| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 26, 24153-24163, June 27, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(cPLA2) and Secretory Phospholipase A2 (sPLA2) in Hydrogen Peroxide-induced Arachidonic Acid Release in Murine Mesangial Cells
ACTIVITY THAT IS RESPONSIBLE FOR ARACHIDONIC ACID RELEASE*
¶ ||
From the
Medical and
Anesthesia Services, Massachusetts General
Hospital, Department of Medicine and Anesthesia, Harvard Medical School, and
¶Harvard-MIT, Division of Health Science and
Technology, Charlestown, Massachusetts 02129-2060
Received for publication, January 14, 2003 , and in revised form, March 25, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
cytosolic
PLA2 (cPLA2) and secretory PLA2s
(sPLA2s) during H2O2-induced arachidonic acid
(AA) release using two types of murine MC: (i)
MC+/+, which lack group IIa and V
PLA2s, and (ii) MC–/–,
which lack groups IIa, V, and IV PLA2s.
H2O2-induced AA release was greater in
MC+/+ compared with
MC–/–. It has been argued that
cPLA2 plays a regulatory role enhancing the activity of
sPLA2s, which act on phospholipids to release fatty acid. Group
IIa, V, or IV PLA2s were expressed in
MC–/– or
MC+/+ using recombinant adenovirus vectors.
Expression of cPLA2 in H2O2-treated
MC–/– increased AA release to a
level approaching that of H2O2-treated
MC+/+. Expression of either group IIa
PLA2 or V PLA2 enhanced AA release in
MC+/+ but had no effect on AA release in
MC–/–. When sPLA2 and
cPLA2 are both present, the effect of
H2O2 is manifested by preferential release of AA
compared with oleic acid. Inhibition of the ERK and protein kinase C signaling
pathways with the MEK-1 inhibitor, U0126, and protein kinase C inhibitor, GF
1092030x, respectively, and chelating intracellular free calcium with
1,2-bis(2-aminophenoyl)ethane-N,N,N',N'-tetraacetic
acid-AM, which also reduced ERK1/2 activation, significantly reduced
H2O2-induced AA release in
MC+/+ expressing either group IIa or V
PLA2s. By contrast, H2O2-induced AA release
was not enhanced when ERK1/2 was activated by infection of
MC+/+ with constitutively active MEK1-DD. We
conclude that the effect of group IIa and V PLA2s on
H2O2-induced AA release is dependent upon the presence
of cPLA2 and the activation of PKC and ERK1/2. Group IIa and
V PLA2s are regulatory and cPLA2 is responsible
for AA release. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
cytosolic PLA2 (cPLA2)
(5,
6), paralogs of this enzyme
(7), and several
Ca2+-independent PLA2s (iPLA2s)
(8). A number of mammalian
sPLA2s have been identified to date (groups IB, IIA, IIC, IID, IIE,
IIF, III, V, X, and XII) and they display distinct yet partially overlapping
tissue distributions (4,
9,
10). Whereas
cPLA2 has a preferential effect on AA-containing membrane
phospholipids as compared with those containing other fatty acids,
sPLA2s do not exhibit acyl chain specificity. The
cPLA2, group IIa, and group V PLA2s have each
been implicated as the primary PLA2 responsible for production of
AA and its metabolites in fibroblastic, endothelial, mast and macrophage
mammalian cell lines (1,
11–15).
Most investigators have concluded that cPLA2 plays a
regulatory role whereas sPLA2 provides most of the AA release that
occurs in response to agonists. Both cPLA2 and the
sPLA2s have been implicated in various physiological and
pathological functions including lipid digestion, release of proinflammatory
mediators, cell proliferation, ischemic injury, inflammatory disease, cancer,
and anti-bacterial defense (6,
9,
16).
Oxidant stress has been implicated in numerous proinflammatory responses in mammalian cells (17–20). Hydrogen peroxide (H2O2) triggers AA release and metabolism in various cell types (21–24). The understanding of which forms of PLA2 are important for AA release and how multiple forms may interact has been hampered by the fact that mammalian cells generally contain more than one form of PLA2. Furthermore, various PLA2 inhibitors and antisense approaches lack specificity and/or efficacy even though much useful information has been derived from these approaches (15, 25, 26). Thus the understanding of specific interactions of PLA2 enzymes contributing to AA release and metabolite production is complex and has been difficult to clarify.
Cellular kinase signal transduction pathways have been implicated in
PLA2 activation and its downstream effects. cPLA2
is regulated post-translationally by phosphorylation and by calcium
(5). cPLA2
contains an N-terminal calcium-dependent phospholipid binding domain.
Mitogen-activated protein kinase (MAPK) cascades and protein kinase C (PKC)
have been implicated in the phosphorylation and activation of
cPLA2
(29–33).
