Protein Kinase C Regulates the Synthesis of Platelet-activating Factor by Human Monocytes Mark R. Elstad, Thomas M. Mcintyre, Stephen M. Prescott, and Guy A. Zimmerman Nora Eccles Harrison Cardiovascular Research and Training Institute and Departments of Internal Medicine (Pulmonary and Cardiology Divisions) and Biochemistry, University of Utah School of Medicine, Salt Lake City, Utah
Human peripheral blood monocytes synthesize the potent lipid autacoid platelet-activating factor (PAF) following appropriate stimulation. We examined the role of protein kinase C (PKC) in regulating the synthesis ofPAF by stimulated monocytes. 4{3-phorboI12-myristate 13-acetate (PMA) and 1,2-dioctanoyl-snglycerol, which directly activate PKC, stimulated the synthesis of PAP. Sphingosine, a long-chain amine that inhibits PKC, blocked both the binding of phorbol esters to monocytes and the synthesis of PAF in response to PMA (half-maximal inhibition at 5 to 10 pM and complete inhibition at 10 to 30 p,M sphingosine). Thus, the activation of PKC was necessary and sufficient for PAF synthesis in response to phorbol ester. Sphingosine also blocked PAF synthesis in response to the calcium ionophore A23187 and opsonized zymosan particles by specific inhibition of PKC. Two other PKC inhibitors, stearylamine and staurosporine, also blocked PAF synthesis following A23187 or opsonized zymosan stimulation. These experiments demonstrated that PKC activation was required for PAF synthesis in response to the calcium signal generated by A23187 or a receptor-mediated agonist, opsonized zymosan. The synthesis of PAF and leukotriene B. were temporally coupled following cell stimulation. Further, production of these two lipid mediators, and the release of arachidonic acid, were inhibited in parallel by sphingosine. Thus, PKC regulates the synthesis of both PAF and leukotriene B. at a common step, probably phospholipase A2 •
Platelet-activating factor (PAF; I-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) is a potent phospholipid mediator that activates leukocytes and platelets, increases vascular permeability, and mediates leukocyte-endothelial cell interactions (1). PAF is synthesized by a variety of cells, including pulmonary mononuclear phagocytes such as monocytes and alveolar macrophages (2-5). This lipid has been implicated in the pathogenesis of inflammatory responses in the lung and elsewhere (6). PAF mediates acute pulmonary vascular injury in rat endotoxemia (7), causes pulmonary hypertension and edema in isolated rabbit lungs (8), and is released into the bloodstream of rabbits during antibody-mediated lung injury (9). Finally, PAF has been implicated as a mediator of human diseases such as bronchial asthma (10). In contrast to many other human cells (11-13), stimulated monocytes release a large fraction of the newly synthesized
(Received in original form April 27, 1990 and in revised form August 22, 1990) Address correspondence to: Mark R. Elstad, M.D., CVRTI, Building 500, University of Utah, Salt Lake City, UT 84112. Abbreviations: fatty acid-free bovine serum albumin, BSA; 1,2-dioctanoylsn-glycerol, diC8; Hanks' balanced salt solution, HBSS; leukotriene B., LTB.; opsonized zymosan, OZ; platelet-activating factor, PAP; 4{3phorbol 12,13-dibutyrate, PDBu; 4a-phorbol 12,13-didecanoate, POD; protein kinase C, PKC; phospholipase A2 , PLA2 ; 4{3-phorbol 12-myristate 13-acetate, PMA. Am. J. Respir. Cell Mol. BioI. Vol. 4. pp. 148-155, 1991
PAF into the fluid phase in vitro (2). This suggests that monocytes may be a major source of the PAF that is found in the blood of individuals with inflammatory disorders or that is released at sites of inflammation (14-16). Further, the regulation of PAF metabolism is altered during the differentiation of human monocytes to macrophages (17, 18). These observations suggest that the control of PAF synthesis by monocytes is of considerable pathophysiologic importance; however, little is known about the molecular mechanisms that are involved. Protein kinase C (PKC) has been implicated in physiologic responses of monocytes such as the respiratory burst (19) and the production of eicosanoids (20). We previously reported that the tumor promoter, 4{3-phorbol 12myristate 13-acetate (PMA), induces the accumulation and release ofPAF by monocytes (2). This contrasts with the inability of PMA to stimulate PAF synthesis by human neutrophils (12, 21) and endothelial cells (22), and suggests an important difference in the regulation of PAF metabolism between these cells. Since PKC is the receptor for PMA (23), these data implied a role for PKC in regulating PAF synthesis by monocytes. Here we report experiments that define the role of PKC in the regulation of PAF production by human monocytes.
