INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Unveiling the mechanisms that control establishment and maintenance of intercellular contacts in the epithelium of the intestine is essential to understanding colon carcinogenesis. There is considerable evidence that protein kinase C (PK-C)¹ isoforms are involved in this process; for example, addition of the phorbol ester phorbol 12-myristate-13-acetate (PMA) has been shown to reduce cell-to-cell contacts in epithelial cells (1). This compound has been widely used to activate PK-C in cells because it is membrane-permeable and causes a maintained stimulation of most members of this extended family (cPK-C and nPK-C) (2, 3). Addition of PMA to intestinal cell subpopulations derived from HT-29 cells, like its addition to many other epithelial cell lines, causes a striking alteration in their morphology and induces scattering of cell colonies (4). This process is characterized by the acquisition of a more fibroblastic phenotype, with lower cell-to-cell adhesion and inactivated E-cadherin (4, 5). In parallel, addition of this compound retards cell growth, an effect that has also been observed in other intestinal cells (6). Previous studies from our group have shown that cell scattering can also be induced by thymeleatoxin, a specific activator of conventional PK-Cs (cPK-Cs) (7). The goal of this study was to identify the PK-C isoform involved in these events and determine whether activation of this isoform in intestinal cells lead to altered tumorigenic properties of HT-29 M6 cells (HT-29 cell subpopulation isolated using 10⁶ methotrexate) in vivo.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Reagents-- PMA, leupeptin and phenylmethylsulfonyl fluoride were supplied by Sigma. Gelatin and plasminogen were from Merck and Boehringer Mannheim, respectively. PK-C inhibitors GF109203X (Gf) and Go 6976 (Go) were purchased from Calbiochem (San Diego, CA); these products were dissolved in dimethyl sulfoxide (Me₂SO) and stored protected from light at 40 °C. [-³²P]ATP was purchased from Amersham Pharmacia Biotech. Monoclonal antibodies against cPK-C and E-cadherin (HECD-1) were obtained from Transduction Laboratories (Lexington, KY) and Zymed Laboratories Inc. (San Francisco, CA), respectively. Prestained SDS-polyacrylamide gel electrophoresis molecular weight markers were from Bio-Rad (Richmond, CA). All of the oligonucleotides used in our assays were synthesized by Amersham Pharmacia Biotech.

Cell Culture-- The properties of the cell line used in this study HT-29 M6 (M6), originally characterized with the name of HT-29 (10⁶ methotrexate), have been extensively described (8, 9). Cells were seeded at a density of 2 × 10⁴ cells/cm² and cultured in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine serum (Life Technologies, Inc.) as described previously (5). Experiments of scattering were performed using cells 4-5 days after plating, when they were 40-50% confluent.

Mutagenesis of cPK-C and Transfection to M6 Cells-- Human cPK-C was obtained from ATCC. Mutation of the alanine situated in position 25 to glutamic acid (A25E) was carried out according to the method of Kunkel et al. (10), using the oligo 5'-CCCGCAAAGGGGAGCTCAGGCAGAAG-3' (corresponding to nucleotides 89-114 of the sequence of human cPK-C). In addition to the change C to A (position 102, in bold), required to replace Ala with Glu, the G present in position 106 was mutated to a C (underlined); this second silent mutation generates a restriction site for SacI, which is useful to recognize the mutated cDNA. The presence of the mutations in the final construction was verified by sequencing using Sequenase (U. S. Biochemical Corp.). Mutated cPK-C, denominated cPK-C(+), was inserted in the NotI site of pOPRSV expression plasmid (Stratagene, La Jolla, CA) and transfected to M6 cells using LipofectAMINE (Life Technologies, Inc.) under the conditions indicated by the manufacturer. After 21 days of selection in Dulbecco's modified Eagle's medium supplemented with G-418 (0.6 mg/ml) (Sigma), resistant clones were isolated. As a control, transfection of HT-29 M6 with pOPRSV plasmid without insert was performed; two clones resistant to G-418 were isolated and analyzed.

