Accelerated Progression of Asbestos-Induced Mesotheliomas in Heterozygous p53+/- Mice

Charles A. Vaslet, Norma J. Messier and Agnes B. Kane,1

Department of Pathology and Laboratory Medicine, Brown University, Providence, Rhode Island 02912

Received January 23, 2002; accepted April 19, 2002


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Asbestos fibers produce diffuse malignant mesotheliomas in chronic rodent inhalation assays or after direct intrapleural or intraperitoneal injection. In vitro models have provided evidence that asbestos fibers are genotoxic carcinogens that can directly or indirectly generate reactive oxygen- and nitrogen-derived species that cause DNA damage. Heterozygous p53+/- mice show an increased incidence and reduced latency of malignant mesotheliomas induced by weekly intraperitoneal injections of crocidolite asbestos fibers. In this study, we investigated whether loss of heterozygosity (LOH) at the p53 tumor-suppressor gene locus contributes to accelerated tumor progression. LOH was found in 50% of the tumors produced in heterozygous p53+/- mice. In contrast to tumors that arise in p53+/+ mice or those that retained one p53 allele, LOH was associated with large tumor masses with central areas of necrosis, local invasion, and penetration of lymphatics. Increased tumor size was not associated with increased levels of cell proliferation as determined by BrdU incorporation, but it was correlated with a reduction in apoptosis as determined morphologically and by the TUNEL assay. Wild-type p53 protein is essential for cell cycle arrest in response to DNA damage and in maintenance of genomic stability. Cell lines established from tumors that showed LOH at the p53 tumor-suppressor gene locus were nearly tetraploid. These results suggest that p53 haplo-insufficiency sensitizes mice to the clastogenic or aneuploidogenic effects of crocidolite asbestos fibers, resulting in a shorter latent period. As solid tumors develop, spontaneous loss of the wild-type allele accompanied by decreased apoptosis and genetic instability is associated with accelerated tumor growth, invasion, and lymphatic dissemination.

Key Words: asbestos fibers; mesothelioma; p53-deficient mice; apoptosis; tumor progression; crocidolite asbestos; genetic instability.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Genetically engineered mice carrying activated oncogenes or inactivated tumor- suppressor genes have been used to assess the role of specific genetic alterations involved in the initiation, promotion, and progression of cancer (Macleod and Jacks, 1999Go). Transgenic mice carrying one copy of a disrupted allele at the p53 tumor suppressor gene locus (p53+/- heterozyotes) have been proposed as a replacement for the two-year rodent carcinogenesis assay (Tennant et al., 1995Go). Heterozygous p53+/- mice develop spontaneous lymphomas, sarcomas, and adenocarcinomas, depending on the genetic background, after a latency of 10 months (Harvey et al., 1993aGo). In contrast, after exposure to genotoxic agents such as ionizing radiation, these heterozygous mice rapidly develop tumors after exposures as short as 26 weeks (Harvey et al., 1993bGo; Lee et al., 1994Go). The tissue specificity of genotoxic carcinogens is similar in both wild-type and p53+/- mice (French et al., 2001Go). These heterozygous p53-deficient mice are useful for human risk assessment, taking into account interindividual variations in genetic sensitivity to carcinogens (Preston, 1996Go).

Asbestos fibers induce malignant mesotheliomas in chronic rodent-inhalation assays, as well as after direct intrapleural or intraperitoneal injection (reviewed in Kane, 1996Go). In vitro models have provided evidence that asbestos fibers have properties of genotoxic as well as nongenotoxic carcinogens. The geometry of long, thin fibers has been associated with genotoxicity in target cells in vitro. The surface reactivity of asbestos fibers and their ability to catalyze iron-dependent generation of oxygen- and nitrogen-derived species has been linked to production of oxidized bases, DNA breaks, mutations, and deletions (reviewed in Kane, 1996Go). Asbestos fibers have also been shown to be mitogenic, both by binding to specific surface receptors and by activating intracellular signaling pathways (reviewed by Robledo and Mossman, 1999Go). Asbestos fibers also cause apoptosis that may trigger compensatory hyperplasia (Broaddus, 1997Go). Activation of signal transduction pathways and induction of apoptosis by asbestos fibers in pleural mesothelial cells are dependent on the generation of reactive oxygen species (Robledo and Mossman, 1999Go).

