| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Biochemistry and Biophysics, Second University of Naples; and
Division of Allergy and Clinical Immunology, University of Naples Federico II, Naples, Italy
1Correspondence: Department of Biochemistry and Biophysics, Second University of Naples, via L. De Crecchio 7, Naples 80138, Italy. E-mail: mluisa.balestrieri{at}unina2.it
ABSTRACT
The transacetylase (TA), reported to be identical to platelet-activating factor (PAF) acetylhydrolase (II), is a multifunctional enzyme with three catalytic activities: lysophospholipid transacetylase (TAL), sphingosine transacetylase (TAS), and acetylhydrolase (AH). We report that TAL activity participates in the control of PAF levels in monocytes and macrophages and that its regulation differs in these two types of cells. In monocytes, LPS or granulocyte-macrophage colony-stimulating factor (GM-CSF) specifically increased the TAL activity. Western blot analysis and enzyme assays on immunoprecipitates revealed that the increased activity can be ascribed to PAF-AH (II) and that both translocation from cytosol to membranes and p38/ERKs-mediated phosphorylation regulate the enzyme activation. Instead, in macrophages differentiated in vitro from monocytes by incubation with FCS, an increase of both TAL and AH activities was observed. Moreover, activation of ERKs and p38 MAP kinase was not required for the up-regulation of PAF-AH (II) in differentiated macrophages. The differences observed in macrophages as compared to monocytes can be explained by 1) p38/ERKs-independent phosphorylation of PAF-AH (II) and 2) appearance of plasma PAF-AH in the course of macrophage differentiation.—Servillo L., Balestrieri C., Giovane A., Pari P., Palma D., Giannattasio G., Triggiani M., Balestrieri M. L. Lysophospholipid transacetylase in the regulation of PAF levels in human monocytes and macrophages.
Key Words: PAF • Inflammation
PERIPHERAL BLOOD MONOCYTES (MO) and tissue macrophage (Mf) are cellular components of the mononuclear phagocyte system that play a key role in the onset and the development of both inflammatory and immune reaction by generating bioactive molecules in response to a variety of stimuli (1
, 2)
. Both human Mo and Mf on appropriate stimulation synthesize and release platelet-activating factor (PAF, 1-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine), a potent proinflammatory phospholipid implicated in anaphylaxis, atherogenesis, septic shock, and other inflammatory processes (3
4
5
6
7
8)
.
In inflammatory cells, the stimulus-mediated PAF synthesis occurs through the remodeling pathway (9)
; in this route lyso-PAF, the immediate PAF precursor, is converted to PAF in a final step via the acetyl-coenzyme A:lysoPAF acetyltransferase (AT), a membrane-bound enzyme that catalyzes the transfer of the acetyl moiety from acetyl-coenzyme A to the free hydroxyl group at the sn-2 position of the lyso-PAF molecule. This enzyme is activated by a reversible phosphorylation mechanism(s) shown to require the activation of p38 MAP kinase (10
, 11)
.
PAF is converted into biologically inactive lyso-PAF by the PAF-acetylhydrolase (PAF-AH), an intracellular enzyme and a lipoprotein associated enzyme in plasma (12
, 13)
.
Blood Mo are thought to differentiate into mature Mf on entering tissues. This event is associated with major changes in the biochemical machinery of the Mo, including the metabolism of lipid mediators (14)
.
In particular, the differentiation of peripheral blood Mo into Mf induces the up-regulation of the secreted PAF-AH activity by enhancing its mRNA expression (15)
. An altered expression or deficiency of plasma PAF-AH has been associated with a number of pathologies such as asthma, stroke, myocardial infarction, brain hemorrhage, and cardiomyopathy (16)
. In light of the importance of the PAF-AH function in the resolution of inflammation, Wu et al. (17)
have recently demonstrated that the Sp family of transcription factors participates in the differential expression of PAF-AH during the maturation of Mo into Mf.
Even though the role of the plasma PAF-AH in the modulation of the PAF levels during Mo/Mf differentiation has been partially elucidated, no data are currently available on other key enzymes involved in PAF metabolism, such as the transacetylase (TA).
The TA transfers the acetyl group from PAF to a variety of substrate acceptors, thus modifying the cellular function of PAF through the generation of diverse lipid mediators (18
19
20)
. This enzyme possesses three separate catalytic activities, namely PAF lysophospholipids transacetylase (TAL), PAF sphingosine transacetylase (TAS), and PAF acetylhydrolase (AH) (21)
. Based on the amino acid sequences predicted from cDNA and peptide sequence correlation analysis, it has been shown that the sequence of the TA purified from both rat kidney membranes and cytosol is identical to that of intracellular bovine PAF-AH (II) (22)
. Moreover, both cytosolic and membrane TA showed the same immunological properties as PAF-AH (II). Similarly to PAF-AH (II), plasma PAF-AH also shows TA activity (22)
, which may contribute to the removal of PAF from LDL particles during LDL oxidation.
