Neoplastic transformation of normal rat embryo fibroblasts by a mutated p53 and an activated ras oncogene induces parathyroid hormone-related peptide gene expression and causes hypercalcemia in nude mice.

Parathyroid hormone-related peptide (PTHRP) is a 141-amino acid protein identified in various carcinomas associated with humoral hypercalcemia of malignancy (HHM). Although the causal role of PTHRP in HHM syndrome has been established, the molecular and cellular mechanism by which PTHRP gene is overexpressed in certain malignancies remains unknown. We have demonstrated in the present study that PTHRP secretion was markedly induced concomitantly with the formation of transformed foci after normal rat embryo fibroblasts (REFs) were co-transfected with an activated ras (ras) and a mutated form of p53 (p53-mt) genes. In either ras- or p53-mt-transfected (nontransformed) cells, only modest or barely detectable secretion of PTHRP was observed, respectively. Northern blot analysis revealed that PTHRP mRNA was markedly induced in fully transformed cells 11 days after transfection with both ras and p53-mt genes. Inhibition of RNA synthesis with actinomycin D resulted in almost complete disappearance of PTHRP mRNA at 2-3 h, suggesting a transcriptional mechanism. Transient transfection experiments revealed that PTHRP promoter activity was induced in ras + p53-mt transfectants. REFs transformed by ras and p53-mt genes and thereby induced to secrete PTHRP in vitro produced aggressively growing tumors associated with HHM syndrome when injected into nude mice. These results suggest that activation of PTHRP gene is closely related to malignant transformation of normal mammalian cells and that ras and p53 may be important regulators of PTHRP gene transcription. The transfection-focus formation system of REFs should provide an excellent model to study the molecular and cellular mechanism underlying concomitant overexpression of PTHRP gene with carcinogenesis.

Parathyroid hormone-related peptide (PTHRP) is a 141-amino acid protein identified in various carcinomas associated with humoral hypercalcemia of malignancy (HHM). Although the causal role of PTHRP in HHM syndrome has been established, the molecular and cellular mechanism by which PTHRP gene is overexpressed in certain malignancies remains unknown. We have demonstrated in the present study that PTHRP secretion was markedly induced concomitantly with the formation of transformed foci after normal rat embryo fibroblasts (REFs) were co-transfected with an activated ras (ras) and a mutated form of p53 (p53-mt) genes. In either ras-or p53-mt-transfected (nontransformed) cells, only modest or barely detectable secretion of PTHRP was observed, respectively. Northern blot analysis revealed that PTHRP mRNA was markedly induced in fully transformed cells 11 days after transfection with both ras and p53-mt genes. Inhibition of RNA synthesis with actinomycin D resulted in almost complete disappearance of PTHRP mRNA at 2-3 h, suggesting a transcriptional mechanism. Transient transfection experiments revealed that PTHRP promoter activity was induced in ras ؉ p53-mt transfectants. REFs transformed by ras and p53-mt genes and thereby induced to secrete PTHRP in vitro produced aggressively growing tumors associated with HHM syndrome when injected into nude mice. These results suggest that activation of PTHRP gene is closely related to malignant transformation of normal mammalian cells and that ras and p53 may be important regulators of PTHRP gene transcription. The transfection-focus formation system of REFs should provide an excellent model to study the molecular and cellular mechanism underlying concomitant overexpression of PTHRP gene with carcinogenesis.
Parathyroid hormone-related peptide (PTHRP) 1 is a 141-amino acid protein identified in various carcinomas as a causative factor for humoral hypercalcemia of malignancy (HHM) (1)(2)(3). PTHRP shows structural and functional similarities to parathyroid hormone, including a sequence homology at the amino-terminal parathyroid hormone-like domain (1)(2)(3). Thus PTHRP overproduced and secreted by certain carcinomas enters the circulation (4), interacts with a common receptor for parathyroid hormone/PTHRP in bone and kidney (5), and causes hypercalcemia in cancer patients. The findings that circulating PTHRP levels are elevated in most patients with HHM (4) and that hypercalcemia is corrected by a neutralizing antibody against PTHRP in animal models of HHM (6) have further strengthened the causal role for PTHRP in the pathogenesis of HHM syndrome.
Accumulating evidence suggests that PTHRP gene is expressed at low levels in a number of normal tissues and plays diverse physiological roles mainly in an autocrine/paracrine fashion (7). However, the exact molecular and cellular mechanism by which PTHRP gene is overexpressed in certain carcinomas remains unknown.
