| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Medicine, Section on Infectious Diseases, Wake Forest University Medical Center, Winston-Salem, NC 27157
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
S, but not ADPS, and displaced by 5-oxoETE
analogues, but not by leukotriene B4, lipoxin
A4, or lipoxin B4. Finally, PMN expressed
PT-sensitive GP 
2 and PT-resistant GP
q/11- and 13-chains; eosinophils
expressed only i2 and q/11. We conclude
that 5-oxoETE activates granulocytes through a unique receptor that
couples preferentially to PT-sensitive GP. The strict dependency of
this putative receptor on PT-sensitive GP may underlie the limited
actions of 5-oxoETE, compared with other CF, and help clarify the
complex relations between receptors, GP, cell signals, and cell
responses. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Classical CF receptors operate through the intermediary of GP. Such
GP-coupled receptors (GPCR) activate cells by disassembling GP into
and ß subunits that in turn regulate various signal pathways. GP
fall into four subfamilies based on the homology of their
-chains: Gs (s- and olf-chains),
Gi (i1-, i2-, i3-,
o-, t1-,
t2-, gust-, and
z-chains), Gq/11
(q-, 11-,
14-, and
15/16-chains), and
G12/13 (12- and
13-chains). There are 6 ß and 12
isoforms, but most ß combinations have similar, whereas -chains
exhibit subfamily-specific effects on cell signaling. In consequence,
GPCR can elicit different responses in the same cell by linking to
different GP subfamilies (16, 17). Based on their
sensitivity to the Gi -chain inhibitor, PT,
GPCR for CF couple to Gi (18, 19).
In genetically engineered cells, however, these receptors are only
partly sensitive to PT and require PT-resistant GP for activity
(20, 21, 22, 23). For instance, transfected C5a (20), IL-8
(21), and LTB4 (23)
receptors must couple to 16-, but do not or
cannot couple to q- or
11-chains to elicit various responses. This
finding raises a dilemma when translated to granulocytes. PMN have
i2- and i3-chains
(24, 25), may (26, 27) or may not
(28) have q- or
11-chains, lack
12-chains, and may contain a trace of
16-chains (27). We are unaware of
any report on PMN 13- or
z-chains (unlike other
Gi -chains, z-chains
are PT resistant (17)). Similarly, Eo have
i2- and q-and/or
11- but no z-chains,
and have not been tested for 13-chains
(29, 30). Studies to date thus find a paucity of
PT-resistant GP appropriate for linking to CF receptors in
granuloctyes, even though such GP may be critical for their actions.
Indeed, the limited activity spectrum of 5-oxoETE could reflect
coupling to a more restricted set of GP than that of other CF. We, in
this study, probe PMN and Eo for PT-resistant GP and compare the PT
sensitivity of 5-oxoETE with other CF. PMN express, we find,
q/11- and 13-, but no
16- or z-chains,
while Eo express q/11-chains. Furthermore,
5-oxoETE analogues act preferentially on and through PT-sensitive GP,
and this preference may explain their limited actions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We prepared [3H]5-oxoETE (250 Ci/mmol)
from [3H]5-HETE (11) and
synthesized 5-oxoETE, 5-oxo-15-(OH)-ETE, rac-5-HETE,
15-HETE, and LTB4 (12). We
purchased: LXA4, LXB4, and
triacsin C (Biomol, Plymouth Meeting, PA); PAF and unlabeled
nucleotides (Bachem Biosciences, Philadelphia, PA); RANTES and IL-8
(R&D Systems, Minneapolis MN); C5a, FMLP, ionomycin, PMA,
antiproteases, agarose, newborn calf serum, and delipidated BSA (Sigma,
St. Louis, MO); PT (List Chemicals, Sunnydale, CA); TNF- (Genzyme,
Boston, MA); MEM (Life Technologies, Grand Island, NY);
D-glucose, L-glucose, and
other chemicals (Fisher Scientific, Woodlawn, NJ);
D-[14C]-2-deoxyglucose
([14C]DOG), 200 Ci/mmol,
[3H]5-HETE, 250 Ci/mmol,
[3H]LTB4, 250 Ci/mmol,
and [35S]GTPS, 1120 Ci/mol (DuPont-New
England Nuclear, Boston, MA); Ready Safe Scintillation fluid (Beckman,
Fullerton, CA); cytochalasin B and diisopropylfluorophosphate (Aldrich
Chemicals, Milwaukee, WI); enhanced chemiluminescence (ECL) kits and
nitrocellulose membranes (Amersham, Arlington Heights, IL); HRP-linked
goat Ab to rabbit IgG (Transduction Labs, Lexington, KY); polyclonal Ab
to a C-terminal peptide of
i1/i2 (catalog number
371776), polyclonal Ab to a C-terminal
peptide of i3 (371729), polyclonal Ab to an
internal peptide of common (371737; reacts
with s, i1,
i2, i3,
o, t, and
z), polyclonal Ab to a C-terminal peptide of
12 (371778), and immunizing peptide for
anti-13 C terminus Ab (Calbiochem, La
Jolla, CA); polyclonal Ab to an N-terminal peptide of
q (catalog number sc-393), polyclonal Ab to a
C-terminal peptide of q/11 (sc-392),
polyclonal Ab to an N-terminal peptide of 11
(sc-394), polyclonal Ab to an N-terminal peptide of
13 (sc-410), polyclonal Ab to a C-terminal
peptide of 16 (sc-7415), and immunizing
peptide for anti-q/11 Ab (Santa Cruz
Biotechnology, Santa Cruz, CA); and mAb to ri2
(catalog number MAB3077) and polyclonal Ab to a C-terminal peptide of
13 (AB1651)(Chemicon International, Temecula,
CA). Polyclonal Ab to an N-terminal peptide of
z and a positive control from lysed human
platelets were a generous gift from Dr. A. Nixon and P. J. Casey,
Departments of Molecular Cancer Biology and Biochemistry (Durham, NC).