Our previous in vitro experiments indicated that in endothelial cells the TA is activated through reversible phosphorylation of the tyrosine and/or serine and threonine residues (23)
and that the three individual catalytic activities of TA are differentially regulated by agonists, posttranslational modifications, and subcellular translocation (24)
. All three TA activities of the purified membrane TA are sensitive to thiol-modifying agents, and only the purified membrane TAS is activated by phosphatidylserine (25)
. Very little is known about the intracellular signals that control TA activity.
Several function of Mo/Mf are regulated by p38 MAP kinase and MAP kinase kinase (MEK)/ERKs cascades. Activation of ERKs induced by various agonists such as GM-CSF, LPS, or MCP-1 correlates with Mo/Mf response toward proliferation or activation (26
27
28)
and with tissue factors or TNF-
expression (29)
. Moreover, activation of ERKs pathway in human Mo/Mf is involved in the CD40-mediated signaling and is crucial for interleukin (IL)-12 suppression by soluble CD40 ligand (30
, 31)
.
We reasoned that the identification of an enzymatic activity that contributes to the PAF degradation with the concomitant production of new lipid mediators would amplify the knowledge about the mechanism(s) that regulate(s) PAF levels during the maturation process of Mo into Mf. Therefore, in the present study, we sought to identify the TA activities in Mo and Mf and to study the possible mechanism(s) by which they are regulated.
The novelty of this study is the demonstration of the TAL role in the down-regulation of the signals elicited by PAF in Mo and in differentiated Mf. An in-depth study on the possible regulatory mechanism(s) indicated that p38/ERKs-mediated phosphorylation and subcellular translocation of the PAF-AH (II) are important events in the signal transduction pathway(s) responsible for the TAL activation. In Mf, the phosphorylating pathway(s) responsible for the up-regulation of PAF-AH (II) activities were p38/ERKs-independent.
Moreover, we demonstrated that in Mf both AH and TAL activities can be also ascribed, besides to PAF-AH (II), to the appearance of plasma AH, amply produced in the course of in vitro differentiation.
MATERIALS AND METHODS
PAF, acetyl-coenzyme A, lysophosphatidylcholine (lyso-GPC or lyso-PAF), dimethylsulphoxide (DMSO), bacterial LPS (LPS), BSA, phospholipase C (PLC) type XI from Bacillus Cereus, benzoic anhydride, Histopaque-1077, PIPES (1x), actinomycin D, leupeptin, sodium orthovanadate, dithiothreitol (DTT, phenylmethylsulfonyl fluoride (PMSF), alkaline phosphatase, and common laboratory chemicals were purchased from Sigma Chemical. Protein A-agarose, recombinant, constitutively activated p38, ERK-1, and ERK-2 were obtained from Upstate Biotechnology. Rabbit polyclonal antibodies against plasma PAF-AH were from Cayman Chemical. Rabbit polyclonal antibodies against PAF-AH (II) and recombinant plasma PAF-AH were a generous gift of Dr. Ten-ching Lee (Oak Ridge Associated Universities). Cruz marker molecular weight standards and goat anti-rabbit IgG (HRP-labeled) were from Santa Cruz Biotechnology. GM-CSF was obtained from PeproTech EC. Both MAP kinase inhibitors, SB203580 and PD98059, were from Alexis Biochemicals. Alkenyl-lyso-glycero-phosphoethanolamine (LPE) was a product from Serdary Research Lab. [3H]acetyl-PAF (13.5 Ci/mmol), [3H]acetate (1.9 Ci/mmol), and [3H]acetyl-CoA (1.54 Ci/mmol) were purchased from NEN Life Science Products. All culture reagents were from Invitrogen. Silica gel plates were from Analtech. Phospholipids were tested for purity by thin-layer chromatography (TLC), and only >95% pure phospholipids were used in the experiments.
Isolation of human Mo
Isolated human Mo were isolated from peripheral blood mononuclear cells (PBMC) of healthy donor by magnetic depletion of T cells, NK cells, B cells, dendritic cells, and basophils (MACS-microbeads monocyte isolation kit, Miltenyi Biotec). Briefly, PBMC were obtained by density gradient centrifugation (400 g for 40 min at 4°C) of 15 ml of leukocyte-rich buffy coat on 20 ml of Histopaque-1077. After centrifugation, the interface cells were carefully removed and transferred to a new conical tube. Cells were washed twice with PIPES (1x), centrifuged at 300 g for 10 min at 4°C, and then suspended in 9 ml of H2O, 3 ml KCl 0.6 M to a final volumes of 50 ml of PIPES (1x). After centrifugation at 300 g for 10 min at 4°C, the pellet was suspended in an appropriate volume of PIPES (1x), counted, and used for the magnetic labeling procedure of the monocyte isolation kit. The purity of the isolated Mo was >98%. Mo were suspended in RPMI 1640 supplemented with 1% DL-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 10% heat-inactivated fetal calf serum (FCS); plated in sixwell-plate at the density of 2.5 x 106 cells /well; and cultured for 12 h at 37°C in 5% CO2 in air. The nonadherent cells were removed after three washes with HBSS-10 mM HEPES before each experiment.