Analysis of the Transfectants-- RNA was obtained from the different clones using the guanidium thiocyanate method (11), treated with DNase I, and reverse transcribed into cDNA using random hexamers. This cDNA was first analyzed by PCR with oligos A1 (5'-TGGACAAACTACCTACAGAGATT-3') and A2 (5'-TCCCAAACCCCCAGATGAAGTCG-3'); oligo A1 corresponds to nucleotides 2638-2660 of pOPRSV, a sequence located downstream from the transcription start point; oligo A2 corresponds to nucleotides 211-189 of cPK-C. To verify the specificity of the amplification, the 275-base pair product of amplification was subjected to electrophoresis, transferred to a Nylon-membrane (Amersham Pharmacia Biotech) and blotted with oligo B1 (5'-GGACCATGGCTGACGTTTTC-3', nucleotides 23-42 of cPK-C) labeled with [-³²P]ATP by T4 polynucleotide kinase. In order to compare the relative expression of the mutated and wild-type forms of cPK-C, the cDNA was amplified quantitatively with oligos B1 and B2 (5'-ACAGCAAACTTGGACTGGAA-3', nucleotides 239-220), which flank the mutation. The product of the PCR was digested with SacI and analyzed as above using oligo A2 labeled with ³²P. After treatment with SacI, wild-type cPK-C generated a fragment of 217 base pairs; the mutated cDNA generated a fragment of 142 base pairs. As a control, the cDNA was subjected to 20 cycles of amplification using two oligonucleotides specific for human actin (sequences 10-30 and 358-338).

Invasion into Embryonic Chick Heart Fragments-- Invasiveness into embryonic chick heart fragments was assayed following the method of Mareel et al. (12), with minor modifications. Briefly, heart fragments of 9-day-old chick embryos were dissected and incubated for 2 days in an orbital shaker set at 100 rpm at 37 °C in an atmosphere of 95% air, 5% CO₂. Precultured heart fragments with a diameter of 0.5 mm were selected under a stereomicroscope. Cell suspensions were brought into contact with precultured heart fragments placed on semisolid agar medium. After incubation for 8 h at 37 °C to allow the cells to attach to the external fibroblastic layer of precultured heart fragments, individual confrontations were transferred to 5-ml Erlenmeyer flasks for further incubation in an orbital shaker (120 rpm) at 37 °C. After 1-4 days, confrontations were fixed and processed for transmission electron microscopy.

Semithin and Ultrathin Sections-- Cell confrontations were fixed with 2.5% glutaraldehyde, treated with 2% OsO₄, and dehydrated washing in ascending series of graded ethanol (once in 50%, once in 70%, twice in 95%, and five times in 100% ethanol). Cells were then embedded in Spur resin (TAAB Laboratories, Aldermaston, United Kingdom). For light microscopy, semithin sections were cut with a LKB ultramicrotome and stained with toluidine blue. For transmission electron microscopy, ultrathin sections of 500-800 Å were placed on uncoated 300-mesh copper grids prior to staining with uranyl acetate (5% in absolute ethanol) and lead citrate and viewed at original magnifications from 4500 to 15,000 in a Hitachi H700 transmission electron microscope operated at 75 kV.

Xenografting-- Cells were trypsinized, washed, and resuspended in sterile phosphate-buffered saline solution at a concentration of 1 × 10⁷ cells/ml. Tumorigenic ability was determined by inoculation of 1 × 10⁶ cells in the subcutis of 5-week-old athymic female nude mice (Criffa, Barcelona, Spain). Viability of the uninjected cells, determined at the end of the process, always exceeded 95%. Tumors were followed externally. After 6 weeks, mice were sacrificed by cervical dislocation, and the site of the injection, as well as the liver, lungs, and lymph nodes, was examined. Tissues were fixed, embedded in paraffin, and analyzed under the microscope after hematoxylin-eosin staining.

Plasminogen Activator Activity-- Activity of plasminogen activators was determined by zymography. Conditioned medium from cells cultured for 24 h in the absence of fetal bovine serum (10%) was centrifuged at 13,000 × g for 15 min at 4 °C. Sample volumes were adjusted on the basis of protein concentration in the corresponding cell lysate, and proteins were separated by SDS-polyacrylamide gel electrophoresis in plasminogen and gelatin-containing gels, as described elsewhere (13). Protease activity was revealed by incubating the gels in 2.5% Triton X-100 for 1 h followed by incubation in 0.1 M glycine, pH 8.3, overnight at 37 °C. After fixing of proteins with methanol/acetic acid/water (30:10:60), gels were stained with 0.1% Amido Black and destained.

