Characterization of a newly established human pancreatic carcinoma cell line, UK Pan-1

BACKGROUND

A highly tumorigenic cell line designated as UK Pan-1 was established in a surgically removed human pancreatic adenocarcinoma and characterized as having many of the genotypic and phenotypic alterations commonly found in pancreatic tumors.

METHODS

The cell line was characterized by its morphology, growth rate in monolayer culture and soft agar, tumorigenicity in nude mice, and chromosomal analysis. Furthermore, the status of p53, Ki-ras mutation and transforming growth factor (TGF)-/receptor expression were determined. The characteristics of UK Pan-1 were compared with those of other commonly used pancreatic carcinoma cell lines.

RESULTS

Quiescent UK Pan-1 cells could be stimulated to proliferate in growth factor free nutrient media, indicating a growth factor independent phenotype. UK Pan- 1 cells grew in soft agar and rapidly formed tumors in nude mice. This cell line possesses a mutation at codon 12 of the c-Ki-ras-2 gene that is commonly found in pancreatic carcinoma. Fluorescence in situ hybridization showed that two alleles of p53 tumor suppressor gene were present in UK Pan-1. However, sequencing analysis revealed a mutation in one allele at exon 8, codon 273 (G to A; Arg to His). Additional growth assays indicated that the cell line was insensitive to negative growth regulation induced by exogenous TGF-beta. Molecular analysis of the TGF-beta signaling pathway showed that UK Pan-1 did not express appreciable levels of the TGF-beta receptor type I, II, or III mRNAs, but did express DPC4 mRNA. Karyotype analysis revealed an 18q21 deletion indicating a possible loss of heterozygosity for DPC4, as well as other chromosomal deletions and rearrangements.

CONCLUSIONS

This study indicates that UK Pan-1 is a highly tumorigenic cell line possessing a molecularly complex pattern of mutations that may be used as a model to further the understanding of the mechanisms responsible for the development of pancreatic carcinoma.

Differential expression of transforming growth factor beta receptors in human pancreatic adenocarcinoma

BACKGROUND

Many cancer cells show resistance to transforming growth factor beta (TGF-beta)-mediated growth inhibition. This resistance to TGF-beta is associated with an increased tumorigenic phenotype.

OBJECTIVE

In this study, we determined whether a loss in expression of TGF-beta receptors or DPC4, an important down stream target of TGF-beta signaling, might account for this lack of TGF-beta sensitivity in pancreatic adenocarcinoma.

MATERIALS AND METHODS

To accomplish this, six established pancreatic cancer cell lines, twenty-six surgically resected tumor specimens of pancreatic adenocarcinoma and ten non-tumor pancreas tissues were analyzed for the mRNA expression of the three TGF-beta receptors (RI, RII, and RIII) and DPC4.

RESULTS

We report here that five of six pancreatic cancer cell lines were not sensitive to TGF-beta. All the ten non-tumor specimens of pancreas showed expression of RI, and DPC4; while nine of ten showed expression of RIII and eight of ten showed expression of RII. Five of six pancreatic cancer cell lines and 23 of 26 tumor specimens showed expression of RI. Two cell lines and about half (46%) of the tumor specimens did not express RII. Only two cell lines showed appreciable levels of RIII expression; while ten of 26 (38%) tumor specimens did not show expression of RIII. DPC4 expression was observed in three of the six (50%) cell lines and 19 of 24 (79%) tumor specimens.

CONCLUSION

This study indicates that apart from the functional loss of DPC4 due to mutations or homozygous deletion, a lack of the TGF-beta receptors, particularly RII and RIII, may contribute to a loss of TGF-beta signaling in a population of pancreatic cancers.

