Osteoprotegerin in prostate cancer bone metastasis.

Osteoprotegerin (OPG), a critical regulator of osteoclastogenesis, is expressed by prostate cancer cells, and OPG levels are increased in patients with prostate cancer bone metastases. The objective of this study was to investigate the effects of OPG overexpression on prostate cancer cells and prostate cancer/bone cell interactions in vitro and in vivo. OPG-transfected C4-2 cells expressed 8.0 ng OPG per mL per 10(6) cells, whereas no OPG was detected in the media of C4-2 cells transfected with a control plasmid. OPG overexpressed by C4-2 cells protected these cells from tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis and decreased osteoclast formation. Subcutaneous OPG-C4-2 and pcDNA-C4-2 tumors exhibited similar growth and take-rate characteristics. However, when grown in bone, tumor volume was decreased in OPG-C4-2 versus pcDNA-C4-2 (P=0.0017). OPG expressed by C4-2 cells caused increases in bone mineral density (P=0.0074) and percentage of trabecular bone volume (P=0.007), and decreases in numbers of osteoblasts and osteoclasts when compared with intratibial pcDNA-C4-2 tumors (P=0.003 and P=0.019, respectively). In summary, our data show that increased expression of OPG in C4-2 cells does not directly affect proliferation of prostate cancer cells but indirectly decreases growth of C4-2 tumors in the bone environment. Our data also show that OPG expressed by C4-2 cells inhibits bone lysis associated with C4-2 bone metastasis, which results in net increases in bone volume. We therefore hypothesize that OPG expressed in prostate cancer patient bone metastases may be at least partially responsible for the osteoblastic character of most prostate cancer bone lesions.

Lineage relationship between LNCaP and LNCaP-derived prostate cancer cell lines.

BACKGROUND: LNCaP and its derivative cell lines, which include C4-2 (and the related C4-2B) and CL1, are used as models of prostate cancer. Unlike LNCaP, the other cell lines show features of progressed disease such as metastatic capability and hormone independence. Analyses were done to determine if C4-2 or CL1 cells were selected from pre-existent subpopulations in LNCaP. METHODS: Prostate cancer cells were characterized by cluster designation (CD) phenotyping. Specific cell populations were sorted by flow cytometry. DNA array analysis was used to probe differential gene expression. RESULTS: CD phenotyping showed that CL1 and C4-2 (and C4-2B) were very dissimilar, and C4-2 was more similar to LNCaP. One common difference between LNCaP and its derivatives was CD26, in which virtually all C4-2 or CL1 cells were CD26(+) but only approximately 10% of LNCaP cells were CD26(+). The CD26(+) subpopulation of LNCaP was isolated and cultured in vitro. After culture, a high percentage of the cells (descended from the sorted cells) were CD26(+), in contrast to those sorted by CD13 or CD44. The cultured CD13 and CD44 populations did not show a high percentage of CD13(+) and CD44(+) cells, respectively. CD13 and CD44 are markers, in addition to CD26, for CL1 but not for C4-2. CONCLUSIONS: C4-2 arose probably from CD26(+) LNCaP cells, while CL1 arose de novo.

Characterization of C4-2 prostate cancer bone metastases and their response to castration.

UNLABELLED: New well-characterized preclinical models of prostate cancer (CaP) bone metastases are needed to improve our understanding of the development of CaP-related bone disease in patients. Here we describe characterization of a model consisting of direct injection of C4-2 cells into tibias. INTRODUCTION: Prostate cancer (CaP) has a high proclivity to metastasize to bone. Development and characterization of preclinical models of CaP bone metastases are of high interest. The objective of this study was to characterize C4-2 bone metastases and their response to castration. MATERIALS AND METHODS: Cell suspensions of C4-2, a subline of LNCaP, were injected directly into the tibias of intact male mice. In groups A (n = 7) and B (n = 5), animals were killed 3 and 8 weeks after injection of C4-2 cells, respectively. In group C (n = 7), animals were castrated 3 weeks after injection and killed 5 weeks after castration. Serum prostate-specific antigen (PSA) levels and bone mineral density (BMD) were measured, and bone histomorphometric analysis was performed. RESULTS: C4-2 cells decreased BMD of the injected tibias by 36.1% and bone volume by 74.1% versus normal tibias. Castration caused a 32.3% drop in serum PSA (p = 0.0438), with a nadir at day 14, after which it began to rise again. Bone destruction in the tumorous tibias of castrated animals was decreased by 15.9% versus tumorous tibias of intact animals (p = 0.0392). However, BMD in the tumorous tibias of castrated mice was still lower than in normal tibias of intact animals. Castration also decreased BMD and bone volume in nontumorous tibias (p = 0.0406 and 0.0232, respectively). CONCLUSIONS: The C4-2 model of bone metastasis recapitulates the response to androgen deprivation observed in CaP patients with bone metastases and is suitable for study of interactions between tumor and bone cells and evaluation of new therapeutic modalities.