Cells and stimuli were suspended in modified Hanks’ buffer containing
1.4 mM CaCl2, 12.5 µg/ml BSA, and 10 mM
D-glucose for chemotaxis; this buffer but with no
glucose for hexose uptake; relaxation buffer for Percoll gradient
fractionation; or binding buffer for GTPS-binding assays
(10).
Cell isolation, PT treatment, and assays of function
We isolated PMN (>95% PMN, <5% Eo, <2% monocytes, <5
platelets/100 cells, no erythrocytes) from human donor blood and Eo
(>95% Eo, <3% PMN, <3% monocytes) from PMN preparations by
CD16-negative selection (4). Cells (2 x
107/ml Ca2+-free buffer)
were treated with PT (37°C) for 2 h, washed twice in
Ca2+-free buffer, and suspended in
Ca2+-containing buffer. Chemotaxis was assayed
under agarose (2.5 gm% (w/v) agarose in MEM plus 2.5% (v/v) newborn
calf serum). Aliquots (10 µl) of stimuli, 5 x
105 cells, and Hanks’ buffer were placed in
60-mm-spaced outer, center, and inner wells. Reactions were incubated
at 37°C for 3 h under an atmosphere of 5%
CO2; fixed in 2.5% gluteraldehyde; stained for
granulocytes; and microscopically examined for distance between the
leading front of 
3 cells to the edge of the well of origin
(4). Results are reported in migration indices, i.e., the
mm of migration to stimuli minus that to buffer. Movements toward
buffer-containing wells averaged 0.05 mm for PMN or 0.02 mm for Eo. To
assay hexose uptake, 6 x 105 PMN were
incubated in 200 µl of glucose-free buffer, challenged (37°C) for
10 min, exposed to 60,000 dpm of [14C]DOG in
100 µl of buffer, and incubated for 1 h. Reactions were diluted
with 1 ml of 4°C glucose-free buffer, placed on ice, centrifuged
(12,000 x g, 5 s, 4°C), suspended in 4°C
buffer, and again centrifuged to obtain pellets that were solubilized
in 2 ml of scintillation fluid and counted for label. Results are in
percentage of added radioactivity that was cell associated. When
indicated, PMN were incubated with D-glucose,
L-glucose, or cytochalasin B for 10 min before
challenge.
Percoll gradients
PMN were treated with 2 mM diisopropylfluorophosphate; washed; suspended (3 x 108 cells) in 7 ml of relaxation buffer; subjected to N2-cavitation; treated with 2 mM EGTA, 50 mM 2-ME, and 1 mM PMSF; freed of nuclei and unbroken cells by low-speed centrifugation; and resolved on discontinuous Percoll gradients into 21 fractions that were enriched with markers for: cytosol (lactate dehydrogenase, fractions 1–4); plasma membranes (surface alkaline phosphatase, fractions 5–8); light Golgi (UDP-galactose: N-acetyl-glucosamine galactosyltransferase, fractions 5–8); secretory vesicles (latent alkaline phosphatase, fractions 7–11); secondary granules (myeloperoxidase, fractions 13–16); heavy Golgi (UDP-galactose:N-acetyl glucosamine galactosyltransferase, fractions 13–16); and primary granules (ß-glucuronidase, fractions 18–21) (10). Cytosol, plasma membrane, and granule fractions were pooled and centrifuged (200,000 x g, 2 h, 4°C). Supernatants from cytosol were treated with pepstatin (25 µg/ml), leupeptin (40 µg/ml), aprotinin (0.05 U/ml), PMSF (1 mM), soybean trypsin inhibitor (10 µg/ml), benzamidine (3 mM), DTT (1 mM), and SDS (1% [w/v]). Pellets were washed twice and suspended to original volume in buffer containing these inhibitors. Where indicated, secondary and primary granule pools were suspended to half original volume and combined. In other studies, PMN or Eo (1.5 x 107 per ml of Hanks’ buffer; no Ca2+) were suspended with the inhibitors given above (including 2 mM diisopropylfluorophosphate); sonicated at 4°C (three 10-s strokes; setting of 2; Heat Systems, Model W220, St. Louis MO); treated with SDS; and centrifuged (240,000 x g, 20 min, 20°C). Supernatants were isolated and pellets brought to original volume in buffer with the same inhibitors.
