| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Urology, University of Texas Southwestern Medical Center, Dallas, Texas 75235-9110
Address all correspondence and requests for reprints to: Dr. Jer-Tsong Hsieh, Department of Urology, University of Texas Southwestern Medical Center, Dallas, Texas 75235-9110. E-mail: hsieh{at}utsw.swmed.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is postulated that luminal epithelium and basal epithelium may share a common lineage during androgen-induced epithelial differentiation (8). Biochemically and physiologically, the effects of androgen on the different subsets of epithelial cells have not yet been clearly defined. However, in prostate gland, androgen deprivation is able to enrich the basal cell population that may contain a stem cell population because of its tremendous proliferation potential (9, 10). To discover any novel androgen-regulated genes associated with the basal cell compartment, we apply differential display-PCR (DD-PCR) (11) to the rat ventral prostate (VP) tissues obtained from both intact and castrated hosts. In this report we describe the isolation, sequencing, and characterization of a novel gene (i.e. C9) from rat VP. C9 is a homolog of both human DOC-2 (deletion in ovarian carcinoma) (12, 13) and mouse p96 phosphoprotein (14), of which the function and regulation are still unknown.
From sequence alignment analysis, DOC-2 belongs to the disabled (dab) gene family. During neuronal differentiation of P19EC cells, expression of dab is induced, and the protein is tyrosine phosphorylated before neurite extension. The expression pattern and phosphorylation of dab are also regulated during mouse embryonic development (15). Mutation of the dab gene in Drosophila has an embryonic lethal phenotype that failed to form a proper axonal connection in the central nervous system (16). Analysis in yeast and in vitro showed that tyrosine-phosphorylated mouse Dab protein binds strongly to the SH2 domains of Src and Fyn, which have been associated with cell transformation (15). The first 180 amino acid residues of the N-terminus of Dab share high homology to the DOC-2 complementary DNA (cDNA) sequence, suggesting that DOC-2 may be a potential signal molecule.
In this study, we found that C9 is up-regulated during androgen deprivation, and the regulation appears to be tissue specific. C9 protein is present in the basal epithelium and surrounding stromal cells of the degenerated prostate gland. By stable transfection of C9 cDNA into C4-2 cells (a derivative of the LNCaP cell line), we demonstrated that C9 inhibits the growth of prostatic epithelial cells in vitro. The potential functional implications of the C9 gene in prostatic glandular development and pathogenesis are discussed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture
LNCaP, C4-2, and NbE cells (18) were routinely cultured in T
medium (80% DMEM, 20% Ham’s F-12, 3 g/liter NaHCO3, 100
U/ml penicillin G, 100 µg/ml streptomycin, 5 µg/ml insulin, 13.6
pg/ml T3, 5 µg/ml transferrin, 0.25 µg/ml
biotin, and 25 µg/ml adenine) containing 5% FBS.
Animal experiments
Male Sprague-Dawley rats, six rats for each group, were
surgically castrated by scrotal incision and maintained under a
standard laboratory environment. Twenty-four hours after orchidectomy,
rats were injected im with sesame oil, testosterone propionate (TP; 500
µg/day), or TP plus 4-hydroxyflutamide (F; 5 mg/day) for a total of 4
days. Rats were killed by CO2 asphyxiation, and their
organs were excised for RNA preparation. For the time-course
experiment, castrated rats were killed on days 1, 5, 10, and 20
postcastration.
RNA purification and DD-PCR
Total cellular RNA was extracted from intact or castrated hosts
as described previously (19) and then digested with deoxyribonuclease I
at 37 C for 1 h. The deoxyribonuclease I-treated RNA (0.2 µg)
were reverse transcribed by Superscript II reverse transcriptase with
T7(dT12)AP4 primers (5'-ACGACTCACTATAGGGCTTTTTTTTTTTTGT-3'). One tenth
of the RT reaction was subsequently amplified by PCR using both
T7(dT12)AP4 primer and M13r-ARP1 primer
(5'-ACAATTTCACACAGGACGACTCCAAG-3') for 4 cycles of 92 C (2 min), 46 C
(30 sec), and 72 C (2 min); 25 cycles of 92 C (15 sec), 60 C (30 sec),
and 72 C (2 min); and finally 1 cycle of 72 C (7 min). The PCR
reactions were performed with [
-35S]deoxy-ATP (1000
Ci/mmol; 10 mCi/ml) and the labeled PCR products were fractionated on a
6% polyacrylamide sequencing gel using the GenomyxLR DNA sequencer
(Genomyx). The candidates for androgen-repressed genes were recovered
from the castrated sample lane and reamplified by PCR using M13 reverse
24-mer primer (5'-AGCGGATAACAATTTCACACAGGA-3') and T7 promoter 22-mer
primer (5'-GTAATACGACTCACTATAGGGC-3') for 4 cycles of 92 C (15 sec), 50
C (30 sec), and 72 C (2 min); 25 cycles of 92 C (15 sec), 60 C (30
sec), and 72 C (2 min); and 1 cycle of 72 C (7 min). The final PCR
products were cloned into the pT7blue-2 PCR cloning vector and
sequenced using the Thermo Sequenase radiolabeled terminator cycle
sequencing kit.