Materials and Methods Materials Leukotriene B. (LTB.) was a gift of Joshua Rokach (Merck Frosst Canada, Quebec, Canada), and mAb 60.3 was the gift
Elstad, McIntyre, Prescott et al.: Protein Kinase C Regulates Monocyte PAF Synthesis
of John Harlan and Patrick Beatty (University of Washington and the Fred Hutchinson Cancer Center, Seattle, WA). The sodium salt of [3H]acetic acid (2.8 Ci/mmol), (3H]phorbol 12,13-dibutyrate (PDBu) (18.9 Ci/mmol), and (3H]arachidonie acid (100 Ci/mmol) were from New England Nuclear (Boston, MA). PAF was from Avanti Polar Lipids (Birmingham, AL), and Hanks' buffered salt solution (HBSS) was from M. A. Bioproducts (Walkersville, MD). Calcium ionophore A23187, fatty acid-free bovine serum albumin (BSA), sphingosine, zymosan, PMA, 4a-phorbol 12,13-didecanoate (PDD), and 1,2-dioctanoyl-sn-glycerol (diC8) were from Sigma Chemical Co. (St. Louis, MO). Monocyte Isolation Monocytes were isolated as described (2). Briefly, mononuclear leukocytes were separated from the blood of healthy volunteers by dextran sedimentation, hypotonic lysis of blood cells, and density-gradient centrifugation. Monocytes were then further purified by countercurrent elutriation. Stock Solutions of Agonists and Inhibitor Opsonized zymosan (OZ) was prepared by incubation of zymosan particles with human serum to allow fixation of complement components as described (24). The suspension was then centrifuged, the serum was removed, and the particles were washed and saved as a stock solution (36 mg OZ/ml PBS). Other stock solutions were prepared as 10-3 M A23187 in dimethylsulfoxide and 10-' M PMA in 1% ethanol. A total of 100 ttM sphingosine in HBSS with 100 ttM BSA was prepared daily from a stock solution of 50 mM sphingosine in ethanol. The control buffer for sphingosine experiments contained the same final concentrations of ethanol and BSA. Assay of PAF Accumulation PAF accumulation was quantified as the incorporation of (3H]acetate into (3H-acetyl]PAF «(3H]PAF), as we have previously described (2). This assay accurately reflects PAF accumulation in monocytes, and other cells, when compared with measurements made by bioassay or other methods (2, II, 12). Suspended monocytes (1.0 to 2.0 X 106 cells in 1.0 ml HBSS with 3.25 mg BSA and 25 ttCi (3H]acetate, unless otherwise indicated) were preincubated with rocking for 5 min at 37° C in a polypropylene tube. The agonist, inhibitor, or control buffer was added and the incubation was carried out at 37° C for the stated time. The reaction was stopped by transferring the cell suspension to a glass tube that contained 1.25 ml chloroform, 2.5 ml acidified methanol (50 mM acetic acid), and 7 ttg unlabeled carrier PAP. The radiolabeled PAF was extracted, separated by thin-layer chromatography, and quantified by scintillation spectrometry. LTB4 and Arachidonic Acid Release The release of LTB. from stimulated monocytes was determined as described (21). Briefly, 4.0 to 8.0 X 106 monocytes were incubated under the conditions for measurement of PAF accumulation (see above) except that (3H]acetate was omitted. Following stimulation, the cell suspension was centrifuged (10,000 X g, 90 s) and the supernatant was recovered. LTB. in the supernatant was then extracted, sepa-
149
rated by high performance liquid chromatography, and quantified by UV absorption at 270 nM in comparison to authentic LTB•. To determine the release of arachidonic acid, suspended monocytes were labeled by incubation with (3H]arachidonie acid (4 X 106 monocytes, 0.5 ttCi (3H]arachidonate, and 1 mg BSA/ml HBSS, 90 min, sr C). The cells were washed twice in HBSS, resuspended (6 X 106 monocytes and 3.25 mg BSA/ml HBSS), preincubated with 30 ttM sphingosine or control buffer for 10 min, and then stimulated with 10 ttM A23187 for 60 min. The reaction was stopped, and the cell-associated and released [3H]arachidonic acid separated by centrifugation (10,000 X g, 90 s). ['Hlarachidonic acid in the fractions was quantitated by scintillation spectrometry. This assay measures the sum of unmetabolized free arachidonate and oxygenated products. (3H]PDBu Binding to Monocytes Competitive binding assays were performed as described (25, 26). Monocytes (1.0 to 2.0 X 106) were preincubated for 5 min at 37° C in 1.0 ml HBSS with 3.25 mg BSA and the stated concentrations of sphingosine. A total of 20 nM (3H]PDBu was then added for an additional 15 min. The monocytes were collected by filtration through a 0.45-ttm metricil membrane filter (Gelman Inc., Ann Arbor, MI) using a Millipore filter apparatus (Millipore Corp., Bedford, MA) and then washed twice with cold HBSS. The filter was dried, and the bound (3H]PDBu was quantitated by scintillation spectrometry (total binding). Specific binding was determined by subtracting the binding obtained in the presence of a 500-fold excess of unlabeled PDBu (nonspecific binding). Translocation of (3H]PDBu Binding Sites from the Cytosol to the Membrane Fraction of Stimulated Monocytes The activation of PKC is associated with translocation of receptors for (3H]PDBu from the cytosol to the membrane (particulate) fraction of stimulated leukocytes (27, 28). We measured PH]PDBu