The p53 protein plays a major role in coordinating cell-cycle arrest and DNA repair or apoptosis induced by oxidants or ionizing radiation. Cells that lack a functional p53 protein as a result of gene deletion, point mutation, or complex formation with viral proteins are defective in cell-cycle arrest and resistant to induction of apoptosis by DNA-damaging agents (Ko and Prives, 1996Go). We have shown that a murine mesothelial cell line that spontaneously acquired a point mutation in the p53 gene during passaging in vitro is defective in the G1 cell cycle checkpoint induced by ionizing radiation. This cell line also shows increased sensitivity to induction of micronuclei by direct exposure to crocidolite asbestos fibers in vitro (Cistulli et al., 1996Go). Therefore, we hypothesized that p53-deficient mice would show increased sensitivity to the genotoxic and carcinogenic effects of crocidolite asbestos fibers in vivo. We have reported previously that heterozygous p53+/- mice show an increased incidence and reduced latency of malignant mesotheliomas induced by crocidolite asbestos fibers (Marsella et al., 1997Go). In this study, we report that 50% of the tumors induced in p53+/- mice show loss of heterozygosity (LOH) at the p53 tumor suppressor gene locus, increased tumor size, and extensive invasion. The accelerated progression of malignant mesothelioma in heterozygous p53+/- mice could be explained by three mechanisms. First, p53-deficient mesothelioma cells could have a selective growth advantage over wild-type or heterozygous p53+/- mesothelioma cells. Second, p53-deficient mesothelial cells may acquire increased genetic instability secondary to loss of cell-cycle checkpoints and accumulation of oxidant-induced DNA damage. Third, p53-deficient mesothelial cells may be resistant to apoptosis. Evidence for or against each of these three mechanisms was evaluated in vitro and in vivo.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals.
Homozygous p53-/-, heterozygous p53+/-, and wild-type p53+/+ male mice derived from the 129/Sv strain on a 75% C57B1/6 background were purchased from Jackson Laboratories (Bar Harbor, ME). These transgenic mice carry a targeted neo insert replacing exons 2–6 of the p53 tumor suppressor gene (Livingston et al., 1992Go). Mesotheliomas were induced by weekly intraperitoneal injections of 200 µg of UICC crocidolite asbestos fibers as described previously (Goodglick et al., 1997Go). Fiber suspensions prepared in phosphate-buffered saline (PBS) were handled under a laminar flow hood equipped with Hepa filters with external exhaust. Mice were housed and treated according to guidelines established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Histopathology.
Complete necropsies were performed on all mice and histopathologic diagnosis of malignant mesotheliomas was carried out as described previously (Goodglick et al., 1997Go). For analysis of cell proliferation, mice were injected with BrdC (1 mg/1ml PBS, ip), which is more soluble than BrdU, one h before sacrifice. BrdC is converted to BrdU in vivo. Paraffin-embedded sections of tumors were processed for immunohistochemistry as described previously (Macdonald and Kane, 1997Go). The thickness of tumors growing on the serosal surfaces was measured at a magnification of 100x using an eyepiece equipped with a micrometer. An in situ assay was developed to visualize micronuclei in mesothelial cells using whole mounts of the diaphragm. In order to distinguish between micronuclei associated with mesothelial cells and debris from inflammatory cells, proliferating mesothelial cells were labeled with BrdU as described above. After 14 h (the cell cycle time of mesothelial cells is 18–24 h), mice were sacrificed and whole mounts of the diaphragm were prepared for BrdU immunohistochemistry. Micronuclei were counted at 40x magnification; at least 1000 proliferating mesothelial cells were counted per diaphragm. Micronuclei were counted if they were in the same focal plane as the nucleus and less than 25% of the diameter of the parent nucleus.

Isolation of mesothelioma cell lines.
Cell lines were isolated from heterozygous p53+/- or wild-type mice after development of ascites or weight loss following weekly intraperitoneal injections of crocidolite asbestos fibers, as described previously (Goodglick et al., 1997Go). Cell lines were established from ascites or peritoneal lavage fluid and analyzed after 2–6 passages in vitro.

Tissue microdissection and DNA extraction.
Fixed, paraffin-embedded tumors were cut into 15 µm sections, mounted on glass microscope slides, and dissected with a #11 surgical blade under an inverted phase-contrast microscope at 10x magnification. Tissue samples were deparaffinized in xylene, followed by 100% ethanol, air-dried, and digested in 100 mM NaCl, 10 mM Tris, pH 8.0, 25 mM EDTA, 0.5% SDS, and proteinase K at 100 µg/ml for 4 h at 370C. DNA was extracted in phenol/chloroform/isoamyl alcohol (25:24:1). The aqueous phase was collected and the salt concentration adjusted to 250 mM NaCl. DNA was ethanol-precipitated overnight at –200C, washed, resuspended in water, and stored at –800C.

PCR and RT-PCR analyses.
Genomic DNA was isolated from microdissected tumors, tail biopsies, or cultured mesothelioma cell lines, as described above. For PCR amplification, 200 ng of genomic DNA was amplified using the following primers: X7: 5`-tatactcagagccggcct-3` and X6.5: 5`-acagcgtggtggtaccttat-3` to amplify the wild-type p53 gene and X7 and neo 19: 5`-ctatcaggdcatagcgttgg-3` to amplify the disrupted p53 allele. For microdissected tumors, only the wild-type p53 allele was amplified, due to limited amounts of DNA. Sufficient tumor samples free of stromal cells were available from only 3 mice. PCR amplification was performed at 950C for 5 min.; 950C, 1 min; 600C, 2 min; and 720C, 3 min for 30–35 cycles, followed by 720C for 10 min. PCR products were separated on a 3% agarose (TAE) gel. Total RNA was isolated from subconfluent cultures of mesothelioma cell lines using RNazolTM, and used for RT-PCR analysis as described previously (Goodglick et al., 1997Go). The following primers were used for p53 RT-PCR: 5`-acaggaccctgtcaccgagacc-3` and 5`-gacctccgtcatgtgctgtgac-3`. Oligonucleotide primers for murine ß-actin were purchased from Clontech Laboratories (Palo Alto, CA).

