Edelfosine protects precultured heart fragments against the invasion of malignant cells through altered sialylation.

1-O-octadecyl-2-O-methylglycero-3-phosphocholine (ET-18-OMe)-treated precultured heart fragments (PHF) are resistant to the invasion of malignant cells. Previous studies have demonstrated that this effect is due to the alterations of the N-linked glycoproteins in PHF after 48-h ET-18-OMe treatment. Moreover, the observed effect was still present seven days after ET-18-OMe was omitted. The present study reveals that approximately 13.4% of the administered ET-18-OMe was taken up by PHF and about 75% of the initial uptake was still present after ET-18-OMe was removed. In addition, we found significant changes in the sialic acid content and sialyltransferase activities in both conditions. Overall, these results clearly demonstrate that the uptake and retention of ET-18-OMe are responsible for the resistance to the invasion of malignant cells due to the altered sialylation of the cell surface glycoproteins in PHF.

Bile acids stimulate invasion and haptotaxis in human colorectal cancer cells through activation of multiple oncogenic signaling pathways.

Bile acids are implicated in colorectal carcinogenesis as evidenced by epidemiological and experimental studies. We examined whether bile acids stimulate cellular invasion of human colorectal and dog kidney epithelial cells at different stages of tumor progression. Colon PC/AA/C1, PCmsrc, and HCT-8/E11 cells and kidney MDCKT23 cells were seeded on top of collagen type I gels and invasive cells were counted after 24 h incubation. Activation of the Rac1 and RhoA small GTPases was investigated by pull-down assays. Haptotaxis was analysed with modified Boyden chambers. Lithocholic acid, chenodeoxycholic acid, cholic acid and deoxycholic acid stimulated cellular invasion of SRC- and RhoA-transformed PCmsrc and MDCKT23-RhoAV14 cells, and of HCT-8/E11 cells originating from a sporadic tumor, but were ineffective in premalignant PC/AA/C1 and MDCKT23 cells. Bile acid-stimulated invasion occurred through stimulation of haptotaxis and was dependent on the RhoA/Rho-kinase pathway and signaling cascades using protein kinase C, mitogen-activated protein kinase, and cyclooxygenase-2. Accordingly, BA-induced invasion was associated with activation of the Rac1 and RhoA GTPases and expression of the farnesoid X receptor. We conclude that bile acids stimulate invasion and haptotaxis in colorectal cancer cells via several cancer invasion signaling pathways.

The role of bile acids in carcinogenesis.

Bile acids play a role in colorectal carcinogenesis as evidenced by epidemiological and experimental studies. Some bile acids stimulate growth of normal colonic and adenoma cells, but not of colorectal cancer cells. Moreover, bile acids stimulate invasion of colorectal cancer cells, at least in vitro. One possible mechanism of action is bile acid-induced DNA binding and transactivation of the activator protein-1 (AP-1) by co-operate activation of extracellular signal-regulated kinases (ERKs) and PKC signaling. In the present paper, we review the mechanisms by which bile acids influence carcinogenesis.

Alkyl-lysophospholipid 1-O-octadecyl-2-O-methyl- glycerophosphocholine induces invasion through episialin-mediated neutralization of E-cadherin in human mammary MCF-7 cells in vitro.

