(Received for publication, May 28, 1996, and in revised form, October 30, 1996)
From the Section on Developmental Genetics, Heritable Disorders Branch, NICHD, National Institutes of Health, Bethesda, Maryland 20892-1830
Cellular migration and extracellular matrix (ECM) invasion are some of the critical steps in embryonic implantation, inflammation, wound healing, and cancer metastasis. Extracellular phospholipases A2 (PLA2s), belonging to either group I (PLA2-I) or group II (PLA2-II), play an essential role in the generation of proinflammatory lipid mediators (e.g. prostaglandins, leukotrienes, etc.). Recent reports indicate that PLA2-I, in addition to its digestive function, has receptor-mediated effects. For instance, PLA2-I, via its receptor, can induce cell proliferation and airway as well as vascular smooth muscle contraction. Here, we report that both porcine pancreatic and Naja naja PLA2s, in a dose-dependent manner, stimulate dramatic invasion of an artificial ECM (MatrigelTM) by NIH 3T3, mouse fibrosarcoma, and sarcoma cells in vitro, but it has no such effect on lymphoma and mastocytoma cells. That this is a receptor-mediated process is strongly suggested by the following findings. (a) While NIH 3T3, mouse fibrosarcoma, and sarcoma cells, which respond to PLA2-I stimulation, express high affinity PLA2-I receptor mRNA and protein, this receptor expression is not detectable in nonresponder lymphoma and mastocytoma cells. (b) There is a close correlation between the Kd values for 125I-PLA2-I binding to its receptor and the ED50 values for PLA2-I-induced ECM-invasion. (c) Catalytically inactivated PLA2-I is as effective as the active enzyme in stimulating invasion. (d) Suppression of PLA2-I receptor expression in responder cells causes inhibition of ECM invasion. (e) Treatment of the same cells with PLA2-I receptor-antibody also produces virtually identical effects. Taken together, our results identify a novel receptor-mediated function of PLA2-I and raises the possibility that extracellular PLA2s play important physiological as well as pathological roles via this receptor. To our knowledge, this is the first report of this novel receptor-mediated effect (i.e. ECM invasion) of PLA2-I on normal as well as malignant cancer cells.
There is compelling evidence to suggest that embryonic implantation (1-3), wound healing, inflammation, and cancer cell metastasis (4-5) require cellular migration and invasion of the ECM.1 However, the molecular mechanism(s) regulating these important cellular events are not yet clearly understood. During the past several years investigations in this area have suggested that cellular migration and invasion are complex processes and may require the coordinated participation of several genes and their products. Current investigations have focused on understanding the receptor-mediated pathways involved in cell motility and invasion (3, 6).
Phospholipases A2 (PLA2s, phosphatidylcholine 2-acylhydrolases; EC 3.1.1.4) catalyze the hydrolysis of the sn-2 ester bond in phospholipids generating free fatty acids, such as arachidonic acid, and lysophospholipids (7-10). Arachidonic acid is the key substrate for the synthesis of potent lipid mediators of inflammation (e.g. prostaglandins, leukotrienes etc.). Low molecular mass (~14 kDa), extracellular PLA2s are classified into two groups, group I (pancreatic) and group II (arthritic) based on their primary structures (11). Initially, group I PLA2s (PLA2-I) were thought to carry out only digestive function. However, the recent discovery of a class of high affinity cell surface PLA2-I receptors, their cDNA cloning and characterization (12-15), and delineation of novel receptor-mediated biological effects have changed this notion. The PLA2 receptor-mediated effects include stimulation of cell proliferation (16, 17), airway and vascular smooth muscle contraction (18, 19), chemokinesis (20), and fertilization (21). The receptor-mediated PLA2-I-induced biological effects, reported so far, appear to be diverse, and it is highly likely that other important cellular events, regulated via this pathway, remain to be discovered.
Here we report that treatment of NIH 3T3, fibrosarcoma, and sarcoma but not lymphoma and mastocytoma cells with both porcine pancreatic and Naja naja PLA2s causes these cells to dramatically invade ECM (MatrigelTM) in vitro in a dose-dependent manner. The Kd values for PLA2-I binding were virtually identical to the ED50 values of invasion by these cells. This effect appears to be mediated via the high affinity PLA2-I receptor as suggested by several observations. We found that the high affinity PLA2-I receptor is expressed on NIH 3T3, mouse fibrosarcoma, and sarcoma cells which respond to PLA2-I stimulation but not on lymphoma and mastocytoma cells which are unresponsive to such stimulation. Transient transfection of NIH 3T3, mouse fibrosarcoma, and sarcoma cells with PLA2-I receptor-antisense S-oligonucleotide (ASPLAR) caused a pronounced inhibition of PLA2-I receptor protein expression, reduced receptor densities, and a dramatic suppression of PLA2-induced ECM invasion. However, transfection of these cells with sense S-oligonucleotide (SPLAR) or with a mixture (1:1) of ASPLAR and SPLAR was totally ineffective. Interestingly, ECM invasion was also blocked when NIH 3T3, fibrosarcoma, and sarcoma cells were pretreated with purified PLA2-I receptor antibody (IgG) prior to PLA2-I stimulation. Most importantly, the results of binding and affinity cross-linking experiments showed that while PLA2-I receptor is expressed on NIH 3T3, fibrosarcoma, and sarcoma cells, it was not detectable on lymphoma or mastocytoma cells. These data suggest that the PLA2-I receptor expression is cell type-specific and raise the possibility that the observed receptor-mediated PLA2-I-induced cellular invasion of ECM may have important physiological as well as pathological roles.