p53 Genomic DNA sequencing.
Genomic DNA, obtained from early passages of mesothelial cell lines isolated from one wild-type and 4 p53+/- mice that retained the wild-type allele, was PCR-amplified for exons 5–8 of the p53 gene, using Platinum Taq DNA Polymerase High Fidelity (Invitrogen, Carlsbad, CA) and primers derived from intron 4 (forward) and exon 11 (reverse). These products were cloned into a pCRII-Topo vector (Invitrogen) and cycle sequenced using p53 intron-specific forward and reverse primers.

Metaphase chromosome spreads.
Mesothelioma cell lines were maintained for 2–6 passages in vitro and exposed to colcemid (0.2 µg/ml for 1.5 h.). Metaphase spreads were prepared as described previously (Marsella et al., 1997Go); at least 20 metaphase spreads were counted to determine chromosome numbers in each cell line.

Apoptosis assays.
Tumor samples from heterozygous p53+/- mice were fixed in Omnifix, embedded in plastic, sectioned at 1 µm, and stained with toluidine blue for histologic analysis of necrosis or apoptosis. Paraffin-embedded tumor tissues were sectioned and analyzed for incorporation of biotinylated dUTP (TUNEL assay kit, Oncor, Inc., Gaithersburg, MD). Apoptotic cells were identified using avidin-peroxidase immunohistochemistry as described previously (Marsella et al., 1997Go).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Increased Sensitivity of p53-Deficient Mice to Induction of Micronuclei by Crocidolite Asbestos
An in situ assay was developed to visualize micronuclei in mesothelial cells using whole mounts of the diaphragm, as illustrated in Figure 1Go. Induction of micronuclei was assayed three days after a single intraperitoneal injection of 200 µg of crocidolite asbestos fibers; this is the peak of mesothelial cell proliferation induced in this murine model (Moalli et al., 1987Go). As summarized in Table 1Go, an increased number of micronuclei was induced in p53-/- mice three days after injection of asbestos fibers, although no increase in micronuclei was observed in saline-injected controls. Wollastonite is a soluble calcium silicate that is not tumorigenic. Injection of 2 mg of wollastonite fibers elicited a comparable inflammatory response as asbestos (Macdonald and Kane, 1997Go); however, the number of micronuclei induced was not statistically elevated over saline-injected controls. The magnitude of the inflammatory response elicited by intraperitoneal injection of asbestos was similar in wild-type and p53-deficient mice (data not shown). Although p53-deficient mice are susceptible to the development of thymomas and sarcomas, they do not develop spontaneous malignant mesotheliomas (Donehower et al., 1992Go). However, because p53-deficient mice showed increased sensitivity to asbestos-induced genotoxicity, we hypothesized that they would be more sensitive to induction of mesotheliomas by weekly intraperitoneal injections of 200 µg of crocidolite asbestos fibers.



View larger version (167K):
[in this window]
[in a new window]
 
FIG. 1. Induction of a micronucleus in a proliferating mesothelial cell identified by BrdU immunohistochemistry on a whole mount of the diaphragm, as described in Materials and Methods; magnification x400.

 

View this table:
[in this window]
[in a new window]
 
TABLE I Induction of Micronuclei in Proliferating Mesothelial Cells in Vivo
 
Malignant Mesotheliomas Induced by Crocidolite Asbestos in Heterozygous p53-Deficient Mice
The time course of induction of malignant mesotheliomas by weekly injections of crocidolite asbestos fibers is shown in Figure 2Go. After a mean latent period of 44 weeks, 76% of the heterozygous p53-deficient mice developed mesotheliomas. In contrast, only 32% of the wild-type mice developed mesotheliomas after a mean latent period of 67 weeks. The incidence of tumors in p53+/- vs. p53+/+ mice was dependent on the genotype (p = 0.003 using Fisher’s exact test). The difference between the mean latent periods was also statistically significant (p = 0.002 using analysis of variance and Fisher’s PLSD test).



View larger version (17K):
[in this window]
[in a new window]
 
FIG. 2. Time of development of malignant mesotheliomas in heterozygous p53+/- or wild-type p53+/+ mice induced by weekly ip injections of crocidolite asbestos fibers (200 µg/1 ml PBS). Mice were sacrificed when they developed massive ascites or weight loss; all abdominal organs were examined histopathologically for the presence of diffuse malignant mesothelioma. These data were originally presented in tabular form in Marsella et al.(1997).