Northern blot analysis
Northern blot analysis was performed as described previously
(7). Briefly, equal amounts of total cellular RNAs (20 µg) were
fractionated by electrophoresis on 0.9% agarose containing 2
M formaldehyde and then transferred onto a Zeta-Probe
membrane (Bio-Rad, Richmond, CA). The blot was hybridized with the
random primer-labeled cDNA probe. Autoradiography was performed using
X-Omat (Eastman Kodak, Rochester, NY) with an intensifying screen at
-80 C.
Isolation of full-length C9 cDNA
C9 cDNA was isolated by RT-PCR with total cellular RNA from VP
of castrated rats. Briefly, the following primers corresponding to
mouse p96 phosphoprotein (P1–P6) or rat C9 (P7) were synthesized: P1,
5'-CCCGTCATGTCTAACGAAGT-3'; P2, 5'-GGGATAATGGCTATGGAGTC-3'; P3,
5'-CCTCCTGACCTAAATAGTCCA-3'; P4, 5'-CTCTGAAAAGGATTCCCCAG-3'; P5,
5'-TCGTTTTTGGTACAACCCCAGCAG-3'; P6,
5'-TGGTCTACATGCTCCTGAGGAATGC-3'; and P7,
5'-ATTGCCTTATGTTTTGAGTTAGACC-3'. The first strand cDNA was
synthesized using Superscript II reverse transcriptase with primers P2,
P4, P6, and P7, respectively. Then one fifth of the RT reactions were
amplified with PCR using the primer sets P1 and P2, P3 and P4, P5 and
P6, and P5 and P7, respectively, for 4 cycles of 92 C (15 sec), 42 C
(30 sec), and 72 C (2 min); 30 cycles of 92 C (15 sec), 55 C (30 sec),
and 72 C (2 min); and 1 cycle of 72 C (7 min). The final PCR products
were cloned into the pT7blue-2 PCR cloning vector and sequenced using
the Thermo Sequenase radiolabeled terminator cycle sequencing kit. At
least three independent clones of each PCR product were sequenced to
eliminate any potential PCR infidelity. The consensus sequence of each
cDNA clone was assembled to generate the full-length C9 cDNA
sequence.
Preparation of antisense riboprobe and RNase protection
assay
To construct a plasmid (PBK-C9-AS) for synthesizing antisense
riboprobe, a 566-bp Bsu36I-PvuII fragment of C9
from 250–815 nucleotides was blunt ended and subcloned into the
SmaI site of the pBluescript plasmid. For RNase protection
assay, either 20 µg total cellular RNA mixed with 30 µg yeast RNA
or 50 µg yeast RNA (negative control) were precipitated with 40,000
cpm of an [-32P]CTP-labeled antisense probe (Amersham,
Arlington Heights, IL). We then added 10 µl preheated (95 C) HybSpeed
Hybridization Buffer to each sample and immediately submerged the
samples in a 95 C water bath to completely dissolve the RNA pellet.
Afterward, the samples were incubated at 68 C for the hybridization
reaction. Ten minutes later, 100 µl RNase A/T1 Mix in HybSpeed RNase
Digestion Buffer (1:100 dilution) were added and further incubated at
37 C for 30 min to digest any unprotected RNA fragments. After the
completion of RNase digestion, 150 µl HybSpeed
Inactivation/Precipitation Mix were added to precipitate the protected
RNA fragments at -20 C for 30 min. The pellet was then dissolved in
the gel loading buffer and fractionated on a 5% denaturing
polyacrylamide gel.