Carcinogenesis is a complex, multistep process, and it has been established that oncogenes and tumor suppressor genes play the fundamental role as the genetic targets, the alteration of which defines each of the distinct steps in the multistep cellular transformation (8,9). The complex process involving the conversion of normal cells to malignant phenotypes can be reconstructed in vitro using the transfection focus formation assay, which has been elaborated to test the oncogenic potentials of various tumor-related genes. Using this assay cytoplasmic oncogenes, represented by ras and src, and nuclear oncogenes, such as myc and mutated p53, have been shown to collaborate with each other to fully transform normal rodent cells by acting in a complementary way, culminating in tumor formation in vivo (10 -12).
In the present study we have demonstrated that PTHRP gene expression is markedly induced in close association with transformation of normal rat embryo fibroblasts (REFs) by co-transfection of an activated ras gene and a mutated p53 gene. Furthermore, the transformed cells produced aggressively growing tumors with hypercalcemia in vivo when inoculated into nude mice. The current reconstruction system of HHM syndrome, starting with normal mammalian cells through the malignant transformation in vitro and tumor formation in vivo, may provide an excellent model to dissect the molecular and cellular events leading to the overexpression of PTHRP gene in certain malignancies.

MATERIALS AND METHODS
Plasmids-pucEJ (13) (courtesy of Dr. Steven F. Dowdy, Whitehead Institute for Biomedical Research, Cambridge, MA) was used to express an oncogenic allele of Ha-ras (EJ-ras), and p53 KH215 (14) (courtesy of Dr. Philip W. Hinds, Harvard Medical School, Boston, MA) with an in-frame insertion of decameric HindIII linker into the KpnI site at amino acids 215-216 of the murine p53 sequence was used to express a mutated form of p53 (p53-mt). ras and p53 were chosen as target DNA since it has been demonstrated that these genes are mutated in a variety of human cancers and that these mutations cooperate with each other to convert normal fibroblasts to malignant phenotypes (12).
Cell Culture and Transfections-Rat embryo fibroblasts (REFs) were prepared from 14-day-old Wistar rat embryos (Tokyo Laboratory Animals Science Co. Ltd., Tokyo, Japan) as described previously (15), and grown for 2 days in Dulbecco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% fetal bovine serum and 60 g/ml kanamycin. Adherent cells were trypsinized and frozen in liquid nitrogen until use.
For transfection experiments REFs were plated at 3-5 ϫ 10 5 /6-cm dish, and a total of 20 g of plasmid DNA consisting of test plasmids or the control plasmid (pSV-SPORT 1, Life Technologies, Inc.) and neo gene plasmid (pcDneo) was transfected on the following day by the calcium phosphate method according to Chen and Okayama (16). After 14 -18 h, cells were washed with phosphate-buffered saline and cultured in fresh medium for an additional day. Then cells were trypsinized, replated into 10-cm dishes, and fed every 3-4 days thereafter. In order to monitor successful transfection, one-tenth of cells were plated into a 6-cm dish and cultured in the presence of 400 g/ml Geneticin (Gibco). The number of foci, the diameter of which was 3 mm or larger, was scored after Giemsa staining 11 days after the transfection. At the indicated intervals, cells and the conditioned medium were harvested separately for RNA analysis and determination of PTHRP concentrations, respectively. Actinomycin D (Sigma) was dissolved in ethanol, and control cells received an equal volume of vehicle.
Animals-Five-week-old male athymic mice were purchased from Clea Japan Inc. (Tokyo, Japan), maintained in sterilized cages, and fed standard rodent chow containing 1.25% calcium and 1.06% phosphate (CE-2, Clea Japan Inc.). Tumorigenicity and the inducibility of hypercalcemia in vivo were studied by inoculating 5-10 ϫ 10 6 cells subcutaneously into the right flank of nude mice. Tumor size was calculated by the following formula: tumor volume (ml) ϭ a ϫ b 2 /2, where a is the length (cm) and b is the width (cm) of the tumor. Blood was drawn every week by retroorbital plexus puncture for determination of blood ionized calcium (Ca 2ϩ ). Blood Ca 2ϩ concentrations were measured by the electrode method using an autoanalyzer (M-634, Chiba Corning Diagnostics Co. Ltd., Tokyo, Japan). At sacrifice blood was drawn by cardiac puncture for determination of plasma PTHRP concentrations. The animals were treated in accordance with Chugai Pharmaceutical's ethical guidelines of animal care, handling, and termination, and the current experimental protocols were approved by the animal care committee of the institution.