Porcine pancreatic PLA2
(PLA2-I) was purchased from Boehringer Mannheim. Naja
naja PLA2 was obtained from Sigma.
Disuccinimidyl suberate (DSS) was purchased from Pierce. Sodium
[125I]iodide (carrier-free, 3.7 GBq/ml) was obtained from
Amersham Corp. and [-32P]deoxycytidine 5
-triphosphate
(222 TBq/mmol) and [
-32P]adenosine 5
-triphosphate
(222 TBq/mmol) were purchased from DuPont NEN. BioCoat
MatrigelTM invasion chambers were from Collaborative
Biomedical.
NIH 3T3, mouse fibrosarcoma (HSDM1C1), sarcoma (sarcoma 180), lymphoma (WEHI22.1), and mastocytoma (P815) cell lines were obtained from ATCC, Rockville, MD. These cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (Life Technologies, Inc.), 100 units/ml penicillin, 100 µg/ml streptomycin, and 2 mM glutamine in a humidified atmosphere of 5% CO2 and 95% air at 37 °C.
ECM Invasion AssayThe ECM invasion assay was performed as described previously (22, 23). Invasion assays were conducted using a commercially available 24-well plate (Collaborative Biomedical, Bedford, MA) which consists of an upper and a lower chamber. The two chambers are divided by a porous filter, the upper surface of which is precoated with a layer of MatrigelTM, an artificial basement membrane. Briefly, the confluent monolayers (NIH 3T3, mouse fibrosarcoma) were harvested with 0.05% trypsin and 0.02% EDTA and centrifuged at 800 × g for 10 min. They were washed with trypsin-neutralizing solution followed by DMEM, 0.1% BSA and resuspended in the same medium. The cells grown in suspension (mouse sarcoma, lymphoma, and mastocytoma) were centrifuged and resuspended at the same conditions as mentioned above. The lower compartment of the invasion chamber was filled with fibroblast-conditioned medium which served as a chemoattractant. Fibroblast conditioned medium was prepared by culturing proliferating NIH 3T3 fibroblasts for 24 h in DMEM followed by filtration of the medium using 0.22-µm cellulose acetate membrane (Corning Glassworks, Corning, NY). The invasion assays were initiated by inoculating the upper chamber with cells (1.6 × 105/well) which were either untreated or treated with varying concentrations of catalytically active or inactivated porcine pancreatic or Naja naja PLA2s (0-250 nM). Inactivation of PLA2s was achieved by treating the purified enzymes with 1 mM 4-bromophenacyl bromide (BPB), an active site inhibitor of PLA2, as described previously (24). After treatment, the cells were incubated in a humidified incubator with 5% CO2 and 95% air at 37 °C for 24 h. We also treated the cells with 3 µM indomethacin, a cyclooxygenase inhibitor in order to rule out the involvement of eicosanoids in inducing invasion. Similarly, EDTA (5 mM) treatment of the cells was also used to determine that metal ion-dependent enzymes (e.g.. metalloproteinases) did not contribute to the observed stimulation of invasion by PLA2-I. The cells which invaded the MatrigelTM and migrated to the lower surface of the filter were fixed in 70% methanol and stained with Giemsa for 3 min. The upper surface of the filters were scraped with moist cotton swab 3-4 times to remove all nonmigrated cells and the MatrigelTM. The chambers were washed three times with water; the migrated cells were counted under an inverted microscope, and photomicrographs (120 ×) were taken by using a Zeiss photomicroscope (Axiovert 405 M). As the commercially available Naja naja PLA2 did not stimulate the migratory effect, we checked the purities of both porcine pancreatic and Naja naja PLA2s by SDS-PAGE and found that Naja naja PLA2 had a 49-kDa protein band in addition to the PLA2 band (14 kDa) in the Coomassie Blue-stained gel. This 14-kDa protein was purified from 49-kDa contaminant by Sephadex G-25 chromatography and used for ECM invasion, binding, and affinity cross-linking experiments.