Generation of C9 antibody
The peptide sequence corresponding to the rat C9 amino acid
residues 705–721 plus an extra cysteine, (C)NQLLNKINEPPKPAPRQ, was
synthesized and used as the antigen to inject into rabbits for
generation of polyclonal antibody by Zymed Laboratory (San Francisco,
CA). To affinity purify the C9 antibody (C9-Ab705), the C9 peptide was
coupled to SulfoLink coupling gel through the sulfhydryl group of the
N-terminal cysteine residue. The rabbit serum containing C9 antibody
was passed through the C9 peptide coupling column and eluted with
ImmunoPure gentle antigen/antibody elution buffer (Pierce).
Western blot analysis
The expression of protein was detected by the enhanced
chemiluminescence method as described previously with some modification
(20). Briefly, cell or tissue lysates were fractionated on a 7.5%
SDS-polyacrylamide gel. After electrotransferring to a nitrocellulose
membrane, the nonspecific binding sites were blocked with 5% dry milk
in PBS containing 0.1% Tween-20 (PBS-T). Then, the membrane was
incubated with the primary antibody for 1 h. The horseradish
peroxidase-labeled secondary antibody was added and incubated for
addition 1 h at room temperature. For autoradiography, film was
exposed as described in the manufacturer’s protocol.
Immunoprecipitation
NbE cells were scraped into 450 µl lysis buffer B [50
mM HEPES (pH 7.5), 150 mM NaCl, 10% glycerol,
1% Triton X-100, 1.5 mM MgCl2, 10 µg/ml
aprotinin, 10 µg/ml leupeptin, 1 mM
phenylmethylsulfonylfluoride, 200 µM sodium
orthovanadate, 0.01 M sodium fluoride, and 1 mM
EGTA] and kept on ice for 30 min (20). After centrifugation, the
supernatants were collected as total cell lysates. The
immunoprecipitation reaction (50 µl protein A/G plus agarose, 450
µl cell lysates, and various amount of C9-Ab705) was performed at 4 C
for 2 h. After three washes with lysis buffer B, the pellets were
dissolved in 1 x SDS sample buffer for Western blot analysis.
Immunohistochemistry
The excised rat prostate tissues were immediately placed in
tissue blocks with OCT compound (Miles, Elkhart, IN) and frozen at -20
C. For immunostaining, tissue sections (5 µm) were placed on
polylysine-coated slides and fixed in a solution containing 95%
methanol and 5% acetic acid at -20 C for 20 min. After rinsing three
times with PBS, the slides were placed in a Superblock solution (Scytek
Laboratories, Logan, UT) for 20 min. Then, the slides were incubated
with affinity-purified polyclonal antibody C9-Ab705 or preimmune serum
overnight at 4 C. The endogenous peroxidase activity was blocked with
3% H2O2 in absolute methanol for 30 min. The
tissue sections were further incubated with a peroxidase-conjugated
goat antirabbit secondary antibody (1:500 dilution in 5% dry milk in
0.01% PBS-T) for 1 h. After several washes with PBS-T buffer, the
enhanced BLACK buffer solution containing diaminobenzidine was used as
a chromogen.
Construction and expression of C9 expression plasmid
A full-length cDNA of C9-p82 was first reconstituted in pT7-blue
2 vector. Briefly, approximately 900 bp of
Bsu36I-BstXI-digested C9-8, about 300 bp of
BstXI-AccI-digested C9-B30, and about 1000 bp of
AccI-HincII-digested C9-B1 were ligated to the
Bsu36I-SmaI-digested pT7blue2-C9-4 resulting in a
plasmid with full-length C9-p82 cDNA (pT7blue-2-C9-p82). To identify
the C9 protein product, the in vitro
transcription/translation reaction was performed as described in the
manufacturer’s protocol. In addition, the mammalian expression plasmid
pC9-p82-neo was obtained by ligation of the approximately 2400-bp
XbaI-PshAI-digested pT7blue2-C9-p82 and the
XbaI-SmaI-digested pCI-neo vector. To construct
the pC9-p59-neo, the AccI-PflMI fragment of
pC9-p82-neo containing p82 cDNA sequence was replaced by the
AccI-PflMI fragment of C9-8-1.
