| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Center for Tissue Regeneration and Repair, Department of Orthopedic Surgery, School of Medicine, University of California, Davis, Sacramento, California
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The profound effects that BMPs have on cell fate and determination necessitate a tight regulation of BMP activities. There are more than a dozen different human BMPs, each of which has its own subset of activities and unique expression patterns. This regulation can occur at different levels of the BMP signaling cascade, either extracellularly, at the cell membrane, or within the cell. A family of extracellular BMP antagonists has recently been discovered. These are secreted proteins that bind to BMPs and reduce their bioavailability for interactions with the BMP receptors. Extracellular BMP antagonists include noggin, follistatin, sclerostatin, chordin, DCR, BMPMER, cerberus, gremlin, DAN, and others (refs. 11, 12, 13, 14, 15, 16 ; reviewed in ref. 17 ). There are several type I and type II receptors that bind to BMPs with different affinities. BMP activity is also regulated at the cell membrane level by receptor antagonists such as BAMBI (18) , which acts as a kinase-deficient receptor. Intracellularly, the regulation of BMP activity at the signal transduction level is even more complex. There are inhibitory Smads (Smad-6 and Smad-7), as well as inhibitors of inhibitory Smads (AMSH and Arkadia). There are proteins that can target either receptors or Smads for degradation (Smurf, SANE, and CHIP) and phosphatases that reduce receptor phosphorylation (GADD34-PP1c). A host of receptor and Smad binding proteins can modulate the nuclear translocation of Smad complexes (EID and DLX) or alter Smad binding to the receptors (XIAP, BRAM, FKBP12, and others). Within the nucleus, transcriptional coactivators and corepressors modulate the transcription of BMP response genes (YY1, Ski, Sno, ECSIT, Tob, CBP/P300, and many more). It is the combined effect of this large symphony of BMP signaling molecules that will ultimately determine cellular response to BMP.
BMP-6 expression is increased in prostate adenocarcinoma, and several studies have shown a correlation between elevated BMP-6 and osteoblastic metastases (19, 20, 21, 22, 23) . At present, there is still much to be learned about the function of BMPs and their antagonists in prostate cancers. In this study, we investigated the expression levels of BMPs, their antagonists, and molecules associated with BMP activity. In particular, we investigated the expression of BMP-6 and the BMP antagonist noggin in prostate cancer cell lines, with special focus on reciprocal feedback regulation and proliferation.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Quantitative Reverse Transcription-Polymerase Chain Reaction.
DNA-free total RNA was prepared with RNeasy mini-columns, using the on-column DNase digestion step (Qiagen, Valencia, CA). Quantitative reverse transcription-polymerase chain reaction (RT-PCR; TaqMan) primers and probes were designed using Primer Express software (Applied Biosystems, Foster City, CA), and their sequences (in 5' to 3' direction) were as follows: BMP6Fwd, CCGTGTAGTATGGGCCTCAGA; BMP6Rev, TCACAACCCACAGATTGCTAGTG; BMP6Probe, CGTGATGTCAAATTCCAGCCAGCCTT; NogginFwd, CCTGGTGGACCTCATCGAA; NogginRev, CAGCGTCTCGTTCAGATCCTT; NogginProbe, ACCCAGACCCTATCTTTGACCCCAGG; GremlinFwd, GCAAAACCCAGCCGCTTA; GremlinRev, GGTTGATGATGGTGCGACTGT; GremlinProbe, CAGACCATCCACGAGGAAGGCTGC; CerberusFwd, ACCATGAGGAAGCTGAGGAGAA; CerberusRev, GCAGGGCTGGTGGCTACA; and CerberusProbe, AGATCTGTTTGTCGCAGTGCCACACC. Real-time quantitative RT-PCR was done using a Sequence Detector 7700 and the EZ-RT-PCR reagents from Applied Biosystems. Expression of each mRNA was normalized to that of glyceraldehyde-3-phosphate dehydrogenase, and the fold change was calculated using the formula 2
Ct as described by Applied Biosystems. Each RT-PCR was done in triplicate, and the results shown are representative of at least two repeated experiments.