 
Distinct morphologic stages in tumor progression have been described in this murine model (Craighead and Kane, 1994Go). Focal, well-localized tumors limited to the serosal surfaces were the earliest mesotheliomas observed in p53+/+ or +/- mice. These tumors spread diffusely over the peritoneal lining or grew as papillary projections within the abdominal cavity. Invasion was limited to the injection site or mesenteries. In 50% of the p53+/- mice, malignant mesotheliomas with extensive invasion were found (Fig. 3Go). Tumors in these mice were larger than those in p53+/+ mice: the tumor thickness on the serosal surfaces ranged between 0.40–0.76 mm (n = 4 mice) compared to 0.20–0.36 mm (n = 4 mice) in the other p53+/- mice and 0.05–0.25 mm (n = 5 mice) in the p53+/+ mice. Lymphatic invasion and penetration through the diaphragm to involve the pleura was also found (Fig. 3Go, lower panel), in addition to metastasis to abdominal lymph nodes. Invasion of the diaphragm and lymphatic metastases have rarely been observed in rodent mesotheliomas induced by intraperitoneal injection of asbestos fibers (Craighead and Kane, 1994Go).



View larger version (133K):
[in this window]
[in a new window]
 
FIG. 3. Histologic appearance of diffuse malignant mesotheliomas in heterozygous p53+/- mice that lost the wild-type allele. Upper panel, solid tumors on the abdominal wall showing central areas of necrosis. Lower panel, infiltration of the diaphragm and penetration into lymphatics; magnification x100. Hematoxylin and eosin stain.

 
Malignant Mesotheliomas Induced by Asbestos in p53+/- Mice Show Loss of the Wild-Type Allele
The accelerated rate of growth and invasion in 50% of mesotheliomas arising in p53+/- mice raises the possibility that the wild-type allele was lost in these tumors. This was investigated using mesothelial cell lines established from these mice and microdissected samples of malignant mesotheliomas identified histologically on serial paraffin-embedded sections (Fig. 4Go, upper and center panels). The genotypes of the donor mouse and the tumor were compared using PCR analysis of DNA extracted from the tail or from early passages of the corresponding cell line. As shown in Figure 4Go (upper panel), primer sets designed to amplify exons 6–7 of the p53 gene generated the expected product at 400 bp in p53+/+ or +/- mice, but not in p53-/- mice. In addition, a second product at 600 bp was detected in p53+/- and -/- mice using primers that amplified the neo-insert in the disrupted p53 allele. Fifty percent of the tumor-cell lines established from the donor p53+/- mice had lost the wild-type p53 allele. This PCR analysis of DNA isolated from cell lines established from p53-deficient mice with asbestos-induced mesotheliomas was confirmed by RT-PCR analysis for p53 expression (Fig. 4Go lower panel). RNA was isolated from early passages of these cell lines, treated with DNase to eliminate priming from the p53 pseudogene, reversed-transcribed, and amplified with primers to detect expression of p53 or actin as a control for integrity of the isolated RNA. All RNA samples showed amplification of actin and 80% showed expression of WT1, which is characteristic of murine malignant mesotheliomas (Goodglick et al., 1977). Expression of p53 corresponded with PCR analysis of DNA extracted from these cell lines in all cases.



View larger version (32K):
[in this window]
[in a new window]
 
FIG. 4. Inactivation of p53 in mesothelioma cell lines isolated from heterozygous p53+/- mice after weekly ip injections of crocidolite asbestos fibers, as described in Materials and Methods. Upper panel, genotypes of mice and tumor cell lines using PCR analysis of DNA; center panel, genotypes of tumors microdissected from wild-type (276) and p53+/- mice (221, tumors 1 and 2 and 236); lower panel, mRNA expression of p53 and actin using RT-PCR analysis of RNA.

 
PCR analysis of mesothelioma cell lines was confirmed by direct genotyping of the primary tumors (Fig. 4Go, center panel). Sections of mesotheliomas free of contaminating stroma were microdissected, deparaffinized, and digested overnight. Phenol-extracted DNA was used to determine the p53 genotype using PCR analysis. A tumor sample microdissected from a p53+/+ mouse showed retention of the wild-type allele. In contrast, 3 tumor samples isolated from 2 p53+/- mice showed loss of the wild-type allele. Mesothelioma cell lines isolated from these mice corresponded to those that showed loss of the wild-type allele using PCR or RT-PCR analysis as described above.

Four of the 5 cell lines derived from p53+/- mice that retained the wild-type allele were analyzed for mutations in exons 5–8 of the p53 gene. No missense mutations or deletions were detected.

Growth of p53-Deficient and Wild-Type Mesothelioma Cells in Vivo
If p53-mutant- or -deficient cells have a selective growth advantage in vivo, we would predict increased mesothelial-cell proliferation in tumors induced by asbestos in p53-deficient mice. Mesothelial-cell proliferation was assessed by BrdU labeling using immunohistochemistry. Sections of mesotheliomas were studied from p53+/+ and +/- mice, including tumors that had lost the wild-type allele. The BrdU labeling index ranged from 6–9%, with no statistically significant difference between p53+/+, +/-, or -/- tumors.