Establishment and characterization of C9 stable
transfectants
C4-2 cells (2 x 105/p35 plate) were
transfected with 2 µg of pCI-neo or pC9-p59-neo by lipofectamine
transfection reagent. Forty-eight hours after transfection, cells were
split 1:10 and were selected for neomycin-resistant clones by 800
µg/ml G-418. Resistant colonies were cloned by ring isolation after 3
weeks of selection.
For determining the in vitro cell growth rate, cells were plated in 24-well plates at a density of 5000 cells/well in 0.5 ml T medium plus 2% TCM (a serum-free defined medium supplement, Celex Co., Minnetonka, MN) and 0.5% FBS. Twenty-four hours later, 0.5 ml T medium plus 2% TCM were added to each well. On the indicated days, cells were subjected to crystal violet assay (21, 22).
For cell cycle analysis, cells were plated at a density of 1.5 x 105 cells/p100 plate in T medium plus 2% TCM and 0.5% FBS. Twenty-four hours later, the medium was replaced with T medium plus 2% TCM and 0.25% FBS. On the seventh day after plating, cells were trypsinized, resuspended in PBS, and fixed with ethanol. The cell suspension was passed through a needle (23-gauge) six to eight times and spun for 5 min at 1,000 rpm. The cell pellet was washed with PBS and resuspended in pepsin solution (0.04% pepsin in 0.1 N HCl). After centrifugation, the cell pellet was further treated with 2 N HCl for 30 min, neutralized with 0.1 M sodium borate, and washed with PBTB solution (PBS containing 0.5% Tween-20 and 0.2% BSA). After the final spin, cells were resuspended in 1 ml PBTB solution containing 10 µg propidium iodide and 5 µg RNase and subjected to flow cytometric analysis. A total of 10,000 events were measured for each sample preparation.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). After excising and eluting from the
gel, we reamplified the C9 fragment and subcloned it into a pT7blue-2
PCR cloning vector. Sequence analysis indicated that the 254-bp C9 cDNA
fragment shared a significant homology (93%) to the 3'-untranslated
region of the mouse p96 phosphoprotein. Based on this information, we
performed RT-PCR to amplify the coding region of rat C9 cDNA using the
upstream primer (P5) from the mouse p96
(5'-TCGTTTTTGGTACAACCCCAGCAG-3') sequence and the downstream primer
(P7) from the C9 sequence (5'-ATTGCCTTATGTTTTGAGTTAGACC-3'). A 1.0-kb
cDNA fragment (B1), as shown in Fig. 2A, was confirmed by sequencing and used as a cDNA probe. A significant
increase in C9 transcripts with approximate sizes of 3.7 kb
(i.e. p82) and 3.0 kb (i.e. p59) was detected in
VP derived from the castrated host (Fig. 1B). These data suggest that
C9 may be regulated by androgen directly or indirectly in rat VP.
|
|
). Then, these primers were used for RT-PCR to generate a
full-length rat C9 cDNA, 3170 bp in size (Fig. 2B). The 2304-bp open
reading frame (ORF) was predicted to encode a protein containing 768
amino acid residues with a calculated molecular mass of 82 kDa (p82)
and a pI of 5.75. The sequence of C9 cDNA is more than 90% identical
to that of mouse p96 protein (14). Noticeably, two additional amino
acid residues (i.e. glycine and serine) were found between
rat nucleotides 1483–1488 compared to the mouse p96 (14). Rat C9 also
shares a significant homology with the human DOC-2 cDNA (12, 13); the
N-terminal amino acid residues share 91% homology between these two
species. In contrast, the diversity is more visible (
57–74%
homology) in the C-terminus (Fig. 2C).
In addition to p82, we obtained a cDNA fragment that had the deletion
between nucleotides 694-1347 from RT-PCR using the primer set P3 and P4
(Fig. 2A). Upon assembly into a full-length cDNA, it contains a 2516-bp
sequence with a 1650-bp ORF (Fig. 2B). This ORF was predicted to encode
a protein containing 550 amino acid residues with a calculated
molecular mass of 59 kDa (p59) and a pI of 5.56. This cDNA clone is
similar to the mouse p67 isoform, which may be derived from the
alternative splicing of C9 mRNA (14). Taken together, our data indicate
that the rat VP expressed at least two C9 isoforms.