1-O-octadecyl-2-O-methyl-glycerophosphocholine (ET-18-OMe) is an analogue of the naturally occurring 2-lysophosphatidylcholine belonging to the class of antitumor lipids. Previously, we demonstrated that ET-18-OMe modulates cell-cell adhesion of human breast cancer MCF-7 cells. In the present study, we tested the effect of ET-18-OMe on adhesion, invasion and localisation of episialin and E-cadherin in MCF-7/AZ cells expressing a functional E-cadherin/catenin complex. The MCF-7/6 human breast cancer cells were used as negative control since their E-cadherin/catenin complex is functional in cells grown on solid substrate but not in suspension. The function of E-cadherin, a calcium-dependent transmembrane cell-cell adhesion and signal-transducing molecule, is disturbed in invasive cancers by mutation, loss of mRNA stability, proteolytic degradation, tyrosine phosphorylation of associated proteins and large cell-associated proteoglycans or mucin-like molecules such as episialin. Episialin, also called MUC1, is an anti-adhesion molecule that by its large number of glycosylated tandem repeats can sterically hinder the adhesive properties of other glycoproteins. ET-18-OMe inhibited the E-cadherin functions of MCF-7/AZ cells as measured by inhibition of fast and slow aggregation and by the induction of collagen invasion. These effects were enhanced by MB2, an antibody against E-cadherin and blocked by monoclonal antibodies (MAbs) 214D4 or M8 against episialin. ET-18-OMe had no influence on tyrosine phosphorylation of beta-catenin and the E-cadherin/catenin complex remained intact. Transcription, translation, protein turnover and cell surface localisation of episialin were not altered. ET-18-OMe induced finger-like extensions with clustering of episialin together with E-cadherin and carcinoembryonic antigen but not with occludin. In cells in suspension, ET-18-OMe caused a shift in the flow-cytometric profile of episialin toward a lower intensity for MCF-7/AZ cells. In contrast with MCF-7/AZ cells, the adhesion-deficient and noninvasive MCF-7/6 cells showed neither morphotypic changes nor induction of aggregation nor invasion in collagen I upon treatment with ET-18-OMe. Co-localisation of episialin with E-cadherin was rarely observed. We conclude that in the human breast cancer cells MCF-7/AZ, E-cadherin and episialin are key molecular players in the regulation of promotion and suppression of cell-cell adhesion and invasion.

Cell aggregation assays.

Invasion of carcinoma cells is the result of a disequilibrium between invasion promoter and invasion suppressor gene products (1). The E-cadherin/catenin complex is the most potent invasion suppressor at the cell membrane of epithelioid cells (2).This complex consists of E-cadherin, a transmembrane glycoprotein of 120 kDa, which is linked to the actin cytoskeleton via the catenins (3). Downregulation of the complex is a common feature in invasive carcinoma cells, and has been recognized at several levels, ranging from genomic mutations to functional deficiencies of an apparently intact complex (4). Cell aggregation assays have been set up to test the functionality of the complex in epithelioid tumor cells. Functional integrity of the complex is a prerequisite for cell-cell adhesion between epithelial cells, and measuring cell aggregation in vitro has thus become another elegant tool to study differences between invasive and noninvasive cell types.

Collagen invasion assay.

Invasion occurs when invasion promoter molecules outbalance the function of invasion suppressors (1). Examples of invasion promoters are cell-matrix adhesion molecules, extracellular proteases, and cell motility factors. In normal tissues, positional stability of the cells is maintained through the counteraction of these invasion promoters by invasion suppressors such as enzyme inhibitors and cell-cell adhesion molecules. Within this context, the interaction of the cancer cells with their surrounding extracellular matrix (ECM) is a determining factor. To study this cell-matrix interaction in vitro, several natural ECM types have initially been applied. Bone (2), salt-extracted cartilage (3), and amnion membrane (4) are examples of devitalized substrata that have been launched in the past to discriminate between invasive and noninvasive cells. Lack of homogeneity of these substrata often made interpretation of invasion difficult, and hampered the reproducibility of those assays (5). To overcome these drawbacks, reconstituted and hence more homogeneous ECMs were developed, and proposed as substrata to test invasiveness. Matrigel (6) (as described in Chapter 7 by Hall and Brooks), and employed also in the assay described in Chapter 8 by Hendrix et al.), Humatrix (7) and collagen type I (8) are today frequently used ECMs in invasion assays. It should, however, be noted that, although these preparations may contain cytokines and growth factors, they are unable to react to the confrontation by cancer cells as a living host tissue does.

Sialylation of E-cadherin does not change the spontaneous or ET-18-OMe-mediated aggregation of MCF-7 human breast cancer cells.

