Toshimasa Yatsuoka

Restoration of SMAD4 by gene therapy reverses the invasive phenotype in pancreatic adenocarcinoma cells.

SMAD4 is a critical cofactor in signal transduction pathways activated in response to transforming growth factor-beta (TGF-β)-related ligands, regulating cell growth and differentiation. The roles played by SMAD4 inactivation in tumours highlighted it as a tumour-suppressor gene. However, restoration of the TGF-β antiproliferative pathway following SMAD4 gene transfer in null-tumour cell lines is controversial. Herein, we report the inhibitory effects of SMAD4 on pancreatic tumour invasion and angiogenesis. Adenoviral transfer of this gene in a panel of SMAD4 homozygous-deleted human pancreatic tumour cell lines restored SMAD4 protein expression and function. Although it did not affect proliferation significantly in vitro, SMAD4 inhibited in vivo tumour growth in immunodeficient mice. In this xenograft setting, differential suppression of tumour growth in vivo was mediated, at least in part, through downregulation of vascular endothelial growth factor and expression of gelatinases. We documented the reduced invasion and angiogenesis histologically and by intravital microscopy, and gained mechanistic insight at the messenger and protein level. Finally, we found a negative reciprocal regulation between SMAD4 and ETS-1. ETS-1 is considered a marker for tumour invasion. Upon SMAD4 deletion, we detected high expression levels of ETS-1 in pancreatic tumour cells, suggesting the shift of the pancreatic tumour toward an invasive phenotype.

Genomic analysis of DUSP6, a dual specificity MAP kinase phosphatase, in pancreatic cancer

DUSP6 (alias PYST1), one of the dual-specificity tyrosine phosphatases, is localized on 12q21, one of the regions of frequent allelic loss in pancreatic cancer. This gene is composed of three exons, and two forms of alternatively spliced transcripts are ubiquitously expressed. Although no mutations were observed in 26 pancreatic cancer cell lines, reduced expressions of the full-length transcripts were observed in some cell lines, which may suggest some role for DUSP6 in pancreatic carcinogenesis.

Association of poor prognosis with loss of 12q, 17p, and 18q, and concordant loss of 6q/17p and 12q/18q in human pancreatic ductal adenocarcinoma

OBJECTIVE:Pancreatic cancer is one of the diseases with the poorest prognosis, but the associated genetic alterations are not yet well understood. The genetic alterations reported to date in pancreatic cancer include frequent mutations of the KRAS, TP53, p16, and SMAD4 genes. Mutation of the TP53 gene was reported to be associated with a poor prognosis. In this study, we analyzed the association of loss of heterozygosity (LOH) with clinicopathological features to attempt to devise effective methods in the future for clinically applying our results to diagnosis and treatment.METHODS:A total of 55 tumors from patients with primary pancreatic ductal carcinomas (34 men and 21 women, mean average age 63.9 yr) in which all the relevant clinical and pathological data were available were analyzed. A total of 46 cases were surgically resected, and nine cases were not. Tumor cells as well as corresponding normal cells were collected by microdissection under a microscope, and DNAs were purified. Allelotype analysis was performed by the PCR-based method, and the results were statistically analyzed.RESULTS:LOH of ≥30% were observed on chromosome arms 17p (47%, 17/36), 9p (45%, 14/31), 18q (43%, 15/35), 12q (34%, 10/29), and 6q (30%, 10/33). LOH of 12q, 17p, and 18q were significantly associated with a poor prognosis. Concordant losses of 6q with 17p and 18q were significantly associated with a poor prognosis. Concordant losses of 6q with 17p and of 12q with 18q were also found.CONCLUSIONS:Because LOH of 12q, 17p, and 18q are significantly associated with a poor prognosis, mutation of the putative tumor suppressor genes on these chromosome arms may play significant roles in the disease progression. Concordant losses of 6q with 17p and of 12q with 18q suggest that protein products of putative tumor suppressor genes on 6q and 12q may function in association with TP53 and SMAD4, respectively.

Overexpression of the p53-inducible brain-specific angiogenesis inhibitor 1 suppresses efficiently tumour angiogenesis.

