|
|
|
|||||||
|
|||||||||
(Received for publication, June 19, 1995; and in revised form, July 11, 1995) From the
Our previous studies have characterized mesenchyme-derived
proteins to identify biologically active proteins and novel markers for
stromal cell paracrine action relative to stromal-epithelial
interactions. Previous reports have characterized properties of a
growth inhibitory activity (to bladder and prostatic epithelial cells),
secreted by U4F fetal rat urogenital sinus mesenchymal cells, not
cross-reactive with antibodies to known cytokines, and provisionally
termed UGIF. The present study reports the characterization,
purification, and biological properties of a 20-21-kDa protein
responsible for UGIF activity. The 20-21-kDa protein (termed
ps20) was purified to near homogeneity, the amino-terminal sequence was
determined, and biological properties were characterized in
vitro. Amino-terminal sequence analysis indicated no direct
matches or regions of homology with known proteins. Purified ps20
induced a linear and saturable inhibition of
thymidine incorporation in PC-3 prostatic
carcinoma cells (half-maximal activity at 2.6 nM), inhibited
cell proliferation (increased population doubling time from 19.8 to
25.8 h), and induced a 210% stimulation in the synthesis of secreted
proteins. These data suggest that ps20 may be a candidate paracrine
effector protein and may play a role in stromal-epithelial cell
interactions in the prostate gland.
The induction of epithelial cell growth and differentiation
patterns by adjacent mesenchymal cells is a common feature in the
organogenesis of many tissues. Such interactions occur in reproductive
tissues including seminal vesicle development(1, 2) ,
in the androgen-induced regression of male mammary gland(3) ,
and in estrogen-induced proliferation of mammary gland
epithelial(4) , vaginal epithelial, and uterine epithelial
cells(5) . Stromal-epithelial interactions have been
particularly well studied in the morphogenesis of fetal urogenital
sinus (UGS) ( In postnatal and
fully differentiated adult tissues, stromal-epithelial interactions
likely maintain histological architecture and differentiated phenotype
relative to ongoing modeling and remodeling processes(6) .
Fully differentiated adult epithelial cells are capable of responding
to a heterotypical (different) stroma. In this regard, adult bladder
transitional epithelium can be induced by urogenital sinus mesenchyme
to change to a prostate-specific epithelial cell phenotype including
the expression of prostate-specific proteins, including androgen
receptor(14, 15) . In addition, abnormal patterns of
epithelial cell proliferation and differentiation in neoplastic disease
progression are also affected by the origin and type of adjacent
stroma. The growth of prostatic Dunning tumor adenocarcinoma was
inhibited by 7-fold, and the morphology was altered to a more normal
phenotype when recombined with UGS mesenchyme and grown in
vivo(16) . Similarly, the recombination of normal UGS
mesenchyme with bladder transitional cell carcinoma resulted in a
change of tumor histopathology to an adenocarcinoma
phenotype(17) . Moreover, the implantation of UGS mesenchyme
directly into the adult mouse prostate gland resulted in an induced
hyperplastic phenotype typical of benign prostatic
hyperplasia(18) . These studies together support the suggestion
that stromal-epithelial interactions are likely to be central in
modulating the progression and histopathology of epithelial cells in
proliferation- and differentiation-related diseases such as cancer and
benign hyperplasia(19) . Although stromal cell induction of
epithelium is critical for organogenesis and growth control in adult
tissues, little is understood regarding paracrine effector molecules,
fundamental molecular mechanisms of stromal-epithelial interactions,
and the basic biology of UGS mesenchymal cells. Indeed, there is
currently no well defined set of parameters or markers to clearly
identify the stages of mesenchymal cell differentiation to adult
stromal cells (fibroblasts, myofibroblasts, or smooth muscle), or the
specific roles of these cell types in continued stromal-epithelial
interactions. To identify putative mediators and markers of UGS
mesenchymal cell action, we have reported previously the development of
organ cultures and mesenchymal cell lines from fetal rat urogential
sinus(20, 21, 22) . Urogential sinus
mesenchymal cell lines (U4F and U4F1) were adapted into chemically
defined medium and characterized for stromal marker
proteins(21, 22) . Initial studies identified in U4F
cell-conditioned medium, a growth inhibitory activity to PC-3 prostatic
carcinoma epithelial cells, NBT-II bladder epithelial cells, and
Mv-Lu-1 mink lung epithelial cells in
vitro(20, 21) . This activity did not cross-react
with a battery of neutralizing antibodies to cytokines (including
transforming growth factor- We report here the identification and
purification of a 20-21-kDa protein secreted from U4F mesenchymal
cells in chemically defined conditioned medium, having biological
properties identical to the previously characterized UGIF activity.
Based on the cell type of origin and the biological activity in
vitro, this protein may be relevant to growth and differentiation
mechanisms of stromal-epithelial interactions and may serve as a useful
marker for the study of mesenchymal cell ontogeny.
The
PC-3 cell line (ATCC CRL 1435, prostatic carcinoma epithelial) and
Mv-Lu-1 cell line (ATCC CCL 64, mink lung epithelial) were received
from American Type Culture Collection (Rockville, MD). PC-3 cells were
cultured in 93% DMEM/Ham's F-12 medium (1:1), supplemented with
7% fetal bovine serum, and Mv-Lu-1 cells in 90% DMEM, supplemented with
10% fetal bovine serum, with each containing penicillin (25 units/ml)
and streptomycin (25 µg/ml). Medium was replaced every 2-3
days. Cultures were passaged at confluence by brief exposure to
trypsin-EDTA (0.25% trypsin, 0.025% EDTA in calcium, magnesium-free
Hanks' salt solution). Cell viability was established by trypan
blue dye exclusion and cells counted with an improved Neubauer type
hemocytometer. All cell lines were routinely tested for mycoplasma
contamination (MycoTect Kit, Life Technologies, Inc.).