binding using minor modifications of previously described methods (27-30). Suspended monocytes (15 X 106 monocytes/1.0 ml HBSS) were exposed to agonist or control buffer for the stated time. The monocytes were disrupted by sonication at 70 W for 25 s on ice (Braunsonic 1510; B. Braun Melsungen AG, Melsungen, West Germany). Nuclei and unbroken cells were sedimented by centrifugation at 600 X g for 10 min. Duplicate 240-ttl aliquots of the supernatant were then subjected to ultracentrifugation (133,000 X g, 15 min) using an air-driven ultracentrifuge (Airfuge>'; Beckman, Palo Alto, CA). The supernatant from ultracentrifugation (cytosol fraction) was transferred to another tube that contained 260 ttl HBSS, and the pellet (membrane fraction) was resuspended in 500 ttl HBSS. To determine total (3H]PDBu binding, the fractions were incubated in a total volume of 1.0ml HBSS with 0.5 mg BSA, 10 ttg phosphatidylserine, and 20 nM (3H]PDBu for 15 min at room temperature. The reaction was stopped by the addition of 7.5 ml cold HBSS with 0.5 mg BSA/ml and filtration through a GF/C filter (Whatman Ltd., UK). The filter was washed with 7.5 ml of the same buffer, and bound (3H]PDBu quantitated by scintillation spectrometry. Specific binding was determined by subtracting binding measured in identi-
150
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 41991
8-.-------------------,
cally treated mixtures that contained a 500-fold excess of cold PDBu (nonspecific binding). The percent binding to the membrane fraction was calculated by dividing the membrane bound radioactivity by the sum of membrane and cytosol radioactivity and multiplying by 100. c:
Expression of Data and Statistics The experiments were performed a minimum of 2 times with cells from different donors. The mean of the determinations, or representative experiments, are presented. Usually more than two experiments were performed, and in such experiments the SD is shown. All results from scintillation spectrometry were rounded off to the nearest 10 cpm.
Cil Q)
0
s,
0
co
6
~ 00 "5 c: 0 E E ::l 0 ....0 « u, i? 0 « a. ....
4
~
2
0 50 IlM dies
:r
>< I:!... E a.
Results PMA and diC8, Molecules That Directly Activate PKC, Stimulate PAF Synthesis by Human Monocytes We have previously reported that PMA (10-6 M) stimulates PAF synthesis by monocytes in a time-dependent manner with maximal accumulation at 90 min (2). The stimulation of this response by phorbol ester implied that the activation of PKC was involved. To clarify the role of PKC, we first characterized the concentration dependence of this response. Monocytes accumulated 140, 560, 1,960, and 1,570 cpm of (3H]PAF/IQ6 cells at 60 min in response to control buffer, 10-7 , 10-6 , and 10-5 M PMA, respectively. Exposure to 10-6 M PDD, a phorbol ester that does not activate PKC, did not stimulate PAF accumulation (data not shown). We then tested the ability of diCS, a water-soluble diacylglycerol that is an activator ofPKC (31, 32), to stimulate PAF accumulation. As shown in Figure I, diCS stimulated the accumulation of PAF in a concentration- and time-dependent manner. The different time courses in response to PMA and diCS are consistent with the fact that diCS is rapidly metabolized and remains in membranes only transiently, whereas PMA causes prolonged activation ofPKC (23,32). Increased PAF accumulation by diCS-stimulated monocytes may reflect greater access, albeit for a shorter duration, of the diCS to PKC. diCS is reported to cause greater release of arachidonic acid from mouse macrophages than PMA (31); however, comparing quantities of PAF accumulation from a small number of experiments using different donors is difficult because we (2) and others (13) have reported a large interassay variation in PAF accumulation by stimulated leukocytes. The mechanism for this is unknown (13). These experiments indicated that direct activation of PKC, by either of two chemically distinct reagents, was sufficient for PAF synthesis by monocytes. We explored this issue further using several experimental strategies. Sphingosine, an Inhibitor of PKC, Inhibits Both Phorbol Ester Binding to Monocytes and PAF Synthesis in Response to Phorbol Ester The stimulation of PAF synthesis by PMA could have been the result of activation of PKC, or the result of an effect on other cellular processes. It has recently been recognized that longchain sphingoid bases, such as sphingosine and sphinganine, are inhibitors of PKC (21, 26, 33, 34). Sphingosine inhibits PKC activation by competitive displacement of endogenous (diacylglycerol) or exogenous (phorbol ester) activators,
f/O~ • -=:::::: o~/:::::::-~~~~u~ffe;r=====:;=:=~:.:::==e_J o
20
40
60
Time (min)
Figure 1. PAF synthesis in response to diC8 is concentration and time dependent. Monocytes (2 x 106 cells in 1.0 ml HBSS with 3.25 mg BSA and 25 jtCi pH]acetate) were stimulated with the indicated concentrations of diC8 for 15 or 60 min. Total PAF accumulation was determined as described (MATERIALS AND METHODS). A second experiment demonstrated the same diC8 concentration and time dependence, although the maximal total PAF accumulation was less (1,200 cpm/lf)" cells).