Genetic Instability of Mesothelioma Cell Lines Derived from p53-Deficient Mice
Tumors that develop in p53-/- mice are characterized by aneuploid and subtetraploid populations and abnormal karyotypes (Donehower et al., 1995Go). Therefore, we hypothesized that malignant mesotheliomas induced by asbestos in p53-deficient mice would show genetic instability. Cell lines were established from ascites or peritoneal lavage fluid collected from mice as described in Cistulli et al.(1996). Metaphase spreads were prepared from cells at passages 2–6. Normal p53+/+ mesothelial cells are nearly diploid (2n = 40). Cell lines derived from all of the p53+/- mice that had lost the wild-type allele were tetraploid (mean chromosome number = 81 ± 3.6). This observation is consistent with a p53-mediated spindle checkpoint (Cross et al., 1995Go). In contrast, cell lines derived from p53+/+ and p53+/- mice that retained the wild-type allele were aneuploid (mean chromosome number = 55 ± 11). The difference between p53+/+ and p53-/- mesothelioma cell lines is statistically significant (p = 0.01). The p53-/- mesotheliomas showed nuclear pleomorphism with binucleate and multinucleated tumor cells and abnormal mitotic figures (Fig. 5Go).



View larger version (132K):
[in this window]
[in a new window]
 
FIG. 5. Histologic section of a diffuse malignant mesothelioma induced in a heterozygous p53+/- mouse that had lost the wild-type allele. Anaplastic tumor cells with multiple nuclei and abnormal metaphases are visible. Original magnification x400. Hematoxylin and eosin staining.

 
Apoptosis in p53+/+ and p53-/- Mesotheliomas
Loss of p53 function has been hypothesized to contribute to aggressive growth during later stages of tumor progression in choroid plexus tumors (Symonds et al., 1994Go) and islet cell carcinomas (Naik et al., 1996Go) in p53-deficient mice. Therefore, we hypothesized that loss of the wild-type allele in heterozygous p53-deficient mice would be associated with decreased apoptosis. The extent of apoptosis in malignant mesotheliomas induced by asbestos in p53-deficient mice was evaluated by the TUNEL assay and nuclear morphology. As illustrated in Figure 6Go, tumors in p53+/+ mice showed isolated cells with crescentic margination of chromatin and numerous apoptotic bodies while tumors that arose in p53+/- mice that showed loss of the wild-type allele had confluent areas of central necrosis. These morphologic changes were consistent with the percentage of cells in apoptosis as detected by the TUNEL assay performed on paraffin-embedded tumors as illustrated in Fig. 7Go: 12 ± 2% of the tumor cells in p53+/- mice that retained the wild-type allele or p53+/+ mice showed apoptosis compared with 1.7±1% of the tumors in p53+/- mice that lost the wild-type allele. This difference is statistically significant (p < 0.05).



View larger version (181K):
[in this window]
[in a new window]
 
FIG. 6. Morphologic evidence of apoptosis or necrosis in diffuse malignant mesotheliomas induced by weekly ip injections of crocidolite asbestos fibers. Upper panel, tumor from a wild-type p53+/+ mouse showing isolated condensed nuclei with fragmentation and margination of chromatin. Lower panel, tumor from a heterozygous p53+/- mouse that had lost the wild-type allele showing confluent areas of necrosis with loss of cell outlines, cytoplasmic swelling, and nuclear debris. Original magnification x400. One µm plastic sections, toluidine blue stain.

 


View larger version (106K):
[in this window]
[in a new window]
 
FIG. 7. Spontaneous apoptosis in a diffuse malignant mesothelioma induced in a wild-type p53+/+ mouse by weekly ip injections of crocidolite asbestos fibers. Apoptotic cells as detected by TUNEL assay show dark staining of the nuclei. Some of the nuclei are pyknotic or show margination of the chromatin; original magnification x400. DAB immunohistochemical reaction as described in Materials and Methods.

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We have previously reported that the incidence of mesotheliomas is statistically increased in p53+/- mice in comparison to wild-type mice with a reduced latent period (Marsella et al., 1997Go). This chronic carcinogenicity assay in heterozygous p53-deficient mice supports the hypothesis that crocidolite asbestos fibers act as a genotoxic carcinogen in this model system. Since crocidolite asbestos fibers produce both persistent mesothelial cell proliferation and chronic inflammation, it is impossible to distinguish between a direct clastogenic or aneuploidogenic effect of asbestos fibers and an indirect genotoxic effect secondary to chronic inflammation characterized by activated macrophages (Macdonald and Kane, 1997Go). The simplest explanation for the increased sensitivity of heterozygous p53-deficient mice to asbestos-induced carcinogenicity is loss of cell cycle checkpoints, resulting in accumulation of oxidant-induced DNA damage and genetic instability (Marsella et al., 1997Go). However, the prolonged latent period before development of malignant mesotheliomas, even in heterozygous p53-deficient mice, suggests that additional mutational events are required for the development of this tumor (Macleod and Jacks, 1999Go).

