We have previously reported that expression of periostin mRNA is markedly reduced in a variety of human cancer cell lines, suggesting that downregulation of periostin mRNA expression is correlated with the development of human cancers. In our study, to clarify the role of the periostin in human bladder carcinogenesis, we examined the expression of periostin mRNA in normal bladder tissues, bladder cancer tissues and bladder cancer cell lines by Northern blot analysis and RT-PCR analysis. Although the expression of periostin mRNA was detected in 100% (5/5) of normal bladder tissues, it was not detected in 3 human bladder cancer cell lines examined. It was also detected in 81.8% (9/11) of grade 1, 40.0% (4/10) of grade 2 and 33.3% (4/12) of grade 3 bladder cancer tissues, indicating that downregulation of periostin mRNA is significantly related to higher grade bladder cancer (p<0.05). To assess the tumor suppressor function of periostin, we investigated the ability of periostin gene to suppress malignant phenotypes of a bladder cancer cell line, SBT31A. Ectopic expression of periostin gene by a retrovirus vector suppressed in vitro cell invasiveness of the bladder cancer cells without affecting cell proliferation and tumor growth in nude mice. Periostin also suppressed in vivo lung metastasis of the mouse melanoma cell line, B16-F10. Mutational analysis revealed that the C-terminal region of periostin was sufficient to suppress cell invasiveness and metastasis of the cancer cells. Periostin may play a role as a suppressor of invasion and metastasis in the progression of human bladder cancers. ' 2005 Wiley-Liss, Inc.

Abstract.We have previously reported that the expression of periostin mRNA is significantly repressed in human bladder cancer tissues, and that periostin plays a role as a suppressive factor for invasion and metastasis in the progression of human bladder cancers. In this study, to clarify the role of alternative splicing of periostin in human bladder carcinogenesis, we examined the expression of wild-type (WT) and spliced variants of periostin mRNA in normal bladder and bladder cancer tissues. Although both WT and spliced periostin mRNA were expressed in all normal bladder tissues examined, no WT periostin mRNA was detected in the examined transitional cell carcinomas (TCCs) of the bladder (0/23) or in bladder cancer cell lines (0/6). Spliced variants of periostin were detected in 48% (11/23) of TCC tissues and 33% (2/6) of bladder cancer cell lines. Two types of spliced periostin (Variants I and II) were successfully isolated from bladder cancer tissues, but Variant I, which is predominantly expressed in bladder cancer tissues, did not show suppressor activity on in vitro invasiveness and in vivo metastasis of cancer cells. Immunohistochemical analysis indicated that strong belt-like expression of periostin protein was observed in the stroma just beneath the normal bladder epithelium, while it was mostly attenuated in bladder cancer tissues. These results indicate that the loss of WT periostin by down-regulation and/or alternative splicing, which produces Variant I, is closely correlated with the development of bladder cancer.

The drs gene was originally isolated as a suppressor against v-src transformation. Expression of drs mRNA was markedly downregulated in a variety of human cancer cell lines and tissues, suggesting that the drs gene acts as a tumor suppressor. In this study, we found that ectopic expression of the Drs protein induced apoptosis in human cancer cell lines. Analyses using deletion mutants of drs revealed that both the C-terminal region and the three consensus repeats in the N-terminal region are essential for the induction of apoptosis. Caspase-12, -9, and -3 were sequentially activated by drs, and specific inhibitors of caspase-3 and -9 suppressed drs-induced apoptosis. The release of cytochrome c from the mitochondria into the cytoplasm was not observed in apoptosis by drs, suggesting that the mitochondrial pathway does not mediate drsinduced apoptosis. Furthermore, we found that the Drs protein can interact with ASY/Nogo-B/RTN-x S , an apoptosis-inducing protein localized in the endoplasmic reticulum, and that coexpression of these genes increased the efficiency of apoptosis. These results indicated that Drs induces apoptosis by a novel pathway mediated by ASY/Nogo-B/RTN-x S , caspase-12, -9, and -3.

The human cyclin D1 gene generates two major isoforms, cyclin D1a and cyclin D1b, by alternative splicing. Although cyclin D1b mRNA is hardly expressed in normal human tissues, it is detected in approximately 60% of human bladder cancer tissues and cell lines. In the present study, to assess the therapeutic ability of cyclin D1b siRNA, we investigated the anti-oncogenic effects of cyclin D1b siRNA on human bladder cancer cell lines, SBT31A and T24, which express cyclin D1b mRNA. Knockdown of cyclin D1b by specific siRNA significantly suppressed cell proliferation, in vitro cell invasiveness and three-dimensional (3D) spheroid formation in these cell lines. Cell cycle analyses revealed that cyclin D1b siRNA inhibited G1-S transition in T24 cells. The increase in the sub-G1 fraction, morphological aberrant nuclei with nuclear fragmentation and caspase-3 activity in SBA31A cells treated with cyclin D1b siRNA showed that cyclin D1b siRNA induced apoptosis. In T24 cells, knockdown of cyclin D1b suppressed the expression of the stem cell marker CD44. Knockdown of cyclin D1b or CD44 suppressed the invasiveness under 3D spheroid culture conditions and expression of N-cadherin. Tumor growth of SBT31A cells in nude mice was significantly inhibited by cyclin D1b siRNA. Taken together, these results indicate that knockdown of cyclin D1b suppresses the malignant phenotypes of human bladder cancer cells via induction of apoptosis and suppression of cancer cell stemness and epithelial-mesenchymal transition. Applying cyclin D1b siRNA will be a novel therapy for cyclin D1b-expressing bladder cancers.

Alternative splicing in the cyclin D1 gene produces cyclin D1b variant which lacks a C-terminal region containing the threonine-286 (T286) phosphorylation site required for nuclear export. We have shown that the expression of the cyclin D1b variant is detected in about 60% of human bladder cancer tissues (15/26) and cell lines (3/5). To examine the role of the cyclin D1b variant in bladder carcinogenesis, we introduced wild-type cyclin D1a, cyclin D1b variant or mutant cyclin D1-T286A cDNAs into a human bladder cancer cell line, SBT991, in which cyclin D1b transcript was not expressed, and compared their oncogenic activities. Ectopic expression of cyclin D1b promoted cell invasiveness and anchorage-independent growth of the cancer cells. On the other hand, cyclin D1-T286A enhanced anchorage-independent growth, but did not promote cell invasiveness. The amount of nuclear-localized cyclin D1b and cyclin D1-T286A was higher than that of nuclear-localized cyclin D1a. In addition, introduction of siRNA specific for cyclin D1b into cells of the T24 bladder cancer cell line, in which cyclin D1b transcript was expressed, significantly suppressed cell invasiveness. Immunoprecipitation analysis revealed that cyclin D1a and cyclin D1-T286A could bind to cyclin-dependent kinase 4 (CDK4) but cyclin D1b has lost its capacity to associate with CDK4. Unlike cyclin D1a and cyclin D1-T286A, expression of cyclin D1b did not enhance phosphorylation of Rb protein in SBT991 cells. These results indicate that cyclin D1b promotes cell invasiveness independent of binding to CDK4 to enhance Rb phosphorylation.