Radio Receptor AssayPLA2-I (25 µg) was radioiodinated using sodium [125I]iodide (2 mCi, carrier-free) and chloramine T as described previously (25). The 125I-PLA2-I was purified by Sephadex G-25 (coarse) followed by Sephadex G-50 (superfine) column chromatography. Fractions containing 125I-PLA2-I were measured by using a gamma counter (ICN Biomedicals, Costa Mesa, CA, model 10/600 plus) with a counting efficiency of approximately 80%. The protein concentration of each fraction was determined spectrophotometrically (280 nM). The specific activity of purified carrier-free mono-iodinated PLA2-I was 95 µCi/µg. For the radio receptor assays, either the nontransfected (NIH 3T3, fibrosarcoma, sarcoma, lymphoma and mastocytoma) or the ASPLAR-transfected (NIH 3T3, fibrosarcoma, and sarcoma) confluent cells grown in 12-well plates were washed once with PBS, pH 7.4, and then incubated with various concentrations of 125I-PLA2-I in 1 ml of Hank's balanced salt solution (HBSS), pH 7.6, containing 0.2% BSA in the absence or presence of either excess porcine pancreatic or Naja naja PLA2 at 4 °C for 2 h. The reactions were stopped by rapid removal of medium containing unbound radiolabeled PLA2-I, and the cells were washed three times with PBS, pH 7.4, and solubilized in 1 N NaOH followed by addition of equal volume of 1 N HCl. The radioactivity was measured by gamma counter. The specific binding was calculated by subtracting the nonspecific binding from the total binding. The binding data were analyzed by Scatchard plot using LIGAND computer program (26).
Affinity Cross-linking ExperimentsConfluent nontransfected
cells grown in six-well plates were washed with PBS, pH 7.4, and
incubated with 125I-PLA2-I (2.9 nM)
in 2.0 ml of HBSS, pH 7.6, containing 0.1% BSA in the absence or
presence of 250 nM each of either porcine pancreatic or
Naja naja PLA2s. After washing twice with PBS,
the cells were further incubated with 0.20 mM DSS in 2.0 ml
of HBSS, pH 7.6, at 4 °C for 20 min. The reaction was terminated by
adding 50 mM Tris-HCl (final concentration) buffer, pH 7.5, and cells were scraped, collected by centrifugation at 10,000 × g for 15 min, and lysed in 60 µl of 1% Triton X-100
solution containing 1 mM PMSF, 20 µg/ml leupeptin, and 2 mM EDTA. The supernatants (40 µl) obtained by
centrifugation at 10,000 × g for 15 min were suspended in SDS sample buffer in the presence of 5% -mercaptoethanol, boiled
for 5 min, and electrophoresed on a 4-20% gradient SDS-polyacrylamide gel (Bio-Rad) according to Laemmli (27). The gels were briefly stained
with Coomassie Blue, dried in a Bio-Rad gel dryer, and autoradiographed
using Kodak X-Omat AR x-ray film.
An antiserum was generated (CYT Immune Sciences, College Park, MD) in the rabbit against a synthetic peptide (Peptide Technologies, Rockville, MD) corresponding to the amino acid sequence Ala185-Phe198 of bovine PLA2-I receptor which is identical to rabbit PLA2-I receptor sequence Ala203-Phe216. The antigenicity of this peptide was predicted by protein-antigen determinant program of PC Gene. The IgG was purified by protein A-affinity chromatography (Akzo Nobel, Rockville, MD). The PLA2-I receptor from either nontransfected or transfected cells were immunoprecipitated by using a kit according to the instructions of the manufacturer (Boehringer Mannheim). Briefly, the cells were washed with ice-cold PBS, lysed with lysis buffer (50 mM Tris-HCl, pH 7.5, containing 150 mM NaCl, 1% Nonidet P-40, 15 µg/ml leupeptin, and 0.5 mM PMSF) on ice, and centrifuged at 12,000 × g for 10 min. The supernatant (1 ml) was incubated with 8 µl of purified PLA2-I receptor antibody (IgG) for 1 h and incubated further with protein A-agarose at 4 °C overnight. Bound complexes were collected by centrifugation, washed, and eluted by boiling in SDS-sample buffer for 5 min. Samples were electrophoresed on 4-20% gradient SDS-polyacrylamide gel and electrotransferred overnight to the nitrocellulose membrane. For Western blot analysis, the membrane containing receptor protein was blocked, washed in PBS-T (PBS containing 0.1% Tween 20), and incubated with purified PLA2 receptor antibody (1:250 dilution). The membrane was washed with PBS-T, incubated with 125I-protein A (ICN Biomedicals, Costa Mesa, CA), washed again, and autoradiographed.
