Regulation of Parathyroid Hormone-Related Peptide by Estradiol: Effect on Tumor Growth and Metastasis in Vitro and in Vivo
http://www.100md.com
内分泌学杂志 2005年第7期
Department of Medicine and Oncology, McGill University Health Centre, Montreal, Quebec, Canada H3A 1A1
Address all correspondence and requests for reprints to: Shafaat A. Rabbani, M.D., McGill University Health Centre, 687 Pine Avenue West, Room H4.61, Montreal, Quebec, Canada H3A 1A1. E-mail: shafaat.rabbani@mcgill.ca.
Abstract
We evaluated the capacity of estradiol (E2) to regulate PTHrP production, cell growth, tumor growth, and metastasis to the skeleton in breast cancer. In estrogen receptor (ER)-negative human breast cancer cells, MDA-MB-231, and cells transfected with full-length cDNA encoding ER (S-30), E2 caused a marked decrease in cell growth and PTHrP production, effects that were abrogated by anti-E2 tamoxifen. E2 also inhibited PTHrP promoter activity in S-30 cells. For in vivo studies, MDA-MB-231 and S-30 cells were inoculated into the mammary fat pad of female BALB/c nu.nu mice. Animals receiving S-30 cells developed tumors of significantly smaller volume compared with MDA-MB-231 tumor-bearing animals. This change in tumor volume was reversed when S-30 cells were inoculated into ovariectomized (OVX) hosts. Inoculation of MDA-MB-231 cells into the left ventricle resulted in the development of lesions in femora and tibia as determined by x-ray analysis. In contrast, these lesions were significantly smaller in volume and number in animals inoculated with S-30, and this lower incidence was reversed in OVX animals. Bone histological analysis showed that the tumor volume to tissue volume ratio was comparable with that seen by x-ray. Immunohistochemical analysis showed that PTHrP production was inhibited in S-30 group and restored to levels comparable to that seen in MDA-MB-231 tumor-bearing animals when S-30 cells were inoculated in OVX animals. Collectively these studies show that E2 production is inversely correlated with PTHrP production and that the growth-promoting effect of PTHrP has a direct impact on tumor growth at both nonskeletal and skeletal sites.
Introduction
BREAST CANCER IS the most frequent malignancy occurring in women and results in a high rate of morbidity and mortality (1, 2). It is unique among malignancies due to the high rate of developing osteolytic skeletal metastases (3). It is now well established that estradiol (E2) plays an important role in mammary development by binding to nuclear estrogen receptors (ER). These actions are mediated by high-affinity receptors that form a receptor complex that can associate with E2 response elements located on several genes to alter cell growth and differentiation (2, 4, 5, 6). Apart form its physiological effects, E2 has also been associated with malignant transformation and induction of tumor cell proliferation (4, 5). The main source of E2 in females are the ovaries, and blockage of these tumor-promoting effects with the anti-E2 tamoxifen (TAM) is one of the most commonly used forms of hormone therapy for early-stage hormone-responsive breast cancer (7). In the initial assessment of patients with breast cancer, determination of the ER status serves as a major criterion for designing the therapeutic strategy in view of the fact that the effectiveness of hormone therapy is dependent upon the expression of functional ER by tumor cells (8, 9). Lack of ER-positive tumor cells is considered to denote a more aggressive tumor phenotype and a poorer prognosis (10). Whereas two ER isoforms (ER and ER?) have been identified, ER is the main form determining hormone dependence because ER? expression is generally found to be low in the majority of breast and ovarian cancers (11, 12). Despite the requirement for ER for an estrogen response, only 60% of ER-positive breast cancer patients respond to hormone therapy with TAM (7). In late stages, when tumor cells are ER negative, they are highly invasive and are often associated with osteolytic skeletal metastases (13, 14, 15). This invasive transformation of tumor cells is achieved via selective induction and production of growth factors, cytokines, and proteases that are believed to promote tumor cell proliferation, invasion, and metastasis to different organs. Breast cancer per se has a particular predilection to metastasize to the skeleton (13, 14, 15). Once these skeletal metastases are established, they result in bone pain due to nerve compression and pathological fractures. The presence of breast tumor cells in the skeleton results in the release of growth factors, cytokines, and proteases from the bone microenvironment that promotes breast cancer-associated osteolysis and results in hypercalcemia of malignancy in at least 15–20% of patients (13, 14, 15, 16, 17).
Whereas a large number of factors have been implicated in promoting hypercalcemia of malignancy, PTHrP has now been shown to be the major pathogenic factor in this syndrome (17, 18, 19). PTHrP appears to exert its actions by binding to a G protein-coupled receptor (type I PTH/PTHrP receptor) that also interacts with the amino-terminal region of PTH (20). PTHrP is produced in a number of malignancies, including breast cancer where it can modulate cell growth and differentiation (21, 22). Since the isolation and characterization of PTHrP, a number of growth factors, cytokines, and steroids have been shown to regulate PTHrP production (23, 24, 25). In a series of studies, we have shown the ability of fetal bovine serum and growth factors to induce PTHrP production, whereas dexamethasone, vitamin D, and its low calcemic analog EB1089 could inhibit PTHrP production (26, 27). More recently, we have shown that androgen was able to decrease PTHrP production by human prostate cancer cells in vitro and to decrease tumor growth in vivo (28).
In the current study, we assessed the ability of E2 to regulate PTHrP production in MDA-MB-231 human breast cancer cells that are E2 insensitive due to the silencing of ER gene transcription and in MDA-MB-231 cells transfected with a full-length cDNA encoding human ER (S-30) (29). Previous studies have shown that in S-30 cells, levels of ER expression and E2 binding affinity are similar to ER expression and ligand binding affinity in MCF-7 human breast cancer cells, which naturally express ER (29). Nevertheless, S-30 cells and MCF-7 cells behave differently with respect to responsiveness to E2 (29). Therefore, we compared the effect of E2 on S-30 and MCF-7 cell growth in vitro and examined the effect of E2 on S-30 cells tumor growth and metastasis to the skeleton in nude mice in vivo using a xenograft model of human breast cancer.
Materials and Methods
Cells and cell culture
MCF-7 and MD-MB-231 human breast cancer cells were obtained from American Type Tissue Culture Collection (Rockville, MD) and maintained in MEM (Life Technologies, Inc., Rockville, MD) supplemented with 10% FBS, 100 U/ml of penicillin-streptomycin sulfate, and 0.2% gentamycin as described previously (29, 30). MDA-MB-231 cells transfected with a functional ER (S-30 cells) were kindly provided by Dr. C. Jordan (Robert H. Lurie Comprehensive Cancer Center, Feinberg School of Medicine, Northwestern University, Chicago, IL) and were maintained in MEM under selection for neomycin resistance with geneticin (G418) (Life Technologies, Inc.) as described (29, 31). MCF-7 cells and S-30 cells have been demonstrated to express approximately 500 fmol of ER per milligram of protein and 450 fmol of ER per milligram of protein, respectively, and to display similar binding affinities (Kd equal to 0.2 nM) (29).
Cell proliferation assay
MCF-7, MDA-MB-231, and S-30 cells were plated in triplicate at a density of 10,000 cells per well in 2 ml media in six-well plates. Cells were grown in FBS or phenol-red-free charcoal-stripped FBS (SFBS) (Invitrogen, Burlington, Ontario, Canada) depleted of E2. Where indicated, cells in stripped serum were treated with either 10–7–10–10 M E2 (Sigma-Aldrich, Oakville, Ontario, Canada) alone or in combination with 10 μM anti-E2 TAM (Sigma-Aldrich). Cells were trypsinized and counted at the times indicated using a Coulter counter (model ZF; Coulter Electronics, Harpenden, Hertfordshire, UK). Cell culture medium was replenished every second day.
RT-PCR and luciferase and ?-galactosidase (?-gal) assays
Total RNA was extracted from MDA-MB-231 and S-30 cells by a single-step method using TRIzol reagent (Invitrogen). The following primers were used to amplify PTHrP: forward 5'-GAGCACCAGCTACTGCATGACAAG-3' and reverse 5'-TGAGTTAGGTATCTGCCCTCGT-3' as previously described (28). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers were used as a control. The amplified PCR product was fractionated on a 1.3% agarose gel and visualized by ethidium bromide staining. Band intensities for PTHrP and GAPDH were quantified using densitometric software (Quantity One; Bio-Rad Laboratories, Inc., Mississauga, Ontario, Canada).
For PTHrP promoter assays, S-30 cells (0.8 x 106) were plated on 35-mm dishes a day before transfection. Transfection was carried out using the lipofectamine method (Invitrogen) (28). The cells were cotransfected with 1 μg pGL3 vector, into which had been cloned the region representing –1954 to +179 of the rat PTHrP promoter upstream of the luciferase reporter gene, and 1 μg pCMV-?-gal reporter plasmid to determine transfection efficiency. The transfected cells were washed the next day and allowed to grow for 24 h in medium containing 10% FBS. After cells were serum deprived for 24 h, they were treated with medium containing FBS, SFBS, or SFBS plus 10–7 E2 alone or in combination with 10 μM anti-estrogen TAM. The cells were harvested with 1x reporter lysis buffer (Promega Corp., Nepean, Ontario, Canada), and luciferase activity in the supernatants was then analyzed as previously described by us using a luminometer (Monolight 2010; BD Biosciences, San Jose, CA). For ?-gal activity, 50-μl lysates were mixed with 200 μl ?-gal assay buffer (24 μM Na2HPO4, 16 μM NaH2PO4, 4 μM KCl, 400 nM MgCl2, 20 μM ?-mercaptoethanol, and 0.3 mg ortho-nitrophenyl-?-D-galactopyranoside per sample) in a 96-well plate and incubated for 4 h at 37 C. The reaction was stopped by adding 0.5 ml of 1 M Na2CO2, and the absorbance was measured at 420 nm in a Vmax plate reader (Molecular Devices, Sunnyvale, CA). Activity was determined by comparison to a standard curve. Luciferase reporter activity in relative luminescence units was normalized to ?-gal activity as described (28). All luciferase assay readings were performed in triplicate and corrected for ?-gal expression levels in each cell population.
Immunoradiometric assay (IRMA)
PTHrP secreted in the conditioned media was measured using a noncompetitive, two-site IRMA (Diagnostic Systems Laboratories, Webster, TX). PTHrP was recognized by an NH2-terminal reactive antibody [raised against PTHrP (1–34)] immobilized on a solid phase (capture antibody), and detected and quantitated with a second I125-radiolabeled COOH-terminal reactive signal antibody [raised against PTHrP (47–86)]. MDA-MB-231 and S-30 cells were plated at a density of 106 cells in 10 ml in 60-mm plates. At 70% confluence, cells were deprived of serum for 24 h and then treated in the presence or absence of E2 for 12 h. The conditioned medium was collected and the final cell number was determined using a Coulter counter (Coulter Electronics). The assay was carried out as per the manufacturer’s specifications. The reaction mixture was incubated at room temperature for 20 h, washed with deionized water, and the radioactivity from each sample was counted on a -counter for 1 min. The detection limit of the assay was 0.3 pmol/liter (3.0 pg/ml).
Animal protocols
For xenograft studies, 5-wk-old BALB/c nu.nu female mice were obtained from Charles River Inc. (St. Constant, Quebec, Canada). Before inoculation, MDA-MB-231 cells grown in serum containing culture medium were washed with Hank’s balanced buffer and centrifuged at 1500 rpm for 5 min. Cell pellets (5 x 105 cells/mouse) were resuspended in 100 μl of Matrigel (Becton Dickinson Labware, Mississauga, Ontario, Canada) and saline mixture (20% Matrigel) and injected into the mammary fat pads of the mice. All animals were numbered and kept separately in a temperature-controlled room on a 12-h light/12-h dark schedule with food and water ad libitum. Five-week-old (15–20 g) female BALB/c nu.nu mice (Charles River) were used throughout the course of the studies. For intravascular tumor cell inoculation, normal and ovariectomized (OVX) animals were anesthetized by xylazine/ketamine mixture and 1 x 105 M DA-MB-231 or S-30 cells were injected in 0.1 ml of cell suspension into the left ventricle of the heart using a 26-gauge needle (30, 32, 33, 34). All animal studies were approved by the institutional review committee.
