Spontaneous Neoplastic Transformation of WB-F344 Rat Liver Epithelial Cells
1998; Elsevier BV; Volume: 153; Issue: 6 Linguagem: Inglês
10.1016/s0002-9440(10)65705-7
ISSN1525-2191
AutoresMichelle J. Hooth, William B. Coleman, Sharon C. Presnell, Kristen M. Borchert, Joe W. Grisham, Gary J. Smith,
Tópico(s)Renal and related cancers
ResumoSeveral studies have shown that cultured rat liver epithelial cells transform spontaneously after chronic maintenance in a confluent statein vitro. In the present study, multiple independent lineages of low-passage WB-F344 rat liver epithelial stem-like cells were initiated and subjected in parallel to selection for spontaneous transformation to determine whether spontaneous acquisition of tumorigenicity was the result of events (genetic or epigenetic) that occurred independently and stochastically, or reflected the expression of a pre-existing alteration within the parental WB-F344 cell line. Temporal analysis of the spontaneous acquisition of tumorigenicity by WB-F344 cells demonstrated lineage-specific differences in the time of first expression of the tumorigenic phenotype, frequencies and latencies of tumor formation, and tumor differentiations. Although spontaneously transformed WB-F344 cells produced diverse tumor types (including hepatocellular carcinomas, cholangiocarcinomas, hepatoblastomas, and osteogenic sarcomas), individual lineages yielded tumors with consistent and specific patterns of differentiation. These results provide substantial evidence that the stochastic accumulation of independent transforming events during the selection regimen in vitro were responsible for spontaneous neoplastic transformation of WB-F344 cells. Furthermore, cell lineage commitment to a specific differentiation program was stable with time in culture and with site of transplantation. This is the first report of a cohort of related, but independent, rat liver epithelial cell lines that collectively produce a spectrum of tumor types but individually reproduce a specific tumor type. These cell lines will provide valuable reagents for investigation of the molecular mechanisms involved in the differentiation of hepatic stem-like cells and for examination of potential causal relationships in spontaneously transformed rat liver epithelial cell lines between molecular/cellular alterations and the ability to produce tumors in syngeneic animals. Several studies have shown that cultured rat liver epithelial cells transform spontaneously after chronic maintenance in a confluent statein vitro. In the present study, multiple independent lineages of low-passage WB-F344 rat liver epithelial stem-like cells were initiated and subjected in parallel to selection for spontaneous transformation to determine whether spontaneous acquisition of tumorigenicity was the result of events (genetic or epigenetic) that occurred independently and stochastically, or reflected the expression of a pre-existing alteration within the parental WB-F344 cell line. Temporal analysis of the spontaneous acquisition of tumorigenicity by WB-F344 cells demonstrated lineage-specific differences in the time of first expression of the tumorigenic phenotype, frequencies and latencies of tumor formation, and tumor differentiations. Although spontaneously transformed WB-F344 cells produced diverse tumor types (including hepatocellular carcinomas, cholangiocarcinomas, hepatoblastomas, and osteogenic sarcomas), individual lineages yielded tumors with consistent and specific patterns of differentiation. These results provide substantial evidence that the stochastic accumulation of independent transforming events during the selection regimen in vitro were responsible for spontaneous neoplastic transformation of WB-F344 cells. Furthermore, cell lineage commitment to a specific differentiation program was stable with time in culture and with site of transplantation. This is the first report of a cohort of related, but independent, rat liver epithelial cell lines that collectively produce a spectrum of tumor types but individually reproduce a specific tumor type. These cell lines will provide valuable reagents for investigation of the molecular mechanisms involved in the differentiation of hepatic stem-like cells and