The effect of N-methyl-N'-nitro-N-nitrosoguanidine on cultured dog gallbladder epithelial cells

The Effect of N-Methyl-N*-Nitro-N-Nitrosoguanidine on Cultured Dog Gallbladder Epithelial Cells RUBEELA MALIK,1 SUNG-KOO LEE,1 CHRISTOPHER E. SAVARD,1 DOLPHINE ODA,2 WAI S. WONG,1 BOB Y. CHAN,1 1 AND SUM P. LEE

Normal dog gallbladder epithelial cells in long-term culture were used as a model to study the morphologic, genetic, and secretory processes associated with the progression to cancer formation. Dog gallbladder epithelial cells cultured on collagen-coated plates grew into polarized monolayers, could be passaged repeatedly, and showed the typical morphological profile of well-differentiated columnar epithelial cells. After cells were exposed to N-methyl-N*-nitro-N-nitrosoguanidine (MNNG) at 1005 mol/L for 48 hours, the treated cells grew on plastic and could be cloned. Flow cytometry revealed emergence of an aneuploid cell subpopulation. In organotypic culture, treated cells showed a pseudostratified appearance, with cellular and nuclear pleomorphism. Large and hyperchromatic nuclei were present as well as increased mitotic rate. The proteins of MNNG-treated dog gallbladder epithelial cells showed increased phosphorylation of tyrosine residues. Treated cells showed a decrease in mucin secretion in response to prostaglandin E2 , manifesting an altered pattern of mucin secretion. Transforming growth factor-beta failed to inhibit cell proliferation in the MNNG-treated cells compared with the prominent inhibition in normal cells. Together, the data reflected changes representing preliminary steps on the pathway to develop cancer cells. Our results indicate that carcinogenic chemicals can cause measurable chromosomal and cellular modifications to normal biliary epithelial cells in vitro. This model may be useful in understanding the sequential steps in carcinogenesis and affords an opportunity to study chromosomal damage, cytokinetics, changes in molecular genetic markers, and expression, as well as cell biological function during cellular transformation. (HEPATOLOGY 1997; 26:1296-1302.) The gallbladder is the most common site for biliary tract cancer. The incidence of gallbladder cancer is reported to be higher in patients with cholelithiasis,1 and bile duct cancer is more common in patients with choledochal cyst, pancreaticobiliary maljunction,2 and sclerosing cholangitis.3 Due, in

Abbreviations: DGBE, Dog gallbladder epithelial cells; MNNG, N-methyl-N*-nitroN-nitrosoguanidine; TGF-b, transforming growth factor beta. From the Departments of 1Medicine and 2Oral Biology, University of Washington and VA Medical Center, Seattle, WA. Received September 19, 1996; accepted June 13, 1997. Supported by a grant from the NIH (DK50246) and in part by the Medical Research Service of the Department of Veterans Affairs, Seattle, WA. Address reprint requests to: Sum P. Lee, M.D., Ph. D., Head, Division of Gastroenterology, AA 103L, Box 356424, University of Washington Medical Center, Seattle, WA 98195. Fax: (206) 764-2232. Copyright q 1997 by the American Association for the Study of Liver Diseases. 0270-9139/97/2605-0031$3.00/0

part, to the relative inaccessibility of the biliary tract and gallbladder, the exact pathogenic mechanism for development of biliary tract cancer is not clearly understood. We reasoned that the ability to chemically induce transformation of a normal biliary epithelial cell line would help elucidate the mechanisms and the early genetic and molecular changes leading to the development of cancer in these tissues. Our laboratory has previously shown the successful culture of epithelial cells from dog gallbladder4 and pancreatic duct5,6 and have applied these cell models in cell physiology and cell biology studies. In this report we describe the establishment of altered cell lines developed from normal dog gallbladder epithelial cells (DGBE) exposed to N-methyl-N*nitro-N-nitrosoguanidine (MNNG), a known chemical carcinogen. Exposure of DGBE to MNNG led to changes in morphology, growth characteristics, and flow cytometry profile. In addition, the number and pattern of proteins phosphorylated on tyrosine residues, a proposed indicator of cell transformation,7,8 was also found to be changed. A decrease in mucin secretion in response to various agonists was also noted indicating an altered cellular function. The anti-proliferative action of the cytokine transforming growth factorbeta (TGF-b) was evident in normal cells but absent in the treated cells. The MNNG treated DGBE present a new in vitro model for the study of the early step-wise changes occurring in biliary epithelial cells, following exposure to chemical carcinogens, during the process of cell transformation. MATERIALS AND METHODS Chemicals and Reagents. Vitrogen, a bovine dermal collagen, was purchased from Celtrix Laboratories (Palo Alto, CA). Tissue culture plates were from Falcon (Lincoln Park, NJ). Chemicals were obtained from Sigma Chemical Co. (St. Louis, MO) except where noted. Monoclonal anti-phosphotyrosine (Sigma Chemical Co.) derived from mouse ascites fluid was used for the phosphotyrosine Assay. Isolation and Culture of Dog Gallbladder Epithelial Cells. Gallbladder epithelial cells were isolated from dog gallbladder by enzyme dissociation and cultured as previously described by Oda et al.4 Culture dishes (60 mm) coated with 1 mL Vitrogen gel (1:1 mixture of Vitrogen and medium) were used to grow stock cultures in Eagle’s minimum essential medium supplemented with 10% fetal bovine serum, 2 mmol/L L-glutamine, 20 mmol/L HEPES, 100 IU/mL penicillin, and 100 ug/mL streptomycin. Medium was changed twice a week, and the cells were passaged when confluent. Carcinogen Treatment. DGBE were grown on Vitrogen-coated 35mm plates. MNNG, freshly dissolved in dimethyl sulfoxide, was diluted with culture medium to graded concentrations of 1007 to 1004 mol/L and added to rapidly dividing as well as confluent DGBE. After 48 hours, the cells were washed once with medium, and fed regular medium. When confluent, the cells were trypsinized and