thereby preventing formation of an active lipid-enzyme complex (33). Others have demonstrated that sphinganine displaces (3H]PDBu from binding sites (PKC) in platelets (33) and neutrophils (34). Using this strategy, we demonstrated that sphingosine inhibited the binding of (3H]PDBu to human monocytes (Figure 2). This suggests that sphingosine blocks the phorbol receptor of monocytes. Further, sphingosine inhibited PAF synthesis in response to PMA (Figure 3). (3H]PDBu binding and PAF synthesis were inhibited by the same concentration of sphingosine. These experiments further supported the idea that PKC activation was required for PAF synthesis by monocytes in response to PMA. PKC Activation Is an Essential Component of the Monocyte's Response to A23187 Having demonstrated that the activation of PKC was sufficient for PAF synthesis by monocytes, we next asked if PKC was required for PAF synthesis in response to other agonists. The activation of PKC results in its translocation from the cytosol to the membrane fraction of the cell (23, 32). PKC translocation (i.e., activation) can be monitored by measurement of (3H]PDBu binding since PKC is the receptor for phorbol esters (27, 2S). As shown in the upper panel of Figure 4, stimulation of monocytes by the calcium ionophore A231S7 resulted in translocation of (3H]PDBu binding sites from the cytosol to the membrane fraction of the cell. These data indicated that an increase in intracellular calcium induces the association of PKC with the cell membrane in monocytes. The translocation of (3H]PDBu binding sites preceded PAF synthesis (Figure 4A), suggesting that PKC mediates this response.
Elstad, McIntyre, Prescott et al.: Protein Kinase C Regulates Monocyte PAF Synthesis
5.0...-------------------,
151
A ID
c
a.~
~Cl
J:I: M ._ -"0 "01: I: .-
::I.e
I
40
0IDS GIS
c: ....
EO
.e~
EL
100
.~
60
:::l
20
7/. . / .
I:
0
75
. //""..,
~:=::Il'Cl
EE ~ .s< ol'Cl c:a: E
...
....
GI
:E
50
LLo ~~
~~
...••••• 0
25
J:
c::!...
-:
• o+------.---------,-------,-J 20 o 10 30
....•..0·
0 0
0 10
20
30
Time (min)
B
Sphingosine Concentration (!IM)
0························
~;
Figure 2. Sphingosine inhibits [3H]PDBu binding to human monocytes. Monocytes (1.4 x 106 cells) were preincubated with the indicated concentration of sphingosine for 5 min followed by the addition of [3H]PDBu for 15 min. The specific binding of [3H]PDBu to the monocytes was determined as described in MATERIALS AND METHODS. The means of duplicate determinations, which varied less than 10%, are shown. A second experiment gave the same result.
40
">«.
100
[3HlPAF
1../ '-.
e
75
50
o
0
i:::=-l'Cl ::IE E::1)( ol'Cl oE c:a: .... u.
0
c:a:~
a.L
25
s:
c::!...
0------.-----..,----,. 0 o 10 20 30 100--------------------,
"3
E
:::l
o
e'E
0 -c o u. '0 c:a: ~ a. ~
u
Figure 4. Translocation of [3H]pDBu binding sites to the cell membrane precedes PAF synthesis by stimulated monocytes. Suspended monocytes were stimulated with either 15 /LM A23l87 (A) or 3.6 mg OZ/ml (B) for the indicated time. [3H]PDBu binding to the membrane fraction (PKC translocation) and total PH]PAF accumulation were determined as described in MATERIALS AND METHODS. This represents two experiments in which PKC translocation preceded PAF accumulation in response to either agonist.
75
I:
.2 iU
Time (min)
50
~
e,
25
o~--_::.======:;::======QJ o
10
20
30
Sphingosine Concentration (!1 M)
Figure 3. Sphingosine inhibits PAF accumulation by stimulated monocytes. Monocytes (2 x 106) were preincubated for 5 min in 1.0 ml HBSS with 3.25 mg BSA, 25 /LCi ['Hlacetate, and the indicated concentration of sphingosine. The monocytes were then stimulated with PMA (1 /LM, 60 min), OZ (3.6 mg/ml, 15 min), or A23l87 (15 /LM, 60 min) and total PAF accumulation was determined as described (MATERIALS AND METHODS). Data are the mean values from four experiments expressed as percentage of stimulated control in the absence of sphingosine. The mean stimulated control values were 1,100 ± 260, 2,540 ± 540, and 2,260 ± 530 cpm [3H]PAF for PMA, A23l87, and OZ, respectively.
Other data supported a role for PKC in transducing the calcium signal generated by A23187. Sphingosine blocked PAF synthesis in response to A23187 (Figure 3). Stearylamine, a related sphingoid base (26,33), blocked PAF synthesis in the same concentration range, with complete inhibition at 30 JA-M (data not shown). Sphingosine may nonspecifically inhibit leukocyte function by permeabilizing the cell under certain conditions (35); however, we observed that treatment of monocytes with sphingosine did not alter trypan blue exclusion (greater than 95% in all cases). To more rigorously test if sphingosine caused a nonspecific inhibitory effect, we examined the ability of PMA to overcome the inhibition caused by sphingosine. As demonstrated in Figure 5, PMA overcame the sphingosine inhibition of A23187-stimulated PAF production. Since a cytotoxic effect of sphingosine would not be reversed by PMA, this provided compelling evidence that sphingosine was not irreversibly injuring the cells. Further, the results were consistent with competition between sphingosine and PMA for binding sites on PKC. An
152
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 41991
5.0
3
\-----.