Northern Blot AnalysisTotal RNA was extracted from NIH
3T3, fibrosarcoma, sarcoma, lymphoma, and mastocytoma cells using the
RNAzol method (28) according to the instructions of the manufacturer
(Tel-Test, Friendswood, TX). RNA concentration was measured
spectrophotometrically using absorption at 260 nm. Twenty micrograms of
total RNA were denatured in formaldehyde and separated by
electrophoresis in 1.5% agarose and 5% formaldehyde gels in 40 mM MOPS, 10 mM sodium acetate, and 1 mM EDTA, pH 7.0, transferred to nylon membranes (Schleicher & Schuell), and cross-linked to the membranes by UV irradiation in a UV
Stratalinker-1800 (Stratagene, La Jolla, CA) for 3 min. Equal amount of
RNA was loaded in each lane for electrophoresis, and the membranes
after transfer were found to have RNA bands of identical intensity when
stained with methylene blue. Hybridization was performed using
-32P-labeled bovine PLA2-I receptor cDNA
probe (a generous gift from Dr. Jun Ishizaki, Shionogi Research Labs,
Osaka, Japan). This 1391-base pair cDNA probe encoding a region
that includes the carbohydrate recognition domain-like domain 4 sequence (nucleotides 1458-2849) was obtained by digesting the plasmid
DNA containing a partial length (nucleotides 765-4950) bovine
PLA2-I receptor cDNA by EcoRI and
PstI. The DNA was purified by low melting agarose gel
electrophoresis, buffer-saturated phenol extraction, and ethanol precipitation. The probe was labeled with [
-32P]dCTP
using random primer-plus extension labeling kit (DuPont NEN) according
to the manufacturer's instructions. Autoradiographs were developed by
exposing the blots to Kodak X-Omat AR x-ray film. After dehybridization
by heating the blots at 70 °C for 2 h in 20 mM
Tris-HCl, pH 8.0, containing 0.2 mM EDTA and 1.0% SDS, the
blots were rehybridized with 32P-labeled
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as control which was
end-labeled with [
-32P]ATP.
For glycosidase treatment, confluent NIH 3T3 cells grown in six-well plates were scraped, collected by centrifugation at 10,000 × g for 15 min, washed, and incubated with N-glycosidase F (3 units/100 µl) in 100 µl of 100 mM Tris-HCl buffer, pH 7.0, at 37 °C for 16 h (29). The samples were centrifuged, washed, and incubated further with 125I-PLA2-I (2.9 nM) in 500 µl of HBSS, pH 7.6, containing 0.1% BSA in the absence or presence of unlabeled porcine pancreatic PLA2 at 4 °C for 2 h. These samples were then treated with 0.20 mM DSS, lysed with 60 µl of 1.0% Triton X-100 containing 1 mM PMSF, 20 µg/ml leupeptin, and 2 mM EDTA. The supernatants collected by centrifugation were mixed with SDS-sample buffer, boiled, electrophoresed, and autoradiographed.
Glycoprotein DetectionThe PLA2-I receptor from NIH 3T3 cells was purified by PLA2-I affinity chromatography followed by gel filtration on Sephadex G-100. Glycosylation of the purified PLA2-I receptor was detected by ECL glycoprotein detection system according to the instructions of the manufacturer (Amersham Corp.). Briefly, confluent NIH 3T3 cells were solubilized in 20 mM Tris-HCl buffer, pH 7.4, containing 1% Triton X-100, 10 µg/ml leupeptin, 2 mM EDTA, and 0.4 mM PMSF by stirring at 4 °C for 4 h. The supernatant was collected by centrifugation at 39,410 × g for 90 min and applied to CNBr-activated Sepharose 4B-coupled PLA2-I affinity column. The PLA2-I receptor protein was eluted from the column using 0.1 M glycine HCl, pH 3.0, containing 0.1% Triton X-100, 10 µg/ml leupeptin, 2 mM EDTA, and 0.4 mM PMSF and neutralized immediately with 2 M Tris-HCl, pH 8.0. The fractions containing PLA2-I receptor were analyzed by 125I-PLA2-I binding assay. The samples were concentrated and purified further by Sephadex G-100 column chromatography using 20 mM Tris-HCl buffer, pH 7.4, containing 0.1% Triton X-100, 10 µg/ml leupeptin, 2 mM EDTA, and 0.4 mM PMSF. The homogeneity of the purified receptor was determined by SDS-PAGE followed by silver staining (Bio-Rad), and its activity was measured by PLA2-binding assay. The purified receptor was then electrophoresed and transferred to nitrocellulose membranes which were treated with 10 mM sodium metaperiodate at room temperature in the dark for 20 min. The membranes were washed, incubated with 0.125 mM biotin hydrazide, and blocked with 5% blocking agent. It was washed, incubated further with horseradish peroxidase conjugated-streptavidin (1:1000 dilution), and receptor protein band was detected by using ECL detection kit (Amersham Corp.).
