Expression of endogenous receptors for neoglycoproteins in Dunning R3327 rat prostatic carcinoma

A proteome study of secreted prostatic factors affecting osteoblastic activity: galectin-1 is involved in differentiation of human bone marrow stromal cells

Prostate cancer cells metastasize to bone causing a predominantly osteosclerotic response. It has been shown that cells from the human prostate cancer cell line PC3 secrete factors that influence the behavior of osteoblast-like cells. Some of these factors with mitogenic activity have been found to be proteins with molecular weights between 20 and 30 kDa, but the identity of the osteoblastic mitogenic factor or factors produced by prostate cancer cells is still unknown. Therefore, the aim of this study was to characterize the protein profile of conditioned medium (CM) from PC3 cells in the molecular weight range from 5 to 30 kDa using proteome analysis. A protein profile of the CM from PC3 cells was performed by two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). Thirty protein spots with molecular weights ranging from 5 to 30 kDa were analyzed by matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS). One of these spots was identified as galectin-1. We examined whether PC3 CM, recombinant galectin-1 alone, or combined with insulin-like growth factor-I (IGF-I) had any effects on the proliferation or differentiation of human bone marrow stromal (hBMS) cells. Furthermore, we tested whether adhesion of PC3 cells to plastic, laminin, fibronectin, and collagen type I was influenced by lactose, which inhibits galectin-1. Galectin-1 (1000 ng/ml) inhibited the proliferation of hBMS cells up to 70 +/- 12% (treated/control) of control in contrast to PC3 CM, which induced hBMS cell proliferation by 3-fold. This effect was abolished by IGF-I. PC3 CM and galectin-1 in concentrations of 10 and 1000 ng/ml increased the alkaline phosphatase (ALP) activity of hBMS cells up to 175 +/- 27%, 137 +/- 8%, and 131 +/- 11%, respectively, compared with ALP activity of untreated cells, and inhibited the secretion of osteocalcin (OC) up to 81 +/- 3%, 93 +/- 1%, and 58 +/- 2%, respectively, compared with OC secretion of untreated cells. These effects were affected by IGF-I. Lactose inhibited adhesion of PC3 cells to plastic, fibronectin, laminin, and collagen type I up to 58 +/- 4%, 30 +/- 12, 72 +/- 9%, and 86 +/- 4%. In conclusion, galectin-1 modulated osteoblastic proliferation and differentiation. These effects were affected by IGF-I. Thus, galectin-1 is likely be involved in the osteoblastic response, caused by prostate cancer cells metastasizing into bone, by affecting the matrix mineralization.

Basal cells of H-Dunning tumor are myoepithelial cells. A comparative immunohistochemical and ultrastructural study with male accessory sex glands and mammary gland

Recent immunohistochemical studies have shown that basal cells in human prostatic epithelium are not myoepithelial cells. Since in the literature the Dunning tumor, originally described as a rat prostate carcinoma derived from the dorsolateral prostate of a Copenhagen rat, was reported to have myoepithelial cells, a comparative immunohistochemical and ultrastructural study was performed in the H-, HIF- and AT3-lines of the Dunning tumor, the male accessory sex glands (ventral, dorsal, lateral prostate, coagulating gland, bulbourethral gland) and the mammary gland of both Copenhagen and Wistar rats. Mono- and polyclonal antibodies directed against intermediate filament proteins (cytokeratin, desmin, vimentin) and the contractile proteins (alpha-actin, muscle type specific myosin, tropomyosin) were used along with phalloidin decoration of F-actin. As in the human prostate, none of the rat prostate lobes in either strains did contain basal cells expressing cytokeratin along with alpha-actin, myosin and tropomyosin Cells representing fully differentiated myoepithelial cells, however, were present as anticipated in the mammary gland, the bulbourethral gland and the H-tumor line of the Dunning tumor. This finding is difficult to reconcile with the contention of a prostatic origin of the H-Dunning tumor. Further studies are required to classify the epithelial parental tissue in order to define the true origin of the H-Dunning tumor and the tumor lines derived thereof.