Immunoradiometric Assay (IRMA) for PTHRP-PTHRP concentrations were determined with a sensitive two-site IRMA for human PTHRP-(1-87) as described (17). In brief, conditioned medium or plasma (200 l) was first incubated with 100 l of affinity-purified anti-PTHRP-(50 -83) polyclonal antibody and a polystylene bead coupled to anti-PTHRP-(1-34) monoclonal antibody for 18 h. After washing with water, the bead was incubated with 125 I-labeled anti-rabbit IgG goat antibody for additional 18 h, and radioactivity associated with the bead was counted after washing with water. The intraassay and interassay variations were 3.4 -5.8% and 4.0 -7.5%, respectively, and the detection limit, as determined by 2 standard deviations above the mean counts of nonspecific binding, was 0.5 pmol/liter (17).
RNA Analysis-Cells were dissolved and tumors were homogenized in guanidinium thiocyanate solution, and total cellular RNA was prepared by pelleting through cesium chloride cushion and quantitated by A 260 . Northern blot analysis was performed under stringent conditions as described previously (18), using a 308-base pair PvuII-SacI cDNA probe for PTHRP (19) (courtesy of Drs. A. E. Broadus and M. Mangin, Yale University, New Haven, CT) and a 2.1-kb BamHI cDNA probe for human ␤-actin (20). The probes were labeled with deoxycytidine 5Ј-[␣-32 P]triphosphate (3000 Ci/mmol, Amersham Corp.) by the random primer technique (Multiprimer labeling kit, Amersham), and used after passage through Sephadex G-50 quick spin column (Boehringer Mannheim). The blots were washed in 0.1 ϫ SSC and 0.1% SDS at 65°C for 1 h, and the washed membranes were exposed to Kodak X-Omat film at Ϫ80°C with intensifying screens.
Transient Transfection and CAT Assay-REFs were transfected with test plasmids or the control DNA and cultured in the presence of Geneticin (500 g/ml) for 11 days. Eighteen hours after replating, cells were co-transfected with 15 g of a PTHRP-CAT construct containing a BamHI-BalI fragment (Ϫ638 to ϩ105 base pairs relative to the transcription start site) of the human PTHRP gene containing the downstream TATA element (21) (courtesy of Drs. A. E. Broadus and M. Mangin, Yale University, New Haven, CT) and 5 g of RSV-␤-galactosidase plasmid by the standard calcium phosphate method according to the manufacturer's instructions (5 Prime 3 3 Prime, Inc., Boulder, CO). After 48 hours of culture, cell extracts were assayed for protein concentration, ␤-galactosidase activity, and CAT activity as described previously (22). The induction of ␤-galactosidase activity did not differ by more than 33% among transfections.
Statistical Analysis-Data were expressed as mean Ϯ S.E., and statistical significance was determined by analysis of variance.

RESULTS AND DISCUSSION
In order to examine the relationship between malignant transformation and PTHRP overproduction, we first attempted to establish an in vitro system in which normal mammalian cells could be converted to transformed cells through introduction of known oncogenes. Normal rat embryo fibroblasts (REFs) have been used in the transfection focus formation assay to test oncogenic potentials of various tumor-related genes and shown to require at least two different types of oncogenes for full transformation (11,12). As shown in Fig. 1, no foci of morphologically altered cells were seen in the control plasmid-or p53-mt-transfected cultures, and only small number of foci were seen 11 days after transfection of ras oncogene alone (4 foci/10-cm dish). In contrast, co-transfection with ras and p53-mt induced the appearance of large numbers of overgrowing foci of cells with transformed morphology 11 days after transfection (80 foci/10-cm dish). These results are in good agreement with the previous observations that an activated ras gene and a mutated p53 gene cooperate with each other in cellular transformation (12).
