The present study showed that a bikunin suppresses the expression of constitutive and PMA-induced uPAR mRNA and protein in a variety of cell types; b an extracellular signal-regulated ki
Trang 1Suppression of urokinase receptor expression by bikunin
is associated with inhibition of upstream targets of extracellular signal-regulated kinase-dependent cascade
Hiroshi Kobayashi1, Mika Suzuki1, Naohiro Kanayama1, Takashi Nishida2, Masaharu Takigawa2and Toshihiko Terao1
1
Department of Obstetrics and Gynecology, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan;2Department
of Biochemistry and Molecular Dentistry, Okayama University Graduate School of Medicine and Dentistry, Okayama, Japan
Our laboratory showed that bikunin, a Kunitz-type protease
inhibitor, suppresses 4b-phorbol 12-myristate 13-acetate
(PMA)- or tumor necrosis factor-alpha (TNFa)-induced
urokinase-type plasminogen activator (uPA) expression in
different cell types In addition to its effects on protease
inhibition, bikunin could be modulating other cellular events
associated with the metastatic cascade To test this
hypo-thesis, we examined whether bikunin was able to suppress
the expression of uPAreceptor (uPAR) mRNAand protein
in a human chondrosarcoma cell line, HCS-2/8, and two
human ovarian cancer cell lines, HOC-I and HRA The
present study showed that (a) bikunin suppresses the
expression of constitutive and PMA-induced uPAR mRNA
and protein in a variety of cell types; (b) an extracellular
signal-regulated kinase (ERK) activation system is necessary
for the PMA-induced increase in uPAR expression, as
PD098059 and U0126, which prevent the activation of
MEK1, reduce the uPAR expression; (c) bikunin markedly
suppresses PMA-induced phosphorylation of ERK1/2 at the concentration that prevents uPAR expression, but does not reduce total ERK1/2 antigen level; (d) bikunin has no ability to inhibit overexpression of uPAR in cells treated with sodium vanadate; and (e) we further studied the inhibition of uPAR expression by stable transfection of HRA cells with bikunin gene, demonstrating that bikunin secretion is nec-essary for inhibition of uPAR expression We conclude that bikunin downregulates constitutive and PMA-stimulated uPAR mRNA and protein possibly through suppression of upstream targets of the ERK-dependent cascade, indepen-dent of whether cells were treated with exogenous bikunin or transfected with bikunin gene
Keywords: bikunin; extracellular signal-regulated kinase; urokinase-type plasminogen activator; uPAreceptor; tumor invasion
Tumor cell invasiveness is a complex, multistep process that
involves cell attachment, the proteolysis of matrix
compo-nents, and the migration of cells through the disrupted matrix
[1] Activation of receptor-bound uPA on the cell surface
appears to play an important role in cancer cell invasion and
metastasis [2] In a number of cancers, the expression of the uPAand uPAR is required for the invasive phenotype [3,4] uPAR binds uPA with high affinity with a Kdof 0.5 nM [5,6] It has been shown previously that increased levels of uPAand uPAR correlate well with higher invasive phenotype [7] Most uPAR protein is concentrated at invasive foci [8]; it accelerates plasmin formation at the cell surface Overexpres-sion of a human uPAR cDNAincreased the ability of tumor cells to penetrate a barrier of reconstituted basement membrane For colon cancer, a high uPAR expression was strongly correlated with the poor prognosis in colon cancer [9] In contrast, exposure to anti-uPAR Ig [10], soluble uPAR [11,12], or stable transfection with antisense uPAR cDNA [13] rescues the invasiveness of tumor cells
The uPAR protein is inducible by epidermal growth factor (EGF), transforming growth factor-beta (TGF-b) [14,15], hepatocyte growth factor (HGF) [16], vascular endothelial growth factor (VEGF) [17], interferon gamma (IFN-c), tumor necrosis factor-alpha (TNF-a) [18], and by the tumor promoter, phorbol ester [5,19] Activation of the protein kinase C pathway by PMAhas been reported to increase uPAR mRNA in certain cell types [20] An increase
in mRNAstability in response to PMAhave been also reported [21]
Bikunin [also known as urinary trypsin inhibitor (UTI)],
a light chain of the interalpha-inhibitor (IaI) family, is a
Correspondence to H Kobayashi, Department of Obstetrics
and Gynecology, Hamamatsu University School of Medicine,
Hamamatsu, Shizuoka, Japan.
Fax: + 81 53 435 2309, Tel.: + 81 53 435 2308,
E-mail: hirokoba@hama-med.ac.jp
Abbreviations: A TF, N-terminal fragment of uPA ; DIP, diisopropyl
fluorophosphate; DMEM, Dulbecco’s minimum Eagle’s medium;
EGF, epidermal growth factor; ELISA, enzyme-linked
immuno-sorbent assay; ERK, extracellular signal-regulated kinase; GraPDH,
glyceroaldehyde-3-phosphate dehydrogenase; HGF, hepatocyte
growth factor; HI-8, a C-terminal domain of bikunin; IaI,
interalpha-inhibitor; IFN-c, interferon-gamma; PMA, 4b-phorbol 12-myristate
13-acetate; TGFb, tumor growth factor-beta; TNFa, tumor necrosis
factor-alpha; uPA, urokinase-type plasminogen activator; uPAR,
uPAreceptor; UTI, urinary trypsin inhibitor; VEGF, vascular
endo-thelial cell growth factor.