Radiological analysis
High-resolution whole-body radiographs of animals were obtained under ketamine/xylazine anesthesia with the animals placed in a prone position against films (18 x 24 cm; Mamoray Screens, AGFA, Mortsel, Belgium) and exposed to x-ray at 25 kV for 5 sec using a Mammo Diagnost UC (Philips, Hamburg, Germany). X-ray films were processed in a Curix Compact processor (AGFA). The area of osteolytic bone metastasis was determined in all of the control and experimental groups of animals by careful evaluation of radiographs at the end of these studies. The analyses were carried out in both femora and tibia of all animals using an image analysis system after capture of radiographic prints. Measurements were taken with the aid of a digitizing tablet connected to an IBM-compatible computer (32, 33, 34). All radiographs were carefully evaluated by at least three investigators, including one radiologist, all of whom were blinded to experimental protocols.
Immunocytochemistry and immunohistochemistry
MDA-MB-231 and S-30 cells, cultured in 24-well plates, were stained immunocytochemically for PTHrP as previously described (28) using the avidin-biotin-peroxidase complex technique. Paraffin embedded tumor samples were cut into 5-μm-thick sections for immunohistochemical analysis. Immunohistochemical staining for PTHrP was performed as previously described using the avidin-biotin-peroxidase complex method (28). Briefly, the sections were dewaxed in xylene and rehydrated through a series of ethanol to water gradients. The sections were incubated in 1% normal goat serum (Vector Laboratories Inc., Burlingame, CA) for 30 min at room temperature before treatment with the primary antibody [rabbit polyclonal antiserum against PTHrP (1–34) from rabbit] at 1:200 dilution overnight at 4 C. Biotinylated goat antirabbit IgG (Vector Laboratories Inc.) was used as the secondary antibody at 1:200 for 30 min at room temperature. The slides were treated with Vectastain ABC-AP kit (Vector Laboratories Inc.), diluted 1:200 for 30 min at room temperature, and subsequently developed with Fast Red TR/Naphthol AS-MX phosphate (Sigma-Aldrich) containing 1 mM levamisole for 10–15 min. The slides were then counterstained with hematoxylin (Fisher Scientific Ltd, Nepean, Ontario, Canada) and mounted with Kaiser’s glycerol jelly. All sections were washed with Tris buffer (pH 7.6) three times for 10 min each after each step. For negative control sections, the primary antibody was omitted.
Images of stained sections were photographed with a Leica digital camera and processed using BioQuant image analysis software, version 6.50.10 (BioQuant Image Analysis Corp., Nashville, TN). The threshold was set by determining the positive staining of control sections and was used to automatically analyze all recorded images of all samples that had been stained in the same session under identical conditions. The area of immunohistochemical stained regions was calculated automatically by the software in each microscopic field. Pixel counts of the immunoreaction product were calculated automatically and were expressed as total density of the integrated immunostaining over a given area. This reflects the relative amount of PTHrP detected (33).
Histological analysis
At the end of these studies, radiologically affected and unaffected long bones were excised and fixed in 10% neutral-buffered formalin for 2 d. These bones were decalcified in 10% EDTA in PBS, pH 7.4, with constant agitation until deemed clear by the ammonium oxalate end point test (32, 33, 34). After decalcification, these specimens were dehydrated through graded ethanol and cleared in methyl salicylate before subjecting to paraffin infiltration. Sections of 5-μm thickness were then embedded in formalin and stained with hematoxylin and eosin (32, 33, 34). To reduce possible variation in analysis, four to six levels of each bone were cut, and each level was assessed to get to the portion of bone with maximum tumor burden, which was considered as a representative section. Histomorphometric measurement of total tumor volume was performed in representative sections of both tibias of the animals inoculated with MDA-MB-231 and S-30 cells. The analysis was performed on the region 1 mm below the growth plate using BioQuant image analysis software (32, 33, 34). The tumor volume and tissue volume were measured, and the percentage of tumor volume in total tissue volume was calculated (32, 33, 34).
Statistical analysis
All results are expressed as mean ± SEM. Statistical significance of the difference in numbers of osteolytic metastases and tumor volume between control and Herceptin-treated groups were analyzed by Mann-Whitney test for nonparametric samples and tumor progression was analyzed by Kaplan-Meier.
Results
Evaluation of the effect of E2 on breast cancer cell growth in vitro
The effect of E2 on breast cancer cell growth was evaluated in a cell proliferation assay in which the cell proliferation rate of MCF-7, MDA-MB-231, and S-30 cells were compared. MCF-7 cells that express endogenous ER showed an increase in their rate of proliferation with E2-containing FBS and a decrease in their rate of proliferation when cultured in phenol red-free MEM containing SFBS that is depleted of E2. Incubation of these cells in SFBS with E2 resulted in a marked increase in cell growth; the effects of E2 were reversed when MCF-7 cells were grown in the presence of E2 and anti-E2 TAM (Fig. 1A). These growth-promoting (MCF-7) and -inhibiting (MDA-MB-231) effects of E2 were dose dependent between 10–10 and 10–7 M (data not shown). When MDA-MB-231 cells were cultured in the presence of FBS they displayed a significantly greater rate of cell proliferation compared with S-30 cells (Fig. 1B). However, incubation of S-30 cells in the presence of culture medium containing SFBS allowed cell growth to increase. Because no such change in cell growth was seen in MDA-MB-21 cells transfected with a mock vector (data not shown), this change in cell growth was attributed to the presence of a functional ER in S-30 cells. Therefore, E2 differently altered cell growth in MCF-7 cells and in the ER-negative MDA-MB-231 cells in which a functional ER was restored. To establish the specificity of the growth-inhibiting effect of E2 in S-30 cells, these cells were incubated in SFBS medium in the presence of either E2 alone or E2 plus anti-E2 TAM. Medium containing SFBS and 10–7 M E2 resulted in a decrease in S-30 cell growth, and these inhibitory effects of E2 in S-30 cells were found to be dose dependent at 10–10–10–7 M as previously reported (29) (data not shown). The specificity of this estrogen-mediated effect was established when coincubation of S-30 cells in SFBS containing 10–7 M E2 and 10 μM anti-E2 TAM was able to alter this rate of S-30 cell growth to levels that were similar to those seen when S-30 cells were incubated in the presence of SFBS alone (Fig. 1B).
FIG. 1. Effect of E2 on human breast cancer cells growth. Human breast cancer cells MCF-7 (A) and MDA-MB-231 and S-30 (B) were plated in six-well plates (104 cells per plate) in MEM containing 10% FBS. S-30 cells were also incubated in phenol red-free MEM containing SFBS in the presence of 10–7 E2 or 10 μM anti-E2 TAM. At different time points, cells in various groups of triplicates were trypsinized and counted using a Coulter counter. Results are representative of three such experiments ± SEM. Significant differences in cell number from control MCF-7 or MDA-MB-231 cells are represented by one asterisk, and from S-30 cells in FBS by two asterisks (P < 0.05).
Effect of E2 on PTHrP production by breast cancer cells in vitro
To examine the effect of E2 on PTHrP production by breast cancer cells, levels of PTHRP mRNA expression by RT-PCR, immunocytochemical evidence of PTHrP (1–34) protein expression in cells and PTHrP (1–86) release into cell-conditioned culture medium were determined. These studies were carried out in MDA-MB-231 and S-30 cells maintained in culture in the presence or absence of E2. After 48 h of incubation, MDA-MB-231 cells incubated with FBS and SFBS and S-30 cells incubated with SFBS alone, SFBS plus 10–7 M E2, and SFBS plus 10–7 M E2 and 10 μM TAM were analyzed. RT-PCR using PTHrP-specific primers showed that PTHrP mRNA expression is higher in MDA-MB-231 cells compared with S-30 cells. When S-30 cells were incubated in the presence of SFBS, a significant increase in PTHrP expression was seen. The addition of 10–7 M E2 reduced this increase, and these effects were abrogated when S-30 cells were coincubated with 10–7 M E2 and 10 μM TAM (Fig. 2). Assessment of PTHrP protein levels by immunocytochemistry showed that PTHrP content was markedly higher in MDA-MB-231 cells compared with S-30 cells when cultures of both cells were incubated in medium containing FBS (Fig. 3A). In contrast, incubation of S-30 cells in the presence of SFBS resulted in significantly higher levels of PTHrP production that were similar to those seen in MDA-MB-231 cells. This increase in PTHrP content was inhibited by 35% when S-30 cells were incubated for 48 h in the presence of 10–7 M E2. The specificity of this E2-mediated effect was established by coincubation of these cells in SFBS containing E2 and TAM. Anti-E2 TAM was effective in reversing the ability of E2 inhibit PTHrP production by S-30 cells (Fig. 3A). Evaluation of conditioned culture medium of the MDA-MB-231 and S-30 cells collected under similar experimental conditions showed results for PTHrP release into conditioned medium, which were similar to those seen by immunocytochemical analysis (Fig. 3B).
FIG. 2. Effect of E2 on PTHrP mRNA expression in MDA-MB-231 and S-30 cells. MDA-MB-231 and S-30 cells were grown to 70% confluence in MEM containing FBS and were incubated overnight in serum-free conditions. Cells were then treated for 12 h in media containing FBS, SFBS, or SFBS with 10–7 E2 and 10 μM TAM. Levels of PTHrP and GAPDH mRNA were determined by RT-PCR as described in Materials and Methods. Results are expressed as the ratio of PTHrP to GAPDH mRNA. Bars represent the means ± SEM of three different experiments. Significant difference in the ratio from control MDA-MB-231 cells are shown by single asterisks (P < 0.05). Significant differences between PTHrP in S-30 cells grown in SFBS vs. S-30 cells grown in FBS are denoted by two asterisks (P < 0.05).
FIG. 3. Effect of E2 on the production of PTHrP by MDA-MB-231 and S-30 cells in vitro. MDA-MB-231 and S-30 cells were grown to confluence in Petri dishes in medium containing FBS, SFBS, or SFBS with E2 and TAM for 48 h. A, Cells from various groups were subjected to immunocytochemical analysis using an antibody directed against PTHrP (1–34). Negative control (NC) represents cells where primary anti-PTHrP antibody was omitted (upper panel). Immunocytochemical reaction was quantified as total gray area and plotted (lower panel). B, Conditioned culture medium of cells from various groups was collected and subjected to IRMA assay to determine the levels of PTHrP production. Results are expressed as picograms of PTHrP(1–86) per 106 cells. Bars are representative of three such experiments ± SEM. Significant differences in PTHrP production from MDA-MB-231 cells are shown by one asterisk, and from S-30 cells by two asterisks (P < 0.05).
The effect of E2 on PTHrP promoter activity was then evaluated using S-30 cells, which had shown a significant decrease in PTHrP production after treatment with E2. S-30 cells were transiently transfected with a vector containing the rat PTHrP promoter inserted upstream of a luciferase reporter gene (pGL3). The pGL3 vector without the PTHrP promoter was used as a control to establish basal levels of luciferase activity. Change in luciferase activity is represented as fold increase over basal promoter activity. Maximum luciferase activity was seen when S-30 cells were incubated in SFBS in the absence of E2. Treatment of these cells with FBS or 10–7 M E2 in SFBS resulted in an approximate 45% decrease in PTHrP promoter activity. In contrast, coincubation of S-30 cells with 10–7 M E2 and 10 μM TAM reversed this decrease in PTHrP promoter activity (Fig. 4).
FIG. 4. Effects of E2 on PTHrP promoter activity. S-30 breast cancer cells were grown to 50–70% confluency. Cells were transiently cotransfected with pGL3 vector containing the rat PTHrP promoter upstream of a luciferase reporter gene and with basic ?-gal vector. Cells were incubated in FBS for 24 h and then incubated in serum-free conditions for 24 h. Subsequently, cells were treated with media containing FBS, SFBS, or SFBS and E2 (10–7) alone or in combination with 10 μM TAM. Luciferase activity was quantified using a luminometer, and results were normalized for transfection efficiency as measured by ?-gal reaction. Values represent fold increase in luciferase activity, compared with control pGL3 vector containing no PTHrP promoter. Bars represent means ± SEM of three different experiments. Significant differences from activity in control S-30 cells cultured in FBS is shown by the asterisk (P < 0.05).
Effect of E2 on tumor growth in vivo
Orthotopic injection of MDA-MB-231 cells into mammary fat pad of 5-wk-old female BALB/c nu.nu mice results in the development of tumors that are palpable by wk 4 after tumor cell inoculation. After this time, these tumors continue to grow for up to 10 wk. Comparison of tumor volumes at weekly intervals after inoculation of animals with MDA-MB-231 cells or S-30 cells showed that tumor volume was significantly smaller after inoculation with S-30 cells compared with MDA-MB-231 cells and remained so throughout the course of these studies. At wk 10 after tumor cell inoculation, S-30 tumors were 66% smaller in volume compared with MDA-MB-231 tumors. To evaluate the effect of E2 ablation on tumor volume, S-30 cells were also inoculated into OVX animals. In these animals S-30 tumors were significantly larger in volume compared with S-30 tumors in non-OVX animals (Fig. 5).
FIG. 5. Effect of E2 on human breast cancer growth in a xenograft model of breast cancer. MDA-MB-231 and S-30 cells (106 cells with 10% Matrigel) were injected via the sc route into the mammary fat pad of female BALB/c nu.nu mice. Additionally, S-30 cells were injected into OVX female animals. Tumors were measured at weekly intervals, and tumor volume was determined according to the equation (length x width2). Results represent the mean ± SEM of two such experiments with 10 animals in each group. Significant difference in tumor volume between S-30 tumors and control MDA-MB-231 tumors is shown by one asterisk, and between S-30 tumors in OVX and non-OVX animals by two asterisks. (P < 0.05).
Effect of E2 on experimental skeletal metastases
The effect of E2 on the development of experimental skeletal metastases was next evaluated. In these studies, MDA-MB-231 cells were inoculated into the left ventricle of 5-wk-old female BALB/c nu.nu mice. Similarly, S-30 cells were inoculated into the left ventricle of normal or OVX female animals. All experimental animals were evaluated for the presence of skeletal metastases by x-ray analysis. In this model, radiological evidence of skeletal metastases in long bones is seen by 5 wk after tumor cell inoculation, and metastases continue to increase in terms of both lesion area and number for up to 10 wk, at which time these animals were killed for further analysis. Radiological analysis of all animals in different groups showed the presence of skeletal metastases in tibia and femur. In these long bones total lesion area was significantly smaller in animals inoculated with S-30 cells compared with animals receiving MDA-MB-231 cells. In contrast, when S-30 cells were inoculated into OVX animals, lesion area was increased and reached levels that were comparable to animals receiving MDA-MB-231 cells (Fig. 6). At the end of these studies (wk 10), long bones were removed and subjected to bone histological analysis. The presence of tumor cells was clearly seen in the bone marrow space where they caused destruction of both trabecular and cortical bone. Determination of tumor volume to tissue volume ratio showed a marked decrease in tumor burden in animals receiving S-30 cells compared with animals inoculated with MDA-MB-231 breast cancer cells. However this decrease in tumor burden was reversed when S-30 cells were inoculated into OVX animals (Fig. 7).
FIG. 6. Effect of E2 on skeletal metastases in a xenograft model of breast cancer. MDA-MB-231 and S-30 human breast cancer cells (106 cells) were inoculated into the left ventricle of female BALB/c nu.nu mice. Additionally, S-30 cells were also inoculated into OVX animals. After tumor cell inoculation, all animals were evaluated at different time points for the development of skeletal metastases using x-ray analysis. Representative radiographs of animals from various groups at wk 10 after tumor cell inoculations are shown (upper panels). Lesion areas of all radiographs were calculated and plotted (lower panel). Results are representative of the mean ± SEM of two experiments with 10 animals in each group. Significant difference in skeletal lesion areas between S-30 and control MDA-MB-231 are shown by one asterisk, and between S-30 cells inoculated into OVX vs. non-OVX animals by two asterisks. (P < 0.05).
FIG. 7. Bone histology and histomorphometric analysis of MDA-MB-231 and S-30 skeletal metastasis. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 cells via the intracardiac route. S-30 cells were also inoculated into OVX animals. At wk 10 after tumor cell inoculation, evidence of skeletal metastases was established by radiographic analysis. All animals were killed, and long bones were removed and subjected to bone histology. Tumor cells (T) filled the marrow cavity and there was an associated loss of trabecular bone (A). Histological analysis of tibia of mice from various groups was carried out to determine tumor volume to tissue volume ratio (TuV/TV) (lower panel). The single asterisk indicates a significant change in TuV/TV ratio in S-30 tumors compared with MDA-MB-231 tumors (P < 0.05). Double asterisks indicate a significant change intumor volume to bone volume ratio in S-30 tumors in OVX vs. S-30 tumors in non-OVX mice (P < 0.05).
Effect of E2 on tumor PTHrP production in vivo
Primary tumors from animals with orthotopic injection and long bones of experimental animals with intracardiac injection were then analyzed for PTHrP production by immunohistochemistry. Ten weeks after orthotopic injection, intense staining for PTHrP was seen in MDA-MB-231 and this positive staining for PTHrP was reduced by 80% in S-30 tumors. Analysis of S-30 tumors isolated from OVX animals showed that the level of PTHrP staining was markedly increased compared with non-OVX animals (Fig. 8). Ten weeks after intracardiac tumor cell inoculation, evaluation of long bones exhibiting radiological and histological evidence of skeletal metastases also showed that the levels of PTHrP production were markedly reduced in the skeletal metastases of control animals inoculated with S-30 cells, and this decrease in PTHrP production was abrogated in long bones of OVX animals inoculated with S-30 cells OVX (Fig. 9).
FIG. 8. Effect of E2 on PTHrP production in primary breast tumors in vivo. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 breast cancer cells. Additionally, S-30 cells were inoculated into OVX animals. Ten weeks after tumor cell inoculation, all animals were killed, and their primary tumors were removed and subjected to immunohistochemical analysis using a PTHrP (1–34) antibody. Negative control (NC) was performed by the omission of primary antibody specific to PTHrP (1–34). Representative photomicrographs (x20) of animals from each group are shown (upper panels). The immunohistochemical reaction was also quantified and plotted (lower panel). Results are representative ± SEM of two experiments where at least five tumors in each group were analyzed. A single asterisk represents a significant difference between PTHrP in S-30 tumors and in MDA-MB-231 tumors (P < 0.05), and double asterisks represent a significant difference between PTHrP in S-30 tumors in OVX mice and in non-OVX mice (P < 0.05).
FIG. 9. Effect of E2 on PTHrP production in skeletal metastases. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 breast cancer cells into left ventricle. Additionally, S-30 cells were injected into OVX animals. Ten weeks after tumor cell inoculation, all animals were killed, and their long bones showing radiological evidence of skeletal metastases were removed and subjected to immunohistochemical analysis using a PTHrP (1–34) antibody. Negative control (NC) was performed by the omission of primary antibody specific to PTHrP (1–34). Areas highlighting bone (B) are marked. Representative photomicrographs (x20) of animals from each group is shown (upper panels). The immunohistochemical reaction was quantified and plotted (lower panel). Results represent the means ± SEM of two experiments where at least five tumors in each group were used. A single asterisk represents a significant difference of PTHrP immunoreactivity in metastases of S-30 tumors and of negative control (NC) vs. MDA-MB-231 tumors (P < 0.05) in OVX mice vs. non-OVX mice (P < 0.05). Double asterisks represent a significant difference of PTHrP immunoreactivity in metastases of S-30 tumors in OVX vs. non-OVX mice (P < 0.05).
Discussion
Estrogens are required for survival and proliferation of normal hyperplasic and malignant breast epithelial cells (9). Estrogen, via its receptors, increases the growth of breast cancer cells in culture by increasing the number of G0/G1 cells entering the cell cycle; however, this may be indirect most likely by inducing autocrine, paracrine, and intracrine growth factors (35). The use of E2 ablation and anti-E2 therapy initially benefits early-stage breast cancer patients; however, the success of these approaches is largely dependent on the ER status of the malignant cells (36). Removal of endogenous E2, for example by the aromatase inhibitors and blocking of the ER with ER antagonists, prevents the growth of hormone-dependent breast cancer cells resulting in initial tumor mass reduction. However, the proliferative capacity of ER-negative tumor cell subpopulations is not inhibited (37). Advanced breast cancer progression may therefore occur when the primary tumor is composed primarily of ER-negative malignant cells that are no longer affected by E2 or antagonism deprivation. Thus, in the later stages of breast cancer, hormone therapy becomes ineffective, requiring the use of other treatment modalities.
MCF-7 cells that naturally express ER exhibit a significant increase in their rate of proliferation after treatment with E2. These findings are compatible with growth-promoting effects of estrogen in breast cancer (29). ER-negative, highly invasive, and metastatic MDA-MB-231 breast cancer cells exhibit a high proliferation rate both in vitro and in vivo. This is compatible with the known aggressiveness of ER-negative breast tumors. However, the stable transfection of MDA-MB-231 cells with a functional human ER to produce S-30 cells caused cellular proliferation to be significantly reduced in vitro (38). In previous studies, S-30 cells have been extensively characterized for the levels and function of ER expression by enzyme assay and ligand binding assays. These studies have shown that in S-30 cells levels of expression of functional ERs and their binding affinity are similar to that seen in MCF-7 cells that naturally express ER (29). However, E2 treatment markedly decreased the expression of ER in MCF-7 cells (39), essentially converting it into an ER-negative cell, whereas this decrease is much less marked in S-30 cells in which ER expression is controlled by a constitutive promoter. Consequently the major effect of E2 may be to decrease cell proliferation via the ER; however, reduction of ER levels by E2 whether by transcriptional and/or posttranscriptional mechanisms may allow the cell to revert to the proliferative status of ER-negative cells (39). Athymic nude mice showed that S-30 cells in vivo also developed significantly smaller tumors and exhibited a later onset of tumor development compared with animals receiving hormone-insensitive MDA-MB-231 cells. Therefore, E2 was found to inhibit cell growth as opposed to promoting the growth of malignant breast cells in vivo as well as in vitro. Persistent ER positivity with E2 responsiveness may therefore be advantageous rather than deleterious in the development and progression of breast cancer.
We also demonstrated here, for the first time, the apparent paradox of sex steroid inhibition of cell growth has also been observed by us previously in a androgen-receptor-positive prostate cancer cell system PC-3T cells in which androgens were found to inhibit cell proliferation rather than stimulate it (28). Thus, PC-3T cells, which were stably transfected with a functional androgen receptor, grew more slowly that the parental PC-3 cells that lack an androgen receptor and are androgen nonresponsive. Androgens were also found by us to decrease PTHrP gene transcription in those studies.
Higher PTHrP expression has been associated with later-stage breast cancer, including metastatic breast cancer, compared with early-stage nonmetastatic breast cancer or with normal breast cells (40, 41). In our studies, S-30 cells were found to express markedly less PTHrP than MDA-MB-231 cells. The specific effect of E2 on PTHrP production was confirmed by incubating S-30 cells in the absence of E2, thus enhancing PTHrP levels, and then demonstrating reduction of these levels when E2 was added. Finally, the E2 effects were reversed with the E2 antagonist TAM. The mechanism of this effect appears to be at least in part direct inhibition of PTHrP gene transcription by diminishing promoter activity. The in vitro inhibiting effects of E2 on PTHrP production were also confirmed in vivo when increased PTHrP levels were observed in tumors of OVX animals inoculated iv with S-30 cells compared with tumors of non-OVX animals inoculated with S-30 cells.
PTHrP is a cytokine that may stimulate cell growth and inhibit apoptosis in different cell types (42, 43, 44, 45, 46) including breast cancer cells (22, 47, 48), and elevated circulating concentrations of PTHrP have been shown to be a prognostic marker for poor survival in cancer (49). Previous studies with MDA-MB-231 cells have demonstrated increased PTHrP production in skeletal metastasis as a cause for increased osteolysis and, similarly, we have implicated PTHrP in local osteolysis in prostate cancer-associated skeletal metastases (50). Increased TGF? release from bone was found to be stimulatory for PTHrP in the studies with MDA-MB-231 cells (51, 52). This is consistent with our previous findings that a variety of growth factors including epidermal growth factor and IGF-I can stimulate PTHrP gene expression (53, 54). In view of the fact that bone is a major reservoir for growth factors such as TGF? and IGF-I, release of these factors during PTHrP-induced osteolysis can further increase PTHrP production by invading tumors as has been previously noted (52). In our studies, reduced PTHrP levels and reduced osteolysis were observed in conditions where E2 action could prevail, i.e. in non-OVX animals inoculated with ER-expressing S-30 cells, whereas reduction of E2 effects in OVX animals was associated with increased PTHrP and increased osteolysis. However, in view of the fact that other growth factors such as TGF and TGF? may be regulated by E2 activation of ER in S-30 cells, the role of such factors in comodulating PTHrP effects on bone osteolysis cannot be ruled out (29, 55).
Our combined in vitro and in vivo data therefore support a role for E2 in regulating PTHrP reduction, which in turn, can mediate breast cancer effects on osteolysis. Our findings may have important implications for the role of PTHrP in breast cancer progression and for new approaches in the future treatment of E2-dependent and E2-independent breast cancer. Hormone therapy in combination with other anticancer therapy may provide the best strategy for the blockade of breast cancer progression and mortality associated with this prevalent disorder.
Acknowledgments
The authors thank Dr. Ida Khalili (Department of Radiology, McGill University Health Centre) for carefully reviewing all x-rays.
References
Margolese TG, Hortobagy GN, Buchholz TA 2003 Neoplasms of the breast. In: Kufe DW, Pollock PE, Weichselbaum RR, Bast Jr RC, Gansler TS, Holland JF, Frei III E, eds. Cancer medicine. 6th ed. Hamilton, Ontario, Canada: BC Decker Inc.; 1879–1970
Bernstein L 2002 Epidemiology of endocrine-related risk factors for breast cancer. J Mammary Gland Biol Neoplasia 7:3–15
Mundy GR 2002 Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer 2:584–593
Mawson A, Lai A, Carroll JS, Sergio CM, Mitchell CJ, Sarcevic B 2005 Estrogen and insulin/IGF-1 cooperatively stimulate cell cycle progression in MCF-7 breast cancer cells through differential regulation of c-Myc and cyclin D1. Mol Cell Endocrinol 229:161–173
Ali S, Coombes RC 2000 Estrogen receptor in human breast cancer: occurrence and significance. J Mammary Gland Biol Neoplasia 5:271–281
Lippman ME, Golan G, Monaco ME, Pinkus L, Engel E 1976 Model systems for the study of estrogen action in tissue culture. J Steroid Biochem 7:1045–1051
Jordan VC 1995 Studies on the estrogen receptor in breast cancer—20 years as a target for the treatment and prevention of cancer. Breast Cancer Res Treat 36:267–285
Diaz LK, Sneige N 2005 Estrogen receptor analysis for breast cancer: current issues and keys to increasing testing accuracy. Adv Anat Pathol 12:10–19
Platet N, Cathiard AM, Gleizes M, Garcia M 2004 Estrogens and their receptors in breast cancer progression: a dual role in cancer proliferation and invasion. Crit Rev Oncol Hematol 51:55–67
Kurosumi M 2003 Significance of immunohistochemical assessment of steroid hormone receptor status for breast cancer patients. Breast Cancer 10:97–104
Saj S, Omoto Y, Shimizu C, Horiguchi S, Watanabe T, Funata N, Hayash S, Gustafsson JA, Toi M 2002 Clinical impact of assay of estrogen receptor betacx in breast cancer. Breast Cancer 9:303–307
Omoto Y, Inoue S, Ogawa S, Toyama T, Yamashita H, Muramatsu M, Kobayashi S, Iwase H 2001 Clinical value of the wild-type estrogen receptor ? expression in breast cancer. Cancer Lett 163:207–212
Yoneda T 1998 Cellular and molecular mechanisms of breast and prostate cancer metastasis bone. Eur J Cancer 34:240–245
Pluijm G, Lowik C, Papapoulos S 2000 Tumour progression and angiogenesis in bone metastasis from breast cancer new approaches to an old problem. Cancer Treat Rev 26:11–27
Reddi AH, Roodman D, Freeman C, Mohla S 2003 Mechanisms of tumor metastasis to the bone: challenges and opportunities. J Bone Miner Res 18:190–194
Coleman RE 2001 Metastatic bone disease: clinical features, pathophysiology and treatment strategies. Cancer Treat Rev 27:165–176
Chirgwin JM, Guise TA 2000 Molecular mechanisms of tumor-bone interactions in osteolytic metastases. Crit Rev Eukaryot Gene Expr 10:159–178
Burtis WJ, Wu T, Bunch C, Wysolmerski JJ, Insogna KL, Weir EC, Broadus AE, Stev AF 1987 Identification of a novel 17,000-dalton parathyroid hormone-like adenyl cyclase-stimulating protein from a tumor associated with humoral hypercalcemia of malignancy. J Biol Chem 262:7151–7156
Moseley JM, Kubota M, Diefenbach-Jagger HD, Wettenhall, REH, Kemp BE, Suva LJ, Rodda CP, Ebeling, PR, Hudson, PJ, Zajac JD, Martin TJ 1987 Parathyroid hormone-related protein purified from a human lung cancer cell line. Proc Nat Acad Sci USA 84:5048–5052
Juppner H, Abou-Samra AB, Freeman M, Kong XF, Schipani E, Richards J, Kolakowski Jr LF, Hock J, Potts Jr JT, Kronenberg HM 1991 A G protein-linked receptor for parathyroid hormone and parathyroid hormone-related peptide. Science 254:1024–1026
Kaiser SM, Sebag M, Rhim JS, Kremer R, Goltzman D 1994 Antisense-mediated inhibition of parathyroid hormone-related peptide production in a keratinocyte cell line impedes differentiation. Mol Endocrinol 8:139–147
Falzon M, Du P 2000 Enhanced growth of MCF-7 breast cancer cells overexpressing parathyroid hormone-related peptide. Endocrinology 141:1882–1892
Falzon M 1996 Serum stimulation of parathyroid hormone-related peptide gene expression in ROS 17/2.8 osteosarcoma cells through transcriptional and posttranscriptional mechanisms. Endocrinology 137:3681–3688
Holt EH, Lu C, Dreyer BE, Dannies PS, Broadus AE 1994 Regulation of parathyroid hormone-related peptide gene expression by estrogen in GH4C1 rat pituitary cells has the pattern of a primary response gene. J Neurochem 62:1239–1246
Sebag M, Henderson J, Goltzman D, Kremer R 1994 Regulation of parathyroid hormone-related peptide production in normal human mammary epithelial cells in vitro. Am J Physiol 267:723–730
Liu B, Goltzman D, Rabbani SA 1993 Regulation of parathyroid hormone-related peptide production in vitro the rat hypercalcemic Leydig cell tumor H-500. Endocrinology 132:1658–1664
Haq M, Kremer R, Goltzman D, Rabbani SA 1993 A vitamin D analogue (EB1089) inhibits parathyroid hormone-related peptide production and prevents the development of malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416–2422
Pizzi H, Gladu J, Carpio L, Miao D, Goltzman D, Rabbani SA 2003 Androgen regulation of parathyroid hormone-related peptide production human prostate cancer cells. Endocrinology 144:858–867
Jiang SY, Jordan VC 1992 Growth regulation of estrogen receptor-negative breast cancer cells transfect with complementary DNAs for estrogen receptor. J Natl Cancer Inst 84:580–591
Guo YJ, Higazi AAR, Arakelian A, Sachais BS, Cines D, Goldfarb RH, Jones TR, Kawan H, Mazar AP, Rabbani SA 2000 A peptide derived from the non-receptor binding region of urokinase plasminogen activator (uPA) inhibits tumor progression and angiogenesis and induces tumor cell death in vivo. FASEB J 14:1400–1410
Levenson AS, Kwaan HC, Svoboda KM, Weiss IM, Sakurai S, Jordan VC 1998 Oestradiol regulation of the components of the plasminogen-plasmin system in MDA-MB-231 human breast cancer cells stably expressing the oestrogen receptor. Br J Cancer 78:88–95
El Abdaimi K, Dion N, Papavasiliou V, Cardinal PE, Binderup L, Goltzman D, Sainte Marie LG, Kremer R 2000 The vitamin D analogue EB1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Research 60:4412–4418
Chen G, Shukeir N, Potti A, Sircar K, Aprikian A, Goltzman D, Rabbani SA 2004 Up-regulation of Wnt-1 and ?-catenin production in patients with advanced metastatic prostate carcinoma. Potential pathogenetic and prognostic implications. Cancer 101:1345–1356
Duivenvoorden WC, Popovic SV, Lhotak S, Seidlitz E, Hirte HW, Tozer RG, Singh G 2002 Doxycycline decreases tumor burden in a bone metastasis model of human breast cancer. Cancer Res 62:1588–1591
Dickson RB, Lippman ME 1995 Growth factors in breast cancer. Endocr Rev 16:559–589
Johnston SRD, Dowsett M 2003 Aromatase inhibitors for breast cancer: lessons from the laboratory. Nat Rev 3:821–831
Miller WR 2004 Biological rationale for endocrine therapy in breast cancer. Best Pract Res Clin Endocrinol Metab 18:1–32
Shao Z, Jiang M, Yu L, Han Q, Shen Z 1997 Estrogen receptor-negative breast cancer cells transfected with estrogen receptor exhibit decreased tumour progression and sensitivity to growth inhibition by estrogen. Chin Med Sci J 12:11–14
Saceda M, Lippman ME, Lindsay RK, Puente M, Martin MB 1989 Role of an estrogen receptor-dependent mechanism in the regulation of estrogen receptor mRNA in MCF-7 cells. Mol Endocrinol 3:1782–1787
Southby J, Kissin MW, Danks JA, Hayman JA, Moseley JM, Henderson MA, Bennett RC, Martin TJ 1990 Immunohistochemical localization of parathyroid hormone-related protein in human breast cancer. Cancer Res 50:7710–7716
Kkohno N, Kitazawa, S, Fukase M, Sakoda Y, Kanbara Y, Furuya Y, Oha O, Ishikawa Y, Saitoh Y 1994 The expression of parathyroid hormone-related protein in human breast cancer with skeletal metastases. Surg Today 24:215–220
Kaiser SM, Laneuville P, Bernier SM, Rhim JS, Kremer R, Goltzman D 1992 Enhanced growth of a human keratinocyte cell line induced by antisense RNA for parathyroid hormone-related peptide. J Biol Chem 267:13623–13628
Massfelder T, Dann P, Wu TL, Vasavada R, Helwig J-J, Stewart AF 1997 Opposing mitogenic and antimitogenic actions of parathyroid hormone-related peptide protein in vascular smooth muscle cells: a critical role for nuclear targeting. Proc Natl Acad Sci USA 169:33–41
Henderson JE, He B, Goltzman D, Karaplis AC 1996 Constitutive expression of parathyroid hormone-related peptide (PTHrP) stimulates growth and inhibits differentiation of CFK2 chondrocytes. J Cell Physiol 169:33–41
Iwamura M, Abrahamsson PA, Foss KA, Wu G, Cockett AT, Deftos LJ 1994 Parathyroid hormone-related protein: a potential autocrine growth regulator in human prostate cancer cell lines. Urology 43:675–679
Burton PBJ, Moniz C, Knight DE 1990 Parathyroid hormone related peptide can function as an autocrine growth factor in human renal cell carcinoma. Res Commun 167:1134–1138
Luparello C, Birch MA, Gallagher JA, Burtis WJ 1997 Clonal heterogeneity of the growth and invasive response of a human breast carcinoma cell line to parathyroid hormone-related peptide fragments. Carcinogenesis 18:23–29
Sepulveda VAT, Shen X, Falzon M 2002 Intracine PTHrP protects against serum starvation-induced apoptosis and regulates the cell cycle in MCF-7 breast cancer cells. Endocrinology 143:596–606
Troung NU, de Edwardes MD, Papavasiliou V, Goltzman D, Kremer R 2003 Parathyroid hormone-related peptide and survival of patients with cancer and hypercalcemia. Am J Med 115:115–121
Rabbani SA, Gladu J, Harakidas P, Jamison B, Goltzman D 1999 Over-production of parathyroid hormone-related peptide results in increased osteolytic skeletal metastasis by prostate cancer cell in vivo. Int J Cancer 80:257–264
Yin J, Selander K, Chirgwin JM, Dallas M, Grubbs BG, Wieser R, Massague J, Mundy GR, Guise TA 1999 TGF-? signaling blockade inhibits PTHrP secretion by breast cancer cells and bone metastases development. J Clin Invest 103:197–206
Kakonen SM, Selander K, Chirgwin JM, Yin JJ, Burns S, Rankin WA, Grubbs BG, Dallas M, Cui Y, Guise TA 2002 Transforming growth factor-? stimulates parathyroid hormone-related protein and osteolytic metastases via Smad and mitrogen-activated protein kinase signaling pathways. J Biol Chem 227:24571–24578
Goltzman D 2001 Extending preclinical models of skeletal manifestations of malignancy to the clinical setting. Semin Oncol 28:9–14
Rabbani SA 2000 Molecular mechanism of action of parathyroid hormone related peptide in hypercalcemia of malignancy: therapeutic strategies. Int J Oncol 16:197–206< href="/cgi/external_ref?access_num=10601566&link_type=MED" >
Jeng MH, Jiang SY, Jordan CV 1994 Paradoxical regulation of estrogen-dependent growth factor gene expression in estrogen receptor (ER)-negative human breast cancer cells stably expressing ER. Cancer Lett 82:123–128(S. A. Rabbani, P. Khalili)
Address all correspondence and requests for reprints to: Shafaat A. Rabbani, M.D., McGill University Health Centre, 687 Pine Avenue West, Room H4.61, Montreal, Quebec, Canada H3A 1A1. E-mail: shafaat.rabbani@mcgill.ca.
Abstract
We evaluated the capacity of estradiol (E2) to regulate PTHrP production, cell growth, tumor growth, and metastasis to the skeleton in breast cancer. In estrogen receptor (ER)-negative human breast cancer cells, MDA-MB-231, and cells transfected with full-length cDNA encoding ER (S-30), E2 caused a marked decrease in cell growth and PTHrP production, effects that were abrogated by anti-E2 tamoxifen. E2 also inhibited PTHrP promoter activity in S-30 cells. For in vivo studies, MDA-MB-231 and S-30 cells were inoculated into the mammary fat pad of female BALB/c nu.nu mice. Animals receiving S-30 cells developed tumors of significantly smaller volume compared with MDA-MB-231 tumor-bearing animals. This change in tumor volume was reversed when S-30 cells were inoculated into ovariectomized (OVX) hosts. Inoculation of MDA-MB-231 cells into the left ventricle resulted in the development of lesions in femora and tibia as determined by x-ray analysis. In contrast, these lesions were significantly smaller in volume and number in animals inoculated with S-30, and this lower incidence was reversed in OVX animals. Bone histological analysis showed that the tumor volume to tissue volume ratio was comparable with that seen by x-ray. Immunohistochemical analysis showed that PTHrP production was inhibited in S-30 group and restored to levels comparable to that seen in MDA-MB-231 tumor-bearing animals when S-30 cells were inoculated in OVX animals. Collectively these studies show that E2 production is inversely correlated with PTHrP production and that the growth-promoting effect of PTHrP has a direct impact on tumor growth at both nonskeletal and skeletal sites.
Introduction
BREAST CANCER IS the most frequent malignancy occurring in women and results in a high rate of morbidity and mortality (1, 2). It is unique among malignancies due to the high rate of developing osteolytic skeletal metastases (3). It is now well established that estradiol (E2) plays an important role in mammary development by binding to nuclear estrogen receptors (ER). These actions are mediated by high-affinity receptors that form a receptor complex that can associate with E2 response elements located on several genes to alter cell growth and differentiation (2, 4, 5, 6). Apart form its physiological effects, E2 has also been associated with malignant transformation and induction of tumor cell proliferation (4, 5). The main source of E2 in females are the ovaries, and blockage of these tumor-promoting effects with the anti-E2 tamoxifen (TAM) is one of the most commonly used forms of hormone therapy for early-stage hormone-responsive breast cancer (7). In the initial assessment of patients with breast cancer, determination of the ER status serves as a major criterion for designing the therapeutic strategy in view of the fact that the effectiveness of hormone therapy is dependent upon the expression of functional ER by tumor cells (8, 9). Lack of ER-positive tumor cells is considered to denote a more aggressive tumor phenotype and a poorer prognosis (10). Whereas two ER isoforms (ER and ER?) have been identified, ER is the main form determining hormone dependence because ER? expression is generally found to be low in the majority of breast and ovarian cancers (11, 12). Despite the requirement for ER for an estrogen response, only 60% of ER-positive breast cancer patients respond to hormone therapy with TAM (7). In late stages, when tumor cells are ER negative, they are highly invasive and are often associated with osteolytic skeletal metastases (13, 14, 15). This invasive transformation of tumor cells is achieved via selective induction and production of growth factors, cytokines, and proteases that are believed to promote tumor cell proliferation, invasion, and metastasis to different organs. Breast cancer per se has a particular predilection to metastasize to the skeleton (13, 14, 15). Once these skeletal metastases are established, they result in bone pain due to nerve compression and pathological fractures. The presence of breast tumor cells in the skeleton results in the release of growth factors, cytokines, and proteases from the bone microenvironment that promotes breast cancer-associated osteolysis and results in hypercalcemia of malignancy in at least 15–20% of patients (13, 14, 15, 16, 17).
Whereas a large number of factors have been implicated in promoting hypercalcemia of malignancy, PTHrP has now been shown to be the major pathogenic factor in this syndrome (17, 18, 19). PTHrP appears to exert its actions by binding to a G protein-coupled receptor (type I PTH/PTHrP receptor) that also interacts with the amino-terminal region of PTH (20). PTHrP is produced in a number of malignancies, including breast cancer where it can modulate cell growth and differentiation (21, 22). Since the isolation and characterization of PTHrP, a number of growth factors, cytokines, and steroids have been shown to regulate PTHrP production (23, 24, 25). In a series of studies, we have shown the ability of fetal bovine serum and growth factors to induce PTHrP production, whereas dexamethasone, vitamin D, and its low calcemic analog EB1089 could inhibit PTHrP production (26, 27). More recently, we have shown that androgen was able to decrease PTHrP production by human prostate cancer cells in vitro and to decrease tumor growth in vivo (28).
In the current study, we assessed the ability of E2 to regulate PTHrP production in MDA-MB-231 human breast cancer cells that are E2 insensitive due to the silencing of ER gene transcription and in MDA-MB-231 cells transfected with a full-length cDNA encoding human ER (S-30) (29). Previous studies have shown that in S-30 cells, levels of ER expression and E2 binding affinity are similar to ER expression and ligand binding affinity in MCF-7 human breast cancer cells, which naturally express ER (29). Nevertheless, S-30 cells and MCF-7 cells behave differently with respect to responsiveness to E2 (29). Therefore, we compared the effect of E2 on S-30 and MCF-7 cell growth in vitro and examined the effect of E2 on S-30 cells tumor growth and metastasis to the skeleton in nude mice in vivo using a xenograft model of human breast cancer.
Materials and Methods
Cells and cell culture
MCF-7 and MD-MB-231 human breast cancer cells were obtained from American Type Tissue Culture Collection (Rockville, MD) and maintained in MEM (Life Technologies, Inc., Rockville, MD) supplemented with 10% FBS, 100 U/ml of penicillin-streptomycin sulfate, and 0.2% gentamycin as described previously (29, 30). MDA-MB-231 cells transfected with a functional ER (S-30 cells) were kindly provided by Dr. C. Jordan (Robert H. Lurie Comprehensive Cancer Center, Feinberg School of Medicine, Northwestern University, Chicago, IL) and were maintained in MEM under selection for neomycin resistance with geneticin (G418) (Life Technologies, Inc.) as described (29, 31). MCF-7 cells and S-30 cells have been demonstrated to express approximately 500 fmol of ER per milligram of protein and 450 fmol of ER per milligram of protein, respectively, and to display similar binding affinities (Kd equal to 0.2 nM) (29).
Cell proliferation assay
MCF-7, MDA-MB-231, and S-30 cells were plated in triplicate at a density of 10,000 cells per well in 2 ml media in six-well plates. Cells were grown in FBS or phenol-red-free charcoal-stripped FBS (SFBS) (Invitrogen, Burlington, Ontario, Canada) depleted of E2. Where indicated, cells in stripped serum were treated with either 10–7–10–10 M E2 (Sigma-Aldrich, Oakville, Ontario, Canada) alone or in combination with 10 μM anti-E2 TAM (Sigma-Aldrich). Cells were trypsinized and counted at the times indicated using a Coulter counter (model ZF; Coulter Electronics, Harpenden, Hertfordshire, UK). Cell culture medium was replenished every second day.
RT-PCR and luciferase and ?-galactosidase (?-gal) assays
Total RNA was extracted from MDA-MB-231 and S-30 cells by a single-step method using TRIzol reagent (Invitrogen). The following primers were used to amplify PTHrP: forward 5'-GAGCACCAGCTACTGCATGACAAG-3' and reverse 5'-TGAGTTAGGTATCTGCCCTCGT-3' as previously described (28). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers were used as a control. The amplified PCR product was fractionated on a 1.3% agarose gel and visualized by ethidium bromide staining. Band intensities for PTHrP and GAPDH were quantified using densitometric software (Quantity One; Bio-Rad Laboratories, Inc., Mississauga, Ontario, Canada).
For PTHrP promoter assays, S-30 cells (0.8 x 106) were plated on 35-mm dishes a day before transfection. Transfection was carried out using the lipofectamine method (Invitrogen) (28). The cells were cotransfected with 1 μg pGL3 vector, into which had been cloned the region representing –1954 to +179 of the rat PTHrP promoter upstream of the luciferase reporter gene, and 1 μg pCMV-?-gal reporter plasmid to determine transfection efficiency. The transfected cells were washed the next day and allowed to grow for 24 h in medium containing 10% FBS. After cells were serum deprived for 24 h, they were treated with medium containing FBS, SFBS, or SFBS plus 10–7 E2 alone or in combination with 10 μM anti-estrogen TAM. The cells were harvested with 1x reporter lysis buffer (Promega Corp., Nepean, Ontario, Canada), and luciferase activity in the supernatants was then analyzed as previously described by us using a luminometer (Monolight 2010; BD Biosciences, San Jose, CA). For ?-gal activity, 50-μl lysates were mixed with 200 μl ?-gal assay buffer (24 μM Na2HPO4, 16 μM NaH2PO4, 4 μM KCl, 400 nM MgCl2, 20 μM ?-mercaptoethanol, and 0.3 mg ortho-nitrophenyl-?-D-galactopyranoside per sample) in a 96-well plate and incubated for 4 h at 37 C. The reaction was stopped by adding 0.5 ml of 1 M Na2CO2, and the absorbance was measured at 420 nm in a Vmax plate reader (Molecular Devices, Sunnyvale, CA). Activity was determined by comparison to a standard curve. Luciferase reporter activity in relative luminescence units was normalized to ?-gal activity as described (28). All luciferase assay readings were performed in triplicate and corrected for ?-gal expression levels in each cell population.
Immunoradiometric assay (IRMA)
PTHrP secreted in the conditioned media was measured using a noncompetitive, two-site IRMA (Diagnostic Systems Laboratories, Webster, TX). PTHrP was recognized by an NH2-terminal reactive antibody [raised against PTHrP (1–34)] immobilized on a solid phase (capture antibody), and detected and quantitated with a second I125-radiolabeled COOH-terminal reactive signal antibody [raised against PTHrP (47–86)]. MDA-MB-231 and S-30 cells were plated at a density of 106 cells in 10 ml in 60-mm plates. At 70% confluence, cells were deprived of serum for 24 h and then treated in the presence or absence of E2 for 12 h. The conditioned medium was collected and the final cell number was determined using a Coulter counter (Coulter Electronics). The assay was carried out as per the manufacturer’s specifications. The reaction mixture was incubated at room temperature for 20 h, washed with deionized water, and the radioactivity from each sample was counted on a -counter for 1 min. The detection limit of the assay was 0.3 pmol/liter (3.0 pg/ml).
Animal protocols
For xenograft studies, 5-wk-old BALB/c nu.nu female mice were obtained from Charles River Inc. (St. Constant, Quebec, Canada). Before inoculation, MDA-MB-231 cells grown in serum containing culture medium were washed with Hank’s balanced buffer and centrifuged at 1500 rpm for 5 min. Cell pellets (5 x 105 cells/mouse) were resuspended in 100 μl of Matrigel (Becton Dickinson Labware, Mississauga, Ontario, Canada) and saline mixture (20% Matrigel) and injected into the mammary fat pads of the mice. All animals were numbered and kept separately in a temperature-controlled room on a 12-h light/12-h dark schedule with food and water ad libitum. Five-week-old (15–20 g) female BALB/c nu.nu mice (Charles River) were used throughout the course of the studies. For intravascular tumor cell inoculation, normal and ovariectomized (OVX) animals were anesthetized by xylazine/ketamine mixture and 1 x 105 M DA-MB-231 or S-30 cells were injected in 0.1 ml of cell suspension into the left ventricle of the heart using a 26-gauge needle (30, 32, 33, 34). All animal studies were approved by the institutional review committee.
Radiological analysis
High-resolution whole-body radiographs of animals were obtained under ketamine/xylazine anesthesia with the animals placed in a prone position against films (18 x 24 cm; Mamoray Screens, AGFA, Mortsel, Belgium) and exposed to x-ray at 25 kV for 5 sec using a Mammo Diagnost UC (Philips, Hamburg, Germany). X-ray films were processed in a Curix Compact processor (AGFA). The area of osteolytic bone metastasis was determined in all of the control and experimental groups of animals by careful evaluation of radiographs at the end of these studies. The analyses were carried out in both femora and tibia of all animals using an image analysis system after capture of radiographic prints. Measurements were taken with the aid of a digitizing tablet connected to an IBM-compatible computer (32, 33, 34). All radiographs were carefully evaluated by at least three investigators, including one radiologist, all of whom were blinded to experimental protocols.
Immunocytochemistry and immunohistochemistry
MDA-MB-231 and S-30 cells, cultured in 24-well plates, were stained immunocytochemically for PTHrP as previously described (28) using the avidin-biotin-peroxidase complex technique. Paraffin embedded tumor samples were cut into 5-μm-thick sections for immunohistochemical analysis. Immunohistochemical staining for PTHrP was performed as previously described using the avidin-biotin-peroxidase complex method (28). Briefly, the sections were dewaxed in xylene and rehydrated through a series of ethanol to water gradients. The sections were incubated in 1% normal goat serum (Vector Laboratories Inc., Burlingame, CA) for 30 min at room temperature before treatment with the primary antibody [rabbit polyclonal antiserum against PTHrP (1–34) from rabbit] at 1:200 dilution overnight at 4 C. Biotinylated goat antirabbit IgG (Vector Laboratories Inc.) was used as the secondary antibody at 1:200 for 30 min at room temperature. The slides were treated with Vectastain ABC-AP kit (Vector Laboratories Inc.), diluted 1:200 for 30 min at room temperature, and subsequently developed with Fast Red TR/Naphthol AS-MX phosphate (Sigma-Aldrich) containing 1 mM levamisole for 10–15 min. The slides were then counterstained with hematoxylin (Fisher Scientific Ltd, Nepean, Ontario, Canada) and mounted with Kaiser’s glycerol jelly. All sections were washed with Tris buffer (pH 7.6) three times for 10 min each after each step. For negative control sections, the primary antibody was omitted.
Images of stained sections were photographed with a Leica digital camera and processed using BioQuant image analysis software, version 6.50.10 (BioQuant Image Analysis Corp., Nashville, TN). The threshold was set by determining the positive staining of control sections and was used to automatically analyze all recorded images of all samples that had been stained in the same session under identical conditions. The area of immunohistochemical stained regions was calculated automatically by the software in each microscopic field. Pixel counts of the immunoreaction product were calculated automatically and were expressed as total density of the integrated immunostaining over a given area. This reflects the relative amount of PTHrP detected (33).
Histological analysis
At the end of these studies, radiologically affected and unaffected long bones were excised and fixed in 10% neutral-buffered formalin for 2 d. These bones were decalcified in 10% EDTA in PBS, pH 7.4, with constant agitation until deemed clear by the ammonium oxalate end point test (32, 33, 34). After decalcification, these specimens were dehydrated through graded ethanol and cleared in methyl salicylate before subjecting to paraffin infiltration. Sections of 5-μm thickness were then embedded in formalin and stained with hematoxylin and eosin (32, 33, 34). To reduce possible variation in analysis, four to six levels of each bone were cut, and each level was assessed to get to the portion of bone with maximum tumor burden, which was considered as a representative section. Histomorphometric measurement of total tumor volume was performed in representative sections of both tibias of the animals inoculated with MDA-MB-231 and S-30 cells. The analysis was performed on the region 1 mm below the growth plate using BioQuant image analysis software (32, 33, 34). The tumor volume and tissue volume were measured, and the percentage of tumor volume in total tissue volume was calculated (32, 33, 34).
Statistical analysis
All results are expressed as mean ± SEM. Statistical significance of the difference in numbers of osteolytic metastases and tumor volume between control and Herceptin-treated groups were analyzed by Mann-Whitney test for nonparametric samples and tumor progression was analyzed by Kaplan-Meier.
Results
Evaluation of the effect of E2 on breast cancer cell growth in vitro
The effect of E2 on breast cancer cell growth was evaluated in a cell proliferation assay in which the cell proliferation rate of MCF-7, MDA-MB-231, and S-30 cells were compared. MCF-7 cells that express endogenous ER showed an increase in their rate of proliferation with E2-containing FBS and a decrease in their rate of proliferation when cultured in phenol red-free MEM containing SFBS that is depleted of E2. Incubation of these cells in SFBS with E2 resulted in a marked increase in cell growth; the effects of E2 were reversed when MCF-7 cells were grown in the presence of E2 and anti-E2 TAM (Fig. 1A). These growth-promoting (MCF-7) and -inhibiting (MDA-MB-231) effects of E2 were dose dependent between 10–10 and 10–7 M (data not shown). When MDA-MB-231 cells were cultured in the presence of FBS they displayed a significantly greater rate of cell proliferation compared with S-30 cells (Fig. 1B). However, incubation of S-30 cells in the presence of culture medium containing SFBS allowed cell growth to increase. Because no such change in cell growth was seen in MDA-MB-21 cells transfected with a mock vector (data not shown), this change in cell growth was attributed to the presence of a functional ER in S-30 cells. Therefore, E2 differently altered cell growth in MCF-7 cells and in the ER-negative MDA-MB-231 cells in which a functional ER was restored. To establish the specificity of the growth-inhibiting effect of E2 in S-30 cells, these cells were incubated in SFBS medium in the presence of either E2 alone or E2 plus anti-E2 TAM. Medium containing SFBS and 10–7 M E2 resulted in a decrease in S-30 cell growth, and these inhibitory effects of E2 in S-30 cells were found to be dose dependent at 10–10–10–7 M as previously reported (29) (data not shown). The specificity of this estrogen-mediated effect was established when coincubation of S-30 cells in SFBS containing 10–7 M E2 and 10 μM anti-E2 TAM was able to alter this rate of S-30 cell growth to levels that were similar to those seen when S-30 cells were incubated in the presence of SFBS alone (Fig. 1B).
FIG. 1. Effect of E2 on human breast cancer cells growth. Human breast cancer cells MCF-7 (A) and MDA-MB-231 and S-30 (B) were plated in six-well plates (104 cells per plate) in MEM containing 10% FBS. S-30 cells were also incubated in phenol red-free MEM containing SFBS in the presence of 10–7 E2 or 10 μM anti-E2 TAM. At different time points, cells in various groups of triplicates were trypsinized and counted using a Coulter counter. Results are representative of three such experiments ± SEM. Significant differences in cell number from control MCF-7 or MDA-MB-231 cells are represented by one asterisk, and from S-30 cells in FBS by two asterisks (P < 0.05).
Effect of E2 on PTHrP production by breast cancer cells in vitro
To examine the effect of E2 on PTHrP production by breast cancer cells, levels of PTHRP mRNA expression by RT-PCR, immunocytochemical evidence of PTHrP (1–34) protein expression in cells and PTHrP (1–86) release into cell-conditioned culture medium were determined. These studies were carried out in MDA-MB-231 and S-30 cells maintained in culture in the presence or absence of E2. After 48 h of incubation, MDA-MB-231 cells incubated with FBS and SFBS and S-30 cells incubated with SFBS alone, SFBS plus 10–7 M E2, and SFBS plus 10–7 M E2 and 10 μM TAM were analyzed. RT-PCR using PTHrP-specific primers showed that PTHrP mRNA expression is higher in MDA-MB-231 cells compared with S-30 cells. When S-30 cells were incubated in the presence of SFBS, a significant increase in PTHrP expression was seen. The addition of 10–7 M E2 reduced this increase, and these effects were abrogated when S-30 cells were coincubated with 10–7 M E2 and 10 μM TAM (Fig. 2). Assessment of PTHrP protein levels by immunocytochemistry showed that PTHrP content was markedly higher in MDA-MB-231 cells compared with S-30 cells when cultures of both cells were incubated in medium containing FBS (Fig. 3A). In contrast, incubation of S-30 cells in the presence of SFBS resulted in significantly higher levels of PTHrP production that were similar to those seen in MDA-MB-231 cells. This increase in PTHrP content was inhibited by 35% when S-30 cells were incubated for 48 h in the presence of 10–7 M E2. The specificity of this E2-mediated effect was established by coincubation of these cells in SFBS containing E2 and TAM. Anti-E2 TAM was effective in reversing the ability of E2 inhibit PTHrP production by S-30 cells (Fig. 3A). Evaluation of conditioned culture medium of the MDA-MB-231 and S-30 cells collected under similar experimental conditions showed results for PTHrP release into conditioned medium, which were similar to those seen by immunocytochemical analysis (Fig. 3B).
FIG. 2. Effect of E2 on PTHrP mRNA expression in MDA-MB-231 and S-30 cells. MDA-MB-231 and S-30 cells were grown to 70% confluence in MEM containing FBS and were incubated overnight in serum-free conditions. Cells were then treated for 12 h in media containing FBS, SFBS, or SFBS with 10–7 E2 and 10 μM TAM. Levels of PTHrP and GAPDH mRNA were determined by RT-PCR as described in Materials and Methods. Results are expressed as the ratio of PTHrP to GAPDH mRNA. Bars represent the means ± SEM of three different experiments. Significant difference in the ratio from control MDA-MB-231 cells are shown by single asterisks (P < 0.05). Significant differences between PTHrP in S-30 cells grown in SFBS vs. S-30 cells grown in FBS are denoted by two asterisks (P < 0.05).
FIG. 3. Effect of E2 on the production of PTHrP by MDA-MB-231 and S-30 cells in vitro. MDA-MB-231 and S-30 cells were grown to confluence in Petri dishes in medium containing FBS, SFBS, or SFBS with E2 and TAM for 48 h. A, Cells from various groups were subjected to immunocytochemical analysis using an antibody directed against PTHrP (1–34). Negative control (NC) represents cells where primary anti-PTHrP antibody was omitted (upper panel). Immunocytochemical reaction was quantified as total gray area and plotted (lower panel). B, Conditioned culture medium of cells from various groups was collected and subjected to IRMA assay to determine the levels of PTHrP production. Results are expressed as picograms of PTHrP(1–86) per 106 cells. Bars are representative of three such experiments ± SEM. Significant differences in PTHrP production from MDA-MB-231 cells are shown by one asterisk, and from S-30 cells by two asterisks (P < 0.05).
The effect of E2 on PTHrP promoter activity was then evaluated using S-30 cells, which had shown a significant decrease in PTHrP production after treatment with E2. S-30 cells were transiently transfected with a vector containing the rat PTHrP promoter inserted upstream of a luciferase reporter gene (pGL3). The pGL3 vector without the PTHrP promoter was used as a control to establish basal levels of luciferase activity. Change in luciferase activity is represented as fold increase over basal promoter activity. Maximum luciferase activity was seen when S-30 cells were incubated in SFBS in the absence of E2. Treatment of these cells with FBS or 10–7 M E2 in SFBS resulted in an approximate 45% decrease in PTHrP promoter activity. In contrast, coincubation of S-30 cells with 10–7 M E2 and 10 μM TAM reversed this decrease in PTHrP promoter activity (Fig. 4).
FIG. 4. Effects of E2 on PTHrP promoter activity. S-30 breast cancer cells were grown to 50–70% confluency. Cells were transiently cotransfected with pGL3 vector containing the rat PTHrP promoter upstream of a luciferase reporter gene and with basic ?-gal vector. Cells were incubated in FBS for 24 h and then incubated in serum-free conditions for 24 h. Subsequently, cells were treated with media containing FBS, SFBS, or SFBS and E2 (10–7) alone or in combination with 10 μM TAM. Luciferase activity was quantified using a luminometer, and results were normalized for transfection efficiency as measured by ?-gal reaction. Values represent fold increase in luciferase activity, compared with control pGL3 vector containing no PTHrP promoter. Bars represent means ± SEM of three different experiments. Significant differences from activity in control S-30 cells cultured in FBS is shown by the asterisk (P < 0.05).
Effect of E2 on tumor growth in vivo
Orthotopic injection of MDA-MB-231 cells into mammary fat pad of 5-wk-old female BALB/c nu.nu mice results in the development of tumors that are palpable by wk 4 after tumor cell inoculation. After this time, these tumors continue to grow for up to 10 wk. Comparison of tumor volumes at weekly intervals after inoculation of animals with MDA-MB-231 cells or S-30 cells showed that tumor volume was significantly smaller after inoculation with S-30 cells compared with MDA-MB-231 cells and remained so throughout the course of these studies. At wk 10 after tumor cell inoculation, S-30 tumors were 66% smaller in volume compared with MDA-MB-231 tumors. To evaluate the effect of E2 ablation on tumor volume, S-30 cells were also inoculated into OVX animals. In these animals S-30 tumors were significantly larger in volume compared with S-30 tumors in non-OVX animals (Fig. 5).
FIG. 5. Effect of E2 on human breast cancer growth in a xenograft model of breast cancer. MDA-MB-231 and S-30 cells (106 cells with 10% Matrigel) were injected via the sc route into the mammary fat pad of female BALB/c nu.nu mice. Additionally, S-30 cells were injected into OVX female animals. Tumors were measured at weekly intervals, and tumor volume was determined according to the equation (length x width2). Results represent the mean ± SEM of two such experiments with 10 animals in each group. Significant difference in tumor volume between S-30 tumors and control MDA-MB-231 tumors is shown by one asterisk, and between S-30 tumors in OVX and non-OVX animals by two asterisks. (P < 0.05).
Effect of E2 on experimental skeletal metastases
The effect of E2 on the development of experimental skeletal metastases was next evaluated. In these studies, MDA-MB-231 cells were inoculated into the left ventricle of 5-wk-old female BALB/c nu.nu mice. Similarly, S-30 cells were inoculated into the left ventricle of normal or OVX female animals. All experimental animals were evaluated for the presence of skeletal metastases by x-ray analysis. In this model, radiological evidence of skeletal metastases in long bones is seen by 5 wk after tumor cell inoculation, and metastases continue to increase in terms of both lesion area and number for up to 10 wk, at which time these animals were killed for further analysis. Radiological analysis of all animals in different groups showed the presence of skeletal metastases in tibia and femur. In these long bones total lesion area was significantly smaller in animals inoculated with S-30 cells compared with animals receiving MDA-MB-231 cells. In contrast, when S-30 cells were inoculated into OVX animals, lesion area was increased and reached levels that were comparable to animals receiving MDA-MB-231 cells (Fig. 6). At the end of these studies (wk 10), long bones were removed and subjected to bone histological analysis. The presence of tumor cells was clearly seen in the bone marrow space where they caused destruction of both trabecular and cortical bone. Determination of tumor volume to tissue volume ratio showed a marked decrease in tumor burden in animals receiving S-30 cells compared with animals inoculated with MDA-MB-231 breast cancer cells. However this decrease in tumor burden was reversed when S-30 cells were inoculated into OVX animals (Fig. 7).
FIG. 6. Effect of E2 on skeletal metastases in a xenograft model of breast cancer. MDA-MB-231 and S-30 human breast cancer cells (106 cells) were inoculated into the left ventricle of female BALB/c nu.nu mice. Additionally, S-30 cells were also inoculated into OVX animals. After tumor cell inoculation, all animals were evaluated at different time points for the development of skeletal metastases using x-ray analysis. Representative radiographs of animals from various groups at wk 10 after tumor cell inoculations are shown (upper panels). Lesion areas of all radiographs were calculated and plotted (lower panel). Results are representative of the mean ± SEM of two experiments with 10 animals in each group. Significant difference in skeletal lesion areas between S-30 and control MDA-MB-231 are shown by one asterisk, and between S-30 cells inoculated into OVX vs. non-OVX animals by two asterisks. (P < 0.05).
FIG. 7. Bone histology and histomorphometric analysis of MDA-MB-231 and S-30 skeletal metastasis. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 cells via the intracardiac route. S-30 cells were also inoculated into OVX animals. At wk 10 after tumor cell inoculation, evidence of skeletal metastases was established by radiographic analysis. All animals were killed, and long bones were removed and subjected to bone histology. Tumor cells (T) filled the marrow cavity and there was an associated loss of trabecular bone (A). Histological analysis of tibia of mice from various groups was carried out to determine tumor volume to tissue volume ratio (TuV/TV) (lower panel). The single asterisk indicates a significant change in TuV/TV ratio in S-30 tumors compared with MDA-MB-231 tumors (P < 0.05). Double asterisks indicate a significant change intumor volume to bone volume ratio in S-30 tumors in OVX vs. S-30 tumors in non-OVX mice (P < 0.05).
Effect of E2 on tumor PTHrP production in vivo
Primary tumors from animals with orthotopic injection and long bones of experimental animals with intracardiac injection were then analyzed for PTHrP production by immunohistochemistry. Ten weeks after orthotopic injection, intense staining for PTHrP was seen in MDA-MB-231 and this positive staining for PTHrP was reduced by 80% in S-30 tumors. Analysis of S-30 tumors isolated from OVX animals showed that the level of PTHrP staining was markedly increased compared with non-OVX animals (Fig. 8). Ten weeks after intracardiac tumor cell inoculation, evaluation of long bones exhibiting radiological and histological evidence of skeletal metastases also showed that the levels of PTHrP production were markedly reduced in the skeletal metastases of control animals inoculated with S-30 cells, and this decrease in PTHrP production was abrogated in long bones of OVX animals inoculated with S-30 cells OVX (Fig. 9).
FIG. 8. Effect of E2 on PTHrP production in primary breast tumors in vivo. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 breast cancer cells. Additionally, S-30 cells were inoculated into OVX animals. Ten weeks after tumor cell inoculation, all animals were killed, and their primary tumors were removed and subjected to immunohistochemical analysis using a PTHrP (1–34) antibody. Negative control (NC) was performed by the omission of primary antibody specific to PTHrP (1–34). Representative photomicrographs (x20) of animals from each group are shown (upper panels). The immunohistochemical reaction was also quantified and plotted (lower panel). Results are representative ± SEM of two experiments where at least five tumors in each group were analyzed. A single asterisk represents a significant difference between PTHrP in S-30 tumors and in MDA-MB-231 tumors (P < 0.05), and double asterisks represent a significant difference between PTHrP in S-30 tumors in OVX mice and in non-OVX mice (P < 0.05).
FIG. 9. Effect of E2 on PTHrP production in skeletal metastases. Female BALB/c nu.nu mice were inoculated with MDA-MB-231 and S-30 breast cancer cells into left ventricle. Additionally, S-30 cells were injected into OVX animals. Ten weeks after tumor cell inoculation, all animals were killed, and their long bones showing radiological evidence of skeletal metastases were removed and subjected to immunohistochemical analysis using a PTHrP (1–34) antibody. Negative control (NC) was performed by the omission of primary antibody specific to PTHrP (1–34). Areas highlighting bone (B) are marked. Representative photomicrographs (x20) of animals from each group is shown (upper panels). The immunohistochemical reaction was quantified and plotted (lower panel). Results represent the means ± SEM of two experiments where at least five tumors in each group were used. A single asterisk represents a significant difference of PTHrP immunoreactivity in metastases of S-30 tumors and of negative control (NC) vs. MDA-MB-231 tumors (P < 0.05) in OVX mice vs. non-OVX mice (P < 0.05). Double asterisks represent a significant difference of PTHrP immunoreactivity in metastases of S-30 tumors in OVX vs. non-OVX mice (P < 0.05).
Discussion
Estrogens are required for survival and proliferation of normal hyperplasic and malignant breast epithelial cells (9). Estrogen, via its receptors, increases the growth of breast cancer cells in culture by increasing the number of G0/G1 cells entering the cell cycle; however, this may be indirect most likely by inducing autocrine, paracrine, and intracrine growth factors (35). The use of E2 ablation and anti-E2 therapy initially benefits early-stage breast cancer patients; however, the success of these approaches is largely dependent on the ER status of the malignant cells (36). Removal of endogenous E2, for example by the aromatase inhibitors and blocking of the ER with ER antagonists, prevents the growth of hormone-dependent breast cancer cells resulting in initial tumor mass reduction. However, the proliferative capacity of ER-negative tumor cell subpopulations is not inhibited (37). Advanced breast cancer progression may therefore occur when the primary tumor is composed primarily of ER-negative malignant cells that are no longer affected by E2 or antagonism deprivation. Thus, in the later stages of breast cancer, hormone therapy becomes ineffective, requiring the use of other treatment modalities.
MCF-7 cells that naturally express ER exhibit a significant increase in their rate of proliferation after treatment with E2. These findings are compatible with growth-promoting effects of estrogen in breast cancer (29). ER-negative, highly invasive, and metastatic MDA-MB-231 breast cancer cells exhibit a high proliferation rate both in vitro and in vivo. This is compatible with the known aggressiveness of ER-negative breast tumors. However, the stable transfection of MDA-MB-231 cells with a functional human ER to produce S-30 cells caused cellular proliferation to be significantly reduced in vitro (38). In previous studies, S-30 cells have been extensively characterized for the levels and function of ER expression by enzyme assay and ligand binding assays. These studies have shown that in S-30 cells levels of expression of functional ERs and their binding affinity are similar to that seen in MCF-7 cells that naturally express ER (29). However, E2 treatment markedly decreased the expression of ER in MCF-7 cells (39), essentially converting it into an ER-negative cell, whereas this decrease is much less marked in S-30 cells in which ER expression is controlled by a constitutive promoter. Consequently the major effect of E2 may be to decrease cell proliferation via the ER; however, reduction of ER levels by E2 whether by transcriptional and/or posttranscriptional mechanisms may allow the cell to revert to the proliferative status of ER-negative cells (39). Athymic nude mice showed that S-30 cells in vivo also developed significantly smaller tumors and exhibited a later onset of tumor development compared with animals receiving hormone-insensitive MDA-MB-231 cells. Therefore, E2 was found to inhibit cell growth as opposed to promoting the growth of malignant breast cells in vivo as well as in vitro. Persistent ER positivity with E2 responsiveness may therefore be advantageous rather than deleterious in the development and progression of breast cancer.
We also demonstrated here, for the first time, the apparent paradox of sex steroid inhibition of cell growth has also been observed by us previously in a androgen-receptor-positive prostate cancer cell system PC-3T cells in which androgens were found to inhibit cell proliferation rather than stimulate it (28). Thus, PC-3T cells, which were stably transfected with a functional androgen receptor, grew more slowly that the parental PC-3 cells that lack an androgen receptor and are androgen nonresponsive. Androgens were also found by us to decrease PTHrP gene transcription in those studies.
Higher PTHrP expression has been associated with later-stage breast cancer, including metastatic breast cancer, compared with early-stage nonmetastatic breast cancer or with normal breast cells (40, 41). In our studies, S-30 cells were found to express markedly less PTHrP than MDA-MB-231 cells. The specific effect of E2 on PTHrP production was confirmed by incubating S-30 cells in the absence of E2, thus enhancing PTHrP levels, and then demonstrating reduction of these levels when E2 was added. Finally, the E2 effects were reversed with the E2 antagonist TAM. The mechanism of this effect appears to be at least in part direct inhibition of PTHrP gene transcription by diminishing promoter activity. The in vitro inhibiting effects of E2 on PTHrP production were also confirmed in vivo when increased PTHrP levels were observed in tumors of OVX animals inoculated iv with S-30 cells compared with tumors of non-OVX animals inoculated with S-30 cells.
PTHrP is a cytokine that may stimulate cell growth and inhibit apoptosis in different cell types (42, 43, 44, 45, 46) including breast cancer cells (22, 47, 48), and elevated circulating concentrations of PTHrP have been shown to be a prognostic marker for poor survival in cancer (49). Previous studies with MDA-MB-231 cells have demonstrated increased PTHrP production in skeletal metastasis as a cause for increased osteolysis and, similarly, we have implicated PTHrP in local osteolysis in prostate cancer-associated skeletal metastases (50). Increased TGF? release from bone was found to be stimulatory for PTHrP in the studies with MDA-MB-231 cells (51, 52). This is consistent with our previous findings that a variety of growth factors including epidermal growth factor and IGF-I can stimulate PTHrP gene expression (53, 54). In view of the fact that bone is a major reservoir for growth factors such as TGF? and IGF-I, release of these factors during PTHrP-induced osteolysis can further increase PTHrP production by invading tumors as has been previously noted (52). In our studies, reduced PTHrP levels and reduced osteolysis were observed in conditions where E2 action could prevail, i.e. in non-OVX animals inoculated with ER-expressing S-30 cells, whereas reduction of E2 effects in OVX animals was associated with increased PTHrP and increased osteolysis. However, in view of the fact that other growth factors such as TGF and TGF? may be regulated by E2 activation of ER in S-30 cells, the role of such factors in comodulating PTHrP effects on bone osteolysis cannot be ruled out (29, 55).
Our combined in vitro and in vivo data therefore support a role for E2 in regulating PTHrP reduction, which in turn, can mediate breast cancer effects on osteolysis. Our findings may have important implications for the role of PTHrP in breast cancer progression and for new approaches in the future treatment of E2-dependent and E2-independent breast cancer. Hormone therapy in combination with other anticancer therapy may provide the best strategy for the blockade of breast cancer progression and mortality associated with this prevalent disorder.
Acknowledgments
The authors thank Dr. Ida Khalili (Department of Radiology, McGill University Health Centre) for carefully reviewing all x-rays.