For direct
cell counting, PC-3 cells were seeded at 4.0
For additional analytical studies and for scaled-up preparative
purification of ps20, samples derived from serum-free, chemically
defined conditioned medium M
Ammonium sulfate
precipitation of proteins from conditioned medium was used as a first
step to concentrate samples. Approximately 90% of growth inhibitory
activity precipitated within the 20-40% saturation range of
ammonium sulfate (data not shown), which was then used for subsequent
procedures. For analysis of UGIF bioactive proteins, all chromatography
buffers were formulated to be both volatile (no salt residues upon
vacuum drying) and bacteriostatic (sterile, nonsupportive of bacterial
growth) so that aliquots from gel filtration or HPLC columns could be
vacuum-dried in sterile vials and used directly for biological assay
(addition to target epithelial cells in culture) without additional
steps of dialysis and sterilization of sample. This strategy allowed
for the biological assay of hundreds of fractions per day. For initial
analysis of UGIF biological activity, precipitated proteins from
conditioned Bfs medium (conditioned by U4F cells for 48 h) and
unconditioned control Bfs medium were resolubilized in ammonium
carbonate buffer. The initial chromatography step to analyze biological
activity utilized ion exchange chromatography. Samples were dialyzed
against 20 mM ammonium carbonate (pH 8.85), applied to and
eluted from a DE-52 anion-exchange column according to the methods
under ``Experimental Procedures.'' Fig. 1shows the
biological activity and A
Figure 1:
Ion exchange
chromatography. Proteins from U4F cell 48-h conditioned medium
(serum-containing media Bfs) or volume-matched fresh Bfs (as control)
were precipitated with (NH
Figure 2:
Analytical gel filtration chromatography.
Proteins from U4F mesenchymal cell conditioned medium (Bfs medium) were
precipitated with (NH
Figure 3:
Analytical reverse phase HPLC. Upper
panel, proteins were chromatographed with a C-18 reverse phase
column as described in the text and under ``Experimental
Procedures,'' and aliquots were vacuum-dried and assayed for
inhibition of [
Fractions from ion exchange
representing the major peak of eluted biological activity (delineated
by the bar underlining fractions in Fig. 1) were
collected and pooled. The pooled sample was dialyzed against 1 M acetic acid (pH 2.5, 4 liters) overnight at 4 °C. Dialyzed
samples were quick frozen, lyophilized, and used either directly or
stored at -20 °C. UGIF activity prepared in this manner was
further analyzed by gel filtration chromatography for assignment of
size using a variety of buffer conditions including 1 M acetic
acid, ammonium carbonate, and ammonium acetate. Of these conditions,
gel filtration chromatography in 1 M acetic acid reduced the
interaction with column matrix optimally and allowed for a reproducible
recovery of an activity peak as shown in Fig. 2. Biological
activity was detected as a single major peak, eluting consistently in
the calculated 18-20-kDa size range (position of molecular mass
markers is shown across the top of graph, Fig. 2).
SDS-PAGE analysis of eluted fractions (Fig. 2, lower
panel) showed the elution pattern of a 20-21-kDa protein to
be directly correlated with the elution peak of biological activity (fraction 56, arrow). To further establish the
correlation of this protein species with peak activity, additional
samples were pooled, chromatographed through C18 reverse phase HPLC
columns, and eluted with a linear gradient of acetonitrile as shown in Fig. 3. In direct agreement with gel filtration, the major peak
of UGIF activity from HPLC was associated with a protein of
approximately 20-21 kDa as analyzed by SDS-PAGE (Fig. 3, lower panel, arrow, peak fraction 93). To
facilitate the scale up of preparations to allow for the purification
of the 20-21-kDa species, a chemically defined (serum-free)
growth medium (medium M
Figure 4:
Identification of the 20-21-kDa
protein in serum-free conditioned medium. Upper panel,
preparative volumes (>600 ml) of serum-free, chemically defined
conditioned medium M
Figure 5:
Preparative gel filtration chromatography. Upper panel, the eluted peak of biological activity from ion
exchange chromatography using serum-free conditioned medium (48 h,
medium M
Figure 6:
Preparative HPLC chromatography. Upper
panel, fractions from gel filtration chromatography as shown in Fig. 5were pooled (fractions 52-55) and used for reverse
phase HPLC as described in the text and under ``Experimental
Procedures.'' Shown is the elution profile of biological activity
with PC-3 cells and acetonitrile elution gradient. Biological activity
eluted with a consistently observed front peak (fraction 22)
and highly variable secondary peak region (fractions
25-34). Lower panel shows the SDS-PAGE analysis and
silver staining of proteins in entire eluted fractions (bar
region). The elution pattern of the 20-21-kDa protein
correlated directly with the initial peak (fractions
20-23) and was purified to near homogeneity. Molecular size
markers (shown in lane 1) are as indicated in Fig. 2.
The fourth purification step utilized
reverse phase HPLC owing to the utility of HPLC in separating proteins
of similar size based on hydrophobic properties. Pooled fractions from
gel filtration chromatography (front consistent peak, Fig. 5, fractions 52-55) were vacuum-dried, resolubilized in 50%
formic acid, and analyzed with reverse phase HPLC as described under
``Experimental Procedures.'' The column was eluted with a
shallow gradient of acetonitrile to produce optimal separation of major
peak versus inconsistent minor peak proteins. Fig. 6shows the biological activity elution profiles and the
corresponding SDS-PAGE analysis associated with peak activity.
Biological activity eluted as a well defined and consistently observed
front peak (fraction 20-23, peak = fraction 22), which was
directly correlated with the elution pattern of the 20-21-kDa
protein species (Fig. 6, lower panel), purified to near
homogeneity as determined by SDS-PAGE analysis and silver staining. The
purification procedure as described yielded approximately 600-650
ng of the 20-21-kDa protein from 600 ml of conditioned medium.