for examination of potential causal relationships in spontaneously transformed rat liver epithelial cell lines between molecular/cellular alterations and the ability to produce tumors in syngeneic animals. Rat liver epithelial cell lines have been used extensively for investigation of the cellular stages of neoplastic transformation in vitro. Such cell lines have been established in culture from both normal rat livers1Tsao M-S Smith JD Nelson KG Grisham JW A diploid epithelial cell line from normal adult rat liver with phenotypic properties of 'oval' cells.Exp Cell Res. 1984; 154: 38-52Crossref PubMed Scopus (371) Google Scholar, 2McMahon JB Richards WL delCampo AA Song MK Thorgeirsson SS Differential effects of transforming growth factor-β on proliferation of normal and malignant rat liver epithelial cells in culture.Cancer Res. 1986; 46: 4665-4671PubMed Google Scholar, 3Tsao M-S Liu C Inhibition of growth of early passage normal rat liver epithelial cell lines by epidermal growth factor.Lab Invest. 1988; 58: 636-643PubMed Google Scholar and from the livers of carcinogen-treated rats.4Braun L Goyette M Yaswan P Thompson NL Fausto N Growth in culture and tumorigenicity after transfection with the ras oncogene of liver epithelial cells from carcinogen-treated rats.Cancer Res. 1987; 47: 4116-4124PubMed Google Scholar, 5Fausto N Thompson HL Braun L Purification and culture of oval cells from rat liver.in: Pretlow II, TG Pretlow TG Cell Separation Methods and Selected Applications. vol 4. Academic Press, FL1987: 45-77Google Scholar Previous studies have shown that neoplastic transformation of these cellsin vitro can be achieved spontaneously,6Lee LW Tsao M-S Grisham JW Smith GJ Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N′-nitro-N-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions.Am J Pathol. 1989; 135: 63-71PubMed Google Scholar, 7Tsao M-S Shepherd J Batist G Phenotypic expression in spontaneously transformed cultured rat liver epithelial cells.Cancer Res. 1990; 50: 1941-1947PubMed Google Scholar, 8Huggett AC Ellis PA Ford CP Hampton LL Rimoldi D Thorgeirsson SS Development of resistance to the growth inhibitory effects of transforming growth factor β1 during the spontaneous transformation of rat liver epithelial cells.Cancer Res. 1991; 51: 5929-5936PubMed Google Scholar through carcinogen treatment,2McMahon JB Richards WL delCampo AA Song MK Thorgeirsson SS Differential effects of transforming growth factor-β on proliferation of normal and malignant rat liver epithelial cells in culture.Cancer Res. 1986; 46: 4665-4671PubMed Google Scholar, 9Tsao M-S Grisham JW Nelson KG Smith JD Phenotypic and karyotypic changes induced in cultured rat hepatic epithelial cells that express the "oval" cell phenotype by exposure to N-methyl-N′-nitro-N-nitrosoguanidine.Am J Pathol. 1985; 118: 306-315PubMed Google Scholar and by transfection with activated oncogenes.10Garfield S Huber BE Nagy P Cordingley MG Thorgeirsson SS Neoplastic transformation and lineage switching of rat liver epithelial cells by retrovirus-associated oncogenes.Mol Carcinog. 1988; 1: 189-195Crossref PubMed Scopus (46) Google Scholar, 11Houck KA Michalopoulos GK Strom SC Introduction of a Ha-ras oncogene into rat liver epithelial cells and parenchymal hepatocytes confers resistance to the growth inhibitory effects of TGF-β.Oncogene. 1989; 4: 19-25PubMed Google Scholar, 12Hampton LL Worland PJ Yu B Thorgeirsson SS Huggett AC Expression of growth-related genes during tumor progression in v-raf-transformed rat liver epithelial cells.Cancer Res. 1990; 50: 7460-7467PubMed Google Scholar, 13Presnell SC Thompson MT Strom SC Investigation of the cooperative effects of transforming growth factor α and c-myc overexpression in rat liver epithelial cells.Mol Carcinog. 1995; 13: 233-244Crossref PubMed Scopus (13) Google Scholar Neoplastic derivatives of rat liver epithelial cells produce a variety of undifferentiated and differentiated tumors after transplantation to subcutaneous or intraperitoneal sites of nude mice or syngeneic rats, including hepatocellular carcinomas, adenocarcinomas, epidermoid carcinomas, hepatoblastomas, and sarcomas.14Tsao M-S Grisham JW Hepatocarcinomas, cholangiocarcinomas, and hepatoblastomas produced by chemically transformed cultured rat liver epithelial cells: a light- and electron-microscopic analysis.Am J Pathol. 