(Received 20 December 2001, revised 20 May 2002,
accepted 21 June 2002)
Trang 2Kunitz-type protease inhibitor [22] and an effective
inhibitor of calcium influx in cell transporter system [23]
Bikunin also inhibits TNF-a-mediated translocation and
activation of PKC [24] as well as PMA-dependent
activation of PKC and MAP kinase cascade [25] We
reported that it inhibits tumor invasion and metastasis
possibly through suppression of cell-associated plasmin
activity and expression of uPAmRNAand protein
However, little is known concerning the potential role of
bikunin in the regulation of uPAR mRNA and its protein
In this paper, the positive modulation of uPAR mRNA
and protein by PMAand the negative regulatory effects
by bikunin on uPAR gene expression in human cancer
cells (ovarian cancer cell lines HOC-I and HRAas well as
chondrosarcoma cell line HCS-2/8) are reported We
undertook the present study to determine the role of
exogenously added bikunin on regulation of extracellular
signal-regulated kinase (ERK)-dependent uPAR
expres-sion Further, bikunin transfection experiments were
carried out to determine whether endogenously produced
bikunin is necessary for inhibition of uPAR expression
possibly through suppression of an upstream target(s) of
ERK phosphorylation
M A T E R I A L S A N D M E T H O D S
Materials
4b-Phorbol 12-myristate 13-acetate (PMA), calphostin C
and staurosporin were purchased from Sigma, St Louis,
MO, USA Human urokinase (high molecular mass
two-chain uPA) was a gift from Yoshitomi Pharmaceutical,
Osaka, Japan Two-chain uPAwas inactivated with
diisopropyl fluorophosphates (DIP) to form DIP-uPA,
as described previously [26] The N-terminal fragment of
uPA(ATF) was purified, as described previously [27]
a1-antitrypsin was from Kaketsuken, Kumamoto, Japan
Polyclonal anti-(phospho-ERK1/2) Ig and anti-(c-Jun) Ig
were from Santa Cruz Biotechnology Polyclonal
anti-MEK1/2 Ig was from Transduction Laboratories Total
ERK was detected using an antibody from Zymed (San
Francisco, CA, USA) A uPA-specific antibody, which
recognizes the N-terminus of uPA(#3471), polyclonal
rabbit anti-uPAR Ig (399R) and a specific ELISA for
uPAR (IMUBIND 893) were obtained from American
Diagnostica (Greenwich, CT, USA) Nonimmune
anti-(mouse/rabbit IgG) Ig and anti-anti-(mouse/rabbit IgG) Ig
conjugated with horseradish peroxidase were from Dako
(Copenhagen, Denmark) Purified bikunin and
recombi-nant C-terminal Kunitz domain II, HI-8 (Thr78-Asn147),
were provided by E Morishita (Mochida Pharmaceutical
Co., Gotenba, Shizuoka) Bikunin derivatives
[O-glycoside-linked N-terminal glycopeptide (bikunin-m1; Ala1-Lys21),
N-glycoside-linked C-terminal tandem Kunitz-domains
(bikunin-m2; Lys22-Leu143), bikunin klacking
O-glyco-side (bikunin-c; Ala1–Leu143), asialo bikunin (bikunin-a;
Ala1–Leu143), bikunin lacking N-glycoside (bikunin-n;
Ala1–Leu143)] were prepared as described previously
[28] The MEK inhibitors, PD098059 and U0126, were
purchased from Calbiochem [32P]dATP random prime
Amersham Japan All reagents used were of analytical
grade
Cell culture The human ovarian cancer cell lines HOC-I [29] and HRA [30], as well as the human chondrosarcoma cell line HCS-2/8 [31], have been described previously The HRAcells were provided by Y Kikuch and the HCS-2/8 cells were a gift from M Takigawa, both at Okayama University, Okay-ama, Japan All cell lines stained negative for mycoplasma contamination Cells were maintained in RPMI-1640 medium (HOC-I and HRA) or Dulbecco’s minimum Eagle’s medium (DMEM) (HCS-2/8) supplemented with penicillin (100 UÆmL)1), streptomycin (100 lgÆmL)1) and 10% heat-inactivated fetal bovine serum (Life Technologies, Inc.; Rockville, MD, USA) at 37C in 5%CO2/air atmo-sphere Before stimulation, cells were washed three times with NaCl/Piand incubated overnight in complete medium containing 1% fetal bovine serum The test drugs were added and incubation was continued for different time lapses After culture, medium was aspirated and cells were harvested and washed extensively Immediately before harvest, cell viability was consistently found to be > 90% The ELISAprocedure we propose measures native IaI and bikunin in human plasma (H Kobayashi, M Suzuki,
Y Hirashima & T Terao, unpublished data) EDTA-plasma from healthy individuals revealed a mean level of 250 and
10 lgÆmL)1of IaI and bikunin, respectively Therefore, the bikunin concentration of culture medium containing 1% fetal bovine serum may correspond to about 0.1 lgÆmL)1
Plasmid construction and transfection into the HRA cell line
The a1 microglobulin-bikunin cDNAwas cloned by PCR using the human liver cDNAlibrary (Clontech, Palo Alto,
CA, USA) as a template with primers (P1: 5¢-ACCG AGCCTCGAGGATATACCAAGGCAGAGGAGC-3¢, P2: 5¢-ACTTGCAGAGCGGCCGCTTGTCAGTTGGA GAAGC-3¢) Cloned cDNAwas inserted into the XhoI– NotI site of the pCIneo vector (Clontech) In order to remove the internal sequence of 500 bases from this vector, an inverted PCR was performed with primers (P3: 5¢-GAGAG TCAGCGCTGCTGTGCTACCCCAAGA-3¢, P4: 5¢-ACT TGGATGTTGTCGGGCGGCGTTGGCACA-3¢) The resulting PCR product was digested with Eco47III and self-ligated Finally, using the vector as a template, bikunin cDNAwas cloned by PCR with primers (P1, P2), and inserted into the SmaI site of pCMV (cytomegalovirus promoter)-IRES (internal ribosome entry site)-bsr (blastic-idin S hydrochloride resistant gene) vector [32] The bikunin expression vector pCMV-bikunin-IRES-bsr and the con-trol vector pCMV-luciferase-IRES-bsr [32] encoding LUC (luciferase) were transfected into HRAcells by the standard calcium phosphate precipitation method [33] The cells were selected in the presence of 10 mgÆmL)1blasticidin S hydro-chloride (Funakoshi Co Ltd, Tokyo, Japan) Resistant clones were obtained after four weeks, and bik+clones and luc+ clones were obtained The cells were subsequently maintained in the presence of 10 mgÆmL)1 blasticidin S hydrochloride