Subsequent analyses showed a consistent purification to near
homogeneity and indicated the 20-21-kDa protein was monomeric in
structure as analyzed in either reducing or nonreducing SDS-PAGE
conditions. Similar to observations from the previous gel filtration
step, a second broad peak or set of peaks (fractions 25-34) of
less well defined and highly variable elution patterns and activity
levels, eluted at higher acetonitrile concentrations. These
inconsistent peaks likely represent variable degrees of 20-21-kDa
protein breakdown products, as they were derived from previous step
20-21-kDa protein peak fractions and were observed only in
scaled-up preparations in later stages of purification. For
microsequence analysis, the 20-21-kDa protein (approximately 700
ng to 1 µg) pooled from multiple purification preparations was
electrophoresed through SDS-PAGE gels, blotted to PVDF membranes, and
analyzed for amino-terminal sequence as described under
``Experimental Procedures.'' A single sequence was detected
with unambiguous assignments made for positions 1-14 and
19-28 as follows:
NH
Figure 7:
Renaturation of ps20 activity from
SDS-PAGE. The ps20 protein isolated by HPLC was electrophoresed through
SDS-PAGE, and the ps20 protein band was excised and renatured by the
guanidine HCl method or the gel crushed and proteins extracted with 1 M acetic acid as described in the text and under
``Experimental Procedures.'' Shown in each panel is the
inhibition of [
To
determine biologically active concentrations, purified ps20 was used
for dose-response assays with target PC-3 cells, and exhibited linear
and saturable dose-response curve as shown in Fig. 8.
Half-maximal activity was observed at 55 ng/ml (2.62 nM) under
these conditions. Maximal (saturable) inhibition was observed at
6.3-8 nM. To correlate inhibition of
[
Figure 8:
Dose-response of purified ps20 activity.
The ps20 protein was purified as described in Fig. 1Fig. 2Fig. 3Fig. 4Fig. 5Fig. 6,
and final protein yield was determined as described under
``Experimental Procedures.'' Increasing concentrations of
purified ps20 were added to PC-3 cells, and cells were assayed for
[
Figure 9:
Inhibition of PC-3 cell proliferation and
stimulation of protein synthesis by purified ps20. Panel A,
PC-3 cells were incubated with either 7.2 nM purified ps20
(maximal active concentration) or vehicle control for 5 days and
counted each 24 h as described under ``Experimental
Procedures.'' Purified ps20 produced a 52-60% decrease in
PC-3 cell proliferation, increased population doubling time from 19.8
to 25.8 h (days 1-3), and achieved confluence at a lower cell
density. Panel B, PC-3 cells were incubated with 7.2 nM purified ps20 or vehicle control for 3 days and pulsed with
[
Data presented in this study shows the characterization and
purification of a novel 20-21-kDa protein (termed ps20) derived
from fetal urogenital sinus mesenchymal cells and exhibiting growth
inhibitory and protein synthesis stimulatory activities in vitro in a dose-dependent and saturable manner, suggesting responses are
mediated through saturable pathways. Amino-terminal sequence
information revealed a unique sequence with no matches to proteins in
data bases, indicating ps20 represents a novel protein species.
Additional studies beyond the scope of the present report are required
to generate probes and determine the specific pattern of ps20
expression in mesenchymal cell ontogeny, the molecular mechanisms of
action in stromal-epithelial interactions, and the extent of target
cell-type specificity. Results here report the activity of ps20 with
PC-3 (human prostatic carcinoma) and Mv-Lu-1 (mink lung epithelial)
assay target cells. Our previous studies with crude preparations showed
identical activities with NBT-II (rat bladder epithelial cells) and
Y-79 cells (human retinoblastoma cells), in addition to PC-3 and
Mv-Lu-1 cells(20, 21) . The full range of specific
cell types responsive to ps20 activities is not yet known and will
require greater yields of native ps20 or recombinant protein to address
fully. The in vitro responses elicited by the ps20 protein
are consistent with a potential function in stromal-epithelial
interactions involving tissue growth and differentiation control in the
prostate gland. Extracellular matrices produced by stromal cells,
direct cell-cell contact, and secretion of paracrine-acting effectors
have each been shown to induce and/or facilitate tissue-specific gene
expression and epithelial cell phenotype in a wide variety of tissues.
Accordingly, the combined influences of paracrine factors, matrix
molecules, and direct cell-cell communication each are likely to
contribute to the overall mechanisms of stromal induction of epithelial
phenotype. The actions of ps20 may be a component of any such
mechanism. Although UGS mesenchyme initially stimulates UGS epithelial
cell proliferation during ductal morphogenesis, an epithelial
differentiation to mature glandular acini, typified by epithelial cell
quiescence (low cell mitosis-turnover) and increased secretory activity
(columnar secretory cells) is subsequently induced by the mesenchyme
stroma(6, 7) . Hence, a growth-inhibitory paracrine
effector protein which also stimulates protein synthesis concurrent
with inhibiting epithelial cell proliferation, as is suggested with
ps20, may play a role in this secondary induction to differentiation. The possible role of ps20 in vivo and the mechanisms of
ps20 action are fully unknown. No possible action or role in vivo can be assigned or ruled out based on studies to date. To define
ps20 further, the cloning of cDNA encoding ps20 protein is required to
determine primary structure and assess likely biological actions. In
addition, the characterization of ps20 expression patterns and
regulatory pathways are required to define molecular regulatory
mechanisms and assess possible functions in stromal cell biology. The
ps20 protein may or may not represent a paracrine effector growth
regulatory protein involved in receptor-mediated pathways.
Alternatively, the ps20 protein may function as a matrix component laid
down by mesenchyme or as an external cell membrane protein involved in
cell-cell or cell-matrix adhesion, thereby affecting
proliferation/differentiation. However, it should be cautioned that the
present report addresses biological properties in vitro only,
and no data yet exist to indicate a similar activity or properties of
ps20 in the fetal urogenital sinus or normal prostate gland in
vivo. Progress to understand both prostate organogenesis and
prostate disease progression requires the elucidation of genes and
proteins involved in stromal cell biology and the mechanisms of
stromal-epithelial interactions. The ps20 protein may be of
significance to the initiation and progression of prostatic diseases,
typified by alterations in epithelial cell proliferation,
differentiation, and patterns of stromal cell histology. The
observation that ps20 growth inhibits the PC-3 carcinoma cell line
derived from a human prostatic adenocarcinoma points to the possibility
that ps20 actions may affect prostatic carcinoma progression.