1987; 127: 168-181PubMed Google Scholar, 15Tsao M-S Zhang X-Y The effects of continuous exposure to epidermal growth factor on the spontaneous transformation of cultured rat liver epithelial cells.Am J Pathol. 1992; 140: 85-94PubMed Google Scholar, 16Williams AO Huggett AC Thorgeirsson SS Pathology of spontaneous and oncogene transformed rat liver epithelial cells and derived tumours in nude mice.Int J Exp Pathol. 1992; 73: 99-114PubMed Google Scholar The differentiated types of tumors reflect the spectrum of liver neoplasms that develop in mammalian liver, suggesting that an undifferentiated stem-like cell may represent the origin of various liver tumors in vivo. The WB-F344 rat liver epithelial cell line has been used to investigate the process of neoplastic transformation in vitro and hepatocarcinogenesis in vivo.14Tsao M-S Grisham JW Hepatocarcinomas, cholangiocarcinomas, and hepatoblastomas produced by chemically transformed cultured rat liver epithelial cells: a light- and electron-microscopic analysis.Am J Pathol. 1987; 127: 168-181PubMed Google Scholar, 17Tsao M-S Grisham JW Nelson KG Clonal analysis of tumorigenicity and paratumorigenic phenotypes in rat liver epithelial cells chemically transformed in vitro.Cancer Res. 1985; 45: 5139-5144PubMed Google Scholar, 18Tsao M-S Grisham JW Phenotypic modulation during tumorigenesis by clones of transformed rat liver epithelial cells.Cancer Res. 1987; 47: 1282-1286PubMed Google Scholar We have shown previously that recovery of spontaneously transformed WB-F344 cells was selectively enhanced from a single parental population when cultures were maintained at confluence with infrequent passaging compared with cultures maintained in exponential growth.6Lee LW Tsao M-S Grisham JW Smith GJ Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N′-nitro-N-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions.Am J Pathol. 1989; 135: 63-71PubMed Google Scholar In the present study, we initiated multiple independent lineages of low-passage WB-F344 cells and subjected them in parallel to selection for spontaneous transformation to determine whether spontaneous acquisition of tumorigenicity was the result of events (genetic or epigenetic) that occurred independently and stochastically, or reflected the expression of a pre-existing alteration within the parental WB-F344 cell line. We reasoned that if the ability to produce tumors resulted from the stochastic accumulation of several spontaneous alterations, a large number of independent cultures established from the WB-F344 parental culture and grown under selection growth conditions would acquire the ability to produce tumors at different times and express different paratumorigenic and tumorigenic phenotypes. However, if a pre-existing heritable alteration, which predisposed the cells to neoplastic transformation, was present in the parental WB-F344 cell line, tumorigenicity would arise in all of the lineages at similar times and/or the tumors would be phenotypically similar. Temporal analysis of the spontaneous acquisition of tumorigenicity by WB-F344 cells demonstrated lineage-specific differences in the time of appearance of tumorigenicity, frequency, and latency of tumor formation and histology of the tumors formed. Although spontaneously transformed WB-F344 cells produced diverse tumor types, individual lineages yielded tumors with consistent and specific patterns of differentiation. These results provide substantial evidence that the stochastic accumulation of independent transforming events during the selection regimenin vitro was responsible for spontaneous neoplastic transformation of WB-F344 cells. Furthermore, commitment to a specific differentiation program was stable with time in culture and with site of transplantation. Diploid WB-F344 rat liver epithelial cells1Tsao M-S Smith JD Nelson KG Grisham JW A diploid epithelial cell line from normal adult rat liver with phenotypic properties of 'oval' cells.Exp Cell Res. 1984; 154: 38-52Crossref PubMed Scopus (371) Google Scholar at passage 4 served as the founder cell population for the initiation of multiple independent