The initial bik+mass cultures were subjected to at least two rounds of subcloning in order to obtain stable bik+ clones DNAsequencing verified the correct insertion of the bikcDNA Finally, HRA bik+tumor cell clones with bik
Trang 3overexpression were confirmed by immunocytochemical
staining and Western blot analysis [34] cDNAsynthesized
from luciferase transfected tumor cells (luc+ clones) was
used as a control
Northern blot analysis
Northern blot analysis was performed using standard
methods Ten micrograms of RNAwere separated in
1.2% agarose gels and blotted onto Hybond N+
mem-branes Prehybridization and hybridization were performed
in 50% formamide at 42C with 5 · 106c.p.m.ÆmL uPAR
cDNAprobe, as described previously [35], and filters were
reprobed with the cDNAfor glyceraldehyde-3-phosphate
dehydrogenase to correct for the amount of RNAloaded
onto the filters A1.1 kb XbaI–EcoRI fragment of uPA R
cDNA[36] was radiolabeled with [32P]dATP via random
hexamer primer extension and used as hybridization probe
After each hybridization, the membranes were washed and
exposed on Kodak BioMax MS-1 film at)70 C Filters
were quantitated by scanning densitometry using a Bio-Rad
model 620Video Densitometer with aID ANALYSTsoftware
package for Macintosh
Western blot
uPAR, phosphorylated and total ERK were detected by
immunoblot analysis To detect uPAR protein, conditioned
media were individually harvested and the remaining
monolayers were scraped and lysed in 50 mMHepes, 0.5M
NaCl, 0.05% Tween-20, 1% Triton X-100, 1 mM
phenyl-methanesulfonyl fluoride, 10 lgÆmL)1 E-64, 10 lgÆmL)1
leupeptin to prepare cell lysates For the detection of ERK
proteins, the cells were extracted with 1.0% Nobidet P-40,
50 mM Hepes, 100 mM NaCl, 2 mM EDTA, 1 lgÆmL)1
leupeptin, 0.4 mgÆmL)1 sodium vanadate, 0.4 mgÆmL)1
sodium fluoride, 5 mgÆmL)1dithiothreitol, pH 7.4 Samples
were stored at)20 C and used only once after thawing
Equal amount of cellular protein (50 or 20 lg per lane) were
subjected to 12% SDS/PAGE and transferred to
poly(viny-lidene difluoride) membranes Filters were probed with
primary antibodies and revealed with a biotinylated
anti-(rabbit/mouse IgG) Ig and avidin-peroxidase Peroxidase
was detected by enhanced chemiluminescence
uPAR protein assay: measurement of ligand binding
DIP-uPAwas radioiodinated with carrier-free Na-125I, as
described previously [37] DIP-uPAwas labeled with125I,
resulting in a specific radioactivity of 4900 c.p.m.Æng)1,
without loss of latent protease activity Binding assays were
performed at 4C as described previously [37] In brief,
confluent monolayers of HRAwere grown in 24-well plate
wells using complete medium supplemented with 10% fetal
bovine serum The cells were washed twice each with
medium supplemented with 1% fetal bovine serum The
cells were maintained overnight medium supplemented with
1% fetal bovine serum Monolayers were cooled to 4C
and washed twice with Tyrode’s-Hepes solution containing
1% fetal bovine serum (washing buffers) Prior to
perform-ing bindperform-ing studies, confluent cultures were subjected to a
mild acid wash (50 mMglycine-HCl, pH 3.0, 0.1MNaCl)
to dissociate uPAR-associated ligands The cells were then
incubated at 4C for 2 h in binding buffer (150 mMNaCl,
10 mMHepes, 2 mMCaCl2, 1 mMMgCl2, 1% fetal bovine serum) supplemented with 10 nM [125I]DIP-uPAand washed four times To determine nonspecific binding, 50-fold higher concentrations of unlabeled DIP-uPAwere added to the incubate Unbound [125I]DIP-uPAwas removed and the contents of the wells were removed using
1M NaOH and radioactivity of the cell lysates was measured using a gamma-counter The cells bound [125I]DIP-uPAin a specific and saturable manner at 10 nM [125I]DIP-uPA Each experimental point was performed in
at least triplicate wells The nonspecific binding, determined
as the percent of input counts bound in the presence of 0.5 lM unlabeled uPA, was approximately 9% and was subtracted from all raw data to give the specific bound counts
Invasion assay Invasion assays were performed essentially as described previously [25] The effects of agents that alter the activity of uPA/uPAR expression, including neutralizing monoclonal antibodies against uPAand uPAR, on the invasiveness of HRAcells were determined by measuring the ability of cells treated with these agents to pass through a layer of the extracellular matrix extract Matrigel coating a filter using chemoinvasion chambers
Statistical analysis Data are presented as mean ± SD All statistical analysis was performed usingSTATVIEWfor Macintosh The Mann– Whitney U-test was used for the comparisons between different groups P < 0.05 was considered significant
R E S U L T S
Induction of uPAR mRNA by PMA in HRA cells Unstimulated cells (HRA, HOC-I and HCS-2/8) expressed different levels of 1.1 kb uPAR transcripts (Fig 1A) uPAR mRNAin HRAcells incubated with PMA(100 nM) for 3 h was increased 7.5-fold as compared with the unstimulated cells, which appeared at 1 h, peaked at 3 h and declined at
24 h (Fig 1B) The effect of PMAon uPAR expression was dose-dependent at PMAconcentrations of 10–100 nM, with
a maximum increase seen after treatment of HRAcells with
100 nMof PMA(Fig 1C) Similar PMAeffects on uPAR mRNAwere found in the two other cancer cell lines (data not shown)
Suppression by bikunin of uPAR mRNA accumulation
by PMA
We showed previously that bikunin plays an important role in signaling of PMA[24] and TNFa [25] It is of interest to determine whether bikunin also affects the expression of PMA-induced uPAR mRNA(Fig 2) When concentrations of bikunin (0.1–1 lM) known to inhibit the uPAproduction [25] were added in the presence of
100 nMPMA(data not shown) to HRAcells, there was a dose-dependent inhibition of 50 and 70% of the uPAR mRNAlevel at concentrations of 0.1 and 1 lM,
Trang 4respectively, as determined by scanning densitometry.