The most striking pathologic finding in this chronic carcinogenicity study in heterozygous p53-deficient mice was accelerated tumor progression. Half of these mice had large tumor masses with central areas of necrosis, extensive local invasion, and penetration of lymphatics; these features are rarely seen in wild-type rodent tumors produced by direct intraperitoneal injection of asbestos fibers (Craighead and Kane, 1994Go). Cell lines were successfully established from 8 of the heterozygous p53+/- mice; 50% of these cell lines showed loss of p53 expression associated with loss of the wild-type allele. LOH at the p53 tumor suppressor gene locus occurs in approximately 50% of the tumors that arise in Li-Fraumeni families, as well as in lymphomas induced by ionizing radiation (Kemp et al., 1994Go) or skin tumors induced by DMBA (Kemp et al., 1993Go). However, not all tumors induced by carcinogens in heterozygous p53-deficient mice develop LOH or other mutations at the p53 tumor suppressor gene locus. For example, neither mammary tumors induced by DMBA (Jones et al., 1997Go) nor lymphomas induced by uv irradiation of the skin (Jiang et al., 2001Go) have any apparent mutations in the remaining wild-type p53 allele. Only 50% of the mesotheliomas induced by asbestos fibers had LOH; therefore, LOH is not essential for the development of mesotheliomas. However, LOH does appear to be associated with accelerated progression of these tumors.

Potential mechanisms responsible for accelerated tumor progression in p53-deficient mice have been explored in other model systems. Increased tumor size could be due to increased cell proliferation, reduced apoptosis, or both. In a mouse mammary tumor model (Wnt1-1TG mice crossed with p53+/+, +/-, or -/- mice), increased tumor size occurred in the Wnt-1TG p53+/- mice and was accompanied by loss of the wild-type allele in 50% of the tumors. These tumors had an increased mitotic index but no change in the level of apoptosis in comparison with the Wnt-1 TG p53+/+ or Wnt-1 TG p53-/- tumors (Jones et al., 1997Go). Although we were unable to measure tumor size directly in this intraperitoneal model because malignant mesotheliomas spread diffusely over the peritoneal surfaces (Craighead and Kane, 1994Go), larger tumor masses and increased tumor thickness were seen in half of the heterozygous p53+/- mice. In contrast to the mouse mammary tumor model, no statistically significant differences in the BrdU labeling index were observed in p53+/+, p53+/-, or p53+/- tumors that had LOH. Instead, a statistically significant decrease in the apoptotic index was observed in p53+/- tumors that had lost the wild-type allele. This observation confirms previous studies in p53-deficient mice; for example, increased growth of choroid plexus tumors in p53-/- transgenic mice expressing a truncated SV40 large-T antigen has been attributed to a reduction in the apoptotic index (Symonds et al., 1994Go).

In addition to induction of apoptosis, wild-type p53 protein is essential for cell-cycle arrest triggered by DNA damage and maintenance of genomic stability. Induction of the G1 cell cycle checkpoint depends on post-translational modification, stabilization, and translocation of the p53 protein to the nucleus where it activates transcription in conjunction with other co-activators (Prives and Hall, 1999Go). Activated transcription of p21 inhibits cyclin-dependent kinases resulting in hypophosphorylated Rb and G1 arrest. Other transcriptional targets of p53 include mdm-2, which binds to p53 itself in an autoregulatory fashion, and GADD45, which interacts with PCNA and prevents entry into S phase (Ko and Prives, 1996Go). In addition to the G1 cell cycle checkpoint, p53 also regulates a G2 checkpoint, duplication of centrosomes, and a spindle checkpoint during mitosis (Fukasawa et al., 1996Go). Therefore, inactivation of p53 function in response to DNA damage would result in unrepaired DNA breaks leading to aneuploidy and multiple rounds of DNA synthesis in the absence of chromosome segregation, resulting in tetraploidy (Cross et al., 1995Go; Ko and Prives, 1996Go).

Histologic examination of mesotheliomas in heterozygous p53-deficient mice that had lost the wild-type allele showed multinucleated tumor cells and abnormal mitotic figures. Mesothelioma cell lines isolated from these mice showed a subteraploid or tetraploid number of chromosomes. Tetraploid or nearly tetraploid tumor cell populations have been reported previously in mammary tumors induced in transgenic Wnt-1 TG p53-/- mice; these tumor cell populations also had evidence of genetic instability as revealed by karyotyping and comparative genomic hybridization (Donehower et al., 1995Go).

It is hypothesized that p53 haplo-insufficiency sensitizes mice to the clastogenic or aneuploidogenic effects of crocidolite asbestos fibers, resulting in a shorter latent period. In solid tumors, spontaneous loss of the wild-type allele, accompanied by decreased apoptosis and accumulation of additional mutations, accelerates tumor growth, invasion, and lymphatic dissemination, as suggested by Macleod and Jacks (1999).