lineages cultured separately under the selective growth conditions. Normal WB-F344 cells are contact inhibited, do not form colonies in soft-agarose, and are nontumorigenic in neonatal Fischer-344 rats.1Tsao M-S Smith JD Nelson KG Grisham JW A diploid epithelial cell line from normal adult rat liver with phenotypic properties of 'oval' cells.Exp Cell Res. 1984; 154: 38-52Crossref PubMed Scopus (371) Google Scholar The founding wild-type WB-F344 cell population served as the control cells for the phenotypic characterization of experimental populations in vitro and determination of their tumorigenic potential in vivo. All cells were cultured in Richter's improved minimal essential medium with zinc option (Irvine Scientific, Santa Ana, CA) supplemented as described previously.6Lee LW Tsao M-S Grisham JW Smith GJ Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N′-nitro-N-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions.Am J Pathol. 1989; 135: 63-71PubMed Google Scholar The experimental design for the generation of spontaneous transformants of WB-F344 cells by maintenance under selective growth conditions (Figure 1) was modified from the protocol described previously.6Lee LW Tsao M-S Grisham JW Smith GJ Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N′-nitro-N-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions.Am J Pathol. 1989; 135: 63-71PubMed Google Scholar Eighteen separate experimental cell populations were plated at a density of 2.9 × 106 cells/150-mm tissue culture dish. The final cell density of each plate was determined at the end of each cycle of selective growth by counting trypsinized cells with a Coulter counter (Coulter Electronics, Hialeah, FL). All of the cells harvested from one plate were divided equally among four 150-mm tissue culture dishes. One of the four dishes became the experimental population for the next selection cycle and was maintained under the selective regimen. The remaining three dishes were grown to confluence to provide cells for cryopreservation, phenotypic characterization, and tumorigenicity assays. Experimental cell populations were subjected to a minimum of 10 cycles of selective growth. The nomenclature for the experimental cell populations incorporates the name of the parental WB-F344 cell line, the lineage identification number (L1 to L20), and the selection cycle number (C1 to C12). Cell morphologies and growth patterns at confluence were evaluated by phase contrast microscopy. For determination of saturation densities in monolayer cultures, cells were plated at a density of 1.0 × 105 cells per 60-mm tissue culture dish and maintained in culture with a medium change every 4 days. At the end of 14 days, the cells were harvested and enumerated with a hemacytometer. For determination of saturation densities at the end of a selection cycle, cells were plated at a density of 2.0 × 104 per well on a 24-well tissue culture dish and maintained in culture with a medium change every 7 days. At the end of 28 days, the cells were harvested and counted. Anchorage-independent growth was assayed as described previously.19Tsao M-S Earp HS Grisham JW Gradation of carcinogen-induced capacity for anchorage-independent growth in cultured rat liver epithelial cells.Cancer Res. 1985; 45: 4428-4432PubMed Google Scholar, 20Coleman WB McCullough KD Esch GL Civalier CJ Livanos E Weissman BE Grisham JW Smith GJ Suppression of the tumorigenic phenotype of a rat liver epithelial tumor cell line by the p11.2-p12 region of human chromosome 11.Mol Carcinog. 1995; 13: 220-232Crossref PubMed Scopus (18) Google Scholar Neoplastic transformation of the independent populations was indicated by the formation of tumors after the subcutaneous transplantation of cells into neonatal syngeneic rats. Tumorigenicity assays were performed as described previously.17Tsao M-S Grisham JW Nelson KG Clonal analysis of tumorigenicity and paratumorigenic phenotypes in rat liver epithelial cells chemically transformed in vitro.Cancer Res. 1985; 45: 5139-5144PubMed Google Scholar, 20Coleman WB McCullough KD Esch GL Civalier CJ Livanos E Weissman BE Grisham JW Smith GJ Suppression of the tumorigenic phenotype of a rat liver epithelial tumor cell line by the p11.2-p12 region of human chromosome 11.Mol Carcinog. 