After treatment with 1 lM bikunin alone (lane 2), 30%
reduction of uPAR mRNAwas observed Similar effects
of bikunin inhibition were found in the other cell lines
(data not shown)
We examined the capacity of truncated proteins [deglyco-sylated bikunin (bikunin-c) and HI-8] and a related protein (a1-antitrypsin) to suppress PMA-stimulated uPAR mRNA expression We showed that bikunin might inhibit uPAR expression by mechanisms different from direct protease inhibition HI-8 is the C-terminal domain of bikunin, which
is active fragment for protease inhibitor but is not recog-nized by the cell-associated bikunin binding sites We could not see any effects of deglycosylated bikunin, HI-8, and a1-antitrypsin on suppressing uPAR expression Similar effects on bikunin specificity were found in the other cell lines (data not shown)
Suppression by bikunin of unstimulated and PMA-induced uPAR protein expression
We further investigated whether bikunin could inhibit unstimulated and PMA-stimulated [125I]DIP-uPAbinding capacity on the cell surface It has been established that PMAtreatment resulted in an increase in the number of binding sites and a decrease of the affinity of the uPAR [38,39] The quantification of uPAR expression by mea-surement of the amount of uPAbound is thus not ideal Actually, it could be underestimating the level of up-regulation Notwithstanding these limitations, in this study, uPAR expression was evaluated by measuring cell-associ-ated [125I]DIP-uPAbinding capacity The dose-dependent ability of bikunin to inhibit expression of uPAR by cells is clearly demonstrated using the [125I]DIP-uPAbinding assay (Fig 3-A) HRA cells bound [125I]DIP-uPAin a specific and saturable manner with a maximum effect at a concen-tration of 10 nM(data not shown) Equivalent Kdvalue was determined ( 1.6 nM), which is consistent with the known binding affinity of uPAfor uPAR [5] In cells treated with PMA, cell-associated uPA-binding capacity was significantly decreased in the presence of 100 nMbikunin The maximal
Fig 1 Induction of uPAR mRNA by PMA in tumor cells (A) Relative levels of uPAR mRNAin HRA, HOC-I, and HCS-2/8 cells Cells were grown to 90% confluence Total cellular RNAwas extracted and separated on 1.2% agarose/formaldehyde gel and transferred to Hybond N +
membrane Filters were hybridized with 32 P-labeled uPAR cDNAor with 32 P-labeled GAPDH cDNAprobe Top, representative autoradiograms; Bottom, Levels of uPAR mRNAexpression as quantified by densitometric scanning (B) Stimulation of uPAR gene expression in HRAcells HRA cells were grown to 90% confluence and then stimulated with PMA(100 n M ) for the indicated periods of time C, PMAstimulates uPAR gene expression in a dose-dependent manner HRAcells were incubated for 3 h with different doses of PMA Results are the mean ± SD of four different determinations, with unlike superscripts (a–d) are different (P < 0.05).
Fig 2 Bikunin specifically suppresses PMA-stimulated uPAR gene
expression HRAcells were incubated for 3 h with or without 100 n M
PMAin the presence or absence of bikunin, its truncated proteins
[deglycosylated bikunin (bikunin-c) and HI-8] and its related protein
(a1-antitrypsin) Total cellular RNAwas extracted and analyzed
for uPAR mRNAexpression by Northern blot analysis and
com-pared with untreated control cells (Ctr) Top, representative
auto-radiograms; Bottom, Levels of uPAR mRNAexpression as quantified
by densitometric scanning Results are the mean ± SD of three
dif-ferent determinations, with unlike superscripts (a–e) are different
(P < 0.05).
Trang 5suppression of PMA-induced uPAR expression was
ob-tained at 1000 nMbikunin Constitutive uPAR expression
without stimulation by PMAwas also affected by 100–
1000 nM bikunin Contrary to bikunin, HI-8 failed to
suppress PMA-stimulated [125I]DIP-uPAbinding at
con-centrations of HI-8 as high as 5000 nM Similar PMAand
bikunin effects on uPAR protein were found in the two
other cancer cell lines (data not shown)
We examined whether bikunin directly inhibits [125
I]DIP-uPAbinding to the cells The stimulated cells were treated
with [125I]DIP-uPAin the presence of several unlabeled
competitors [uPA, ATF, bikunin, deglycosylated bikunin
(bikunin-c), and HI-8] As shown in Fig 3B, we found that
bikunin and its derivatives did not directly inhibit [125
I]DIP-uPAbinding to the uPAR on the cell surface
In a parallel experiment, we measured the uPAR levels in
cells stimulated with or without PMAusing a specific
ELISA for uPAR (data not shown) The levels of uPAR
protein in unstimulated and PMA-stimulated HRA cells
were 4.5 ± 0.53 and 18.0 ± 2.48 ng per 106cells,
respec-tively, demonstrating that, after stimulation, uPAR protein
levels increased about fourfold The levels of uPAR protein
in PMA-stimulated cells treated with bikunin (100 and
1000 nM) were 12.3 ± 0.89 and 10.0 ± 0.97 ng per 106
cells, respectively Thus, the dose-dependent ability of
bikunin to inhibit expression of uPAR protein by cells
was also demonstrated using ELISA
Results obtained after exposing the cells to bikunin before
and after stimulation by PMAare presented in Fig 4
Preincubation of the cells with bikunin during 2 h before
100 nMPMAstimulation results in a
concentration-depen-dent inhibition of the induction of [125I]DIP-uPAbinding
(which is associated with uPAR protein expression) At
concentrations of 100 and 1000 nM, [125I]DIP-uPAbinding
is inhibited by 35 and 50%, respectively In contrast, no
significant decrease of uPAR expression is observed when
bikunin is added to the medium 2 h after stimulation by
PMA More than 90% of the control value still remains at
the highest bikunin concentration (1000 n ) tested Cell
viability, monitored by LDH leakage in the culture medium and trypan blue dye exclusion test, is not altered under the different exposure conditions (data not shown) These experiments demonstrated that a marked decrease of uPAR expression is observed when bikunin is added to the medium before stimulation by PMA
The effects of bikunin and its derivatives on uPAR protein expression
To further determine which domains of bikunin are sufficient to suppress uPAR levels in cells, we determined
Fig 3 Effects of PMA and bikunin on functional uPAR protein expression in HRA cells measured by the [125I]DIP-uPA binding assay (A) HRA cells pretreated with bikunin (0, 10, 100, 1000, and 5000 n M ) or HI-8 (5000 n M ) for 2 h were incubated with or without 100 n M PMAfor 12 h.
