Expression of the castration-induced clusterin protein is incompatible with the survival of human prostate cancer cells in tissues and in cell culture. To investigate the fate of human prostate epithelial cells, when engineered to maintain expression of clusterin protein, we have used an IRES-hyg vector and hygromycin selection. PC-3 prostate tumour cells were substantially more sensitive to clusterin expression than nonmalignant PNT1a cells, showing multiple phenotypic changes including cell cycle arrest and increased apoptosis. The results strengthen the hypothesis that clusterin expression is proapoptotic. Expression of exogenous clusterin in both cell types resulted in its relocation from the cytoplasm and a nuclear accumulation of the protein, as was also seen in the same cells when apoptosis was induced by etoposide treatment. To survive clusterin expression, the PC-3 tumour cells developed apoptosis-inhibitory properties. This could have significance for the resistance of prostate cancers to chemo/radiotherapy, where clusterin overexpression is observed.

Normal (PNT2-C2) and metastatic (PC-3) prostate cell lines were grown in Matrigel to observe the effects on morphology and phenotype in comparison to monolayer culture. In monolayer cultures, PNT2-C2 showed typical round/cuboidal epithelial morphology, with tight cell associations, whereas in Matrigel they formed smooth spheroids, tightly packed with cells. In both monolayer and Matrigel, PNT2-C2 had a differentiated luminal epithelial phenotype with high expression of cytokeratin 8, prostate specific antigen (PSA), prostate specific membrane antigen (PSMA), E-cadherin and desmoglein. In contrast, PC-3 cells possessed an epithelial/mesenchyme morphology in monolayer with loose cell to cell contact and pseudopodial extensions. Immunohistochemical phenotyping indicated the cells were undifferentiated, expressing high levels of vimentin, β1 integrin, CD44 and low expression of cytokeratin 8. In Matrigel they formed smooth and irregular spheroids, which had a lumen surrounded by a single cell layer. Matrigel also influenced the expression of PSA, PSMA and CD44. These results indicate that Matrigel culture can induce morphological differentiation of prostate cancer cells which initially had a basal phenotype. © 2001 Cancer Research Campaign http://www.bjcancer.com

BACKGROUND:Epithelial cells are critically dependent upon cell-matrix and cell-cell adhesion for growth and survival. Anoikis is programmed cell death caused by disruption of cell-substrate adhesion in normal epithelial cells.METHODS:We studied the induction of anoikis in vitro in two cell lines; HaCaT and SW742. PI3K, JAK2 and PKC are key elements in signalling pathways regulating cell survival, and using specific inhibitors we also examined their potential role in the induction of anoikis.RESULTS:When prevented from adhesion by culture on polyHEMA, HaCaT cells underwent apoptosis selectively from the proliferating population; surviving cells underwent cell cycle arrest. In SW742 cells anoikis also occurred, but was balanced by increased cycling. The effects of specific kinase inhibitors indicated that both Janus kinase 2 and protein kinase C partially protect HaCaT cells from anoikis through inducing cell cycle arrest of surviving nonadherent cells; inhibition of Phosphatidylinositol 3-kinase did not induce cycling in HaCaTs prevented from adhesion but did stimulate anoikis. SW742 cells showed markedly different responses: Janus kinase 2 inhibition activated apoptosis directly, Phosphatidylinositol 3-kinase inhibition stimulated both cell cycling and apoptosis, while protein kinase C inhibition stimulated cycling but inhibited apoptosis.CONCLUSIONS:Susceptibility to cell death in adhesion-prevented epithelial cells may thus be regulated by signalling pathways involving Phosphatidylinositol 3-kinase, Janus kinase 2 and protein kinase C. The ability of epithelial tumour cells to invade and metastasize may therefore result from disruption of these pathways.

