Chimeric tumor suppressor 1, a p53-derived chimeric tumor suppressor gene, kills p53 mutant and p53 wild-type glioma cells in synergy with irradiation and CD95 ligand.

Adenoviral chimeric tumor suppressor 1 (CTS1) gene transfer was evaluated as a novel approach of somatic gene therapy for malignant glioma. CTS1 is an artificial p53-based gene designed to resist various pathways of p53 inactivation. Here, we report that an adenovirus encoding CTS1 (Ad-CTS1) induces growth arrest and loss of viability in all glioma cell lines examined, in the absence of specific cell cycle changes. In contrast, an adenovirus encoding wild-type p53 (Ad-p53) does not consistently induce apoptosis in the same cell lines. Electron microscopic analysis of Ad-CTS1-infected glioma cells reveals complex cytoplasmic pathology and delayed apoptotic changes. Ad-CTS1 induces prominent activation of various p53 target genes, including p21 and MDM-2, but has no relevant effects on BCL-2 family protein expression. Although Ad-CTS1 strongly enhances CD95 expression at the cell surface, endogenous CD95/CD95 ligand interactions do not mediate CTS1-induced cell death. This is because Ad-CTS1 promotes neither caspase activation nor mitochondrial cytochrome c release and because the caspase inhibitors, z-val-Ala-DL-Asp-fluoromethylketone (zVAD)-fmk or z-Ile-Glu-Thr-Asp- fluoromethylketone (z-IETD)-fmk, do not block CTS1-induced cell death. Ad-CTS1 synergizes with radiotherapy and CD95 ligand in killing glioma cells. In summary, Ad-CTS1 induces an unusual type of cell death that appears to be independent of BCL-2 family proteins, cytochrome c release, and caspases. CTS1 gene transfer is a promising strategy of somatic gene therapy for malignant glioma.

Claudin-1 and claudin-5 expression and tight junction morphology are altered in blood vessels of human glioblastoma multiforme.

The aim of the study was to characterize the interendothelial junctions in tumor microvessels of five cases of human glioblastoma multiforme. In addition to morphological analysis, tumors were screened for the expression of junctional proteins, such as occludin, claudin-1, ZO-1 and catenins. The expression of the tight junction protein claudin-1 was lost in the majority of tumor microvessels, whereas claudin-5 and occludin were significantly down-regulated only in hyperplastic vessels. As shown by freeze-fracture analysis, under the conditions of tumor growth tight junction particles of endothelial cells were almost exclusively associated with the exocytoplasmic fracture face, providing evidence for a switch of the particles from the protoplasmic to the external leaflet of the endothelial membrane. These results suggest a relationship between claudin-1 suppression and the alteration of tight junction morphology, which is likely to correlate with the increase of endothelial permeability. Underlining the undifferentiated state of tumor microvessels, plakoglobin, a crucial protein for mature endothelial junctions, was not detectable in most microvessels, whereas beta-catenin was abundantly labeled. In this context, it is of particular interest that the majority of microvascular pericytes were negative for alpha-smooth muscle actin, which is a marker of differentiated pericytes, although pericytes were frequently found in electron micrographs. In conclusion, the data suggest that the increase in microvascular permeability in human gliomas, contributing to the clinically severe symptoms of brain edema, is a result of a dysregulation of junctional proteins.

Transepithelial electrical resistance and tight junctions of human gingival keratinocytes.

Human gingival keratinocytes (HGKs) were studied by means of freeze-fracture technique, conventional electron microscopy and the transepithelial electrical resistance for the investigation of intercellular contacts. For the purpose of comparison, MDCK cells and HaCat cells were also included. An unexpected finding was the presence of tight junctions in the HGKs. In vivo the tight junctions, which were of low complexity and P-face-associated, were co-distributed with desmosomes; in one case, the strands ran directly through desmosomal plaques. Where tight junctions and desmosomes occurred together, no gap junctions were seen. In contrast, where no tight junctions were present, gap junctions and desmosomes were co-localized. However, the unfavourable fracture planes through the tissue did not allow a clearcut allocation of gap junction/tight junction occurrence to certain strata. In vitro, HGKs also expressed tight junctions which formed networks of low complexity and high P-face association. Whereas desmosomes were highly expressed, gap junctions were not observed in cultured keratinocytes. Transepithelial electrical resistances (TEER) of cultured HGKs were higher than the values in low resistance-MDCK cells and HaCat cells but considerably lower than the values in high resistance MDCK cells, supporting the fundamental correlation between tight junction morphology and TEER. The results with this cell culture model of the human gingiva provide some valuable information about in vitro differentation and concommittent changes in cellular contacts of human gingival keratinocytes.