MUC4 expression increases progressively in pancreatic intraepithelial neoplasia

Pancreatic adenocarcinoma is believed to develop from histologically identifiable intraductal lesions known as pancreatic intraepithelial neoplasias (PanINs) that undergo a series of architectural, cytologic, and genetic changes, a progression model similar to the adenoma-carcinoma sequence in the colon. The apomucin MUC4 has been implicated in invasive pancreatic adenocarcinoma. MUC4 expression is not detectable at the RNA level in normal pancreas but is detectable at high levels in invasive pancreatic adenocarcinoma. We documented the pattern of expression of MUC4 in PanINs by studying a series of 71 PanIN lesions immunohistochemically using a new monoclonal antibody to MUC4. Five (17%) of 30 PanIN-1 lesions, 10 (36%) of 28 PanIN-2 lesions, 11 (85%) of 13 PanIN-3 lesions, and 25 (89%) of 28 invasive adenocarcinomas labeled with the MUC4 antibody used in the study. In addition, afew nonneoplastic lesions labeled with the MUC4 antibody, including reactive ducts in chronic pancreatitis, atrophic ducts filled with inspissated secretions, and ducts showing squamous metaplasia. Our data help establish the patterns of MUC4 expression in neoplastic precursors in the pancreas and add further support to the progression model for pancreatic adenocarcinoma.     

Telomere lengths of translocation-associated and nontranslocation-associated sarcomas differ dramatically

Sarcomas can be divided into those with specific translocations displaying monotonous cytomorphology, and those with complex karyotypes and marked cellular pleomorphism. Telomeres contain terminal DNA sequence repeats that maintain chromosomal stability. Telomeres shorten with cell division and may become dysfunctional leading to chromosomal instability. Using a fluorescence in situ hybridization/immunofluorescence method to assess telomere lengths in archival tissues we analyzed these two types of sarcomas using paraffin-embedded primary tumor specimens. Tissues from nine sarcomas with characteristic translocations (two synovial sarcomas, two alveolar rhabdomyosarcomas, two desmoplastic round cell tumors, and one each of infantile fibrosarcoma, myxoid liposarcoma, cellular congenital mesoblastic nephroma) and nine without (four malignant fibrous histiocytomas, two leiomyosarcomas, one pleomorphic rhabdomyosarcoma, one dedifferentiated chondrosarcoma, and one malignant peripheral nerve sheath tumor) were analyzed. In all (nine of nine) cases with specific translocations, which generally have few karyotypic abnormalities, telomere lengths were similar to or reduced compared to surrounding nonneoplastic tissues. In contrast, telomeres in cases lacking specific translocations, which generally contain complex karyotypes, were often found to be dramatically lengthened and heterogeneous. In addition to markedly elongated telomeres, seven of nine (78%) complex cases exhibited large brightly stained regions corresponding to a specific type of promyelocytic leukemia nuclear body found in immortalized cells that maintain telomeres in a telomerase-independent manner [alternative lengthening of telomeres (ALT) pathway]. This phenotype is unlike that of epithelial neoplasms that typically display complex karyotypes with abnormally short telomeres maintained by the enzyme telomerase. The discovery of heterogeneous telomere lengths and evidence of the ALT pathway in the majority of sarcomas with complex karyotypes supports the existence of a telomere maintenance pathway incapable of full karyotypic stabilization in pleomorphic sarcomas. These findings provide additional molecular-genetic evidence supporting the dichotomous grouping of sarcomas into those with characteristic signature translocations without extensive additional karyotypic abnormalities, and those without such signature translocations that typically display very complex karyotypes, and point to telomere dysfunction as a plausible contributor to the chromosomal aberrations found in complex sarcomas.     

The V599E BRAF mutation is uncommon in biliary tract cancers.