~
c::
.~
(;
3~
Eb :> ...
g ><
r: /e
(i) Ql
>. 0 0
c::
2
0
E
oo '"0
Human peripheral blood monocytes synthesize the potent lipid autacoid platelet-activating factor (PAF) following appropriate stimulation. We examined the role of protein kinase C (PKC) in regulating the synthesis ofPAF by stimulated monocytes. 4{3-phorboI12-myristate 13-acetate (PMA) and 1,2-dioctanoyl-snglycerol, which directly activate PKC, stimulated the synthesis of PAP. Sphingosine, a long-chain amine that inhibits PKC, blocked both the binding of phorbol esters to monocytes and the synthesis of PAF in response to PMA (half-maximal inhibition at 5 to 10 pM and complete inhibition at 10 to 30 p,M sphingosine). Thus, the activation of PKC was necessary and sufficient for PAF synthesis in response to phorbol ester. Sphingosine also blocked PAF synthesis in response to the calcium ionophore A23187 and opsonized zymosan particles by specific inhibition of PKC. Two other PKC inhibitors, stearylamine and staurosporine, also blocked PAF synthesis following A23187 or opsonized zymosan stimulation. These experiments demonstrated that PKC activation was required for PAF synthesis in response to the calcium signal generated by A23187 or a receptor-mediated agonist, opsonized zymosan. The synthesis of PAF and leukotriene B. were temporally coupled following cell stimulation. Further, production of these two lipid mediators, and the release of arachidonic acid, were inhibited in parallel by sphingosine. Thus, PKC regulates the synthesis of both PAF and leukotriene B. at a common step, probably phospholipase A2 •
Platelet-activating factor (PAF; I-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) is a potent phospholipid mediator that activates leukocytes and platelets, increases vascular permeability, and mediates leukocyte-endothelial cell interactions (1). PAF is synthesized by a variety of cells, including pulmonary mononuclear phagocytes such as monocytes and alveolar macrophages (2-5). This lipid has been implicated in the pathogenesis of inflammatory responses in the lung and elsewhere (6). PAF mediates acute pulmonary vascular injury in rat endotoxemia (7), causes pulmonary hypertension and edema in isolated rabbit lungs (8), and is released into the bloodstream of rabbits during antibody-mediated lung injury (9). Finally, PAF has been implicated as a mediator of human diseases such as bronchial asthma (10). In contrast to many other human cells (11-13), stimulated monocytes release a large fraction of the newly synthesized
(Received in original form April 27, 1990 and in revised form August 22, 1990) Address correspondence to: Mark R. Elstad, M.D., CVRTI, Building 500, University of Utah, Salt Lake City, UT 84112. Abbreviations: fatty acid-free bovine serum albumin, BSA; 1,2-dioctanoylsn-glycerol, diC8; Hanks' balanced salt solution, HBSS; leukotriene B., LTB.; opsonized zymosan, OZ; platelet-activating factor, PAP; 4{3phorbol 12,13-dibutyrate, PDBu; 4a-phorbol 12,13-didecanoate, POD; protein kinase C, PKC; phospholipase A2 , PLA2 ; 4{3-phorbol 12-myristate 13-acetate, PMA. Am. J. Respir. Cell Mol. BioI. Vol. 4. pp. 148-155, 1991
PAF into the fluid phase in vitro (2). This suggests that monocytes may be a major source of the PAF that is found in the blood of individuals with inflammatory disorders or that is released at sites of inflammation (14-16). Further, the regulation of PAF metabolism is altered during the differentiation of human monocytes to macrophages (17, 18). These observations suggest that the control of PAF synthesis by monocytes is of considerable pathophysiologic importance; however, little is known about the molecular mechanisms that are involved. Protein kinase C (PKC) has been implicated in physiologic responses of monocytes such as the respiratory burst (19) and the production of eicosanoids (20). We previously reported that the tumor promoter, 4{3-phorbol 12myristate 13-acetate (PMA), induces the accumulation and release ofPAF by monocytes (2). This contrasts with the inability of PMA to stimulate PAF synthesis by human neutrophils (12, 21) and endothelial cells (22), and suggests an important difference in the regulation of PAF metabolism between these cells. Since PKC is the receptor for PMA (23), these data implied a role for PKC in regulating PAF synthesis by monocytes. Here we report experiments that define the role of PKC in the regulation of PAF production by human monocytes.
Materials and Methods Materials Leukotriene B. (LTB.) was a gift of Joshua Rokach (Merck Frosst Canada, Quebec, Canada), and mAb 60.3 was the gift
Elstad, McIntyre, Prescott et al.: Protein Kinase C Regulates Monocyte PAF Synthesis
of John Harlan and Patrick Beatty (University of Washington and the Fred Hutchinson Cancer Center, Seattle, WA). The sodium salt of [3H]acetic acid (2.8 Ci/mmol), (3H]phorbol 12,13-dibutyrate (PDBu) (18.9 Ci/mmol), and (3H]arachidonie acid (100 Ci/mmol) were from New England Nuclear (Boston, MA). PAF was from Avanti Polar Lipids (Birmingham, AL), and Hanks' buffered salt solution (HBSS) was from M. A. Bioproducts (Walkersville, MD). Calcium ionophore A23187, fatty acid-free bovine serum albumin (BSA), sphingosine, zymosan, PMA, 4a-phorbol 12,13-didecanoate (PDD), and 1,2-dioctanoyl-sn-glycerol (diC8) were from Sigma Chemical Co. (St. Louis, MO). Monocyte Isolation Monocytes were isolated as described (2). Briefly, mononuclear leukocytes were separated from the blood of healthy volunteers by dextran sedimentation, hypotonic lysis of blood cells, and density-gradient centrifugation. Monocytes were then further purified by countercurrent elutriation. Stock Solutions of Agonists and Inhibitor Opsonized zymosan (OZ) was prepared by incubation of zymosan particles with human serum to allow fixation of complement components as described (24). The suspension was then centrifuged, the serum was removed, and the particles were washed and saved as a stock solution (36 mg OZ/ml PBS). Other stock solutions were prepared as 10-3 M A23187 in dimethylsulfoxide and 10-' M PMA in 1% ethanol. A total of 100 ttM sphingosine in HBSS with 100 ttM BSA was prepared daily from a stock solution of 50 mM sphingosine in ethanol. The control buffer for sphingosine experiments contained the same final concentrations of ethanol and BSA. Assay of PAF Accumulation PAF accumulation was quantified as the incorporation of (3H]acetate into (3H-acetyl]PAF «(3H]PAF), as we have previously described (2). This assay accurately reflects PAF accumulation in monocytes, and other cells, when compared with measurements made by bioassay or other methods (2, II, 12). Suspended monocytes (1.0 to 2.0 X 106 cells in 1.0 ml HBSS with 3.25 mg BSA and 25 ttCi (3H]acetate, unless otherwise indicated) were preincubated with rocking for 5 min at 37° C in a polypropylene tube. The agonist, inhibitor, or control buffer was added and the incubation was carried out at 37° C for the stated time. The reaction was stopped by transferring the cell suspension to a glass tube that contained 1.25 ml chloroform, 2.5 ml acidified methanol (50 mM acetic acid), and 7 ttg unlabeled carrier PAP. The radiolabeled PAF was extracted, separated by thin-layer chromatography, and quantified by scintillation spectrometry. LTB4 and Arachidonic Acid Release The release of LTB. from stimulated monocytes was determined as described (21). Briefly, 4.0 to 8.0 X 106 monocytes were incubated under the conditions for measurement of PAF accumulation (see above) except that (3H]acetate was omitted. Following stimulation, the cell suspension was centrifuged (10,000 X g, 90 s) and the supernatant was recovered. LTB. in the supernatant was then extracted, sepa-
149
rated by high performance liquid chromatography, and quantified by UV absorption at 270 nM in comparison to authentic LTB•. To determine the release of arachidonic acid, suspended monocytes were labeled by incubation with (3H]arachidonie acid (4 X 106 monocytes, 0.5 ttCi (3H]arachidonate, and 1 mg BSA/ml HBSS, 90 min, sr C). The cells were washed twice in HBSS, resuspended (6 X 106 monocytes and 3.25 mg BSA/ml HBSS), preincubated with 30 ttM sphingosine or control buffer for 10 min, and then stimulated with 10 ttM A23187 for 60 min. The reaction was stopped, and the cell-associated and released [3H]arachidonic acid separated by centrifugation (10,000 X g, 90 s). ['Hlarachidonic acid in the fractions was quantitated by scintillation spectrometry. This assay measures the sum of unmetabolized free arachidonate and oxygenated products. (3H]PDBu Binding to Monocytes Competitive binding assays were performed as described (25, 26). Monocytes (1.0 to 2.0 X 106) were preincubated for 5 min at 37° C in 1.0 ml HBSS with 3.25 mg BSA and the stated concentrations of sphingosine. A total of 20 nM (3H]PDBu was then added for an additional 15 min. The monocytes were collected by filtration through a 0.45-ttm metricil membrane filter (Gelman Inc., Ann Arbor, MI) using a Millipore filter apparatus (Millipore Corp., Bedford, MA) and then washed twice with cold HBSS. The filter was dried, and the bound (3H]PDBu was quantitated by scintillation spectrometry (total binding). Specific binding was determined by subtracting the binding obtained in the presence of a 500-fold excess of unlabeled PDBu (nonspecific binding). Translocation of (3H]PDBu Binding Sites from the Cytosol to the Membrane Fraction of Stimulated Monocytes The activation of PKC is associated with translocation of receptors for (3H]PDBu from the cytosol to the membrane (particulate) fraction of stimulated leukocytes (27, 28). We measured PH]PDBu binding using minor modifications of previously described methods (27-30). Suspended monocytes (15 X 106 monocytes/1.0 ml HBSS) were exposed to agonist or control buffer for the stated time. The monocytes were disrupted by sonication at 70 W for 25 s on ice (Braunsonic 1510; B. Braun Melsungen AG, Melsungen, West Germany). Nuclei and unbroken cells were sedimented by centrifugation at 600 X g for 10 min. Duplicate 240-ttl aliquots of the supernatant were then subjected to ultracentrifugation (133,000 X g, 15 min) using an air-driven ultracentrifuge (Airfuge>'; Beckman, Palo Alto, CA). The supernatant from ultracentrifugation (cytosol fraction) was transferred to another tube that contained 260 ttl HBSS, and the pellet (membrane fraction) was resuspended in 500 ttl HBSS. To determine total (3H]PDBu binding, the fractions were incubated in a total volume of 1.0ml HBSS with 0.5 mg BSA, 10 ttg phosphatidylserine, and 20 nM (3H]PDBu for 15 min at room temperature. The reaction was stopped by the addition of 7.5 ml cold HBSS with 0.5 mg BSA/ml and filtration through a GF/C filter (Whatman Ltd., UK). The filter was washed with 7.5 ml of the same buffer, and bound (3H]PDBu quantitated by scintillation spectrometry. Specific binding was determined by subtracting binding measured in identi-
150
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 41991
8-.-------------------,
cally treated mixtures that contained a 500-fold excess of cold PDBu (nonspecific binding). The percent binding to the membrane fraction was calculated by dividing the membrane bound radioactivity by the sum of membrane and cytosol radioactivity and multiplying by 100. c:
Expression of Data and Statistics The experiments were performed a minimum of 2 times with cells from different donors. The mean of the determinations, or representative experiments, are presented. Usually more than two experiments were performed, and in such experiments the SD is shown. All results from scintillation spectrometry were rounded off to the nearest 10 cpm.