A novel method of generating prostate cancer metastases from orthotopic implants.

BACKGROUND: Spontaneous metastasis following implantation at the orthotopic site is a highly desired feature in prostate cancer (CaP) models, since it would enable studies of mechanisms associated with tumor cell dissemination. METHODS: LuCaP 23.8 and LuCaP 35, hormone-sensitive CaP xenografts that express PSA and the wild-type androgen receptor, were grown orthotopically in SCID mice. When tumor volumes reached approximately 250-500 mg, primary tumors were removed to allow micrometastases to grow. RESULTS: Using this procedure we generated macroscopic metastases (>20 mg) in 71% (LuCaP 23.8) and 100% (LuCaP 35) of animals, respectively. CONCLUSIONS: These models may be used to evaluate new treatment modalities and study mechanisms associated with development of metastases.

Haploinsufficiency and reduced expression of genes localized to the 8p chromosomal region in human prostate tumors.

Cytogenetic and molecular studies have suggested that deletion or rearrangement of sequences that map to the short arm of chromosome 8 may be permissive for tumorigenesis in several organ systems, and in human prostate tumors in particular. In this study, we hypothesized that genes deleted for one copy and localized to the 8p chromosomal region may be transcriptionally down-regulated or ablated in affected human prostate tumor tissues. To test this hypothesis, we used cDNA microarray analysis to determine the transcriptional profiles for 259 transcribed sequences mapping to the 8p chromosomal region for seven human prostate tumor xenografts, completely characterized for numerical and structural alterations on chromosome 8, and five normal human prostate tissues. These experiments identified 33 genes differentially expressed between normal and malignant prostate tissues, the majority of which (28/33, 85%) were transcriptionally down-regulated in malignant compared to normal human prostate tissues. These findings, that haploinsufficiency and transcriptional down-regulation for genes mapping to 8p are largely coincident in human prostate tumors, should provide a powerful tool for the identification of tumor-suppressor genes associated with human prostate cancer initiation and progression.

LuCaP 35: a new model of prostate cancer progression to androgen independence.

BACKGROUND: Generation of suitable in vivo models is critical for understanding of processes associated with development and progression of prostate cancer (CaP). METHODS: Lymph nodes containing metastatic androgen-independent CaP were implanted into athymic mice. A xenograft designated LuCaP 35 and its hormone-independent variant LuCaP 35V were established and characterized. RESULTS: LuCaP 35 is an androgen-sensitive, prostate-specific antigen (PSA)-producing xenograft. It expresses the wild-type androgen receptor and exhibits deletions in chromosome 8p, but not in chromosome 10. The response of LuCaP 35 to androgen ablation is similar to that observed in man. Using recurring LuCaP 35 tumors we have also established an androgen-insensitive variant of LuCaP 35. CONCLUSIONS: The availability of hormone-dependent and -independent variants of LuCaP 35, which exhibit many properties analogous to those of CaP in man, provides an excellent model system to study the processes associated with development of androgen independence and to evaluate new treatment modalities.

Molecular characterization of prostatic small-cell neuroendocrine carcinoma.

OBJECTIVES: A subset of prostate carcinomas is composed predominantly, even exclusively, of neuroendocrine (NE) cells. In this report, we sought to characterize the gene expression profile of a prostate small cell NE carcinoma by assessing the diversity and abundance of transcripts in the LuCaP 49 prostate small cell carcinoma xenograft. METHODS: We constructed a cDNA library (PRCA3) from the LuCap 49 prostate small cell xenograft. Single pass DNA sequencing of randomly selected cDNA clones followed by sequence assembly and annotation produced a library of Expressed Sequence Tags (ESTs) representing the LuCaP 49 transcriptome. Comparative sequence analysis with ESTs derived from prostate adenocarcinoma libraries was performed using statistical algorithms designed to identify differentially expressed sequences. Putative NE cell-specific genes were further examined by Northern analysis. RESULTS: Sequence assembly and analysis identified 1,447 distinct genes expressed in the LuCaP 49 cDNA library. These include cDNAs encoding the NE markers secretogranin (SCG2), CD24, and ENO2. Northern analysis revealed that three additional genes, ASCL1, INA, and SV2B are expressed in LuCaP 49 but not in various prostate cancer cell lines or xenografts. Fifteen genes were identified with a statistical probability (P > 0.9) of being up-regulated in LuCaP 49 small cell carcinoma relative to prostate adenocarcinoma (two primary prostate adenocarcinomas and the LNCaP prostate adenocarcinoma cell line). CONCLUSIONS: Prostate small cell carcinoma expresses a diverse repertoire of genes that reflect characteristics of their NE cell of origin. ASCL1, INA, and SV2B are potential molecular markers for small cell NE tumors and NE cells of the prostate. This small cell NE carcinoma gene expression profile may yield insights into the development, progression, and treatment of subtypes of prostate cancer.