Novel mutations in the polyadenine tract of the transforming growth factor beta type II receptor gene are found in a subpopulation of human pancreatic adenocarcinomas

In this study, we determined the incidence of microsatellite instability (MIN) in pancreatic adenocarcinoma and determined whether MIN might target, for mutations, the simple nucleotide repeats of the transforming growth factor beta type II receptor (TGFBR2) gene. Forty-eight surgically resected pancreatic tumor tissue samples and two normal pancreas tissue samples were analyzed in this study. Microsatellite analysis was performed for six loci in 14 of the 48 tumor specimens for which we had matching normal genomic DNA. Only four of the 14 tumors (29%) were MIN-positive as determined by the presence of microsatellite variations in more than one locus. Interestingly, eight of the 14 specimens (57%) showed microsatellite variations or loss of heterozygosity at D18S34, suggesting that this locus may be a critical region of genetic instability in pancreatic tumorigenesis. Of the 48 tumors, only two (4%) showed mutations in the polyA region, one of the MIN-targeted sites of the TGFBR2 gene. DNA sequence analysis of these two specimens showed the presence of a two-base deletion in one tumor specimen and the other tumor specimen showed a base substitution in the polyA tract at codon 128 of the TGFBR2 gene. The fact that these mutations occurred in the polyA tract of some pancreatic tumors suggests that a subpopulation of these tumors may be susceptible to MIN-targeted mutations. The incidence of these mutations are low and similar to that reported for nonhereditary, sporadic colon cancers.

Nucleolar p120 is expressed as a delayed early response gene and is inducible by DNA-damaging agents

Regulation of the expression of the growth-related nucleolar p120 protein was examined in serum-deprived and stimulated nontransformed and SV40-transformed WI-38 human fibroblasts. In quiescent cells, transcriptional activity of the p120 gene was very low or undetectable, and the steady-state levels of the p120 mRNA and the p120 protein were also negligible. The transient expression of the p120 gene in the cell cycle was detected in middle G1-phase after the expression of the early response genes and before the expression of the DNA-synthesis genes. Protein synthesis was required for the induction of p120 expression in serum-stimulated cells. The increased level of p120 mRNA in middle G1-phase was attributed to an increased transcription rate of the p120 gene, and not to a change in p120 mRNA stability. The calculated half-life of p120 mRNA was unchanged (1.8 +/- 0.2 hr) in all four cell conditions tested; i.e., in middle G1- or S-phase cells and in exponentially growing normal or transformed cells. Transcription rate of the p120 gene was correlated with the steady-state levels of either p120 protein or p120 mRNA. A sharp increase in p120 mRNA level occurred in both normal and transformed cells treated with actinomycin D used to examine p120 mRNA stability. This induction of p120 mRNA expression was seen in early G1-phase, but not in quiescent cells, or in middle to late G1-phase when cells expressed the highest level of p120 mRNA. The same expression pattern was seen by treatment with chlorambucil, another DNA-damaging agent. The conclusions of these studies are that the expression of p120 (1) is serum inducible in a fashion characteristic of the delayed early response gene products, (2) requires the presence of newly synthesized proteins, (3) is regulated transcriptionally, and (4) can be induced by DNA-damaging agents.

Altered transcription control is responsible for the increased level of proliferation-associated P120 in rapidly growing breast carcinoma

Transcriptional and post-transcriptional regulation of proliferation-associated nucleolar P120 protein expression was examined in 1 normal and 5 malignant breast cell lines. The 6 breast cell lines could be placed into 3 categories on the basis of in vitro growth rate. BT549 and HBL100 grew rapidly; MCF-7/6, MCF-7/AZ and Hs578T grew at a moderate rate; and Hs578N normal epithelia grew slowly. There was a significant correlation between the growth rate, measured by percentage of S-phase fraction of cells or doubling time of cell numbers and the steady-state levels of either P120 protein or P120 mRNA. Next, the mechanisms responsible for the increased level of P120 expression, which is associated with rapidly growing breast carcinoma, were examined. The stability of P120 mRNA was measured by densitometry of quantitative Northern blots of mRNAs from cells treated with actinomycin D. Before the expected decay, a sharp increase in P120 mRNA level was detected shortly after inhibiting the overall RNA or protein synthesis. The calculated half-life of P120 mRNA was very similar (1.8 +/- 0.2 hr) in all 6 cell lines examined. The transcription rate of the P120 gene was determined by densitometric analysis of quantitative nuclear run-off assays. A significant positive correlation was found between the transcription rate of the P120 gene and the steady-state levels of either P120 protein or P120 mRNA. Our conclusion is that the expression of P120 protein is transcriptionally regulated in these breast cells. Therefore, the characteristically high level of P120 protein and mRNA found in most breast carcinomas is due to the altered transcription rate of the P120 gene in transformed cells.