Balsinde and Dennis (34) and
Hernandez et al. (35)
reported the involvement of extracellular signal-regulated kinases (ERKs) as
mediators of cross-talk between sPLA2s and cPLA2
in polymorphonuclear leukocytes and macrophages. Others have reported that
sPLA2s can activate MAPK cascades and PKC, which subsequently
activate cPLA2
(30,
32,
36).
H2O2 has been reported to increase the activity of MAPK
cascades (23) and stimulate
cPLA2 activity in smooth muscle cells
(37).
Renal mesangial cells are an important source of both eicosanoid and
reactive oxygen species generation in the kidney during normal and
pathological states. Reactive oxygen species have been implicated in the
response of mesangial cells to hyperglycemia
(38) and increased levels of
prostaglandins are characteristic of diabetic and other glomerulopathies
(39,
40). The role of
PLA2s in the generation of eicosanoids and propagation of
inflammation has been extensively studied by several groups using mesangial
cells derived from rats. Our laboratory has produced a mouse line with the
cPLA2 gene mutated
(41). The mouse strains
(C57b/6 and SV/129) used to construct the
cPLA2–/– strain
have spontaneous null mutations in the gene encoding group IIa PLA2
(42). In addition to group IIa
PLA2, our murine mesangial cells do not express group V
PLA2 under quiescent or stimulated conditions, unlike rat mesangial
cells, which are known to synthesize group IIa and V PLA2s upon
stimulation of the cells with cytokines
(43). Two types of MC were
used: (i) cPLA2+/+ MC
(MC+/+), which lack group IIa and V
PLA2s and (ii)
cPLA2–/– MC
(MC–/–), which lack group IIa, V,
and IV PLA2s. We expressed group IV, IIa, and V
PLA2 proteins in MC–/–
and MC+/+ with recombinant adenoviral
vectors. Using this approach we dissect the specific roles played by
cPLA2 and the sPLA2s in the mediation of
H2O2-induced AA release in mesangial cells.
To better define the cross-talk between cPLA2 and
sPLA2s during oxidant stress, we examined the effect of expression
of various forms of PLA2 on H2O2-induced AA
release in murine mesangial cells. This is the first time that the cross-talk
between cPLA2 and sPLA2s in a mammalian cell has
been studied by utilizing recombinant adenovirus and cPLA2
knockout MC, which lack group IIa, V, and IV PLA2s. We
report here that group IIa and V PLA2s potentiate
H2O2-induced AA release in a cPLA2-,
PKC-, and ERK1/2-dependent manner. Activation of ERK1/2 is necessary but not
sufficient for H2O2-mediated AA release. We conclude
that in murine mesangial cells, cPLA2 is the major enzyme
responsible for AA release whereas sPLA2 serves an amplifying
role.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
polyclonal antibody was provided by Dr. Andrey Cybulsky.
Peroxidase-conjugated, goat anti-rabbit immunoglobulin was from DAKO
(Carpinteria, CA). Total ERK1/2 and p38 kinase antibodies were from Santa Cruz
Biotechnology, Inc. (Santa Cruz, CA). Phospho-specific (Ser505)
cPLA2 polyclonal antibody and phospho-specific ERK1/2 and
p38 kinase antibodies were from Cell Signaling Technology (Beverly, MA).
Protein measurements based on Bradford's assay were performed with reagents
from Bio-Rad.
Generation of Primary Murine Mesangial Cells—Primary murine
MC were cultivated from wild type and cPLA2 knockout mice.
Cell lines were generated from kidneys taken from 3
cPLA2+/+ and 3
cPLA2–/–
littermates. The cortices of each mouse were dissected under sterile
conditions. The glomeruli were isolated by mechanical disruption, passaged
through 140 µm and then collected on a 46-µm sieve, followed by
centrifugation. Following isolation the glomeruli were treated with 1 mg/ml
collagenase IV for 30 min at 37 °C. Clones with apparent MC morphology
were used for further processing. The cells were grown in RPMI 1640 (Cellgro)
supplemented with 10% (v/v) fetal bovine serum (Invitrogen) at 37 °C in
95% air, 5% CO2. MC exhibit the typical stellate morphology.
Moreover, they stain positively for the intermediate filaments desmin and
vimentin, which are considered to be specific for myogenic cells. Passages
6–10 of MC were used for the reported experiments.
Analysis of mRNA Expression of Group V PLA2 in MC—Total RNA was extracted from mesangial cells by using Ultraspec (Biotecx, Houston, TX) according to the manufacturer's instructions. Primers derived from the 5' (5'-CAGGGGGCTTGCTAGAACTCAA-3') and 3' (5'-AAGAGGGTTGTAAGTCCAGAGG-3') ends of the coding region of the mouse group V PLA2 were used in a reverse transcriptase-polymerase chain reaction to clone the mouse group V PLA2 cDNA. The reverse transcriptase reaction was carried out after 10 min incubation at 70 °C of denatured template and dNTP with 10 pmol of reverse primer and then incubated with Moloney murine leukemia virus reverse transcriptase (Stratagene, La Jolla, CA) for 1 h at 42 °C to generate a cDNA template for PCR. The PCR was carried out for 40 cycles of 95 °C for 1 min, 50 °C for 1 min, and 72 °C for 1 min in buffer with 0.5 units of Taq polymerase. The amplified products were resolved in 2% agarose gels and visualized with ethidium bromide. The expression of glyceraldehyde-3-phosphate dehydrogenase was used as internal control.