B, Competitive inhibition of solid-phase [125I]DIP-uPAbinding to HRAmonolayer cells by unlabeled competitors The PMA-stimulated cells (100 n M , 12 h) were treated with 10 n M [ 125 I]DIP-uPAin the presence of 1000 n M unlabeled competitors [uPA, amino-terminal fragment of uPA (ATF), bikunin, deglycosylated bikunin (bikunin-c) and HI-8] Levels of uPAR expression as quantified by [ 125 I]DIP-uPAbinding assay The percent fractions bound on the surface of the cells treated with or without 100 n M PMAcorrespond to 4 or 1.4%, respectively, of [ 125
I]DIP-uPAadded Results are the mean ± SD of three different determinations, with unlike superscripts (a–f) are different (P < 0.05).
Fig 4 Effect of exposure of cells to bikunin before and after stimulation
by PMA Levels of uPAR expression as quantified by [125I]DIP-uPA binding assay Values represent means ± SD of three experiments (A) 10 n M bikunin; (B) 100 n M bikunin; and (C) 1000 n M bikunin a–f, means ± SD with unlike superscripts are different (P < 0.05) Results are representative of two separate experiments.
Trang 6whether the downregulation observed with intact bikunin
could be induced by bikunin derivatives Dose-dependent
suppression of uPAR in cells treated with PMA in response
to treatment with intact bikunin or bikunin derivatives is
shown in Table 1 The levels of uPAR protein in cell lysates
were determined using a specific ELISAfor uPAR
Bikunin-a and bikunin-n effectively suppressed uPAR
levels Bikunin-a and bikunin-n were essentially equipotent
to intact bikunin with respect to the inhibition of uPAR
levels as judged by ELISA, with ID50values of 200 nM
Other bikunin derivatives had no discernible effect on
uPAR levels in cells
Suppression by bikunin of ERK1/2 activity in the
unstimulated and PMA-stimulated HRA cells
Recent studies demonstrated that activation of a
PMA-dependent signal pathway involves a relay of
phosphoryla-tion of several proteins making up the MAP kinase pathway
[40] uPAR gene expression is modulated by multiple signal
transduction pathways EGF and PMAcause increased
uPARtranscription [15] An increased level of
phosphory-lated ERK is associated with increased level of cell-surface
uPAR, which results in enhanced invasion [10] To
deter-mine the role of bikunin in the regulation of ERK1/2
phosphorylation, the unstimulated and PMA-stimulated
HRAcells were analyzed for the phosphorylation of ERK1/2
(Fig 5) Immunoblotting of cell extracts with anti-ERK1/2
Ig indicated the presence of immunoreactive proteins
(44 kDa ERK1 and 42 kDa ERK2) The HRAcells
express primarily ERK1 and some ERK2, as has been
previously reported for HRAcells [24] A
nti-(phospho-ERK) Ig showed a strong kinase activity in the
PMA-stimulated cells corresponding in size to ERK1, while
unstimulated cells contained low ERK1 activity We found
that, in the stimulated cells, phosphorylation of ERK1/2 is
modified within 30 min of induction by PMAand then
returned to the uninduced state after 5 h (data not shown)
We further demonstrated that the high level of uPAR
expression may have aided in the use of HRAcell line as a
model system, since levels of HRAcell ERK1/2 activation
were typically high Compared to HRAcells, lower levels of
phosphorylated ERK1/2 were found in HOC-I and HCS-2/8
cells (data not shown)
When HRAcells were preincubated with bikunin for 2 h,
we could detect suppression of phosphorylation of ERK1/2
in a dose-dependent manner The results, based on densi-tometric scanning, show that 1 lMbikunin inhibits constit-utive and PMA-triggered phosphorylation of ERK1/2, by
50 and 70%, respectively However, 5 lMHI-8 failed
to change significantly the expression of phosphorylated
Table 1 Dose-dependent suppression of uPAR expression and ERK phosphorylation stimulated by PMA after intact and truncated bikunin treatment
of HRA cells The relative amount of PMA-induced cell-associated uPAR expression suppressed in response to increasing concentrations of intact and truncated bikunins is shown The amount of uPAR accumulated in the cells treated without or with 100 n M PMAin the absence of competitor corresponds to 4.5 ± 0.53 and 18.0 ± 2.48 ng per 10 6 cells, respectively The amount of ERK phosphorylation in cell lysates treated with 100 n M
PMAin the absence of competitor corresponds to 100% as judged by a densitometric scan The data of antitryptic activity, cell binding and uPA suppression are from [28,] These data are representative of two independent experiments.