Both C9 isoforms are up-regulated during prostate
degeneration
It is known that androgen is a major regulator of prostate
degeneration/regeneration. To characterize the effect of androgen on
the induction of C9 gene expression, castrated rats were injected daily
with TP (500 µg/day) or with TP plus the antiandrogen F (5 mg/day)
24 h after castration for a total of 4 days. As shown in Fig. 1B, TP administration suppressed C9 gene expression to its basal level.
However, F can antagonize the TP effect by maintaining elevated C9 mRNA
levels, as seen in the VP of castrated rats. On the same blot, both a
known androgen-inducible gene (i.e. C3) (26) and an
androgen-repressed gene (i.e. TRPM-2) (4) were used as the
internal controls. Therefore, these data suggested that up-regulation
of C9 gene expression paralleled the prostate degeneration process.
To further study the regulation of C9 isoforms in rat VP, RNase
protection assay was performed to determine whether C9 isoforms have
any different response to castration-induced prostate degeneration. The
BssSI-digested fragment of the PBK-C9-AS plasmid containing
Bsu36I-BssSI C9 cDNA fragments was used as the
template to synthesize the C9 antisense riboprobe (Fig. 2A). With this
riboprobe, both p82 and p59 transcripts can be detected clearly by the
difference in size of the protected fragments. For example, 402- and
280-bp protected fragments were expected for p82 and p59, respectively.
As shown in Fig. 3, we found that both
p82 and p59 RNA transcripts in VP were expressed significantly higher
in the castrated host than in the intact animal, indicating that both
p82 and p59 were up-regulated during prostate degeneration. We also
noticed that p59 mRNA appeared to be the predominant isoform in VP.
|
, the steady state levels of C9
mRNA were not detected during the first 24 h after castration.
However, the expression of C9 mRNA peaked on day 5 postcastration, and
the elevated levels of C9 mRNA were persistent throughout the entire
castration period (i.e. 20 days postcastration). The time
course of any gene activities, such as TRPM-2, involved in the
apoptosis of degenerated prostate should be exhibited in a transient
expression pattern; that is, no signal can be detected 1 week
postcastration. In contrast, the patterns of both C9 mRNA and protein
expression are very different from that of TRPM-2 expression,
suggesting that C9 protein may have a physiological function other than
apoptosis in the degeneration process of the prostate gland.
Furthermore, an androgen-induced gene, C3, showed a dramatic decrease
in expression shortly after castration.
|
, the expression of p59
protein was elevated during castration and reached the highest level on
day 20 postcastration. The p82 protein is barely detectable in Western
blot analysis, indicating that p82 was a minor isoform in VP,
consistent with the data from Northern blot analysis. Taken together,
these results are consistent with the Northern blot analysis and
indicate that up-regulation of C9 may not be associated the apoptotic
luminal cells during prostate degeneration.
Expression of C9 is predominantly associated with basal prostatic
epithelial cells
It is known that the prostate gland is composed of both epithelial
and stromal compartments. However, androgen deprivation changes the
epithelial/stromal ratio dramatically (28). Therefore, it is critical
to determine which cell types express the C9 gene in degenerated VP. An
attempt to use the p96 antibody (Transduction Laboratories) for
immunostaining resulted in weak or no staining. Therefore, we decided
to generate a rabbit polyclonal antibody (Ab705) against the C9 peptide
sequence corresponding to amino acid residues 705–721.
Immunoprecipitation of the prostatic epithelial NbE cell extracts with
Ab705 followed by immunodetection with the p96 antibody resulted in two
major protein bands corresponding to the p82 and p59 isoforms (Fig. 5A). After affinity purification of the
rabbit antiserum, Ab705 reacted with two proteins from NbE cell
extracts with molecular masses identical to that observed with the p96
antibody (Fig. 5B). We also detected an additional protein with a
molecular mass lower than that of p59, suggesting that an additional C9
isoform may be present in rat VP. In the presence of increasing amounts
of synthetic peptide antigen, Ab705 failed to detect both p82 and p59
in NbE cells (Fig. 5C). However, Ab705 detected the expression of our
C9 cDNA expression plasmids in LNCaP cells (data not shown). Therefore,
these results confirmed the specificity of Ab705 to C9 protein.