Understanding the expression patterns of ps20 in prostatic disease
(benign prostatic hyperplasia and prostatic carcinoma) and defining
molecular mechanisms of action will be of importance in assessing any
potential role of this protein in prostatic disorders. Of additional
interest is the possibility that ps20 may represent a new marker for
stromal cell ontogeny and functional differentiation. Our preliminary
studies to date ( This report serves as the
initial characterization of the ps20 protein and associated biological
properties. The full extent of ps20 activity and relevance to prostate
gland biology awaits more extensive characterization of cDNA encoding
this protein and determination of specific biological activity in
vivo.
Volume 270,
Number 37,
Issue of September 15, pp. 22058-22065, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
)to mature prostate
gland(6, 7) . UGS epithelial cells progress through a
specific morphogenesis pattern (prostatic glandular acini) only when
recombined with UGS mesenchymal
cells(8, 9, 10) . Similar studies addressing
the development of differentiated epithelium in skin(11) ,
gut(12) , and lung (13) have shown a likewise potent
inductive nature of stromal cells in directing tissue-specific
epithelial growth and differentiation patterns.
s, interferons, and interleukins) and
was provisionally termed urogenital sinus-derived growth inhibitory
factor (UGIF) pending identification of biologically active
components(21) . Crude UGIF activity acted to inhibit
epithelial cell proliferation in a linear and saturable manner,
stimulate synthesis of secretory proteins, and alter epithelial
phenotypic morphology in vitro, and it was nontoxic/reversible
in mechanisms of action(20, 21) . Gel filtration
chromatography analyses of crude UGIF activity showed a consistent
elution pattern (peak of activity) assigned to the 18-22-kDa
calculated size range(21) . Based on additional
physicochemical, biological, and immunological properties, UGIF
activity could not be ascribed to a previously identified protein or
factor(20, 21) . The present study was conducted to
characterize, identify, and purify the protein(s) primarily responsible
for the previously described UGIF activity and characterize biological
activity in vitro.
Materials
The following materials were
purchased: DMEM, Ham's F-12 medium, Hanks' solution,
penicillin, and streptomycin from Life Technologies, Inc.; Nu-Serum and
epidermal growth factor from Collaborative Research (Lexington, MA);
fetal bovine serum from HyClone (Logan UT); MCDB-110 basal medium,
insulin, trypsin type II, crystal violet dye, trypan blue dye,
testosterone, and sodium lauryl sulfate from Sigma; methanol,
trichloroacetic acid, EDTA, ammonium carbonate, and ammonium sulfate
from Fisher; glacial acetic acid from Chempure (Curtin Matheson,
Houston, TX); [methyl-
H]thymidine and
[S]methionine from ICN Radiochemicals (Irvine,
CA); aqueous counting scintillant (BCS, no. NBCS104) from Amersham
Corp.; Bio-Gel P-100 (100-200 mesh) and P-30 (100-200 mesh)
chromatography gels, glycine, and acrylamide from Bio-Rad;
diethylaminoethyl cellulose (DE-52) ion exchange resin and glass fiber
filters (934-AH, 2.4 cm) from Whatman; 96 (no. 76-003-05) and 24 (no.
76-033-05) well culture plates from Flow Laboratories (McLean, VA); 25
cm
(no. 25100) culture flasks from Corning (Corning, NY);
96-well plates (no. 3072), 24-well plates (no. 3847), and 75-cm culture flasks (no. 3024) from Falcon (Becton Dickinson) (Oxnard,
CA); dialysis tubing no. 3 (M
cutoff =
3,500) from Spectrapore (Los Angeles, CA); polyvinylidene difluoride
(PVDF) Immobilon-P membranes from Millipore (Bedford, MA); and
Pro-Blott membranes from Promega (Madison, WI).Cell Culture
The U4F urogenital sinus mesenchymal
cell line, derived originally from UGS organ explants(20) , was
cultured in medium Bfs (90% DMEM, 5% fetal bovine serum, 5% Nu-Serum, 5
µg/ml insulin, 0.5 µg/ml testosterone, 100 units/ml penicillin,
and 100 µg/ml streptomycin), which was replaced every 48 h as
previously reported(21) . Upon confluence, U4F cultures
typically formed multicellular spheroids (multicellular, spherical
domes) after 20-30 days in culture, with UGIF activity first
detectable in conditioned medium from spheroid-containing
cultures(21) . To generate sufficient quantities of conditioned
medium for analysis of UGIF activity, 75-cm culture flasks
were seeded with U4F cells yielding 100-200 spheroids per flask
by 30 days in culture. U4F spheroid cultures were maintained for
approximately 3-6 months in this manner with conditioned media
collected every 48 h, pooled, quick frozen, and used as starting
material for analysis of UGIF activity. For preparative purification of
UGIF protein(s), spheroid cultures were switched to serum-free,
chemically defined medium M (MCDB-110 basal medium plus 5
µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml selenium, 0.1
µg/ml epidermal growth factor, and 0.5 µg/ml testosterone),
formulated empirically to support U4F proliferation and secretion of
UGIF activity as reported elsewhere(21) . Conditioned medium
was collected every 48 h, clarified by centrifugation, and either used
immediately, or stored at -20 °C until subsequent use.Cell Proliferation Assays
Cell proliferation was
determined by direct cell counting of viable cells and with a
[H]thymidine incorporation assay following minor
modifications of procedures we reported
previously(20, 21, 22, 23) . For the
[H]thymidine incorporation assay, PC-3 or Mv-Lu-1
target epithelial cells were seeded at 8.1 
10 cells/148 µl/well in 96-well plates and allowed to attach for
24 h. Wells received a 52-µl aliquot of sample to be tested
(previously vacuum dried and resolubilized in growth medium in sterile
conditions) (200 µl/well final total volume) and were allowed to
incubate for an additional 24 h. Cultures were pulsed with
[H]thymidine (2 µCi/ml) during the final 3 h.