Other Methods-- To determine cell dissociation, the assay developed by Nagafuchi et al. (14) was used, with the modifications described previously (7). The extent of dissociation was represented by the index N_TC/N_TE, where N_TC and N_TE are the total particle number after treatment of cells with trypsin in the presence or absence of calcium, respectively. E-cadherin association to the cytoskeleton was determined analyzing by Western blot the presence of this protein in Triton-soluble and -insoluble fractions, prepared as described by Nelson and co-workers (15). Autoradiograms were quantified by scanning densitometry (Hoefer GS-300 Scanning Densitometer).

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Inhibitors of cPK-C Isoforms Block Scattering of M6 Cells Induced by PMA-- We have previously reported that PMA induces remarkable morphological changes in M6 cultures (4). After addition of this phorbol ester, M6 colonies scatter, and cells acquire a fibroblastic aspect. These effects are observed shortly after the addition of this phorbol ester at concentrations as low as 20 nM. Among the properties altered by PMA, cells lose homotypic adhesion, and concomitantly, E-cadherin is inactivated (5). These effects of PMA are prevented if the cells are simultaneously incubated with the PK-C inhibitor Gf (2 µM) (7). Gf is a bisindolymaleimide derivative that selectively inhibits PK-C isoforms; in vitro, it shows a ranked order of potency of cPK-Cs > nPK-Cs > atypical PK-Cs (16). As shown in Table I, the action of PMA on cell scattering, homotypic adhesion, or E-cadherin inactivation was blocked by similar concentrations of Gf. The minimal dose of Gf required to prevent morphological scattering induced by PMA was 0.5 µM; in the presence of this concentration, cells incubated with PMA presented a similar phenotype to untreated controls (not shown). The effect on this inhibitor on the loss of homotypic adhesion was also estimated; IC₅₀ was obtained with approximately 70 nM Gf, a concentration lower than the IC₅₀ for nPK-Cs  or  activities (210 and 130 nM, respectively), determined by in vitro protein kinase assays. Concomitantly with the effect on homotypic aggregation, low doses of Gf (0.5 µM) PMA also blocks the decrease in E-cadherin-associated to cytoskeleton caused by PMA (Table I).

Expression of an Activated Form of cPK-C Causes Scattering of M6 Cells-- The use of these inhibitors suggests that a cPK-C is responsible for triggering the process of scattering and loss of cell-to-cell adhesion. We have previously described that the only cPK-C expressed in HT-29 M6 cells is cPK-C (7). The presence of cPK-C, and not of cPK-Cs  or , was demonstrated both by Western blot with specific monoclonal antibodies and by reverse transcription-PCR analysis, using two oligos corresponding to consensus sequences of these three isoforms. Therefore, to prove the role of cPK-C in cell scattering, we have overexpressed this isoform in M6 cells.

View larger version (140K): [in this window] [in a new window]   Fig. 1.   Morphology of cPK-C(+) transfectants. Cells were grown in complete medium for 6 days; when indicated, PMA (100 nM) was added at the time of seeding, and Gf (1 µM) was added during the last 2 days of incubation. In the experiment shown in the bottom row, cells were incubated with PMA (10 nM) for 4 h. Pictures were taken under a phase-contrast microscope at × 200 magnification, except in the case of A4 cells incubated with PMA, which was taken at × 350. No morphological differences respect to the control were observed in C1 and C2 clones, in cells incubated with Gf, or with PMA and Gf added simultaneously (results not shown).

View larger version (37K): [in this window] [in a new window]   Fig. 2.   PCR analysis of the transfectants. cDNA was obtained from the different clones and analyzed by PCR using oligos A1 and A2 (PCR A), B1 and B2 (PCR B), or actin-specific (Actin) as described under "Experimental Procedures." A scheme of the construction with the position of the oligos is shown. The product of PCR A was transferred to a nylon membrane and analyzed with oligo B1 labeled with 32P; the product of PCR B was digested with SacI and analyzed with oligo A2. The figure shows representative results of PCRs A and B (autoradiograms) and of the amplification of the actin fragment used as control (ethidium bromide staining). The three amplifications were performed at low number of cycles (20-25 cycles) to avoid saturation of the reaction. In PCR B, a control of digestion of the plasmid with SacI was included; although it is not shown, the bottom fragment was not observed in M6 or A1 cells.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   uPA activity in cPK-C(+) transfectants. Media conditioned for 24 h from the different cell clones or M6 cells were treated when indicated with PMA because seeding were subjected to SDS-polyacrylamide gel electrophoresis in plasminogen and gelatin-containing gels, and protease activity was measured as described under "Experimental Procedures." In addition of a protease of 55 kDa, corresponding to uPA, a minor band of lower molecular mass (probably consequence of the processing of uPA) was also detected in A4 cells.

Expression of cPK-C(+) Modifies Other Cellular Properties Associated with Scattering: Transfectants Show Higher Mobility, Lower Cell-to-Cell Adhesion, and Decreased Levels of E-cadherin Compared to Controls-- Scattering is associated with an increase in the mobility of cells. To determine whether expression of cPK-C enhanced mobility, a wound was inflicted in the epithelial monolayer, and the ability of the cells to fill the gap was determined. After 24 h, control M6 cells had not moved into the denuded area (Fig. 4). Incubation with 100 nM PMA but not with 10 nM PMA induced the cells to significantly fill the free space. Expression of cPK-C(+) correlated with the extent of the healing; A4 cells showed higher mobility than A3, and these in turn showed higher mobility than A2. A2 and A3 were sensitive to low doses of PMA (10 nM) that did not induce the movement of M6 cells (Fig. 4).

View larger version (67K): [in this window] [in a new window]   Fig. 4.   cPK-C(+) expression increases mobility of M6 cells. Cell mobility was determined using an assay of wound healing. Confluent cell clones were wounded with two cross-shaped scratches; PMA was added when indicated, and the ability of the cells to refill the gaps was examined after 24 h. The × 50 magnification micrographs show the results of one representative experiment of three performed.

View this table: [in this window] [in a new window]   Table II Expression of cPK-C(+) decreases homotypic aggregation NTC/NTE index was determined from cells 50% confluent as described. When indicated, cell medium was supplemented with PMA for 24 h. The mean ± S.D. of three independent experiments is shown.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   cPK-C(+) transfectants present lower levels of E-cadherin associated to the cytoskeleton. Fractions soluble in Triton X-100 or only in SDS were prepared from different clones or M6 cells treated, when indicated, with PMA (100 nM) since seeding. Equivalent amounts of both fractions were analyzed by Western blot with an anti-E-cadherin monoclonal antibody; only a band of 120 kDa, corresponding to this protein, was detected. The figure shows a representative experiment of three performed.

Cell Growth Is Retarded by the Activation of cPK-C(+)-- Another of the features that appear in M6 cells treated with PMA is a retardation in cell growth (4). As observed in Fig. 6, this effect of the phorbol ester requires concentrations significantly higher than those required to induce scattering. For instance, addition of 20 nM PMA, the lowest concentration that induced scattering, did not have any effect on the proliferation of HT-29 M6 cells. Growth of the different clones was measured either in the absence or in the presence of 20 and 100 nM PMA. Control cells (C1 and C2) grew with times of duplication similar to those of M6 cells and showed similar sensitivity to PMA. Minor differences were found between control and A2 cells either in the absence of PMA or in the presence of the two different concentrations of this compound. As shown in Fig. 6, growth of A3 cells in absence of phorbol ester was similar to control. However, these cells were much more sensitive to the phorbol ester; 20 nM PMA inhibited proliferation of A3 by 50% versus less than 5% to M6 or C1 cells. The highest cPK-C(+)-expressing clone, A4, proliferated at a lower rate and responded to low doses of PMA with a total block in proliferation (Fig. 6).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Expression of cPK-C(+) retards cell growth. Proliferation of cells was measured during the exponential phase of growth directly by cell counting in the absence or in the presence of the indicated concentrations of PMA. The figure shows the mean of three different of experiments, which did not differ by more than 5%. The number in parentheses corresponds to the duplication time of the cells in the different conditions studied.