1995; 13: 220-232Crossref PubMed Scopus (18) Google Scholar WB-F344 cells of the founding cell population were transplanted similarly as controls. Clonal populations of tumorigenic cells were established from tumors produced by the heterogeneous spontaneously transformed lineages as described previously.6Lee LW Tsao M-S Grisham JW Smith GJ Emergence of neoplastic transformants spontaneously or after exposure to N-methyl-N′-nitro-N-nitrosoguanidine in populations of rat liver epithelial cells cultured under selective and nonselective conditions.Am J Pathol. 1989; 135: 63-71PubMed Google Scholar The nomenclature for the tumor cell lines incorporates the name of the heterogeneous cell population that produced the tumor (L1–20, C1–12), the tumor identification number (T1-T6), and the subclone identification number.1Tsao M-S Smith JD Nelson KG Grisham JW A diploid epithelial cell line from normal adult rat liver with phenotypic properties of 'oval' cells.Exp Cell Res. 1984; 154: 38-52Crossref PubMed Scopus (371) Google Scholar, 2McMahon JB Richards WL delCampo AA Song MK Thorgeirsson SS Differential effects of transforming growth factor-β on proliferation of normal and malignant rat liver epithelial cells in culture.Cancer Res. 1986; 46: 4665-4671PubMed Google Scholar, 3Tsao M-S Liu C Inhibition of growth of early passage normal rat liver epithelial cell lines by epidermal growth factor.Lab Invest. 1988; 58: 636-643PubMed Google Scholar, 4Braun L Goyette M Yaswan P Thompson NL Fausto N Growth in culture and tumorigenicity after transfection with the ras oncogene of liver epithelial cells from carcinogen-treated rats.Cancer Res. 1987; 47: 4116-4124PubMed Google Scholar, 5Fausto N Thompson HL Braun L Purification and culture of oval cells from rat liver.in: Pretlow II, TG Pretlow TG Cell Separation Methods and Selected Applications. vol 4. Academic Press, FL1987: 45-77Google Scholar To confirm their tumorigenic potential and to evaluate their differentiation potential at a different transplantation site, 5 × 106 cells from each clonal tumor cell line were injected intraperitoneally into adult (3-month old) male Fischer-344 rats as described previously.20Coleman WB McCullough KD Esch GL Civalier CJ Livanos E Weissman BE Grisham JW Smith GJ Suppression of the tumorigenic phenotype of a rat liver epithelial tumor cell line by the p11.2-p12 region of human chromosome 11.Mol Carcinog. 1995; 13: 220-232Crossref PubMed Scopus (18) Google Scholar Studies involving the use of animals were carried out in accordance with federal and institutional guidelines put forth by the National Institutes of Health and the Institutional Animal Care and Use Committee of the University of North Carolina at Chapel Hill. Sections of tumor tissue were fixed in buffered formalin, processed for paraffin sections, and stained with hematoxylin and eosin (H&E) for histological analysis. Formalin-fixed tissues were fixed additionally in 3% glutaraldehyde in 0.15 mol/L sodium phosphate buffer and processed for transmission electron microscopy. Low-passage WB-F344 cells did not produce tumors during 1 year after subcutaneous transplantation into neonatal syngeneic rats. None of the lineages produced tumors before selection cycle 8, but 5/18 lineages, 7/18 lineages, and 1/18 lineages first displayed tumorigenic potential at selection cycles 8, 10, and 11, respectively (Table 1). Five lineages (L10, L13, L17, L18, and L19) were not tumorigenic at any selection cycle. Tumor incidence increased and/or latency of tumor formation decreased with increasing numbers of selection cycles in 6/7 lineages (Table 1) that were tumorigenic at more than one selection cycle. Tumorigenic potential was not related to the number of accumulated population doublings (Table 2). On average, nontumorigenic lineages accumulated 20.1 ± 1.1 (n = 5) population doublings, and tumorigenic lineages accumulated 20.7 ± 0.6 (n = 13) population doublings during the course of 10 selection cycles.Table 1Tumorigenicity of Transplanted Spontaneous Transformants of WB-F344 Rat Liver Epithelial CellsLineageSelection cycle*Selection cycle following which cells were transplanted into syngeneic rats.Subcutaneous tumors†Fraction represents number