Intracellular Calcium Signals Regulating Cytosolic
Phospholipase A2 Translocation to Internal
Membranes*,
John H.
Evans
,
Diane M.
Spencer,
Adam
Zweifach§¶, and
Christina C.
Leslie
**
From the Program in Cell Biology, Department of
Pediatrics, and ¶ Department of Immunology, National Jewish
Medical and Research Center, Denver, Colorado 80206 and the Departments
of § Physiology and Pathology, University of Colorado
School of Medicine, Denver, Colorado 80262
Received for publication, January 31, 2001, and in revised form, May 23, 2001
 |
ABSTRACT |
Increased intracellular Ca2+
concentrations ([Ca2+]i) promote cytosolic
phospholipase A2 (cPLA2) translocation to
intracellular membranes. The specific membranes to which
cPLA2 translocates and the [Ca2+]i
signals required were investigated. Plasmids of EGFP fused to
full-length cPLA2 (EGFP-FL) or to the cPLA2 C2
domain (EGFP-C2) were used in Ca2+/EGFP imaging experiments
of cells treated with [Ca2+]i-mobilizing
agonists. EGFP-FL and -C2 translocated to Golgi in response to
sustained [Ca2+]i greater than ~100-125
nM and to Golgi, ER, and perinuclear membranes (PNM) at
[Ca2+]i greater than ~210-280 nM.
In response to short duration [Ca2+]i transients,
EGFP-C2 translocated to Golgi, ER, and PNM, but EGFP-FL translocation
was restricted to Golgi. However, EGFP-FL translocated to Golgi, ER,
and PNM in response to long duration transients. In response to
declining [Ca2+]i, EGFP-C2 readily dissociated
from Golgi, but EGFP-FL dissociation was delayed. Agonist-induced
arachidonic acid release was proportional to the
[Ca2+]i and to the extent of cPLA2
translocation. In summary, we find that the differential translocation
of cPLA2 to Golgi or to ER and PNM is a function of
[Ca2+]i amplitude and duration. These results
suggest that the cPLA2 C2 domain regulates differential,
Ca2+-dependent membrane targeting and that the
catalytic domain regulates both the rate of translocation and enzyme residence.
| |
INTRODUCTION |
The Ca2+-sensitive cytosolic phospholipase
A2
(cPLA2)1 is found
in most cells and tissues and hydrolyzes phospholipids containing arachidonate at the sn-2 position to liberate arachidonic
acid (AA), an important regulator of diverse cell functions and a
precursor of potent inflammatory lipids (1). Structural studies of
cPLA2 have shown that it contains an N-terminal
calcium-dependent lipid binding (CaLB or C2) domain and a
C-terminal catalytic domain (2). The cPLA2 C2 domain is
structurally similar to other C2 domains (3-6), has a high affinity
for Ca2+ relative to C2 domains from other proteins (6),
and serves to localize cPLA2 to membrane for access to its
phospholipid substrate (4, 7). cPLA2 exhibits
receptor-mediated control, and a variety of Ca2+-mobilizing
agonists regulate cPLA2 activity by promoting its binding
to membrane through the C2 domain (8). The cPLA2 C2 domain
is required for membrane binding in vitro (9) and is necessary and sufficient in vivo for the translocation of
cPLA2 following an increase in
[Ca2+]i (10, 11).
Ca2+-dependent regulation of membrane binding
in vitro is dependent on the binding of two Ca2+
ions by Ca2+ binding loops (5, 6), and translocation
in vivo is abolished by mutation of Ca2+-binding
residues in the C2 domain (10, 11). These studies demonstrate that
translocation of cPLA2 is determined solely by the C2
domain in response to an increase in [Ca2+]i.
Although work on Ca2+-dependent binding of
cPLA2 to phospholipid vesicles (9) and natural membranes
(9, 12) has been carried out in vitro, there is little
information on Ca2+-dependent binding to
specific cellular membranes in intact cells.
The target of cPLA2 translocation following an increase in
[Ca2+]i is usually characterized as the nuclear
membrane based on the general distribution of indirectly labeled
cPLA2 (13-16) or GFP-tagged cPLA2 (10, 11, 17)
in the cell. The endoplasmic reticulum (ER) and the Golgi have also
been suggested as targets of cPLA2 translocation in
experiments using supraphysiological [Ca2+]i
changes brought about by high dose ionophore (10, 13, 16, 18). However,
there has been little evidence for assigning a specific organelle as
the target of Ca2+-induced translocation, and few studies
have attempted localization of cPLA2 with high spatial
resolution (16). Previous studies have determined, however, that
cPLA2 does not translocate to the plasma membrane.
A number of Ca2+-mobilizing agonists regulate
cPLA2 activity and AA release (8). For example, in MDCK
cells, P2Y2 (P2U) receptor activation and an inositol
1,4,5-trisphosphate-mediated intracellular [Ca2+]i release mediate
cPLA2-dependent AA release (19, 20). Other
studies have associated cPLA2-mediated AA release with an
agonist-induced, extracellular [Ca2+]i influx,
resulting in a sustained elevation in [Ca2+]i
(17, 21). In addition to a [Ca2+]i increase,
mitogen-activated protein kinase activity is involved in
cPLA2 regulation and is required for AA release in MDCK
cells (22). In other cell types, both an increase in [Ca2+]i and mitogen-activated protein
kinase-dependent phosphorylation is required for maximal
cPLA2 activation (11, 23). However, the
[Ca2+]i signals that regulate cPLA2
translocation or AA release have not been investigated rigorously in
live cells.
To determine the specific [Ca2+]i requirements
for translocation of cPLA2 and to verify the identity of
the membranes to which cPLA2 translocates, we used MDCK
cells expressing full-length cPLA2 or the cPLA2
C2 domain fused to an EGFP marker. Using immunocytochemical methods and
a live cell dye, we identified Golgi, ER, and a perinuclear membrane
(PNM) as targets of cPLA2 translocation. Simultaneous EGFP/Fura2 imaging of cells in response to sustained
[Ca2+]i of increasing amplitudes revealed that
full-length cPLA2 and the cPLA2 C2 domain
exhibited similar [Ca2+]i requirements for
translocation to Golgi, ER, and PNM but that the
[Ca2+]i threshold for translocation to the Golgi
was lower than it was to the ER and PNM. Imaging experiments in
response to transient [Ca2+]i signals of
different duration showed that the cPLA2 C2 domain
translocated more rapidly to membrane than did full-length cPLA2. In response to [Ca2+]i
declines, full-length cPLA2 remained associated with membrane targets longer than the cPLA2 C2 domain. AA
release was found to be proportional to the extent of translocation and
to the [Ca2+]i. These results demonstrate that
the Golgi, ER, and PNM are principle targets of cPLA2
translocation. They also reveal a differential membrane targeting
mechanism that is dependent on [Ca2+]i amplitude
and duration. These results suggest that the C2 domain is involved in
differential, Ca2+-dependent membrane targeting
and that the catalytic domain retards translocation and is required for
prolonged enzyme residence at the membrane.
| |
EXPERIMENTAL PROCEDURES |
EGFP-cPLA2 Fusion Constructions--
DNA encoding
human cPLA2 was cloned into the vector pCR2.1 as previously
described (11). The pCR2.1 was cut with KpnI and BamHI, and the fragment containing the cPLA2
cDNA was cloned into pGEM4Z (Promega). Using the polylinker
SacI and PstI sites, the cPLA2-containing fragment was cloned into pEGFP-C3
(CLONTECH) to create pEGFP-cPLA2
full-length (pEGFP-FL). For construction of pEGFP-C2, the C2 region of
cPLA2 (encoding Ser17 to Met148)
was amplified by polymerase chain reaction from pEGFP-FL using the
primers 5'-CTCAAGCTTTCCCACAAGTTTACGGTA-3' (which contains a
HindIII restriction site) and
5'-GATCCCGGGCATACTAAATCGTAGGTC-3' (which contains a SmaI
restriction site). The PCR product was cut with HindIII and
SmaI and cloned into pEGFP-C3 using the HindIII and SmaI sites. Empty vector (pEGFP-C3) was used for
control. The constructions were confirmed by sequencing.