1296

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

hepa

WBS: Hepatology

HEPATOLOGY Vol. 26, No. 5, 1997

MALIK ET AL.

plated into 96-well plate without Vitrogen at approximately 100 cells per well. One of the wells that grew to confluency on this plate was passaged to 35-mm plates without Vitrogen coating. At the same time, untreated DGBE cells were also seeded onto plastic plates. All further passages of MNNG-treated DGBE cells were grown on plastic. Medium was changed twice a week, and the cells were passaged when confluent (every 10-14 days) using trypsin (2.5 g/L) and ethylenediaminetetraacetic acid (1 g/L) treatment. After twelve passages, some actively dividing cells were exposed again to 1005 mol/L MNNG, as described previously, and cloned to develop multiple cell lines derived from single cells. After trypsinization, both untreated and twice-treated MNNG-treated cells were pipetted repeatedly to disrupt cell clumps into single cells. After using a hemocytometer and trypan blue dye exclusion method to determine cell count and viability, the cells were plated out at 0.5 cells per well of a 96-well plate without Vitrogen. Of the 80 wells plated with the MNNG-treated cells, 7 wells developed one colony. These cells were passaged to 35-mm plastic plates and were maintained as separate subcloned cell lines. Human gallbladder adenocarcinoma cells were cultured on plastic dishes as described by Purdum et al.9 To test whether the cells would form tumors in immunodeficient animals, 5 1 106 of untreated DGBE, MNNG-treated DGBE, and human gallbladder carcinoma cells were injected subcutaneously into separate athymic nude mice, as approved by the Animal Use Committee of the Seattle VA Medical Center. Organotypic Culture. Control and MNNG-treated cells at various passages were grown in organotypic culture as described.10 In brief, trypsinized epithelial cells were plated onto the top of a solid gel consisting of type 1 rat-tail collagen (Collaborative Biomedical Products, Bedford, MA) and human dermal fibroblasts. After epithelial cell attachment, the gels were submerged in supplemented medium containing 10% fetal bovine serum and fed daily. After 5 to 7 days, the gels were lifted to the liquid/air interphase by transferring them to metal grids placed on organ culture plates, and the gels were fed exclusively through the basal surface. The epithelial cells received all their nutritional support from the basolateral side as the medium diffused through the collagen/fibroblast matrix. After 5 days, the gels were fixed in Hollande’s fixative for thin-section light microscopy or prepared for electron microscopy.11 Thin sections of paraffin-embedded cells were stained by hematoxylin and eosin. Flow Cytometry. Cells were washed and trypsinized. Cell pellets were resuspended in 0.5 mL of 4*,6-diamidino-2-phenylindole (DAPI) staining solution containing 10% dimethylsulfoxide and stored at 0707C. The stained cell suspension was thawed before flow analysis. Samples were analyzed with epiillumination flow system designed by GOHDE (ICP21; Ortho Instruments, Westwood, MA). The procedure described by Rabinovitch et al.12 was followed for staining and flow analysis. Phosphotyrosine Assay. Subconfluent monolayers of cells were rapidly washed twice with 150 mmol/L NaCl; 3.3 mmol/L KCl at pH 7.4, and the buffer was aspirated completely. Two milliliters of sample buffer (0.5 mol/L TRIS Buffer, 2% SDS, 10% glycerol, 5% 2mercaptoethanol; 0.4% bromophenol blue) at 1007C were applied.13 The cell lysate was collected and boiled for 5 minutes, sheared 10 times through a 22-gauge needle and another 10 times through a 27-gauge needle, and stored at 0707C before use. Protein concentration was determined using a modified Lowry Assay (Sigma Chemical Co.). Approximately 10 ug of total protein was run on 7.5% SDSpolyacrylamide gels and then transferred to polyvinylidene membrane (Immobilon-P, Millipore, Bedford, MA). The transfer membrane was blocked for 2 hours, at 47C, in blocking buffer (10 mmol/ L TRIS [pH 7.5], 100 mmol/L NaCl, 0.1% Tween 20 containing 3% bovine serum albumin). The blot was incubated with 2 ug/mL of the anti-phosphotyrosine antibody for 2 hours at room temperature. After washing in TBS-Tween 20, a horseradish peroxidase-labeled secondary antibody was used to detect bound immunoglobulins. The blots were developed using an enhanced horseradish peroxidase-luminol chemiluminescence reaction (ECL, Amersham, Ar-