Western blots
Pools of cell fractions were placed in boiling water for 5 min. Samples (30–80 µl, equal to 1–3 x 106 cells) were run on SDS-PAGE (5% loading/10% separating gels (37.5:1 bis-polyacrylamide)) to 9 cm at 35 mA for 5 h and electrotransferred to nitrocellulose at 100 mA for 16 h. The blots were washed in TBS (10 mM Tris, 100 mM NaCl, pH 7.4) four times; blocked with 1–3% (w/v) BSA in TBS for 1 h; washed twice in TBS; treated with a 1/2000 dilution of primary Ab for 1.5 h; washed twice in TBS; treated with 1/2000 dilutions of peroxidase-tagged secondary Ab for 1 h; washed four times with TBS containing 0.2% (v/v) Tween-20; and visualized by ECL (12). For some studies, fractions were run on SDS-PAGE (5% stacking/12% separating gels (37.5:1 bis-polyacrylamide)) to 14 cm for 4 h at 50 mA; transferred to nitrocellulose (500 mA over 3–5 h); washed twice in TBS; blocked for 1 h with 3% (w/v) Carnation nonfat dry milk in TBS plus 0.1% (v/v) Tween 20; washed twice in TBS/0.2% Tween 20; incubated with a 1/2000 dilution of primary Ab in 1% (w/w) BSA/TBS/0.2% Tween-20 for 2–3 h at 20°C; washed twice in TBS/0.2% Tween 20; incubated with 1/2000 dilutions of secondary Ab in 3% dry milk/TBS/0.2% Tween 20; washed with TBS/0.2% Tween 20; and visualized by ECL. Blots were stripped of Ab by a 30-min incubation at 50°C with 100 mM mercaptoethanol and 2% SDS in 65 mM Tris-HCl buffer (pH 6.7).
Assay of GP activity
PMN (1 x 108 per ml of HBSS buffer
(no added Ca2+ or Mg2+))
were incubated at 37°C for 2 h with 0–2 µg/ml of PT. Cells
were washed and suspended in isotonic saline (4°C), incubated (5
x 107/ml) with 2 mM diisopropylfluorophosphate
for 5 min, twice washed in isotonic saline, and suspended at 2 x
108/ml in HBSS buffer (no
Ca2+ or Mg2+) with 2 mM
EGTA, 50 mM 2-ME, and 1 mM PMSF. Suspensions were sonicated and
centrifuged (100,000 x g; 1 h; 4°C); pelleted
membranes were taken up in binding buffer. Aliquots (100 µl) of
membranes equivalent to 3 x 105 PMN were
incubated at 30°C with 50 pM [35S]GTPS
(±100 µM GTPS), 320 nM GDP, 50 µg BSA, and a stimulus, and
passed through G/FC filters. Filters were washed five times with 1 ml
of buffer (50 mM triethanolamine, 4 mM MgCl2, pH
7.4, 4°C); air dried; incubated for 10 min with methanol (250 µl);
suspended in 2 ml of scintillation fluid; and counted for
radioactivity. We defined specific binding as the fraction of added
radiolabel bound by membranes incubated with 50 pM
[35S]GTPS minus that bound by membranes
incubated with 50 pM [35S]GTPS plus 100 µM
GTPS. Results are reported as the specific binding of
[35S]GTPS by membranes incubated with a
stimulus minus the specfic binding by membranes incubated with all
reagents except a stimulus.
Ligand-binding assays
PMN were treated with diisopropylfluorophosphate, PMSF, and EGTA; sonicated; and centrifuged, as described above. Membranes from 2 x 108 cells were suspended in 5 ml of Hanks’ buffer with 1.4 mM CaCl2 and 0.7 mM MgCl2; treated (37°C) with 12.5 µM triacsin C for 30 min; and incubated with 100 pM of [3H]5-oxoETE or [3H]LTB4 in the presence or absence of other agents. Reactions were passed through GF/C filters and washed five times with 1 ml of Hanks’ buffer (no glucose, 1.4 mM Ca2+, and 0.7 mM Mg2+, 4°C). Filters were air dried, treated with methanol (250 µl) for 10 min, mixed with 2 ml of scintillation fluid, and counted for radioactivity.