References
Margolese TG, Hortobagy GN, Buchholz TA 2003 Neoplasms of the breast. In: Kufe DW, Pollock PE, Weichselbaum RR, Bast Jr RC, Gansler TS, Holland JF, Frei III E, eds. Cancer medicine. 6th ed. Hamilton, Ontario, Canada: BC Decker Inc.; 1879–1970
Bernstein L 2002 Epidemiology of endocrine-related risk factors for breast cancer. J Mammary Gland Biol Neoplasia 7:3–15
Mundy GR 2002 Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer 2:584–593
Mawson A, Lai A, Carroll JS, Sergio CM, Mitchell CJ, Sarcevic B 2005 Estrogen and insulin/IGF-1 cooperatively stimulate cell cycle progression in MCF-7 breast cancer cells through differential regulation of c-Myc and cyclin D1. Mol Cell Endocrinol 229:161–173
Ali S, Coombes RC 2000 Estrogen receptor in human breast cancer: occurrence and significance. J Mammary Gland Biol Neoplasia 5:271–281
Lippman ME, Golan G, Monaco ME, Pinkus L, Engel E 1976 Model systems for the study of estrogen action in tissue culture. J Steroid Biochem 7:1045–1051
Jordan VC 1995 Studies on the estrogen receptor in breast cancer—20 years as a target for the treatment and prevention of cancer. Breast Cancer Res Treat 36:267–285
Diaz LK, Sneige N 2005 Estrogen receptor analysis for breast cancer: current issues and keys to increasing testing accuracy. Adv Anat Pathol 12:10–19
Platet N, Cathiard AM, Gleizes M, Garcia M 2004 Estrogens and their receptors in breast cancer progression: a dual role in cancer proliferation and invasion. Crit Rev Oncol Hematol 51:55–67
Kurosumi M 2003 Significance of immunohistochemical assessment of steroid hormone receptor status for breast cancer patients. Breast Cancer 10:97–104
Saj S, Omoto Y, Shimizu C, Horiguchi S, Watanabe T, Funata N, Hayash S, Gustafsson JA, Toi M 2002 Clinical impact of assay of estrogen receptor betacx in breast cancer. Breast Cancer 9:303–307
Omoto Y, Inoue S, Ogawa S, Toyama T, Yamashita H, Muramatsu M, Kobayashi S, Iwase H 2001 Clinical value of the wild-type estrogen receptor ? expression in breast cancer. Cancer Lett 163:207–212
Yoneda T 1998 Cellular and molecular mechanisms of breast and prostate cancer metastasis bone. Eur J Cancer 34:240–245
Pluijm G, Lowik C, Papapoulos S 2000 Tumour progression and angiogenesis in bone metastasis from breast cancer new approaches to an old problem. Cancer Treat Rev 26:11–27
Reddi AH, Roodman D, Freeman C, Mohla S 2003 Mechanisms of tumor metastasis to the bone: challenges and opportunities. J Bone Miner Res 18:190–194
Coleman RE 2001 Metastatic bone disease: clinical features, pathophysiology and treatment strategies. Cancer Treat Rev 27:165–176
Chirgwin JM, Guise TA 2000 Molecular mechanisms of tumor-bone interactions in osteolytic metastases. Crit Rev Eukaryot Gene Expr 10:159–178
Burtis WJ, Wu T, Bunch C, Wysolmerski JJ, Insogna KL, Weir EC, Broadus AE, Stev AF 1987 Identification of a novel 17,000-dalton parathyroid hormone-like adenyl cyclase-stimulating protein from a tumor associated with humoral hypercalcemia of malignancy. J Biol Chem 262:7151–7156
Moseley JM, Kubota M, Diefenbach-Jagger HD, Wettenhall, REH, Kemp BE, Suva LJ, Rodda CP, Ebeling, PR, Hudson, PJ, Zajac JD, Martin TJ 1987 Parathyroid hormone-related protein purified from a human lung cancer cell line. Proc Nat Acad Sci USA 84:5048–5052
Juppner H, Abou-Samra AB, Freeman M, Kong XF, Schipani E, Richards J, Kolakowski Jr LF, Hock J, Potts Jr JT, Kronenberg HM 1991 A G protein-linked receptor for parathyroid hormone and parathyroid hormone-related peptide. Science 254:1024–1026
Kaiser SM, Sebag M, Rhim JS, Kremer R, Goltzman D 1994 Antisense-mediated inhibition of parathyroid hormone-related peptide production in a keratinocyte cell line impedes differentiation. Mol Endocrinol 8:139–147
Falzon M, Du P 2000 Enhanced growth of MCF-7 breast cancer cells overexpressing parathyroid hormone-related peptide. Endocrinology 141:1882–1892
Falzon M 1996 Serum stimulation of parathyroid hormone-related peptide gene expression in ROS 17/2.8 osteosarcoma cells through transcriptional and posttranscriptional mechanisms. Endocrinology 137:3681–3688
Holt EH, Lu C, Dreyer BE, Dannies PS, Broadus AE 1994 Regulation of parathyroid hormone-related peptide gene expression by estrogen in GH4C1 rat pituitary cells has the pattern of a primary response gene. J Neurochem 62:1239–1246
Sebag M, Henderson J, Goltzman D, Kremer R 1994 Regulation of parathyroid hormone-related peptide production in normal human mammary epithelial cells in vitro. Am J Physiol 267:723–730
Liu B, Goltzman D, Rabbani SA 1993 Regulation of parathyroid hormone-related peptide production in vitro the rat hypercalcemic Leydig cell tumor H-500. Endocrinology 132:1658–1664
Haq M, Kremer R, Goltzman D, Rabbani SA 1993 A vitamin D analogue (EB1089) inhibits parathyroid hormone-related peptide production and prevents the development of malignancy-associated hypercalcemia in vivo. J Clin Invest 91:2416–2422
Pizzi H, Gladu J, Carpio L, Miao D, Goltzman D, Rabbani SA 2003 Androgen regulation of parathyroid hormone-related peptide production human prostate cancer cells. Endocrinology 144:858–867
Jiang SY, Jordan VC 1992 Growth regulation of estrogen receptor-negative breast cancer cells transfect with complementary DNAs for estrogen receptor. J Natl Cancer Inst 84:580–591
Guo YJ, Higazi AAR, Arakelian A, Sachais BS, Cines D, Goldfarb RH, Jones TR, Kawan H, Mazar AP, Rabbani SA 2000 A peptide derived from the non-receptor binding region of urokinase plasminogen activator (uPA) inhibits tumor progression and angiogenesis and induces tumor cell death in vivo. FASEB J 14:1400–1410
Levenson AS, Kwaan HC, Svoboda KM, Weiss IM, Sakurai S, Jordan VC 1998 Oestradiol regulation of the components of the plasminogen-plasmin system in MDA-MB-231 human breast cancer cells stably expressing the oestrogen receptor. Br J Cancer 78:88–95
El Abdaimi K, Dion N, Papavasiliou V, Cardinal PE, Binderup L, Goltzman D, Sainte Marie LG, Kremer R 2000 The vitamin D analogue EB1089 prevents skeletal metastasis and prolongs survival time in nude mice transplanted with human breast cancer cells. Cancer Research 60:4412–4418
Chen G, Shukeir N, Potti A, Sircar K, Aprikian A, Goltzman D, Rabbani SA 2004 Up-regulation of Wnt-1 and ?-catenin production in patients with advanced metastatic prostate carcinoma. Potential pathogenetic and prognostic implications. Cancer 101:1345–1356
Duivenvoorden WC, Popovic SV, Lhotak S, Seidlitz E, Hirte HW, Tozer RG, Singh G 2002 Doxycycline decreases tumor burden in a bone metastasis model of human breast cancer. Cancer Res 62:1588–1591
Dickson RB, Lippman ME 1995 Growth factors in breast cancer. Endocr Rev 16:559–589
Johnston SRD, Dowsett M 2003 Aromatase inhibitors for breast cancer: lessons from the laboratory. Nat Rev 3:821–831
Miller WR 2004 Biological rationale for endocrine therapy in breast cancer. Best Pract Res Clin Endocrinol Metab 18:1–32
Shao Z, Jiang M, Yu L, Han Q, Shen Z 1997 Estrogen receptor-negative breast cancer cells transfected with estrogen receptor exhibit decreased tumour progression and sensitivity to growth inhibition by estrogen. Chin Med Sci J 12:11–14
Saceda M, Lippman ME, Lindsay RK, Puente M, Martin MB 1989 Role of an estrogen receptor-dependent mechanism in the regulation of estrogen receptor mRNA in MCF-7 cells. Mol Endocrinol 3:1782–1787
Southby J, Kissin MW, Danks JA, Hayman JA, Moseley JM, Henderson MA, Bennett RC, Martin TJ 1990 Immunohistochemical localization of parathyroid hormone-related protein in human breast cancer. Cancer Res 50:7710–7716
Kkohno N, Kitazawa, S, Fukase M, Sakoda Y, Kanbara Y, Furuya Y, Oha O, Ishikawa Y, Saitoh Y 1994 The expression of parathyroid hormone-related protein in human breast cancer with skeletal metastases. Surg Today 24:215–220
Kaiser SM, Laneuville P, Bernier SM, Rhim JS, Kremer R, Goltzman D 1992 Enhanced growth of a human keratinocyte cell line induced by antisense RNA for parathyroid hormone-related peptide. J Biol Chem 267:13623–13628
Massfelder T, Dann P, Wu TL, Vasavada R, Helwig J-J, Stewart AF 1997 Opposing mitogenic and antimitogenic actions of parathyroid hormone-related peptide protein in vascular smooth muscle cells: a critical role for nuclear targeting. Proc Natl Acad Sci USA 169:33–41
Henderson JE, He B, Goltzman D, Karaplis AC 1996 Constitutive expression of parathyroid hormone-related peptide (PTHrP) stimulates growth and inhibits differentiation of CFK2 chondrocytes. J Cell Physiol 169:33–41
Iwamura M, Abrahamsson PA, Foss KA, Wu G, Cockett AT, Deftos LJ 1994 Parathyroid hormone-related protein: a potential autocrine growth regulator in human prostate cancer cell lines. Urology 43:675–679
Burton PBJ, Moniz C, Knight DE 1990 Parathyroid hormone related peptide can function as an autocrine growth factor in human renal cell carcinoma. Res Commun 167:1134–1138
Luparello C, Birch MA, Gallagher JA, Burtis WJ 1997 Clonal heterogeneity of the growth and invasive response of a human breast carcinoma cell line to parathyroid hormone-related peptide fragments. Carcinogenesis 18:23–29
Sepulveda VAT, Shen X, Falzon M 2002 Intracine PTHrP protects against serum starvation-induced apoptosis and regulates the cell cycle in MCF-7 breast cancer cells. Endocrinology 143:596–606
Troung NU, de Edwardes MD, Papavasiliou V, Goltzman D, Kremer R 2003 Parathyroid hormone-related peptide and survival of patients with cancer and hypercalcemia. Am J Med 115:115–121
Rabbani SA, Gladu J, Harakidas P, Jamison B, Goltzman D 1999 Over-production of parathyroid hormone-related peptide results in increased osteolytic skeletal metastasis by prostate cancer cell in vivo. Int J Cancer 80:257–264
Yin J, Selander K, Chirgwin JM, Dallas M, Grubbs BG, Wieser R, Massague J, Mundy GR, Guise TA 1999 TGF-? signaling blockade inhibits PTHrP secretion by breast cancer cells and bone metastases development. J Clin Invest 103:197–206
Kakonen SM, Selander K, Chirgwin JM, Yin JJ, Burns S, Rankin WA, Grubbs BG, Dallas M, Cui Y, Guise TA 2002 Transforming growth factor-? stimulates parathyroid hormone-related protein and osteolytic metastases via Smad and mitrogen-activated protein kinase signaling pathways. J Biol Chem 227:24571–24578
Goltzman D 2001 Extending preclinical models of skeletal manifestations of malignancy to the clinical setting. Semin Oncol 28:9–14
Rabbani SA 2000 Molecular mechanism of action of parathyroid hormone related peptide in hypercalcemia of malignancy: therapeutic strategies. Int J Oncol 16:197–206< href="/cgi/external_ref?access_num=10601566&link_type=MED" >
Jeng MH, Jiang SY, Jordan CV 1994 Paradoxical regulation of estrogen-dependent growth factor gene expression in estrogen receptor (ER)-negative human breast cancer cells stably expressing ER. Cancer Lett 82:123–128(S. A. Rabbani, P. Khalili)