Total PLA2 and iPLA2 Assay—MC were washed with phosphate-buffered saline and lysed by sonication in a buffer containing 150 mM NaCl, 1 mM EDTA, and 50 mM Tris/HCl at pH 9. The lysate was centrifuged at 3,000 rpm for 30 min at 4 °C. Total PLA2 activity was assayed by measuring the amounts of free fatty acid released from the substrate [14C]PE. Each reaction mixture consisted of an aliquot of the required sample, 75 mM Tris-HCl, pH 7.4, 5 mM CaCl2, and 2 µM substrate. After incubation for 30 min at 37 °C, the [14C]PE released was extracted by Dole's method and the radioactivity was counted (44).
For the iPLA2 assay, lysate was preincubated with various concentrations of BEL (iPLA2 inhibitor) for 10 min and then iPLA2 activity was assayed by measuring the amounts of free fatty acid released from the substrate [14C]PE. Each reaction mixture consisted of an aliquot of the sample, 75 mM Tris-HCl, pH 7.4, 5 mM EGTA, and 2 µM substrate. After incubation for 30 min at 37 °C, the [14C]PE released was extracted by Dole's method and the radioactivity was counted.
Arachidonic and Oleic Acid Release—Confluent MC in 12-well plates were labeled for 14–16 h either with 0.15 µCi of [3H]AA (specific activity: 1 Ci = 37 GBq) or 0.15 µCi of [14C]OA (specific activity: 1 Ci = 38 GBq) in 1 ml of RPMI 1640 with 0.2% (v/v) fetal bovine serum. After labeling, the medium was removed, and cells were washed three times with RPMI containing 0.2% bovine serum albumin. To measure H2O2-induced AA or OA release, cells were exposed to 75 µM H2O2 or vehicle in RPMI, 0.2% bovine serum albumin for 3 to6hat37 °C in 95% air, 5% CO2. The medium was removed and centrifuged to remove detached cells. The cells were solubilized with 1 ml of 0.5% NaOH. The radioactivity in 800 µl of supernatant and cells was measured in a liquid scintillation counter. The amount of [3H]AA or [14C]OA released into the medium was expressed as a percentage of the total (cell-associated plus released).
SDS-PAGE and Western Blot Analysis—Total MC extracts were
harvested with lysis buffer (20 mM HEPES, pH 7.4, 2 mM
EGTA, 1 mM dithiothreitol, 1 mM NaVO4, 1%
Triton X-100, 10% glycerol, 2 µM leupeptin, 400 µM
phenylmethylsulfonyl fluoride, 50 mM 
-glycerophosphate) and
mixed with 6x sample buffer. Fifteen micrograms of cell extracts were
subjected to SDS-PAGE (10% acrylamide gel), and proteins were transferred onto
Immobilon-P membranes for 1 h at 400 mA using a Bio-Rad transblot apparatus.
The transfer buffer used was 50 mM Tris-HCl, pH 7.4, 384
mM glycine, 0.01% SDS, and 20% methanol. After the transfer, the
membrane was blocked with a buffer containing: 1x phosphate-buffered
saline, 5% nonfat dry milk, and 0.5% Tween 20. The membrane was incubated with
primary antibodies and horseradish peroxidase-conjugated secondary antibody,
respectively. Proteins were visualized with an enhanced chemiluminescence
detection system (PerkinElmer Life Sciences).
Measurement of Intracellular Free Calcium Concentration—Intracellular free Ca2+ concentration ([Ca2+]i) was determined with the Ca2+-sensitive fluorescent dye Fura-2 according to Cheung et al. (45) with modification. Cells grown on coverslips coated with bovine collagen type I were rinsed with phosphate-buffered saline and loaded with 3 µM Fura-2AM in Earle's balanced salt solution. Pluronic F-127 (20%) at 1:1000 (v/v) dilution was added to Fura-2AM to facilitate cell loading. In addition, 2 mM probenecid was added to minimize intracellular compartment transport or extrusion of Fura-2-free acid. Cells were loaded with Fura-2AM for 1 h at 37 °C and washed 2–3 times with Earle's balanced salt solution containing probenecid. The coverslips were positioned in a quartz cuvette containing 3.5 ml of Earle's balanced salt solution with probencid for fluorescence analysis using a Shimadzu RF-5000 spectrofluorophotometer (Shimadzu, Columbia, MD). [Ca2+]i was calculated as equal to Kd (224 nM) x (R – Rmin)/(Rmax – R) according to Grykiewicz et al. (46) as described previously. Fluorescence emission was monitored at 505 nm. R is the ratio (F1/F2) of the fluorescence at excitation 340 nm to that at excitation 380 nm.