Cell Culture--
MDCK cells obtained from ATCC were cultured in
Dulbecco's modified Eagle's medium containing 10% fetal
bovine serum, 100 units/ml penicillin, 100 µg/ml streptomycin, 0.292 mg/ml glutamine (growth medium) in 5% CO2 at 37 °C.
Subconfluent cells (1 × 104 cells/cm2)
were transfected with 1 µg of the relevant plasmid using Fugene-6 (Roche Molecular Biochemicals) in Dulbecco's modified Eagle's medium
containing 0.2% bovine serum albumin, 100 units/ml penicillin, 100 µg/ml streptomycin, 0.292 mg/ml glutamine (labeling medium) following
the manufacturer's protocol. For studies using transient transfection,
MDCK cells were grown on a glass-bottomed 35-mm tissue culture plate
(MatTek, Ashland, MA), transfected, and used within 24 h. Stable
lines of transfected cells were generated by growing transfected cells
in growth medium for 3 days, supplementing the growth medium with 5 mg/ml Geneticin (antibiotic G418-sulfate), and culturing for an
additional 2 weeks in Geneticin. Cells expressing EGFP fluorescence
were selected using a fluorescence-activated cell sorter. The
EGFP-positive cells were maintained in growth medium supplemented with
5 mg/ml Geneticin. For imaging studies, stably transfected cells were
plated on MatTek dishes at 1 × 104
cells/cm2 in growth medium with G418, incubated overnight,
changed into labeling medium (without G418) to quiesce the cells,
incubated overnight, and used the next day.
Immunoblotting--
Stable transfectants were grown on 100-mm
dishes at 1 × 104 cells/cm2 in growth
medium. Cells were scraped into ice-cold lysis buffer: 50 mM HEPES, pH 7.4, 150 mM sodium chloride, 1.5 mM magnesium chloride, 10% glycerol, 1% Triton X-100, 1 mM EGTA, 200 µM sodium vanadate, 10 mM tetrasodium pyrophosphate, 100 mM sodium
fluoride, 10 µg/ml leupeptin, and 10 µg/ml aprotinin. Lysates were
centrifuged at 15,000 × g for 15 min, and protein
concentration of the supernatant was determined by the bicinchoninic
acid method. Laemmli electrophoresis sample buffer (5×) was added to
the lysates, and SDS-polyacrylamide gel electrophoresis and
immunoblotting were performed using 35 µg of lysate protein (24).
Immunocytochemistry--
Transiently transfected cells on MatTek
plates were washed with Hanks' balanced salt solution containing 25 mM HEPES, pH 7.4 (HHBSS) and stimulated with either 100 µM ATP or 10 µM ionomycin (IONO). Cells
were fixed 45 s after stimulation in PBS containing 3.2%
paraformaldehyde and 3% sucrose. Cells were then permeabilized with
0.1% Triton X-100 in PBS for 15 min, washed in PBS, and blocked with
10% fetal bovine serum in PBS for 30 min. Cells were then labeled with
either a 1:100 dilution of anti-golgin 97 antibody (Molecular Probes,
Inc., Eugene, OR) or with a 1:100 dilution of anti-nucleoporin 97 antibody (Transduction Laboratories) in PBS containing 10% fetal
bovine serum for 60 min. After washing with PBS containing 10%
fetal bovine serum, cells were stained with a TRITC-conjugated
secondary antibody (Jackson ImmunoResearch). Cells were visualized
using a Nikon diaphot inverted microscope with a 60×, 1.4 NA oil
immersion lens and a Photometrics CCD camera using FITC filters for the
EGFP fluorescence and TRITC filters for the antibody fluorescence.
Images were acquired with IP Labs software (Scanalytics, Inc.) and
processed with Adobe Photoshop.
Dual Imaging Digital Microscopy of EGFP Translocation and
[Ca2+]i Changes--
Stably transfected
cells grown on MatTek plates were washed with HHBSS containing 1 mM probenecid and incubated with 10 µM Fura2-AM (Calbiochem) in HHBSS, 1 mM probenecid, and 1%
Me2SO for 1.5 h at 37 °C. Cells were then washed
with HHBSS containing 1 mM probenecid and imaged after a
30-min incubation for de-esterification of the Fura2-AM. Single-cell
imaging was performed on a Nikon inverted microscope using a 40×, 1.3 NA oil immersion objective, Fura2 and FITC filter sets (Chroma), and a
CCD camera (Cooke). Image acquisition and analysis were performed with
SlideBook software (Intelligent Imaging Innovations). For calcium
clamping experiments, images of Fura2 and EGFP fluorescence at 340 and
380 nm and 480 nm, respectively, were taken at each
extracellular [Ca2+]. For time lapse imaging, Fura2/EGFP
image sets were taken at 4-11-s intervals. The
[Ca2+]i was determined using the equation,
[Ca2+]i = KD × 
× (R 
Rmin)/(Rmax R) (25), where R was determined from the average
background-corrected pixel values of the Fura2 fluorescence from the
cytoplasmic area of individual cells from 340- and 380-nm image pairs;
Rmin, Rmax, and were
determined by in situ calibration at the conclusion of each
experiment; and a value of 224 nM was used for the
KD of Fura2. EGFP fluorescence was determined for
different regions of interest by calculating the
Ft/F0, where
Ft is the background- and bleach-corrected EGFP
fluorescence at time t, and F0 is the
background-corrected EGFP fluorescence at time 0. Profile plots of EGFP
fluorescence were performed using NIH Image software (developed by the
National Institutes of Health and available on the Internet at
rsb.info.nih.gov/nih-image). Time lapse EGFP imaging not involving
[Ca2+]i measurements (see Figs. 2, 3,
J-L, and 4, D-G and K-N) were
performed using an inverted Nikon microscope using a 60×, 1.4 NA oil
immersion objective, a Photometrics CCD camera, FITC filters (Chroma),
and IP Labs software.
Measurement of AA Release--
The protocol for determining AA
release is essentially as described (24). Briefly, MDCK cells were
freshly plated in 24-well plates at 2 × 104
cells/well (1 × 104 cells/cm2) and
incubated in growth medium overnight. Cells were then washed with
labeling medium and labeled with 0.2 µCi of
[3H]arachidonic acid/well in labeling medium overnight.