Bikunin Bikunin-m1 Bikunin-m2 Bikunin-c Bikunin-a Bikunin-n HI-8 Molecular mass (kDa) 40 10–30 21 25 39 35 8 Antitryptic activity (IC 50 ; lgÆmL) 1.6 >30 1.1 1.4 1.5 1.9 0.9 Cell binding (IC 50 ) nM 12 30 350 15 15 9 >1000 uPAsuppression
Cell lysate (IC 50 ; nM) 20 >1000 >1000 >1000 30 5 >1000 Conditioned medium (IC 50 ; nM) 28 >1000 >1000 >1000 35 40 >1000 uPAR suppression 200 >1000 >1000 >1000 200 200 >1000 ERK suppression 100 >1000 >1000 >1000 100 100 >1000
Fig 5 PMA-stimulated cells demonstrate increased levels of activated ERK1/2All cells were extracted when 90% confluent Cells pretreated with bikunin or HI-8 for 2 h were incubated with PMAfor 30 min and then extracted to assess ERK phosphorylation Cells were solubilized
in lysis buffer supplemented with protease inhibitors Equal amounts
of cellular protein (50 lgÆlane)1) were loaded in each lane, subjected to SDS/PAGE, and electrotransferred to PVDF membrane for detection with antiphospho-ERK1/2 (active) and anti-ERK1/2 Ig (total) Immunoblot analysis was performed to detect phosphorylated and total ERK1/2 in unstimulated and PMA-stimulated cells Results are representative of two separate experiments Top, representative immunoblotting; Bottom, Levels of phosphorylated ERK1 expression
as quantified by densitometric scanning.
Trang 7proteins Bikunin did not reduce total ERK1/2 antigen
level Therefore, these results show that bikunin inhibits
both constitutive and PMA-induced phosphorylation of
ERK1/2 at the concentration (0.1–1 lM) that prevents
uPAR expression Again, similar effects on ERK
phospho-rylation were found in the two other cell lines (data not
shown)
Domain specificity of bikunin on ERK activation
Dose-dependent suppression of ERK phosphorylation in
the stimulated cells in response to pretreatment with intact
bikunin or bikunin derivatives is shown in Fig 6 and
Table 1 Exposure of cells to bikunin, bikunin-a and
bikunin-n resulted in decrease in the amount of ERK
phosphorylation in cell lysates as judged by
immunoblot-ting, indicating that ID50was about 200 nM Other bikunin
derivatives had no significant effect on suppression of ERK
phosphorylation The inhibition curves for bikunin-induced
suppression of uPAR expression and ERK activation are
shifted towards higher bikunin concentrations in relation to
that for bikunin-induced suppression of uPAexpression
ERK activation is associated with uPAR expression
In order to determine whether the ERK activation system is
necessary for the increase in uPAR mRNAexpression, we
cultured these cells in the presence of PD098059 or an
alternative MEK1 inhibitor U0126, which prevents the
activation of MEK1 [41,42] HRAcells were treated for 12
and 3 h with varying concentrations of the inhibitors and
100 nMPMA, respectively PD098059 (Fig 7A, lanes 2–4) and U0126 (Fig 7C, lane 4) reduced the uPAR mRNA expression in a dose-dependent manner Bikunin showed no additive effect on PD098059-mediated suppression of uPAR expression (not shown) The reduced amount of uPAR mRNAreflected a diminished amount of uPAR protein as determined by the [125I]DIP-uPAbinding assay and ELISA
Fig 6 Dose-dependent inhibition by bikunin and its derivatives of ERK phosphorylation in the PMA-stimulated HRA cells Cells pretreated with bikunin or its derivatives for 2 h were incubated with 100 n M
PMA for 30 min and then extracted (20 lgÆlane)1) to assess ERK phosphorylation Experiments were performed twice with similar results.
Fig 7 Levels of uPAR mRNA expression in unstimulated or PMA-stimulated cells treated with bikunin and/or several inhibitors A, Cells were pretreated with PD098059 (20, 40, or 60 l M ), sodium vanadate (0.1 m M ), or bikunin (1000 n M ) or vehicle for 9 h before the addition of 100 n M
PMA After cells were incubated for 3 h, total cellular RNA was extracted and analyzed for uPAR mRNAexpression by Northern blot analysis (B) Unstimulated cells were treated with PD098059 (60 l M ) or vehicle for 12 h C, Cells were pretreated with staurosporin (50 n M ), PD098059 (60 l M ), U0126 (20 l M ), or with a combination of each agent 9 h before the addition of 100 n M PMA Top, representative autoradiograms; Bottom, Levels of uPAR mRNAexpression as quantified by densitometric scanning A, Results are the mean ± SD of three different determi-nations with unlike superscripts (a–e) are different (P < 0.05) (B) and (C), experiments were performed twice with similar results.