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

1297

lington Heights, IL) and detected with Kodak X-OMAT AR film (Kodak, Rochester, NY). Mucin Secretion Assay. Mucin assays were performed as described by Kuver et al.14 Briefly, confluent cells were labeled with N-[3H] acetyl-D-glucosamine (ICN Biomedicals, Irvine, CA), a mucin precursor, for 24 hours. The cells were washed with sterile phosphatebuffered saline followed by addition of serum-free medium. Four hours after adding agonists, the apical medium was collected and the cells harvested by trypsinization. The release of labeled glycoproteins into the apical medium and intracellular glycoprotein labeling was measured by scintillation counting after protein precipitation by 10% trichloroacetic acid/1% phosphotungstic acid. TGF-b Growth Inhibition Experiments. For growth experiments, untreated and MNNG-treated cells were plated at a density of 50,000 cells per well on Vitrogen-coated and plastic 24-well plates, respectively. After 24 hours, TGF-b1 (Calbiochem, San Diego, CA) was added in concentrations ranging from 0.1 ng/mL to 10 ng/mL in triplicate. Cell proliferation was measured by a nonradioactive colorimetric method using tetrazolium salts (CellTiter 96, Promega, Madison, WI). Forty-eight hours after adding the TGF-b1, 150 uL of tetrazolium dye in phosphate-buffered saline was added to each well. After incubation for 4 hours, 1 mL of solubilization solution was added. After 1 hour, the solution was transferred to a cuvette and read at the optical density of 570 nm, using a spectrophotometer. To confirm these results, cells were counted using a hemocytometer, after Trypan Blue staining. RESULTS Epithelial Cells After MNNG Treatment. The confluent monolayers of cultured dog gallbladder epithelial cells consisted of tall columnar cells with granular cytoplasm and erect oval shaped nuclei, as described in Oda et al.4 These cells have been in culture for over 2 years and greater than 70 passages without obvious changes in morphology. The response of these cells to different doses of MNNG was investigated. MNNG at 1004 mol/L was toxic to all nonconfluent cells and resulted in development of very few large cells in the confluent monolayers after 48 hours. Marked toxicity by MNNG was seen at the concentration of 1005 mol/L, however many cells were noted to survive. The surviving cells continued to divide, eventually showing significant morphological alterations. These changes appeared more striking in the nonconfluent (rapidly dividing) cells as opposed to confluent cells. DGBE, confluent, and actively dividing, exposed to MNNG concentration of 1006 and 1007 mol/L appeared relatively normal. In subsequent experiments, only cells treated by a MNNG concentration of 1005 mol/L were used. On the next passaging, the cells showed a preference for plastic as growth surface and did not grow as fast on Vitrogen-coated plates. Uncoated plastic dishes were used for subsequent passages of the treated cells. Simultaneous control experiments were performed where untreated cells were plated on to plastic dishes. After initial attachment, the cells plated on plastic stopped dividing and became enlarged, multi-nucleated, vacuolated, and died over a 4-week period. No untreated cell survived the change in growth conditions. MNNG-treated cells, growing as monolayers on plastic, appeared larger, cuboidal in shape, compared with untreated control cells that could grow only on Vitrogen-coated culture dishes. There were approximately 1.5 1 105 cells in a confluent 60-mm plate. The cells given a second exposure to 1005 mol/L MNNG and then subcloned into monoclonal cell lines shared similar characteristics. Untreated DGBE cells could not be subcloned. The MNNG-treated cells have been in

hepa

WBS: Hepatology

1298 MALIK ET AL.

HEPATOLOGY November 1997

culture for longer than 1 year and more than 28 passages without additional morphological changes. Neither the untreated DGBE nor the MNNG-treated cells from various passages formed subcutaneous tumors in athymic nude mice over a 3-month period. On the other hand, the human gallbladder carcinoma cells consistently produced tumors within 1 month after injection. Organotypic Cultures. When grown in an organotypic culture system, control cells were columnar and palisaded in appearance and maintained their polarity (Fig. 1A). The cells grew as a monolayer consisting of tall columnar cells with oval, erect nuclei, and were evident as a single layer by light microscopy. Intercellular vacuoles partially filled with mucus were noted in some areas. Transmission electron microscope showed columnar cells with well-developed microvilli projecting into the lumen. Earlier and later passages of these cells grown on organotypic culture had identical morphology. Distinct morphological differences were noted in MNNGtreated cells (Fig. 1B). They contained large, centrally located and hyperchromatic nuclei. The cells were generally cuboidal in shape, but showed pleomorphism. The columnar pattern was replaced by multiple layers of low cuboidal cells. A hint of stratification was noted. The overall morphology of the MNNG-treated cells was suggestive of dysplastic and premalignant morphology. The morphology of human gallbladder adenocarcinoma grown on organotypic culture was similar to MNNG treated cells, but with a more pronounced pseudostratification appearance. On the cellular level, cells were hyperchromatic and pleomorphic. These characteristics were consistent with malignant transformation. Flow Cytometry. Flow cytometry was performed on control DGBE- (Fig. 2) and MNNG-treated DGBE cells (Fig. 3) using trout erythrocytes as a standard. Mean G1 peaks for control cells and MNNG-treated cells were 54.8 and 80 respectively. The DNA index (2N) of DGBE was 2.4N and that of MNNGtreated DGBE was 3.9N indicating emergence of an aneuploid population in the MNNG-treated cells. Flow cytometry also showed that 44% of the control cells were in S phase 3 days after plating (Fig. 2A) followed by reduced S phase activity in confluent cells at day 10 to approximately 14% (Fig. 2B), which went down to under 5% after multiple days of confluence. MNNG-treated cells had 32% of cells in S phase 3 days after plating (Fig. 3A) and maintained a high number (23%) of the cells in S phase after the cells were confluent at day 9 (Fig. 3B), and they maintained this higher % S phase over many days. Phosphotyrosine Assay. Figure 4 shows the Western blot analysis using anti-phosphotyrosine antibody of normal cells (row A) and different cell lines of MNNG-treated DGBE cells (rows B through D), as well as the human gallbladder cancer cell line (row E). The number and amount of proteins phosphorylated on their tyrosine residues were consistently increased in MNNG-treated DGBE as compared with normal DGBE cells. Several novel phosphotyrosyl proteins were identified in MNNG-treated cells with molecular weights of 90, 100, and 120 kd, which were not detected in normal DGBE cells. There were only minor differences between the various cloned MNNG-treated cell lines. Mucin Secretion. The incorporation of the mucin precursor into the cell and the amount of released labeled glycoprotein was not greatly different between the DGBE- and MNNG-

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

FIG. 1. Thin sections of organotypic cultured control and MNNG-treated dog gallbladder epithelial cells and human gallbladder carcinoma cells. (A) Normal control dog cells where cells were low columnar with an apparent layer of mucin on the apical surface. (B) MNNG-treated cells where the cells had large and hyperchromatic nuclei and formed a disorganized architecture. (C) Human gallbladder carcinoma cells with multi-layer appearance and clear pleomorphism consistent with malignancy. (Original magnification 1400.)

treated cells based on mg protein. The MNNG-treated cells showed a small but statistically significant decrease in mucin secretion in response to PGE2 , whereas no significant in-

hepa

WBS: Hepatology

HEPATOLOGY Vol. 26, No. 5, 1997

MALIK ET AL.

1299

FIG. 2. Representative DNA histograms obtained from dog gallbladder epithelial cells in culture using flow cytometry. Cells were analyzed serially during the course of culture at (A) day 3 and (B) day 9 after passage.

crease or decrease was noted in response to both vasoactive intestinal peptide and epinephrine, compared with controls (Table 1). These results did not differ between MNNGtreated cells grown on plastic or Vitrogen. TGF-b Growth Experiments. The inhibition of cell growth by TGF-b1 of both DGBE- and MNNG-treated DGBE cells was dose dependent, but the concentration profile for this TGF-b1 inhibition was distinct between these two cell types. At concentrations of 10 ng/mL, 1 ng/mL, and 0.1 ng/mL, TGF-b1 inhibited DGBE by 51.6% ({1.2 SE), 50.5% ({0.7), and 30.2% ({1.8) respectively, compared with controls, in the colorimetric assay (Fig. 5). However, MNNG-treated DGBE cell proliferation was not significantly inhibited by TGF-b1 at the same concentrations. TGF-b1 at higher concentrations than 10 ng/mL did inhibit cell proliferation of MNNG-treated DGBE cells. Direct counting of the cells confirmed the colorimetric assay results.

DISCUSSION

Chemical mutagenesis is considered to play an important role in the development of cancers within the gastrointestinal tract and related organs.15 Exposure to chemical carcinogens results in genetic changes including activation of various oncogenes leading to immortality, anchorage-independent growth and ultimately neoplastic transformation in a stepwise manner.16-18 Live animal models have been used to describe the effect of carcinogens on various tissues, but this can often be cumbersome, time consuming and expensive. In this report, we have described cell lines derived from exposing dog gallbladder epithelial cells to MNNG, a potent gastrointestinal chemical carcinogen.19,20 The MNNG-treated DGBE cells showed an altered anchorage requirement by being able to grow on plastic. To show that this change in growth characteristics was because of MNNG and not secondary to spontaneous conversion of the cells, untreated

FIG. 3. Representative DNA histograms obtained from MNNG-treated dog gallbladder epithelial cells in culture using flow cytometry. Cells were serially analyzed at (A) day 3 and (B) day 9 after passage.

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

hepa

WBS: Hepatology

1300 MALIK ET AL.

HEPATOLOGY November 1997

FIG. 4. Detection of proteins with phosphorylated tyrosines in cultured epithelial cells. Proteins were analyzed by Western blotting using Phosphotyrosine antibodies. (A) Control dog gallbladder epithelial cells. (B-D) Various cloned cell lines of MNNG-treated dog gallbladder epithelial cells. (E) Cells derived from well-differentiated human gallbladder mucosal carcinoma.

DGBE were also plated on plastic dishes. The untreated cells plated onto plastic dishes did not survive nor could they be subcloned, indicating that it was the exposure to the MNNG that altered the cell’s growth capabilities. Subsequent passaging has shown continued ability of these cells to grow on plastic indicating that this change is permanent. Because the control and MNNG-treated DGBE cells were cultured on different substrates (plastic- and Vitrogen-coated dishes, respectively), the differences in their morphology could be attributed to their distinct growth conditions. To compare their morphological appearance under uniform growth parameters, the control and MNNG-treated cells, along with cultured human gallbladder cancer cells, were grown using the organotypic culture technique. Control cells in organotypic culture maintained their polarity, had welldeveloped microvilli on the apical surface, and displayed prominent apically directed secretory vesicles. These features are characteristic of a normal gallbladder epithelium in vivo. In contrast, the MNNG-treated cells in organotypic culture formed multi-layered epithelium with some morphology

TABLE 1. The Effect of Different Agonists on Mucin Secretion of MNNG-Treated DGBE Treatment (umol/L)

Control PGE2 1.0 VIP 1.0 0.1 Epinephrine 1.0

% of Control

n

100 { 2.6

7

88 { 2.0

4

111 { 15.1 105 { 13.9

6 3

113 { 11.0

4

NOTE. Values are mean { SE. Total number of treatments Å n. Treatments were performed in duplicate in each assay, and each treatment repeated in at least two separate assays. The % of control was determined for each individual treatment with the same number of controls. There was no statistical difference between treatments and control except for the small decrease in secretion in response to PGE2 (P õ .01) as determined by unpaired Student’s t test.

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

FIG. 5. The effect of TGF-b1 on growth of normal and MNNG-treated dog gallbladder epithelial cells. The cell division of normal cells was greatly inhibited by low doses of TGF-b1, while MNNG-treated cells were resistant to the growth inhbitory effects of TGF-b1. TGF-b1 was added to the cells at concentrations of 10, 1, and 0.1 ng/mL. The values are expressed as a percentage of controls incubated without TGF-b1. The results are the mean of triplicate measurements { SE. The assay was repeated three times with the same results. The values for normal cells treated with TGF-b1 were all statistically õ untreated control values (P õ .01).

characteristics resembling human gallbladder cancer cells. Exposure of cultured DGBE to MNNG led to distinct morphological changes that persisted on subsequent passages indicating that the effect of MNNG was long-term and inheritable and therefore could not be attributed to short-term toxicity. The DNA content of MNNG-treated cells was compared with control DGBE using flow cytometric analysis. The MNNG-treated cells had a DNA index of 4N in 60% to 70% of cell population. This DNA index indicates the presence of an aneuploid cell population, indicating partial genomic loss or duplication and confirming a change in the DNA content. Presence of an aneuploid cell population is considered to be pathological and represents one of the spectrum of genomic alterations reflecting steps towards progression to invasive cancer.21,22 Comparison of the S or proliferative phase revealed a persistently higher proliferative phase in the MNNGtreated cells after the cells achieved confluence suggesting that the cells exposed to MNNG have lost some of the contact-mediated inhibition of cell division seen in normal cells. Studies have shown that signal transduction processes leading to cell proliferation are partly mediated through tyrosine phosphorylation of various intracellular proteins.7,8 Also, tyrosine specific protein kinase activation has been described in oncogene-induced malignant transformation.23,24 Cells transformed by these oncogenes contain increased levels of proteins phosphorylated on tyrosine residues. In the normal counterparts of these transformed cells, tyrosine phosphorylation accounts for a very small percentage of the total phosphorylated amino acids, the majority being phosphorylated serine and threonine. Hence, it has been proposed to use phosphorylated tyrosines as markers of transformation of human tumors in which tyrosine kinase is activated.13 In this report, we showed an increase in the number and amount of phosphotyrosine-containing proteins in dog gallbladder cells treated with MNNG compared with control DGBE. In the human gallbladder carcinoma cells, the levels of the tyro-

hepa

WBS: Hepatology

HEPATOLOGY Vol. 26, No. 5, 1997

MALIK ET AL.