Cells were stimulated with the agonist of choice, and the medium was
collected at appropriate time points. The medium was centrifuged at
500 × g for 5 min, and the amount of radioactivity in
the supernatant was determined by scintillation counting. Cells were
scraped in 0.5 ml of 0.1% Triton X-100 for determining the total
cellular radioactivity.
| |
RESULTS |
Cytosolic PLA2 Translocates to the Golgi, Endoplasmic
Reticulum, and a Perinuclear Membrane--
Previous immunocytochemical
and GFP fusion protein studies have consistently described the pattern
of cPLA2 translocation in response to a saturating
[Ca2+]i as perinuclear (10, 11, 13-17), but the
specific membranes in the perinuclear region targeted by
cPLA2 have not been definitively identified. To identify
the membrane domains to which cPLA2 translocates in
response to an increase in [Ca2+]i,
EGFP-cPLA2 fusion proteins were constructed (Fig. 1A) and expressed in MDCK
cells. Previous studies have shown that EGFP fused to the amino
terminus of cPLA2 does not affect its function (11), and
cPLA2 tagged at the N (10) or C terminus (17) with GFP
gives similar translocation patterns in stimulated cells. As shown in
Fig. 1B, the EGFP fusion of full-length cPLA2 (EGFP-FL) was expressed in MDCK cells at approximately the same level
as the endogenous cPLA2. The EGFP fusion containing the cPLA2 C2 domain (EGFP-C2, amino acids 17-148) was also
expressed in MDCK cells at the same level as endogenous
cPLA2. Structural studies and a previous GFP fusion protein
study showed that amino acids 1-16 do not contribute to the C2 domain
structure (2, 4, 6) and are not required for translocation of the C2
domain (10). Expression of the EGFP vector alone did not affect the expression of endogenous cPLA2. Rapid translocation of the
EGFP-FL fusion protein in response to saturating
[Ca2+]i was elicited by bath application of the
Ca2+ ionophore IONO (10 µM) and observed in
live cells by time-lapse fluorescence microscopy (Fig. 1C).
EGFP-FL distribution was mostly cytoplasmic in resting cells but not
totally homogeneous, since a dark reticulated structure was apparent in
the cytoplasm (Fig. 1C). EGFP-FL fusion protein translocated
from the cytoplasm to three distinct intracellular regions in response
to an increase in [Ca2+]i: a juxtanuclear region,
a perinuclear region, and a reticulated region in the cytoplasm. A
similar pattern of translocation was observed with the EGFP-C2 fusion
protein, but no translocation of the empty EGFP vector was observed
after stimulation with IONO (not shown). The fluorescence patterns of
the translocated EGFP-FL and -C2 fusions are similar to what has been
observed in previous studies using high dose ionophore to elicit a
supraphysiological [Ca2+]i increase and
translocation (10, 11, 17), and these results confirm the observation
that the C2 domain is necessary and sufficient for
Ca2+-mediated translocation and intracellular targeting of
cPLA2.
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 1.
EGFP-cPLA2 fusion protein
constructs and their expression in MDCK cells. See the Fig. 1
movie at http://www.jbc.org. A, the crystal structure of
cPLA2 has shown that the protein consists of two domains:
an N-terminal C2 domain and a C-terminal catalytic domain. For
fluorescence microscopy translocation studies in MDCK cells, the
full-length (EGFP-FL) or C2 domain (EGFP-C2) of human cPLA2
was fused to the C terminus of EGFP. Empty vector (EGFP) was used for
control. B, a representative immunoblot of protein lysates
from stably transfected MDCK cells. The EGFP-FL fusion protein runs at
an apparent molecular mass of ~125 kDa, and the EGFP-C2 fusion
protein runs at ~50 kDa. C, time lapse images of EGFP-FL
translocation in response to saturating [Ca2+]i
at times after the addition of 10 µM IONO.
|
|
To confirm the identity of the intracellular membranes targeted by
cPLA2, we used immunocytochemical methods and a live cell organelle-specific dye. Cells expressing EGFP-FL and -C2 were stimulated with the P2Y2 agonist ATP or IONO to elicit an increase in
the [Ca2+]i, fixed, and stained with antibodies
for a cis-Golgi marker (golgin 97) or a nuclear membrane
marker (nucleoporin 62). An increase in EGFP-FL fluorescence in a
juxtanuclear position 45 s after treatment with 100 µM ATP was observed in >90% of cells (Fig.
2A) and overlapped completely
with a region identified by the cis-Golgi marker (Fig. 2,
B and C). As a comparison with the physiological
agonist and with previously published cPLA2 localization studies, we also used high dose ionophore as an agonist. In response to
10 µM IONO, a similar fluorescence increase in a
juxtanuclear region was observed (Fig. 2D) that corresponded
to the location of the cis-Golgi marker (Fig. 2,
E and F). Similar results were found in
EGFP-C2-expressing cells treated in the same fashion (not shown). No
translocation was apparent in EGFP-FL- or EGFP-C2-expressing cells that
were not stimulated with ATP or IONO prior to fixation (not shown). In
response to 10 µM IONO, EGFP-FL and -C2 fluorescence was
observed also in a distinct ring (Fig. 2G), which was found to delimit the area of the nucleus (Fig. 2, H and
I). Although this is suggestive of cPLA2
translocation to the nuclear membrane, the spatial resolution of these
studies cannot rule out translocation to closely apposed ER. In studies
using the live cell dye ER Tracker, dark areas of EGFP-C2 fluorescence
in unstimulated cells (Fig. 2J, arrowheads) were
coincident with areas of ER Tracker fluorescence (Fig. 2K).
After stimulation with 1 µM IONO (Fig. 2L),
the pattern of EGFP-C2 fluorescence was completely coincident with the
ER Tracker signal (Fig. 2, K and L).
Interestingly, ER Tracker also produced a perinuclear ring that could
indicate close apposition of the ER and nuclear membranes or staining
of the outer nuclear membrane with the dye. Because previous studies
differentiate between ER and nuclear membrane, we refer to the ring as
perinuclear membrane (PNM). The ER Tracker also stained juxtanuclear
membranes in the presumed area of the Golgi. This is to be expected,
because it has been shown that ER and Golgi membranes are in close
apposition (26). Similar results for ER Tracker experiments were
obtained in EGFP-FL-transfected cells. These studies suggest that
cPLA2 targets Golgi, ER, and PNM after translocation and
that targeting is a function of the C2 domain.
View larger version (48K):
[in this window]
[in a new window]
|
Fig. 2.
EGFP-FL translocation to Golgi, PNM, and ER
is elicited by ATP and IONO. EGFP-FL-transfected cells were
stimulated with 100 µM ATP (A-C) or 10 µM IONO in HHBSS (D-I). Cells were fixed at
45 s, permeabilized, blocked, and incubated with an antibody to
golgin 97, a cis-Golgi protein, or to nucleoporin 62, a
constituent of the nuclear membrane. After washing, cells were
incubated with a TRITC-labeled secondary antibody.
Images were made with FITC optics for visualizing EGFP
fluorescence and TRITC optics for the secondary antibody. In response
to ATP, an increase in EGFP-FL fluorescence was clear at a juxtanuclear
location (A). Indirect labeling of golgin 97 exhibited a
similar juxtanuclear fluorescence (B), which was entirely
coincident with the EGFP-FL signal (C). In response to IONO,
a similar fluorescence pattern was observed for the EGFP signal
(D) and the golgin 97 signal (E), and the two
signals were found to be coincident (F). In response to IONO
stimulation, there was an increase in EGFP-FL fluorescence
(G) in a ring around the nucleus (H and
I). Similar results were found in EGFP-C2-transfected cells.