Western Blotting.
Human prostate cancer biopsies were obtained from Dr. Ralph de Vere White (Department of Urology, University of California Davis Medical Center) and homogenized in reducing SDS-PAGE sample loading buffer that contained 8 mol/L urea. After centrifugation for 5 minutes at 12,000 x g, the supernatants were boiled for 5 minutes and then electrophoresed on precast 4–12% gradient gels (Invitrogen). Cultures of 200,000 cells were lysed in 100 µl of 1x SDS-PAGE sample buffer, and 40 µl were loaded into each lane of a SDS-PAGE precast cell (corresponding to 80,000 cells per lane for even protein loading), which was then processed as described above. After transfer to Immobilon-P (Millipore, Bedford, MA), the membranes were blocked with 2% bovine serum albumin and then incubated with the anti-noggin (Regeneron Pharmaceuticals) and anti–BMP-6 antibodies at 2 to 10 µg/ml and anti-p21 (BD Biosciences, San Jose, CA) and anti–phospho-Smad (Cell Signaling, Beverly, MA) at the dilutions recommended by the manufacturers. Detection was done by either horseradish peroxidase-conjugated goat antirabbit secondary antibodies and enhanced chemiluminescence (ECL; Amersham Biosciences, Piscataway, NJ) or alkaline phosphatase-conjugated goat antibodies and Enhanced ChemiFluorescence and quantified on a Storm-860 scanner (Amersham Biosciences).
Immunohistochemistry.
A slide array of 80 characterized, paraffin-embedded 2-mm human prostate biopsies (Imgenex, San Diego, CA) was characterized by Imgenex using the Gleason grading scale. The slide array was incubated with 25 µg/ml mouse anti-noggin monoclonal antibody using the VectaStain ABC-AP kit according to the manufacturer’s protocols (Vector Laboratories, Burlingame, CA).
Cell Proliferation Assay.
Cells were seeded in media supplemented with 10% FBS at a density of 1.5 x 105 cells per well in 6-well dishes and allowed to attach overnight. The culture medium was replaced with media supplemented with 10% charcoal-stripped FBS for 24 hours, and then BMP-6 and dihydrotestosterone [DHT (in media supplemented with 10% charcoal-stripped FBS)] were added for a period of 7 days. Media were replaced and supplemented with BMP-6 and DHT every 2 to 3 days. Cells were lysed by the addition of 500 µl 0.5% SDS solution, and DNA was isolated using the phenol/chloroform method. The relative DNA concentration was obtained using the PicoGreen assay (Molecular Probes, Eugene, OR) and quantified on a Storm-860 scanner.
Microarray Analysis.
DU-145 cells were seeded in media supplemented with 10% FBS at a density of 3 x 106 cells per 150-cm2 flask (approximately 20,000 cells per cm2) and allowed to attach overnight. The culture medium was replaced with media supplemented with 1% charcoal-stripped FBS for 24 hours, and then BMP-6 was added in media supplemented with 1% charcoal-stripped FBS. DNA-free total RNA was prepared using RNEasy mini-columns, using the on-column DNase digestion step. RNA samples were processed for analysis on Affymetrix U133 2.0 Plus arrays by the Gene Expression Resource core facility at University of California Davis Medical Center. These arrays cover the entire human genome and include more than 47,000 gene transcripts.
Public Databases.
