Cell cycle perturbation by sodium butyrate in tumorigenic and non-tumorigenic human urothelial cell lines assessed by flow cytometric bromodeoxyuridine/DNA analysis.

The effect of sodium butyrate on cell proliferation was studied in eight human urothelial cell lines differing in transformation grade (TGr): Hu 1752 (mortal, TGr I); HCV29 (immortal and tumorigenic, TGr II); HCV29T, T24, T24A, T24B, Hu 961A and Hu 1703He (tumorigenic, TGr III). In all cell lines, except Hu 1752, addition of 4 mM sodium butyrate at 18 h after replating resulted in a significantly decreased population of adherent cells after a further 24-48 h. This might partially be explained by detachment of cells, probably mainly S phase cells, from the substrate in the lines HCV29, HCV 29T, Hu 961A and Hu 1703He. Flow cytometric DNA analysis of the adherent cell population showed that all TGr II and III urothelial cell lines were DNA aneuploid, and that butyrate perturbed the cell cycle distribution in these cell lines, mainly by a decrease of the S phase fraction. Flow cytometric bromodeoxyuridine (BrdUrd)/DNA analysis of continuously BrdUrd labelled cultures, using a 'washless' BrdUrd/DNA staining technique, showed that butyrate inhibited the G0/1-S phase transition, indicated by a delayed depletion of BrdUrd negative G0/1 cells in the cell lines HCV29, HCV29T, T24B, Hu 961A and Hu 1703He. BrdUrd/DNA analysis further showed that butyrate inhibited the G2M-G0/1 phase transition, indicated by a pronounced accumulation of BrdUrd positive G2M cells in the cell lines HCV 29T, T24B, Hu 961A and Hu 1703He. Microscopy of HCV29T and Hu 961A cells indicated that this block did not occur in mitosis. The parental cell line T24 and the cell line T24A did not show an accumulation of BrdUrd negative G0/1 cells or BrdUrd positive G2M cells like that occurring in the derived cell line T24B.

Identity of tumorigenic human urothelial cell lines and 'spontaneously' transformed sublines.

Restriction fragment length polymorphism (RFLP) analysis, comparative marker chromosome analysis, and polymorphic enzyme analysis was carried out on a total of eight human urothelial cell lines and sublines selected according to our knowledge of their HLA-A,B phenotype. RFLP analysis and cytogenetic analysis showed that the cell lines Hu1703He, Hu1922, and T24 are genuine cell lines of different origin. The identity of Hu1703He could not be confirmed by its isozyme phenotype which was identical to the T24 phenotype. RFLP analysis and isozyme analysis revealed that three cell lines, Hu456, Hu549, and Hu961a, and two transformed sublines, HCV-29Tmv and Hu609Tmv, are sublines of T24. A common origin of Hu456, Hu549, Hu961a, HCV-29Tmv, and Hu609Tmv was confirmed by marker chromosome analysis. However, the T24 origin of these cytogenetically related cell lines was not supported by chromosome analysis of T24. RFLP analysis and HLA phenotyping of two tumorigenic and invasive sublines isolated from a culture of non-tumorigenic Hu609 cells showed that non-tumorigenic Hu609 cells can transform 'spontaneously' in vitro into tumorigenic Hu609T cells. The results emphasise the need for careful monitoring and screening of cell lines for their identity using more than one identification parameter.

Identity of non-malignant human urothelial cell lines classified as transformation grade I (TGrI) and II (TGrII).

Restriction fragment length polymorphism (RFLP) and marker chromosome or isozyme analysis were used to study the identity of 7 human urothelial cell lines selected according to their HLA-A,B phenotype. The results confirmed that the cell lines Hu961b, Hu1125, Hu1694, Hu1752, Hu609 and HCV-29 were genuine cell lines of different origin, while Hu1734 and HCV-29 were of the same origin. Hu609 was recorded to be derived from a female. However, the presence of a Y chromosome indicated either false labelling or cross-contamination. Analysis of a series of frozen stocks of Hu609 cultures showed that "malignant transformation" of these cultures was due to cross-contamination.

