis of H-ras mutations is needed in future investigations.

While high percentages of carcinogen-induced HBs have been found with gene mutations and abnormal localization of β-catenin, in the majority of cases B6C3F1 mice were used (de La Coste et al., 1998; Devereux et al., 1999; Anna et al., 2000, 2003; Hayashi et al., 2003). We employed Crj:BDF1 mice in our previous (Kakuni et al., 2004) and present study and there may be clear strain differences in gene mutations induced by carcinogens. Buchmann et al. (1991) demonstrated a 56% incidence rate for mutations of the H-ras gene in liver tumors induced by N-nitrosodiethylamine, but none in equivalent lesions due to this carcinogen in C57BL/6 mice. Incidence rates of H-ras mutations in both spontaneously developed and urethan-induced liver tumors were found to be more than 50% in C3H and B6C3F1 mice but only 7–9% in Balb/c mice (Dragani et al., 1991). Furthermore, there is a clear strain dependence regarding K-ras mutations in ethylnitrosourea-induced lymphomagenesis (Shimada et al., 2003).

In conclusion, liver neoplasms induced by pNA in Crj:BDF1 mouse appear to completely lack functional abnormalities in β-catenin, and our results raise a possibility that there is a β-catenin-independent pathway for development of HBs, as well as hepatocellular adenomas and carcinomas.