For ER staining, EGFP-C2-transfected cells were transferred to HHBSS
and incubated with ER Tracker. ER Tracker was imaged with
4,6-diamidino-2-phenylindole optics. Cells were stimulated with
1 µM IONO, and time lapse images were recorded using FITC
optics. In unstimulated cells, dark areas in the EGFP-C2 fluorescence
(J, arrowheads) coincided with the distribution
of ER tracker fluorescence (K, arrowheads). After
the IONO addition, EGFP-C2 fluorescence produced a reticulated pattern
in the cytoplasm (L, arrowheads) that was
completely coincident with ER Tracker fluorescence (K) in
the same cell. ER Tracker fluorescence also localized as a ring around
the nucleus and in the area of the Golgi (K). Similar
results with ER Tracker were observed in EGFP-FL-transfected
cells.
|
|
Co-localization of cPLA2 with a cis-Golgi Marker
following Disruption of the Golgi--
To study the association of
cPLA2 with Golgi in more detail, we investigated the effect
of agents that disrupt Golgi structure on cPLA2
translocation. Cells expressing EGFP-C2 were first incubated in
brefeldin A (BFA) for 3 h and then stimulated with ATP, fixed, and
stained with the cis-Golgi marker antibody. BFA disrupts
Golgi structure by promoting the retrograde trafficking of Golgi
components to the ER (27-29). Many Golgi matrix proteins, however,
remain with Golgi remnants following BFA treatment (30). In
immunocytochemical studies, BFA treatment disrupted normal juxtanuclear
EGFP-C2 fluorescence following stimulation and gave instead a punctate
distribution (Fig. 3A), which
was similar to the distribution of the Golgi marker (Fig. 3,
B and C). This distribution is similar to the distribution of Golgi proteins such as 
-mannosidase II and
-1,4-galactosyltransferase (27-29), but unlike the Golgi matrix
protein GM130 (30), following BFA treatment. In time lapse imaging
studies, unstimulated cells expressing EGFP-C2 and treated with BFA
exhibited a diffuse cytoplasmic distribution of EGFP fluorescence (Fig.
3D) but exhibited a punctate EGFP fluorescence pattern
following stimulation with ATP (Fig. 3E). Following wash-out
of BFA and readdition of ATP, the cell regained a normal juxtanuclear
pattern of EGFP fluorescence consistent with reformation of the Golgi
(Fig. 3, F and G). Similar experiments were
carried out using the microtubule inhibitor nocodazole (NOC). NOC
disrupts Golgi structure by stimulating retrograde transport of Golgi
components to the ER and then reformation of dispersed Golgi
"ministacks" (31, 32). In EGFP-C2-expressing cells, NOC caused a
punctate pattern of EGFP fluorescence following ATP stimulation (Fig.
3H) and a similar pattern of staining as the cis-Golgi marker (Fig. 3I), which overlapped with
the EGFP fluorescence (Fig. 3J). Again, the NOC results are
similar to the pattern of Golgi component distribution seen in
previously published studies (31, 32). In time lapse studies, NOC had
no effect on the distribution of fluorescence in resting cells (Fig.
3K) but caused a dispersed pattern of fluorescence in
ATP-stimulated cells consistent with translocation to Golgi
ministacks (Fig. 3L) that was reversed by wash-out of
the drug (Fig. 3, M and N). The findings using BFA and NOC suggest that a component of the Golgi membrane may be
involved in the Ca2+-mediated targeting of
cPLA2.
View larger version (64K):
[in this window]
[in a new window]
|
Fig. 3.
Disruption of Golgi structure with BFA
or NOC. See the Fig. 3 movies at http://www.jbc.org.
EGFP-C2-transfected cells were incubated with 10 µg/ml BFA
(A-C) or 20 µM NOC
(H-J) for 3 h at 37 °C in HHBSS. Cells
were stimulated with 100 µM ATP and stained with
anti-golgin 97 antibody as described in Fig. 2. ATP stimulation of
BFA-treated cells resulted in a punctate distribution of EGFP-C2
fluorescence (A) that was similar to the pattern of golgin
97 distribution (B), resulting in a considerable overlap of
the two fluorescence signals (C). In time lapse imaging of
cells incubated with 10 µg/ml BFA for 3 h and stimulated with
100 µM ATP (D and E), EGFP-C2
fluorescence exhibited a punctate distribution, which was reversed by a
2-h wash-out of the drug and readdition of ATP (F and
G). ATP stimulation of NOC-treated cells resulted in a
punctate distribution of EGFP-C2 fluorescence consistent with the
ministack formation (H) that was similar to the pattern of
golgin 97 distribution (I), resulting in a considerable
overlap of the two fluorescence signals (J). In time lapse
imaging of cells incubated with 20 µM NOC for 3 h
and stimulated with 100 µM ATP (K and
L), EGFP-C2 fluorescence exhibited a punctate distribution,
which was reversed by a 1-h wash-out of the drug and readdition of ATP
(M and N).
|
|
Targeting of cPLA2 to Different Membrane Domains Is
Ca2+-dependent and Is a Function of the C2
Domain--
Previous immunofluorescence and GFP-cPLA2
fusion protein translocation studies have demonstrated translocation of
cPLA2 in response to increases in
[Ca2+]i, usually in response to a saturating
[Ca2+]i brought about by the Ca2+
ionophore A23187 or IONO (10, 13-17, 33). In another study, platelet-activating factor was used to increase the
[Ca2+]i in cells overexpressing the
platelet-activating factor receptor and expressing a
GFP-cPLA2 construct, but no simultaneous quantification of
fluorescence change and [Ca2+]i was made to
correlate specific [Ca2+]i with cPLA2
translocation (17). To precisely determine the
[Ca2+]i required for translocation to the
different organelles, the [Ca2+]i in cells was
quantified using the Ca2+-specific dye Fura2 simultaneously
with the EGFP fluorescence at the different organelle membranes.
Initially, calcium clamping experiments were carried out, whereby
transfected cells were incubated in an EGTA/calcium buffer to deplete
the internal Ca2+ stores and then treated with 10 µM IONO to make the cell membranes permeable to
Ca2+. Because it was not possible to clamp the
[Ca2+]i to predetermined values by manipulating
the extracellular [Ca2+] ([Ca2+]e),
the [Ca2+]e was increased in discreet steps by
adding CaCl2 to the bath solution, and the
[Ca2+]i was determined empirically using
ratiometric measurements of Fura2 fluorescence. Once allowed to
equilibrate with a specific [Ca2+]e, the
[Ca2+]i remained constant or "clamped" for
several minutes and allowed for the quantification of cPLA2
translocation at "steady state" [Ca2+]i
conditions. The EGFP fluorescence signal at regions of interest
corresponding to the ER, PNM, and Golgi were monitored and quantified
at each step increase in the [Ca2+]i. Images of
EGFP-FL fluorescence from a representative calcium clamping experiment
are shown in Fig. 4. As shown in Fig. 4A, plot profiles were made along vectors traversing
representative areas of the Golgi, ER, and PNM at each
[Ca2+]i in the same cell. Plot profiles represent
the EGFP fluorescence intensity from a 2-pixel-wide area along the
length of the vector. In cells clamped at a
[Ca2+]i of 15 nM (Fig.
4B), the Golgi profile (Fig. 4G, left traces) shows a relatively flat fluorescence. Also visible
in the cytoplasm of the cell was the location of the ER, visualized by
the lower fluorescence intensity relative to the cytoplasm (a
"negative image") from the exclusion of EGFP from the area of the
ER (Fig. 5, A and
B). This area of lower fluorescence is evident in the ER
profile (Fig. 4G, center traces) and
has been found to be completely coincident with the fluorescence signal of ER Tracker (Fig. 2L). Because the EGFP fluorescence was
greater in the cytoplasm than the nucleoplasm at 15 nM
(Fig. 4, A and B), the intensity of the PNM
profile is seen to decrease as it crosses the nuclear membrane (Fig. 4,
A and G, right plots). At a
[Ca2+]i of 231 nM (Fig.
4C), a fluorescence increase was observed in the area of the
Golgi (Fig. 4, C and G), and the fluorescence increased further at 377 nM (Fig. 4, D and
G). However, at [Ca2+]i of 231 and 377 nM, there was no fluorescence increase at the ER, relative
to the cytoplasm, or at the PNM (Fig. 4G), and the
"negative" image of the ER was still visible in the cytoplasm (Fig.