This genetically engineered murine model is highly relevant to the pathogenesis of human malignant mesotheliomas. Although p53 mutations, rearrangements, or deletions are rare in either human or rodent malignant mesotheliomas (Lechner et al., 1997Go; Murthy and Testa, 1999Go), some investigators have demonstrated SV40 T-antigen and viral sequences in up to 80% of human mesothelioma tissues examined (reviewed in Carbone et al., 1999Go). Antisense constructs directed against large T-antigen have been shown to inhibit proliferation of human mesothelioma cell lines (Waheed et al., 1999Go). Although these findings are still controversial (Pipas and Levine, 2001Go), this murine model demonstrates a potential role for inactivation of p53 function in the development of asbestos-induced malignant mesotheliomas. This murine model provides the opportunity for additional mechanistic studies to investigate whether SV40 virus and asbestos fibers are cocarcinogens.


    ACKNOWLEDGMENTS
 
We acknowledge the excellent secretarial assistance of Marilyn May. This research was supported by NIH Grant R01 ES03721 from the National Institute of Environmental Health Sciences.


    NOTES
 
1 To whom correspondence should be addressed. Fax: (401) 863-9008. E-mail: agnes_kane{at}brown.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Broaddus, V. C. (1997). Asbestos, the mesothelial cell and malignancy: A matter of life or death. Am. J. Respir. Cell. Mol. Biol. 17, 657–659.[Free Full Text]

Carbone, M., Fisher, S., Powers, A., Pass, H. I., and Rizzo, P. (1999). New molecular and epidemiological issues in mesothelioma: Role of SV40. J. Cell Physiol. 180, 167–172.[ISI][Medline]

Cistulli, C. A., Sorger, T., Marsella, J. M., Vaslet, C. A., and Kane, A. B. (1996). Spontaneous p53 mutation in murine mesothelial cells: Increased sensitivity to DNA damage induced by asbestos and ionizing radiation. Toxicol. Appl. Pharmacol.141, 264–271.[ISI][Medline]

Craighead, J. E., and Kane, A. B. (1994). The pathogenesis of malignant and nonmalignant serosal lesions in body cavities consequent to asbestos exposure. In The Mesothelial Cell and Mesothelioma (M.-C. Jaurand and J. Bignon, Eds.), pp. 79–102. Marcel Dekker, New York.

Cross, S. M., Sanchez, C. A., Morgan, C. A., Schimke, M. K., Ramel, S., Idzerda, R. L., Raskind, W. H., and Reid, B. J. (1995). A p53-dependent mouse spindle checkpoint. Science267, 1353–1356.[ISI][Medline]

Donehower, L. A., Godley, L. A., Aldaz, C. M., Pyle, R., Shi, Y.-P., Pinkel, D., Gray, J., Bradley, A., Medina, D., and Varmus, H. E. (1995). Deficiency of p53 accelerates mammary tumorigenesis in Wnt-1 transgenic mice and promotes chromosomal instability. Genes Dev.9, 882–895.[Abstract]

Donehower, L. A., Harvey, M., Slagle, B. L., McArthur, M. J., Montgomery, C. A. Jr., Butel, J. S., and Bradley, A. (1992). Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature356, 215–221.[ISI][Medline]

French, J. E., Lacks, G. D., Trempus, C., Dunnick, J. K., Foley, J., Mahler, J., Tice, R. R., and Tennant, R. W. (2001). Loss of heterozygosity frequency at the Trp53 locus in p53-deficient (+/-) mouse tumors is carcinogen- and tissue-dependent. Carcinogenesis 22, 99–106.[Abstract/Free Full Text]

Fukasawa, K., Choi, T., Kuriyama, R., Rulong, S., and Vande Woude, G. F. (1996). Abnormal centrosome amplification in the absence of p53. Science271, 1744–1747.[Abstract]

Goodglick, L. A., Vaslet, C. A., Messier, N. J., and Kane, A. B. (1997). Growth factor responses and protooncogene expression of murine mesothelial cell lines derived from asbestos-induced mesotheliomas. Toxicol. Pathol. 25, 565–573.[ISI][Medline]

Harvey, M., McArthur, M. J., Montgomery, C. A., Jr., Bradley, A., and Donehower, L. A. (1993a). Genetic background alters the spectrum of tumors that develop in p53-deficient mice. FASEB J.7, 938–943.[Abstract/Free Full Text]

Harvey, M., McArthur, M. J., Montgomery, C. A., Jr., Butel, J. S., Bradley, A., and Donehower, L. A. (1993b). Spontaneous and carcinogen-induced tumorigenesis in p53-deficient mice. Nat. Genet.5, 225–229.[ISI][Medline]

Jiang, W., Ananthaswamy, H. N., Muller, H. K., Ouhtit, A., Bolshakov, S., Ullrich, S. E., El-Naggar, A. K., and Kripke, M.L. (2001). UV irradiation augments lymphoid malignancies in mice with one functional copy of wild-type p53. Proc. Natl. Acad. Sci. U.S.A.98, 9790–9795.[Abstract/Free Full Text]

Jones, J. M., Attardi, L., Godley, L. A., Laucirica, R., Medina, D., Jacks, T., Varmus, H.E., and Donehower, L. A. (1997). Absence of p53 in a mouse mammary tumor model promotes tumor-cell proliferation without affecting apoptosis. Cell Growth Differ.8, 829–838.[Abstract]

Kane, A. B. (1996). Mechanisms of mineral fibre carcinogenesis. In Mechanisms of Fibre Carcinogenesis (A. B. Kane, P. Boffetta, R. Saracci, and J. D. Wilbourn, Eds.), pp. 11–34. IARC Scientific Publications No. 140. International Agency for Research on Cancer, Lyon, France.