Oligonucleotide TreatmentsMurine PLA2-I
receptor antisense (ASPLAR, 5 TAA CCA TTG CAC CAT CAG GC 3
) and
sense (SPLAR, 5
GCC TGA TGG TGC AAT GGT TA 3
) oligonucleotides with
phosphorothioate linkages were synthesized. (Bio-Synthesis, Lewisville,
TX). These oligonucleotides were purified by column chromatography, and
purity was checked by electrophoresis using denaturing polyacrylamide
gels. NIH 3T3, fibrosarcoma, and sarcoma cells were transfected with
Lipofectin-oligonucleotide complex (30) according to the instructions
of the manufacturer (Life Technologies, Inc.). Briefly, 5 µg of
Lipofectin was mixed with 1.5 µg each of SPLAR, ASPLAR, or a mixture
(1:1) of the two oligonucleotides in 200 µl of serum-free medium
(SFM) and incubated for 15 min at room temperature. The cells were
washed with SFM, and 1 ml of Lipofectin-oligonucleotide complex
containing SFM was added to each well and mixed by gentle agitation.
The cells were incubated further at 37 °C for 12 h. The control
cells received either SFM or Lipofectin alone. Cell viability after
treatment was detected by trypan blue dye exclusion test. After
incubation, SFM containing oligonucleotides was removed, and the cells
were refed with growth medium containing 10% fetal calf serum and
cultured further for an additional 48 h. A separate experiment was
also performed in which varying doses of ASPLAR oligonucleotides
(0-2.0 µg) were used to determine a dose-dependent
response. These cells were used for Western blot analysis, receptor
binding assay, and ECM invasion assay. Additionally, we performed
experiments in which PLA2-I receptor antibody (IgG) was
used to treat the cells in order to ascertain whether
PLA2-induced ECM invasion is mediated via its receptor. A
nonspecific IgG treatment served as a negative control.
The ECM invasion assay was carried out
by pretreating the NIH 3T3, fibrosarcoma, sarcoma, lymphoma, and
mastocytoma cells (1.6 × 105/well) in DMEM containing
0.1% BSA with either Naja naja or porcine pancreatic
PLA2 (catalytically active or inactivated with BPB). A
montage of typical ECM invasion of NIH 3T3 cells treated with varying
concentrations of porcine pancreatic PLA2 (0-250
nM) is shown in Fig. 1A
(frames a-d). It is clear that porcine pancreatic PLA2 treatment of the cells caused pronounced increase in
ECM invasion while a basal level of 17.9% invasion was observed in the
untreated controls. Fig. 1B shows the dose-response curve of
cellular invasion (% control) as a function of treatment of the cells
with varying concentrations of porcine pancreatic PLA2, assuming the invasion induced by 250 nM PLA2 as
100%. The invasion induced by PLA2 is clearly
dose-dependent and reached a plateau at PLA2
concentration of 100 nM. Since the commercially available Naja naja PLA2 was found to have a 49-kDa
contaminant protein, we purified this PLA2 further by
Sephadex G-25 column chromatography. Fig. 2A
shows the results of SDS-PAGE of porcine pancreatic (lane 1), nonpurified (lane 2), and purified (lane
3) Naja naja PLA2s, respectively. Different
concentrations of this purified Naja naja PLA2
(0-250 nM) was used for migration assay, and a plot of
migrated cells (% control) versus different concentrations
of Naja naja PLA2 is shown in Fig.
2B. Similarly, the fibrosarcoma and sarcoma but not the
lymphoma and mastocytoma cells, when treated with either varying
concentrations of porcine pancreatic or Naja naja PLA2 (0-250 nM), manifested identical patterns
of invasiveness, and treatment of the cells with catalytically active
or BPB-inactivated PLA2-I also yielded similar results
(data not shown). Use of myoglobin (as a nonspecific protein control),
indomethacin, or EDTA had no effect on the ECM invasion of these cells,
and when unconditioned medium was used in the lower chamber as
chemoattractant no invasion occurred (data not shown). Interestingly,
when bee venom or Crotalus adamenteus PLA2s were
used for migration no effect was observed (data not shown). These
results suggest that PLA2-induced ECM invasion is both
PLA2 and cell type-specific process.
Binding and Affinity Cross-linking Studies
In order to
delineate whether the PLA2-I-mediated ECM invasion is
exerted via its cell surface receptor, we first sought to determine
whether these cells expressed functional high affinity PLA2-I receptor. Accordingly, we performed the saturation
binding experiments using 125I-porcine pancreatic
PLA2. We limited the incubation to 2 h because the
plateau was reached within this time (data not shown). The binding data
were analyzed by using LIGAND computer program (26). Scatchard analysis
yielded a straight line, indicating the presence of a single class of
binding site with dissociation constants (Kd) of 0.5 and 2.25 nM and maximum binding capacities (Bmax) of 5.4 and 9.0 fmol/106 cells
for porcine pancreatic and Naja naja PLA2s,
respectively, using NIH 3T3 cells (Fig. 3, A
and B). The use of fibrosarcoma and sarcoma cells with
porcine pancreatic PLA2 as unlabeled ligand yielded
Kd values of 0.76 and 0.73 nM and
Bmax of 4.0 and 6.0 fmol/106 cells,
respectively (Fig. 3, C and D). When purified
Naja naja PLA2 was used as unlabeled ligand in a
saturation binding assay on fibrosarcoma and sarcoma cells, we also
obtained Kd values which were virtually identical to
those observed with NIH 3T3 cells (i.e. 2-3 nM)
(data not shown). Affinity cross-linking of
125I-PLA2-I using DSS on NIH 3T3, fibrosarcoma,
and sarcoma cells yielded a radiolabeled receptor protein band with a
molecular mass of ~180 kDa as resolved by SDS-PAGE under reducing
conditions (Fig. 4 A-C, lanes 2). This
molecular weight was deduced by subtracting the calculated molecular
mass of PLA2 (~14 kDa) from the apparent molecular mass
of radiolabeled, cross-linked protein band (~194 kDa). This band was
virtually abolished when unlabeled porcine pancreatic PLA2
(Fig. 4 A-C, lane 3) or Naja naja
PLA2 (Fig. 4A, lane 4) was used as competing
ligands. As expected, in the absence of DSS no high molecular weight
radiolabeled protein band was detected (Fig. 4, lane 1).