|
, A and B, C9 was detected weakly
in the columnar luminal epithelial cells of the intact host. On the
other hand, no staining was detected in the stroma. On day 5
postcastration (Fig. 6, C and D), the luminal epithelial cells
underwent apoptosis. The expression of C9 in VP appeared to localize in
the remaining cuboidal basal cells. Furthermore, an elevated C9
staining intensity (Fig. 6, E and F) was observed in the basal cells
after prolonged castration (i.e. 20 days). In these
remaining basal cells, we noticed a heterogeneous staining of C9,
suggesting that there were at least two subsets of basal cell
population expressing different levels of C9. We also detected weak
staining in the stromal cells surrounding the basal cells, suggesting
that C9 may be involved in prostatic epithelial-stromal cells
interaction. Consistent with Western blot analysis, these data
indicated that C9 protein increased during prostate degeneration and
the pattern of C9 expression switched from the luminal epithelia to the
remaining basal cells and surrounding stromal cells after a mass loss
of luminal cells induced by castration.
|
, A and B, similar patterns of C9 gene
expression were observed in both SV and DLP, but not in CG. However, in
androgen-independent organs such as kidney, we observed an abundant C9
mRNA in both intact and castrated rats. Overall, we failed to detect
any effect of androgen on C9 gene expression in kidney (Fig. 7) and
other androgen-independent organs (data not shown). On the same blot,
the glyceraldehyde-3-phosphate dehydrogenase gene was used as the
internal control for equal loading. Taken together, these data
indicated that the regulation of C9 gene by androgen was tissue
specific.
|
, we detected the expression of p82 and
p59 proteins from both pC9-p82-neo and pC9-p59-neo, respectively. Both
proteins comigrated with the C9 protein endogenously expressed in NbE
cells, a prostatic epithelial cell line derived from the VP of the
Noble rat.
|
). Forty-eight hours after transfection, we harvested
cell extracts from the control (pCI-neo) and C9 transfected cells and
analyzed the expression of C9 protein by Western blot analysis. As
shown in Fig. 8B, we detected the expression of p82 and p59 from
pC9-p82-neo- and pC9-p59-neo-transfected cells, respectively. Taken
together, these data indicated that these two C9 cDNA constructs should
be functional, and the protein products corresponded to the endogenous
C9 protein expressed in prostatic epithelial cells.
Inhibitory effect of p59 on the growth of prostatic epithelial
cells
Recent data suggest that C9 may be a potential tumor suppressor in
human ovarian carcinoma (12). With respect to the predominant
expression of p59 in rat VP, we decided to transfect the pC9-p59-neo
vector into the C4-2 cells, a tumorogenic subline derived from LNCaP
cells (17), to study its function. Basically, p59 showed an adverse
effect on cell survival; the majority of cells failed to grow. However,
we were able to select two individual clones (p59-18 and p59-23), based
on their elevated C9 levels (Fig. 9A) and
different clonal origins (Fig. 9B), for the in vitro growth
assay. As shown in Fig. 9C, both p59-18 and p59-23 clones showed a
slower growth rate compared with those of the plasmid control cells
(neo-4) and parental C4-2 cells. The doubling times of p59-18 and
p59-23 under the same experimental conditions were significantly longer
than those of the control cells (60 and 70 h vs. 43 and
40 h; Table 1). To determine the
impact of p59 expression on the cell cycle transition of C4-2 cells, we
determined the cell cycle distribution of these transfectants by flow
cytometric analysis. Consistent with the doubling time, p59-18 and
p59-23 clones showed an increase in cell number in the
G1/G0 phase (68% and 79%, respectively; Table 1). Interestingly, after continuous passage for 2 months, we obtained a
revertant (i.e. p59-18R) from the p59-18 clone with no
detectable p59 (Fig. 9A). The in vitro growth rate of this
revertant appeared to be the same as that of the parental C4-2 cells
(Fig. 9C). These data indicated that C9 can inhibit the growth of
prostatic epithelial cells in vitro.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, we first observed that regulation of the C9 gene was mediated by androgen. However, we cannot rule out the possibility that this androgen effect is indirect. We also showed that C9 is expressed weakly in the luminal epithelial cell compartment of intact rat prostate. However, unlike those apoptosis-related genes associated with luminal cells, the expression of C9 switched to the remaining basal cells of the degenerated prostate during castration. Recently, we have shown a similar expression pattern with other genes, such as cytokeratin 8 (6) and C-CAM1 (7). These genes are consistently detected in the basal cells of the degenerated prostate gland after prolonged castration. Therefore, C9 and this subset of genes may have a unique functional role in controlling the homeostasis of prostatic epithelium. In addition, C9 was detected in the stromal compartment of the prostate after androgen deprivation. It is known that stromal cells play a role in development and glandular formation in prostatic epithelia (33). Therefore, it is likely that C9 may be a common denominator underlying the stromal-epithelial interaction of prostate development.