In vitro models of normal and malignant human prostate are currently limited to a few well established cell lines that, with a single exception (LNCaP), fail to express the androgen receptor (AR) - a common characteristic of prostatic epithelium grown in culture. To investigate the molecular mechanism of action of the non-steroidal antiandrogen Casodex (bicalutamide) against wild-type AR, we have established a transient AR expression model in non-tumorigenic prostate cells of both epithelial and mesenchymal origin. In this model, both dihydrotestosterone and Casodex can effectively transport the AR protein into the nucleus of prostate cells. Whereas the natural ligand, dihydrotestosterone, stabilises the receptor, the AR is rapidly degraded at a nuclear location when the transfected cells are treated with Casodex. In contrast, whereas the mutant AR in the LNCaP line is also degraded on Casodex treatment over the same time period, its intracellular targeting is defective.

The prostate cancer cell lines PC3 and LNCaP have been shown to lack expression of the tumour suppressor gene MMAC1/PTEN, in contrast to the immortalized non-tumorigenic epithelial lines PNT1a and PNT2. We have measured the effects of reintroduction of wild type (wt) and mutant MMAC1 genes on to these genetic backgrounds, using gene constructs expressing either wt MMAC1 or various mutants deficient in the dual specificity phosphatase domain of the protein. Over-expression of wild type PTEN protein induced cell shrinkage and rounding, but did not result in increased levels of classical apoptosis. Permanently transfected lines containing the MMAC1 gene could only be obtained from the PNT cells, as PTEN expression resulted in rapid loss of both tumour lines. In contrast, mutation of the phosphatase domain resulted in partial attenuation of the phenotypic effects of MMAC1 after transient transfection, and also allowed the derivation of permanent tumour cell lines containing the mutated MMAC1 gene. The results suggest that re-expression of wt PTEN is incompatible with survival of human prostate cancer cells in vitro, and that the full biological activity of this common tumour suppressor requires functions additional to the established protein and lipid phosphatase activities in epithelial systems. © 2000 Cancer Research Campaign

The human tumour suppressor gene PTEN/MMAC1/TEP1 encodes a lipid and protein phosphatase. Using RT-PCR, alternatively spliced forms of PTEN mRNA, encoding full-length PTEN and two forms of the protein truncated at the C-terminal end, were detected in normal human tissue. Cultured tumour and non-tumour cell lines show similar splicing patterns.

The presence of prostate-specific antigen (PSA)-positive cells has previously been demonstrated in the peripheral blood of prostate cancer patients by flow cytometry (FC), but the identity of these cells has not been established. In this study, the reverse transcriptase polymerase chain reaction (RT-PCR) was compared with analytical FC in an attempt to detect and characterise these cells. Peripheral blood was obtained from 12 patients with newly diagnosed and untreated prostate cancer and five controls. Nine of the 12 patients with prostate cancer (75%) had circulating PSA-positive cells as shown by FC. Only one of those patients (11.1%) was found to express PSA mRNA by RT-PCR. The absence of PSA mRNA in the majority of samples showing PSA-positive cells suggests that they do not represent haematogenous micrometastases. PSA-positive cells in the blood could represent monocytes that express PSA, either following binding/phagocytosis of free serum PSA or phagocytosis of tumour cells.