of rats with tumor over total number of rats transplanted with cells.Latency (weeks, mean ± SEM)‡Mean time to detection and harvest of 1 cm. diameter tumors.Range (weeks)§Range of tumor latencies.181/7 (14%)15107/7 (100%)14 ± 0.7313–182108/8 (100%)22 ± 0.5618–231110/10 (100%)18 ± 0.9214–233111/7 (14%)52483/6 (50%)35 ± 4.1027–4098/9 (89%)30 ± 2.8522–44109/9 (100%)18 ± 0.0718682/11 (18%)53 ± 0.3653107/7 (100%)16 ± 0.5513–177105/18 (28%)35 ± 1.4331–40116/8 (75%)42 ± 4.4329–518103/9 (33%)35 ± 2.6731–409102/19 (11%)43 ± 10.5732–5411102/10 (20%)50 ± 2.2147–5212109/9 (100%)32 ± 1.0930–401485/10 (50%)35 ± 1.2930–371681/10 (10%)52106/14 (43%)38 ± 4.3427–5220107/7 (100%)19 ± 0.8416–21111/10 (10%)51Lineages and selection cycles that were negative for tumor formation are not displayed in this table. There were no differences in tumor formation for individual lineages between male and female host rats (data not shown).* Selection cycle following which cells were transplanted into syngeneic rats.† Fraction represents number of rats with tumor over total number of rats transplanted with cells.‡ Mean time to detection and harvest of 1 cm. diameter tumors.§ Range of tumor latencies. Open table in a new tab Table 2Phenotypic Characteristics of Spontaneous Transformants of WB-F344 Rat Liver Epithelial Cells after 10 Cycles of Selective GrowthLineageMorphological patternSaturation density (28-day)Anchorage-independent growthTumorigenic potentialAccumulated population doublingsWB controlI2.40 × 105No0/8 0%NAL1III5.80 × 105*Statistically significant differences from the WB control (n = 2).No7/7 100%23.0L2III6.30 × 105*Statistically significant differences from the WB control (n = 2).No8/8 100%19.4L3II3.40 × 105Yes0/8 0%15.6L4IIINDNo9/9 100%23.0L6III7.50 × 105*Statistically significant differences from the WB control (n = 2).No7/7 100%21.3L7I6.00 × 105*Statistically significant differences from the WB control (n = 2).No5/18 28%21.2L8IINDNo3/9 33%22.0L9II3.80 × 105*Statistically significant differences from the WB control (n = 2).Yes2/19 11%22.1L10I2.60 × 105No0/7 0%20.7L11IIINDNo2/10 20%21.7L12IV4.60 × 105*Statistically significant differences from the WB control (n = 2).Yes9/9 100%19.2L13I5.00 × 105*Statistically significant differences from the WB control (n = 2).No0/8 0%20.6L14IVNDNo0/10 0%17.4L16IIINDNo6/14 43%22.4L17IV3.20 × 105Yes0/5 0%20.4L18I3.60 × 105No0/8 0%16.2L19IVNDNo0/9 0%22.7L20III1.90 × 105No7/7 100%21.3Refer to the text for a description of each morphological pattern. Positive growth in soft agar was indicated by the formation of 0.05-mm-diameter colonies by more than 0.1% of the 1.25 × 104 cells plated under standard conditions. The number of population doublings per selection cycle = 1n[(nf)/(ni)]/1n2 where nf is the final number of cells present on the dish at the end of a selection cycle and niis the initial number of cells seeded at the beginning of the selection cycle. ND, not done.* Statistically significant differences from the WB control (n = 2). Open table in a new tab Lineages and selection cycles that were negative for tumor formation are not displayed in this table. There were no differences in tumor formation for individual lineages between male and female host rats (data not shown). Refer to the text for a description of each morphological pattern. Positive growth in soft agar was indicated by the formation of 0.05-mm-diameter colonies by more than 0.1% of the 1.25 × 104 cells plated under standard conditions. The number of population doublings per selection cycle = 1n[(nf)/(ni)]/1n2 where nf is the final number of cells present on the dish at the end of a selection cycle and niis the initial number of cells seeded at the beginning of the selection cycle. ND, not done. Four cell lineages showed no evidence of morphological transformation in culture during 10 to 12 cycles of selective growth. These lineages were indistinguishable morphologically from confluent cultures of parental WB-F344 cells (pattern I, Figure 2A). The morphologies and growth patterns of the other 14 lineages could be classified into three distinct patterns of abnormal growth (Table 2 and Figure 2) that became apparent as early as selection cycle 5 and were maintained throughout subsequent selection