| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Geriatric Research, Education, and Clinical Center and Research Service, Veterans Affairs Medical Center (L.R.Q., C.M.P.-S., K.L.B., G.A.H., B.A.R.); and Departments of Medicine (L.R.Q., C.M.P.-S., G.A.H., B.A.R.), Molecular and Cellular Pharmacology (K.L.B.), Biochemistry and Molecular Biology (G.A.H.), and Neurology (B.A.R.), and Sylvester Comprehensive Cancer Center (L.R.Q., C.M.P.-S., K.L.B., G.A.H., B.A.R.), University of Miami School of Medicine, Miami, Florida 33101; Genentech, Inc. (R.H.S.), South San Francisco, California 94080; and Good Samaritan Cancercare and Research Center (R.C.O.), Puyallup, Washington 98371
Address all correspondence and requests for reprints to: Bernard A. Roos, M.D., University of Miami School of Medicine, P.O. Box 016960 (D-503), Miami, Florida 33101. E-mail: broos{at}med.miami.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hepatocyte growth factor (HGF; also known as scatter factor) is a mesenchymal protein with mitogenic, motogenic, and morphogenic effects on nonneoplastic as well as neoplastic epithelial cell types (8, 9). HGF’s pleiotropic effects are mediated via its receptor (MET), the transmembrane tyrosine kinase encoded by the MET protooncogene (10). Unlike vitamin D, which usually slows growth, HGF stimulates the proliferation of most normal and neoplastic cells (9). However, in vitro studies have shown that some carcinomas, such as hepatoma, melanoma, and breast carcinoma, are paradoxically inhibited by HGF (11, 12, 13). In normal prostate, HGF is produced by stromal cells nearest to the basal epithelial cells, which have more HGF receptor and less androgen receptor compared with luminal prostate epithelium (14, 15, 16). During prostate growth the basal cells respond to a series of induced stromal growth factors, including HGF, which regenerate differentiated luminal secretory epithelium with high sensitivity to and dependence on androgen (14, 17, 18, 19). MET is increased early in epithelial neoplasia, including prostatic intraepithelial neoplasia (15, 20), and remains highly expressed in virulent carcinomas and derived cell lines (15, 20, 21).
The prominence of MET expression in prostate carcinoma in concert with our previous finding of HGF and vitamin D synergistic action in increasing alkaline phosphatase activity in cartilage (22) prompted our investigation of HGF’s action in prostate cancer and HGF’s possible interactions with vitamin D. We surveyed HGF’s effects on cell morphology and growth of the androgen-sensitive LNCaP cells and on three androgen-refractory cell lines (DU 145, PC-3, and ALVA-31). Based on that survey, we pursued more detailed characterization of the separate and combined effects of HGF and vitamin D in ALVA-31, the cell line our initial survey found to be most inhibited by the growth factor-vitamin D combination. We observed that a significant inhibition was obtained by combining low doses of the agents, amounts that by themselves had no growth effect. Similarly, the HGF-vitamin D combination, but neither agent alone, decreased the fraction of cells in G1 and increased the fraction in later cell cycle phases. These results support slowing of cell cycle events by cooperative actions of vitamin D and HGF at loci beyond the G1/S contact point.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture
Human prostate carcinoma cell lines (24) (DU 145; ATCC HTB81),
LNCaP (25) (ATCC CRL 1740), and PC-3 (26) (ATCC CRL 1435) were obtained
from the American Type Culture Collection (Manassas, VA).
ALVA-31 cells were obtained as previously described (27). All cells
were cultured in RPMI 1640 medium (Life Technologies, Inc., Grand Island, NY) supplemented with 7.5% FBS
(HyClone Laboratories, Inc., Logan, UT), 100 IU/ml
penicillin, and 100 µg/ml streptomycin at 37 C in a humidified
atmosphere of 5% CO2 in air.
Androgen receptor expression
Because it had previously been reported that ALVA-31 cells
express the androgen receptor (27), we assessed the expression of this
receptor using the RNase protection technique (28). Human androgen
receptor messenger RNA (mRNA) levels were measured with a
32P-labeled T7 polymerase-synthesized
antisense RNA probe from StuI-digested
HindIII-EcoRI/Bluescript-KS. Human
glyceraldehyde-3-phosphate dehydrogenase mRNA was measured with a
32P-labeled T7 polymerase-synthesized antisense
RNA probe from StyI-digested pTRIPLEscript-
glyceraldehyde-3-phosphate dehydrogenase (Ambion, Inc.,
Austin, TX). Forty micrograms of total RNA were hybridized at 37 C for
16 h and digested with RNase mixture (Ambion, Inc.)
for 30 min at 37 C. RNase digestion products were separated by
electrophoresis on polyacrylamide-urea gels and analyzed by
autoradiography. Androgen receptor mRNA was present in LNCaP (positive
control), absent in DU 145 (negative control), and absent in PC-3 and
ALVA-31 cells.