We next examined the expression of PTHRP gene in the transfection-focus formation system. The conditioned medium was harvested during the last 3 days, and PTHRP concentrations were determined by a sensitive and specific IRMA for human PTHRP-(1-87) (17) since no immunoassay specific to rat PTHRP was available. Amino acid sequences of PTHRP are remarkably conserved among various species up to the residue 111 (1), and it has been reported that IRMA for human PTHRP-(1-74) is capable of detecting circulating PTHRP derived from FIG. 1. Transformation of normal rat embryo fibroblasts after transfection with EJ-ras and/or mutated p53 genes. REFs were prepared as described previously (15). After plating at a density of 3 ϫ 10 5 cells/6-cm dish, transfection was carried out with pSV-SPORT 1 (15 g) as a control DNA (Control), the expression construct for mutated p53 (p53-mt, 15 g), the plasmid for EJ-ras (ras, 15 g), or both EJ-ras (12 g) and mutated p53 (3 g) (ras ϩ p53-mt) together with 5 g of pcDneo plasmid as described under "Materials and Methods." The photographs were taken 11 days after transfection in a representative experiment. Similar results were obtained in several independent experiments. a rat tumor associated with hypercalcemia (23). As shown in Fig. 2, cells transfected with the control vector or p53-mt alone secreted barely detectable amounts of PTHRP, and cells transfected with ras oncogene secreted larger amounts of PTHRP, although not statistically significant. In contrast, fully transformed cells after co-transfection with ras and p53-mt genes secreted markedly elevated amounts of PTHRP per the same number of cells (Fig. 2). Dilution of the medium conditioned by ras ϩ p53-mt transformants revealed a response curve parallel to that of human PTHRP-(1-87) standard in the IRMA, and immunoreactive PTHRP was barely detectable in plasma from non-tumor-bearing nude rats or mice (data not shown). These results suggest that the IRMA used in the current study crossreacted with rat PTHRP, although we cannot rule out the possibility that our assay underestimated the concentrations of rat peptide.
The time course of the induction of PTHRP secretion after transfection of the oncogenes is shown in Fig. 3. Cells cotransfected with ras and p53-mt genes secreted gradually increasing but modest amounts of PTHRP until day 8 and were markedly induced to secrete large amounts of PTHRP between 8 and 11 days after transfection. Again, cells transfected with ras alone secreted much smaller amounts of PTHRP during the same period (Fig. 3). In good agreement with the results of PTHRP secretion, the expression of PTHRP mRNA was markedly induced in fully transformed cells transfected with both ras and p53-mt, and only a faint band corresponding to rat PTHRP transcript was visible in cells transfected with ras alone (Fig. 4A). PTHRP mRNA was not readily detectable in control or p53-mt-transfected cells.
Malignant cells are known to produce a variety of cytokines, such as hematopoietic colony-stimulating factors, interleukin 1, and interleukin 6, and it has been reported that transfection of an activated ras oncogene into normal human fibroblasts induces these cytokine genes mainly through a post-transcriptional mechanism (24). The PTHRP gene has certain characteristics of an early response gene, including the presence in PTHRP mRNAs of the AUUUA instability motifs at the 3Јuntranslated regions (25). In fact it has been demonstrated that PTHRP mRNA turns over rapidly in vitro (26) and in vivo (27) with the reported half-life ranging from 30 min to 2 h. In order to determine whether the induction of PTHRP mRNA in ras ϩ p53-mt transfectants occurred at a transcriptional or post-transcriptional level, the disappearance rate of PTHRP mRNA was assessed after replating the fully transformed cells. As shown in Fig. 4B, PTHRP mRNA level was markedly decreased at 2 h and disappeared at 3 h after the addition of actinomycin D, an inhibitor of RNA synthesis, suggesting that the expression of PTHRP gene was induced mainly at a transcriptional level. Furthermore, when transient transfection ex-  (Control, 15 g), p53 mutant (p53-mt, 15 g), EJ-ras alone (ras, 15 g), or EJ-ras (12 g) plus p53 mutant (3 g) (ras ϩ p53-mt) together with pcDneo plasmid (5 g), and the conditioned medium was collected at the indicated intervals after transfection. PTHRP concentrations were determined by IRMA using human PTHRP-(1-87) as the standard. Data are shown as mean Ϯ S.E. (n ϭ 3-12), and asterisks indicate significant difference from the control group (p Ͻ 0.001). Quantitatively similar results were obtained in three separate experiments.

FIG. 4. Concomitant induction of PTHRP mRNA with malignant transformation after transfection with EJ-ras and p53-mt genes.
A, REFs were transfected with pSV-SPORT 1 (Control), p53 mutant (p53-mt), EJ-ras alone (ras), or both EJ-ras and p53-mt genes (ras ϩ p53-mt) together with pcDneo plasmid, and total cellular RNA was prepared 11 days after transfection. Northern blot analysis was performed using PTHRP and ␤-actin cDNA probes as described under "Materials and Methods." RNA prepared from a human carcinoma associated with humoral hypercalcemia (38) served as a positive control (Human Ca), and the sizes of the multiple transcripts of human PTHRP mRNA are indicated. Similar results were confirmed in three separate experiments. B, REFs were co-transfected with EJ-ras (12 g) and p53-mt (3 g), and 11 days after transfection cells were trypsinized, split into eight 10-cm dishes, and then cultured in the absence (Control) or presence of actinomycin D (AcD, 10 g/ml). Total cellular RNA was prepared at the indicated times after RNA synthesis was inhibited with actinomycin D, and Northern blot analysis was performed using PTHRP and ␤-actin cDNA probes as described under "Materials and Methods." Control cells received an equal volume of vehicle (ethanol). periments were performed using a PTHRP-CAT reporter construct, CAT activity was markedly induced only in fully transformed cells with both ras and p53-mt genes (Fig. 5). These results not only confirm the transcriptional mechanism of PTHRP gene induction but suggest that some trans-acting cellular factor is induced in close association with malignant transformation by ras and p53-mt. The PTHRP promoter fragment used in the current transfection experiments contains binding sites for various transcription factors, including Sp 1, Ets, and AP-2 (28 -30). Identification of a trans-acting factor as well as a DNA element responsible for PTHRP gene activation in ras ϩ p53-mt-transformed cells will help gain further insight into the molecular mechanism underlying PTHRP overproduction in certain malignancies.
We finally examined whether cells fully transformed by ras and p53-mt genes and thereby induced to produce large amounts of PTHRP in vitro were indeed capable of forming tumors and causing HHM syndrome in vivo. Nude mice were inoculated with ras ϩ p53-mt transfectants and monitored for subsequent appearance of subcutaneous tumors and blood Ca 2ϩ concentrations. As shown in Fig. 6 (left panel), animals inoculated with ras-transfected cells did not produce tumors or develop hypercalcemia, whereas those with the same number (1 ϫ 10 7 ) of cells co-transfected with ras and p53-mt genes produced aggressively growing tumors and developed hypercalcemia. The ability of ras ϩ p53-mt transfectants to induce tumors associated with marked hypercalcemia in vivo was further confirmed in three out of four nude mice inoculated with 5 ϫ 10 6 cells in a separate experiment (Fig. 6, right panel). The fourth animal showed marginal hypercalcemia. As shown in Fig. 7A, peak values of blood Ca 2ϩ were significantly higher in nude mice inoculated with ras ϩ p53-mt transfectants than those in age-matched control animals, and the hypercalcemic nude mice displayed markedly elevated plasma PTHRP concentrations as compared with controls (0.6 Ϯ 0.02 pM) (Fig. 7B). Plasma PTHRP concentrations in these hypercalcemic animals were comparable with those in patients with HHM determined with the current IRMA (17). Evidently the tumors associated with hypercalcemia expressed high levels of a single transcript for rat PTHRP mRNA (Fig. 7C).
In summary, we have demonstrated in the present study that introduction of two types of oncogenes, EJ-ras and p53-mt, was necessary and sufficient not only for the conversion of normal mammalian cells into malignant cells but for the induction of PTHRP gene transcription in vitro and the development of HHM syndrome in vivo. A family of ras genes encode highly homologous proteins with molecular mass of 21 kDa, which play pivotal roles in the transduction of extracellular signals to the nucleus (31). It has been shown in a human keratinocyte model that a progressive dysregulation of PTHRP gene expression occurs during the conversion from established phenotypes conferred by human papilloma virus 16 to malignant transformation induced by ras oncogene (32). In addition, Li and Drucker (33) have recently reported that transfection of either EJ-Ha-ras or v-src oncogene into established cell lines, NRK 49F (fibroblasts) and RCB 2.2 (osteoblast-enriched cell line), induced PTHRP mRNA expression at a transcriptional level, although it remained to be proven whether the increased production of PTHRP after transfection with either EJ-ras or v-src alone was sufficient to cause HHM syndrome in vivo. We have observed in the current study that the transfection of EJ-ras alone into REFs resulted in a modest increase in FIG. 5. Induction of PTHRP promoter activity in transformed cells with ras and p53-mt. REFs were transfected with pSV-SPORT 1 (Control), p53 mutant (p53-mt), EJ-ras alone (ras) or both EJ-ras and p53-mt genes (ras ϩ p53-mt) together with pcDneo plasmid. Eleven days after transfection cells were trypsinized and split into 6-cm dishes, and transient transfection experiments were performed using a PTHRP-CAT construct and calcium phosphate technique as described under "Materials and Methods." CAT activity of cell extracts was assayed on thin-layer chromatography. Similar results were confirmed in two separate experiments.