Ligand metabolism
Mixed membranes from sonicated PMN were treated with 12.5 µM triacsin C and incubated with 100 pM of [3H]5-oxoETE or [3H]LTB4 as in ligand-binding assays. After 20 min, reactions were made pH 3–4 with HCl; extracted with chloroform:methanol (2:1, v/v, per volume of membranes); and analyzed by TLC (11). Virtually all recovered radioactivities migrated with the appropriate 5-oxoETE or LTB4 standard, and therefore had not been appreciably metabolized.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
illustrates these responses as they occurred after 1 h of
challenge. Cytochalsin B, which blocks the glucose transporter, and
D-glucose, but not L-glucose, abrogated the
uptake of [14C]DOG in both resting and
challenged cells (Table I). Thus, the
responses of Fig. 1 are due to the stimulation of a constitutively
active, stereospecific hexose carrier. PT blocked the actions of CF,
but did not alter hexose uptake in resting cells (data not shown) nor
inhibit agents that operate independently of GP viz, PMA,
ionomycin, or TNF- (Fig. 2). Among the
CF, however, PAF had minimal, and C5a, FMLP, IL-8, and
LTB4 only partial sensitivity to PT. Indeed, PMN
treated with even 16 µg/ml of the toxin took up significant amounts
of DOG in response to these CF. In sharp contrast, 1 µg/ml of PT
almost completely abrogated the action of 5-oxoETE (Fig. 2) as well as
rac-5-HETE (data not shown). PT likewise only partly blocked
PMN and Eo chemotactic responses to C5a, FMLP, PAF, and
LTB4 (Fig. 3). In
studies not shown, it also blocked PMN migration to 2.5 or 25 nM IL-8
by less than 70% at 16 µg/ml and Eo migration to 25 or 100 nM RANTES
by less than 60% at 4 µg/ml (16 µg/ml of PT clumped Eo). The toxin
again had a far greater impact on 5-oxoETE, totally blocking the
responses of both cell types to the eicosanoid at >1 µg/ml (Fig. 3).
We conclude that 5-oxoETE analogues exhibit a singularly high
sensitivity to PT. We next determined whether this effect is realized
at the level of GP.
|
|
|
|
S (50 pM),
membranes from sonicated PMN incorporated radioactivity progressively
over 60 min, after which apparent equilibrium occurred. This
incorporation was due to the specific binding of the label since >1
µM of GTPS displaced [35S]GTPS by
>90% (data not shown). GDP had a similar effect, reducing
[35S]GTPS binding by 50% and 90% at 300
and 1000 µM, respectively (data not shown). Given these results, the
binding of label to membranes incubated with 50 pM of
[35S]GTP plus 320 nM of GDP reflects
competition between [3H]GTPS and GDP for a
common acceptor (10, 31). In the presence of 320 µM of
GDP, CF stimulated the incorporation of radioactivity into membranes,
i.e., increased the binding of [35S]GTPS
relative to GDP. The most active agent, LTB4, was
3-fold more powerful than 5-oxoETE or rac-5-HETE (Fig. 4) and 1.3- to 2.5-fold more powerful
than FMLP, C5a, IL-8, or PAF (data not
shown). To
examine the effect of PT on these responses, we treated PMN with PT for
2 h, sonicated the cells, isolated their membranes, and stimulated
these membranes with an optimally effective and at least one lower
concentration of each CF. Membranes from PMN treated with 0.5 or 2
µg/ml PT had greatly reduced responses to 5-oxoETE compared with
those to FMLP, C5a, IL-8, PAF, or LTB4 (Fig. 5).
Unlike the latter agents, then, 5-oxoETE induces the exchange of
[35S]GTPS for GDP preferentially in
PT-sensitive GP. This could be due to a paucity of 5-oxoETE binding
sites if the Gi subfamily has a predilection over
other subfamilies for receptor coupling. That is,
Gi heterotrimers might tie up a sparsely
represented receptor type, effectively excluding it from interacting
with other heterotrimers (31). We accordingly enumerated
5-oxoETE and LTB4 binding sites.
|
|
and Table II).
[3H]LTB4 binding to the
membranes followed a similarly rapid (equilibrium reached within 10
min) and reversible (>98% with 1 µM of LTB4)
pattern. LTB4, but not 5-oxoETE, displaced
[3H]LTB4 (Table II).
Although there was a large difference in the specific binding of
5-oxoETE and LTB4, Scatchard analysis revealed
that this resulted from the 30-fold greater
Kd of high- and low-affinity receptors
for LTB4, relative to the comparable
Kd values for 5-oxoETE. 5-OxoETE
actually bound to more high- and low-affinity sites than
LTB4 (Fig. 7).