sine-phosphorylated proteins were also high, indicating an increase in their tyrosine kinase activity. Preliminary results indicate that cyclin D1, a cell cycle check point gene, is overexpressed in the treated cells (Nadir A, personal communication, May 1996).25,26 Gallbladder and intestinal epithelial cells are covered by a coating of mucus, of which the major constituent is mucin, a glycoprotein. Mucin secretion is considered to be a unique property of columnar epithelial cells. Studies have shown that mucin secretion in normal cells is regulated by various secretagogues such as prostaglandins, vasoactive intestinal peptide, and epinephrine via an adenosine 3*,5*-cyclic monophosphate signal transduction pathway,14,27 as well as bile salts.28 However, regulation of mucin secretion has been shown to be altered in various human carcinoma cell lines.29 As previously described, normal DGBE cells show increased mucin secretion in response to vasoactive intestinal peptide, PGE2 , and epinephrine.14 However, the MNNG-treated cells showed no change in mucin secretion in response to epinephrine or vasoactive intestinal peptide. On the other hand, PGE2 was observed to significantly decrease mucin secretion in these treated cells. The altered mucin secretion response to various secretagogues indicates that the MNNG-treated cells have undergone modifications resulting in a change in its basic cellular function. Gallbladder cells are targets for various kinds of cytokines, and a further understanding of actions of cytokines on the biology of gallbladder epithelial cells is important. TGF-b is a cytokine that shows regulatory potential on growth, differentiation and morphogenesis.30-33 A readily measurable biological activity of TGF-b is its anti-proliferative action on various cell types.34 Normally, TGF-b arrests epithelial cells in the G1 phase of their cell cycle.34-36 Although many normal epithelial cells are inhibited by TGF-b, their transformed counterparts are often resistant to its inhibitory effects.37-43 The resistance of these transformed cell lines to TGF-b supports the idea that the loss of the TGF-b induced growth inhibition might be an important step during carcinogenesis.37 The TGF-b receptor complex is a new addition to the roster of human tumor suppresser genes, and genetic mutations that inactivate this tumor suppresser are believed to be important in the genesis of multiple different malignancies.44 The significant inhibition of cell growth of the cultured dog gallbladder epithelial cells, but not of the MNNG-treated cells, by TGF-b1 again shows a major alteration between these two cell types. The MNNG-treated cells may not have developed into fully malignant cancer cells because they did not form tumors in athymic mice. However, these cells were exposed to one carcinogen so there may not have been enough genetic alteration to create truly cancerous cells. Although tumorigenesis in immunodeficient mice is a behavior of most malignant cells, there are examples of cultured carcinoma cells failing to evolve into tumors in immunodeficient mice.45 Although no major differences were observed between the original MNNG cell line and the cell lines after a second MNNG exposure, the progress to true cancer cells may require additional exposure to carcinogens other than MNNG or the use of various clonal selection procedures. Cancer is understood to be a clonal selection process where those cells that lose control over their cell division or fail to respond to apoptosis signals have an advantage over other cells. The differences in the various parameters measured, i.e., attach-

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

1301

ment capabilities, morphology, ploidy, mucin secretion, tyrosine phosphorylation, and TGF-b response, have all been implicated in the development of various cancers. In each of these cases studied, the MNNG-treated cells showed impressive differences, compared with untreated cells. While the human carcinoma cells have undergone malignant change, the MNNG-treated cells share many of the characteristics of transformed cells but apparently without complete malignant transformation. In conclusion, normal dog gallbladder epithelial cells provide an excellent model to study the early genetic events involved in the process of carcinogenesis. The addition of a carcinogen to these cells leads to alterations in morphology, growth, and physiological characteristics, all preliminary steps on the pathway to transformation, clearly indicating that these cultured cells can be chemically modified. This in vitro system may afford an opportunity to better understand the evolution of genetic markers,46 cell kinetics, morphology, and cell biological function such as mucin synthesis and secretion during the process of carcinogenesis.47-49 This system may also be useful in the study of putative carcinogenic or anti-carcinogenic effects of test compounds, and at the same time reduce the need for using live animals. Acknowledgment: We thank Dr. Peter Rabinovitch for his assistance with flow cytometry; Dr. Preston Purdum for the human gallbladder cancer cells; and Audrey Wass for the electron microscopy. REFERENCES 1. Shukla VK, Tiwari SC, Roy SK. Biliary bile acids in cholelithiasis and carcinoma of the gallbladder. Eur J Cancer Prev 1993;2:155-160. 2. Funabiki T, Suigiue K, Matsubara T, Amano H, Ochiai M. Bile acids and biliary carcinoma in pancreaticobiliary maljunction. Keio J Med 1991;40:118-122. 3. Broome U, Lofberg R, Veress B, Ljusk SE. Primary sclerosing cholangitis and ulcerative colitis: Evidence for increased neoplastic potential. HEPATOLOGY 1995;22:1404-1408. 4. Oda D, Lee SP, Hayashi A. Long-term culture and partial characterization of dog gallbladder epithelial cells. Lab Invest 1991;64:682-692. 5. Oda D, Savard CE, Nguyen TD, Eng L, Swenson ER, Lee SP. Dog pancreatic duct epithelial cells: Long-term culture and characterization. Am J Path 1996;148:977-985. 6. Oda D, Savard CE, Eng L, Lee SP. The effect of N-methyl-N*-nitro-Nnitrosoguanidine (MNNG) on cultured dog pancreatic duct epithelial cells. Pancreas 1996;12:109-116. 7. Comoglio PM, Di Renzo MF, Gaudino G, Ponzetto C, Prat M. Tyrosine Kinase and control of cell proliferation. Am Rev Respir Dis 1990;142: S16-S19. 8. Saggioro D, Ferrancini R, Di Renzo MF, Naldini L, Chieco-Bianchi L, Comoglio PM. Protein phosphorylation at tyrosine residues in v-abl transformed mouse lymphocytes and fibroblasts. Int J Cancer 1986;37: 623-628. 9. Purdum PP, Ulissi A, Hylemon PB, Shiffman ML, Moore EW. Cultured human gallbladder epithelia. Methods and partial characterization of a carcinoma-derived model. Laboratory Investigation 1993;68:345-353. 10. Merrick DT, Blanton RA, Gown AM, McDougall JK. Altered expression of proliferation and differentiation markers in human papillomavirus 16 and 18 immortalized epithelial cells grown in organotypic culture. Am J Pathol 1992;140:167-177. 11. Karnovski MJ. A formaldehyde-gluteraldehyde fixative of high osmolarity for use in electron microscopy [Abstract]. J Cell Biol 1965;27:137A. 12. Rabinovitch PS, O’Brien K, Simpson M, Callis JB, Hoehn H. Flow cytogenetics. High-resolution ploidy measurements in human fibroblast cultures. Cytogenet Cell Genet 1981;29:65-76. 13. Pories SE, Weber TK, Simpson H, Greathead P, Steele G, Summerhayes IC. Immortalization and neoplastic transformation of normal rat colon epithelium: An in vitro model of colonic neoplastic progression. Gastroenterology 1993;146:1346-1355. 14. Kuver R, Savard CE, Oda D, Lee SP. Prostaglandin E generates intracellular cAMP and accelerates mucin secretion by cultured dog gallbladder epithelial cells. Am J Physiol 1994;267:G998-G1003.

hepa

WBS: Hepatology

1302 MALIK ET AL.

HEPATOLOGY November 1997

15. Thomas G. Genetics and molecular biology of gastrointestinal cancer. In: Ahlgreen J, Macdonald J, eds. Gastrointestinal Oncology. Philadelphia: Lippincott, 1992:559-569. 16. Paraskeva C, Corfield AP, Harper S, Hague A, Audcent K, Williams AC. Colorectal carcinogenesis: Sequential steps in the in vitro immortalization and transformation of human colonic epithelial cells (review). Anticancer Res 1990;10:1189-1200. 17. Rhim JS. Neoplastic transformation of human epithelial cells in vitro. Anticancer Res 1989;9:1345-1366. 18. Druker BJ, Mamon HJ, Roberts TM. Oncogenes, growth factors, and signal transduction. NEJM 1989;321:1383-1390. 19. Kodama M, Kodama T, Kodama M. Is N-methyl-N*-nitro-N-nitrosoguanidine a hormonal carcinogen? (review). Anticancer Res 1991;11:941946. 20. Lee LW, Tsao MS, Grisham JW, Smith GJ. Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N*-nitroN-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions. Am J Path 1989;135:6371. 21. Orfao A, Ciudad J, Gonzalez M, Lopez A, Del-Mar-Abad M, Paz-Bouza JI, Cruz JJ, et al. Flow cytometry in the diagnosis of cancer. Scand J Clin Lab Invest Suppl 1995;221:145-152. 22. Crissman JD, Visscher DW, Sarkar FH. Premalignant lesions of the upper aerodigestive tract: Biomarkers of genetic alterations, proliferation, and differentiation. J Cell Biochem Suppl 1993;17F:192-198. 23. Giordano S, Di Renzo MF, Cirillo D, Naldini L, Chiado’Piat L, Comoglio PM. Proteins phosphorylated on tyrosine as markers of human tumor cell lines. Int J Cancer 1987;39:482-487. 24. Kamps MP, Sefton BM. Identification of multiple novel polypeptide substrates of the v-src, v-yes, v-fps, v-ros, and v-erb-B oncogenic tyrosine protein kinases utilizing antisera against phosphotyrosine. Oncogene 1988;2:305-315. 25. Motokura T, Arnold A. Cyclin D and oncogenesis. Curr Opin Genet Dev 1993;3:5-10. 26. Hunter T, Pines J. Cyclins and cancer : Cyclin D and CDK inhibitors come of age. Cell 1994;79:573-582. 27. Wahlin T, Thornell E, Jivegard L, Svanvik J. Effects of intraluminal prostaglandin E2 in vivo on secretory behavior and ultrastructural changes in mouse gallbladder epithelium. Gastroenterology 1988;95: 1632-1635. 28. Klinkspoor JH, Kuver R, Savard CE, Oda D, Azzouz H, Tytgat GNJ, Groen AK, et al. Model bile and bile salts accelerate mucin secretion by cultured dog gallbladder epithelial cells. Gastroenterology 1995;109: 264-274. 29. Yedgar S, Eidelman O, Malden E, Roberts D, Etcheberrigaray R, Goping G, Fox C, et al. Cyclic AMP-independent secretion of mucin by SW1116 human colon carcinoma cells. Differential control by Ca2/ ionophore A23187 and arachidonic acid. Biochem J 1992;283:421-426. 30. Hofler P, Wehrle I, Bauer G. TGF-b induces an inhibitory effect of normal cells directed against transformed cells. Int J Cancer 1993;54: 125-130. 31. Barnard JA, Lyons RM, Moses HL. The cell biology of transforming growth factor beta. Biochem Biophys Acta 1990;1032:79-87. 32. Massague J. The transforming-growth factor beta family. Ann Rev Cell Biol 1990;6:597-641.