Effect of nucleolar P120 expression level on the proliferation capacity of breast cancer cells

Steady-state level of nucleolar P120 protein and P120 mRNA was compared to the doubling time and S-phase fraction in human breast cancer cell lines growing exponentially and in similar cells treated with a single dose of P120 antisense oligodeoxynucleotides. The study included six breast cancer cell lines and one nontransformed breast cell line with doubling times from 1.1 to 5.5 days and with S-phase fractions from 35 to 9%. P120 expression level was determined by densitometric computerized evaluation of protein and mRNA blots and with a quantitative 32P-reverse transcriptase-polymerase chain reaction method developed for small-scale samples. In the slowest growing normal cell line, P120 expression level was only about 10% of the level found in the most rapidly growing cancer cell line. The amount of P120 mRNA was highly correlated with the amount of P120 protein (P = 0.0001), indicating that P120 accumulation is regulated in these cells primarily at a transcriptional level. There was also a significant positive correlation between the level of P120 protein/mRNA and doubling time of cell lines (P = 0.0008) or percentage of S-phase cells (P = 0.210). P120 antisense oligomer treatment decreased the growth rate of cells in a dose-dependent manner, and the inhibition reached 70% at 100 microM concentration. Both P120 mRNA and P120 protein levels were also decreased by approximately 70% in cells treated with 100 microM P120 antisense oligomer. Slowly growing cells exhibited 50% inhibition by treatment at a proportionally lower concentration of P120 antisense oligomer than fast growing cells. This study shows that the expression of P120, measured either at the protein or the mRNA level, correlates with proliferation rate, identifying P120 as a cell proliferation marker.

Cell cycle regulated expression of nucleolar antigen P120 in normal and transformed human fibroblasts

Normal and SV40 virus-transformed WI-38 human lung fibroblasts were serum starved and refed, or synchronized by double thymidine block and released from the block. At different time points in the cell cycle, steady state levels of P120 mRNA and P120 protein content of the cells were determined by densitometric scans of Northern and Western blots. At the same time points, [3H]thymidine uptake was measured and flow cytometric analysis performed for DNA content and P120 antigen staining. Levels of P120 protein and P120 mRNA were approximately 4 times greater in non-synchronous, exponentially growing transformed cells than in similarly growing normal cells. Early G1-phase cells, synchronized either with serum deprivation or with metabolic block, contained only a trace amount of P120 protein and mRNA. The P120 gene was transcribed early in G1 and P120 protein synthesis initiated in middle G1. A dramatic increase of P120 protein level occurred in S-phase with a corresponding mRNA peak preceding the P120 protein peak. These results indicate that P120 is overexpressed in transformed WI-38 cells and that P120 is temporally regulated during the cell cycle of both transformed and normal fibroblasts. The dramatic increase in P120 protein expression at the G1 to S boundary suggests that P120 may play a role in the regulation of cell cycle and increased nucleolar activity that is associated with cell proliferation.

Antisense-mediated specific inhibition of P120 protein expression prevents G1- to S-phase transition

A pentadecadeoxyribonucleotide (5'-AAAGCCCCCCACCAC), complementary to a splice junction site of mRNA for human proliferation-associated nucleolar protein P120, inhibited expression of the P120 gene and the mitogen-induced proliferation of human lymphocytes. The inhibition of P120 gene expression and proliferation was concentration dependent and reached 90% at 200 microM, as measured by [3H]thymidine uptake and by densitometric scanning of Northern (mRNA) and Western (protein) blots of P120. Inhibition was not observed in cells treated with the correspondent nonsense oligomer. P120 antisense oligomer treatment prevented S-phase entry of mitogen-stimulated lymphocytes, as determined by flow cytometric analysis, but did not block G0-G1 transition assessed by morphological blast transformation and induction of [3H]uridine incorporation. Results of this study suggest that P120 expression may be required for the upregulation of nucleolar function necessary for cell proliferation.