PLCε^−/− mice were obtained from the Laboratory of Experimental Molecular Biology at Kobe University (Kobe, Japan) and maintained at the 302 Hospital of the Chinese People's Liberation Army (Beijing, China). PLCε^−/− homozygous mice and wild-type (PLCε^+/+) littermates were obtained by crossing C57BL/6J PLCε^+/− and 129S4 PLCε^+/− mice. Mice carrying inactivated PLC allele (PLCε⁻) were generated using in-frame deletion of an exon encoding the catalytic X domain, as described previously (14). All animals were housed under standard conditions (22±1°C; 12-h light/dark cycle; 50–55% humidity) with free access to food pellets and tap water. Experiments were performed on 6–8 week-old male mice, and a total of 72 PLCε^+/− and 72 PLCε^+/− mice were used in the present study. The mean weight of the animals was 20.3±0.1 g for the PLCe⁺/⁺ mice and 20.3±0.15 g for the PLCe⁻/⁻ mice. Disparate groups of mice were used in each experiment. Animal experiments were approved by the Animal Care and Use Committee of The 463 hospital of Chinese People's Liberation Army (Shenyang, China).

Antibodies used included anti-mouse PLCε antibody (Dako Cytomation, Copenhagen, Denmark), rabbit anti-mouse polyclonal anti-vascular endothelial growth factor-A (VEGF-A) antibody (cat. no. ab51745; Abcam, Cambridge, MA, USA), rabbit anti-mouse poly-clonal anti-GAPDH antibody (cat. no. sc-25778; Santa Cruz Biotechnologies, Dallas, TX, USA), rabbit anti-mouse anti-cyclooxygenase-2 (COX-2) antibody (cat. no. 160126; Cayman, Ann Arbor, MI, USA) and goat anti-rabbit immunoglobulin G-horseradish peroxidase conjugated secondary antibodies (Santa Cruz Biotechnologies).

A BBN-induced mouse model of bladder cancer was prepared according to the method described by Vecchione et al (19) with certain modifications. Briefly, six-week-old male PLCε⁻/⁻ mice (n=72) and PLCε⁺/⁺ mice (n=72) were sub-divided into BBN treatment groups (n=48) and control groups (n=24) without treatment of BBN (Table I). BBN-treated mice were given tap water containing 0.1% BBN for 12 weeks. Thereafter, they had access to tap water without BBN. Control mice were given water without BBN throughout the experiment. Mice were sacrificed at 8, 12 and 18 weeks after the cessation of BBN treatment. Bladder specimens were harvested and analyzed for pathology [hematoxylin-eosin (HE; Beyotime Institute of Biotechnology, Shanghai, China) staining and ultrastructural assessment] and protein (western blot and immunofluorescence).

The mice were anesthetized with sodium pentobarbitone (40 mg/kg, i.p.; Sigma-Aldrich, St. Louis, MO, USA) and then transcardially perfused with 10 ml 0.9% saline, followed by 30 ml 0.1 M phosphate buffer (PB; pH 7.4) containing 4% paraformaldehyde for 5 min. At necropsy, urinary bladders were removed and placed in 4% paraformaldehyde for HE staining immunofluorescence or in 2.5% glutaraldehyde (Sigma-Aldrich) for ultrastructural study. For HE staining, each bladder was dissected, processed for routine paraffin embedding, cut into 5 µm sections and mounted onto polylysine-coated slides. Sections were de-waxed in xylene (Sinopharm, Shanghai, China), re-hydrated in a descending series of ethanols and processed for routine HE. For ultrastructural study, sections were washed with 0.01 M phosphate-buffered saline (PBS), post-fixed with 10 g/l OsO₄ in PB for 45 min, de-hydrated through a graded ethanol series and propylene oxide, and flat-embedded with Epon-812 (Nanjing Tansi Technology Co., Ltd., Nanjing, China). The sections were examined under a light microscope (AE31; Motic, Xiamen, China) and regions containing all layers of the bladder were investigated under an electron microscope. Tissue samples from the selected regions were cut into sections on an ultramicrotome (EM UC7; Leica, Wetzlar, Germany) and prepared for electron microscopic analysis (H-7100; Hitachi, Tokyo, Japan).

Protein was extracted from each sample using radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology). The concentrations of proteins were detected using the bicinchoninic acid assay (Beyotime Institute of Biotechnology). Protein samples were separated by 10% SDS-PAGE and then transferred onto nitrocellulose membranes. Then membranes were blocked with 5% non-fat milk followed by incubation with antibodies against COX2 (1:500), VEGF-A (1:5,000) and GAPDH (1:1,000), respectively, at 37°C for another 1.5 h. Membranes were subsequently washed with Tris-buffered saline containing Tween 20 (TBST) and incubated with the corresponding secondary antibodies conjugated with horseradish peroxidase (1:5,000) for 45 min at 37°C. Following washing with TBST, a RapidStep™ ECL reagent (Millipore, Bedford, MA, USA) and x-ray film were used to capture images of the blots.