Although the physiological role of C9 in prostate development is not clear, sequence analysis of C9 reveals that, overall, more than 80% sequence homology exists between rodent and human C9 (i.e. DOC-2) cDNA. Such evolutionary conservation in sequence indicates its potential role in maintaining normal cellular physiology. Mok et al. showed loss of DOC-2 expression in human ovarian cancer (12). Our preliminary data also indicated that C9 protein was often undetected or down-regulated in several human prostate cancer lines (data not shown), whereas restored C9 expression decreased the in vitro growth rate of prostate cancer cells. This experimental evidence indicated that C9 may be a negative growth regulator and a potential tumor suppressor. With respect to the presence of C9 in the basal cells of the degenerated prostate, we hypothesize that C9 plays a homeostatic role in preventing the outgrowth of prostatic epithelium. Therefore, alteration of C9 may potentially initiate hyperplastic growth. A detailed analysis of C9 expression in both developing prostate and prostate cancer specimens is warranted.
How C9 affects prostatic cell proliferation is not completely understood. The increase in the G1 cell population of our C9-p59 stable transfectants suggests that C9 may play a role in the signal transducing pathway leading to cell arrest. Structural analysis of C9 protein motifs indicates that C9 may interact with another molecule(s) to elicit its activity. For example, the C-terminus of C9 has several proline-rich sequences (i.e. residues 504–512, residues 619–624, residues 663–667, and residues 714–719) that have the potential to interact with proteins bearing the SH3 domain (34). Such interaction is known to be a critical step in the signaling network elicited by protein tyrosine kinase receptor (35). Therefore, it is likely that the adapter protein bearing the SH3 domain(s), such as Grb2, may be sequestered by C9, resulting in intervention of the epidermal growth factor receptor signaling pathway. This interaction may explain the growth inhibitory effect of C9 in the p59 stable transfectants. Several potential ERK phosphorylation sites mapped on C9 (i.e. residues 504–507 and residues 512–515) also suggest that C9 may be a substrate for mitogen-activated protein kinase, which is known to mediate the mitogenic signal into the cell nucleus (36). Moreover, the first 180 amino acid residues of the N-terminus share 54% homology to mouse and Drosophila disabled (Dab) protein, which has been implicated in neural regeneration (15, 16). Noticeably, this region contains a phosphotyrosine interaction domain that may interact with other tyrosine-phosphorylated proteins (37). Therefore, it will be worthwhile to characterize the putative C9 interacting protein(s) and gain insight into the role of C9 in cell cycle regulation.
Other than the protein-protein interaction leading to signal transduction, protein phosphorylation is another important regulatory mechanism of signaling (38). Xu et al. has shown that colony-stimulating factor-1 and several mitogens induced p96 phosphorylation in a mouse macrophage cell line (14). Similarly, stimulation of quiescent NbE cells by epidermal growth factor and other growth factors changes the phosphorylation status of C9 protein (Tseng, C.-P., unpublished data). These data suggest a strong possibility that phosphorylation is operative in prostatic epithelial cells under mitogenic stimulation. These observations suggest that the activity of the C9 protein may be regulated at least partially by other uncharacterized protein kinase(s)/phosphatase(s). Detailed characterizations of these phosphorylation events are essential for understanding the regulation and function of C9.
From the RNase protection assay and Western blot analysis, our study indicates that at least two C9 isoforms are expressed in the normal prostate gland and in a nontumorogenic rat prostatic epithelial cell line (NbE). In addition to the expected p82 and p59 isoforms detected by the RNase protection assay, we observed an additional band at approximately 120 bp from the castrated sample. It is possible that an additional isoform of C9 may be present in the degenerated VP. Xu et al. reported that two alternative splicing forms of mouse p96 (i.e. p93 and p67) were found during mouse cDNA library screening (14). The p67 is similar to the p59 found in rat VP. In contrast, we did not detect p93; if this isoform exists, we should have detected two fragments (i.e. 220 and 121 bp) with the riboprobe used in the RNase protection assay. However, we only observed a 120-bp band from the autoradiogram, suggesting that there may be a different isoform present in rat VP. Similarly, the C9 protein in NbE cells detected by Western blot revealed that an additional protein with a molecular mass smaller than that of p59 immunoreacted with the Ab705 antibody. We are currently identifying this isoform, which may arise from alternative splicing of the C9 gene or another member of the C9 gene family. As the impact of alternative splicing on the function of gene products has been well documented (39, 40, 41), it is conceivable that each isoform may have distinct biological functions. Further characterization of individual C9 isoforms in VP is warranted.