Evidently, then, 5-oxoETE binding sites do not interact appreciably
with LTB4, LXA4, or
LXB4, and the different PT sensitivities of
5-oxoETE and LTB4 are not explained by a simple
quantitative difference in their respective binding sites.
|
|
|
-chains. Ab to
common (data not shown),
ri2 (Fig. 8,
left panel), and i2 C terminus
(data not shown) reacted with a 42-kDa band in particulate but not
soluble fractions. Anti-i3 C terminus Ab,
which reacts with i2 and
i3, and Ab to the N terminus of
i2, also detected this band (data not shown).
We further observed that Ab to the common C terminus of
q and 11 recognized a
particulate but not soluble band at 44 kDa (Fig. 8, left
panel). Blots probed with anti-ri2
Ab, stripped, and probed with Ab to q/11, or
blots probed with these Ab in the opposite sequence revealed that
i2 moved faster than
q/11. Ab to the N terminus of
q and Ab to the N terminus of
11 also reacted with the latter band (data not
shown). We therefore identify the 44-kDa Ag as
q/11. Finally,
anti-13 C terminus Ab detected a band at
43 kDa in particulate but not soluble PMN fractions (Fig. 8, right panel). Sequential probing and stripping experiments
indicated this band moved between i2 and
q/11. However, fractions stored at -20°C
instead of examined immediately after isolation had a band at 41 in
addition to 43 kDa that reacted with Ab to the C terminus of
13. Only the 43-kDa band reacted with Ab to
13 N terminus (Fig. 8, right
panel). Thus, PMN express i2 and
q/11, but not 16.
They also express 13, which, it appears, is
subject to N-terminal degradation during storage.
|
i2 Ab, as well as a 44-kDa band
reacting with anti-q/11 Ab. Ab to the C
terminus of 13 detected a 41-kDa band in
plasma membrane (Fig. 9, upper
panel). Since Ab to the N terminus of 13
did not detect this band (data not shown), it is likely that
13 suffered N-terminal degradation during the
several hours required to fractionate PMN on Percoll gradients. In any
event, the bands detected with Ab to i2,
q/11, and 13 migrated
with their counterparts from sonicated PMN (data not shown), and
q/11 and 13 were
further distinguished from each other by immunizing peptides. These
peptides blocked reactions to the cognate, but not opposite Ab (Fig. 9, lower panel). Finally, anti-z
Ab, while readily reacting with a 42-kDa band in the lysates of human
platelets, detected no such band in PMN Percoll gradient fractions. Ab
to 12 and Ab to 16
also gave negative results in Percoll gradient as well as sonicated PMN
fractions (data not shown). Evidently, then,
i2 and q/11 reside in
PMN plasma membranes and secondary granules, whereas
13 localizes only to plasma membranes. PMN do
not have measurable amounts of z,
12, or 16, at least
when examined with the methods and Ab used in this study.
|
i2 and
q/11 Ab, respectively. The latter bands
migrated with their counterparts from PMN (Fig. 8
, left
panel). However, Eo from six of six donors did not exhibit a band
reacting with Ab to the N terminus or C terminus of
13 (results not shown). Thus, Eo membranes
express i2 and q/11,
but not 13. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
i2 (Fig. 8, upper
panel). We cannot agree, however, with a report (28)
finding no q/11 in PMN. Rather, we confirm
studies (26, 27) finding PMN have a band that reacts with
Ab to the common C terminus of q and
11 (Fig. 8, left panel). The band
also: reacted with Ab to the N terminus of 11
and Ab to the N terminus of q (these termini
are structurally similar); moved at 44 kDa, similar to authentic
q and 11
(27); localized to plasma membranes and secondary granules
(Fig. 9, upper panel); did not react with Ab in the presence
of immunizing peptide (Fig. 9, lower panel); and occurred in
the particulate fraction of sonicated Eo (Fig. 8, left
panel). In addition, PMN, but not Eo, had a plasmalemma band
reactive with two different Ab to 13. This
band migrated at 43 kDa (Fig. 8, left panel), the same rate
found in other cells (32), and was not detectable in the
presence of immunizing peptide (Fig. 9, lower panel).
Finally, PMN (data not shown), similar to Eo (30), lacked
a band reacting with Ab to z. PMN also failed
to express material reactive with Ab to 12 or
16. Tenailleau et al. (27), while
not finding 12 in PMN, did detect a faint
doublet with Ab that, unlike our anti-16
C-terminal Ab, was directed to the N terminus of
16. This may indicate that
16 is degraded at its C terminus during cell
processing. However, the findings of Tenailleau et al. could be due to
monocyte contamination since these cells express high levels of
16 (27). In any event, PMN
express at the very least PT-sensitive i2 and
PT-resistant q/11,
13, and perhaps a trace of
16, while Eo express
i2 and q/11. These
results allow a possibility that the PT-resistant responses actuated by
transfected CF receptors are mediated by an unsuspected coupling of
these receptors to 13, the presence of which
was not sought in recipient model cells (9, 10, 11, 12, 13). They may
also pertain to PMN.
C5a, FMLP, PAF, LTB4, IL-8, and RANTES elicited
PMN and Eo chemotaxis and PMN hexose uptake responses that were only
partly inhibited by PT (Figs. 2 and 3). Although these results contrast
with findings that granulocyte responses to CF are sometimes totally
blocked by PT, four points merit emphasis. First, PT can have
nonspecific effects. We found, for example, that 16 µg/ml of PT
irreversibly clumped Eo. This led to total blockage of Eo random
motility and migration responses, effects that cannot be related simply
to GP inhibition. Second, the efficacy of PT in PMN (32, 33) and Eo (34, 35, 36) sometimes depends on CF dosage,
with high concentrations of CF typically overcoming the PT inhibition,
even if total, that occurs at a lower CF concentration. Note that we
observed PT completely blocked optimal doses of 5-oxoETE at levels only
weakly blocking even suboptimal levels of other CF. Third, PT has
response-selective effects (32, 33, 34, 35, 36, 37), e.g.,
Ca2+ transient responses to
LTB4 are not reduced under conditions in which
migration responses are fully blocked by a given amount of PT
(37). Fourth, in any given assay, PT may inhibit some, but
not other, GPCR. This is evidenced by studies in which PAF-induced
Ca2+ transients in PMN and Eo (32, 36, 37) and hexose uptake in PMN (Fig. 2) are not altered by PT at
levels that block other CF. Indeed, CF commonly exhibit a wide range of
PT sensitivities, and within this range PAF often occupies a position
of extreme PT resistance (19, 22). We find that 5-oxoETE
behaves oppositely to PAF in that PT blocked PMN and Eo responses to it
at levels far below those comparably inhibiting the more classical CF.
Indeed, PT fully suppressed responses to 5-oxoETE or
rac-5-HETE, but not to C5a, FMLP, PAF,
LTB4, IL-8, or RANTES (Figs. 2 and 3). Similar
results occurred in experiments that measured the coupling of CF
receptors to heterotrimeric G proteins more directly.
The mixed membranes from sonicated PMN reversibly bound
[3H]5-oxoETE and
[3H]LTB4. 5-OxoETE,
5-oxo-15-hydroxy-ETE, and rac-5-HETE, but not
LTB4 or 15-HETE, displaced
[3H]5-oxoETE, while LTB4,
but not 5-oxoETE, displaced
[3H]LTB4 (Fig. 6, Table II). GTPS inhibited the
binding of both radioligands, whereas ATPS had no such effect (Table II). These results parallel those observed with partially purified
plasma membranes from PMN (11). Since FMLP, C5a, PAF, and
IL-8 did not displace 5-oxoETE binding to plasma membranes
(11), and since LXA4 and
LXB4 did not alter 5-oxoETE binding to mixed
membranes (Fig. 6), the 5-oxoETE binding site appears distinct from
those for other granulocyte-stimulating eicosanoids and CF.
5-OxoETE induced these same mixed membrane preparations to
incorporate [35S]GTPS (Fig. 4), a property
characteristic of the GTP/GDP exchange reaction conducted by GPCR, such
as those for CF. Indeed, other CF stimulated a similar exchange, but
their effects proved insensitive to PT under conditions that fully
abrogated the action of 5-oxoETE (Fig. 5). Taken together, these
results suggest PMN have a plasma membrane GPCR for 5-oxoETE. This
putative GPCR resembles those for other CF in coupling to GP, but is
unique in binding 5-oxoETE analogues rather than other eicosanoids or
CF and in exhibiting exquisite PT sensitivity in selected functional
and G protein activity assays. We stress that our PT treatment regimen
used a limited incubation time (PMN incubated for >2 h, ± PT, showed
reduced responses to all CF), and accordingly may not have inactivated
all PT-sensitive GP. Nonetheless, the data are compatible with the
notion that putative 5-oxoETE receptors operate through
Gi, while receptors for other CF have actions
that go beyond this GP subfamily.
GPCR coupled to Gi, but not those coupled to
Gq, activate MAPK, even though ß-chains bear
the responsibility for regulating MAPK in COS-7 and Rat-1 cells
(38, 39, 40). Similarly, Gi-coupled, but
not Gq-coupled or
Gs-coupled, GPCR elicit chemotactic responses,
although it is ß-chains that mediate the motility responses of
HEK293 cells (41, 42). Such results imply that GPCR,
including those for IL-8, can act independently of GP in genetically
engineered cells (42). Differences in the PT sensitivities
of classical CF and 5-oxoETE would be explained if GPCR for the former
agents have GP-independent effects, while the putative 5-oxoETE
receptor acts strictly through GP. It is also possible that some
Gi-chains, because of compartmentalization,
contribute to function in PT-treated cells. In this scenario, PT (which
under our conditions attacks only plasma membrane GP) would partly
inhibit classical CF if they operate through both surface membrane
and secondary granule-associated GP (Fig. 9), yet totally inhibit
5-oxoETE if it acts exclusively on plasmalemma GP. Alternatively, PT
might target only a portion of plasma membrane Gi
-chains, reducing their net activity below some threshold level
needed to transmit the action of 5-oxoETE, but not below that needed by
classical CF. These and other explanations do not stem from a
difference in receptor density since receptors for 5-oxoETE are
numerically comparable with those for LTB4 (Fig. 7) and, by extrapolation, PAF and FMLP (43).
In conclusion, PMN membranes contain a site that binds 5-oxoETE
analogues, but not LTB4,
LXA4, LXB4, PAF, FMLP, C5a,
or IL-8. This site resembles the receptors for other CF in that it
couples to GP, as indicated by the effects of guanosine nucleotides on
5-oxoETE binding and the ability of 5-oxoETE analogues to stimulate
GDP/GTP exchange. We hypothesize this site is a novel GPCR. The
putative 5-oxoETE GPCR nonetheless differs from the GPCR for other CF
receptors since it activates mainly, if not exclusively, PT-sensitive
GP and displays uniquely high sensitivity to PT in PMN and Eo. The
latter cells have PT-resistant q/11-,
13-, and/or, perhaps,
16- in addition to PT-sensitive
i2- and, probably,
i3-chains. We accordingly also hypothesize the
putative 5-oxoETE receptor preferentially couples to the
Gi subfamily, while receptors for other CF couple
to not only this, but also G12/13 and/or
Gq/11 subfamilies of GP. However, we cannot
exclude a possibility that GPCR for the latter CF, but not 5-oxoETE
receptors, have effects on PMN and Eo that bypass GP and thereby attain
PT resistance. Regardless of this or other possibilities, the
distinctive GP coupling of putative 5-oxoETE GPCR may explain the
limited granulocyte-stimulating actions of 5-oxoETE analogues, and
prove useful to probe the connections between particular receptors, GP
subfamilies, cell signal pathways, and cell responses.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Joseph T. O’Flaherty, Department of Medicine, Section on Infectious Diseases, Wake Forest University Baptist Medical Center, Medical Center Boulevard, Wake Forest University Medical Center, Winston-Salem, NC 27157. E-mail address:
3 Abbreviations used in this paper: 5-oxo-ETE, 5-oxo-6,8,11,14-(E,Z,Z,Z)-eicosatetraenoate; CF, chemotactic factor; [14C]DOG, D-[14C]-2-deoxyglucose; ECL, enhanced chemiluminescence; Eo, eosinophil; GP, heterotrimeric G protein; GPCR, G protein-coupled receptor; 15-HETE, 15(S)-hydroxy-5,8,12,14-(Z,Z,Z,E)-eicosatetraenoate; LTB4, leukotriene B4; LXA4, lipoxin A4; LXB4, lipoxin B4; MAPK, mitogen-activated protein kinase; 5-oxo-15-hydroxy-ETE, 5-oxo-15(S)-hydroxy-6,8,11,14-(E,Z,Z,E)-eicosatetraenoate; PAF, platelet-activating factor; PMN, polymorphonuclear neutrophil; PT, pertussis toxin; rac-5-HETE, 5-(rac)-hydroxy-6,8,11,14-(E,Z,Z,Z)-eicosatetraenoate.
Received for publication November 12, 1999. Accepted for publication January 7, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
I and 16, but not q or 11 G-protein subunits. Biochem. J. 335:15.
in human neutrophils. J. Biol. Chem. 263:10958. stimulates increased plasma membrane guanine nucleotide binding protein activity in polymorphonuclear leukocytes. J. Leukocyte Biol. 57:500.[Abstract]
with interleukin-8 receptors. J. Biol. Chem. 271:12783. subunits. Nature 369:418.[Medline]
activation of p21ras. Proc. Natl. Acad. Sci. USA 91:12706. subunit of Gi, not by activating Gq or Gs. Proc. Natl. Acad. Sci. USA 94:14489.i is not required for chemotaxis mediated by Gi-coupled receptors. J. Biol. Chem. 274:2824.This article has been cited by other articles:
|
|
 |
|
  C. Cossette, P. Patel, J. R. Anumolu, S. Sivendran, G. J. Lee, S. Gravel, F. D. Graham, A. Lesimple, O. A. Mamer, J. Rokach, et al. Human Neutrophils Convert the Sebum-derived Polyunsaturated Fatty Acid Sebaleic Acid to a Potent Granulocyte Chemoattractant J. Biol. Chem., April 25, 2008; 283(17): 11234 - 11243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Koike, H. Obinata, A. Yamamoto, S. Takeda, H. Komori, F. Nara, T. Izumi, and T. Haga 5-Oxo-Eicosatetraenoic Acid-Induced Chemotaxis: Identification of a Responsible Receptor hGPCR48 and Negative Regulation by G Protein G12/13. J. Biochem., March 1, 2006; 139(3): 543 - 549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Mercier, C. Morin, M. Cloutier, S. Proteau, J. Rokach, W. S. Powell, and E. Rousseau 5-Oxo-ETE regulates tone of guinea pig airway smooth muscle via activation of Ca2+ pools and Rho-kinase pathway Am J Physiol Lung Cell Mol Physiol, October 1, 2004; 287(4): L631 - L640. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. B. Stamatiou, C.-C. Chan, G. Monneret, D. Ethier, J. Rokach, and W. S. Powell 5-Oxo-6,8,11,14-eicosatetraenoic Acid Stimulates the Release of the Eosinophil Survival Factor Granulocyte/Macrophage Colony-stimulating Factor from Monocytes J. Biol. Chem., July 2, 2004; 279(27): 28159 - 28164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Brink, S.-E. Dahlen, J. Drazen, J. F. Evans, D. W. P. Hay, G. E. Rovati, C. N. Serhan, T. Shimizu, and T. Yokomizo International Union of Pharmacology XLIV. Nomenclature for the Oxoeicosanoid Receptor Pharmacol. Rev., March 1, 2004; 56(1): 149 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Jones, S. Holden, L. Tenaillon, U. Bhatia, K. Seuwen, P. Tranter, J. Turner, R. Kettle, R. Bouhelal, S. Charlton, et al. Expression and Characterization of a 5-oxo-6E,8Z,11Z,14Z-Eicosatetraenoic Acid Receptor Highly Expressed on Human Eosinophils and Neutrophils Mol. Pharmacol., March 1, 2003; 63(3): 471 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T. O'Flaherty, L. C. Rogers, B. A. Chadwell, J. S. Owen, A. Rao, S. D. Cramer, and L. W. Daniel 5(S)-Hydroxy-6,8,11,14-E,Z,Z,Z-eicosatetraenoate Stimulates PC3 Cell Signaling and Growth by a Receptor-dependent Mechanism Cancer Res., December 1, 2002; 62(23): 6817 - 6819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Channon, K. A. Miselis, L. A. Minns, C. Dutta, and L. H. Kasper Toxoplasma gondii Induces Granulocyte Colony-Stimulating Factor and Granulocyte-Macrophage Colony-Stimulating Factor Secretion by Human Fibroblasts: Implications for Neutrophil Apoptosis Infect. Immun., November 1, 2002; 70(11): 6048 - 6057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kato, H. Kimura, Y. Motegi, A. Tachibana, H. Minakami, A. Morikawa, and H. Kita Platelet-Activating Factor Activates Two Distinct Effector Pathways in Human Eosinophils J. Immunol., November 1, 2002; 169(9): 5252 - 5259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hosoi, Y. Koguchi, E. Sugikawa, A. Chikada, K. Ogawa, N. Tsuda, N. Suto, S. Tsunoda, T. Taniguchi, and T. Ohnuki Identification of a Novel Human Eicosanoid Receptor Coupled to Gi/o J. Biol. Chem., August 23, 2002; 277(35): 31459 - 31465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. E. L. Stubbs, P. Schratl, A. Hartnell, T. J. Williams, B. A. Peskar, A. Heinemann, and I. Sabroe Indomethacin Causes Prostaglandin D2-like and Eotaxin-like Selective Responses in Eosinophils and Basophils J. Biol. Chem., July 12, 2002; 277(29): 26012 - 26020. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Hevko, R. C. Bowers, and R. C. Murphy Synthesis of 5-Oxo-6,8,11,14-eicosatetraenoic Acid and Identification of Novel omega -Oxidized Metabolites in the Mouse Macrophage J. Pharmacol. Exp. Ther., April 13, 2001; 296(2): 293 - 305. [Abstract] [Full Text]
|
|
|
|
|
|
R. C. Bowers, J. Hevko, P. M. Henson, and R. C. Murphy A Novel Glutathione Containing Eicosanoid (FOG7) Chemotactic for Human Granulocytes J. Biol. Chem., September 22, 2000; 275(39): 29931 - 29934. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05,
F-distribution), the numbers per cell (Rt), and
Kd values of which are given in each
panel.