4C). At 540 and 592 nM
[Ca2+]i, the fluorescence intensities at the
Golgi, ER, and PNM increased (Fig. 4G), and the reticulated
fluorescence indicative of translocation to ER and a distinct
perinuclear ring was observed (Fig. 4, E and F).
This result suggests that cPLA2 translocates to Golgi
membranes at lower [Ca2+]i values than those
required for translocation to the ER and PNM.
View larger version (69K):
[in this window]
[in a new window]
|
Fig. 4.
Translocation of EGFP-FL at sustained
[Ca2+]i. Stably transfected cells expressing
EGFP-FL, loaded with Fura2, were subjected to step increases in the
[Ca2+]e of an EGTA-buffered bath solution that
resulted in step increases in the [Ca2+]i of the
cells (calcium clamp conditions). Fura2 ratio pair and EGFP images were
taken at each [Ca2+]i step, and the
[Ca2+]i was determined. B-F, EGFP-FL
fluorescence images at the indicated [Ca2+]i
values are shown. G, plot profiles along vectors (shown in
A) of the EGFP intensity across regions of ER, Golgi, and
the PNM were produced at each [Ca2+]i step
(B-F). Intensity measurements across ER also included
surrounding regions of cytoplasm (cyt). Intensity
measurements across the PNM also included surrounding regions of
cytoplasm (cyt) and nucleoplasm (nuc). Data are
representative of 16 cells from four experiments.
|
|
View larger version (68K):
[in this window]
[in a new window]
|
Fig. 5.
Correlation of EGFP-C2 translocation with
[Ca2+]i in response to an ATP-induced
[Ca2+]i transient. See the Fig. 5 movie at
http://www.jbc.org. Stably transfected cells expressing EGFP-C2 were
loaded with Fura2 and stimulated with 100 µM ATP by bath
application. Time lapse images of EGFP and Fura2 fluorescence were
recorded; the relative EGFP fluorescence at regions of interest
corresponding to the Golgi, PNM, and ER were measured; and
[Ca2+]i was determined for each image set.
A, EGFP-C2 fluorescence images at times after ATP
application. B, graph of [Ca2+]i and
EGFP fluorescence over time for the cell shown in A. Results
are representative of 28 cells from seven independent
experiments.
|
|
Analysis of 49 cells from additional calcium clamping experiments
revealed that a minimum [Ca2+]i of between 100 and 125 nM is required for EGFP-FL and -C2 translocation to
the Golgi (Table I). As the
[Ca2+]i was increased above 125 nM,
the EGFP fluorescence at the Golgi continued to increase. A minimum
[Ca2+]i of between 210 and 280 nM was
required for EGFP-FL and -C2 translocation to the ER and PNM (Table I).
At [Ca2+]i above 280 nM, EGFP
fluorescence increased at the Golgi, ER, and PNM. It is important to
note that at ~280 nM, EGFP fluorescence at ER and PNM is
just measurable, while the EGFP fluorescence at Golgi is intense.
Interestingly, once the [Ca2+]i threshold for
translocation to Golgi or to the ER and PNM was exceeded, no further
increase in fluorescence at the membranes was observed without an
increase in the [Ca2+]i, suggesting that the
fraction of cPLA2 translocated to membrane is proportional
to the [Ca2+]i. Together, these results suggest
that membrane targeting of cPLA2 is a function of
[Ca2+]i amplitude and that the
Ca2+-dependent, differential targeting of
internal membranes is a property of the cPLA2 C2
domain.
View this table:
[in this window]
[in a new window]
|
Table I
[Ca2+]i requirements for EGFP-FL and -C2
translocation to intracellular membranes
EGFP-FL and -C2-transfected cells, loaded with Fura2 and treated with
10 µM IONO in a calcium/EGTA buffer ("calcium clamp"
conditions), were subjected to dual GFP/Fura2 imaging (see "Materials
and Methods"). Intracellular [Ca2+]i was increased
in a stepwise fashion by adjusting the [Ca2+]e by
bath application of CaCl2. Translocation to Golgi, ER and PNM
and the [Ca2+]i was assessed at each
[Ca2+]e step for individual cells. The values below
represent the range of the five lowest [Ca2+]i values
where fluorescence was detected at the indicated membranes (EGFP-FL, 16 cells from four independent experiments; EGFP-C2, 33 cells from three
independent experiments).
|
|
Translocation of cPLA2 in Response to Transient
[Ca2+]i Changes--
Transient increases in
[Ca2+]i are induced in MDCK cells in response to
a variety of agonists, including ATP (34, 35), low dose IONO, and the
Ca2+-ATPase inhibitor thapsigargin (TG) (35). ATP and TG
induce release of Ca2+ from intracellular pools, while IONO
releases Ca2+ from internal pools and stimulates an
extracellular Ca2+ influx (35). We used these agonists to
evaluate the translocation characteristics of the
EGFP-cPLA2 fusion proteins from time lapse recordings of
Fura2 and EGFP fluorescence signals. Representative frames from a time
lapse recording of EGFP-C2 translocation in a single cell in response
to 100 µM ATP are shown in Fig. 5A. In the
resting cell, the area of the nuclear membrane and ER was clearly
visible as a "negative image." Following stimulation, an increase
in fluorescence at the Golgi, ER, and PNM was observed. The
fluorescence at the ER and PNM then declined, while the fluorescence continued to increase at the Golgi. The Golgi fluorescence then declined to resting levels at later time points. The fluorescence intensities at the Golgi, ER, and PNM and the
[Ca2+]i change in the cell are shown graphically
in Fig. 5B. The graph reveals that EGFP fluorescence
increases at the Golgi, ER, and PNM with an increase in the
[Ca2+]i but declines at the ER and PNM earlier
than at the Golgi. We hypothesize that the ATP-mediated
Ca2+ release from ER stores produces a high local
[Ca2+]i that mediates translocation to the ER and
PNM. As the [Ca2+]i declined between 30 to
60 s, the EGFP-C2 fluorescence decreased at the ER and PNM,
reflecting the higher [Ca2+]i required for ER and
PNM association (Table I) and concomitantly increased at the Golgi. At
90 s, the Golgi fluorescence declined with approximately the same
slope as the decline in [Ca2+]i, until the Golgi
fluorescence and [Ca2+]i returned to resting
values at 360 s.
In contrast to EGFP-C2, time lapse imaging of EGFP-FL fluorescence
change in response to an ATP-mediated [Ca2+]i
increase showed a transient increase in fluorescence at the Golgi but
no measurable increase at the ER or PNM. In the time lapse images shown
in Fig. 6A, the distribution
of EGFP-FL in the resting cell was similar to that of EGFP-C2, and the
"negative image" of the ER was clearly visible. After stimulation,
a fluorescence increase was observed at the Golgi by 20 s.