of incubation. The assay was terminated by fixing cell monolayers in situ with methanol:acetic acid (3:1) (200 µl/well) for
10 min (22 °C), washing with 100% methanol (5 min, 22 °C)
followed by a wash with 5% trichloroacetic acid (5 min, 22 °C), and
a three times sequential wash with methanol (22 °C, 200
µl/well). Plates were then allowed to air dry for 5-10 min
under a heat lamp. Plates could either be stored indefinitely at 22
°C or processed immediately. For processing, monolayers were
hydrolyzed by 1 N NaOH (200 µl/well, 5 min, 22 °C),
and aliquots (180 µl) were added to scintillation vials containing
180 µl of 1 N HCl to neutralize pH. Radioactivity was
determined by scintillation counting. Results are presented as the
reciprocal of incorporated counts/min (1/cpm) where indicated in the
figure legends, to reflect inhibition of
[H]thymidine incorporation as a peak of activity.
Results from all comparative assays are presented as the mean of n 
3 tests ± S.E. (error bars), and results were analyzed
for significance using Student's t test. 10 cells/148 µl/well in 96-well plates and allowed to attach for
24 h. Wells received 52-µl aliquots of test sample and were allowed
to incubate for 5 days. Cultures received fresh medium plus/minus test
sample every 48 h. At each 24-h interval, cells were released by
exposure to 0.25% trypsin, 0.025% EDTA in calcium, magnesium-free
Hanks' solution (30 µl/well) for 4 min. To each well were
added 70 µl of PC-3 cell growth medium plus 7% v/v fetal bovine
serum, 40 µl of the cell suspension were incubated with trypan blue
(4 min, 22 °C), and total and viable cells were counted as
described previously(20) .Protein Synthesis Assay
Protein synthesis was
assayed by incorporation of [S]methionine
according to procedures published previously(20) . PC-3 cells
were seeded and processed identically to the cell counting procedure.
On day 2-4 of exposure to test samples, wells received
[
S]methionine (10 µCi/ml) for 24 h. The
medium (200 µl) was harvested and added to an equal volume of 20%
trichloroacetic acid for 1 h at 2 °C to precipitate proteins.
Aliquots (350 µl) from each sample were added to glass fiber
filters (934-AH, 2.4 cm, Whatman) fitted to a filter vacuum manifold
(Millipore). Filters were washed sequentially three times with 2 ml of
EtOH (-20 °C) to remove unincorporated label, filters were
added to scintillation vials, and incorporated radioactivity was
determined by scintillation counting. Cell number was counted from each
corresponding set of wells using the direct cell counting method, and
results were expressed as disintegrations/min
[
S]methionine incorporation/10
cells
to standardize results to cell number.Ion Exchange Chromatography
Samples were dialyzed
(Spectrapore no. 3, M cutoff = 3,500) with
20 mM ammonium carbonate buffer (pH 8.85, 4 °C) overnight,
and applied directly to a DE-52 anion exchange column (1.5 9.5
cm, hydrostatic pressure = 30 cm) equilibrated in the same
buffer. The column was washed with 3 bed volumes of the same buffer and
eluted with a linear gradient of ammonium carbonate (20-300
mM, pH 8.89, 150 ml) and 5 ml/fraction collected. Absorbance
at 280 nm (A) and conductivity were determined
for each fraction. Aliquots (25 µl) of each fraction were directly
vacuum-dried in sterile microcentrifuge tubes and resolubilized in
sterile growth medium Bfs (65 µl), and 52-µl aliquots/well were
assayed directly for PC-3 cell growth activity using the
[
H]thymidine incorporation assay.Gel Filtration Chromatography
For analytical
studies to assign a protein species to the peak of biological activity,
50-200 ml of U4F conditioned medium Bfs (serum-containing) were
dialyzed against 1 M acetic acid (pH 2.25) and lyophilized.
Lyophilized proteins were resolubilized in 1 M acetic acid
(1-3 ml) and chromatographed through a Bio-Gel P-200 or P-100 gel
filtration column (2.5 70 cm), equilibrated in 1 M acetic acid, by gravity flow (45 cm), and 3-ml fractions were
collected. Aliquots (100 µl) were vacuum-dried and resolubilized in
65 µl of Bfs growth medium, and 52-µl aliquots were added to
PC-3 cells for [H]thymidine incorporation assay. were processed through the ion
exchange chromatography step as described. Pooled fractions from the
biologically active peak were prepared by dialysis (Spectrapore no. 3
tubing, 3,500 M cutoff) against 4 liters of 1 M acetic acid (pH 2.25) overnight at 4 °C. Dialyzed
samples were frozen and lyophilized and either used immediately or
stored at -20 °C until use. Lyophilized samples were
solubilized in 1 M acetic acid (1 ml) and applied to either a
Bio-Gel P-100 column or P-30 column (1.4 70 cm) equilibrated in
1 M acetic acid (pH 2.25). Proteins were eluted with a
hydrostatic pressure of 55 cm, and 1.4-ml fractions were collected.