Construction of Recombinant Adenovirus Vectors Carrying the cDNA of
Group IIa and V PLA2—The system for
generation of recombinant adenoviruses has been described previously
(47). Briefly the mouse V
PLA2 cDNA (500-base pair HindIII/XbaI restriction
fragment) and human IIa PLA2 cDNA (780-base pair
KPN1/XhoI restriction fragment) were subcloned into a
shuttle vector (pAdTrack-CMV). The shuttle vector plasmid was linearized with
PmeI (restriction endonuclease) and transformed together with
supercoiled adenoviral backbone vector (pAdEasy-1) into Escherichia
coli strain BJ5183 by electroporation in a Bio-Rad Gene Pulser
electroporator. Recombinants were selected with kanamycin (50 µg/ml) and
screened by restriction endonuclease digestions (PacI, SpeI,
and BamHI). The recombinant adenoviral construct was transformed into
DH 5 cells for large scale amplification by electroporation and was
purified by CsCl banding. For production of adenoviruses in mammalian cells,
the recombinant adenoviral construct linearized by PacI was
transfected into 293 cells using LipofectAMINE and Opti-MEM (Invitrogen). The
process of viral production was monitored by visualization of the green
fluorescent protein (GFP) expression, which is incorporated into the viral
backbone. Viruses were harvested after 7–10 days. The viruses were
purified by CsCl banding and viral particles were measured by optical density
(OD) (viral particles = OD x dilution factor (usually 10) x
1012). The recombinant adenoviral vectors carrying the cDNA of
cPLA2 (Ad-cPLA2), cDNA of GFP/E.
coli LacZ gene encoding -galactosidase (Ad-GFP/LacZ) as an
adenovirus control, and cDNA of constitutively active MEK1-DD (Ad-MEK1-DD)
were constructed as previously described in our laboratory
(48,
49).
Introduction of PLA2 Enzymes into Primary
Murine Mesangial Cells—Subconfluent MC were infected with
adenoviral vectors at varying levels of infection, as reflected by plaque
forming units/cell, for 48 h in RPMI 1640 with 2% (v/v) fetal bovine serum.
The adenovirus-mediated gene transfers were followed by the expression of GFP
under UV light for Ad-IIa and VPLA2s and Ad-GFP/LacZ, and by
Western blot analysis of cell lysate for Ad-cPLA2 and
Ad-MEK1-DD, respectively. After confirming the infection, cells were used for
experiments.
Statistical Analysis—Data are expressed as the mean ± S.E. Statistical differences among the groups were calculated on raw data using the analysis of variance test. Significance was tested using Student's t test between groups. A value of p < 0.05 was chosen to determine statistical significance. Each experiment was performed in triplicate and independently three to five times.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
and Its Effect
on H2O2-induced AA
Release in
MC–/–—MC–/–
were infected with Ad-cPLA2. Seventy-two hours after
infection, total cell extracts were collected to confirm expression of the
cPLA2 in MC–/–
by Western blotting. Fig.
1A shows the expression of cPLA2 in a
dose-dependent manner after Ad-cPLA2 infection at different
multiplicity of infections (m.o.i.) in
MC–/–.
|
At the sites of inflammation, H2O2 levels can reach
relatively high local concentrations (range from 0.1 to 1.0 mM) in
the presence of activated polymorphonuclear leukocytes
(50–52).
To study the effect of cPLA2 on
H2O2-induced AA release from MC,
MC–/– infected with varying
m.o.i. of Ad-cPLA2, were stimulated with 75 µM
H2O2 for periods of 3 and 6 h and AA released into media
was measured (Fig.
1B). The elevation of AA release was dependent on
Ad-cPLA2 m.o.i. and exposure time to
H2O2. Expression of cPLA2 in
H2O2-treated
MC–/– significantly increased AA
release at 3 and 6 h. H2O2 at a concentration of 75
µM did not result in cellular injury as monitored by trypan blue
staining and lactate dehydrogenase release (data not shown).
Expression of Group IIa and V PLA2s in
MC—The mouse strains (C57b/6 and SV/129) used to construct the
cPLA2–/– strain
have spontaneous null mutations in the gene encoding group IIa
PLA2. Furthermore, MC derived from
cPLA2+/+
(MC+/+) and
cPLA2–/–
(MC–/–) mice, which have mixed
C57b/6 and SV/129 backgrounds, do not express group V PLA2 mRNA
under quiescent or stimulated conditions
(Fig. 2A).
MC+/+ and
MC–/– were infected with
Ad-sPLA2. Seventy-two hours after infection, total cell extracts
were collected to confirm expression of the sPLA2s in MC by Western
blotting. Fig. 2B
shows the dose-dependent expression of either group IIa or V PLA2s
after Ad-IIa PLA2 or Ad-V PLA2 infection at different
m.o.i. in MC+/+. Infection with adenovirus
expressing the GFP/LacZ enzyme had no effect upon the levels of either of the
PLA2s. This indicates that the infection process is not associated
with endogenous sPLA2 expression.