Two phorbol ester receptor affinities in partially transformed human urothelial cells and decrease of receptor binding in desensitized cells.

The presence of specific binding sites for phorbol esters was studied in a transformed but non-tumorigenic human urothelial cell line HCV-29 by assay of specific binding of 3H-phorbol-12,13-dibutyrate (3H-PDBu) to intact living cells. 3H-PDBu bound specifically to HCV-29 cells in a saturable and competitive manner. Scatchard plot analysis of specific binding yielded a curved plot consistent with two binding sites with Kd of 11 nM and 102 nM, respectively. At saturation the corresponding PDBu binding capacities (Bmax) were 8.8 pmol/10(6) cells (5.2 x 10(6) molecules bound per cell) and 2.8 pmol/10(6) cells (1.7 x 10(6) molecules bound per cell). 3H-PDBu binding was displaced by biologically active phorbol ester tumor promoters such as 12-O-tetradecanoylphorbol-13-acetate (TPA) and mezerein, but not by tumor promoters such as L-tryptophan, anthranilic acid and sodium saccharin. In cells desensitized by pretreatment with 1 microgram/ml (2 microM) TPA or PDBu for 24 h the level of binding was reduced to 28% of the level in non-exposed cells. The ability of desensitized cells to bind 3H-PDBu was gradually restored within 5-6 days. At the same time the cells became sensitive to the morphological alteration induced by PDBu. This suggests that desensitization of HCV-29 cells is due to a decreased receptor-ligand binding capacity probably associated with down regulation of the phorbol ester receptors.

Age-related decrease in transplantability of human tumours in nu/nu mice.

Experiments were designed to assess age-related changes in the transplantability of human tumours xenografted in congenitally athymic (nu/nu) mice. It has been found that the number of progressively growing human tumour xenografts decreased significantly with increasing age of BALB/c nu/nu recipients. These findings, taken together with a previously recognized increase in the frequency of endogenous interleukin 2 (IL-2)-producing cells with age of nu/nu mice, prompted us to investigate whether administration of exogenous IL-2 to young adult nu/nu mice could change the transplantability of human tumours in the mice. Peritumoral administration of exogenous interleukin 2 to 8-week-old nu/nu mice inhibited the growth of the human tumour xenografts. In vitro activation of nu/nu splenocytes with exogenous Il-2 resulted in the generation of killer cells which have been found to be cytolytic when allowed to react with human tumour targets in 51Cr cytotoxicity assay. In addition, it has been found that the percentage of IL-2-activated Thy 1.2+ and ASGM1+ cells substantially increased with increasing age of nu/nu spleen cell donors. These findings are compatible with the hypothesis that the observed age-related decrease in takes of human tumour xenografts might be determined by the increasing level of IL-2 production and subsequent maturation of IL-2-dependent effector cells.

Expression of polymorphic HLA-A,B epitopes in human urothelial cell lines.

Malignant human urothelial cell lines propagated in vitro have previously been demonstrated to express low amounts of monomorphic HLA-A,B,C as compared to premalignant urothelial cells. In this study the expression of polymorphic HLA-A,B epitopes in human urothelial cell lines have been investigated in greater detail. The expression of HLA-B locus coded epitopes in malignant TGrIII cells was demonstrated to be low or absent as compared to pre-malignant TGrII or slightly transformed TGrI cells, suggesting a mechanism by which malignant cells could escape from the host immune response. The extreme polymorphism of HLA-A,B,C antigens suggests that HLA typing could be used as a method to identify the origin of cell lines which is essential in the study of the process of malignant transformation in vitro, or when correlating in vitro data with clinical observations of the patient. Two urothelial cell lines classified as slightly transformed (TGrI) and two as pre-malignant (TGrII) could, according to their expression of polymorphic HLA-A,B epitopes, be identified as genuine independent cell lines. One TGrII cell line previously designated Hu1734 shared the same HLA-A,B phenotype as the genuine HCV29 (TGrII) cell line and is therefore suspected to be a subline of the latter. Out of 18 cell lines and sublines classified as TGrIII the fidelity of four (Hu1922 and three sublines of T24) was proved by their HLA-A,B phenotype. Mistaken identity either by contamination or false labelling of the cultures was suspected in samples of six TGrIII cell lines and sublines. In four of these the HLA-A,B type characteristic for the T24 cell line could be demonstrated, but in general HLA typing of TGrIII cell lines as a method to identify the origin of the individual cell lines failed, primarily due to the decreased expression of HLA-A,B,C antigens and the apparent selective loss of HLA-B locus coded antigens.