Interestingly, in affinity cross-linking studies, using lymphoma and
mastocytoma cells, we could not detect any such ~194-kDa band (Fig.
4, D and E). These results indicate that
PLA2-I receptor, while readily detectable on NIH 3T3,
fibrosarcoma, and sarcoma cells, was undetectable on lymphoma and
mastocytoma cells.
Characterization of PLA2-I Receptor by Western and Northern Blot Analyses
The cell extracts from NIH 3T3,
fibrosarcoma, sarcoma, lymphoma, and mastocytoma were
immunoprecipitated with purified anti-PLA2-I receptor
antibody and resolved by SDS-PAGE. The receptor protein band was
electrotransferred to nitrocellulose membrane, incubated with
anti-PLA2-I receptor antibody followed by detection with 125I-Protein A. Fig. 5A shows
that this peptide antibody specifically recognizes the
PLA2-I receptor protein (~180 kDa) on NIH 3T3 (lane 1), fibrosarcoma (lane 2), and sarcoma (lane
3) cells but not on lymphoma (lane 4) and mastocytoma
(lane 5) cells. Other proteins within the same molecular
weight range, used as controls, did not cross-react with this antibody
(data not shown). This finding also suggests that PLA2-I
receptor on these murine cells has the same or at least an
immunologically similar epitope to those of the rabbit and bovine
PLA2 receptors. Total RNA was extracted from NIH 3T3,
fibrosarcoma, sarcoma, lymphoma, and mastocytoma cells using the RNAzol
method and tested for the presence of PLA2-I receptor
mRNA by Northern blot analysis. GAPDH gene expression was also
studied in order to check the integrity and the quality of the RNAs and
to rule out any error during gel loading of the samples. Data presented
in Fig. 5B (upper panel) clearly demonstrate that
in all of these cell types (lanes 1-3) except lymphoma and mastocytoma (lanes 4 & 5) a 4.2-kilobase mRNA band
hybridized with the bovine PLA2-I receptor cDNA probe.
Fig. 5B (lower panel) shows the expression of
GAPDH mRNA.
Deglycosylation Studies and Glycoprotein Detection
Confluent
NIH 3T3 cells were pretreated with N-glycosidase F,
incubated with 125I-PLA2-I in the absence or
presence of unlabeled porcine pancreatic PLA2 for binding,
and then cross-linked with DSS. A ~180-kDa receptor protein band
(Fig. 6A, lane 1) was specifically displaced
by unlabeled porcine pancreatic PLA2 (Fig. 6A, lane
2) when cells were not treated with N-glycosidase F. However, the same cells when pretreated with this enzyme and incubated
with 125I-PLA2-I in the absence or presence of
unlabeled porcine pancreatic PLA2 failed to produce this
protein band (Fig. 6A, lanes 3 and 4). This
result suggests that N-linked glycosylation(s) is essential for PLA2 binding. Same results were obtained when
fibrosarcoma and sarcoma cells were used (data not shown). Since our
results indicate that the carbohydrate moiety of the PLA2-I
receptor is essential for ligand binding, we further confirmed the
glycosylation status of this purified receptor by using an ECL
glycoprotein detection system (Amersham Corp.). The data presented in
Fig. 6B confirm the presence of a ~180-kDa glycosylated
receptor protein band. We also studied the effect of deglycosylation of
purified PLA2-I receptor from NIH 3T3 cells by
N-glycosidase F which yielded virtually identical results as
described above (data not shown).
Inhibition of ECM Invasion, PLA2 Receptor Protein Expression, and Receptor Density by PLA2-I Receptor Antisense S-Oligonucleotide
To determine whether
PLA2-I-induced invasion occurred via its receptor, we
transfected NIH 3T3, fibrosarcoma, and sarcoma cells with Lipofectin
containing various oligonucleotides. After 12 h of transfection,
viability of these cells was checked by trypan blue dye exclusion test,
and it was found that >90% of the cells were viable. Nontransfected
and cells transfected with Lipofectin alone were used as controls. The
ECM invasion assay was performed by treating the transfected or
nontransfected cells with 100 nM porcine pancreatic
PLA2. The effects of transfection with Lipofectin alone,
Lipofectin containing antisense, or sense oligonucleotides using NIH
3T3 (A), fibrosarcoma (B), and sarcoma (C) cells, respectively, were shown in Fig.