Fig. 2C shows that
there is an increase in total PLA2 activity in
Ad-IIaPLA2, Ad-V PLA2-infected unstimulated MC. No
increase in activity is seen in cells infected with Ad-GFP/LacZ compared with
uninfected cells. LY311727, which selectively inhibits group IIa
(53) and group V
PLA2 activity (54),
completely inhibited the increased total PLA2 activity in the
Ad-IIa- or Ad-V PLA2-infected cells.
|
Role of Cellular Group IIa and V PLA2s in
the H2O2-induced AA
Release—To evaluate potential interactions between
sPLA2s and cPLA2 during the
H2O2-induced AA release, sPLA2s enzymes were
expressed in MC+/+ and
MC–/– using recombinant
adenoviral vectors encoding group IIa PLA2 (Ad-IIa PLA2)
or V PLA2 (Ad-V PLA2).
Fig. 3A shows that
H2O2-induced AA release is significantly increased when
MC+/+ are infected with either Ad-IIa
PLA2 (m.o.i. = 50) or Ad-V PLA2 (m.o.i. = 50). Ad-IIa or
Ad-V PLA2s has no effect, however, on AA release in
MC–/–
(Fig. 3B). Neither
group IIa nor V PLA2s have any effect on unstimulated AA release
levels in MC+/+ and
MC–/–.
Fig. 3C shows that
Ad-GFP/LacZ (m.o.i. = 40) infection had no effect on
H2O2-induced AA release when compared with non-infected
groups. Thus adenovirus infection itself had no effect upon AA release.
|
Role of Cellular sPLA2 in the
H2O2-induced OA
Release— Because, unlike cPLA2, sPLA2
does not have a preferential effect on AA-containing membrane phospholipids,
we compared the release of OA and AA when cPLA2 and
sPLA2 are both present in MC+/+.
Fig. 4 shows that there is a
small increase in H2O2-induced OA release at 6 h in Ad-V
PLA2 (m.o.i. = 50) infected MC+/+
compared with Ad-IIa PLA2 (m.o.i. = 50) infected
MC+/+. However, this difference is not
statistically significant and overall OA release is not significantly
increased in the presence of either with sPLA2. When
sPLA2s and cPLA2 are present in
MC+/+, H2O2 has a much
greater effect on AA release than it does on OA release.
|
Effect of PKC and MEK1 Inhibition on
H2O2-induced ERK
Activation and cPLA2 Ser505
Phosphorylation—H2O2 is known to activate the
ERK1/2 signaling pathway and cPLA2
(23). Phosphorylation of
ERK1/2 and cPLA2 were evaluated at various time points after
treatment of confluent MC with 75 µM H2O2.
ERK activity was deduced from Western blotting with antibodies specific for
the phosphorylated, activated forms of ERK1/2. The phosphorylation of
cPLA2 was deduced from Western blotting with antibody
specific for the phosphorylated Ser505 of
cPLA2.
The level of phosphorylation of ERK1/2 reached a peak within 30 min after
exposure to 75 µM H2O2 in both
MC+/+ and
MC–/–
(Fig. 5A). There was
no significant quantitative difference in degree of phosphorylation of ERK1/2
comparing MC+/+ and
MC–/–. MEK1 is a dual specificity
kinase that phosphorylates and activates ERK1/2 on threonine and tyrosine
residues (55,
56). The phosphorylation of
ERK1/2 was completely prevented by the presence of the MEK1 inhibitor, U0126
(10 µM) (Fig.
5B). Because cPLA2 is a target of the
MEK1/ERK signaling cascade, the effect of MEK1 inhibition by U0126 on
cPLA2 phosphorylation was evaluated. cPLA2
phosphorylation peaked 30 min after exposure of cells to
H2O2. The phosphorylation of cPLA2 was
significantly inhibited with U0126 treatment
(Fig. 5B).
|
PKC has been shown to activate MAP kinase
(57,
58). The phosphorylation of
ERK1/2 was unaffected at 30 min and inhibited by the presence of the PKC
inhibitor, GF 1092030x (5 µM) at 60 and 180 min after
H2O2 addition. By contrast the phosphorylation of
cPLA2 was significantly reduced with GF 1092030x treatment
at each of the time points (Fig.
5C). Neither H2O2 nor U0126/or GF
1092030x had any effect on total ERK1/2 and cPLA2
expression. Because there is no effect of GF 1092030x at 30 min on ERK1/2
phosphorylation when the phosphorylation of cPLA2 is down,
the effect of PKC on cPLA2 phosphorylation is independent of
ERK1/2 at this time point.
Role of ERK1/2 in H2O2-induced AA Release—To evaluate the effect of inhibition of the ERK signaling pathway with U0126 upon the H2O2-induced AA release, mesangial cells were preincubated with U0126 (10 µM) and then treated with H2O2. U0126 significantly reduced the H2O2-induced AA release from MC+/+ (Fig. 6A). As discussed above (Fig. 3A), the expression of group IIa or V PLA2s in MC+/+ enhances the AA release induced by H2O2. Interestingly, U0126 completely inhibited H2O2-induced AA release in the presence of either sPLA2 (Fig. 6B).
|
Role of Constitutively Activated ERK1/2onH2O2-induced AA Release—To further evaluate the role for activation of ERK1/2 in H2O2-induced AA release, MC+/+ were infected with an adenoviral vector carrying constitutively active MEK1-DD (Ad-MEK1-DD). Fig. 7A reveals augmentation of the phosphorylation of ERK1/2 in a dose-dependent manner after Ad-MEK1-DD infection in MC+/+. However, Ad-MEK1-DD had no effect on MC+/+ basal AA release (Fig. 7B). Furthermore, H2O2-induced AA release was not increased by activated ERK1/2 in MC+/+ (Fig. 7B). Thus the activation of ERK1/2 alone is not sufficient to augment H2O2-induced AA release.
|
Role of PKC in H2O2-induced AA Release—To evaluate the effect of inhibition of the PKC signaling pathway with GF 1092030x upon the H2O2-induced AA release, mesangial cells were preincubated with GF 1092030x (5 µM) and then treated with H2O2. GF 1092030x significantly reduced the H2O2-induced AA release from MC+/+ (Fig. 8A). GF 1092030x greatly inhibited H2O2-induced AA release in the presence of either sPLA2 (Fig. 8B). Treatment of MC with U0126 and GF 109203x simultaneously did not have any synergistic effect on H2O2-induced AA release in the presence of either sPLA2.
|
Role of Intracellular Free Calcium in
H2O2-induced AA
Release—H2O2 is known to increase the
[Ca2+]i
(59). An increase in
[Ca2+]i is a critical step for
cPLA2 to translocate from the cytosol to perinuclear
membranes for initiation of stimulus-coupled AA release
(60,
61).
Fig. 9A demonstrates
that continuous treatment with H2O2 induces an elevation
of [Ca2+]i in MC. The concentration
of [Ca2+]i reached a peak at 90 min
after exposure to 75 µM H2O2 in
MC+/+ and remained elevated until at least
120 min. To evaluate the effect of inhibition of
H2O2-mediated
[Ca2+]i increase on the
H2O2-induced AA release, MC were preincubated with the
intracellular calcium chelator (BAPTA-AM) and then treated with
H2O2. BAPTA-AM significantly reduced the
H2O2-induced AA release from
MC+/+ in the presence of either
sPLA2 (Fig.
9B). These data are consistent with the importance of the
release of [Ca2+]i from
intracellular stores for H2O2-induced AA release in the
presence of either sPLA2.
|
Because BAPTA-AM has been shown to affect MAPK pathways in other cells, we
evaluated whether ERK1/2 phosphorylation was affected by BAPTA-AM treatment
(62–64).
In Fig. 9C, we
demonstrate that H2O2-induced ERK1/2 phosphorylation is
inhibited by the presence of the BAPTA-AM. Thus, BAPTA-AM significantly
reduces ERK1/2 phosphorylation and H2O2-induced AA
release in MC+/+. Therefore
[Ca2+]i may be important in two ways
for increased functional activity of cPLA2. It is necessary
for activation of ERK1/2 and hence cPLA2 phosphorylation and
facilitates cPLA2 translocation to the membrane.
Role of iPLA2 and p38 MAPK in H2O-induced AA Release—To evaluate a potential role for iPLA2,H2O2-induced AA release was examined after inhibition of iPLA2 with BEL (65). Fig. 10A shows that BEL significantly inhibited iPLA2 activity. MC+/+ and MC–/– were preincubated with BEL and then treated with H2O2. There was no difference in H2O2-induced AA release in MC in the presence or absence of iPLA2 inhibitor (Fig. 10C), or inhibition of the p38 signaling pathway with SB203580 (Fig. 10C). The phosphorylation of p38 kinase was significantly inhibited with SB203580 treatment (Fig. 10B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
), leading to the absence of two forms of
PLA2. Furthermore, reverse transcriptase-PCR experiments revealed
that murine MC–/– and
MC+/+ do not express group V PLA2
under basal or stimulated conditions. Previous studies have made extensive use
of transfection techniques to express different PLA2s in cultured
cells (44,
67). When transient
transfection is used, this approach leads to variable levels of expression
from cell to cell with nonphysiological expression levels in some cells. We
employed an adenoviral infection technique in these experiments. This
technique allows rather precise control of the level of protein expression and
better uniformity of expression among the cells. We have been able to
demonstrate clearly that lipase activity is directly related to the amount of
PLA2 enzyme expressed. We have also shown that the infection
process itself has no discernable effect upon endogenous PLA2
activities or cellular viability and does not alter the cellular
phenotype.