Aliquots (100 µl) from each fraction were vacuum-dried in sterile
microcentrifuge tubes and resolubilized in 65 µl of medium Bfs, and
52-µl aliquots were added to PC-3 cultures for
[H]thymidine incorporation assay.Reverse Phase High Performance Liquid
Chromatography
All samples for HPLC were first chromatographed
through either ion exchange chromatography and/or gel filtration
chromatography prior to HPLC analysis. Samples (pooled fractions from
bioactivity peak) were either vacuum dried or lyophilized and
solubilized in 50% formic acid (0.5 ml). Samples were applied (three
consecutive 150-µl applications) to a Waters C-18 reverse phase
column fitted to an HPLC system composed of a Waters 712 WISP automatic
injector, and Beckman 412 controller, 110A pumps, and 165 variable
wavelength detector. Proteins were eluted with a 1.0 ml/min/fraction
flow rate with 0.1% trifluoroacetic acid in a linear acetonitrile
gradient as indicated in the text. Aliquots (100 µl) from each
fraction were vacuum-dried, resolubilized in 65 µl of medium Bfs,
and 52-µl aliquots added to either PC-3 cells or Mv-Lu-1 cultures
for the [H]thymidine incorporation assay.SDS-PAGE Electrophoresis
Samples were
vacuum-dried, resolubilized in Laemmli sample buffer (24) and
analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) following the
procedures of Laemmli(24) . Samples were heated (5-10
min, 95 °C), 60-100-µl sample aliquots loaded per lane,
and electrophoresed through the stacking gel at 40 mAmp and through the
running gel (15% acrylamide) at 70-mAmp continuous current. Proteins
were visualized using silver staining procedures as described
previously(22) . Protein determinations of highly purified
samples below detectable range with Bradford analysis (<1 µg)
were determined by comparative analysis of sample lanes with sets of
protein standards 50-500 ng/lane in SDS-PAGE gels and an
assignment was made based on relative staining density after silver
staining. For renaturation of activity after SDS-PAGE, gels were washed
in double-distilled water, protein standards position was determined by
incubation in 1 M KCl at 4 °C for visualization of opaque
bands, and protein bands of interest were precisely excised in
1-2-mm slices. Excised proteins were renatured following minor
modifications of the guanidine-HCl method (25) or were directly
eluted from gels with 1 M acetic acid overnight (agitation, 22
°C), vacuum-dried, and processed for biological assay as described. Protein Sequencing
Substantially purified ps20
protein samples were electrophoresed through SDS-PAGE gels as
described, except 100 mM thioglycolic acid was added to the
cathode buffer, and gels were preelectrophoresed for 20 min at 50 V to
reduce amino-terminal blockage. The gel was incubated in transfer
buffer (25 mM Tris-base, 192 mM glycine, pH 8.5) for
5 min and assembled with Immobilon-P or Pro-Blott PVDF membranes in a
Bio-Rad Trans-Blot cell apparatus according to the manufacturer's
recommendations. Proteins were electroblotted using 30 V for 18 h at 4
°C. Proteins were visualized by Coomassie staining, the ps20
protein bands were excised, and the sequence was determined using
Applied Biosystems 477A and 773A protein sequencers.
Assignment of UGIF Activity to a Protein
Species
Initial studies were conducted to determine whether a
putative protein species or set of proteins could be assigned to the
peak or peaks of UGIF biological activity in order to assess
feasibility of purification and focus on a particular protein for
preparative purification. Accordingly, gel filtration chromatography
and reverse phase HPLC were used to analyze UGIF activity, and eluted
proteins were analyzed with SDS-PAGE. UGIF activity was harvested from
U4F mesenchymal cell-conditioned medium following conditions reported
previously(20, 21) . UGIF activity in conditioned
medium was first detectable from confluent U4F cultures having small
multicellular spheroids, which formed at approximately 25 days of
culture, and activity continued to increase with the development of
larger U4F cell spheroids as reported previously(21) .
Identical to previous reports, UGIF activity in CM collected from these
long term cultures did not cross-react with neutralizing antibodies to
a variety of cytokines including transforming growth factor-s,
interleukin-6, and interferons(21) . protein elution
profiles. The elution of a major growth inhibitory peak was observed
from the 48-h conditioned medium preparations. In comparison, the
control (fresh medium) sample did not produce a growth inhibitory peak
and exhibited an otherwise similar baseline activity and A
elution pattern. These studies indicated the
UGIF activity peak was produced by U4F cells and was not a constitutive
component of fresh Bfs medium.
)SO and
chromatographed through a DE-52 anion exchange column. Aliquots (100
µl) from each fraction were vacuum-dried and assayed for inhibition
of [H]thymidine incorporation in PC-3 cells as
described under ``Experimental Procedures.'' Activity was
plotted as reciprocal of incorporated counts/min (1/cpm) to illustrate
inhibition of [H]thymidine incorporation as a
peak of activity. With conditioned medium samples, activity eluted as a
single peak and fractions were pooled and processed for analytical gel
filtration chromatography and HPLC as described in Fig. 2Fig. 3. Fresh medium preparations were negative for
biological activity. Bottom panel shows the corresponding A pattern of total
protein.
)SO,
chromatographed through DE-52 as described in Fig. 1, and
analyzed with gel filtration chromatography, and fractions were assayed
for activity with PC-3 cells as described in Fig. 1and under
``Experimental Procedures.'' Upper panel, elution
profile of biological activity from P-100 gel filtration column.
Biological activity eluted as a single peak (maximum activity at fraction 56) associated with the 18-20-kDa size region.
The elution position of molecular size markers are shown across the top of the graph: bovine serum albumin (66 kDa)
chymotrypsinogen A (27.5 kDa) soybean trypsin inhibitor (20.1 kDa), and
lysozyme (14.3 kDa). Lower panel, SDS-PAGE analysis of eluted
fractions. Fractions (underlined by bar, top
panel) from gel filtration were vacuum dried, electrophoresed
through a 15% acrylamide gel, and stained with the silver method as
described. The elution pattern of a 20-21-kDa species (arrow, fraction 56, bottom panel)
correlated directly the position of eluted bioactivity peak. Molecular
size markers are shown in lanes 1 and 10: myoglobin
fragment I (8.16 kDa), lysozyme (14.3 kDa), myoglobin (16.9), soybean
trypsin inhibitor (20.1 kDa), carbonic anhydrase (29 kDa), ovalbumin
(43 kDa), and bovine serum albumin (66
kDa).
H]thymidine incorporation in
Mv-Lu-1 cells as described under ``Experimental Procedures.''
Biological activity eluted as a major peak associated with fraction 93. Lower panel, SDS-PAGE analysis of eluted fractions from HPLC (bar region, upper panel) shows the elution pattern
of the a 20-21-kDa protein species correlating directly to the
peak of biological activity. Molecular size markers are as described in Fig. 2.