We have investigated the role of sialylation on cell-cell adhesion mediated by E-cadherin. Two MCF-7 human breast cancer cell variants were studied: MCF-7/AZ cells showed a spontaneous cell-cell adhesion in the fast and slow aggregation assay. whereas the adhesion deficient MCF-7/6 cell variant failed to form larger aggregates, suggesting that E-cadherin was not functional under the conditions of both assays. We measured the sialyltransferase activities using Galbeta1-3GalNAcalpha-O-benzyl and Galbeta1-4GlcNAcalpha-O-benzyl as acceptor substrates as well as mRNA levels of four sialyltransferases, ST3Gal I, ST3Gal III, ST3Gal IV, ST6Gal I, using multiplex RT-PCR in MCF-7 cell variants. The alpha2-6 and alpha2-3 sialylation of E-cadherin was investigated by immuno-blot using Sambucus nigra agglutinin and Maackia amurensis agglutinin. Compared to the adhesion-proficient MCF-7/AZ cells, the adhesion-deficient MCF-7/6 cell line apparently lacks ST6Gal I mRNA, has a lower ST3Gal I mRNA, a lower ST3Gal I sialyltransferase activity, and no alpha2-3 linked sialic acid moieties on E-cadherin. The potential anti-cancer drug 1-O-octadecyl-2-O-methylglycero-3-phosphocholine (ET-18-OMe, 48 h, 25 microg/ml) belonging to the class of alkyllysophospholipids restored the E-cadherin function in the adhesion-deficient MCF-7/6 cells as evidenced by an increased aggregation. ET-18-OMe caused loss of ST6Gal I mRNA in MCF-7/AZ cells but no changes of sialyltransferase activities or sialic acid moieties on E-cadherin could be observed. We conclude that Ca2+-dependent, E-cadherin-specific homotypic adhesion of MCF-7/AZ or MCF-7/6 cells treated with ET-18-OMe was not affected by sialylation of E-cadherin.

Activation of the E-cadherin/catenin complex in human MCF-7 breast cancer cells by all-trans-retinoic acid.

All-trans-retinoic acid (RA), like insulin-like growth factor I (IGF-I) and tamoxifen, inhibit invasion of human MCF-7/6 mammary cancer cells in vitro. For tamoxifen and for IGF-I, activation of the invasion-suppressor function of the E-cadherin/catenin complex was shown to be the most probable mechanism of the anti-invasive action. We did a series of experiments to determine whether the anti-invasive effect of RA also implicated the invasion-suppressor E-cadherin/catenin complex. Human MCF-7/6 mammary and HCT-8/R1 colon cancer cells, both with a dysfunctional E-cadherin/catenin complex, were treated with RA and the function of the complex was evaluated through Ca(2+)-dependent fast aggregation. Fast aggregation of both MCF-7/6 and HCT-8/R1 cells was induced by 1 microM RA. This effect was abolished by antibodies against E-cadherin. RA-induced fast aggregation was not sensitive to cycloheximide, tyrosine kinase inhibitors or antibodies against IGF-I or against the IGF-I receptor. RA did not stimulate IGF-I receptor phosphorylation or alter the E-cadherin/catenin complex, as evidenced by immunoprecipitation. RA up-regulates the function of the invasion-suppressor complex E-cadherin/catenin. Its action mechanism is different from that of IGF-I. RA may act as an anti-invasive agent with unique mechanisms of action.

E-cadherin expression and in vitro invasion of canine mammary tumor cells.

E-cadherin is considered to be an invasion suppressor molecule. We have studied the expression and function of E-cadherin in three cell lines derived from a dog mammary tumor, namely SH15, SH24, and SH27. In monolayer culture the cell lines can be distinguished by their morphotype: epithelioid (SH15), fibroblast-like (SH24) and intermediate type (SH27). SH27 was unable to form colonies in collagen gel in contrast to SH15 and SH24. All three cell lines expressed the E-cadherin antigen, as evident from immunocytochemistry, and alpha-, beta-, and gamma-catenins as evident from immunoprecipitation with E-cadherin antibody. Only SH27 showed E-cadherin-dependent aggregation, and little invasion into collagen type 1 gels, in contrast to SH15 and SH24 cells. However, in the precultured embryonic chick heart assay all three cell lines were invasive, demonstrating that invasion depends upon the microenvironment. We assume that in the embryonic chick heart, factors were present or were induced by the SH27 cells, interfering with the function of E-cadherin.

Transition from the noninvasive to the invasive phenotype and loss of alpha-catenin in human colon cancer cells.