Zoledronic acid exhibits inhibitory effects on osteoblastic and osteolytic metastases of prostate cancer.

PURPOSE: In this study we have examined the effects of zoledronic acid (ZA), a new-generation bisphosphonate, on prostate cancer (CaP) cells in vitro, and on both osteoblastic and osteolytic CaP metastases in animal models. EXPERIMENTAL DESIGN: In vitro, CaP cells were treated with ZA, and the effects on proliferation, cell cycle, and apoptosis were determined. In vivo, PC-3, and LuCaP 23.1 s.c. and tibial tumors were treated with ZA. Effects on bone and tumor were determined by histomorphometry and immunohistochemistry. RESULTS: ZA decreased proliferation of CaP cells, and caused G(1) arrest and apoptosis of CaP cells in vitro. In vivo, s.c. CaP tumor growth was not affected by ZA. However, growth of osteoblastic and osteolytic metastases of CaP was inhibited significantly in vivo. Matrix metalloproteinase-2, matrix metalloproteinase-9, and Cathepsin K levels were decreased in osteolytic bone metastases after ZA administration. CONCLUSIONS: In conclusion, we have shown that ZA has significant antitumor effects on CaP cells in vitro and in vivo. Antiosteolytic activity and the antitumor effects of this compound could benefit CaP patients with bone metastases.

Establishment and characterization of osseous prostate cancer models: intra-tibial injection of human prostate cancer cells.

BACKGROUND: To improve the therapy of advanced prostate cancer (CaP), it is critical to develop animal models that mimic CaP bone metastases. Unlike the human disease, CaP xenograft models rarely metastasize spontaneously to bone from the orthotopic site of primary tumor growth. METHODS: Single-cell suspensions of LNCaP, PC-3, LuCaP 35, and LuCaP 23.1 CaP cells were injected directly into tibia of SCID mice. Immunohistochemistry and bone histomorphometrical analyses were performed to characterize these osseous-CaP models. RESULTS: LuCaP 23.1 yields an osteoblastic response, LNCaP yields mixed lesions, and LuCaP 35 and PC-3 result in osteolytic responses. We have detected osteoprotegerin, RANK ligand, parathyroid hormone-related protein, and endothelin-1, proteins associated with bone growth and remodeling, in the CaP cells grown in the bone. CONCLUSIONS: These animal models can be used to study biological interactions, pathways, and potential therapeutic targets, and also to evaluate new agents for treatment and prevention of CaP bone metastasis.

Inhibition of androgen-independent growth of prostate cancer xenografts by 17beta-estradiol.

PURPOSE: Estrogen treatment has long been known to be of benefit in prostate cancer (CaP), but its mechanism was thought to involve merely a reduction in androgen levels. However, new evidence indicates that estrogen may exert effects on CaP cells in the absence of androgens. EXPERIMENTAL DESIGN: Implantation of CaP xenografts (LuCaP 35, LuCaP 49, LuCaP 58, LuCaP 73, PC-3, and LNCaP) into intact and ovariectomized female mice was done to characterize growth and take rates in the absence of androgens. Ovariectomized female mice were supplemented with 17beta-estradiol, and LuCaP 35 CaP xenograft take and growth rates were determined. Reverse transcription-PCR was used to evaluate the presence of the estrogen receptor messages in CaP xenografts. RESULTS: We have observed significant inhibition of CaP growth in intact versus ovariectomized female animals in five of six CaP xenograft lines. 17beta-Estradiol supplements given to ovariectomized female mice led to inhibition of tumor establishment and diminished growth of LuCaP 35 similar to that observed in intact female mice. Using reverse transcription-PCR, we have shown that these xenografts express the estrogen receptor beta message. CONCLUSIONS: We have determined that 17beta-estradiol supplementation causes inhibition of CaP growth in an animal model by mechanisms that are independent of androgen action. This gives rise to the possibility that estrogen therapy may be of potential use with hormone-refractory cancers. The xenograft models we describe herein may be useful as well in elucidating the pathways mediating the androgen-independent effects of estrogen on CaP.