7. For comparison, the invasion of nontransfected cells
treated with 100 nM porcine pancreatic PLA2 was
considered 100%. We found that there was a 62% inhibition of ECM
invasion in NIH 3T3 cells transfected with ASPLAR (A),
whereas the same treatment of fibrosarcoma (B) and sarcoma
(C) cells yielded 46 and 50% inhibition,
respectively. The NIH 3T3, fibrosarcoma, and sarcoma cells were
also individually transfected with varying concentrations of
ASPLAR, and maximum inhibition was reached when these cells were
transfected with 1.5 µg of ASPLAR (data not shown). None of the
transfections with sense oligonucleotide, used as controls, had any
effect on invasion of these cells. Interestingly, when all cell types
mentioned above were individually pretreated with purified
PLA2-I receptor antibody (IgG) and then incubated with 100 nM porcine pancreatic PLA2, the inhibition of
ECM invasion was 41% for fibrosarcoma (B), 46% for sarcoma
(C), and 50% for NIH 3T3 cells (data not shown). A preimmune IgG was used for pretreatment of these cells as control which
did not have any effect on ECM invasion. Virtually identical results
were obtained when we transfected all three cell types with different
oligonucleotides and performed ECM invasion assay with 100 nM Naja naja PLA2 (data not shown).
Taken together, these results strongly suggest that PLA2-I
mediated ECM invasion is dependent on the expression of functional
PLA2 receptor on the cell surface. We also checked the
receptor protein expression by immunoprecipitation followed by
Western blot analysis after transfecting the NIH 3T3 cells with
Lipofectin, Lipofectin containing SPLAR, or ASPLAR. These results
are shown in Fig. 7D which demonstrate that approximately
50% of inhibition of receptor protein expression was achieved with
ASPLAR treatment of the cells (lane 3) compared with
Lipofectin alone (lane 1) or SPLAR (lane 2).
Identical results were obtained when fibrosarcoma or sarcoma cells were
used (data not shown). We also transfected all three cell types with
different oligonucleotides and measured the receptor binding of
125I-PLA2-I using either porcine pancreatic or
Naja naja PLA2 as unlabeled, competing ligand.
Using porcine pancreatic PLA2 as unlabeled ligand, the
receptor densities on NIH 3T3, fibrosarcoma, and sarcoma cells
transfected with ASPLAR were reduced to 2.5, 1.8, and 2.9 fmol/106 cells, respectively. However, there was no change
in receptor densities when cells were treated with Lipofectin alone or
with SPLAR. Essentially the same levels of reduction in receptor
densities were observed when Naja naja PLA2 was
used as unlabeled ligand (data not shown).
We have demonstrated that treatment of NIH 3T3, fibrosarcoma, and sarcoma cells with porcine and Naja naja but not bee venom and C. adamanteus PLA2s induce dramatic invasion of the ECM (MatrigelTM). However, such response could not be elicited in lymphoma and mastocytoma cells. The signal for this PLA2-induced response appears to be transduced via its high affinity receptor on responder cells (i.e. NIH 3T3, fibrosarcoma, and sarcoma). These findings add a novel dimension to the previously reported list of diverse receptor-mediated cellular functions of PLA2 (16-21). We have used several criteria to determine that in responder cells induction of ECM invasion by porcine and Naja naja PLA2s is a receptor-mediated process. Initially, we found that both porcine pancreatic and Naja naja PLA2s stimulate NIH 3T3, fibrosarcoma, and sarcoma cells to invade a layer of MatrigelTM, an artificial ECM, in a dose-dependent manner and that both catalytically active as well as BPB-inactivated PLA2s were capable of inducing this effect. Use of indomethacin did not affect PLA2-induced ECM invasion suggesting that lipid mediators (e.g. prostaglandins) generated by enzymatic pathways (i.e. cyclooxygenase) downstream from PLA2 catalysis are not involved. Based on these results we sought to determine whether this effect is mediated by PLA2-I receptor. In order to answer this question it was necessary to delineate that these cells expressed functional PLA2-I receptor. Thus, we characterized the PLA2-I receptor in these cells by 125I-PLA2-I affinity cross-linking and binding studies. The results of affinity cross-linking experiments yielded a receptor protein with a molecular mass of ~180 kDa. As expected, 125I-PLA2-I binding was readily displaced by unlabeled porcine pancreatic and Naja naja PLA2s in NIH 3T3, fibrosarcoma, and sarcoma cells with high affinities (Kd = 0.5-2.25 nM). In a saturation binding assay, catalytically inactivated PLA2s also displaced 125I-PLA2-I binding. Moreover, the binding of 125I-PLA2-I to these cell surfaces was not displaced by bee venom or C. adamanteus venom PLA2s (data not shown). Most interestingly, while porcine pancreatic and Naja naja PLA2s dramatically stimulated cellular invasion of ECM, bee venom and C. adamanteus venom PLA2s failed to do so (data not shown). Taken together, these results suggest that porcine pancreatic and Naja naja PLA2s are specific and potent inducers of cellular invasion of the ECM via its high affinity receptor.