Kemp, C. J., Donehower, L. A., Bradley, A., and Balmain, A. (1993). Reduction of p53 gene dosage does not increase initiation or promotion but enhances malignant progression of chemically induced skin tumors. Cell74, 813–822.[ISI][Medline]

Kemp, C. J., Wheldon, T., and Balmain, A. (1994). p53-Deficient mice are extremely susceptible to radiation-induced tumorigenesis. Nat. Genet.8, 66–69.[ISI][Medline]

Ko, L. J., and Prives, C. (1996).p53: Puzzle and paradigm. Genes Dev.10, 1054–1072.[ISI][Medline]

Lechner, J. F., Tesfaigzi, J., and Gerwin, B. I. (1997). Oncogenes and tumor-suppressor genes in mesothelioma—a synopsis. Environ. Health Perspect. 105(Suppl. 5), 1061–1067.[ISI][Medline]

Lee, J. M., Abrahamson, J. L. A., Kandel, R., Donehower, L. A., and Bernstein, A. (1994). Susceptibility to radiation-carcinogenesis and accumulation of chromosomal breakage in p53 deficient mice. Oncogene9, 3731–3736.[ISI][Medline]

Livingston, L. R., White, A., Sprouse, J., Livanos, E., Jacks, T., and Tlsty, T. D. (1992). Altered cell cycle arrest and gene amplification potential accompany loss of wild-type p53. Cell70, 923–935.[ISI][Medline]

Macdonald, J. L., and Kane, A. B. (1997). Mesothelial cell proliferation and biopersistence of wollastonite and crocidolite asbestos fibers. Fundam. Appl. Toxicol. 38, 173–183.[ISI][Medline]

Macleod, K. F., and Jacks, T. (1999). Insights into cancer from transgenic mouse models. J. Pathol. 187, 43–60.[ISI][Medline]

Marsella, J. M., Liu, B. L., Vaslet, C. A., and Kane, A. B. (1997). Susceptibility of p53-deficient mice to induction of mesothelioma by crocidolite asbestos fibers. Environ. Health Perspect.105(Suppl. 5), 1069–1072.[ISI][Medline]

Moalli, P. A., Macdonald, J. L., Goodglick, L. A., and Kane, A. B. (1987). Acute injury and regeneration of the mesothelium in response to asbestos fibers. Am. J. Pathol. 128, 426–445.[Abstract]

Murthy, S. S., and Testa, J. R. (1999). Asbestos, chromosomal deletions, and tumor suppressor gene alterations in human malignant mesothelioma. J. Cell Physiol. 180, 150–157.[ISI][Medline]

Naik, P., Karrim, J., and Hanahan, D. (1996). The rise and fall of apoptosis during multistage tumorigenesis: Down-modulation contributes to tumor progression from angiogenic progenitors. Genes Dev. 10, 2105–2116.[Abstract]

Pipas, J. M., and Levine, A. J. (2001). Role of T-antigen interactions with p53 in tumorigenesis. Semin. Cancer Biol.11, 23–30.[ISI][Medline]

Preston, R. J. (1996). Interindividual variations in susceptibility and sensitivity: Linking risk assessment and risk management. Toxicology 111, 331–341.[ISI][Medline]

Prives, C., and Hall, P. A. (1999). The p53 pathway. J. Pathol.187, 112–126.[ISI][Medline]

Robledo, R., and Mossman, B. (1999). Cellular and molecular mechanisms of asbestos-induced fibrosis. J. Cell Physiol. 180, 158–166.[ISI][Medline]

Symonds, H., Krall, L., Remington, L., Saenz-Robles, M., Lowe, S., Jacks, T., and Van Dyke, T. (1994).p53-Dependent apoptosis suppresses tumor growth and progression in vivo. Cell78, 703–711.

Tennant, R. W., French, J. E., and Spalding, J. W. (1995). Identifying chemical carcinogens and assessing potential risk in short-term bioassays using transgenic mouse models. Environ. Health Perspect. 103, 942–950.

Waheed, I., Guo, Z. S., Cen, G. A., Weiser, T. S., Nguyen, D. M., and Schrump, D. S. (1999). Antisense to SV0 early region induces growth arrest and apoptosis in T-antigen-positive human pleural mesothelioma cells. Cancer Res. 59, 6068–6073.