) was developed empirically to lower
protein complexity in the starting conditioned medium. The M medium supported U4F spheroid growth and production of UGIF
activity as described under ``Experimental Procedures.'' To
determine whether the 20-21-kDa protein was responsible for UGIF
activity in conditioned M medium, increased volumes (up to
600 ml) of conditioned media were prepared, chromatographed through ion
exchange chromatography, and analyzed with reverse-phase HPLC as shown
in Fig. 4. Biological activity eluted as a major consistent peak
and a minor inconsistent peak under these conditions. SDS-PAGE analysis (Fig. 4, lower panel) showed the elution of a
20-21-kDa protein species in a pattern correlating directly with
the major peak of biological activity (peak fraction 89, arrow), in agreement with previous results in serum-containing
medium. The minor peak of activity was heterogeneous in elution pattern
and a protein species could not be assigned. Subsequent analyses
focused exclusively on the 20-21-kDa protein.
(U4F cells, 48 h) were processed as
described in Fig. 1, chromatographed with a C-18 reverse phase
column, and eluted with acetonitrile as described in the text. Aliquots
were vacuum-dried and assayed with PC-3 cells as in described in Fig. 1Fig. 2Fig. 3. Biological activity eluted as
a major peak associated with fraction 89 and a minor, variable peak
associated with fraction 96. Lower panel, SDS-PAGE analysis of
eluted fractions (bar region, upper panel) shows the
elution pattern of the a 20-21-kDa species (fraction 89, arrow) directly correlated with the peak pattern of biological
activity as shown in the upper panel. Molecular size markers
are shown in lane 1 and are as described in Fig. 2.
Purification of ps20
Subsequent purification
procedures utilized conditioned M (chemically defined)
medium exclusively and were scaled up to specifically purify the
20-21-kDa protein species. For preparative purification using
increased volumes of M conditioned medium (600 ml), the
initial two steps of ammonium sulfate precipitation and ion exchange
chromatography were used (as described in Fig. 1) which produced
identical elution patterns of activity. For the third purification
step, pooled peak lyophilized material from the ion exchange step was
resolubilized in 1 M acetic acid (1 ml) and applied to either
a P-100 or P-30 (Bio-Rad) gel filtration column (1.4 70 cm)
equilibrated in 1 M acetic acid, and each fraction was
processed and assayed for biological activity as described under
``Experimental Procedures.'' As observed previously with HPLC
of conditioned M medium, biological activity eluted as two
separate peak regions when analyzed with either P-100 or P-30 gel
filtration chromatography. Shown in Fig. 5is the elution
profile from a scaled-up preparation. The earlier eluting (front) peak
was consistently observed in the 20-23-kDa size range (molecular
size markers across top of graph), and a second peak of highly
variable activity levels corresponded to a lower 9-14-kDa size
region. Further SDS-PAGE analysis of eluted fractions (Fig. 5, lower panel) showed the front peak of biological activity
consistently (n > 15) corresponded to the elution pattern
of the 20-21-kDa protein (arrow, fraction 53).
A set of lower molecular mass proteins (8-17 kDa) eluted in
patterns correlating to the second, highly variable peak of activity (Fig. 5, fractions 57-63). This variable,
secondary peak region of activity was inconsistent from preparation to
preparation and was observed only in scaled-up purification procedures
from gel filtration and HPLC steps. Proteins or fragments responsible
for any secondary peaks of activity have not yet been fully determined,
and their analysis was beyond the scope and focus of this report. Final
purification and activity analyses focused exclusively on the
20-21-kDa species.
, preparative volumes, 600 ml) were
chromatographed through a P-30 gel filtration column and aliquots
assayed for biological activity with PC-3 cells as described in the
text and under ``Experimental Procedures.'' Activity eluted
in two peaks. An early front peak consistently eluted at fraction
53, in the 18-21-kDa size region (molecular size markers
shown across top, as in Fig. 2). A set of secondary,
variable and inconsistent peaks eluted in fractions 58-63 as
shown. Lower panel, fractions (bar region, upper
panel) were analyzed by SDS-PAGE and silver staining. Note the
elution pattern of the 20-21-kDa protein (fraction 53, arrow) directly correlated with the front activity peak as
shown in the upper panel. The front peak material was pooled
(fractions 52-55) and used for reverse phase HPLC (Fig. 6). Molecular size markers are shown in lane 1 and as indicated in Fig. 2.
-Thr-Trp-Glu-Ala-Met-Leu-Pro-Val-Arg-Leu-Ala-Glu-Lys-Ser-Xaa-Xaa-Xaa-Xaa-Val-Ala-Ala-Thr-Gly-Xaa-Arg-Gln-Pro-His.
Analysis with PDB, SwissProt, PIR, SPUpdate, GenPept, and GPUpdate data
bases indicated no regions of direct match or homology with previously
characterized proteins. The 20-21-kDa protein was hereafter
referred to as ps20 (20-kDa prostate stromal protein).Biological Activity of ps20
To confirm growth
inhibitory activity was endogenous to the ps20 protein, renaturation
experiments were conducted with purified ps20 protein extracted from
SDS-PAGE gels of samples generated from HPLC steps. Samples purified
following the above steps were electrophoresed through an SDS-PAGE gel.
The ps20 protein band was precisely excised (1-mm width gel band) as
well as control gel bands containing either no loaded protein or
protein controls (11-kDa range). The samples were either electroeluted
from the gel slices, dialyzed, and renatured following modifications of
the guanidine-HCl method(25) , or were passively eluted from
crushed gel slices in 1 M acetic acid and dialyzed following
procedures described under ``Experimental Procedures.'' The
ps20 protein renatured with either of these procedures produced
significant (p < 0.01) growth inhibitory activity relative
to controls as shown in Fig. 7. Renaturation of ps20 extracted
with 1 M acetic acid produced a 15.4% inhibition of
[H]thymidine incorporation over control (Fig. 7A). Guanidine-HCl renaturation of ps20 (Fig. 7B) produced up to a 30.3% inhibition relative to
controls. With either procedure in any particular experiment,
renaturation efficiency after SDS-denaturation was maximally
15-20% relative to activity levels of samples prior to SDS-PAGE,
similar to expected efficiencies of renaturation for proteins extracted
and processed from denaturing SDS-PAGE gels(25) .