AID

Hepa 0044

/

5p28$$$861

10-13-97 12:12:20

33. Sporn MB, Roberts AB. Transforming growth factor-b: Recent progress and new challenges J Cell Biol 1990;119:1017-1021. 34. Geng Y, Weinberg RA. Transforming growth factor b effects on expression of G1 cyclins and cyclin-dependent protein kinases. Proc Natl Acad Sci U S A 1993;90:10315-10319. 35. Laiho M, DeCaprio JA, Ludlow JW, Livingston DM, Massague J. Growth inhibition of TGF -b linked suppression of retinoblastoma protein phosphorylation. Cell 1990;62:175-185. 36. Howe PH, Draetta G, Leof EB. Transforming growth factor b 1 inhibition of p34 cdc2 phosphorylation and histone H1 kinase activity is associated with G1/S-phase growth arrest. Mol Cell Biol 1991;11:1185-1194. 37. Manning AM, Williams AC, Game SM, Paraskeva C. Differential sensitivity of human colonic adenoma and carcinoma cells to transforming growth factor b (TGF -b): conversion of adenoma cell line to a tumorigenic phenotype is accompanied by a reduced response to the inhibitory effect of TGF-b . Oncogene 1991;6:1471-1476. 38. Cofffey RJ Jr., Sipes NJ, Bascom CC, Graves-Deal R, Pennington CY, Weissmann BE, Moses HL. Growth modulation of mouse keratinocytes by transforming growth factors. Cancer Res 1988;48:1596-1602. 39. Game SM, Stone A, Matthews JB, Scully C, Prime SS. Differentiation of malignant oral rat keratinocytes reflects changes in EGF and TGF- b receptor expression but not growth factor dependence. Carcinogenesis 1991;12:409-416. 40. Knabbe C, Lippmann ME, Wakefield LM, Flanders FC, Kasid A, Derynck R, Dickson RB. Evidence that transforming growth factor- b is a hormonally regulated negative growth factor in human breast cancer cells. Cell 1987;48:417-428. 41. McMahon JB, Richard WL, del Campo AA, Song MH, Thorgeirsson SS. Differential effects of transforming growth factor- b on proliferation of normal and malignant rat liver cells in culture. Cancer Res 1986;46: 4665-4671. 42. Masui T, Wakefield LM, Lechner JF, LaVeck MA, Sporn MB, Harris CC. Type beta transforming growth factor is the primary differentiationinducing serum factor for normal human bronchial epithelial cells. Proc Natl Acad Sci U S A 1986;83:2438-2442. 43. Kimchi A, Wang XF, Weiberg RA, Cheifetz S, Massague J. Absence of TGF-b receptors and growth inhibitory responses in retinoblastoma cells. Science 1988;240:196-199. 44. Markowitz SD, Roberts AB. Tumor suppressor activity of the TGF -b pathway in human cancers. Cytokine Growth Factor Rev 1996;7:93102. 45. Chang SE. Human oral keratinocyte cultures and in vitro model systems for studying oral carcinogenesis. In: Johnson NW, ed. Oral Cancer. Cambridge: Cambridge University Press, 1991:340-363. 46. Tahara E. Genetic alterations in human gastrointestinal cancers. The application to molecular diagnosis. Cancer 1995;75:1410-1417. 47. Wistuba II, Sugio K, Hung J, Kishimoto Y, Virmani AK, Roa I, AlboresSaavedra J, et al. Allele-specific mutations involved in the pathogenesis of endemic gallbladder carcinoma in Chile. Cancer Res 1995;55:25112515. 48. Teh M, Wee A, Raju GC. An immunohistochemical study of p53 protein in gallbladder and extrahepatic bile duct/ampullary carcinomas. Cancer 1994;74:1542-1545. 49. Watanabe M, Asaka M, Tanaka J, Kurosawa M, Kasai M, Miyazaki T. Point mutations of K-ras gene codon 12 in biliary tract tumors. Gastroenterology 1994;107:1147-1153.

hepa

WBS: Hepatology