Differences in adhesion of urothelial cells to a panel of substrates are characteristic for established human urothelial cell lines differing in their transformation grade.

Eight human urothelial cell lines established and propagated in the Fibiger Institute, differing in their transformation grade as described by Christensen et al were brought into suspension and contacted with a panel of seven substrates, ie. Con. A, Fibronectin, WGA, Collagen IV, Laminin, PNA and artificial basal membrane. Cell adhesion was measured on big cell populations with the use of flow cytophotometry, measured samples contained nonadhering cells and fluorescent beads, added as an internal standard. It was found that the cell lines differed from each other by the kinetics of cell adhesion to the chosen panel. A higher percentage of cells belonging to established cell lines adhered to PNA, WGA or Con A coated surfaces than to laminin, fibronectin or collagen IV. The difference in adherence of cells to fibronectin, laminin or collagen IV coated surfaces makes differentiation between the single cell lines possible. Discriminant analysis of the total patterns of adhesion of all cell lines was used for comparison. The effect of trypsinization on the structure of the cell surface, accompanied by possible changes in adherence, was evaluated by comparison of the immediate results with those obtained with cells after 3 hours of recovery at 37 degrees C. It was found that all cell lines showed a better adherence to laminin after recovery. This is consistent with the published data on the high sensitivity of laminin receptors to trypsinization. On the other hand, the receptors for PNA, WGA and Con A were less expressed after the recovery period, as evidenced by a decreased adherence of the cells to these three substrates. The effect of recovery of cells on the amount of their receptors for fibronectin and collagen IV was different in different cell lines, but the direction of change for this pair of receptors was identical in a given cell line. In some instances the recovery period had no effect whatsoever on cell adherence. The different pattern of adhesion of cells belonging to established cell lines to the chosen panel of substrates can be used as a parameter for characterization of single cell lines, but it does not correlate with their grade of transformation.

Reduced HLA-A,B,C expression in tumourigenic v-raf transfected human urothelial cells.

Tumourigenic (TGrIII) human urothelial cells grown in vitro have previously been demonstrated to have a markedly decreased expression of beta 2-microglobulin and HLA-A,B,C antigens as compared to non-tumourigenic (TGrII) human urothelial cell lines. Furthermore, during 'spontaneous' in vitro transformation of a non-tumourigenic (TGrII) human urothelial cell line Hu609 into a tumourigenic (TGrIII) subline Hu609T/LLH, changes in morphology and tumourigenicity have been demonstrated to be accompanied by a decreased HLA-A,B,C expression. After malignant transformation of the non-tumourigenic (TGrII) human urothelial cell line HCV29 by DNA transfection with the v-raf oncogene, four sublines could be isolated. In this study we have investigated these sublines for their expression of membrane bound HLA-A,B,C antigens and provide further evidence that an inverse relationship exists between tumourigenicity and monomorphic HLA-A,B,C expression. Treatment of the cells with recombinant human interferon alpha for 3 days increased the expression of HLA-A,B,C antigens by 50-150% indicating that at least some of the reduced HLA-A,B,C expression could be due to decreased synthesis of HLA-A,B,C antigens. All the transfected cell lines overexpress v-raf and c-myc.

Alpha-2-macroglobulin is not synthesized by human urothelial cell lines in vitro.

We have analyzed 14 human urothelial cell lines in two different transformation grades for the production of a tumor-associated-alpha-2-macroglobulin. It has been shown in radioimmunoprecipitation and immunoblotting tests that human urothelial cells in vitro do not synthesize the wide-spectrum proteinase inhibitor alpha-2-macroglobulin. Four of the tested cell lines were additionally tested for the expression of specific mRNA which resulted in negative findings. The significance of the absence of alpha-2-macroglobulin in the human urothelial cells is discussed.