In this study, we have demonstrated: (i) H2O2-induced
AA release is increased with introduction of either group IIa or V
PLA2s into MC+/+, (ii) more AA
than OA is released in response to H2O2 when
sPLA2s and cPLA2 are both present in
MC+/+, and (iii) the effect of
sPLA2s on H2O2-induced AA release is
dependent upon the presence of cPLA2 and activation of PKC
and ERK1/2 in MC.
An important question is whether sPLA2s enhance the activity of
cPLA2, or whether cPLA2 enhances the
activity of sPLA2s when both enzymes are present in mesangial
cells. There are several possible ways in which these PLA2s can
interact. The first possibility is that sPLA2 and
cPLA2 work independently and both play a major role in AA
release. Our data show that this is not the case because infection with Ad-IIa
or Ad-V PLA2s had no effect on
MC–/– AA release. A second
possibility is that cPLA2 enhances sPLA2 activity
where sPLA2 is the major enzyme directly acting on phospholipid to
promote AA release. sPLA2s display very distinct heparanoid and
membrane binding properties. Group IIa sPLA2 is known to act poorly
against intact cellular membranes because of its preference for anionic
phospholipids and its weak binding capacity to the phosphatidylcholine.
Several other activation mechanisms have been proposed to account for
sPLA2 interactions with cells. Membrane pertubation by phospholipid
scramblase has been proposed to contribute to sensitization of cells toward
the action of sPLA2
(68). Other studies have
demonstrated that cPLA2 is required for group IIa or V
PLA2s to act properly
(26,
67,
69,
70). The
cPLA2 inhibitor methylarachidonyl fluorophosphonate
prevented both early and late phases of AA release, whereas the
sPLA2 inhibitor LY311727 only prevented the late phase release in
lipopolysaccharide-primed P388D1 cells
(34). Kuwata et al.
(71) also reported that
PGE2 production in 3Y1 rat fibroblast was completely dependent upon
the induction of group IIa PLA2, but could be prevented by
inhibiting cPLA2 with arachidonyl trifluoromethyl ketone. In
bone marrow mast cells, cPLA2 is necessary for both early
and late phase responses (72).
Furthermore, the production of 12/15 lipooxygenase metabolites, which is
downstream of cPLA2, has been proposed to play an important
role as a mechanism by which cPLA2 regulates
cytokine-induced group IIa PLA2 gene expression and facilitation of
group IIa PLA2-mediated membrane hydrolysis
(73). It has been reported
that endogenously expressed groups IIa and V PLA2 can liberate AA
from the cell membrane in a heparan sulfate proteoglycan-dependent manner in
agonist-stimulated cells (heparan sulfate proteoglycan shuttling pathway)
(74,
75). Group V and X
PLA2 are also known to release cellular AA independently of the
heparan sulfate proteoglycan shuttling pathway in unstimulated cells
(75,
76). Our data show that
neither Ad-IIa nor Ad-V PLA2 infection alone increase the basal
level of AA release in MC+/+ and
MC–/–, indicating that the
presence of cPLA2 does not enhance basal sPLA2
activity.
Another possibility is that sPLA2 enhances the activity of
cPLA2, which is the primary enzyme for AA release. When both
forms of PLA2s were present in
MC+/+, H2O2 had a much
greater effect on AA release than it did on OA release. This result strongly
suggests that sPLA2s are regulatory, enhancing the activity of
cPLA2 that is responsible for AA release in mesangial cells.
In addition, our data show that the effect of cPLA2 on
H2O2-induced AA release is dependent on the level of
adenoviral-mediated cPLA2 expression in
MC–/–, which do not express group
IIa or V PLA2s. This result indicates that cPLA2
alone can be a major enzyme for H2O2-induced AA release
in MC–/– without enhancement from
sPLA2.
Many signaling molecules have been implicated in AA release including
GTP-binding proteins, ERK, p38, ionic channels, phospholipase D, and
phopholipase C. H2O2-induced lipid peroxidation is
capable of increasing [Ca2+]i
through either physical disruption of ionic homeostasis through membrane
alteration, or by release of Ca2+ from intracellular
stores (59).
H2O2 has also been reported to increase the activity of
MAPK cascades. cPLA2 contains a calcium-lipid binding domain
and consensus phosphorylation sites for ERK1/2 and PKC
(5,
30,
77). An increase in
[Ca2+]i and activation of MAPKs will
stimulate cPLA2 activity and induce hydrolysis of membrane
phospholipids. H2O2 activates PKC in a number of cell
types (78). PKC activation has
been shown to be upstream of ERK1/2 activation
(57,
58). In addition
lysophosphatidylcholine and AA, products of the catalytic action of
sPLA2, have been shown to activate ERK1/2
(36). Hernandez et
al. (35) have
demonstrated that sPLA2 elicits biochemical signaling by
interaction with a membrane receptor in 132N1 astrocytoma cell line. This
interaction activates MAPKs and cPLA2, which subsequently
causes release of AA and mitogenesis. In addition, Xu et al.