Isolation of human MET RNA by RT-PCR
Total RNA was isolated from PC-3 human prostate cancer cells by
the LiCl-urea method (29) and treated with deoxyribonuclease
(RNase-free). RT-PCR was performed for MET using the oligonucleotide
primers 5'-GGTTGCTGATTTTGGTCATGC-3' (forward, residues 3905–3925 bp)
and 5'-TTCGGGTTGTAGGAGTCTTCT-3' (reverse, residues 4146–4166 bp) (30).
After the RT reaction, PCR was carried out in a DNA thermal cycler
(Perkin-Elmer Corp./Cetus, Palo Alto, CA) under the
following conditions: 1-min denaturing at 94 C, 1-min annealing at 55
C, 2-min extension at 72 C for 35 cycles, and 7-min extension at 72 C.
The expected 261-bp PCR fragment was cloned into
EcoRV-digested Bluescript-KS (treated with Taq
polymerase), and the correct sequence was confirmed by DNA sequencing
to give plasmid hMET/BS.
RNase protection analysis
RNA was isolated from the human prostate cancer cell lines
LNCaP, PC-3, DU 145, and ALVA-31 by the LiCl-urea method (29). Human
MET mRNA levels were measured with 32P-labeled T7
polymerase-synthesized antisense RNA probe from
HindIII-digested hMET/BS. Human cyclophilin mRNA was
measured with a 32P-labeled T3
polymerase-synthesized antisense RNA probe from pTRI-cyclophilin
(Ambion, Inc.). Ten micrograms of total RNA were
simultaneously hybridized with the MET and cyclophilin probes at 37 C
for 16 h and digested with RNase mixture (Ambion, Inc.) for 30 min at 37 C. RNase digestion products were analyzed
by electrophoresis on polyacrylamide-urea gels and autoradiography. The
protected RNA fragments are 261 nucleotides (MET) and 103 nucleotides
(cyclophilin).
Western blot analysis
Total protein lysates from LNCaP, PC-3, DU 145, and ALVA-31
cells were prepared as previously described (31) except without
boiling. After separation of 10 µg protein by SDS-PAGE, proteins were
transferred by electrophoresis to Immobilon-P membrane (Millipore Corp., Bedford, MA) and incubated in 5% nonfat dry milk, PBS,
and 0.05% Tween-20 for 1 h. Rabbit polyclonal antibodies specific
for human MET (1/1000 dilution; h-MET C-12, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were diluted in 5%
nonfat dry milk, PBS, and 0.05% Tween-20 and incubated overnight at 4
C. Membranes were washed in PBS and 0.05% Tween-20 (three times, 10
min each time) and incubated with horseradish peroxidase-conjugated
secondary antibody (antirabbit; 1/1000 dilution; Roche Molecular Biochemicals, Indianapolis, IN) for 1 h, washed in PBS and
0.05% Tween-20, and analyzed by exposure to x-ray film (X-Omat,
Eastman Kodak Co., Rochester, NY) using enhanced
chemiluminescence (ECL, Amersham Pharmacia Biotech,
Arlington Heights, IL).
MET activation (P-Tyr) measurement by KIRA (kinase receptor
activation) assay
Cells were plated in 10-cm dishes with RPMI medium
supplemented with 7.5% FBS. After incubation overnight at 37 C,
semiconfluent cultures were washed three times with PBS, then treated
for 10 min with HGF at 10 ng/ml diluted in serum-free RPMI 1640
with 1 mg/ml BSA. Cells were rinsed and lysed in 1 ml PBS, 0.2% Triton
X-100, 10 µg/ml aprotinin, 5 mM sodium fluoride, 2
mM orthovanadate, and 0.2 mM
phenylmethylsulfonylfluoride for 30 min at room temperature. The lysate
was cleared by centrifugation, and the supernatant was collected; the
protein was quantified by the bicinchoninic acid protein assay
(Pierce Chemical Co., Rockford, IL). One
hundred-microliter aliquots were transferred in duplicate to
a Maxisorp 96-well plate that had been coated with 5 µg/ml
protein A-purified rabbit IgG antibody against
MET-extracellular domain (anti-MET/ECD IgG) (32). Cell lysates were
incubated for 2 h at room temperature. The bound P-Tyr was
detected after another 2-h incubation with biotin-conjugated
antiphosphotyrosine antibody, followed by horseradish
peroxidase-streptavidin and development with tetramethylbenzidine
peroxidase solution (32). The reaction was stopped with 1 M
phosphoric acid, and MET activation (i.e. phosphorylated
tyrosine) was determined based on optical density at 450–690 nm in an
automatic plate reader.