However, unlike EGFP-C2 (Fig. 5), no increase in fluorescence was
observed at the ER and PNM, although the [Ca2+]i
increase was of approximately the same magnitude and duration as the
[Ca2+]i increase of the EGFP-C2-transfected cell
(~600 nM for ~120 s, compare Figs. 5B and
6C), and the minimum [Ca2+]i required
for translocation to ER and PNM was exceeded (Table I). EGFP-FL
fluorescence at the Golgi remained elevated, while the
[Ca2+]i declined, with an elevated fluorescence
at the Golgi measurable at later times (Fig. 6, A and
C). However, when the same cells were treated with 1 µM IONO after a 30-min wash-out and recovery period,
EGFP-FL fluorescence was seen to move to the ER and PNM as well as to
the Golgi (Fig. 6B). The increase in EGFP-FL fluorescence at
the Golgi was slower with IONO than with ATP, probably due to the
difference between the rate of [Ca2+]i increase
(Fig. 6, C and D). The more extensive
translocation in response to IONO correlated with a much longer
[Ca2+]i transient (Fig. 6D) and was
similar to the pattern resembling that of EGFP-C2 translocation in
response to ATP (Fig. 5B). At later times after 1 µM IONO, the EGFP-FL fluorescence was diminished at the
ER but remained associated with the Golgi (Fig. 6, B and
D).
View larger version (81K):
[in this window]
[in a new window]
|
Fig. 6.
Correlation of EGFP-FL translocation with
[Ca2+]i in response to ATP- and IONO-induced
[Ca2+]i transient. See the Fig. 6 movies at
http://www.jbc.org. Stably transfected cells expressing EGFP-FL were
prepared as in Fig. 5. A, time lapse images of EGFP
fluorescence after the addition of 100 µM ATP are shown
at the indicated time points. C, quantitative analysis of
the EGFP translocation to Golgi and the [Ca2+]i
for the center cell is shown in A. B, following a
30-min wash-out of ATP, the same cells were treated with 1 µM IONO, and time lapse images of EGFP translocation at
the indicated time points are shown. D, analysis of EGFP
translocation and the [Ca2+]i is shown for the
cell in B. Results are representative of 19 cells from four
independent experiments.
|
|
Similar results to those found with ATP were found following
stimulation of EGFP-FL- or EGFP-C2-transfected cells with 2 µM TG (Fig. 7, A
and C, and Fig. 7, B and D,
respectively). The time lapse images in Fig. 7A show that
EGFP-FL translocation in response to TG resulted in translocation to
the Golgi, but little translocation was evident to ER or PNM.
Thapsigargin induced a [Ca2+]i increase and
translocation to Golgi that is similar to that elicited by ATP (compare
Fig. 6C with Fig. 7C). As with ATP stimulation
(Fig. 6C), EGFP-FL fluorescence remained associated with the
Golgi, while the [Ca2+]i declined (Fig. 7,
A and C).
View larger version (90K):
[in this window]
[in a new window]
|
Fig. 7.
EGFP-FL, but not EGFP-C2, remains associated
with Golgi following an IONO- or TG-stimulated transient
[Ca2+]i increase. See the Fig. 7 movies at
http://www.jbc.org. Stably transfected cells expressing EGFP-FL or -C2
were loaded with Fura2 and stimulated with 2 µM TG
(panels A and C and panels
B and D, respectively) or 1 µM IONO
(E and F, respectively) by bath application. Time
lapse images of EGFP and Fura2 fluorescence were recorded, the EGFP
fluorescence at the Golgi was measured, and
[Ca2+]i was determined for each image set.
Representative images from the time lapse recording of EGFP-FL and
EGFP-C2 translocation in response to 2 µM TG are shown in
A and B. Graphs of [Ca2+]i
and EGFP fluorescence over time corresponding to the image sets in
A and B are shown in C and
D. Graphs of [Ca2+]i and EGFP
fluorescence over time are shown for representative, single, EGFP-FL-
and EGFP-C2-transfected cells treated with 1 µM IONO
(EGFP-FL/1 µM IONO, 45 cells from 11 experiments;
EGFP-C2/1 µM IONO, 10 cells from one experiment;
EGFP-FL/TG, eight cells from one experiment; EGFP-C2/TG, 27 cells from
three experiments).
|
|
In contrast to results using EGFP-FL, time lapse images of
EGFP-C2-transfected cells in Fig. 7B show translocation to
Golgi, ER, and PNM in response to TG, although the
[Ca2+]i increase in the particular cell analyzed
was lower (Fig. 7D) than that observed in the
EGFP-FL-transfected cell (Fig. 7C). The difference between
the amplitudes of the [Ca2+]i transients is
independent of the construct expressed and highlights the heterogeneity
of the [Ca2+]i response usually found in cell
populations (34, 36). As with stimulation with ATP (Fig. 5), EGFP-C2
fluorescence at the Golgi decreased at the same rate as the decline in
[Ca2+]i (Fig. 7D).
In contrast to stimulation with ATP or TG, stimulation with 1 µM IONO resulted in [Ca2+]i
transients of much longer duration (8-9 min with IONO versus 1-2 min with ATP or TG; Fig. 7, E and
F). Also in contrast to ATP or TG stimulation, EGFP-FL
fluorescence was observed at the ER and PNM (similar to that shown in
Figs. 1C and 6B) in addition to the Golgi in
response to 1 µM IONO. Although the IONO-induced [Ca2+]i increase was approximately the same
magnitude as the ATP- or TG-induced increase, where no ER or PNM
fluorescence was observed (Figs. 6A and 7C), the
duration was much longer (~540 s for IONO versus 120-240
s for ATP and TG). Therefore, EGFP-FL can translocate to ER and PNM as
well as Golgi but only if the [Ca2+]i transient
is of long duration, as was shown in Fig. 6. Consistent with the
results from ATP or TG stimulation, EGFP-FL fluorescence in response to
1 µM IONO remained associated with the Golgi, although
the [Ca2+]i returned to resting level (Fig.
7E).
Similar to the results from ATP- or TG-stimulated cells, EGFP-C2
translocation in response to 1 µM IONO was observed at
the Golgi, ER, and PNM, and the decrease in fluorescence at the Golgi declined at the same rate as the decrease in the
[Ca2+]i (Fig. 7F).
These results suggest that the association of the EGFP-FL with target
membranes occurs more slowly than association of the EGFP-C2, even when
the minimum [Ca2+]i requirement for translocation
is satisfied. They also show that dissociation from membranes is much
slower for the EGFP-FL than for the EGFP-C2 in response to a decrease
in [Ca2+]i.
Translocation of cPLA2 Correlates with Amount of AA
Release--
Cytosolic PLA2 is responsible for
receptor-mediated AA release from a variety of cell types (8). Several
studies suggest, however, that a transient
[Ca2+]i release is insufficient for AA release
and that a sustained [Ca2+]i is required (17,
21). A latency period between the translocation of cPLA2 to
membranes and enzymatic activity has also been suggested (17). In light
of our findings of a rapid translocation of EGFP-FL to Golgi in
response to agonist, release of AA was assayed in MDCK cells in
conditions reflecting our imaging studies. To replicate the conditions
that produced transient elevations in [Ca2+]i, we
assayed AA release over time from cells stimulated with 1 µM IONO or 100 µM ATP (Fig.
8A). AA release in response to
a saturating [Ca2+]i elicited by 10 µM IONO was also assayed. AA release from cells in these
conditions was rapid, with significant AA release measured at 30 and
60 s. Time courses of AA release in response to 1 µM
IONO or ATP were similar and reached a plateau by 3-5 min.
Interestingly, the slope of AA release from 0 to 30 s in response
to ATP stimulation was consistently greater than that for 1 µM IONO (Fig. 8B) and may be related to the
greater rate of [Ca2+]i increase and faster
translocation of cPLA2 in response to ATP (Fig. 7). These
results suggest that there is little latency between cPLA2
translocation to membrane and subsequent enzymatic activity. AA release
in response to 10 µM IONO was much greater than in
response to 1 µM IONO or ATP, possibly reflecting the massive cPLA2 translocation stimulated by saturating
[Ca2+]i conditions brought about by 10 µM IONO (Fig. 1C). A similar difference in the
magnitude of AA release between natural agonists and high dose
ionophore has also been observed in RBL cells (16).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 8.