Twenty-seven normal and 52 cancerous prostate samples were processed on Affymetrix U95A microarrays by Singh et al. (24)
, and the data were made publicly available.1
We analyzed these files with dChip 1.32
to determine the presence or absence of each gene listed and whether there was differential regulation in tumor versus normal biopsies. SAGE and electronic Northern expression data, when available, were taken from the Cancer Genome Anatomy Project databases.3
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
|
|
|
. In the genes detected, there is considerable agreement between the prostate biopsies and the DU-145 cell line. However, comparison of extended BMP family and related signaling molecules showed that many genes that were present in the DU-145 cell line were not detectable in RNA from the prostate biopsies. We were particularly interested in the genes that negatively regulate BMP signaling because elevated BMP-6 expression correlated with bone metastasis of prostate cancer. Noggin, Smad-6, Smad-7, and BAMBI are negative regulators of BMP signaling whose expression was increased on BMP-6 treatment of DU-145 cells. BAMBI was also up-regulated in cancerous prostates. The regulation of noggin by BMP-6 will be presented in greater detail below.
BMP-6 and Noggin Are Present in Prostate.
To confirm the presence of BMP-6 and noggin in prostate at the protein level, immunoblots of prostate cancer homogenate were probed with antibodies against human BMP-6 and noggin. Both of these proteins were expressed in the homogenized prostate cancer biopsies (Fig. 1A
,Lanes 1 and 2, respectively).
|
shows four representative images. Staining in normal prostate was observed primarily in epithelial cells, with little or no staining in the stromal cells. In high-grade cancers, noggin protein was localized to areas dense with carcinoma cells. In samples with mixed areas of connective tissue and carcinoma, noggin was localized to the carcinoma cells, and there was little or no staining in the connective tissue. Using this technique, we were not able to demonstrate a difference in noggin staining intensity among normal epithelium and low-grade and high-grade cancer cells.
BMP-6 Induces Noggin Expression.
Recombinant human BMP-6 protein added to cultures of DU-145 and LNCaP prostate cancer cell lines for 48 hours resulted in a dose-dependent increase in noggin mRNA, as shown in Fig. 2A
. This confirms the previous observations made in rat osteoblasts (25)
that noggin can be induced by BMP-6 in certain cell types. The mRNA levels of two additional extracellular BMP antagonists, gremlin and cerberus, were examined in the same samples and showed no significant changes in the presence of BMP-6 protein (Fig. 2C)
. In LNCaP prostate cancer cell culture, the BMP-6–induced increase of noggin mRNA was not dependent on the presence of androgens (data not shown). The induction of noggin mRNA was observed in both the LNCaP (androgen receptor-positive) and DU-145 (androgen receptor-negative) cell lines, providing further evidence that this is an androgen-independent event (Fig. 2A)
.
|
, Lanes 3 and 4). Noggin protein was detected in the cells and their associated extracellular matrix, but not in the cell culture supernatant.
Noggin Inhibits BMP-6 Activity.
Having shown that BMP-6 reproducibly up-regulates noggin at the mRNA and protein levels, we designed experiments to determine whether noggin could act as an antagonist to BMP-6. Previously, it had been shown that noggin inhibits the activity of BMP-2, -4, and -7, but the effect of noggin on BMP-6 had not been demonstrated. Increasing amounts of noggin were added to cells cultured in the presence of 100 ng/ml BMP-6 for 48 hours, and quantitative RT-PCR was used to measure the up-regulation of noggin mRNA. Noggin protein at higher doses (1–3 µg/ml) reduced BMP-6 activity by 50% to 100% in both LNCaP and DU-145 cells. As a control, in the absence of BMP-6 protein, noggin protein had no discernible effect (Fig. 2C)
.
BMP-6 Activates Negative Feedback Mechanisms.
Because elevated BMP-6 expression in prostate correlates with poor patient prognosis, we were interested in studying the expression of negative regulators of BMP activity. As described above, the profound effects of BMPs on cells and tissues have given rise to many levels of mechanisms to regulate these effects, including negative feedback loops that are activated by BMP-6. Table 2
is a summary of microarray data showing that BMP-6 activates several negative feedback mechanisms when added to DU-145 cell culture. Inhibitory Smad-6 and Smad-7 are up-regulated and function to block BMP signal transduction by interfering with the activation of receptor Smads (Smads 1, 5, and 8; ref. 26
). Smurf-2 is up-regulated by BMP-6 and functions by increasing the degradation rate of receptor Smads, thus attenuating BMP signaling (27)
. BAMBI was up-regulated by BMP-6 and acts as a kinase-deficient type I receptor unable to activate Smads (18)
. These proteins are examples of negative feedback regulation that reduce or weaken BMP signaling.