Cil Q)
0
s,
0
co
6
~ 00 "5 c: 0 E E ::l 0 ....0 « u, i? 0 « a. ....
4
~
2
0 50 IlM dies
:r
>< I:!... E a.
Results PMA and diC8, Molecules That Directly Activate PKC, Stimulate PAF Synthesis by Human Monocytes We have previously reported that PMA (10-6 M) stimulates PAF synthesis by monocytes in a time-dependent manner with maximal accumulation at 90 min (2). The stimulation of this response by phorbol ester implied that the activation of PKC was involved. To clarify the role of PKC, we first characterized the concentration dependence of this response. Monocytes accumulated 140, 560, 1,960, and 1,570 cpm of (3H]PAF/IQ6 cells at 60 min in response to control buffer, 10-7 , 10-6 , and 10-5 M PMA, respectively. Exposure to 10-6 M PDD, a phorbol ester that does not activate PKC, did not stimulate PAF accumulation (data not shown). We then tested the ability of diCS, a water-soluble diacylglycerol that is an activator ofPKC (31, 32), to stimulate PAF accumulation. As shown in Figure I, diCS stimulated the accumulation of PAF in a concentration- and time-dependent manner. The different time courses in response to PMA and diCS are consistent with the fact that diCS is rapidly metabolized and remains in membranes only transiently, whereas PMA causes prolonged activation ofPKC (23,32). Increased PAF accumulation by diCS-stimulated monocytes may reflect greater access, albeit for a shorter duration, of the diCS to PKC. diCS is reported to cause greater release of arachidonic acid from mouse macrophages than PMA (31); however, comparing quantities of PAF accumulation from a small number of experiments using different donors is difficult because we (2) and others (13) have reported a large interassay variation in PAF accumulation by stimulated leukocytes. The mechanism for this is unknown (13). These experiments indicated that direct activation of PKC, by either of two chemically distinct reagents, was sufficient for PAF synthesis by monocytes. We explored this issue further using several experimental strategies. Sphingosine, an Inhibitor of PKC, Inhibits Both Phorbol Ester Binding to Monocytes and PAF Synthesis in Response to Phorbol Ester The stimulation of PAF synthesis by PMA could have been the result of activation of PKC, or the result of an effect on other cellular processes. It has recently been recognized that longchain sphingoid bases, such as sphingosine and sphinganine, are inhibitors of PKC (21, 26, 33, 34). Sphingosine inhibits PKC activation by competitive displacement of endogenous (diacylglycerol) or exogenous (phorbol ester) activators,
f/O~ • -=:::::: o~/:::::::-~~~~u~ffe;r=====:;=:=~:.:::==e_J o
20
40
60
Time (min)
Figure 1. PAF synthesis in response to diC8 is concentration and time dependent. Monocytes (2 x 106 cells in 1.0 ml HBSS with 3.25 mg BSA and 25 jtCi pH]acetate) were stimulated with the indicated concentrations of diC8 for 15 or 60 min. Total PAF accumulation was determined as described (MATERIALS AND METHODS). A second experiment demonstrated the same diC8 concentration and time dependence, although the maximal total PAF accumulation was less (1,200 cpm/lf)" cells).