The brain-specific angiogenesis inhibitor 1 gene has been isolated in an attempt to find fragments with p53 “functional” binding sites. As reported herein and by others, brain-specific angiogenesis inhibitor 1 expression is present in some normal tissues, but is reduced or lost in tumour tissues. Such data and its particular structure prompted the hypothesis that brain-specific angiogenesis inhibitor 1 may act as a mediator in the local angiogenesis balance. We herein demonstrate that brain-specific angiogenesis inhibitor 1 over-expression suppresses tumour angiogenesis, delaying significantly the human tumour growth in immunodeficient mice. The inhibitory effect of brain-specific angiogenesis inhibitor 1 was documented using our intravital microscopy system, strongly implicating brain-specific angiogenesis inhibitor 1 as a mediator in the control of tumour angiogenesis. In contrast, in vitro tumour cell proliferation was not inhibited by brain-specific angiogenesis inhibitor 1 transfection, whereas some level of cytotoxicity was assessed for endothelial cells. Immunohistochemical analysis of tumour samples confirmed a reduction in the microvessel density index in brain-specific angiogenesis inhibitor 1-overexpressing tumours. At messenger level, moderate changes could be detected, involving the down-regulation of vascular endothelial growth factor and collagenase-1 expression. Furthermore, brain-specific angiogenesis inhibitor 1 expression that was lost in a selection of human cancer cell lines could be restored by wild-type p53 adenoviral transfection. Brain-specific angiogenesis inhibitor 1 should be considered for gene therapy and development of efficient drugs based on endogenous antiangiogenic molecules.

Identification of three commonly deleted regions on chromosome arm 6q in human pancreatic cancer.

Pancreatic cancer has one of the poorest prognoses among malignant diseases. To understand its molecular mechanisms, we studied allelic losses on the long arm of chromosome 6. Using 55 paired DNAs of tumors and their corresponding normal tissues and 30 microsatellite markers that spanned the entire 6q chromosome arm, we found three distinct regions of common allelic loss: region A, a less than 500‐kb region bordered by D6S449 and D6S283 on 6q21 with a loss of heterozygosity (LOH) frequency of 69% (38/55); region B, a 7‐cM region bordered by D6S292 and D6S308 on 6q23–q24 with a LOH frequency of 60% (33/55); and region C, a 13‐cM region bordered by D6S305 and D6S264 with a LOH frequency of 51% (28/55). We further focused on region A and constructed a physical map using yeast artificial chromosome (YAC) clones, their derived cosmid clones, and bacterial artificial chromosome (BAC) clones. Region A was completely covered by three overlapping BAC clones. Our results in the present study should shed light on the cloning and characterization of tumor suppressor genes in pancreatic carcinogenesis. Genes Chromosomes Cancer 25:60–64, 1999.

Prognostic value of KRAS and BRAF mutations in curatively resected colorectal cancer.

AIM To investigate the prognostic role of KRAS and BRAF mutations after adjustment for microsatellite instability (MSI) status in Japanese colorectal cancer (CRC) population. METHODS We assessed KRAS and BRAF mutations and MSI status in 813 Japanese patients with curatively resected, stage I-III CRC and examined associations of these mutations with disease-free survival (DFS) and overall survival (OS) using uni- and multivariate Cox proportional hazards models. RESULTS KRAS and BRAF mutations were detected in 312 (38%) of 812 and 40 (5%) of 811 tumors, respectively. KRAS mutations occurred more frequently in females than in males (P=0.02), while the presence of BRAF mutations was significantly associated with the female gender (P=0.006), proximal tumor location (P<0.001), mucinous or poorly differentiated histology (P<0.001), and MSI-high tumors (P<0.001). After adjusting for relevant variables, including MSI status, KRAS mutations were associated with poorer DFS (HR=1.35; 95%CI: 1.03-1.75) and OS (HR=1.46; 95%CI: 1.09-1.97). BRAF mutations were poor prognostic factors for DFS (HR=2.20; 95%CI: 1.19-4.06) and OS (HR=2.30; 95%CI: 1.15-4.71). Neither the BRAF by MSI interaction test nor the KRAS by MSI interaction test yielded statistically significant results for DFS and OS. CONCLUSION KRAS and BRAF mutations are associated with inferior survival, independent of MSI status, in Japanese patients with curatively resected CRC.