|
Again, we wanted to extend these findings beyond the DU-145 cell culture model and ask which of these BMP signaling genes are involved in prostate cancer. To a large extent, the DU-145 cancer cell model system agrees with both the SAGE and cancer versus normal prostate microarray data, although the data from the DU-145 cell arrays (Affymetrix 133 2.0 Plus chips) are much more complete than the chips used by Singh et al. (24)
(Affymetrix U95A chips) and also more sensitive than the SAGE technology. The cancer versus normal prostate comparisons reveal information about BMP regulation beyond what is possible with DU-145 cell culture microarrays. For example, DAN, an extracellular BMP antagonist protein, was down-regulated in prostate cancers compared with normal prostate both on microarray and SAGE data sets and was present but not affected by BMP-6 in DU-145 cell culture. These comparisons are presented in Table 2
.
In summary, we have demonstrated that BMP-6 induces noggin expression in prostate cancer cell culture at the mRNA and protein levels. Recombinant noggin protein inhibits this activity of BMP-6 in a prostate cancer cell culture system. In addition to noggin, BMP-6 activates several other negative (and some positive) feedback mechanisms that may attenuate BMP signal transduction. The cancer cell model data are in agreement with data derived from comparing normal and cancerous prostate.
BMP-6 Induces Smad Phosphorylation.
Many studies have documented the function of BMPs in the context of bone development and embryogenesis. However, the function of BMP-6 in the adult prostate remains unknown. To extend the documented correlation between elevated BMP-6 expression and tumor metastases, we wanted to (a) examine the BMP-6 signal transduction pathway, (b) identify some of the downstream targets, and (c) characterize the effects of BMP-6 on the biology of prostate cancer cells.
The first few steps in BMP signal transduction involve binding of BMP to its receptors, which are activated and subsequently phosphorylate Smads. These pathways are often disrupted in cancer cells (29)
. To test whether Smads are phosphorylated on BMP-6 treatment of DU-145 and LNCaP cells, immunoblots were performed on cell lysates and probed with antibodies that specifically recognize only phosphorylated Smad-1, -5, and -8. BMP-6 treatment induced Smad phosphorylation within 30 minutes, and elevated levels of phosphorylated Smads continued to be detected to at least 96 hours of BMP-6 treatment, as shown in Fig. 3A
. All samples were compared with untreated cells cultured for the same length of time under otherwise identical conditions. The amount of unphosphorylated Smad-1 protein remained essentially identical between BMP-6–treated and control cells. Thus, the Smad phosphorylation pathway is intact in these cells and used for BMP signal transduction in response to exogenous BMP-6.
|
. Two lines of evidence indicated that this growth-inhibitory effect is not dependent on the presence of androgens. First, the effect was seen with or without addition of DHT to the medium. Second, the growth-inhibitory effect of BMP-6 was also seen in androgen-independent DU-145 cells (data not shown).
BMP-6 Induces p21/CIP Expression.
The next set of experiments were done to gain insight into how BMP-6 may control cell proliferation. Microarray data showed that BMP-6 treatment caused a strong up-regulation of p21/CIP mRNA within 6 hours. P21/CIP is a cell cycle control protein shown to mediate BMP-dependent growth reduction in other cell types (30)
and to be induced by BMP-2 and -4 in prostate (31)
.