Expression of cPK-C(+) Stimulates Invasion of Chick Embryo Heart Fragments-- The results obtained so far have indicated that activation of cPK-C induced the acquisition of a fibroblastic phenotype, with low functionality of E-cadherin and enhanced secretion of uPA, two alterations that have been related to a more invasive phenotype of epithelial cells (21, 22). Therefore, we analyzed whether transfection of cPK-C(+) had indeed enhanced the invasive properties of M6 cells, using the assay of invasion of embryonic chick heart fragments. Single-cell suspensions were added to the precultured heart fragments, and samples were processed after different periods of incubation. Histological analysis of cultures revealed striking differences between control cells and cells expressing cPK-C(+). M6 cells, or clone C1 cells, which behaved identically, gave rise to compact colonies that progressively occupied the peripheral parts of the heart fragment without invading the myocardial tissue (Fig. 7, G and H). Only after long times of confrontation (6 days) were small clusters of cells seen inside the heart tissue (not shown). In contrast, A4 cells were highly invasive; spindle-shaped A4 cells could be observed penetrating the heart tissue as early as 8 h after the initiation of the assay (Fig. 7, A and B). After 24 h, A4 cells had invaded the heart fragment and replaced extensive areas of this tissue (Fig. 7, C and D). These cells did not survive well in the co-culture; at longer periods of time (3-6 days), most of the A4 cells showed symptoms of cell damage, such as clumping of chromatin and shrinkage of cytoplasm (not shown). The invasive capability of A3 cells was also determined; these cells invaded more slowly than A4 cells, although an extensive colonization of the fragment was observed by 24 h (Fig. 7, E and F).

View larger version (136K): [in this window] [in a new window]   Fig. 7.   Expression of cPK-C(+) stimulates invasion of chick heart fragments by M6 cells. The figure shows light micrographs of semithin sections from confrontations between precultured embryo heart fragments (marked H) and clones A4 (A-D), A3 (E and F), or M6 cells (G and H) fixed after 8 h (A and B), 24 h (C-F), or 3 days (G and H). T, M6 cells; arrowheads in A and B: A4 cells; asterisk in H: fibroblast showing signs of cell damage. Magnifications: A, C, E, and G: × 150; B, D, F, and H, × 600.

View larger version (145K): [in this window] [in a new window]   Fig. 8.   Alterations in cell ultrastructure in cells expressing cPK-C(+) after invasion of EHFs. The figure shows representative transmission electron micrographs from some of the samples analyzed in Fig. 7. A, two A4 cells (T) actively progressing inside the heart tissue. H, myoblasts (× 4500). B, detail from A showing an A4 cell (T) with a marked hypertrophy of rough endoplasmic reticulum cisternae (asterisk). Myofilaments (M) can be distinguished in neighboring myoblasts (H). Arrowheads, fuzzy material; arrow, Z-line (× 15,000). C, M6 cells did not invade the heart tissue (H) after 3 days of culture. Arrow, mucosecretory granules; arrowheads, microvilli (× 5000).

Growth and Characteristics of Tumors Originated by M6 Cells Are Affected by Expression of cPK-C(+)-- We have demonstrated that activation of cPK-C induces two alterations that are presumably antagonistic in the process of tumorigenesis: enhanced cell invasion and retarded proliferation. In order to analyze the tumorigenic ability of cells expressing cPK-C(+), the different clones and controls were xenografted into the subcutis of nude mice. Expression of cPK-C(+) was associated to a decrease in the size of the tumors originated by M6 cells (Fig. 9). In contrast to control C1 or A1 cells, which originated large tumors after eight weeks, A4 cells did not form macroscopic tumors at this late time (Table III). Absence of tumors was confirmed by microscopic analysis of the area of implantation after hematoxylin-eosin staining. No evidence of tumor cells was observed when the analysis was performed at longer (10 weeks) or shorter (10 days) times after grafting. Implantation of these cells in the spleen gave similar results (not shown).

View larger version (167K): [in this window] [in a new window]   Fig. 9.   Histological analysis of tumors. The figure shows representative tumors obtained from different xenografts stained with hematoxylin-eosin. Tumors A and B were obtained from animals injected with C1 cells, C and D from animals injected with with A2 cells, and E-G from animals injected with A3 cells. Details are shown to the right of the indicated tumors; they are labeled as the letter corresponding to the tumor followed by a number. Magnifications: A-G × 12.5; G1, × 40;, B1, D1, E1, and E2, × 100.