Assay of cell proliferation
Cell proliferation was assessed by cell counting. To test
the separate and combined antiproliferative effects of
1,25-(OH)2D and HGF, cells were seeded at a
density of 5000 cells/6-cm dish. After incubation for 24 h, medium
was replaced with fresh medium containing vehicle (ethanol, final
concentration of 0.01%), 1,25-(OH)2D (to a final
concentration of 1, 10, or 100 nM), HGF (added in medium to
a final concentration of 1, 10, or 20 ng/ml), or a combination of HGF
and 1,25-(OH)2D. After 3 days, the medium was
changed and replenished, and on the sixth day, cells were harvested by
trypsinization and counted with a Neubauer hemocytometer (Hausser
Scientific, Hursham, PA). Cell numbers of each experiment were
derived from the mean value of triplicate wells in an experiment.
Nearly all (>98%) cells under all treatment conditions excluded
trypan blue.
Analysis of cell cycle effects
Cell cycle distribution and changes with HGF and
1,25-(OH)2D were estimated for ALVA-31 cells by
analytical flow cytometry (33). To accumulate sufficient cells for cell
cycle analyses, cells were seeded at 30,000 cells/10-cm dish (twice the
density used for proliferation studies). They were treated 24 h
later with ethanol (0.01%), HGF (10 ng/ml),
1,25-(OH)2D (10 nM), or a combination
of HGF (10 ng/ml) and 1,25-(OH)2D (10
nM). After 6 days of treatment, cells were trypsinized and
washed twice with ice-cold PBS containing 0.1% glucose, then fixed by
dropwise addition of 70% ethanol. After at least 12 h of ethanol
fixation, DNA was stained with propidium iodide (50 µg/ml) for 30
min. RNase (100 U/ml) was included in the staining solution to degrade
double stranded RNA. Analyses were performed with a FACScan unit
(Becton Dickinson and Co., San Jose, CA). Excitation was
at 488 nm, with emission measured at 630 nm. Distribution of cells with
respect to their DNA content was analyzed for 5000 cells for each test
condition. The relative proportions of cells in various cell cycle
phases were estimated by compartment analysis of DNA fluorescence using
cell cycle analysis software from the manufacturer to set cut-offs for
G1, S, and G2/M (33).
Statistical analysis
Statistical analysis was performed by ANOVA. For single
comparisons of the difference between means, unpaired Student’s
t test was applied. Significance levels are indicated in the
figure legends.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
Similarly, MET protein was not detectable in LNCaP by Western blot,
whereas it was present in equivalent amounts in the other cell lines
(Fig. 1).
|
).
|
). After 6 days of
treatment with HGF alone, no growth inhibition was noted in LNCaP or
PC-3, but ALVA-31 and DU 145 were inhibited 49% and 35%, respectively
(Table 1). Vitamin D alone markedly inhibited LNCaP (71% decrease from
control), with much smaller effects on PC-3 (8% decrease) and ALVA-31
(42% decrease) and no significant effect on DU 145 (9% increase).
Although the addition of HGF together with vitamin D did not further
inhibit LNCaP, HGF addition did further inhibit each of the
androgen-unresponsive cell lines: PC-3 showed a 15% decrease with the
combination vs. an 8% decrease with vitamin D alone, DU 145
showed a 35% decrease with the combination vs. no
inhibition with vitamin D alone, and ALVA-31 manifested a 78% decrease
with the combination vs. 42% with vitamin D alone. Because
combining HGF with vitamin D caused as great an inhibition in the
androgen-unresponsive ALVA-31 cells as the marked inhibition seen with
vitamin D treatment of the androgen-sensitive LNCaP cells, we selected
ALVA-31 for further study.
|
).
Inhibition with the HGF-vitamin D combination was significantly more
potent than with either agent alone at all concentrations, with the
maximal effect (>85% decrease) of the combination always
significantly greater than that achieved by either agent alone.
Although no significant inhibition occurred with 1
nM vitamin D or 1 ng/ml HGF, combining the two
agents at these low concentrations produced a significant inhibition
(35% decrease). Ten-fold higher concentrations of HGF or vitamin D
were needed to achieve comparable inhibition with only one agent (Fig. 3).
|
HGF-vitamin D combination treatment alters cell cycle
distribution
Cells grown in 10 nM vitamin D or in 10 ng/ml HGF did
not show significant changes in cell cycle distribution. However,
with the HGF-vitamin D combination, the percentage of cells in
G1 decreased from 60 ± 3% to 50 ±
2.5%, and the fraction of cells in later cell cycle phases
(S/G2/M) increased from 39 ± 3.2% to
48 ± 1% (Fig. 4). Cell cycle
analysis performed at 2, 3, and 6 days of treatment failed to show any
lower DNA mass fluorescence peaks, which argues against DNA
fragmentation and apoptosis. The generalized increase in later cell
cycle phases suggests that the growth inhibitory effects involved cell
cycle slowing at points beyond the G1/S
checkpoint.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inhibition of prostate cancer cells with HGF seems paradoxical in view of its classical role as a paracrine growth stimulator (9) as well as previous investigations that suggested an adverse role for HGF in prostate carcinoma based on several findings (14, 15). First, MET expression appears to correlate with prostate tumor invasion and spread (15). Second, growth stimulation by HGF was reported in nonneoplastic rat prostate cells (14) and in one androgen-refractory (DU 145) human prostate cancer cell line (34). Third, increased invasion potential in the Matrigel invasion chamber assay was stimulated by HGF (38). Although we observed growth inhibition at higher plating densities, the greater effects on growth were seen at lower plating densities. However, the difference in cell density is unlikely to explain the finding of growth stimulation (34) or lack of inhibition (38) by HGF in DU 145 cells. The most likely reason for the difference from our results was the lack of serum in their experiments (34, 38). Therefore, we compared the effect of HGF on DU 145 cells grown under full serum and serum-free conditions. Despite an HGF-induced growth inhibition in the presence of serum, we found that HGF had no significant effect on the growth of DU 145 cells in serum-free medium (unpublished data). A similar discrepancy of growth effects in the presence and absence of serum has been reported for LNCaP cells with vitamin D treatment (4).
The lack of mitogenic, motogenic, and morphogenic responses of LNCaP cells to HGF was not surprising in view of the absence of MET RNA transcript and protein by RNase protection and Western blot, respectively. Although all three androgen-independent cell lines (ALVA-31, DU 145, and PC-3) showed equivalent MET activation (tyrosine phosphorylation) with HGF, PC-3 lacked the antimitogenic and morphogenic responses. A similar lack of proliferative and invasive response of PC-3 to HGF was reported by Nishimura et al. (38). Because signals given to epithelial cells by HGF are mediated through the MET receptor tyrosine kinase, the lack of response by PC-3 despite the receptor’s activation points to a defect(s) in the downstream signaling pathway, which might be true for other malignant cell types.
We selected ALVA-31 cells for studies of the HGF and vitamin D interaction because they show clear-cut growth inhibition with either agent alone. The cooperative interaction of HGF with vitamin D was revealed by the significant inhibition resulting from the combination of low doses of these agents, which by themselves exerted no effect on growth. In addition, although neither HGF nor vitamin D alone significantly altered cell cycle distribution, their combination did, with an accumulation of cells in S/G2/M phases. These results are consistent with earlier reports that despite significant antiproliferative effects on androgen-refractory cells, vitamin D alone did not alter cell cycle distribution (39). Our finding that the HGF-vitamin D combination treatment decreased the fraction of cells in G1 and increased the fraction in later cell cycle phases (S, G2, and M), in concert with the marked growth inhibition, suggests that the slowed cell cycle progression with these factors involves cooperative, but distinct, actions at loci beyond the G1/S checkpoint. Down-regulation of cyclin B and/or cyclin- dependent kinase-1, which regulate late cell cycle events, is a plausible mechanism at which future work will be directed. Slowing of cell cycle progression through these later phases was previously noted in T47D breast cancer cells by vitamin D, an agent considered to exert its inhibition via G1/S slowing (40). In our experiments, cell loss through apoptotic mechanisms seems unlikely, given the large inhibition observed and the lack of any evidence for apoptosis during serial microscopy or from flow cytometric analyses (33). Furthermore, despite the lower cell number under treatment conditions, the number of cells present at the end of the experiment still represents a 10-fold increase over the number of cells initially plated. The reversibility of the antiproliferative effect with treatment withdrawal also argues strongly against apoptosis.