Trang 8(data not shown) The expression of uPAR mRNAin
unstimulated cells was slightly inhibited by treatment of the
cells with PD098059 (Fig 7B) or U0126 alone (not shown)
These results suggest that, in unstimulated cells, MEK1
inhibition in itself does not significantly affect uPAR
mRNAlevels, although the PMA-stimulated ERK activity
downregulation by bikunin is suggested to be the major
mechanism through which bikunin works
The overexpression of uPAR mRNAin the PMA
-stimulated cells was also inhibited by treatment of the cells
with PKC inhibitors, calphostin C (data not shown) and
staurosporin, as measured by Northern blotting (Fig 7C)
However, MEK1 inhibitors showed no additive effect on
PKC inhibitor-mediated suppression of uPAR
overexpres-sion We next examined whether stimulation of ERK1
activity was associated with increased uPAR mRNA
expression For this, HRAcells were treated with a protein
tyrosine phosphatase inhibitor (sodium vanadate) [43]
0.1 mM sodium vanadate significantly increased uPAR
mRNA2.3-fold (Fig 7A, lane 5) Even in the presence of
bikunin, however, the level of ERK phosphorylation was
not decreased back to the basal level observed without
sodium vanadate, suggesting that bikunin has no ability to
inhibit overexpression of uPAR mRNAin HRAcells
treated with sodium vanadate We showed that effects of
these inhibitors do not reflect a decrease in cell viability over
the time frame of these experiments (data not shown)
cDNA transfection of bikunin
In order to determine whether bikunin expression in cells
transfected with bikunin gene is necessary for the decrease in
ERK phosphorylation, followed by suppression of uPAR
mRNAexpression, HRAcells were transfected to express
bikunin gene As shown in Fig 8, bikunin was not detected
in extracts (data not shown) and conditioned media of HRA
and luc+ clone, by immunoblot analysis, but was easily
identified in conditioned medium of bik+clone
We examined whether cDNAtransfection of bikunin
does not alter levels of representative signaling proteins in
bik+clone, luc+clone, and HRAcells Total antigenic levels
of MEK1/2, ERK1/2, and c-Jun were equivalent in the
different cell types (data not shown), suggesting that bikunin
transfection did not nonspecifically inhibit cellular
expres-sion of proteins which impact on MAP kinase cascade
Down-regulation of uPAR mRNA level
in bikunin-transfected cells
Figure 9Ashows that the intensity of the uPAR mRNA
band was much lower in bik+clone than in HRAand luc+
clone Scanning autoradiograms of the hybridization
sig-nals with a laser densitometer and normalization with
the glyceraldehyde-3-phosphate signal showed a fourfold
decrease in uPAR mRNAin bik+clone as compared with
HRAand luc+clone
Immunoreactive uPAR protein was detected in these
cells The bik+clone expressed significantly low levels of
immunoreactive uPAR protein with molecular mass of
50 kDa under nonreducing conditions (Fig 9B) Based on
densitometric acanning, the expression of uPAR at the
protein level in the parental cells was reduced by 50–60% in
bik+clone
Expression of uPAR in HRA cells is associated with
an active ERK1/2
AS we showed in this study that exogenous bikunin suppresses PMA-induced uPAR expression via an inhibi-tion of ERK1/2 phosphorylainhibi-tion, studies were undertaken
to determine whether the level of ERK phosphorylation is decreased in bik+ clone in the presence or absence of exogenous PMAstimuli (Fig 10) Anti-phospho-ERK antibody showed a strong kinase activity in the parental and luc+clone, high uPAR expressor, corresponding in size
to ERK1 On the other hand, bik+ clone, low uPAR expressor, contained low ERK1/2 kinase activity It is unlikely that the large difference in ERK activity between the parental and bik+clone is a consequence of different growth rates, as we were unable to show a reproducible decrease in proliferation of the bik+clone (data not shown) The results show that not only exogenously applied bikunin (1 lM) but also bikunin gene transfection significantly inhibits constitutive and PMA-triggered phosphorylation of ERK1/2 by about 70% Similar effects of bikunin on suppression of MAP kinase activation were found in the two other cancer cell lines (data not shown)
The effect of bikunin and HI-8 as well as neutralizing antibodies against uPA and uPAR on HRA cell invasiveness
Our previous experiments showed that treatment with PMA produced a significant stimulation of the invasiveness of HRAcells in a dose-dependent manner, with a maximum stimulation at 100 n PMA[25] Figure 11 shows the effect
Fig 8 Expression of bikunin protein in HRA cells transfected with cDNA coding for human bikunin Detection of bikunin protein by Western blot Equal amounts (10 lLÆlane)1) of conditioned media derived from the same number of cells of parental HRA(lane 1), luc +
clone (lane 2), and bik+clone (lane 3) were applied to SDS/12% polyacrylamide gel electrophoresis followed by Western blot with antibik antibody Expression of the 50–80 kDa bik protein is up-regulated in bik + clones Molecular mass standards are indicated at the left Results are representative of two separate experiments.
Trang 9of adding increasing concentrations of antibodies or bikunin and HI-8 on the invasiveness of the PMA-stimulated cells (left panel) and bik+ clone (right panel) The PMA-stimulated cell invasion was specifically reversed by concurrent treatment with either neutralizing anti-uPAIg or anti-uPAR Ig, as well as with bikunin These data support that HRAcells leading to invasion is induced through the upregulation of the uPA/uPAR system Bikunin, but not HI-8, which could induced change in uPA/uPAR expres-sion, could in turn modify the invasive behavior of HRA cells However, bikunin had no additive effect on antibody-mediated suppression of cell invasiveness The bik+clone invasion was also reversed by concurrent treatment with either neutralizing anti-uPAIg or anti-uPAR Ig, but not with exogenous bikunin
D I S C U S S I O N
Several authors have reported that the serine protease uPA and its receptor uPAR play a key role in the invasive and metastatic capacity of tumor cells [3,4,44] Phorbol ester and a variety of growth factors including EGF, FGF, and VEGF, which upregulate uPAR synthesis, also stimulate ERK activity [15,45–49] Further, ERK may represent an essential step in the pathway by which cytokine and integrins promote cellular motility [50] It has been estab-lished that PMAinduces ERK activity in a number of systems [45], however, the signaling mechanism by which PMAmodulates uPAR expression is not completely understood The effect of PMAon the expression of other genes has been ascribed to activation of the classical pathway (Rasfi c-Raf1 fi ERK signaling cascade) [51]
An alternative pathway consists of the sequential activation
of Rac1, MEKK1, c-Jun N-terminal kinase kinase (JNKK)
Fig 9 Downregulation of uPAR level in bikunin-transfected cells (A) Downregulation of uPAR mRNAlevel in bikunin-transfected cells Ten micrograms of total RNAisolated from HRAcells, bik + cells, and luc + cells, shown in the upper panel, was electrophoresed in a 1.2% agarose/ formaldehyde gel and then transferred to Hybond N + membrane The membrane was then hybridized with a radiolabeled cDNAprobe specific for uPAR mRNA The same blot was stripped and hybridized with radiolabeled GAPDH cDNAto check for equality of loading uPAR mRNAlevels were measured by scanning autoradiograms with a laser densitometer, and relative hybridization signals were calculated by assigning an arbitrary value of 100 to the highest intense signal seen by Northern blot analysis correlated for mRNAloading inequalities Results are the mean ± SD of three different determinations with unlike superscripts (a and b) are different (P < 0.05) (B) Expression of uPAR protein in bikunin transfected cells Cell lysates of control and bikunin transfected cell clones treated with or without PMAwere used to analyze uPAR expression by Western blot Results are representative of two separate experiments.