Treatment with BMP-6 caused an increase of cell cycle control protein p21/CIP, which was detected at the mRNA level in DU-145 cells and confirmed at the protein level in both DU-145 and LNCaP cells. LNCaP cells expressed a higher basal level of p21 protein than DU-145 cells, and in both cell lines, there was a time-dependent increase of p21/CIP protein with BMP-6 treatment, as shown in Fig. 3B
. P21/CIP protein levels were detectably increased after 90 minutes of BMP-6 treatment and remained elevated at least until 96 hours after BMP-6 treatment compared with cells cultured for the same length of time without BMP-6 in otherwise identical media. Thus we show evidence that the decreased proliferation of BMP-6–treated cells may be due in part to increased levels of p21/CIP protein.
A number of additional cell cycle control proteins were regulated at the mRNA level by BMP-6 treatment in DU-145 cells, as shown in Table 3
. Most of these changes are consistent with reduced proliferation as a result of BMP-6 treatment. Cyclin-dependent kinase inhibitors p21/CIP, p18, and p19 were up-regulated, and cyclin-dependent kinases were down-regulated.
|
. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In this study, a comprehensive list of genes involved in generating and regulating the activities of BMPs was compiled. We tested the expression of these genes in DU-145 cells and analyzed how their expression is regulated by BMP-6. To relate these cell culture observations to whole prostates, we searched publicly available data from National Center for Biotechnology Information and Massachusetts Institute of Technology, and we found that the cell culture model in many ways mimics what occurs in the prostate. Of the top 50 genes with high expression in tumor and the top 50 genes with high expression in normal prostate identified by Singh et al. (24) , 13 are either up- or down-regulated by BMP-6 in DU-145 cell culture. Comparisons between cell culture and whole prostate microarray data, although useful to extend the validity of cell culture observations, must also be viewed critically because of the many cell types contributing to RNA from whole prostates.
Special attention was paid to antagonists of BMP activity, especially noggin. Prostate cancer cells were shown to express BMP-6 and noggin at the mRNA and protein levels. Noggin was localized predominantly in normal epithelial and carcinoma cells in prostate tissue biopsies. Surprisingly, it was not detected in cell culture unless the cells were stimulated with BMP-6. Together with the observation that noggin was not detectable in the cell culture supernatant, this may indicate that there is very limited diffusion of the protein away from the cells in which it is produced. It will also be interesting to study whether the BMP-6 produced by prostate carcinoma can diffuse past the noggin bound to the heparan sulfate proteoglycan in the extracellular matrix (37 , 38) . Because little is known about the relative binding affinities of noggin to the various BMP family members, it remains to be determined whether noggin specifically inhibits BMP-6 availability in prostate epithelium or whether it functions as a general antagonist of all BMPs.
BMP-6 inhibited the proliferation of DU-145 and LNCaP cells. We demonstrated that BMP-6 causes changes in the expression of cell cycle control proteins. Specifically, BMP-6 caused strong up-regulation of p21/CIP, which may explain in part the reduced proliferation seen in BMP-6–treated cells. This effect was not androgen dependent because BMP-6 reduced proliferation and increased p21/CIP expression in both DU-145 and LNCaP cells, and in the presence and absence of DHT. DU-145 cells, in addition to being androgen independent, have a missense mutation in the p53 gene (39) . This indicates that BMP-6 induces p21/CIP protein and mRNA expression in a p53-independent manner.
The antiproliferative response of prostate cancer cells to BMP-6 seems counterintuitive, given the established correlation between aggressive cancer and elevated BMP-6 expression. However, the BMP-6–induced increase in Id protein expression may explain in part the increased resistance to apoptosis (40) and the correlation between BMP overexpression and cancer progression. In prostate cancers metastatic to bone, the effect of the BMP-6 in the surrounding bone environment may be related to the osteosclerotic nature of the metastatic lesions.
The findings that BMP-6 up-regulated noggin expression at mRNA and protein levels and, conversely, that noggin inhibited BMP-6 function have potential therapeutic implications for noggin in ameliorating prostate bone metastases and attendant pain and morbidity. The challenge for the future is to dissect BMP-6 signaling in prostate cancer, both in the prostate cancer in situ and in the bone microenvironment in the immediate vicinity of the metastases.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: A. Hari Reddi, Center for Tissue Regeneration and Repair, Department of Orthopedic Surgery, School of Medicine, University of California, Davis, 4635 Second Avenue, Sacramento, CA 95817. Phone: 916-734-3311; Fax: 916-734-5750; E-mail: ahreddi{at}ucdavis.edu
1 http://www.broad.mit.edu/cgi-bin/cancer/datasets.cgi. 