Androgen-responsive prostate epithelial cells, such as LNCaP, are markedly and irreversibly inhibited by vitamin D, with decreased G1/S transit that results in accumulation of cells in the G1 phase of the cell cycle, suggesting G1/S arrest (39, 41). The development of vitamin D resistance in androgen-unresponsive prostate cancer cells could signal their diminished ability for G1/S checkpoint regulation. Although decreasing vitamin D receptor expression during tumor progression to androgen unresponsiveness could explain some of the apparent vitamin D resistance (42), neither vitamin D receptor loss nor diminished transcriptional activity fully accounts for this difference, because the androgen-unresponsive ALVA-31 cells have higher vitamin D receptor number and activity than the androgen-responsive LNCaP cells (7). Each androgen-unresponsive cell line has well-documented mutations (43, 44) involving Rb, p53, p21, and/or other antioncogenes critical to G1/S checkpoint control by vitamin D and other hormones (44, 45, 46). Such mutations may provide a reason for why vitamin D is less effective in inhibiting G1/S transit in androgen-unresponsive cell lines (39, 47, 48). Another reason is the loss of androgen modulation of cell cycle control. Restoring androgen receptor activity has been shown to reinstitute G1/S checkpoint control (49). Although no one regulatory or genetic alteration is likely to disrupt G1/S control, cumulative changes may result in loss of cell cycle control.
In conclusion, although most of the previous work on vitamin D and HGF-mediated growth inhibition involved effects on early cell cycle events, our results suggest that in androgen-insensitive prostate cancer cells, HGF and vitamin D can use distinct mechanisms to slow cell cycle progression at loci beyond the G1/S checkpoint. Prostate neoplasia is associated with increased MET (HGF receptor) immunostaining (15). Such heightened MET expression together with the paradoxical inhibition by HGF we observed in vitro might offer an opportunity to slow the progression of disease that has advanced in the face of androgen deprivation. Similarly, other pleiotropic growth factors might be used to inhibit aggressive prostate cancers (which lost control of G1/S) if they gain or retain their ability to inhibit later in the cell cycle. A combination of such agents with diverse, albeit minor, cooperative inhibitory actions at these later cell cycle loci might ultimately serve as an important adjuvant to vitamin D and other therapeutic agents in advanced prostate cancer. Notwithstanding the novel potential benefits our in vitro results imply, any clinical use of HGF in advanced prostate cancer will have to address problems attached to mode, duration, and cost of HGF administration (50) as well as potential adverse effects related to angiogenesis. The prospects for HGF as adjunct to vitamin D therapy are currently being addressed in preclinical xenograft studies.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 29, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,25-dihydroxyvitamin
D3 in seven human prostatic carcinoma cell lines.
Clin Cancer Res 1:997–1003[Abstract]
,25-dihydroxyvitamin D3 in the
human prostate cancer cell line LNCaP involves reduction of
cyclin-dependent kinase 2 activity and persistent G1 accumulation.
Endocrinology 139:1197–1207,25-Dihydroxyvitamin D3 actions in LNCaP
human prostate cancer cells are androgen-dependent. Endocrinology 138:3290–3298This article has been cited by other articles:
 |
|
 |
|
  N. Shinomiya and G. F. Vande Woude Suppression of Met Expression: A Possible Cancer Treatment: Commentary re: S. J. Kim et al., Reduced c-Met Expression by an Adenovirus Expressing a c-Met Ribozyme Inhibits Tumorigenic Growth and Lymph Node Metastases of PC3-LN4 Prostate Tumor Cells in an Orthotopic Nude Mouse Model. Clin. Cancer Res., 14: 5161-5170, 2003. Clin. Cancer Res., November 1, 2003; 9(14): 5085 - 5090. [Full Text] [PDF]
|
|
|
|
 |
|
 F. TOSETTI, N. FERRARI, S. DE FLORA, and A. ALBINI Angioprevention': angiogenesis is a common and key target for cancer chemopreventive agents FASEB J, January 1, 2002; 16(1): 2 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Gmyrek, M. Walburg, C. P. Webb, H.-M. Yu, X. You, E. D. Vaughan, G. F. Vande Woude, and B. S. Knudsen Normal and Malignant Prostate Epithelial Cells Differ in Their Response to Hepatocyte Growth Factor/Scatter Factor Am. J. Pathol., August 1, 2001; 159(2): 579 - 590. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