To further establish whether porcine pancreatic and Naja naja PLA2-induced invasion occurred via its specific, high affinity receptor, we took advantage of the antisense technology. In separate experiments, NIH 3T3, fibrosarcoma, and sarcoma cells were transfected with PLA2-receptor sense (SPLAR) and antisense (ASPLAR) S-oligonucleotides and measured the invasion, receptor protein expression, and binding. The results showed that ASPLAR transfection of the NIH 3T3, fibrosarcoma, and sarcoma cells drastically inhibited both porcine pancreatic and Naja naja PLA2-induced invasion. These results indicate that PLA2-I receptor is involved in transducing PLA2-induced signal for invasion. This conclusion was further supported by the fact that treatment of the cells with a PLA2-I receptor-specific antibody (IgG) but not with preimmune IgG abolished this effect of porcine pancreatic and Naja naja PLA2s. Immunoprecipitation and Western blot analyses indicated that the expression of PLA2 receptor protein is reduced at least by 50%, when all three cell types (NIH 3T3, fibrosarcoma, and sarcoma) were individually transfected with ASPLAR compared with cells transfected with Lipofectin alone or SPLAR. Binding data also showed that the receptor densities were dramatically reduced in all three cell types mentioned above, when these cells were transfected with ASPLAR. In addition, the similarity between ED50 values for invasion and the Kd values for PLA2-I binding provide compelling evidence that PLA2-I-induced ECM invasion is a receptor-mediated process.
Since we first found normal fibroblasts such as NIH 3T3 cells expressed PLA2 receptor and both pancreatic and Naja naja PLA2 treatment stimulated dramatic invasion of the ECM, we sought to determine whether malignant cancer cells which generally invade the ECM during metastasis express this receptor. Accordingly, we determined the presence or absence of PLA2-I receptor on two different groups of cancer cells, (a) fibrosarcoma and sarcoma and (b) lymphoma and mastocytoma. The results of these studies revealed that while the cancer cells of fibroblastic lineage such as fibrosarcoma and sarcoma cells had high levels of this receptor, it was not detectable on lymphoma or mastocytoma cells. Moreover, the cells that expressed the PLA2-I receptor (fibrosarcoma and sarcoma) responded to PLA2-I treatment by dramatic stimulation of ECM invasion, whereas the ones in which the receptor was not detectable (e.g. lymphoma and mastocytoma) did not respond to this treatment. These results strongly suggest that the observed PLA2-I-mediated stimulation of ECM invasion is receptor-mediated and raise the possibility that PLA2-I may play critical roles in augmenting the invasiveness of certain types of malignant tumor cells.
Although PLA2-I is generally considered to be a digestive enzyme, in light of its receptor-mediated functions, it now appears that this enzyme may have other physiological roles. The levels of secretory PLA2s are increased in several inflammatory diseases (31, 32). The cDNAs of a number of proinflammatory secretory PLA2s have been cloned and characterized (33, 34), and these enzymes are believed to play critical roles in inflammatory processes (reviewed in Refs. 31, 35-37). Since immune cells, in response to various antigens (including infectious agents) invade through the ECM of the vascular bed to arrive at the site of inflammation (or infection), it is possible that increased levels of circulating PLA2-I first activates these cells to invade the vascular ECM. This is followed by interaction of other chemotactic agents (e.g. bacterial lipopolysaccharide, cytokines, arachidonic acid, various eicosanoids, etc.) present at the site of inflammation which cause these cells to leave the circulation and migrate toward an increasing gradient of chemoattractants to finally arrive at the site of local inflammation. The novel biological effect of PLA2-I (i.e. induction of ECM invasion), which we observed in our present study, may be important in several physiological events, such as embryogenesis, tissue remodeling during inflammatory response, and wound healing. Moreover, our findings may also have implications on pathological conditions such as tumor cell invasion during metastasis. In sum, we have demonstrated, for the first time, a novel receptor-mediated effect of porcine pancreatic and Naja naja PLA2s which indicates that this enzyme with its cytokine-like properties plays important physiological as well as pathological roles, in addition to its well recognized digestive function.
We thank Drs. Jean DeB Butler and Sondra W. Levin for critical review of the manuscript and Rick Dreyfuss and Shauna Everett of Medical Arts and Photography Branch, NIH, for their expert technical assistance in photomicrography. We are also grateful to Dr. Jun Ishizaki, for his generous gift of a partial length bovine PLA2-I receptor cDNA clone.