(79) have shown the important
role of PKC and MAPK in regulating phosphorylation of cPLA2
and AA release in primary murine astrocytes. Our findings are consistent with
previously reported findings that cPLA2 is regulated by
ERK1/2 (29,
30). Our studies reveal that
H2O2 induces an increase in
[Ca2+]i and the PKC and ERK1/2
signaling pathways are critical for cellular sPLA2 to alter
H2O2-induced AA release in primary murine MC. When
ERK1/2 was constitutively activated by infection of
MC+/+ with Ad-MEK1-DD, there was no
difference in basal and H2O2-induced AA release between
non-infected and Ad-MEK1-DD-infected MC+/+.
Thus ERK1/2 activation alone is insufficient for full activation of
cPLA2 in the absence of group IIa and V PLA2s.
Our studies using the PKC inhibitor indicate that PKC activation is also
critical for H2O2-induced AA release. It is not clear
which isoform of PKC is involved in the cross-talk between sPLA2
and cPLA2. The PKC inhibitor GF 109203x is known to be more
selective for the isoform than for other PKC isoforms
(80,
81), suggesting the relevance
of this isoform. A number of years ago, we had shown that agents that activate
PKC could activate high molecular mass soluble PLA2 activity in
mesangial cells as long as there was an increase in
[Ca2+]i
(82,
83). Subsequent to the
identification of this activity as cPLA2, other
investigators have reported that PKC is involved in the regulation of
cPLA2 (29,
31,
79,
84–86).
Nemenoff et al. (29)
demonstrated direct phosphorylation in a cell-free system. It is not clear
whether PKC activity phosphorylates cPLA2 in our study. Our
data suggest that the effect of PKC on cPLA2 is mediated by
ERK1/2-dependent and independent pathways. In
Fig. 5C, we observe a
reduction in ERK1/2 and cPLA2 phosphorylation states at 60
and 180 min following H2O2 treatment in the presence of
GF 1092030x. However, because there is no effect of GF 1092030x at 30 min
after H2O2 treatment on ERK1/2 phosphorylation when the
phosphorylation of cPLA2 is decreased, the effect of PKC on
cPLA2 phosphorylation is independent of ERK1/2 at this time
point. The possibility of direct cPLA2 phosphorylation by
PKC continues to be an important subject for further investigation. It has
been reported that iPLA2
(87,
88) and p38
(89,
90) are involved in
stimulus-coupled AA release in some cell types. Our data indicate, however,
that both pathways are not involved in H2O2-induced AA
release in murine MC.
Fig. 11 shows a schematic
illustration of a cross-talk model involving cPLA2 and
sPLA2s. The treatment of cells with hydrogen peroxide leads to
activation of PKC and ERK1/2 and increases in
[Ca2+]i. These events will increase
cPLA2 activity acting in synergy. The
H2O2-activated cPLA2 will increase
intracellular free AA, which may be utilized by 12/15 lipooxygenase to
generate products that sensitize cell membranes to sPLA2s. Then
sPLA2s will act on the membrane via the heparan sulfate
proteoglycan glypican shuttling pathway. These events will enhance the
activation for sPLA2 of the PKC and ERK pathways which will, in
turn, further potentiate cPLA2 activation. The activated
cPLA2 will release AA.
|
In conclusion, the effect of group IIa and V PLA2s on
H2O2-induced AA release is dependent upon the presence
of cPLA2 and the activation of PKC and ERK1/2. The
activation of ERK1/2 is necessary but not sufficient to augment
H2O2-induced AA release. Our data are consistent with a
model whereby sPLA2 enhances the activity of
cPLA2, which then acts to release AA when both enzymes are
present in mesangial cells and supports the critical importance of
cPLA2 as the primary AA-releasing enzyme.
|
FOOTNOTES |
|---|
|| To whom correspondence should be addressed: MRB4, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115. Tel.: 617-732-6020; Fax: 617-582-6010; E-mail: joseph_bonventre{at}hms.harvard.edu.
1 The abbreviations used are: PLA2, phospholipase A2;
cPLA2, cytosolic phospholipase A2;
sPLA2, secretory phospholipase A2; ERK1/2, extracellular
signal-regulated kinases 1/2; MAPK, mitogen-activated protein kinase; MEK-1,
ERK kinase; PKC, protein kinase C; MC, mesangial cells; m.o.i., multiplicity
of infection; BAPTA,
1,2-bis(2-aminophenoyl)ethane-N,N,N',N'-tetraacetic
acid; [14C]PE,
1-palmitoyl-2-[14C]linoleoyl-sn-glycero-3-phosphoethanolamine;
OA, oleic acid; iPLA2, Ca2+-independent
phospholipase A2; GFP, green fluorescent protein;
[Ca2+]i, intracellular
Ca2+; AA, arachidonic acid; BEL, bromoenol lactone.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
,
p < 0.005 as compared with non-Ad-cPLA2