Recombinant human interferon gamma exerts an anti-proliferative effect and modulates the expression of human leukocyte antigens A,B,C and DR in human urothelial cell lines.

In this study we have treated three malignant (TGrIII) and two pre-malignant (TGrII) urothelial cell lines with recombinant human interferon gamma (rHu-INF gamma). The malignant cells (HCV29-T112C1, Hu1703He and T24) were inhibited in growth by more than 50% after treatment with 100-1000 units of rHu-INF gamma/ml for 4 days as compared to untreated controls. The growth of the pre-malignant cell lines (HCV29 and Hu609) was not influenced to the same extent in the presence of rHu-INF gamma in the culture medium. Treatment with rHu-INF gamma increased the expression of monomorphic human leukocyte antigens (HLA) A,B,C as well as beta 2-microglobulin in all the cell lines tested, as demonstrated using a quantitative immunofluorescence assay. The tumourigenic cell lines increased their expression of HLA in a dose-dependent way, whereas treatment of the non-tumourigenic cells with higher concentrations of rHu-INF gamma than 10 units/ml, did not increase the HLA-A,B,C expression further. None of the cell lines expressed HLA-DR unless treated with rHu-INF gamma. No correlation between tumourigenicity and the dose of rHu-INF gamma required for "de novo" induction of HLA-DR could be demonstrated. After removal of rHu-INF gamma from the medium, the expression of HLA-DR gradually decreased in less than 14 days, indicating that the expression of HLA-DR is not constitutive but dependent upon the presence of rHu-INF gamma. We conclude that human urothelial cells grown in vitro are sensitive to the anti-proliferative and major-histocompatibility-complex-modulating effects of rHu-INF gamma, and that malignant urothelial cells are more sensitive than pre-malignant cells. Finally, our data indicate a possible role for rHu-INF gamma in the management of human bladder cancer.

Pyruvate kinase inhibited by L-cysteine as a marker of tumorigenic human urothelial cell lines.

It was found that a decrease in electrophoretic mobility of pyruvate kinase (PK) isoenzyme, and an increase of the sensitivity of this enzyme to L-cysteine, were markers of immortalization and tumorigenic properties, respectively, in human urothelial cell lines characterized by different grades of transformation (TGr) in vitro.

Spreading of cells on various substrates evaluated by Fourier analysis of shape.

In order to evaluate in mathematical terms the morphological changes occurring in the course of cell spreading, Fourier analysis of shape was applied. Human urothelial Hu 961 b cells plated on type IV collagen, fibronectin, laminin, glass and bovine serum albumin (BSA) were studied. Fourier parameters describing cell shape as well as surface areas covered by the cells on the substrate were subjected to statistical analysis. Using analysis of variance and discriminant analysis it was found that parameters describing cell shape (both gross shape of cells and their fine scale contour foldings) possessed a higher power of discrimination between the cells spread on various substrates than the differences in cell surface areas. In the course of observation (75 and 150 min) the highest number of attached cells and highest degree of spreading were found when cells were plated on type IV collagen. Moderate alterations in cell shape and moderate increase of surface area were seen in the group of cells seeded on fibronectin, whereas the cells plated on laminin, glass and BSA revealed a moderate increase of surface area, but no changes in their shape were observed. The differences in attachment of cells and in the degree of their spreading might be due to the variation in expression of plasma membrane receptors for various substrates. The Fourier analysis of cell shape coupled with measurement of surface area is a good tool for quantitative evaluation of cell spreading and can be used for discrimination between cells spread on different substrates.

The Thomsen-Friedenreich antigen on human urothelial cell lines detectable by peanut lectin and monoclonal antibody raised against human glycophorin A.