Fig 10 Bikunin gene transfection inhibits constitutive and
PMA-trig-gered phosphorylation of ERK1/2All cells were extracted when 90%
confluent Cells were stimulated with PMAfor 30 min and then
extracted to assess ERK1/2 Immunoblot analysis to detect
phos-phorylated and total ERK1/2 in unstimulated and PMA-stimulated
cells Results are representative of two separate experiments Top,
representative immunoblotting; Bottom, Levels of phosphorylated
ERK1 expression as quantified by densitometric scanning Results are
representative of two separate experiments.
Trang 10and the c-Jun N-terminal kinase (JNK) subset of MAPK
[52] Recent studies also demonstrated that the
PMA-dependent stimulation of uPAR gene expression requires a
JNK1-dependent and -independent signaling modules [53]
Tumor dormancy is induced by downregulation of uPAR
[54] It has been reported thata 70% reduction in the
uPAR level in human carcinoma HEp3 cells induced a
protracted state of tumor dormancy in vivo Therefore,
treatment of uPAR-rich cells, which maintain high ERK
activity in vivo, with reagents interfering with the uPAR
signal to ERK activation, mimic the in vivo dormancy
induced by downregulation of uPAR With these in mind,
we investigated the regulation by bikunin of ERK activation
as a signaling molecule in PMA-induced uPAR
overexpres-sion in highly invasive human ovarian cancer cell lines and
human chondrosarcoma cell line
We previously reported that bikunin expression is
associated with a less malignant phenotype [24] There is
evidence that bikunin significantly prevented pulmonary
metastasis of mouse Lewis lung carcinoma 3LL cells [55]
Our ongoing study shows that transfection of the highly
invasive and metastatic HRAcell line lacking bikunin
expression with bikunin-encoding constructs causes a
marked decrease in their metastatic ability (Suzuki &
Kobayash, unpublished data) We have clearly
demon-strated in the recent studies [28,56–59] that bikunin can
suppress PMA-stimulated upregulation of uPA mRNAand
protein via a specific receptor for bikunin, which results
in bikunin-mediated suppression of cell invasiveness The
precise mechanism by which expression of bikunin and
decrease in the metastatic ability of the bikunin transfectants
might be linked to the expression of uPAhas been explored
in our laboratory However, nothing is known about the
mechanism by which bikunin would modulate the uPAR
expression
The present study showed that (a) the effect of PMAon
uPARexpression is time- and dose-dependent; (b)
exoge-nously added bikunin is able to suppress the expression of
constitutive and PMA-induced uPAR mRNAand protein;
(c) a marked decrease of uPAR expression is observed when
bikunin is added to the medium before stimulation by
PMA; (d) PMA promotes cellular uPAR expression, at
least, by activating an ERK-dependent signaling pathway:
the ERK activation system is necessary for the increase in
uPAR expression, since PD098059 and U0126 reduce the
uPAR expression; (e) bikunin markedly suppresses
PMA-induced phosphorylation of ERK1/2 at the concentration that prevents uPAR expression, but does not reduce total ERK1/2 antigen level; (f) the inhibition curves for bikunin-induced suppression of uPAR expression and ERK activa-tion are shifted towards higher bikunin concentraactiva-tions in relation to that for bikunin-induced suppression of uPA expression; (g) the inhibition of ERK1/2 by bikunin depends on receptor binding, since HI-8 does not inhibit ERK1/2 phosphorylation: O-glycoside-linked core protein without N-glycoside, that is the active domain for suppres-sion of uPAexpressuppres-sion, is required for bikunin-mediated suppression of uPAR production and ERK phosphoryla-tion; (h) stimulation of ERK1 activity by sodium vanadate
is associated with increased uPAR expression: bikunin has
no ability to inhibit overexpression of uPAR in cells treated with sodium vanadate, indicating that bikunin is able to suppress uPAR mRNAand protein possibly through inhibition of upstream components of ERK activation in MAP kinase cascade: a set of consecutive signaling mole-cules which teleologically alter the program of gene expression; (i) bikunin transfection experiments demon-strated that uPAR expression is associated with an active ERK and the bikunin expression is necessary for inhibition
of uPAR expression possibly through suppression of ERK phosphorylation; (j) not only exogenous bikunin but also bikuningene transfection markedly inhibits PMA-triggered phosphorylation of ERK1/2; and (k) the HRAcells leading
to invasion is mainly through upregulation of uPA/uPAR expression: suppression by neutralizing antibodies to uPA/ uPAR of cell invasion is almost complete, while inhibition
by bikunin alone is partial Bikunin showed no additive effect on neutralizing antibody-mediated suppression of invasion Therefore, the present results clearly show that bikunin (O-glycoside-linked core protein without N-glyco-side) specifically inhibits expression of uPAR mRNAand protein possibly through suppression of an upstream target(s) of the ERK-dependent cascade
Arecent publication demonstrated that binding of uPA
to uPAR activates ERK1/2, which is required for increased cellular motility in breast cancer cells [26] Furthermore, uPA[60] and uPAR [49] expression are increased in response to multiple factors that activate ERK1/2 It is thus possible that uPAmight regulate expression of uPA itself and its own receptor via ERK activation Our previous and present results demonstrated that bikunin specifically inhibits constitutive and inducible gene expression including
Fig 11 Suppression of invasiveness in PMA-stimulated HRA cells and bik+clone by treat-ment with antibodies and bikunin HRAcells and bik+clone (5 · 104 cells) were placed in the top wells of a chemoinvasion chamber apparatus with the neutralizing antibodies against uPA(10 lgÆmL)1) and uPA R (10 lgÆmL)1), bikunin (0.5 l M ) and HI-8 (1 l M ) in the presence of 100 n M PMA Each point represents the mean of measurements made on two independent wells This is a representative experiment selected from two performed.