 |
INTRODUCTION |
Secretory phospholipase A2
(sPLA2)1
comprises a large family of lipolytic enzyme, in which 10 isozymes
(groups IB, IIA, IIC, IID, IIE, IIF, III, V, X, and XII) have been
identified in mammals (1, 2). Group I, II, V, and X sPLA2s
are closely related in that they have in common a molecular mass of
14-16 kDa, a conserved active site, Ca2+-binding site,
6-8 disulfides, and a conserved three-dimensional structure. Group III
and group XII sPLA2s each have unique structural characteristics, and display homology to the I/II/V/X
sPLA2s only in the active site and Ca2+-binding
loop (3, 4).
Group IB (sPLA2-IB), or pancreatic sPLA2, is
abundantly present in pancreatic juice, where it plays a role in
digestion of dietary phospholipids, and is also expressed in trace
amounts in several nondigestive organs (5). sPLA2-IIA,
known as an inflammatory-type sPLA2, is detected in a
variety of tissues and cells and is highly up-regulated in response to
inflammatory stimuli both in vitro and in vivo
(6-11). Accumulating evidence suggests that sPLA2-IIA
plays an augmentative role in stimulus-coupled arachidonic acid (AA)
release and subsequent cyclooxygenase (COX)-mediated prostaglandin (PG)
generation (10-16). This enzyme has been also implicated in cell
growth and death (17), atherosclerosis (18), tumorigenesis (19),
degranulation (20), anti-coagulation (21), and defense against bacteria
(22-24). sPLA2-IIC is highly expressed in rodent testes,
but only a pseudogene for this enzyme has been found in the human
genome (25). sPLA2-IID and -IIE are structurally most
related to sPLA2-IIA, showing nearly 50% homology to each other (26-29). sPLA2-IID augments stimulus-induced AA
release in a manner similar to sPLA2-IIA (16) and its
expression is also regulated by proinflammatory stimulus (27).
sPLA2-IIF has an unusually long, proline-rich C-terminal
extension containing a free cysteine (28, 30). The AA-releasing
function of this enzyme has not yet been investigated.
sPLA2-V undergoes stimulus-dependent induction
in various tissues and immune cells (31-35) and promotes PG generation
often more potently than does sPLA2-IIA (12-16, 32-34). The genes for sPLA2-IIA, IIC, IID, IIE, IIF, and V
sPLA2s are clustered at the same chromosome locus (30),
suggesting that they have arisen from a common ancestor by gene
duplication. Based on this and the fact that these 6 sPLA2s
have similar structural features, they are often referred to as the
group II subfamily sPLA2s. sPLA2-X has
structural characteristics of both group I and II enzymes (36) and
releases cellular AA potently even under the conditions where the cells
are resistant to sPLA2-IB, -IIA, and -V (15, 16, 37-40).
The elevated expression of sPLA2-X in some colon
adenocarcinoma neoplastic cells suggests its possible participation in
the COX-2-dependent development of colorectal cancer
(40).
sPLA2s display very distinct membrane and heparanoid
binding properties, which dictate their behaviors in various mammalian cells. The prototypic isozyme, sPLA2-IIA, binds very poorly
to zwitterionic phosphatidylcholine (PC) and thereby acts poorly on the
PC-rich external leaflet of the plasma membrane of resting cells (41,
42). In activated cells, however, this enzyme is sorted into the
caveolin-rich vesicular and perinuclear compartments during its
secretion and internalization processes through the association with
glypican, a glycosylphosphatidylinositol-anchored form of the heparan
sulfate proteoglycan (HSPG). After proper sorting, the enzyme releases
AA from the rearranged membrane microdomains that may be rich in
anionic or oxidized phospholipids (14-16, 43-45). Basic amino acid
clusters located over most of the surface of sPLA2-IIA are
crucial for its association with HSPG and therefore for its cellular
AA-releasing function (42, 46). This pathway, designated as the
HSPG-shuttling pathway, is utilized by several other
heparin-binding sPLA2s, such as sPLA2-IID and
sPLA2-V (12, 13, 16). Conversely, HSPGs can also exert a
negative regulatory effect on these heparin-binding sPLA2s
by facilitating their internalization and degradation in some instances
(47, 48).
In contrast, sPLA2-X, which displays high affinity for
PC-rich membranes but shows no affinity for heparanoids, can release AA
from the PC-rich external surface of the plasma membrane with no
dependence on HSPG (15, 16, 37). This mechanism is called the
external plasma membrane pathway. sPLA2-V, which shows
high affinity toward PC-rich membranes, also utilizes this pathway to
release AA according to the type of cell (16, 33-35, 47). In addition,
the functions of several sPLA2s, such as
sPLA2-IB and -X, can be also modified by the M-type
sPLA2 receptor, which transduces sPLA2-mediated
cell activation signals (49, 50) or promotes sPLA2
internalization and degradation (51).
In the present study, we have examined the regulation of AA release and
eicosanoid biosynthesis by sPLA2-IIF, the newest member of
the group II subfamily of sPLA2s. Our results reveal that
the modes of action of sPLA2-IIF share some features of
those of other group II subfamily sPLA2s and
sPLA2-X, but sPLA2-IIF displays some novel
features. The expression of sPLA2-IIF is elevated in various tissues during inflammatory responses in vivo,
suggesting that, as has been proposed for other group II subfamily
sPLA2s, sPLA2-IIF may play a role in inflammation.
| |
EXPERIMENTAL PROCEDURES |
Materials--
Human embryonic kidney (HEK) 293 cells (Human
Science Research Resources Bank) and rat basophilic leukemia (RBL)-2H3
cells (Riken Cell Bank) were cultured in RPMI 1640 medium (Nissui
Pharmaceutical Co.) containing 10% (v/v) fetal calf serum (FCS;
Bioserum) as described previously (12-16). The cDNAs for mouse and
human sPLA2-IIFs (28, 30), human COX-1 and COX-2 (13),
human microsomal PGE2 synthase (mPGES) (52), rat glypican-1
(14), and porcine 12/15-lipoxygenase (LOX) (43) were described
previously. HEK293 cells stably expressing mouse sPLA2-IIA,
mouse sPLA2-IID, human sPLA2-V, human
sPLA2-X, human COX-2, and human mPGES were described
previously (13-16, 52). BALB/c and C57BL/6 mice were from Nippon
Bio-supply Center.
The enzyme immunoassay kits for PGE2 and LTC4
and the COX-2 inhibitor NS-398 were purchased from Cayman Chemicals.
The rabbit anti-human COX-1 and goat anti-human COX-2 antibodies were
purchased from Santa Cruz. A23187 was purchased from CalBiochem. The LOX inhibitor nordihydroguaiaretic acid (NDGA) was from BIOMOL. Human
IL-1
was purchased from Genzyme. LipofectAMINE PLUS and LipofectAMINE 2000 reagents, Opti-MEM medium, geneticin, and TRIzol reagent were obtained from Invitrogen. Fluorescein
isothiocyanate-conjugated anti-mouse IgG and horseradish
peroxidase-conjugated anti-goat IgG were purchased from
Zymed Laboratories Inc. Mouse monoclonal anti-FLAG
antibody, lipopolysaccharide (LPS; Salmonella minnesota Re
595), and heparin were from Sigma. Mouse IgE anti-trinitrophenyl and
trinitrophenyl-conjugated bovine serum albumin were provided by Dr. H. Katz (Harvard Medical School). Heparin-Sepharose was purchased from
Amersham Biosciences. Zeocin, hygromycin, and the pCR3.1 and
pcDNA3.1 series of vectors containing a neomycin-, zeocin-, or
hygromycin-resistant gene were from Invitrogen.
Antisera for sPLA2-IIF--
Recombinant human
sPLA2-IIF was produced in Escherichia coli after
in vitro refolding (30). The method for the recombinant expression in E. coli, refolding, and purification of mouse
sPLA2-IIF will be reported
elsewhere.2 SDS-PAGE
demonstrated that recombinant mouse and human sPLA2-IIFs exist as a mixture of monomer and homodimer. Reduction by
dithiothreitol gave only the monomeric protein, implying that a free
cysteine residing in the long C-terminal extension contributes to
formation of an intermolecular disulfide-linked homodimer.2
Anti-mouse and human sPLA2-IIF antisera were prepared in
rabbits by Cocalico Biologicals Inc., as described previously (23). Antisera did not cross-react with other sPLA2s
(sPLA2-IB, -IIA, -IIC, -IID, -IIE, -IIF, -V, and -X) when
50 ng quantities of these enzymes were analyzed by immunoblotting.
Establishment of Transfectants--
Establishment of HEK293
transformants was performed as described previously (12-16). Briefly,
1 µg of plasmid (sPLA2 cDNA subcloned into the
pCDNA3.1 vector) was mixed with 2 µl of LipofectAMINE PLUS in 100 µl of Opti-MEM medium for 30 min and then added to cells that had
attained 40-60% confluence in 12-well plates (Iwaki Glass) containing
0.5 ml of Opti-MEM. After incubation for 6 h, the medium was
replaced with 1 ml of fresh culture medium. After overnight culture,
the medium was replaced with 1 ml of fresh medium and culture was
continued at 37 °C in an incubator flushed with 5% CO2
in humidified air. The cells were cloned by limiting dilution in
96-well plates in culture medium supplemented with 1 mg/ml geneticin.
After culture for 3-4 weeks, wells containing a single colony were
chosen, and the expression of each protein was assessed by RNA
blotting. The established clones were expanded and used for the
experiments as described below.
To establish double transformants expressing
sPLA2-glypican or sPLA2-12/15-LOX in
combination, cells expressing each sPLA2 were subjected to
a second transfection with glypican cDNA subcloned into
pCDNA3.1/Zeo(+) or 12/15-LOX cDNA subcloned into
pCDNA3.1/Hyg(+) using LipofectAMINE PLUS. Three days after the
transfection, the cells were used for the experiments or seeded into
96-well plates and cloned by culturing in the presence of 50 µg/ml
zeocin or hygromycin to establish stable transformants.
To assess functional coupling between sPLA2-IIF and either
of the two COX isozymes, cells stably expressing sPLA2-IIF
were transfected with COX-1 or COX-2 subcloned into pCDNA3.1 using LipofectAMINE 2000. Three days after the transfection, the cells were
activated with A23187 to measure PGE2 generation and were subjected to immunoblotting to examine COX-1 or COX-2 expression (see below).
RBL-2H3 cells were seeded into 150-mm diameter dishes and cultured for
2~3 days to subconfluency. The cells (107 cells) were
harvested, washed twice with Opti-MEM, and suspended in 400 µl of
Opti-MEM. The cells were mixed with each cDNA (2~5 µg) and
subjected to electroporation (BTX electroporator ECM600, at 200 V pulse
amplitude, 900 µF capacitance). After culturing for 2 days, the cells
were resuspended in 10 ml of culture medium containing 800 µg/ml
geneticin and seeded into 96-well plates (100 µl/well). After culture
for 2 weeks, single colonies were expanded into 12-well plates. After
reaching confluence, the expression of sPLA2 was assessed
by RNA blotting. As a control, cells transfected with the empty
pCDNA3.1 vector were used.
Measurement of sPLA2 Activity--
sPLA2
activity was assayed by measuring the amounts of free radiolabeled
fatty acids released from the substrate
1-palmitoyl-2-[14C]arachidonoyl-phosphatidylethanolamine
(2-AA-PE), 1-palmitoyl-2-[14C]linoleoyl-PE (2-LA-PE),
2-AA-PC, or 2-LA-PC (Amersham Biosciences). Each reaction mixture
(total volume 250 µl) consisted of appropriate amounts of the
required sample, 100 mM Tris-HCl (pH 7.4), 4 mM CaCl2, and 10 µM substrate. After incubation
for 10-30 min at 37 °C, [14C]AA was extracted, and
radioactivity was quantified, as described previously (46).
Heparin Binding--
Affinity of sPLA2s for
heparin-Sepharose was assessed as described previously (12). Briefly,
~25 ml of culture supernatants of HEK293 transfectants were applied
to a heparin-Sepharose column (1 × 5 cm) pre-equilibrated with 10 mM Tris-HCl, pH 7.4, containing 150 mM NaCl
(TBS) at a flow rate of 20 ml/h. After extensive washing with TBS, the
bound proteins were eluted using 10 mM Tris-HCl, pH 7.4, with a 0.15-1 M NaCl gradient. The PLA2
activities of each fraction was measured as described above.
RNA Blotting--
Approximately equal amounts (~5 µg) of
total RNA obtained from the cells were applied to separate lanes of
1.2% (w/v) formaldehyde-agarose gels, electrophoresed, and transferred
to Immobilon-N membranes (Millipore). The resulting blots were then
probed with the respective cDNA probes that had been labeled with
[32P]dCTP (Amersham Biosciences) by random priming
(Takara Biomedicals). All hybridizations were carried out as described
previously (12-16, 46). Equal loading of RNA in each lane was verified
by reprobing the same membranes with glyceraldehyde-3-phoshate
dehydrogenase (GAPDH) cDNA (CLONTECH) (not shown).
SDS-PAGE/Immunoblotting--
Lysates from 105 cells
were subjected to SDS-PAGE using 10% gels under a reducing condition.
The separated proteins were electroblotted onto nitrocellulose
membranes (Schleicher and Schuell) using a semi-dry blotter
(MilliBlot-SDE system; Millipore). After blocking with 3% (w/v) skim
milk, the membranes were probed with the respective antibodies
(1:20,000 dilution for COX-1 and 1:5,000 dilution for COX-2) for 2 h, followed by incubation with horseradish peroxidase-conjugated anti-goat IgG (1:5,000 dilution) for 2 h, and were visualized using the ECL Western blot system (Amersham Biosciences) (12-16).
Reverse Transcription-Polymerase Chain Reaction
(RT-PCR)--
Synthesis of cDNA was performed using 0.5 µg of
total RNA from mouse tissues and avian myeloblastosis reverse
transcriptase, according to the manufacturer's instructions supplied
with the RNA PCR kit (Takara Biomedical). Subsequent amplification of
the cDNA fragments was performed using 1 µl of the
reverse-transcribed mixture as a template with mIIF-5' primer and
mIIF-
C primer (see below). The PCR condition was 94 °C for
30 s and then 35 cycles of amplification at 94 °C for 5 s
and 68 °C for 4 min, using the Advantage cDNA polymerase mixture
(CLONTECH). Expression of GAPDH was assessed by 25 cycles of PCR amplification using specific primers
(CLONTECH). The PCR products were analyzed by 1%
agarose gel electrophoresis with ethidium bromide. The gels were
further subjected to Southern blot hybridization using mouse
sPLA2-IIF cDNA as a probe.
Construction of sPLA2-IIF
Mutants--
sPLA2-IIF mutants were produced by PCR with
the Advantage cDNA polymerase mixture. The condition of PCR was 25 cycles at 94 °C, 55 °C, and 72 °C for 30 s each. The
primers used were as follows: mIIF-5' primer,
5'-ATGAAGAAATTCTTTGCCATC-3'; mIIF-3'-primer, CTAGGTTGAGACAGGGGTCGC-3'; mIIF-C primer, 5'-TTAGCAGTTGGGTGTGGGGCC-3'; mIIF-C136S primer, 5'-GAAGTCACCTCTGGCATG GC-3'; mIIF-C136S AS primer,
5'-GCCATGCCCAGAGGTGACTTC-3'; hIIF-5' primer,
5'-ATGAAGAAGTTCTTCACCGTG-3'; hIIF-3' primer,
5'-CTAGGGAGGGGCGGGGGGCGC-3'; hIIF-C primer,
5'-GTAGCAGGTGACCTCCTCAGG-3'; mIIF-FLAG primer, 5'-TTACTTGTGATCGTCGTCCTTGTAGTCGGTTGAGACAGGGGTCGC-3';
mIIF-C primer, 5'-CTACTTGTGATCGTCGTCCTTGTAGTCTTAGCAGTTGGGTGTGG-3';
hIIF-FLAG primer,
5'-TTACTTGTGATCGTCGTCCTTGTAGTCCTAGGGAGGGGCGGGGG-3'; and hIIF-C-FLAG primer,
5'-CTACTTGTGATCGTCGTCCTTGTAGTCGCTGCAGTTGGGCGTGG-3' (FLAG sequence underlined). To prepare the C mutant
constructs, PCR was preformed with mIIF-5' and mIIF-C primers or
with hIIF-5' and hIIF-C primers using species-matched
sPLA2-IIF cDNA as a template. As for the C-terminal
FLAG-tagged constructs, mIIF-5' or hIIF-5' primer and mIIF-FLAG,
mIIF-C-FLAG, hIIF-FLAG, or hIIF-C-FLAG primers were mixed with
species-matched sPLA2-IIF cDNA and PCR was carried out.
To construct the C136S mutant, the first PCR was conducted with mIIF-5'
and mIIF-C136S AS primers or with mIIF-3' and mIIF C136 primers using
mouse sPLA2-IIF cDNA as a template. The resulting two
primary PCR fragments were mixed, denatured at 94 °C for 5 min,
annealed at 37 °C for 30 min and then 55 °C for 2 min, and
extended at 72 °C for 4 min during each cycle. A secondary PCR
product with specific mutation was obtained after 25 additional PCR
cycles with mIIF-5' and mIIF-3' primers. Each PCR product was ligated
into the pCR3.1 and was transfected into Top10F' supercompetent cells
(Invitrogen). The plasmids were isolated and sequenced using a
Taq cycle sequencing kit (Takara Biomedicals) and an
autofluorometric DNA sequencer DSQ-1000L (Shimadzu) to confirm the sequences.
Activation of HEK293 Cells--
HEK293 cells (5 × 104/ml) were seeded into each well of 24- or 48-well
plates. To assess fatty acid release (12-16), [3H]AA or
[3H]oleic acid (OA) (both from Amersham Biosciences) (0.1 µCi/ml) was added to the cells in each well on day 3, when they had
nearly reached confluence, and culturing was continued for another day. After three washes with fresh medium, 250 µl (24-well plate) or 100 µl (48-well plate) of RPMI 1640 with or without 10 µM
A23187 with 1% FCS or 1 ng/ml IL-1 and/or 10% FCS was added to
each well, and the amount of free [3H]AA or
[3H]OA released into the supernatant was measured. The
percentage release was calculated using the formula [S/(S + P)] × 100, where S and P are the radioactivity measured in the supernatant
and cell pellet, respectively. The supernatants from replicate cells were subjected to the PGE2 enzyme immunoassay.
To assess transcellular PGE2 biosynthesis (13), two cell
populations (2.5 × 104 cells/ml for each) were added
to the same wells of 48-well plates and cultured for 4 days. Then the
cells were stimulated with IL-1 in medium containing 10% FCS for
4 h and PGE2 released into the supernatants was quantified.
Activation of RBL-2H3 Cells--
The cells (5 × 104 cells/ml) were seeded into 24-well plates and cultured
for 2 days in 1 ml of culture medium. Then the cells were sensitized
with 1 µg/ml IgE anti-trinitrophenyl in culture medium for 30 min,
washed twice with culture medium, and activated for 10 min at 37 °C
with 10 ng/ml trinitrophenyl-conjugated bovine serum albumin as an
antigen in culture medium (16, 20). After harvesting the supernatants,
the remaining cells were collected and disrupted by two freeze-thawing
cycles. Release of LTC4 was assessed by enzyme immunoassay
according to the manufacturer's instruction.
Confocal Laser Microscopy--
Cells grown on collagen-coated
cover glasses (Iwaki Glass) were fixed with 3% paraformaldehyde for 30 min in phosphate-buffered saline (PBS). After three washes with PBS,
the fixed cells were sequentially treated with 1% (w/v) bovine serum
albumin (for blocking) and 0.2% (v/v) Triton X-100 (for
permeabilization) in PBS for 1 h, with anti-FLAG antibody (1:500
dilution) for 1 h in PBS containing 1% bovine serum albumin, and
then with fluorescein isothiocyanate-goat anti-mouse IgG (1:500
dilution) for 1 h in PBS containing 1% bovine serum albumin.
After six washes with PBS, the cells were mounted on glass slides using
Perma Fluor (Japan Tanner), and the sPLA2 signal was
visualized using a laser scanning confocal microscope (IX70; Olympus),
as described previously (14).
LPS Treatment of Mice--
LPS (5 mg/kg) was administered
intraperitoneally to 4-week-old male C57BL/6 mice (Nippon Bio-Supply
Center). After 24 h, mice were sacrificed by bleeding, their
organs were removed, and RNA was extracted by homogenization in TRIzol
reagent by 10 strokes of a Potter homogenizer at 1,000 rpm.
Mouse Ear Atopic Dermatitis--
Five repeated topical
applications of 2,4-dinitrobenzene (DNFB) to the ears of BALB/c, but
not C57BL/6, mice result in contact hypersensitivity of the ears and
significant elevation of serum IgE level, accompanied by increased
TH1 response for the onset of skin dermatitis and
TH2 response in the lymph node (53). Briefly, mouse ears
were painted with 25 µl of 0.15% (w/v) DNFB or vehicle
(acetone:olive oil 3:1) once a week. The ears were removed 24 h
after the fifth painting and subjected to RNA extraction. Replicate ear sections were fixed by formalin, embedded in paraffin, and stained with hematoxylin and eosin to verify the progress of
inflammation. All the procedures and analyses of other parameters are
detailed elsewhere (53).
Immunohistochemistry--
The fresh synovial tissues of human
rheumatoid arthritis were obtained at synovectomy, and promptly placed
in optimal cutting temperature compound (Sakura Industry), frozen, and
then stored at 
80 °C for 1 week. Thin sections (4 µm thickness)
from the frozen tissues were fixed in acetone for 10 min and dried in
air for 10 min. The samples were pretreated with 0.3% (v/v)
H2O2 for 5 min in TBS and then with 5% (w/v)
skim milk in TBS containing 0.02% (v/v) Tween 20. After washing with
TBS containing 0.02% Tween 20 three times, they were sequentially
incubated for 30 min with anti-human sPLA2-IIF antiserum
for 10 min with biotinylated anti-mouse immunoglobulin, and for 10 min
with peroxidase-conjugated streptavidin at room temperature. After
washing, the sections were visualized with diaminobentidine
tetrahydrochloride. Biotinylated anti-mouse immunoglobulin,
peroxidase-conjugated streptavidin and diaminobentidine
tetrahydrochloride were supplied in the LSAB kit (Dako).
Statistical Analysis--
Data were analyzed by Student's
t test. Results are expressed as the mean + S.E., with
p = 0.05 as the limit of significance.
| |
RESULTS |
Enzymatic Properties--
cDNAs for wild-type (WT) mouse or
human sPLA2-IIF, a mutant in which the unique portion of
the C-terminal extension was entirely deleted (C), and another
mutant in which the extra Cys residue in the long C-terminal extension
that likely allows formation of a disulfide-linked homodimer was
replaced by Ser (C136S), the structures of which are illustrated in
Fig. 1A, were each transfected into HEK293 cells. After selection with geneticin, clones stably expressing WT and mutant enzymes at comparable levels, as assessed by
RNA blotting (Fig. 1B), were used in subsequent studies.
Since results with mouse and human enzymes turned out to be similar, we
hereafter report the results mostly with the mouse enzyme.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1.
Establishment of HEK293 transfectants stably
expressing WT and mutant mouse sPLA2-IIF and their in
vitro enzymatic activity. A,
structures of the WT and two mutant forms (C and C136S) of mouse
sPLA2-IIF (mIIF). B, expression of the WT and
two mutant forms of mIIF in HEK293 transfectants, as assessed by RNA
blotting. Each lane contains 5 µg of RNA. C, in
vitro enzymatic activity of the WT and two mutant forms of mIIF
toward 2-AA-PE. Equal portions (25 µl) of culture supernatants of
1 × 106 cells grown in 2 ml of medium in 6-well
plates were taken for the PLA2 assay. D,
substrate specificity of the WT enzyme, using 2-AA-PE, 2-LA-PE,
2-AA-PC, and 2-LA-PC as substrates.
|
|
When the in vitro enzymatic properties of the WT and mutant
sPLA2-IIF were examined using culture supernatants of the
transfectants as enzyme sources, the activity of the C mutant toward
2-AA-PE was slightly lower than, whereas the activity of the C136S
mutant was comparable with, that of the WT enzyme (Fig. 1C).
The WT enzyme (Fig. 1D) and both mutants (data not shown)
hydrolyzed PE ~8 times more efficiently than PC, with only a modest
preference for AA to linoelic acid at the sn-2 position. In
comparison, the PE versus PC hydrolytic ratios of human
sPLA2-IIA, -V, and -X were approximately >100:1, 2:1, and
1:1, respectively, under the same assay condition (12-15). The WT and
mutant enzymes showed a similar Ca2+ dependence with an
optimal pH at 7-9 (data not shown). Studies with highly pure
preparations of WT and the C mutant of sPLA2-IIF, produced as recombinant proteins in E. coli, showed that
both proteins had comparable activity on
1-palmitoyl-2-oleoyl-phosphatidylserine vesicles or
1-palmitoyl-2-oleoyl-phosphatidylglycerol (less than 1.3-fold
difference) vesicles.2 These data are consistent with our
current studies using HEK293 cell transfectants showing that both the
WT enzyme and the C mutant have similar specific activities on
phospholipid vesicles in vitro.
Cellular Functions--
When [3H]AA- or
[3H]OA-prelabeled HEK293 transfectants were cultured for
4 h, FCS-dependent releases of [3H]AA
and [3H]OA by sPLA2-IIF-expressing cells were
increased 2- and 1.7-fold, respectively, relative to those by replicate
control cells (Fig. 2A).
[3H]AA, but not [3H]OA, release by the
sPLA2-IIF-expressing cells was markedly augmented when IL-1
was added to FCS (Fig. 2A). IL-1-stimulated
[3H]AA release by the sPLA2-IIF-expressing
cells proceeded gradually over 8 h of culture (data not shown),
with a concomitant increase in PGE2 generation over the
same time period (Fig. 2B). IL-1-stimulated PGE2
generation by the sPLA2-IIF-expressing cells was completely blocked by the COX-2 inhibitor NS-398 (see below), indicating absolute
dependence of this delayed PGE2 generation on endogenous COX-2. The sPLA2-IIF-expressing cells expressed more COX-2
(Fig. 2C) and its downstream mPGES (Fig. 2D) than
did the control cells after IL-1 stimulation. The expression of COX-2
peaked at 1 h and then declined gradually (Fig. 2C),
whereas that of mPGES, which was already visible even before IL-1
stimulation, increased to reach a maximum by 1 h and maintained a
plateau level thereafter (Fig. 2D), in the
sPLA2-IIF-expressing cells. The enhanced expression of
COX-2 (14-16) and mPGES (Fig. 2D) was also observed with
the cells transfected human sPLA2-V as well as those
transfected with other group II subfamily sPLA2s, such as
sPLA2-IIA and -IID, but not those transfected with
sPLA2-X (data not shown). When HEK293 cells transfected
with sPLA2-IIF were co-cultured with those transfected with
COX-2 (the transcellular PG biosynthesis assay (13)), PGE2 production was severalfold higher than that produced by the
sPLA2-IIF or COX-2 single transfectants (Fig.
2E), indicating that sPLA2-IIF secreted from one
cell can act on neighboring COX-2-expressing cells. Furthermore,
PGE2 generation by the COX-2/mPGES co-transfectants was
further augmented when co-cultured with the
sPLA2-IIF-expressing cells (Fig. 2E).
sPLA2-IIF is thus capable of supplying AA to the downstream
sequential biosynthetic enzymes, COX-2 and mPGES, for delayed
PGE2 production in autocrine and paracrine fashions.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2.
AA release and PGE2 generation by
sPLA2-IIF in HEK293 cells. A, delayed fatty
acid release. Parental HEK293 cells and cells expressing mouse
sPLA2-IIF (mIIF) were prelabeled with [3H]AA
or [3H]OA, washed, and then cultured for 4 h with
1% FCS ( FCS), 10% FCS (+ FCS), or 10% FCS plus 1 ng/ml IL-1 to
assess [3H]AA or [3H]OA release (mean ± S.E., n = 4). B, time course of
PGE2 generation by the control and mIIF-transfected cells
stimulated for the indicated periods with 10% FCS plus 1 ng/ml
IL-1. C, induction of COX-2. RNAs obtained from the
control and mIIF-transfected cells, which were stimulated for the
indicated periods with 10% FCS plus 1 ng/ml IL-1, were subjected to
RNA blotting using COX-2 and mIIF cDNAs as probes. D,
induction of mPGES. Control, human sPLA2-V (hV)-transfected
and mIIF-transfected cells were cultured for the indicated periods with
10% FCS plus 1 ng/ml IL-1, and mPGES expression was assessed by RNA
blotting. The bottom panel shows 18 S ribosomal RNA stained
with ethidium bromide to verify equal sample loading in each lane.
E, transcellular PGE2 generation. Control,
COX-2-expressing or COX-2/mPGES-coexpressing cells were cultured with
(+) or without () mIIF-expressing cells for 4 h in the presence
of 10% FCS plus 1 ng/ml IL-1. PGE2 released into the
supernatants was quantified. F, immediate AA release.
Control and mIIF-expressing cells, prelabeled with
[3H]AA, were stimulated for the indicated times with 10 µM A23187 to assess [3H]AA release.
G, immediate PGE2 production. Control and
mIIF-expressing cells were transfected with either COX-1 or COX-2.
Three days after the transfection, the cells were stimulated for 30 min
with A23187 to assess PGE2 generation (top
panel). Lysates of remaining cells were subjected to
immunoblotting to verify the expression of COX-1 or COX-2 protein. In
B-G, representative results of three to four independent
experiments are shown.
|
|
A23187-induced immediate AA release (0-30 min) was also markedly
increased in the sPLA2-IIF-expressing cells over the
control cells (Fig. 2F). When either COX-1 or COX-2 was
co-transfected with sPLA2-IIF (the expression of each COX,
as assessed by Western blotting, is shown in Fig. 2G,
bottom), A23187-induced immediate PGE2
production via each COX was markedly augmented in the
sPLA2-IIF-expressing cells relative to replicate control
cells (Fig. 2G).
The truncated sPLA2-IIF mutant C increased
A23187-induced immediate AA release and IL-1-induced delayed AA release
and PGE2 generation only modestly, whereas the mutant C136S
was as active as WT in both immediate and delayed responses (Fig.
3, A-C). A catalytically
inactive sPLA2-IIF mutant, G29S, in which Gly29
in the conserved Ca2+-binding site was replaced by Ser, was
unable to elicit AA release under all conditions tested (data not
shown).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3.
The unique C-terminal extension is essential
for cellular functions of sPLA2-IIF. A-C,
parental HEK293 cells and cells transfected with the WT or two mutant
forms (C and C136S) of mouse sPLA2-IIF (mIIF)
were stimulated for 30 min with A23187 (A) or for 4 h
with 10% FCS plus 1 ng/ml IL-1 (B and C) to
assess AA release (A and B) and PGE2
production (C) (mean ± S.E., n = 4).
D, RBL-2H3 cells transfected with the WT and C mIIF were
sensitized with IgE and activated with a suboptimal concentration of
hapten-specific antibody for 30 min to assess LTC4
production (mean ± S.E., n = 3).
Inset, expression of the WT and C mIIF was assessed by
RNA blotting.
|
|
When WT and C mutant sPLA2-IIF enzymes were transfected
into rat mastocytoma RBL-2H3 cells (the expression of WT and mutant C was verified by Northern blotting (Fig. 3D,
inset)), the WT enzyme increased
IgE/antigen-dependent immediate production of LTC4 markedly over the replicate mock-transfected cells,
whereas LTC4 production by the C-transfected cells was
minimal (Fig. 3D). Given the fact that the C mutant has
PLA2 activity nearly comparable with that of the WT enzyme
in vitro (Fig. 1C), these observations suggest
that the C-terminal extension unique to sPLA2-IIF may be
crucial for its cellular function.
Cellular Actions of sPLA2-IIF Occur Independently of
HSPG--
More than 95% of the expressed sPLA2-IIF was
secreted from the HEK293 transfectants into the culture medium, and
washing the cells with 1 M NaCl, which solubilizes the
HSPG-associated form of several group II subfamily sPLA2s
from cell surfaces (12-16), did not alter the recovery of
sPLA2-IIF. This result suggests that sPLA2-IIF
has a very weak or no affinity for heparanoids. To confirm this, we
tested the binding of this enzyme to heparin-Sepharose. When the pooled
culture supernatant of the sPLA2-IIF-expressing cells was
applied to the heparin-Sepharose column, the activity of
sPLA2-IIF was recovered exclusively in the flow-through
fraction (Fig. 4A, top). In
comparison, sPLA2-IID, used as an example of a
heparin-binding sPLA2, bound tightly to the column (Fig.
4A, middle) and nearly 40% of the enzyme
expressed in HEK293 cells was recovered only after cells were washed
with 1 M NaCl (16). The sPLA2-IIF mutant C136S
also failed to bind to the heparin-Sepharose column (data not shown),
whereas the truncated mutant C showed a weak heparanoid affinity,
with as much as 10% being associated with the column (Fig.
4A, bottom). This increase in heparin affinity of
the C mutant is likely because the extra C-terminal extension, which
is rich in acidic amino acids (28, 30), is absent.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 4.
sPLA2-IIF acts on cells
independently of HSPG. A, heparin-Sepharose
chromatography. Pooled culture supernatants of sPLA2-IIF
(mIIF) (top), mouse sPLA2-IID
(mIID) (middle), and mIIF-C mutant
(bottom) expressing cells were applied to the
heparin-Sepharose column, and the bound proteins were eluted with a
linear gradient of NaCl from 0.15 to 1 M (dashed
line). An aliquot of each fraction was taken for PLA2
assay. B, effects of glypican co-transfection on AA release
by various sPLA2s. Control cells and cells expressing mouse
sPLA2-IIA (mIIA), mIIF, or human sPLA2-X (hX)
were transfected with glypican, and [3H]AA release 4 h after stimulation with 10% FCS plus 1 ng/ml IL-1 was assessed.
Values are mean ± S.E. of three independent experiments
(p < 0.05 versus mIIA without glypican
co-transfection). The expression of glypican was verified by RNA
blotting (not shown). C, effects of heparin on AA release by
various sPLA2s. Control cells and cells expressing mIIA,
mIIF, or hX were preincubated with 400 µg/ml heparin overnight and
then stimulated for 4 h with 10% FCS plus 1 ng/ml IL-1 to
assess [3H]AA release. Values are mean ± S.E. of
three independent experiments (p < 0.05 versus mIIA without heparin treatment).
|
|
Consistent with the heparin-nonbinding feature, the AA-releasing
function of sPLA2-IIF was not affected by coexpression with the HSPG glypican (Fig. 4B), which acts as a functional
adaptor for the heparin-binding group II subfamily sPLA2s
(IIA, IID, and V) (14, 16). Indeed, under the same experimental
condition, AA release by sPLA2-IIA, the function of which
largely depends on HSPG, was significantly augmented, whereas that by
sPLA2-X, which utilizes the HSPG-independent pathway, was
unaffected, by glypican coexpression (Fig. 4B), as reported
previously (14, 16). Furthermore, IL-1-stimulated AA release and
PGE2 generation by sPLA2-IIF was insensitive to
exogenous heparin (Fig. 4C). Under the same condition, AA
and PGE2 release by sPLA2-IIA was markedly suppressed (due to transfer of the cell surface-bound
sPLA2-IIA to soluble heparin (14, 16)), whereas AA release
by sPLA2-X was insensitive to exogenous heparin (Fig.
4C). These results imply that sPLA2-IIF acts on
cells in a cellular HSPG-independent manner.
To determine the cellular sites of sPLA2-IIF action, we
performed immunocytostaining of HEK293 cells transfected with mouse and
human sPLA2-IIFs that were C-terminal tagged with the FLAG epitope. Their expression levels were verified by Northern blotting and
enzyme activity (data not shown). Confocal laser microscopic analyses
revealed that the outline of the cells was intensely stained for human
(Fig. 5A) and mouse (Fig.
5B) sPLA2-IIFs, indicating their location on the
plasma membrane. There were no detectable cytoplasmic punctate and
perinuclear signals, which were prominent when the other group II
subfamily sPLA2s utilizing the HSPG-shuttling pathway, such
as sPLA2-IIA, IID, and V, were immunostained (14, 16).
Washing the cells with heparin did not affect the plasma membrane
staining of sPLA2-IIF (data not shown). In contrast, the
plasma membrane signal for the truncated mutant C was very faint
(Fig. 5C).
View larger version (51K):
[in this window]
[in a new window]
|
Fig. 5.
Confocal microscopic analysis of
sPLA2-IIF in HEK293 cells. Cells transfected with
human sPLA2-IIF (hIIF) (A), mouse
sPLA2-IIF (mIIF) (B), and mIIF mutant C
(C), which were tagged with the FLAG epitope, were fixed,
permeabilized, incubated sequentially with anti-FLAG antibody and
fluorescein isothiocyanate-conjugated anti-mouse IgG, and then
visualized by confocal laser microscopy (top panels).
Overlay of fluorescent signals on the phase-contrast visions
(bottom panels) clearly indicates that the WT hIIF and mIIF
are located on the plasma membrane.
|
|
sPLA2-IIF Action Is Facilitated by 12/15-LOX
Metabolites--
The fact that the cellular AA releasing functions of
sPLA2-IIF (Fig. 2A) as well as the other group
II subfamily sPLA2s (IIA, IID, and V) (12-16) are markedly
augmented by appropriate stimuli (IL-1 and A23187 in this case) argues
that their phospholipid-hydrolytic actions are, in common, facilitated
by cellular membrane rearrangement that occurs during cell activation.
Since lipid oxidation is one of the key events leading to membrane
rearrangement (10, 44, 45) and since the increased AA release by
sPLA2-IIA or sPLA2-V transfectants is
attenuated by treatment with several antioxidants that inhibit
12/15-LOX (10), we examined the effect of NDGA, a 12/15-LOX-inhibitable
antioxidant (10), on the AA-releasing function of
sPLA2-IIF. As illustrated in Fig.
6A, IL-1-stimulated [3H]AA release and PGE2 generation by HEK293
cells expressing sPLA2-IIF and sPLA2-IIA, as
well as those expressing sPLA2-V and sPLA2-IID (data not shown), were markedly reduced by NDGA. Analysis by thin layer
chromatography confirmed that the released radioactivity was largely
(>90%) associated with AA (data not shown), indicating that the
suppression by NGDA was not due to the inhibition of the release of LOX
products. In contrast, AA release by sPLA2-X-expressing cells was largely NDGA-insensitive (Fig. 6A), consistent
with the notion that this enzyme does not require membrane
rearrangement for its cellular action (15, 16, 37-40). Furthermore, AA
release by sPLA2-IIF, like that by sPLA2-IIA
(10), was markedly augmented following 12/15-LOX co-transfection
(Fig. 6B). The COX-2 inhibitor NS-398, which almost
completely blunted PGE2 generation, did not alter AA
release by these sPLA2s (Fig. 6A). Collectively,
these results suggest that certain 12/15-LOX, but not COX, reaction products may facilitate membrane rearrangement and resultant AA releasing function of the group II subfamily sPLA2s.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6.
Cellular functions of sPLA2-IIF
are facilitated by 12/15-LOX metabolites. A,
effects of NDGA and NS-398 on AA release (left) and
PGE2 production (right) by various
sPLA2s. Control cells and cells expressing mouse
sPLA2-IIA (mIIA), mouse sPLA2-IIF
(mIIF), or human sPLA2-X (hX) were
incubated for 4 h with 10% FCS plus 1 ng/ml IL-1 in the
presence or absence of 10 µM NDGA or 1 µM
NS-398 to assess [3H]AA release and PGE2
production. Values are mean ± S.E. of four to six separate
experiments (p < 0.05 versus without drug
treatments in each group). PGE2 generation by hX was not
assessed, since this enzyme does not produce PGE2 unless
COX-2 is enforcibly transfected (15). B, effects of
12/15-LOX overexpression on AA release. Control cells and cells
expressing mIIA or mIIF were transfected with 12/15-LOX, and then
stimulated for 30 min with A23187 to assess [3H]AA
release. Values are mean ± S.E. of three independent experiments
(p < 0.05 versus without 12/15-LOX
transfection in each group).
|
|
Induction of sPLA2-IIF during
Inflammation--
C57BL/6 mice were injected with LPS
intraperitoneally, and the expression of sPLA2-IIF in
several tissues before and 24 h after the injection was examined
by RT-PCR, followed by Southern blotting. Although the basal expression
level of sPLA2-IIF was undetectable or barely detectable in
the tissues examined (brain, heart, liver, colon, and testis), it was
markedly increased in these tissues after LPS treatment (Fig.
7A).
View larger version (45K):
[in this window]
[in a new window]
|
Fig. 7.
Induction of sPLA2-IIF expression
in the mouse during inflammation. RNAs obtained from various
tissues of mice with or without 24 h treatment with LPS
(A) and from ears of mice with or without five repeated
treatments with DNFB (B) were subjected to RT-PCR for mouse
sPLA2-IIF (35 cycles), which was visualized by Southern
hybridization using sPLA2-IIF cDNA probe, and GAPDH (25 cycles), which was detected in agarose gels by ethidium bromide
staining.
|
|
The expression of sPLA2-IIF in mouse ears with or without
five repeated treatments with DNFB, which induces a chronic
inflammation similar to human atopic dermatitis (53), was next examined
by RT-PCR/Southern blotting. In the ears of BALB/c mice, which
represent a hypersensitive strain displaying profound TH1
and TH2 responses in this atopic dermatitis model (53),
there was a marked elevation in the expression of sPLA2-IIF
after DNFB challenge (Fig. 7B). In contrast, its expression
was unchanged in the DNFB-treated ears of C57BL/6 mice (Fig.
7B), a low responder strain (53).
Finally, we performed immunohistochemical staining of human
rheumatoid arthritic tissues using a polyclonal
anti-sPLA2-IIF antiserum. Intense enzyme staining was seen
in capillary endothelial cells (Fig. 8,
A and B), synovial lining cells (Fig. 8,
A and C), synovial sublining cells (Fig. 8,
A and D), and plasma cells (i.e.
larger cells) (Fig. 8, B-D) in the intima. Some lymphocytes were also weakly positive, whose cytosol was scant (Fig.
8B). Control antiserum did not stain these tissue sections
at all (data not shown). Identification of the above cell variety was
judged from conventional hematoxylin and eosin staining serially
sectioned to the specimen used for immunohistochemistry.
View larger version (140K):
[in this window]
[in a new window]
|
Fig. 8.
Immunohistochemistry of rheumatoid arthritis
tissue for sPLA2-IIF. A, villous synovial
tissue including synovial lining cells, subsynovial lining cells,
plasma cells and capillary endothelial cells (×36). Panels
B-D show higher magnification (×361). Arrows indicate
synovial lining (black) and sublining (green)
cells, capillary endothelial cells (red), plasma cells
(yellow), and lymphocytes (blue).
|
|
| |
DISCUSSION |
Current evidence suggests that the cellular AA releasing function
of various sPLA2s is crucially influenced by their ability to bind to the PC-rich lipid interface for some enzymes (15, 16,
33-35, 37-40, 47) and to anionic heparanoids for others (12-16, 46).
sPLA2-V and -X, which bind well to PC-rich vesicles in vitro (35, 37), are capable of binding to the PC-rich
outer leaflet of the plasma membrane of unstimulated cells
leading to the release of AA and OA (15, 16, 33-35, 37-40, 47). The
AA released by these sPLA2s may pass across the plasma
membrane, diffuse into the cytosol (or possibly with an aid of fatty
acid-binding proteins), and reach the perinuclear COX and 5-LOX enzymes
for eicosanoid synthesis (transcellular AA movement is also possible). The heparin binding, basic enzymes sPLA2-IIA, -IID, and -V
bind to the HSPG glypican and are sorted into caveolin-rich vesicular and perinuclear compartments in activated cells (12-16,
46). This localization may spatially and temporally allow efficient transfer of AA to adjacent COX and 5-LOX enzymes. Membrane
rearrangement that may occur during cell activation is also a
prerequisite event for the function of the HSPG-shuttled
sPLA2s, where exposure of anionic and/or oxidative
phospholipids may facilitate their interfacial membrane association and
subsequent hydrolysis (10, 12-16, 44-46). Moreover, this perturbed
membrane microdomain appears to be rich in AA relative to other fatty
acids, since these sPLA2s release AA in preference to OA
(12, 15) despite the fact that they display almost no fatty acyl
selectivity in vitro (6, 7, 12, 26, 27, 31, 35). This
pathway is supported by a recent study by Cho and co-workers (54), who
demonstrated the HSPG- and catalytic activity-dependent
internalization of and subsequent substrate hydrolysis at the
perinuclear membrane by exogenously added sPLA2-V and
sPLA2-IIA (in the latter case, cell membrane perturbation
by cytokine priming is essential).
The present functional analyses of sPLA2-IIF reveal that
this enzyme behaves in a manner similar, but not identical, to other group II subfamily sPLA2s. In the HEK293 cell system,
sPLA2-IIF releases AA in preference to OA in response to
IL-1 and augments IL-1-stimulated COX-2 expression and attendant
PGE2 production (Fig. 2, A-D). The expression of
mPGES, a terminal enzyme acting downstream of COX-2 in the delayed
PGE2-biosynthetic pathway (52), is also augmented by
sPLA2-IIF and by other group II subfamily sPLA2s (Fig. 2D), implying that the increased AA
release and the up-regulation of the two sequential downstream enzymes
(COX-2 and mPGES) contribute to the amplification of the
PGE2-biosynthetic cascade. Although the mechanisms for the
induction of COX-2 and mPGES by the group II subfamily
sPLA2s are unclear, requirement of their catalytic
activities for this event (15) suggests that certain reaction products
spatiotemporally generated by the sPLA2 action in
compartmentalized sites may in turn trigger the COX-2/mPGES induction
machinery. sPLA2-IIF also elicits A23187-induced immediate
AA release and, if COX-1 is expressed by transfection, can be coupled
with COX-1 for immediate PGE2 synthesis (Fig. 2, F and G). Furthermore, AA release by
sPLA2-IIF is attenuated by the LOX-inhibitable antioxidant
NDGA and, conversely, facilitated by introduction of the
membrane-oxidizing enzyme 12/15-LOX (Fig. 6). All of these aspects are
commonly observed with the signaling group II subfamily
sPLA2s such as IIA, IID, and V (10).
However, unlike these heparin-binding group II subfamily
sPLA2s, sPLA2-IIF does not bind to heparanoids,
and its function does not depend on the HSPG glypican (Fig. 4). The
heparin-nonbinding feature of sPLA2-IIF is consistent with
the fact that this enzyme is an acidic protein (calculated pI of 5.4 and 4.7 for the mouse and human enzymes, respectively), whereas other
group II subfamily sPLA2s are basic (1, 2, 25-30). Likely
due to its failure to bind cellular HSPG, sPLA2-IIF is
largely secreted into the extracellular medium. Of interest, an
immunocytochemical study demonstrates that at least some
sPLA2-IIF remains associated with the plasma membrane (Fig.
5), rather than in the vesicular and perinuclear components in which
the heparin-binding sPLA2s are commonly distributed (10,
14, 16). Thus, sPLA2-IIF is the first example among the
"group II" sPLA2s that is capable of functioning on the
plasma membrane with no interaction with HSPG.
As noted above, the plasma membrane action of sPLA2-X and
-V depends essentially on their high capacity to interact with PC-rich membranes (16, 35, 37-40, 47). Studies of the action of
sPLA2-IIF on phospholipid vesicles in vitro
shows that it hydrolyzes PE ~8 times more efficiently than PC (Fig.
1D), yet this PC hydrolyzing activity is still far more
efficient than that of sPLA2-IIA (30). Mouse and human
sPLA2-IIFs bind to PC vesicles in vitro with
affinities comparable with those of sPLA2-V and
-X.2 Modest and parallel increases in AA and OA release by
sPLA2-IIF-expressing HEK293 cells cultured in 10% FCS
(Fig. 2A) is compatible with marked increases in AA and OA
release in sPLA2-X-expressing cells (15) and suggests that
sPLA2-IIF, like sPLA2-X, acts on the plasma
membrane. Interaction with PC in the plasma membrane is also supported
by the finding that sPLA2-IIF, which is exocytosed rapidly
from RBL-2H3 cells by degranulation, elicits LTC4
production (Fig. 3D), an event occurring predominantly
through the external plasma membrane pathway (16). However, it is also
possible that sPLA2-IIF releases AA from IgE/Ag-stimulated
RBL-2H3 cells by acting on perturbed membrane microdomain on the
surface, as discussed below.
Despite these facts, the ability of sPLA2-IIF to bind PC in
the outer plasma membrane appears insufficient for optimal action of
this enzyme, since unlike sPLA2-X that can elicit the full AA-releasing response without an additional stimulation (15, 16,
37-40), sPLA2-IIF requires an appropriate stimulus, such as IL-1, to do so (Fig. 2A). Thus, it is tempting to
speculate that cell activation signaling may lead to formation of a
perturbed microdomain on the plasma membrane, to which
sPLA2-IIF preferentially attacks. This putative microdomain
for sPLA2-IIF action may contain anionic and oxidized
phospholipids, and also be enriched in AA-containing phospholipids
because the IL-1-stimulated response is AA-selective (Fig.
2A).
Site-directed mutagenesis of sPLA2-IIF in two cell models
(i.e. HEK293 and RBL-2H3) shows that the unique C-terminal
extension is essential for its cellular function (Fig. 3), even though
the truncation of this region does not profoundly affect the in
vitro enzyme activity (Fig. 1C). It may be noted that
the plasma membrane staining, which appears with the WT enzyme, is not
observed if this C-terminal domain is deleted (Fig. 5). The plasma
membrane-acting enzyme, sPLA2-X, does not give a clear cell
outline staining (16), probably because the dissociation constant for
sPLA2-X on PC-rich vesicles in vitro is ~100
µM (i.e. 50% of sPLA2-X is
vesicle-bound in the presence of 100 µM PC),2
and the plasma membrane phospholipid concentration in these cellular studies is estimated to be <10 µM. These results raise
the intriguing possibility that the unique C-terminal extension is
required for this enzyme to tightly interact with the plasma membrane.
The mutation of a free cysteine residing in the middle portion of this
domain does not alter the in vitro or cellular functions of
the enzyme (Figs. 1C and 3), indicating that homodimer
formation or possible heterodimer formation with other cellular
proteins via this cysteine residue is nonessential. Another notable
characteristics of this extra C-terminal extension is the abundance of
acidic and proline residues (28, 30). However, it is currently unclear whether this unique feature is responsible for the plasma membrane distribution of the enzyme.
Taken together, we prefer the model in which sPLA2-IIF can
act on the plasma membrane through a combined interaction of its interfacial binding surface (that which surrounds the opening of the
catalytic site slot) and the C-terminal extension. Cell activation-directed transbilayer movement of anionic phospholipids and
accelerated oxidation may further increase its ability to interact with
the plasma membrane and thereby increase AA release. However, the
possibility that a small, undetectable, fraction of enzyme is present
in internalized perinuclear compartments, localized through an unknown
pathway, cannot be ruled out. Regardless of the mechanisms involved,
our present results clearly indicate that sPLA2-IIF is an
additional member of the signaling PLA2s that can control
cellular AA metabolism.
As are other group II subfamily sPLA2s (6-11, 27, 29, 32),
sPLA2-IIF is a stimulus-inducible enzyme, whose expression is markedly induced in various tissues during at least some forms of
inflammation (Figs. 7 and 8). The presence of potential binding motifs
for several stimulus-activated transcription factors, such as C/EBP,
CREB, NF
B, and AP-1, within ~2.5 kb upstream of the human
sPLA2-IIF gene, as revealed by nucleic data base
search, also supports this notion. Although high level
sPLA2-IIF expression has been reported in the adult mouse
testis (28), in this study its expression in this tissue is low (Fig.
7A), probably due to differences in mouse strains or ages.
Our present data indicate that its expression is induced (even though
at lower expression levels than sPLA2-V (32)) in a variety
of tissues after LPS challenge (Fig. 7A). Increased
expression of sPLA2-IIF in the atopic dermatitic ears of
mice (Fig. 7B), together with the fact that it is expressed
in murine cultured mast cells (20) and that it increases
LTC4 production in a rat mast cell line (Fig. 3D), argues that this enzyme may be involved in exacerbation
of the allergic response.
sPLA2-IIF is also detected in human rheumatoid arthritic
tissues, in which synovial lining cells in the intima, capillary endothelial cells, and plasma cells are intensely stained (Fig. 8).
Previous work has shown that sPLA2-IIA is abundantly
present in rheumatoid arthritic tissues (6-8, 43), yet the
presence of other related sPLA2 enzymes have not been
addressed. An immunohistochemical study has demonstrated that
sPLA2-IIA exists in fibroblastic and macrophage-like cells
in the synovial lining and subsynovial lining layers, perineural cells,
endothelial cells, mast cells, vascular smooth muscle cells, and
extracellular matrices (43), observations that are largely consistent
with our own immunohistochemical
study.3 Furthermore, our
preliminary analyzes using the sPLA2 group-specific antisera show that other sPLA2s, particularly those
belonging to the group II subfamily, are also present in rheumatoid
arthritic tissues.3 Co-distribution of related
sPLA2s in rheumatoid synovial epithelium is suggestive of
their redundant functions in AA release, whereas sPLA2-IIF
expressed in the plasma cells, in which other sPLA2s were
not detected,3 might play a unique immunoregulatory role.
Even though sPLA2-IIF is capable of augmenting AA
metabolism in the HEK293 and RBL-2H3 cell overexpression system and is
induced during inflammation, there is still no direct evidence that
this enzyme indeed plays a role in lipid mediator production under pathophysiological conditions in vivo. There may be some
novel functions of sPLA2s other than the regulation of AA
metabolism. For instance, sPLA2-IIA has a bactericidal
activity, which may be a major physiological function of this isozyme
(22-24). It is unlikely that sPLA2-IIF plays a similar
anti-bacterial role, since this enzyme is rather acidic (28, 30),
whereas cationic property is a prerequisite for the anti-bacterial
function of sPLA2s (22-24). Several sPLA2s
elicit several cellular functions through binding to the
sPLA2 receptor, an action that occurs independently of the
catalytic activity (49, 50), although it is currently unknown whether
sPLA2-IIF binds to the receptor. The high expression of
sPLA2-IIF in embryonic tissues and testis (28, 30) may reflect a unique functional role for this enzyme. The presence of
multiple sPLA2 enzymes in local environments forces us to
explore the precise functions of each enzyme in more detail. This
should be of great value for the development of novel sPLA2
inhibitors as drugs with a broad or specific spectrum of therapeutic
and prophylactic activities.
§,
,
TOP
ABSTRACT
INTRODUCTION