The Thomsen-Friedenreich (TF) antigen is a cryptic disaccharide structure on human erythrocytes which can be exposed by neuraminidase treatment and which is supposed to be expressed in an unmasked form on some carcinoma cells. For its detection in addition to auto-, allo- and heteroantisera, PNA (peanut lectin) is being applied. In the present studies the mouse monoclonal antibody (MoAb) raised to asialoglycophorin from human erythrocytes was used. The MoAb 22.19 is of mouse IgM isotype and is specifically binding to beta-D Gal-1-3 alpha-D GalNAc. The human urothelial cell lines maintained and characterized earlier were analyzed using indirect immunofluorescence assays. Among the spectrum of cell lines tested, five out of six cell lines belonging to the transformation grade III category (invasive in vitro and tumorigenic in nude mice) expressed TF antigen. The relationship between expression of TF antigen and other earlier defined biological traits related to malignant phenotype is discussed.

Local administration of cells containing an inserted IL-2 gene and producing IL-2 inhibits growth of human tumours in nu/nu mice.

We have prepared a retroviral expression construct, pPS-IL-2, in which human IL-2 cDNA has been inserted into the polylinker region, and have used the retroviral vector to introduce the functional IL-2 gene into a fibroblast cell line, RAT-1. Peritumoral administration of IL-2-producing RAT-1 cells into congenitally athymic (nu/nu) mice carrying subcutaneous transplants of human carcinoma cells inhibited the growth of the human tumour xenografts.

Defect in lectin-induced interleukin 2 production by peripheral blood lymphocytes of patients with invasive urinary bladder carcinoma.

The production of interleukin 2 (IL-2) by phytohaemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMC) from 21 patients with transitional cell carcinoma of the urinary bladder (BTCC) and 16 control blood donors was measured with a solid phase enzyme immunoassay based on the dual antibody immunometric sandwich principle. PBMC from patients with invasive BTCC (grade III-IV) showed a defect in the production of IL-2. The concentration of IL-2 in the supernatants of PBMC cultures from these patients was substantially lower (0.4 +/- 0.1 U/ml) than that observed in the supernatants of PBMC cultures from patients with non-invasive BTCC, grade II (1.5 +/- 0.7 U/ml), and from tumour-free controls (1.4 +/- 0.8 U/ml). These results suggest an immune dysfunction based on quantitatively impaired IL-2 production in patients with invasive BTCC and indicate that exogenous IL-2 could be used as an immunological response modifier for the treatment of these patients.

Correlation between classification of human urothelial cell lines and HLA-A,B,C expression.

Quantitative changes in major histocompatibility class I antigen expression in tumour cells are believed to affect the host immune response against the tumour. In tumourigenic (TGrIII) human urothelial cell lines the apparent loss of polymorphic HLA-A,B epitopes has previously been demonstrated. In the present study, 3 non-tumourigenic (TGrII) and 6 tumourigenic (TGrIII) human urothelial cell lines have been investigated for their quantitative expression of monomorphic HLA-A,B,C and B2-microglobulin. Evidence is provided that an inverse correlation exists between tumourigenicity and HLA-A,B,C and B2-microglobulin expression. Furthermore, treatment of the cells with neuraminidase partly restored the expression of monomorphic HLA-A,B,C suggesting that at least some of the observed quantitative differences could be due to masking of the membrane bound HLA antigens by sialic acid-containing glycoconjugates.

In vitro studies of human bladder cancer.

Explantation and cultivation in vitro of 55 human biopsies of histologically normal urothelial tissue, and 63 biopsies from transitional cell carcinomas (TCC) revealed a primary success rate of about 75% and a secondary success rate of about 50% with no significant differences, between normal and TCC derived cultures. While only half of the subcultured biopsies of normal origin were able to grow for more than 4 passages all TCC derived cultures were able to do so. Based on morphology, life span, invasive growth into co-cultured fragments of embryonic chick heart, and tumorigenicity in nude mice the following transformation grades (TGr) were defined: Normal (TGr 0), non-malignant with prolonged, but finite life span (TGr I), premalignant with infinite life span but without tumorigenic and invasive potential (TGr II), malignant with infinite life span and tumorigenic and invasive properties (TGr III). Differentiation between TGr I - TGr III was possible by several other parameters, including cytological characteristics, molecular and immunological surface structure, relative oncogene expression, pyruvate kinase isoenzymes, DNA Repair, and growth pattern. However, in addition to these common characteristics, the individual cell lines of the various categories of transformation showed differences that allowed the identification of well defined cell lines and sublines.