H]thymidine incorporation (1/cpm)
in PC-3 cells induced by renatured ps20 (column C) relative to
controls (columns A, B, and D) as described
below. Panel A, gel extracted in 1 M acetic acid. A, media control; B, extracted gel (no protein)
control; C, extracted ps20 protein; D, acetic acid
control. Panel B, gel-extracted proteins were renatured with
the guanidine HCl method. A, media control; B,
extracted gel control (no protein); C, extracted ps20 protein; D, extracted gel control (arbitrary 11-kDa protein band).
Values are mean of n = 3 separate determinations
± S.E. Column C is statistically significant (p < 0.01) in each experiment.
H]thymidine incorporation with inhibition of
cell proliferation, PC-3 cells were incubated with 7.2 nM ps20
or vehicle control for 5 days, and cells were counted every 24 h as
shown in Fig. 9A. Under these conditions, PC-3 cell
proliferation was inhibited by 52-60% relative to control at any
particular time point, in general agreement with the maximal percent
inhibition of [H]thymidine incorporation
(approximately 70%) at 7.2 nM shown in Fig. 8. Under
these conditions, the population doubling time of subconfluent PC-3
cells (days 1-3) was increased from an average of 19.8 h in
control cultures to 25.8 h in ps20-treated cultures. In addition,
ps20-treated cultures attained confluence at lower cell densities (days
3-5), due to a change in cell shape to a larger, more spread-out
phenotype as discussed below. The ratios of viable to nonviable cells
were identical between control and ps20 treated cultures indicating the
results were not due to increased cell death or toxicity, in agreement
with our previous reports using crude UGIF activity(21) . The
inhibition of cell proliferation was associated with a stimulation of
protein synthesis. As shown in Fig. 9B, purified ps20
(7.2 nM) stimulated the synthesis of secreted proteins from
PC-3 cells by 210.4% relative to control on a per cell basis. In
addition, the ps20-treated (7.2 nM) PC-3 cells assumed a more
spread-out cell shape with increased pseudopodia and filopodia cell
extensions compared to control cultures (data not shown). The effects
of purified ps20 on growth inhibition, stimulation of protein
synthesis, and alteration in morphology are each consistent with
previous reports on UGIF activity in crude conditioned
medium(20, 21) . Based on these data, the activity
ascribed to ps20 protein likely accounts for the previously observed
UGIF activity in crude conditioned medium.
H]thymidine incorporation as described under
``Experimental Procedures.'' Incorporated
[H]thymidine (cpm) was plotted as a function of
ps20 concentration. One-half maximal activity was determined at 2.62
nM with maximal activity at 6.3-8 nM. Values
are mean of n = 3 separate determinations,
±S.E.
S]methionine (10 µCi/ml) the final 24 h,
cells were counted, and proteins from the medium were assayed for
[
S]methionine incorporation as described under
``Experimental Procedures.'' Presented are
disintegrations/min of incorporated
[
S]methionine/10
cells. Values are
mean of n = 3 separate determinations, ±
S.E.
)indicate that significant ps20 expression
is limited to mesenchymal cells and adult stromal cells expressing
smooth muscle differentiation marker proteins and is not observed in
either fibroblast or epithelial cells. Recent studies indicate that
prostate smooth muscle represents the major androgen-regulated stromal
cell type in the postnatal prostate gland (26, 27) .
Prostate smooth muscle cells evolve from androgen receptor-positive
mesenchymal cells, such as U4F cells, in the immediate proximity to
developing pockets of epithelial islands(28) . Accordingly,
ps20 expression may correlate with ontogeny of mesenchymal
differentiation to smooth muscle and may provide insight to mechanisms
of mesenchymal/smooth muscle actions.
))
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
ABSTRACT
This article has been cited by other articles:
|
|
 |
|
  S. Madar, R. Brosh, Y. Buganim, O. Ezra, I. Goldstein, H. Solomon, I. Kogan, N. Goldfinger, H. Klocker, and V. Rotter Modulated expression of WFDC1 during carcinogenesis and cellular senescence Carcinogenesis, January 1, 2009; 30(1): 20 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Alvarez, J. Reading, D. F. L. King, M. Hayes, P. Easterbrook, F. Farzaneh, S. Ressler, F. Yang, D. Rowley, and A. Vyakarnam WFDC1/ps20 Is a Novel Innate Immunomodulatory Signature Protein of Human Immunodeficiency Virus (HIV)-Permissive CD4+ CD45RO+ Memory T Cells That Promotes Infection by Upregulating CD54 Integrin Expression and Is Elevated in HIV Type 1 Infection J. Virol., January 1, 2008; 82(1): 471 - 486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. McAlhany, S. J. Ressler, M. Larsen, J. A. Tuxhorn, F. Yang, T. D. Dang, and D. R. Rowley Promotion of Angiogenesis by ps20 in the Differential Reactive Stroma Prostate Cancer Xenograft Model Cancer Res., September 15, 2003; 63(18): 5859 - 5865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Gerdes, T. D. Dang, M. Larsen, and D. R. Rowley Transforming Growth Factor-{beta}1 Induces Nuclear to Cytoplasmic Distribution of Androgen Receptor and Inhibits Androgen Response in Prostate Smooth Muscle Cells Endocrinology, August 1, 1998; 139(8): 3569 - 3577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Larsen, S. J. Ressler, B. Lu, M. J. Gerdes, L. McBride, T. D. Dang, and D. R. Rowley Molecular Cloning and Expression of ps20 Growth Inhibitor. A NOVEL WAP-TYPE "FOUR-DISULFIDE CORE" DOMAIN PROTEIN EXPRESSED IN SMOOTH MUSCLE J. Biol. Chem., February 20, 1998; 273(8): 4574 - 4584. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |