Development and progression of prostate cancer is a multistep process of cumulative genetic damage, acquired during a life-time. However, the altered genotype acts against an appropriate background of epigenetic control mechanisms. Several mechanisms of mitotically heritable, epigenetic control of differential gene transcription have been noted, such as stromal-epithelial and cell-cell interactions. In prostate cancer, an important, supporting and/or inhibiting role of stromal-epithelial interaction has been implicated in tumor growth, angiogenesis and metastasis, which includes cell proliferation, adhesion and motility. Within these processes, data mainly obtained in systems other than the prostate have shown a crucial (regulatory) role of proteoglycans (PGs) acting at the level of cell-cell and cell-pericellular matrix interactions. Although little information has been recorded from normal, benign hyperplastic and malignant prostate tissue, PGs are components of both the cell surface and the extracellular matrix (ECM) that form associations with other molecules, such as fibronectin and laminin. On the basis of cell-ECM adhesion/interaction as a prerequisite for both cell proliferation and motility, and the involvement of PGs, the purpose of this study was to investigate the possible biological relevance of (free) glycosaminoglycans (GAGs), as major functional substructures of PGs, on cell adhesion of a series of human prostatic cell lines cultured in vitro. The effects of a series of exogenously applied GAGs on cell adhesion and proliferation were studied in the human cell lines LNCaP. DU 145 and PC-3, cultured on tissue culture plastic as substratum. The applied GAGs were the natural GAGs, heparin, heparan, dermatan, chondroitin-4 and chondroitin-6 sulfate, and the semisynthetic, GAG-like pentosan polysulfate (PPS). Addition of GAGs (1–300 μg/ml) to cultures that were allowed to adhere for 24 h prior to GAG addition did not affect cell proliferation. In contrast, whereas the natural GAG added during cell adhesion had no effect. PPS strongly inhibited proliferation of LNCaP and DU145, but not the less anchorage-dependent PC-3 cells. Under the latter conditions, after 6 days of culturing the IC50 of proliferation were determined to be < 1 and 50 μg PPS/ml for LNCaP and DU145, respectively, corresponding with a profound effect on cell morphology. Direct measurements of cell adhesion confirmed that, in contrast to the natural GAGS, PPS inhibited cell adhesion. In conclusion, the interference of a nonnatural, GAG-like structure with cell adhesion may be interpreted as the involvement of PGs of the cell surface in cell adhesion, possibly affecting the various processes (proliferation, angiogenesis and metastasis) of prostate tumor progression. Although similar interferences of nonnatural GAGs with cell-adhesion-associated proliferation of anchorage-dependent cells remain to be established under in vivo conditions, this type of compounds deserves further attention as a tool with which to study the role of cell adhesion in the progression of prostate cancer and as a potential candidate for the development of a stromal-epithelial targeted therapy. [ABSTRACT FROM AUTHOR]