In summary, we have isolated and characterized two full-length C9 isoform cDNAs that have a unique pattern of gene regulation in rat VP and other urogenital organs. We have also addressed the potential function of C9 in the control of prostatic epithelial cell proliferation. Our study suggests that C9 may be a negative growth regulator in prostate development and carcinogenesis. Further characterization of the C9 molecule may increase our understanding of the mechanisms underlying prostate gland homeostasis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 18, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J.-T. Hsieh, J. A. Karam, and W. Min Genetic and Biologic Evidence that Implicates a Gene in Aggressive Prostate Cancer J Natl Cancer Inst, December 19, 2007; 99(24): 1823 - 1824. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Holmes, A. Flett, D. Coudreuse, H. C. Korswagen, and J. Pettitt C. elegans Disabled is required for cell-type specific endocytosis and is essential in animals lacking the AP-3 adaptor complex J. Cell Sci., August 1, 2007; 120(15): 2741 - 2751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Karam, S. F. Shariat, H.-Y. Huang, R.-C. Pong, R. Ashfaq, E. Shapiro, Y. Lotan, A. I. Sagalowsky, X.-R. Wu, and J.-T. Hsieh Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma of the Bladder Clin. Cancer Res., August 1, 2007; 13(15): 4400 - 4406. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Romero, L. L. Yanes, A. F. de Rodriguez, M. W. Plonczynski, B. L. Welsh, J. F. Reckelhoff, E. P. Gomez-Sanchez, and C. E. Gomez-Sanchez Disabled-2 Is Expressed in Adrenal Zona Glomerulosa and Is Involved in Aldosterone Secretion Endocrinology, June 1, 2007; 148(6): 2644 - 2652. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. H. Kumbar, A. VanBergen, C. Ocampo, S. Muangmingsuk, A. J. Griffin, and M. Gupta Adapter molecule DOC-2 is differentially expressed in pressure and volume overload hypertrophy and inhibits collagen synthesis in cardiac fibroblasts J Appl Physiol, May 1, 2007; 102(5): 2024 - 2032. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-L. Huang, J.-C. Cheng, A. Stern, J.-T. Hsieh, C.-H. Liao, and C.-P. Tseng Disabled-2 is a novel {alpha}IIb-integrin-binding protein that negatively regulates platelet-fibrinogen interactions and platelet aggregation J. Cell Sci., November 1, 2006; 119(21): 4420 - 4430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-C. Pong, R. Roark, J.-Y. Ou, J. Fan, J. Stanfield, E. Frenkel, A. Sagalowsky, and J.-T. Hsieh Mechanism of Increased Coxsackie and Adenovirus Receptor Gene Expression and Adenovirus Uptake by Phytoestrogen and Histone Deacetylase Inhibitor in Human Bladder Cancer Cells and the Potential Clinical Application. Cancer Res., September 1, 2006; 66(17): 8822 - 8828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.-H. Yang, Z. Fazili, E. R. Smith, K. Q. Cai, A. Klein-Szanto, C. Cohen, I. R. Horowitz, and X.-X. Xu Disabled-2 Heterozygous Mice Are Predisposed to Endometrial and Ovarian Tumorigenesis and Exhibit Sex-Biased Embryonic Lethality in a p53-Null Background Am. J. Pathol., July 1, 2006; 169(1): 258 - 267. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Zhoul, G. Hernandez, S.-W. Tu, C.-L. Huang, C.-P. Tseng, and J.-T. Hsieh The Role of DOC-2/DAB2 in Modulating Androgen Receptor-Mediated Cell Growth via the Nongenomic c-Src-Mediated Pathway in Normal Prostatic Epithelium and Cancer Cancer Res., November 1, 2005; 65(21): 9906 - 9913. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