Loss of epithelioid organization in carcinoma cell lines has been related to invasiveness and poor differentiation of tumors. We investigated the invasion in vitro of various human colon cancer cell lines. Most cell lines were noninvasive into chick heart fragments, and this correlated with an epithelioid morphotype. Only cell lines COLO320DM, SW620, and variants of HCT-8 and DLD-1 were invasive and nonepithelioid. We examined in these cell lines whether invasiveness was related to changes in the structure and function of the E-cadherin/catenin complex. E-cadherin functions as an invasion suppressor and as a cell-cell adhesion molecule when linked to the cytoskeleton via alpha-catenin plus beta- or gamma-catenin. All noninvasive cell lines showed E-cadherin linked to these catenins. The E-cadherin-dependent cell-cell adhesion function in these cell lines was demonstrated by two assays in vitro. It was interesting that all invasive cell lines showed a dysfunctional E-cadherin/catenin complex. COLO320DM, SW480, and SW620 cells were defective in E-cadherin expression, whereas the invasive variants of HCT-8 and DLD-1 lacked the alpha-catenin protein. From clonal epithelioid HCT-8 cultures with functional E-cadherin/catenin complexes, we subcloned, repeatedly, round cell variants that were again invasive and expressed no alpha-catenin protein. Our data suggest that reproducible transformations toward a more invasive phenotype in HCT-8 cells are associated with down-regulation of alpha-catenin. The mechanisms of this transformation and the level of alpha-catenin down-regulation are currently investigated.

Removal of sialic acid from the surface of human MCF-7 mammary cancer cells abolishes E-cadherin-dependent cell-cell adhesion in an aggregation assay.

MCF-7 human breast cancer cells express E-cadherin and show, at least in some circumstances, E-cadherin-dependent cell-cell adhesion (Bracke et al., 1993). The MCF-7/AZ variant spontaneously displays E-cadherin-dependent fast aggregation; in the MCF-7/6 variant, E-cadherin appeared not to be spontaneously functional in the conditions of the fast aggregation assay, but function could be induced by incubation of the suspended cells in the presence of insulinlike growth factor I (IGF-I) (Bracke et al., 1993). E-cadherin from MCF-7 cells was shown to contain sialic acid. Treatment with neuraminidase was shown to remove this sialic acid, as well as most of the sialic acid present at the cell surface. Applied to MCF-7/AZ, and MCF-7/6 cells, pretreatment with neuraminidase abolished spontaneous as well as IGF-I induced, E-cadherin-dependent fast cell-cell adhesion of cells in suspension, as measured in the fast aggregation assay. Treatment with neuraminidase did not, however, inhibit the possibly different, but equally E-cadherin-mediated, process of cell-cell adhesion of MCF-7 cells on a flat plastic substrate as assessed by determining the percentage of cells remaining isolated (without contact with other cells) 24 h after plating.

Capillary gas chromatography of hexadecylphosphocholine in Caco-2T cells and cell culture media.

The gas-chromatographic determination of hexadecylphosphocholine (HePC), an experimental antitumor agent of the alkyllysophospholipid group, in Caco-2T cell culture and cell culture media is described. The Caco-2T cells were treated with HePC at a concentration of 40 micrograms/ml (9.8 microM) and the uptake of the drug into the cells (calculated per milligram protein) was measured after 48 h culture (37 degrees C, 10% CO2). Also, a reversibility test for another 48 h was carried out in which the retention of the drug was measured. The toxicity of HePC on Caco-2T cells in viability assays was determined. Before the capillary gas-chromatographic determination, sample cleanup was performed by solid-phase extraction (SPE) on a weak cation-exchange column of the CBA (carboxylic acid) type. For quantitation, racemic 1-O-octadecyl-2-O-methylglycero-3-phosphorylcholine (ET-18-OMe) was added as internal standard, followed by derivatization with trimethylsilylbromide. The results showed that HePC taken up by the cells during 48 h of treatment was still detectable 48 h after removal of the drug from the medium.

Tamoxifen restores the E-cadherin function in human breast cancer MCF-7/6 cells and suppresses their invasive phenotype.

Tamoxifen is an antiestrogen used in adjuvant therapy of breast carcinoma and could potentially prevent the development of mammary cancer. While it is widely clinically used, its exact mechanisms of action are not yet fully elucidated. MCF-7/6 cells are estrogen receptor-positive invasive human breast cancer cells with a functionally inactive cell surface E-cadherin. In this study, we report that tamoxifen, and to a lesser extent its metabolites 4-OH-tamoxifen and N-desmethyltamoxifen, restore the function of E-cadherin in MCF-7/6 cells. In an aggregation assay, 10(-6) M tamoxifen significantly increases the aggregation of MCF-7/6 cells. This effect is abrogated by a monoclonal antibody against E-cadherin (HECD-1), is fast (within 30 min), and does not require de novo protein synthesis. Tamoxifen was also found to inhibit the invasion of MCF-7/6 cells in organ culture. Our data is the first demonstration that tamoxifen can activate the function of an invasion suppressor molecule and suggest that the restoration of E-cadherin function may contribute to the therapeutic benefit of tamoxifen in breast cancer patients.

An anti-invasive concentration of the alkyl-lysophospholipid ET-18-OCH3 enhances the motility of embryonal chick heart cells cultured on solid substrate.

Pretreatment of embryonal chick heart fragments with ET-18-OCH3 is known to induce resistance to invasion by several malignant cell lines. Embryonal chick heart fragments or cell suspensions prepared from such fragments were explanted on solid substrate and treated in medium with 10 micrograms/ml ET-18-OCH3 or with drug-free medium (control) for 48 h. This medium was washed away and replaced by drug-free fresh medium. Twenty-four to 48 h later the fast plasma membrane movements (involved in ruffling, blebbing, fast shape change and fast translocation) were quantified using a simple method based on subtracting two video images taken with an interval of 28 s. The ET-18-OCH3-treated cells showed a higher intensity of fast plasma membrane movements than control cells. Cells around a treated explant did not show the same radial alignment as in controls, suggesting loss of contact inhibition of movement. Cells from a cell suspension derived from a treated fragment showed faster translocation on solid substrate and faster shape change. We speculate that increased motility of host cells may be involved in resistance to invasion.

Micro-encapsulation of MDCK-ras-e cells prevents loss of E-cadherin invasion-suppressor function in vivo.

The invasion-suppressor molecule E-cadherin mediates Ca(2+)-dependent cell aggregation and prevents invasion. E-cadherin-positive Madin-Darby canine kidney (MDCK) cells that were non-invasive in vitro formed, upon i.p. injection, tumors that were invasive. Differentiated tubular tumor areas showed an intense immuno-signal for E-cadherin at intercellular contacts, whereas undifferentiated structures did not. Cell lines derived from such tumors turned out to be invasive in vitro and showed decreased Ca(2+)-dependent cell aggregation but no change in E-cadherin immunopositivity. This combination of phenotypes indicated a loss of the E-cadherin invasion-suppressor function. Micro-encapsulation of i.p.-injected cells prevented the loss of the E-cadherin invasion-suppressor function. We concluded that this loss in vivo was dependent upon immediate contacts between tumor cells and host cells or upon host factors that could not cross the capsule membrane.

Enlarged cell-associated proteoglycans abolish E-cadherin functionality in invasive tumor cells.

Mouse and dog epithelial cell lines, expressing high levels of the Ca(2+)-dependent cell-cell adhesion molecule E-cadherin in vitro, generated invasive and metastatic tumors in athymic mice. From these tumors, neoplastic cell lines were isolated. All ex vivo isolates retained high expression levels of E-cadherin at their surface. Nevertheless, some showed a fusiform morphotype, were defective in Ca(2+)-dependent cell aggregation, and were invasive in vitro, indicating that E-cadherin was not functional. Cell-associated proteoglycans were found to be enlarged in these variants as compared to their counterparts with functional E-cadherin. Treatment of the cells with the drug 4-methylumbelliferyl beta-D-xyloside specifically reduced the amount and size of cell-associated proteoglycans. This same drug induced an epithelial morphotype, increased Ca(2+)- and E-cadherin-dependent cell aggregation, and abrogated invasiveness without influencing E-cadherin expression levels. Our results indicate that enlarged proteoglycans can prevent the homophilic binding of E-cadherin, probably by steric hindrance. This is one more mechanism by which carcinomas may counteract invasion-suppressor genes and acquire malignancy.

Insulin-like growth factor I activates the invasion suppressor function of E-cadherin in MCF-7 human mammary carcinoma cells in vitro.