PURPOSE:The purpose of this study was to assess qualitatively the expression of adhesion molecules by human retinal pigment epithelium (RPE) and to study their regulation by inflammatory cytokines. These molecular events and the role played by inflammatory cytokines are important for selective lymphocyte trafficking into the eye during uveitis.METHODS:Expression of intracellular adhesion molecule-1 (ICAM-1), endothelial leukocyte adhesion molecule-1 (ELAM-1), platelet endothelial cell adhesion molecule-1 (PECAM-1), and vascular adhesion molecule (VCAM-1) by early passage human RPE cells was assessed by flow cytometry. In addition, the regulation of the expression of these molecules by the inflammatory cytokines interleukin-1 beta (IL-1 beta), IL-6, tumor necrosis factor alpha (TNF alpha), and interferon gamma (IFN gamma) was determined. Reverse transcription-polymerase chain reaction (RT-PCR) was used to characterize further adhesion molecule expression.RESULTS:Flow cytometric analysis determined that ICAM-1 was constitutively expressed on RPE cell lines and that in the presence of TNF alpha, IFN gamma, and IL-1 beta, there was a median fold increase in expression of 4.4, 5.4, and 4.4, respectively. In contrast, flow cytometric analysis of ELAM-1, PECAM-1, and VCAM-1 indicated that these adhesion molecules were not constitutively expressed at the cell surface; only the expression of VCAM-1 was upregulated by the presence of cytokines. The results of RT-PCR on RPE cells indicated that mRNA for all the adhesion molecules was present constitutively in some RPE cultures. After activation with IFN gamma, TNF alpha, and IL-1 beta, RT-PCR analysis showed that the number of RPE cell lines expressing all the adhesion molecules increased.CONCLUSIONS:ICAM-1 expression is markedly upregulated by inflammatory cytokines. Although mRNA for other adhesion molecules is expressed in RPE cells and is enhanced by inflammatory cytokines, this does not necessarily reflect the cell surface protein expression. Thus, the expression of adhesion molecules by RPE cells, and the subsequent recruitment of specific leukocytes, may be determined by the local cytokine environment.

2D-gel electrophoresis was used to investigate protein synthesis in leukaemic cells from a series of 15 chronic lymphocytic leukaemia (CLL) patients, and in non-malignant B-cell populations from different sources. The protein synthesis profiles of CD5+ B-cells from umbilical cord blood and from tonsil were determined, and the levels of expression of their proteins were observed to be similar to the CLL cells. The CD5-cells from cord blood resembled peripheral blood B-lymphocytes, and the protein synthesis profile of CD5-cells from tonsils was very complex. One protein was also identified which consistently appeared to be synthesised at a low level in CD5+ B-cells from tonsil but which was always more prominant in CLL cells and other non-malignant B-lymphocytes. On the basis of these data it is possible that the closest non-malignant counterpart to CLL is the CD5+ B-lymphocyte from cord blood.

Objective. The aim of this study was to establish whether reactive oxygen species, generated during oxidation of amines, catalyzed by polyamine oxidase (PAO) and diamine oxidase (DAO) in cervical secretions may play a role in the etiology of cervical cancer.Methods. Cervical mucus was obtained from women attending the gynecological outpatient department: 139 with and 154 without cytological evidence of cervical intraepithelial neoplasia were recruited. The mucus was freeze dried in liquid nitrogen, weighed, and later resuspended for assay of PAO and DAO concentrations using a chemiluminescence method. The two groups were compared by group sequential analysis using PEST3 software.Results. Patients with a colposcopic diagnosis of a high-grade squamous intraepithelial lesion (SIL) had significantly higher enzyme activities than control cases (LNPAO 1.37 (0.37) versus 1.18 (0.35): Student t test: P < 0.001; LNDAO 1.37 (0.36) versus 1.15 (0.37): Student t test: P < 0.001).Conclusion. It is probable that this rise in enzyme activity precedes cytological changes and plays some part in the etiology of cervical cancer, as the cells that undergo premalignant change are normally squamous in origin, whereas mucus is a product of columnar epithelium. Higher enzyme activity in patients with SIL than in controls may be a reflection of higher risk of exposure to amine substrates in semen through multiple sexual partners.

Protein synthesis profiles of leukaemic cells from 15 chronic lymphocytic leukaemia (CLL) patients were analysed by 2D-electrophoresis of 35S-methionine labelled proteins. This series of CLL included patients with stage A (7), B (4) and C (4) disease. Although the protein synthesis profiles were similar in all cases, some consistent differences were noted between the different stages. The levels of synthesis of three proteins (approximately 35 kD size) were of particular interest. Two of these were always expressed in stage C CLL but either infrequently or not at all in stage A or B CLL. By contrast a third protein was expressed at a much reduced level in stage C compared to stages A or B. This type of analysis could prove invaluable for identifying proteins whose expression was intimately associated with the evolution of CLL.