Clinicopathological characteristics and prognostic impact of colorectal cancers with NRAS mutations

At present, molecular markers of colorectal cancer (CRC), including KRAS, NRAS and BRAF mutations, and the microsatellite status are evaluated for the development of personalized treatments. However, clinicopathological and molecular characteristics and the prognostic role of NRAS mutations remain unclear. In the present study, a total of 1,304 consecutive stage 0-IV CRC tumor samples were analyzed for KRAS (exon 2, 3 and 4), NRAS (exon 2 and 3) and BRAF (exon 15) mutations. Multivariate analysis was performed to assess the prognostic impact of NRAS mutations. KRAS, NRAS and BRAF mutations were identified in 553 (42.4%), 35 (2.7%), and 59 (4.5%) of 1,304 CRC cases, respectively. Tumors with NRAS mutations were more frequently located in the distal colorectum compared with those with KRAS or BRAF mutations. Multivariate analysis indicated that KRAS and BRAF mutations were found to be associated with poor prognosis [hazard ratio (HR)=1.44, 95% confidence interval (CI), 1.18-1.76 and HR=2.09; 95% CI, 1.33-3.28, respectively], whereas NRAS mutations were associated with a trend toward favorable prognosis (HR=0.53; 95% CI, 0.27-1.03). Characteristics and prognosis of CRC with NRAS mutations are different from those with KRAS or BRAF mutations.

Chromosome 12, frequently deleted in human pancreatic cancer, may encode a tumor-suppressor gene that suppresses angiogenesis.

Several lines of evidence have suggested that the long arm of chromosome 12 may carry a tumor-suppressor gene(s) that plays a role in pancreatic ductal carcinogenesis. We have previously found a significant association between loss of heterozygosity of the 12q arm and a poor prognosis in pancreatic cancer patients. In this study, we introduced a normal copy of chromosome 12 into some pancreatic ductal carcinoma cells. Both anchorage-dependent and -independent proliferations as well as invasiveness were similar throughout the hybrid clones when compared with their corresponding parental cells. In sharp contrast, significant suppression of tumorigenesis was observed after inoculation of the hybrid clones into nude mice. Measurements made up to 1 month later showed that there was a significant delay in the growth of tumors into which the introduced normal copy of chromosome 12 had been restored. More significantly, using our dorsal skin chamber and an intravital microscopy system experiment in SCID mice, we demonstrated and visualized directly that implantation of the hybrids failed to promote the angiogenic phenotype encountered in the parental cells. Gene expression profiling using the complementary DNA microarray system identified a set of 24 genes differentially expressed between the hybrids and parental cells. An additional set of 18 genes was also identified that were differentially expressed between the hybrid clone that lost its growth-suppression activity and one that retained such activity. Another set of 25 genes mapped on 12q was detected that showed high expression levels in the hybrid clones retaining growth-suppressive activity. In summary, this study provides the first functional evidence of the existence of an additional tumor-suppressor gene(s) on chromosome 12, whose absence is responsible for the pathogenesis in pancreatic ductal carcinogenesis.

Combined Microsatellite Instability and BRAF Gene Status As Biomarkers for Adjuvant Chemotherapy in Stage III Colorectal Cancer

The clinical relevance of combined microsatellite instability (MSI) and BRAF status for adjuvant treatment in stage III colorectal cancer (CRC) remains elusive.

Isolation and Characterization of the Human Gene Homologous to the Drosophila Headcase (hdc) Gene in Chromosome Bands 6q23-q24, a Region of Common Deletion in Human Pancreatic Cancer

Deletions of the long arm of chromosome 6 (6q) are one of the most common chromosomal abnormalities in multiple human malignancies. Previously, we have identified three commonly deleted regions on 6q (6q21, 6q23-q24, and 6q26) in pancreatic cancer by loss of heterozygosity studies, suggesting the presence of one or more tumor suppressor genes on this chromosome arm. Using a combination of database search and cDNA library screening, we successfully isolated a transcript from 6q24. This mRNA encodes a protein consisting of 543 amino acids with homology to the Drosophila headcase (hdc) gene and, thus, is designated as hHDC. Northern analysis identified a ubiquitously expressed 5.6-kb transcript. Seventeen (81%) of 21 pancreatic cancer cell lines and four (80%) of five renal cell carcinoma cell lines showed low level expression, suggesting that the hHDC gene may play an important role in some human cancers.