PubMed Health⌕ Search

Biomedical subjects

Carsten Friedrich

Publications and source records attributed to Carsten Friedrich.

2 recordsLinked to original sources

Overexpression of Midkine in malignant peripheral nerve sheath tumor cells inhibits apoptosis and increases angiogenic potency.

Malignant peripheral nerve sheath tumors (MPNSTs) are aggressive soft tissue tumors arising sporadically although more frequently in patients with Neurofibromatosis type 1. Prognosis remains dismal as chemo- and radiotherapy have not been shown to be successful. The heparin-binding growth factor, Midkine (MK), is implicated in the tumorigenesis of benign and plexiform neurofibromas, and thereof arising MPNSTs. MK is mitogenic, anti-apoptotic, angiogenic and can promote tumorigenicity in several cell types. Thus, we investigated the role of MK in malignant biology and tumorigenicity in MPNSTs by stable transfection into MPNST cell lines. Overexpression of MK in the MPNST cell line, S462, increased cell viability and protected cells from apoptosis under serum deprivation, but did not induce proliferation. In addition, MK-transfected S462 cells were partially protected from vincristine-induced cell death. Conditioned medium of MK-transfected S462 cells was a potent mitogen for human umbilical venous endothelial cells. Furthermore, MK overexpression in S462 cells was accompanied by higher levels of VEGF mRNA. Yet, stable overexpression of MK in S462 as well as in ST88-14 cells was not sufficient to promote xenograft tumor growth in nude mice. However, increasing survival and enhanced angiogenic potency of MK-transfected S462 cells highlight the importance of developing specific inhibitors for MK as part of new therapeutic concepts against MPNSTs.

Animals↗

Visualisation and navigation methods for typed protein-protein interaction networks.

Protein-protein interactions form large and complex networks. Their visualisation can aid biologists in gaining new insights about the processes in cells and is, therefore, very useful for building sophisticated research tools. Often standard force-directed graph drawing algorithms are used for the visualisation of these networks. However, currently available visual interfaces to biological databases only show general interactions and cannot cope well with more complex networks with different types of interactions. This paper presents a new approach to the visual analysis of protein-protein interaction networks. It uses a combination of circular and force-directed graph drawing algorithms to compute visual representations of protein networks depending on the type of the selected interaction. Smooth transitions between subsequent drawings enable users to explore different functional clusters in these networks without getting lost in the entire network. The visualisation system has been tested with data from the BRITE database.

Algorithms↗