Protein synthesis was analysed in leukaemic cells from 10 chronic lymphocytic leukaemia (CLL) patients by 2D-gel electrophoresis of 14C-labelled proteins. There appeared to be only minor differences between each of the CLL samples, but there was evidence that the level of expression of a few of the proteins might have correlated to the stage of the disease. Comparison of the CLL samples to populations of normal B-lymphocytes demonstrated marked differences in protein synthesis between the leukaemic and non-malignant cells. We subsequently used the fluorescence activated cell sorter (FACs) to separate CD5+ from CD5- B-lymphocytes, but observed that the protein synthesis exhibited by these two populations was essentially the same, and both were very different to that observed in CLL cells. The significance of these observations with respect to the origins of CLL is discussed.

Aberrant expression of c-myc has been implicated in the development of colorectal carcinomas. We have used monoclonal antibodies 6E10 and 9E10, raised against mid-sequence and C-terminal peptides of the c-myc protein, to study the distribution of myc protein in normal and diseased bowel at the light microscope and ultrastructural levels. Normal mucosa showed staining only of some nuclei in the proliferative zones of crypts. In adenomas, staining varied from predominantly nuclear to pancellular to focal or pancytoplasmic. Moderately well differentiated areas of carcinomas gave strong focal cytoplasmic staining, while in poorly differentiated tumours staining was pancytoplasmic. Electron microscopy with these antibodies detected myc protein associated with dense chromatin and, where cytoplasmic staining occurred, with polyribosomes. Tumours showed a reduced staining of nuclear pores compared with normal tissue. Comparison of staining patterns with 6E10 and 9E10 in normal tissue, adenomas, and tumours suggests that tumour progression is associated with an accumulation of cytoplasmic c-myc protein, perhaps resulting from alterations to the C-terminus which reduce the efficiency of nuclear targeting of the protein and thus disrupt the regulation of the cell cycle.

Over-expression and abnormal intracellular location of the product of the oncogene c-myc in colonic dysplasia and neoplasia may be related to alterations in epigenetic mechanisms controlling the functioning of this gene. We have investigated the methylation patterns of the c-myc oncogene in human colorectal tissue representing various stages of dysplasia and neoplasia, including metastasis to liver, omentum and lymph node. Comparison of normal and neoplastic tissues from the same patient showed a decrease in methylation in a specific CCGG site in the third exon of c-myc through the progression from normal via dysplastic to neoplastic and metastatic tissue. Quantitative analysis revealed that in colonic adenocarcinomas an average of 66.1% and in metastatic deposits 83.1% of the myc gene DNA was hypomethylated at this site, as compared to a value of 9.2% in normal colonic mucosa. Adenomatous polyps showed an average value of 50.5% and hyperplastic polyps, 24.8%. The results suggest that partial hypomethylation of the c-myc gene third exon is associated with cell proliferation, and that deregulation of proliferation may be linked to the high levels of hypomethylation, presumably involving both copies of the gene in some cells, which occur at a relatively early stage in neoplastic progression.

We have identified two components of human seminal plasma which suppress natural killer (NK) cell activity in vitro. Dialysis and gel filtration experiments have shown both components to be of low molecular weight. The first will suppress NK cells following a short period of pretreatment, but this suppression is dependent upon the presence of bovine serum in the medium and is directly related to a loss of cell viability as measured by trypan blue dye exclusion. We suggest that this molecule is a polyamine. The second factor will not suppress NK activity following pretreatment of lymphocytes, but is a potent suppressor when added for the duration of the assay. This suppression is completely bovine serum independent, unrelated to toxicity and appears to be mediated by prostaglandin E2. The relevance of these results to a clinical situation is discussed.