View this table: [in this window] [in a new window]   Table III Expression of cPK-C decreases tumorigenicity of M6 cells Mice were injected subcutaneously in the flank with 106 cells; presence of tumors was analysed as described 8 weeks later. ND, tumor not detected.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

Several conclusions can be drawn from our results. First, it is evident that activation of cPK-C in M6 cells induces the acquisition of a fibroblastic phenotype, denominated scattered. Cell scattering is a consequence of co-ordinated changes that decrease cell-to-cell interactions, increase the adhesion to the extracellular matrix, and stimulate the migration of the cells. The basis of these phenotypic changes is alteration in the function of several proteins; for instance, inactivation of E-cadherin and increased uPA activity have been related to loss of homotypic adhesion and enhanced mobility, respectively (22, 23). We show here that expression of an activated form of cPK-C is sufficient to induce the acquisition of the scattered phenotype and to alterate the functionality of both E-cadherin and uPA.

Expression of cPK-C(+) also increased the ability of M6 cells to invade chick heart embryo fragments. Similar data have been obtained with other epithelial cells: overexpression of cPK-C in breast cells induced the acquisition of a fibroblastic phenotype and enhanced the proliferation rate and the tumorigenic and metastatic capabilities (24). However, differently to the results in breast cells, activation of cPK-C in intestinal M6 cells decreased proliferation and tumorigenicity. Similar data were obtained by Weinstein and co-workers (25) in HT-29 cells after transfection of another cPK-C, cPK-C. Based on these data and the localization of this enzyme along the crypt-villus axis, a role for cPK-C in the negative control of cell growth in intestinal epithelium has been suggested (6, 26, 27). Association of cPK-C to the membrane has been detected by immunofluorescence in the mid-crypt region, where cells cease proliferation; this alteration has been suggested to reflect an increased activity of this enzyme, although this conclusion has not been definitively proved.

It is remarkable that activation of the same enzyme, cPK-C, exerts two different actions on M6 cells: it increases invasion and retards cell growth, two effects that seem antagonistic in tumor development. The results presented in this report demonstrate that although both effects are triggered by activation of the same PK-C isoform, they are differentially sensitive to the extent of this activation. A moderate stimulation of the enzyme (for instance, as present in A3 cells) causes an increase in the ability of these cells to infiltrate surrounding tissues and a decrease in the growth of the tumor; on the other hand, when the activation of the enzyme is extensive (A4 cells), the growth inhibitory effect is predominant, and the tumor is unable to develop. These effects mimic what happens in M6 cells treated with PMA: cell scattering and loss of cell-to-cell contacts require low doses of this phorbol ester, whereas higher concentrations are required to affect cell growth. Similar results were obtained with other intestinal cell lines with respect to the sensitivity of these two parameters to PMA. In all of these cell lines (WiDr, Caco-2, HRT18, SW620, and SWCo15), maximal inhibition of growth (26-52%, depending on the cell lines) required concentrations of the phorbol ester about 100 nM, whereas scattering was observed at much lower doses.²

Several lines of evidence indicate that alterations in PK-C may be involved in malignant transformation of colon in humans. Colon adenocarcinomas show a reduction in PK-C activity compared with normal adjacent mucosa, indicating down-regulation of this enzyme in the tumors (28-30). Experimental models of colon carcinogenesis have provided evidence of changes in PK-C both in premalignant and malignant epithelial cells; these changes include translocation of PK-C and subsequent down-regulation of this enzyme (31). Alterations in the content of specific PK-C isozymes have been also described in colon tumors with respect to normal tissue either in human samples or in experimental animal models (32, 33); however, different laboratories have reported contrary results of these analyses (32, 34, 35). Therefore, at the present, it is not clear whether the initial activation or the later down-regulation are related to colon tumorigenesis, neither the role of specific isoforms in this process.

The results presented here can help to explain the role of cPK-C in colon carcinogenesis. This enzyme might exert a dual action: 1) a moderate activation of cPK-C would take place at the first stages of carcinogenesis, being involved in the loss of function of E-cadherin, and 2) a later down-regulation and inactivation of cPK-C would result in an uncontrolled growth of the primary tumor or the metastasis. According to this model, studies performed with cell lines or tumors that have progressed beyond the first stage would not show any positive effect of PK-C activation in tumor development. Experiments directed to the validation of this model using transgenic animals are currently in progress in our laboratory.

While this work was being written, the article by Rosson et al. (36) came to our notice; it describes the involvement of cPK-C in the scattering of LLC-PK1 cells induced by PMA. Their conclusion was obtained by expression of the wild-type form and a negative mutant of cPK-C in these cells. Although overexpression of a dominant negative mutant does not demonstrate involvement of cPK-C in cell scattering induced by PMA, the results presented by these authors are totally consistent with those described in this work.