The calcium-dependent cell-cell adhesion molecule E-cadherin has been shown to counteract invasion of epithelial neoplastic cells. Using three monoclonal antibodies, we have demonstrated the presence of E-cadherin at the surface of human MCF-7/6 mammary carcinoma cells by indirect immunofluorescence coupled to flow cytometry and by immunocytochemistry. Nevertheless, MCF-7/6 cells failed to aggregate in a medium containing 1.25 mM CaCl2, and they were invasive after confrontation with embryonic chick heart fragments in organ culture. Treatment of MCF-7/6 cells with 0.5 microgram ml-1 insulin-like growth factor I (IGF-I) led to homotypic aggregation within 5 to 10 min and inhibited invasion in vitro during at least 8 days. The effect of IGF-I on cellular aggregation was insensitive to cycloheximide. However, monoclonal antibodies that interfered with the function of either the IGF-I receptor (alpha IR3) or E-cadherin (HECD-1, MB2) blocked the effect of IGF-I on aggregation. The effects of IGF-I on aggregation and on invasion could be mimicked by 1 microgram ml-1 insulin, but not by 0.5 microgram ml-1 IGF-II. The insulin effects were presumably not mediated by the IGF-I receptor, since they could not be blocked by an antibody against this receptor (alpha IR3). Our results indicate that IGF-I activates the invasion suppressor role of E-cadherin in MCF-7/6 cells.

Evidence for abrogation of oncogene-induced radioresistance of mammary cancer cells by hexadecylphosphocholine in vitro.

Hexadecylphosphocholine (HePC), an experimental and clinical antitumour agent of the alkyllysophospholipid group, was tested for its radiosensitising effect on a panel of nine human mammary cancer cell lines in vitro. Growth inhibition by ionising radiation and recovery from it were not influenced by pretreatment with HePC in most cases, except for two cell lines expressing an activated ras oncogene. In the latter we found an enhanced radioresistance that was abolished by pretreatment with HePC. Our results suggest that HePC may act as a radiosensitiser for cells carrying an activated ras oncogene.

Role of the host tissue in the anti-invasive activity of the alkyllysophospholipid, ET-18-OCH3, in vitro.

The alkyllysophospholipid, racemic-l-O-octadecyl-2-O-methylglycero-3- phosphocholine (ET-18-OCH3) was previously shown to inhibit invasion of malignant cells into precultured heart fragments (PHF) in vitro. In particular, pretreatment of PHF with 10 micrograms ET-18-OCH3 for 48 h was sufficient to induce in the host tissue resistance towards invasion by mouse MO4 cells. Resistance was obvious when MO4 cells were confronted either immediately (the pretreatment experiment) or after withdrawal of the drug 7 days prior to confrontation (the reversibility experiment). In the present study, the survival of PHF cells in the pretreatment and reversibility experiments was similar to that of untreated PHF cells as determined by the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide (MTT) test and by the PHF explantation test. The effective anti-invasive concentration was 6 micrograms/ml in the pretreatment experiment while 3 micrograms/ml was sufficient to inhibit invasion in the reversibility experiment. Induction of resistance towards invasion in pretreated PHF was shown to occur not only with MO4 cells but also with mouse LLC-H61 Lewis lung carcinoma and mouse BW-O-Li1 T-lymphoma cells. The increase in molecular weight of N-linked cell surface glycosylpeptides (N-GP) of PHF was apparent in the pretreatment experiment and was enhanced in the reversibility experiment. This effect was completely abolished in cells obtained from pretreated PHF which were converted into a cell suspension and further cultured as a monolayer on tissue culture plastic without drug for 7 days. The results reported here provide additional evidence for the causal involvement of N-GP of the PHF host tissue in the anti-invasive activity of ET-18-OCH3 in vitro.

Multiple differences between wild-type B16 melanoma cells and a wheat germ agglutinin resistant clone.

A low-metastatic, glycosylation-defective variant of the B16 murine melanoma was obtained by Tao and Burger (1977) through selection with wheat germ agglutinin. We found that variant and parental (wild-type) cell lines were equally invasive when confronted with precultured embryonic chick heart fragments in vitro. Also, a short-term in vivo arrest assay showed no significant differences. After intravenous injection, wild-type cells killed the recipient mice faster than did the variant cells. We were able to confirm the changes in glycosylation at the enzyme level. In addition, we showed that the pattern of endogenous lectins was strikingly different, at least at the quantitative level. We also looked at another set of receptor proteins, namely receptors for neurotransmitters coupled to adenylate cyclase. The response to the vasoactive intestinal peptide and prostaglandins was lower in the variant cells, which also had a delayed response to cholera toxin. Although most of the data can be explained by altered glycosylation in the variant cells, the large number of differences between variant and parent cells makes it difficult to identify the biochemical basis of altered metastatic behaviour. This might also be the case with other pairs of cells differing in metastatic activity.