thereby preventing formation of an active lipid-enzyme complex (33). Others have demonstrated that sphinganine displaces (3H]PDBu from binding sites (PKC) in platelets (33) and neutrophils (34). Using this strategy, we demonstrated that sphingosine inhibited the binding of (3H]PDBu to human monocytes (Figure 2). This suggests that sphingosine blocks the phorbol receptor of monocytes. Further, sphingosine inhibited PAF synthesis in response to PMA (Figure 3). (3H]PDBu binding and PAF synthesis were inhibited by the same concentration of sphingosine. These experiments further supported the idea that PKC activation was required for PAF synthesis by monocytes in response to PMA. PKC Activation Is an Essential Component of the Monocyte's Response to A23187 Having demonstrated that the activation of PKC was sufficient for PAF synthesis by monocytes, we next asked if PKC was required for PAF synthesis in response to other agonists. The activation of PKC results in its translocation from the cytosol to the membrane fraction of the cell (23, 32). PKC translocation (i.e., activation) can be monitored by measurement of (3H]PDBu binding since PKC is the receptor for phorbol esters (27, 2S). As shown in the upper panel of Figure 4, stimulation of monocytes by the calcium ionophore A231S7 resulted in translocation of (3H]PDBu binding sites from the cytosol to the membrane fraction of the cell. These data indicated that an increase in intracellular calcium induces the association of PKC with the cell membrane in monocytes. The translocation of (3H]PDBu binding sites preceded PAF synthesis (Figure 4A), suggesting that PKC mediates this response.
Elstad, McIntyre, Prescott et al.: Protein Kinase C Regulates Monocyte PAF Synthesis
5.0...-------------------,
151
A ID
c
a.~
~Cl
J:I: M ._ -"0 "01: I: .-
::I.e
I
40
0IDS GIS
c: ....
EO
.e~
EL
100
.~
60
:::l
20
7/. . / .
I:
0
75
. //""..,
~:=::Il'Cl
EE ~ .s< ol'Cl c:a: E
...
....
GI
:E
50
LLo ~~
~~
...••••• 0
25
J:
c::!...
-:
• o+------.---------,-------,-J 20 o 10 30
....•..0·
0 0
0 10
20
30
Time (min)
B
Sphingosine Concentration (!IM)
0························
~;
Figure 2. Sphingosine inhibits [3H]PDBu binding to human monocytes. Monocytes (1.4 x 106 cells) were preincubated with the indicated concentration of sphingosine for 5 min followed by the addition of [3H]PDBu for 15 min. The specific binding of [3H]PDBu to the monocytes was determined as described in MATERIALS AND METHODS. The means of duplicate determinations, which varied less than 10%, are shown. A second experiment gave the same result.
40
">«.
100
[3HlPAF
1../ '-.
e
75
50
o
0
i:::=-l'Cl ::IE E::1)( ol'Cl oE c:a: .... u.
0
c:a:~
a.L
25
s:
c::!...
0------.-----..,----,. 0 o 10 20 30 100--------------------,
"3
E
:::l
o
e'E
0 -c o u. '0 c:a: ~ a. ~
u
Figure 4. Translocation of [3H]pDBu binding sites to the cell membrane precedes PAF synthesis by stimulated monocytes. Suspended monocytes were stimulated with either 15 /LM A23l87 (A) or 3.6 mg OZ/ml (B) for the indicated time. [3H]PDBu binding to the membrane fraction (PKC translocation) and total PH]PAF accumulation were determined as described in MATERIALS AND METHODS. This represents two experiments in which PKC translocation preceded PAF accumulation in response to either agonist.
75
I:
.2 iU
Time (min)
50
~
e,
25
o~--_::.======:;::======QJ o
10
20
30
Sphingosine Concentration (!1 M)
Figure 3. Sphingosine inhibits PAF accumulation by stimulated monocytes. Monocytes (2 x 106) were preincubated for 5 min in 1.0 ml HBSS with 3.25 mg BSA, 25 /LCi ['Hlacetate, and the indicated concentration of sphingosine. The monocytes were then stimulated with PMA (1 /LM, 60 min), OZ (3.6 mg/ml, 15 min), or A23l87 (15 /LM, 60 min) and total PAF accumulation was determined as described (MATERIALS AND METHODS). Data are the mean values from four experiments expressed as percentage of stimulated control in the absence of sphingosine. The mean stimulated control values were 1,100 ± 260, 2,540 ± 540, and 2,260 ± 530 cpm [3H]PAF for PMA, A23l87, and OZ, respectively.
Other data supported a role for PKC in transducing the calcium signal generated by A23187. Sphingosine blocked PAF synthesis in response to A23187 (Figure 3). Stearylamine, a related sphingoid base (26,33), blocked PAF synthesis in the same concentration range, with complete inhibition at 30 JA-M (data not shown). Sphingosine may nonspecifically inhibit leukocyte function by permeabilizing the cell under certain conditions (35); however, we observed that treatment of monocytes with sphingosine did not alter trypan blue exclusion (greater than 95% in all cases). To more rigorously test if sphingosine caused a nonspecific inhibitory effect, we examined the ability of PMA to overcome the inhibition caused by sphingosine. As demonstrated in Figure 5, PMA overcame the sphingosine inhibition of A23187-stimulated PAF production. Since a cytotoxic effect of sphingosine would not be reversed by PMA, this provided compelling evidence that sphingosine was not irreversibly injuring the cells. Further, the results were consistent with competition between sphingosine and PMA for binding sites on PKC. An
152
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 41991
5.0
3
\-----.
~
c::
.~
(;
3~
Eb :> ...
g ><
r: /e
(i) Ql
>. 0 0
c::
2
0
E
oo '"0