cycles. Three of the lineages expressed a growth pattern characterized by foci of large polygonal cells separated by cords of smaller cells with scant cytoplasm that established a boundary around the foci (pattern II, Figure 2B). In more than one-third of the cell lineages, a subpopulation of smaller densely packed cells formed a second layer on top of the attached monolayer of apparently normal cells (pattern III, Figure 2C). In four lineages, the growth pattern was characterized by the presence of free-floating aggregates of rounded cells suspended above a monolayer that resembled the parental WB-F344 cells (pattern IV, Figure 2D). In general, lineages that displayed growth pattern I were nontumorigenic, whereas lineages that displayed growth patterns II and III were tumorigenic. Lineages that displayed growth pattern IV were variably tumorigenic. The majority of the populations did not form any colonies in soft agar (Table 2). Therefore, anchorage-independent growth was not correlated with the expression of tumorigenicity by spontaneous transformants of WB-F344 cells. Final cell densities of the individual cell populations were evaluated at the end of each cycle of selection to detect the emergence of subpopulations that had acquired a growth advantage in the environment of confluent cell culture. The final cell density of 7/11 (64%) of the tumorigenic lineages fluctuated minimally through selection cycle 8 but increased dramatically between selection cycles 8 and 12, attaining cell numbers 4- to 14-fold above those observed at selection cycle 1 (Figure 3A). In contrast, 6/7 (86%) of the nontumorigenic lineages demonstrated only slight fluctuations in final cell density over the 12 cycles of selective growth (Figure 3B). Therefore, the progressive increase in the final cell density over the course of the late selection cycles was a good predictor of tumorigenic potential at selection cycle 10, and conversely, minimal variation in final cell density was associated with a lack of tumorigenic potential. Saturation densities of the individual populations at selection cycle 10 after 28 days in culture provided a measure of loss of contact inhibition of the populations after prolonged maintenance under confluent conditions. A Baysian analysis21Galen RS Statistics.in: Sonnenwirth AC Jarett L ed 8. Gradwohl's Clinical Laboratory Methods and Diagnosis. vol 1. CV Mosby, St. Louis1980: 41-68Google Scholar of saturation densities determined that a critical value of 3.6 × 105 cells/cm2 provided the best separation of tumorigenic and nontumorigenic populations at this selection cycle (correct assignments = 83%; sensitivity = 86%; specificity = 80%). Six of seven populations that attained this density were tumorigenic, whereas four of five of the populations that did not attain this density were nontumorigenic (Figure 4). Although the increases in saturation density were statistically significant, none of the populations attained densities greater than threefold that of the parental WB-F344 cells. At least two independent clonal cell lines were isolated from tumors produced by each of seven spontaneously transformed lineages of WB-F344 cells. Almost all (14/16) of the clonal tumor cell lines exhibited a loss of contact inhibition at confluence, demonstrated by significant increases in saturation densities relative to parental WB-F344 cells (Table 3). However, none of the cell lines attained densities greater than threefold that of the parental WB-F344 cells. One-half of the clonal tumor cell lines were capable of anchorage-independent growth although the average colony-forming efficiency of these lines was only 3.4% (Table 3). Despite the low capacity for anchorage-independent growth, all of the clonal tumor cell lines were tumorigenic at the intraperitoneal transplantation site (Table 3). In general, the latency of tumor formation for individual tumor cell clones derived from the same tumor was similar and was significantly shorter than that observed for the parental lineages from which they were originally derived.Table 3Phenotypic Characterization and Tumorigenicity of Transplanted Clonal Tumor Cell Lines Established form Spontaneous Transformants of WB-F344 Rat Liver Epithelial CellsTum
Referência(s)