Received 6/24/04. Revised 9/ 2/04. Accepted 9/ 9/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Ye, H. Kynaston, and W. G. Jiang Bone Morphogenetic Protein-9 Induces Apoptosis in Prostate Cancer Cells, the Role of Prostate Apoptosis Response-4 Mol. Cancer Res., October 1, 2008; 6(10): 1594 - 1606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Blish, W. Wang, M. C. Willingham, W. Du, C. E. Birse, S. R. Krishnan, J. C. Brown, G. A. Hawkins, A. J. Garvin, R. B. D'Agostino Jr., et al. A Human Bone Morphogenetic Protein Antagonist Is Down-Regulated in Renal Cancer Mol. Biol. Cell, February 1, 2008; 19(2): 457 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Daibata, Y. Nemoto, K. Bandobashi, N. Kotani, M. Kuroda, M. Tsuchiya, H. Okuda, T. Takakuwa, S. Imai, T. Shuin, et al. Promoter Hypermethylation of the Bone Morphogenetic Protein-6 Gene in Malignant Lymphoma Clin. Cancer Res., June 15, 2007; 13(12): 3528 - 3535. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Theriault, T. G. Shepherd, M. L. Mujoomdar, and M. W. Nachtigal BMP4 induces EMT and Rho GTPase activation in human ovarian cancer cells Carcinogenesis, June 1, 2007; 28(6): 1153 - 1162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Schwaninger, C. A. Rentsch, A. Wetterwald, G. van der Horst, R. L. van Bezooijen, G. van der Pluijm, C. W.G.M. Lowik, K. Ackermann, W. Pyerin, F. C. Hamdy, et al. Lack of Noggin Expression by Cancer Cells Is a Determinant of the Osteoblast Response in Bone Metastases Am. J. Pathol., January 1, 2007; 170(1): 160 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Vessella and E. Corey Targeting factors involved in bone remodeling as treatment strategies in prostate cancer bone metastasis. Clin. Cancer Res., October 15, 2006; 12(20): 6285s - 6290s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Moll, C. Millet, D. Noel, B. Orsetti, A. Bardin, D. Katsaros, C. Jorgensen, M. Garcia, C. Theillet, P. Pujol, et al. Chordin is underexpressed in ovarian tumors and reduces tumor cell motility FASEB J, February 1, 2006; 20(2): 240 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Dai, J. Keller, J. Zhang, Y. Lu, Z. Yao, and E. T. Keller Bone Morphogenetic Protein-6 Promotes Osteoblastic Prostate Cancer Bone Metastases through a Dual Mechanism Cancer Res., September 15, 2005; 65(18): 8274 - 8285. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and LNCaP (
) prostate cancer cell cultures for 48 hours, in the presence of 1% charcoal-stripped serum and 100 ng/ml DHT. Fold increase of noggin mRNA was measured by quantitative RT-PCR. B, increasing amounts of recombinant noggin were added to DU-145 cells for 48 hours in the presence of 100 ng/ml BMP-6 and 100 ng/ml DHT. BMP-6 activity was measured as increased noggin mRNA by quantitative RT-PCR. C, recombinant BMP-6 or noggin was added to cultures of DU-145 cells for 48 hours, and fold increase of noggin, gremlin, and cerberus mRNA was measured by quantitative RT-PCR. D, cells (2 x 105) were cultured for 7 days in a base medium of RPMI 1640 with 10% charcoal-stripped FBS and with the addition of either 100 ng/ml BMP-6, 5 nmol/L DHT, or both BMP-6 and DHT. DNA content measurement was done using PicoGreen.