AA release in response to transient or
sustained [Ca2+]i increase. MDCK cells were
labeled with [3H]AA overnight, washed with HHBSS
containing 0.05% bovine serum albumin, and stimulated with the agonist
of choice in HHBSS containing 0.2% bovine serum albumin. Samples were
collected at the indicated time points. Results are expressed as
percentage of total cellular AA release. A, MDCK cells were
stimulated with 100 µM ATP, 1 µM IONO, or
10 µM IONO. AA release in response to ATP or 1 µM IONO was of similar magnitude but was ~4-fold
greater in response to 10 µM IONO. B, a larger
view of the AA release from 0 to 60 s for ATP and 1 µM IONO. Results shown are representative of three
independent experiments. C, to determine AA release in
calcium clamping conditions, cells were preincubated for 30 min in
HHBSS containing 3 mM EGTA and then stimulated with one of
three EGTA/calcium buffers containing 10 µM IONO. The
EGTA/calcium buffers used typically resulted in
[Ca2+]i of 1) less than 15 nM (low
[Ca2+]i), 2) between 110 and 190 nM
(intermediate [Ca2+]i), or 3) greater than 700 nM (high [Ca2+]i) in the calcium
clamping experiments (see Fig. 4). Control cells were incubated in high
[Ca2+]i buffer without IONO. MDCK cells exhibited
little AA release in response to incubation in low
[Ca2+]i buffer or when IONO was excluded. At
intermediate [Ca2+]i, AA release was 6% at
180 s. At high [Ca2+]i, AA release was
greater than 9% at 180 s. Results shown are representative of
three independent experiments.
|
|
To assay AA release in conditions replicating the calcium clamping
experiments (Fig. 4), cells were first incubated for 30 min in 3 mM EGTA/HHBSS to deplete internal Ca2+ stores.
Cells were then stimulated with 10 µM IONO in one of three buffered EGTA/Ca2+ solutions, which typically
resulted in 1) a [Ca2+]i of less than 15 nM (low [Ca2+]i), 2) a
[Ca2+]i between 110 and 190 nM
(intermediate [Ca2+]i), and 3) a
[Ca2+]i greater than 700 nM (high
[Ca2+]i) in our calcium clamping experiments
(Table I). We would expect that in low [Ca2+]i no
translocation of endogenous cPLA2 would occur; at
intermediate [Ca2+]i, translocation would be
mainly to the Golgi; and in high [Ca2+]i
conditions, translocation would be to Golgi, ER, and PNM. Cells were
also subjected to the high [Ca2+]i, but without
the addition of IONO for control. As shown in Fig. 8C, there
was little AA release in cells stimulated at low
[Ca2+]i with IONO or in cells in high
[Ca2+]i without IONO, and AA release was
similar to the basal release of unstimulated cells (Fig.
8A). Cells stimulated in intermediate [Ca2+]i levels gave a rapid, large AA release
that reached a plateau by 180 s. AA release in response to high
[Ca2+]i was greater but also slowed between 180 and 300 s. These results demonstrate that AA release in MDCK cells
is dependent on and proportional to the [Ca2+]i.
Interestingly, in all conditions (Fig. 8, A and
C), AA release reached a plateau between 3 and 5 min even
under conditions where cPLA2 remains associated with
membrane. This may be due to inactivation of the enzyme (37, 38).
| |
DISCUSSION |
We examined the translocation of EGFP-tagged cPLA2
fusion proteins in MDCK cells in response to sustained and transient
Ca2+ signals elicited by ATP, IONO, or TG. Previous studies
have generally described the pattern of cPLA2 translocation
as perinuclear (11, 15, 17) or consistent with translocation to the
nuclear membrane or envelope (10, 13, 14, 16). Fewer studies have
suggested that cPLA2 translocation was consistent with the
ER (10, 13, 16) or Golgi (10, 18). A study on RBL cells, however, found that cPLA2 preferentially targeted the nuclear envelope and
failed to describe any targeting of the Golgi (16). In this report, using immunofluorescence methods, we established that full-length cPLA2 and the cPLA2 C2 domain translocate to
the Golgi and to a perinuclear membrane. Using the fluorescent live
cell dye ER Tracker, we concluded that cPLA2 also
translocates to ER. Because the co-localization was identical for the
full-length cPLA2 and the C2 domain, we conclude that
membrane targeting is determined by the C2 domain, as previously
suggested (10, 11).
By disrupting Golgi structure in cells by incubation with BFA or NOC,
we were able to provide more evidence that a prominent target of
translocation is the Golgi. BFA promotes retrograde trafficking of
Golgi membrane components to the ER (27-29). Recent reports have
demonstrated that some Golgi proteins, such as GM 130, are present in
Golgi remnants after BFA treatment, but others, such as the Golgi
markers -mannosidase II and -1,4-galactosyltransferase, are
dispersed (30). In our experiments, the C2 domain translocation and the
Golgi marker co-localized and displayed a diffuse pattern in the cells.
Thus, the distribution of translocated cPLA2 after BFA
treatment resembles distribution of -1,4-galactosyltransferase and
-mannosidase II more than GM 130. In response to Golgi disruption by
NOC, the C2 domain translocation co-localized with a Golgi marker in a
pattern consistent with the ministack formation. Time lapse experiments
with BFA and NOC demonstrated that the dispersed distribution of
translocated C2 domain was reversible and that the normal Golgi
translocation was restored following wash-out of the drugs. Thus,
despite multiple rounds of Golgi component trafficking to ER and back
to Golgi, translocated cPLA2 C2 domain remains associated
with Golgi markers. Moreover, AA release from cells was only slightly
suppressed by treatment with BFA or NOC (data not shown). Whether the
targeted component of the Golgi membrane is an endogenous protein or is
phospholipid-dependent is unknown but is under active
investigation. cPLA2 has been found to bind various
proteins in vitro, including vimentin (15), annexin I (39),
and p11 (40), suggesting a role for interacting proteins in its regulation.
A surprising finding was that translocation to Golgi or to the ER and
PNM is a function of [Ca2+]i amplitude. This is
the first description of differential membrane targeting of a C2
domain-containing protein by [Ca2+]i amplitude.
Translocation of full-length cPLA2 and the C2 domain to
Golgi was initiated by a sustained [Ca2+]i
greater than 100-125 nM, but translocation to the ER or
PNM required a sustained [Ca2+]i greater than
210-280 nM. These [Ca2+]i values
(100-250 nM) for Golgi translocation are in the same range
as the [Ca2+] values required for distribution of
cPLA2 to cell membranes (12) and for binding full-length
cPLA2 and the C2 domain to natural membranes and
phosphatidylcholine-containing liposomes (9). It remains unclear,
however, why Golgi membranes are preferentially targeted at lower
[Ca2+]i. Although there may be some
cPLA2 translocation to ER and PNM at low
[Ca2+]i that we are unable to detect using our
methods, the density of cPLA2 at Golgi membranes at low
[Ca2+]i, and hence the fluorescence, is clearly
greater than at the ER or PNM. Even at [Ca2+]i
above the threshold for translocation to the ER and PNM, the majority
of membrane fluorescence resides at the Golgi. Translocation to the ER
and PNM, however, may be physiologically important particularly with a
sustained increase in [Ca2+]i, because the
ER/Golgi surface area ratio is possibly as high as 8:1 (41). The
AA-metabolizing enzymes prostaglandin H synthase-1 and 5-lipoxygenase
have been localized to the ER and nuclear envelope (42) and to the
nuclear membrane (43), respectively, and it would be of interest to
determine if these enzymes target the Golgi in light of our
observations of cPLA2 targeting.