Detection of PLCε, COX2 and VEGF-A was performed on consecutive sections obtained from paraffin-embedded tissues using an immunofluorescence double labeling method. Sections were dried at room temperature, de-paraffinized, re-hydrated and then treated with 2% hydrogen peroxide at 37°C for 5 min. Antigen retrieval was performed by pepsin (Biosharp, Hefei, China) treatment at 40°C for 10 min. After blocking with 5% bovine serum albumin, the sections were incubated with anti-PLCε and COX2 (1:200) or VEGF-A antibody (1:500) overnight at 4°C. Sections were washed with 0.01 M PBS and incubated with fluorescein isothiocyanate-labeled anti-goat immunoglobulin (Ig)G antibody (1:500; Sigma-Aldrich) for PLCε and Texas Red-labeled anti-rabbit IgG antibody (1:1,000; Molecular Probes, Breda, Netherlands) for COX2 and VEGF-A at 37°C for 2 h. The sections were thoroughly rinsed in 0.01 M PBS between and after the incubation steps. The sections were examined using confocal laser-scanning microscopy (FV1000; Olympus, Tokyo, Japan). Digital images were captured using EZ-C1 3.50 software (Nikon).

All statistical analyses were performed using SPSS 10.0 software (SPSS, Inc., Chicago, IL, USA). The Kaplan-Meier test was used to analyze survival data. The incidence and multiplicity of urinary bladder cancer were analyzed by the Fisher's exact and Wilcoxon's rank-sum tests, respectively. Each experiment was repeated three times and representative results are presented.

In the present study, one hundred and forty-four mice were used, out of which seven died during the experiment (5%; Group A, 1; group B, 4; group D, 2). These mice showed no significant difference from others in weight and activity levels prior to their death. A higher susceptibility of PLCε^+/+ mice to BBN-induced tumor development was observed. The tumors of PLCε^−/− mice were smaller than those of PLCε^+/+ mice at all time-points examined (Fig. 1A–D). In Group B, eight mice (18.18%) had normal urinary bladder mucous membranes, eight mice (18.18%) showed atypical hyperplasia and 63.6% (28/44) presented with neoplastic lesions (with in situ or invasive carcinoma), and in group D, these numbers were 26.09% (12/46), 30.43% (14/46) and 43.5% (20/46), respectively (Table I). Pre-neoplastic lesions in the urothelium adjacent to advanced tumors were frequently observed. By contrast, in the control groups A and C, atypical hyperplasia and neoplastic lesions were not observed. The size of the tumors in group B was significantly larger than that in group D (Fig. 1E and F). No significant differences were observed between groups A and C.

In the present study, pathological changes in the development of BBN-induced bladder carcinoma in PLCε^−/− and PLCε^+/+ mice were monitored. The following pathological changes were observed by light microscopy: Smooth mucosa of bladder walls were present in groups A and C without the presence of ulcers, congestion or neoplasms (Fig. 2A and B). The morphology of the mucosa of the bladder walls in groups B and D changed gradually with increasing time of BBN intake. In general, tumor formation in PLCε^−/− mice was delayed and its incidence was reduced compared with that in PLCε^+/+ mice. Electron microscopy further confirmed the morphological findings in the experimental groups: While no differences in morphology of the bladder transitional epithelium were observed between groups A and C (Fig. 2E and F), the following pathological changes were visible in group B at week 12: Chromatin was distributed across the nuclear membrane, nucleolar hypertrophy was present and fiber structure around the nuclei was disordered (Fig. 2G). However, heterogenous transitional cells were observed in the mucosa of mice in group D at week 12 (Fig. 2H). All of the above results demonstrated that knockout of PLCε attenuated BBN-induced bladder tumorigenesis.

In order to investigate the underlying mechanisms of the involvement of PLCε in tumor development, the present study examined the expression of representative inflammatory and angiogenesis-associated proteins in tumors from the bladders of mice in groups B and D. Previous studies have revealed that PLCε is important in TPA-induced skin inflammation and tumor promotion (17,20). Inflammation is known to exert its highest effects on tumorigenesis in the later stage, in which tumors progress to high-grade adenomas (21–23). Therefore, the present study investigated the role of PLCε in inflammation associated with late-stage bladder tumorigenesis. In grade-matched tumors from PLCε^+/+ and PLCε^−/− mice at week 18, the expression of COX-2 and VEGF-A was assessed; these proteins were selected to be analyzed as representative signaling molecules due to their intrinsic functions in inflammation and angiogenesis, respectively, and because they have been implicated in bladder tumorigenesis (24). Western blot analysis revealed that the expression levels of COX-2 and VEGF-A were influenced by the PLCε background of the mice, with upregulation in the BBN-treated PLCε^+/+ mice, but not in BBN-treated PLCε^−/− mice (Fig. 3). In addition, the expression of COX-2 and VEGF-A was detected by immunofluorescence analysis. Groups A and B showed stronger green fluorescence (PLCε) than groups C and D, which was consistent with their genetic backgrounds. Consistent with the western blot results, marked increased in the VEGF-A signals in tumor epithelial cells of high-grade adenomas in PLCε^+/+ mice were observed, which were less apparent in PLCε^−/− mice (Fig. 4A). The expression patterns of COX-2 were similar to those of VEGF-A, with increased expression in groups B and D (Fig. 4B). However, a locally elevated level of COX-2 was also identified. These results demonstrated that the protective role of PLCε knockout against bladder carcinogenesis may be associated with inflammatory responses and angiogenesis.