Activating point mutations of the BRAF oncogene have been identified in several solid tumors, most commonly in cutaneous melanomas and papillary carcinomas of the thyroid. A specific point mutation--V599E--accounts for the overwhelming majority of these mutational events. We explored the frequency of the V599E BRAF mutation in biliary tract cancers. In all, 62 archival biliary tract cancers, including 15 gallbladder cancers, 15 extrahepatic, and 10 intrahepatic cholangiocarcinomas from the United States, and 22 gallbladder carcinomas from Chile were analyzed for the V599E mutation of the BRAF gene using three distinct methods: direct sequencing, a primer extension method (Mutector assay), and the highly sensitive quantitative Gap Ligase Chain Reaction. The common V599E mutation was not identified in any of the 62 biliary cancer samples using these three methods of detection. The V599E somatic mutation of the BRAF gene is absent in biliary tract cancers, at least in the two geographic populations (United States and Chile) examined. Activation of the RAS/RAF/MAP kinase pathway in biliary tract cancers is likely to be secondary to oncogenic RAS mutations, or due to mutations of the BRAF gene at nucleotide positions not explored in the current study.     

Immunohistochemical labeling for the Dpc4 gene product is a specific marker for adenocarcinoma in biopsy specimens of the pancreas and bile duct.

We immunohistochemically labeled 72 biopsy specimens from the extrahepatic biliary tree and pancreas for Dpc4 protein and correlated expression with histologic diagnosis and patient follow-up. Specimens were classified histologically as follows: nonneoplastic, 35; neoplastic, 22; atypical, 15. Loss of expression of Dpc4 protein was identified in 12 specimens; 11 were histologically diagnostic of carcinoma. The 12th specimen was from a patient whose biopsy specimen initially was diagnosed as "atypical," but clinical follow-up revealed adenocarcinoma. Of the 12 atypical biopsy specimens with intact expression for Dpc4, follow-up later revealed that 10 were adenocarcinoma. Loss of expression of Dpc4 protein was never identified in a benign specimen. Immunohistochemical labeling for the Dpc4 gene product is a specific marker of carcinoma in biopsy specimens of the pancreas and extrahepatic bile ducts and is marginally helpful in classifying atypical specimens. The sensitivity for carcinoma is low. This latter finding is not unexpected, because the DPC4 tumor suppressor gene is inactivated in only about half of pancreatic and biliary malignant neoplasms. Importantly, loss of Dpc4 expression has been reported in in situ carcinomas, suggesting that loss of expression should not be equated with invasive carcinoma.     

Retinoblastoma pathway dysregulation causes DNA methyltransferase 1 overexpression in cancer via MAD2-mediated inhibition of the anaphase-promoting complex

We have examined the mechanism of normal DNA methyltransferase 1 (DNMT1) degradation as well as its mechanism of dysregulation in cancer. We have previously reported that DNMT1 protein levels were elevated and abnormally stabilized because of defective degradation through its N-terminal destruction domain. Here, we report that DNMT1 was abnormally stabilized in several cancer cell lines and that, in cells with normal DNMT1 destruction, depletion of CDC20 or FZR1 (two substrate recognition adaptor components of the anaphase-promoting complex) resulted in stabilization of DNMT1 that was partially dependent on the N-terminal destruction domain, thus implicating this cell cycle regulator in the destruction of DNMT1. MAD2, an inhibitor of CDC20, was shown to stabilize DNMT1 levels, and overexpression of MAD2, a consequence of retinoblastoma (RB) pathway dysregulation, was shown to correlate with impaired G(1) phase DNMT1 destruction and RB inactivation by hyperphosphorylation in several normal and cancer cell lines. Furthermore, in a series of 85 cases of human breast cancer, a moderately strong, but highly significant, correlation between MAD2 and DNMT1 immunohistochemical staining was observed, yielding a Spearman rank order correlation coefficient of 0.37 (P<0.001). This suggests that RB pathway inactivation, a common dysfunction in cancer cells, may be the underlying cause of DNMT1 dysregulation.