FIG. 6. EJ-ras-and p53-mt-transformed cells induce hypercalcemia in nude mice. Left panel, REFs were transfected with EJ-ras alone (15 g, open circles) or EJ-ras (12 g) plus p53-mt (3 g) genes (solid circles) together with pcDneo plasmid (5 g). Eleven days after transfection, cell number was counted, and 1 ϫ 10 7 cells were injected into nude mice. Right panel, a similar experiment was performed with 5 ϫ 10 6 cells after transfection with both EJ-ras and p53-mt. Tumor size (upper panel) and blood Ca 2ϩ concentrations (lower panel) were determined at the indicated intervals. Tumor volume was calculated from the length (cm) and the width (cm) of the tumor as described under "Materials and Methods." Shaded areas indicate the normal range of blood Ca 2ϩ in age-matched non-tumor-bearing nude mice.
FIG. 7. PTHRP expression in nude mice inoculated with ras ؉ p53-mt transfectants. Blood Ca 2ϩ (A) and plasma PTHRP concentrations (B) in nude mice inoculated with ras ϩ p53-mt transfectants (n ϭ 3) and age-matched control animals (n ϭ 10) were determined as described under "Materials and Methods." Data are shown as mean Ϯ S.E., and asterisks indicate significant difference from the control group (p Ͻ 0.001). C, total RNA was extracted from tumors of hypercalcemic nude mice (n ϭ 3), and Northern blot analysis was performed using PTHRP and ␤-actin cDNAs as described under "Materials and Methods." PTHRP secretion, although not statistically significant (Figs. 2  and 3), and a slight induction of PTHRP mRNA on Northern analysis (Fig. 4). Taken together, it is conceivable that the activation of intracellular signal transduction pathways by ras oncogene seems to be important for PTHRP gene expression.
Although ras mutations have been detected in a variety of human cancers, notably adenocarcinoma of the pancreas, the colon, and the lung and thyroid tumors (34), most of these tumors are not frequently associated with HHM syndrome except for pancreas cancer (1-3), raising the possibility that ras mutations are not sufficient for full induction of PTHRP gene expression. In our experimental system malignant transformation by both ras and p53-mt, but not by transfection of ras alone, was associated with a marked induction of PTHRP gene transcription in vitro and the reconstitution of HHM syndrome in vivo. Taken together, it is tempting to hypothesize that the development of HHM requires another genetic alteration, such as a mutation of p53 gene in the current system or the introduction of human papilloma virus 16 DNA in the keratinocyte model (32). It remains to be elucidated, however, what alteration in cellular functions the second mutation confers upon transformed cells. Our preliminary analysis showed that ras ϩ p53-mt transfectants contained a modestly increased amount (1.5-2-fold) of ras mRNA compared with cells transfected with ras alone (data not shown), which may suggest that the effect of a mutated p53 gene is mediated, at least in part, through the up-regulation of ras gene expression.
The p53 tumor-suppressor gene encodes a 53-kDa nuclear phosphoprotein involved in the control of cell proliferation (35,36), and its mutations have been detected in approximately half of almost all types of human cancers, as diverse as colon and breast carcinomas, lymphomas and leukemias, lung and esophageal carcinomas, and hepatocellular carcinoma (37). Although the biochemical activities of wild-type p53 and its mutated forms are not entirely clear, there is accumulating evidence suggesting that p53 binds DNA specifically and affects the expression of various genes and that these functions of p53 are interfered with by its mutations (35,36). The current system may be useful to examine the possible role for p53 mutation in the activation of PTHRP gene during the process of cellular transformation. Also, the association of HHM syndrome with the occurrence of p53 and ras mutations in human tumors remains to be elucidated.
In conclusion, the results of the present study suggest that the overproduction of PTHRP and development of HHM syndrome occur during the multistep transformation of normal mammalian cells to malignant phenotypes and that ras and p53 may be important upstream regulators of PTHRP gene transcription. Since our experimental system can reproduce HHM syndrome in a relatively short period of time (11 days in vitro and 4 -5 weeks in vivo), it should provide an excellent model to further dissect the molecular and cellular events involved in the overexpression of PTHRP gene in certain malignancies.