In addition to investigating targeting in response to sustained
[Ca2+]i, we investigated translocation in
response to transient [Ca2+]i changes. Curiously,
in contrast to the rapid translocation of the C2 domain to the three
membrane domains in response to short duration
[Ca2+]i transients, full-length cPLA2
exhibited translocation primarily to Golgi, although the
[Ca2+]i threshold for translocation to ER and PNM
was greatly exceeded. Full-length cPLA2 did translocate,
however, to all three membrane domains in response to a long duration
[Ca2+]i transient. Full-length cPLA2
may diffuse more slowly than the C2 domain alone or may translocate as
part of a larger, slower diffusing protein complex. Another possibility
is that full-length cPLA2 must first dissociate from a
sequestering agent in the cytoplasm before translocation. Whatever the
mechanism responsible for the observed difference between the
full-length cPLA2 and the C2 domain, the difference
suggests that the catalytic domain exerts an additional regulatory
level on translocation. A consequence of this slower diffusion is that
a predominant target of cPLA2 translocation in response to
a physiological [Ca2+]i change is the Golgi.
Full-length cPLA2 remained associated with the Golgi for a
prolonged period after transient increases in
[Ca2+]i returned to base line, but the C2 domain
did not. Hirabayashi et al. (17) also found that, following
a sustained IONO- or platelet-activating factor-induced
[Ca2+]i increase, cPLA2 remained
associated with intracellular membranes following the addition of EGTA
to the medium. We hypothesize that the prolonged residence of
full-length cPLA2 after translocation and a decrease in the
[Ca2+]i is a function of the catalytic domain or
is the result of cooperative effects between the C2 and catalytic
domains. The Ca2+-independent membrane association may be
related to cPLA2 "trapping" on vesicles in
vitro that has been reported (38). Further experimentation will
determine if this prolonged association of cPLA2 with the target membrane at low [Ca2+]i is a function of
the [Ca2+]i amplitude driving translocation,
duration at the membrane, or phosphorylation state of cPLA2
or is the result of catalytic activity and product formation in the membrane.
AA release studies under conditions similar to the calcium clamping
experiments demonstrated that AA release is
Ca2+-dependent and proportional to the
[Ca2+]i. AA release with increasing
[Ca2+]i generally reflected the increased
fraction of cPLA2 at the membrane, as we have shown in the
calcium clamping experiments. We expect that, in the intermediate
[Ca2+]i conditions, the predominant membrane
target is Golgi and that the AA release in these conditions represents
release from Golgi phospholipids. AA release in response to transient [Ca2+]i increases elicited by ATP, where
translocation was predominantly at the Golgi, was similar to that
generated by the intermediate [Ca2+]i. In
response to a nonphysiological [Ca2+]i increase
elicited by 10 µM IONO, the AA release was supermaximal
and probably reflected the contribution of AA release from ER and PNM
as well as from Golgi phospholipids. A superphysiological AA release
caused by a nonphysiologic increase in [Ca2+]i
has been shown previously (16) and may represent the majority of AA
release studies where a high dose of ionophore is used as the stimulus.
In all [Ca2+]i conditions (sustained
[Ca2+]i elevations of different magnitudes or
transient [Ca2+]i increases of long or short
duration), most of the AA release occurred by 3 min and reached a
plateau between 3 and 5 min, and the time to plateau was independent of
the amount of AA released. The time course of this apparent
cPLA2 inactivation is very similar to that described in
previous studies (37, 38), which showed a great reduction in
cPLA2 activity 2-3 min after binding to membrane.
ATP and IONO induced a rapid (<30 s) increase in AA. Interestingly,
the initial rate of AA release was greatest using ATP, which elicits a
[Ca2+]i increase by rapidly mobilizing
intracellular Ca2+ stores (35). These results contrast
sharply with reports suggesting that [Ca2+]i
transients caused by an intracellular Ca2+ release are not
sufficient to generate AA release (17, 21, 23). In addition, these
results contrast with an observation that a
[Ca2+]i increase with a duration of 120 s
was insufficient for AA release in transfected CHO cells, which the
authors ascribed to a latency period between membrane binding and AA
release (17). In another report of AA release in MDCK D1
cells, a lag of greater than 1 min was observed in response to
epinephrine (22). The difference between the rapid AA release observed
in our study and the lag in AA release observed by others may involve
the state of mitogen-activated protein kinase activation (22, 44) or the phosphorylation state of cPLA2 (45).
Our results demonstrate that cPLA2 is preferentially
targeted to the Golgi in response to physiological, short duration
[Ca2+]i transients. Interestingly,
PLA2 activity has been implicated in Golgi formation (18,
29) and tubulation (29), in membrane remodeling (46), and in membrane
trafficking events including recycling between Golgi and ER (28, 32,
47), endocytosis (48), exocytosis (49), and trafficking of plasma
membrane and secretory proteins (18, 50). It is not clear from most reports whether cPLA2 or other PLA2s (or
multiple PLA2 species) are involved in the described
processes. However, cPLA2 has been directly implicated in
the trafficking of plasma membrane proteins (18), ER to Golgi vesicle
trafficking (47), and maintenance of Golgi structure (18). In one
report (18), expression of a GFP-cPLA2 fusion protein and
an increase in cPLA2 activity was observed to disrupt Golgi
structure. However, in that report, the GFP-cPLA2 chimera
as well as a GFP-cPLA2 C2 domain chimera was observed to be
constitutively associated with membrane, an observation that we and
others do not make (10, 11, 14, 17). In addition, in our model,
disruption of Golgi structure did not occur following cPLA2
translocation of several minutes. At resting
[Ca2+]i, cPLA2 is cytosolic, and it
is unlikely that it would be involved in constitutive membrane
trafficking in unstimulated cells. However, it is intriguing to
hypothesize that cPLA2 translocation to Golgi may be
important for receptor-mediated membrane trafficking events.
| |
FOOTNOTES |
*
This work was supported by National Institutes of Health
Grants HL61378 and HL34303 (to C. C. L.) and Individual National Research Service Award HL10507 (to J. H. E.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
**
To whom correspondence should be addressed: Dept. of Pediatrics,
National Jewish Medical and Research Center, 1400 Jackson St., Denver,
CO 80206. Tel.: 303-398-1214; Fax: 303-270-2155; E-mail:
lesliec@njc.org.
The on-line version of this article (available at
http://www.jbc.org) contains movies for Figs. 1, 3, and 5-7.
Published, JBC Papers in Press, May 24, 2001, DOI 10.1074/jbc.M100943200
| |
ABBREVIATIONS |
The abbreviations used are:
cPLA2, cytosolic phospholipase A2;
PLA2, phospholipase A2;
AA, arachidonic acid;
BFA, brefeldin A;
GFP, green fluorescent protein;
EGFP, enhanced GFP;
HHBSS, HEPES-buffered Hanks' balanced salt solution;
IONO, ionomycin;
NOC, nocodazole;
PNM, perinuclear membrane(s);
TG, thapsigargin;
ER, endoplasmic reticulum;
MDCK, Madin-Darby canine kidney;
PBS, phosphate-buffered saline;
FITC, fluorescein isothiocyanate;
TRITC, tetramethylrhodamine isothiocyanate.
| |
REFERENCES |
| 1.
|
Leslie, C. C.
(1997)
J. Biol. Chem.
272,
16709-16712
|
| 2.
|
Dessen, A.,
Tang |
TOP
ABSTRACT
INTRODUCTION