Culture of Mo-derived Mf
Adherent Mo were cultured in RPMI 1640 supplemented with 1% L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and various concentrations of FCS (0, 2, 5, and 20%) for 1, 3, 7, 10, and 12 days at 37°C in 5% CO2 in air to induce differentiation to the Mf phenotype. After the desired time of culture, Mf were washed twice with HBSS-10 mM HEPES and used for the experiments. Mf-specific markers associated with differentiation were ß-D-glucuronidase activity and changes in the expression of cell surface antigens. ß-D-Glucuronidase activity in cell lysates was measured by a colorimetric assay (32)
. The phenotype of cultured Mf was also confirmed through FACS analyses performed with specific fluorochrome-conjugated antibodies from Becton-Dickinson [anti-CD14, anti-CD11c and anti-human leukocyte antigen (HLA)-DR] and from Caltag (anti-CD86). Mo-derived Mf at various days of culture in the presence of FCS were used to determine the enzyme activities or the PAF accumulation. When indicated, Mf were washed twice with PBS and starved for 12 h before treatments.
Cell treatments
Adherent Mo were stimulated in the presence of HBSS-10 mM HEPES with or without different concentration of agonists, LPS, or GM-CSF, at 37°C for various times, as indicated in the figures. When Mo were treated to inhibit mRNA synthesis, cells were preincubated with actinomycin D (3 µg/ml) for 15 min at 37°C before stimulation with LPS. At the end of the incubations, the media were removed and the cells were rinsed twice with 2 ml of HBSS-10 mM HEPES. Homogenates prepared by sonicating cells at 30% power out put for 15 s x 5 times in the homogenizing buffer (0.25 M sucrose, 100 mM Tris-HCl pH 7.3, 1 µg/ml leupeptin, and 1 mM dithiotreitol) were used to perform enzyme assays. The protein content of the cell homogenates was determined by the Lowry method (33)
. To investigate PAF metabolism in Mo-derived Mf, cells were washed twice with HBSS-10 mM HEPES and cell homogenates were prepared as described above.
Experiments to evaluate the effect of the MAP kinase inhibitors SB 203580 and PD 98059 that, respectively, block the activation of p38 MAP kinase and ERK pathways, were conducted by incubating Mo or Mf in HBSS-10 mM HEPES with or without 50 µM PD 98059 or 25 µM SB 203580; DMSO was used as inhibitor vehicle and its final concentration in the media was <0.1%. Cells were incubated with SB 203580 for 30 min at 37°C in 5% CO2 in air or for 60 min with PD 98059 (20 min at 4°C, 20 min at room temperature, and 20 min at 37°C in 5% CO2 in air). After incubation, cells were activated with specific stimuli and used for the enzyme assays. Viability of cells was >95% as assessed by trypan blue exclusion test.
In vitro activation of TAL
Cell homogenates were prepared from Mo treated with or without LPS (20 ng/ml) for 15 min and from Mf starved for 12 h before activation in presence of 20% FCS for 1 h. Homogenates from untreated cells (100 µg, 
65 µl) were immediately incubated at 37°C with 2 µg of recombinant, activated p38, ERK-1, or ERK-2 in the presence of 500 µM Mg2+ and 40 µM ATP in a final volume of 150 µl. After 15 min, the entire 150 µl of kinase reaction mixture was used to assay the TAL and AH activities.
Alkaline phosphatase treatment
Cell homogenates (100 µg) from Mf at different days of differentiation were incubated with 1.5 µg of calf intestinal alkaline phosphatase (2300 U/mg protein) for 15 min at 37°C in the presence of 50 mM Tris-HCl pH 9.0, 150 mM NaCl, 1 mM PMSF, and 1 mM EDTA. After incubation with alkaline phosphatase, cell homogenate proteins were extracted for 1 h at 4°C by adding 1% Triton X-100, 1% sodium deoxycholate, and 0.1% SDS and then used for Western blot.
Western blot analysis
Cell were lysed at 4°C for 1 h in lysis buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1 mM PMSF, 1 mM EDTA, 5 µg/ml aprotinin, 5 µg/ml leupeptin, 1% Triton X-100, 1% sodium deoxycholate, and 0.1% SDS) and centrifuged at 10,000 g for 10 min. The protein content of the supernatants was measured by Lowry method (33)
. Aliquots containing 50 µg of cell lysate proteins or 1 µg of standard protein, i.e., purified PAF-AH (II) (21)
or recombinant plasma PAF-AH, were subjected to SDS-12.5% PAGE under reducing conditions and blotted onto nitrocellulose membrane filters. The blots were blocked for 2 h at 37°C in 2% BSA in 1 mM Na2HPO4, 1.7 mM NaH2PO4, 150 mM NaCl, and 0.05 Tween-20 (PBS-T). The nitrocellulose membranes were washed once with PBS-T for 15 min; three times for 5 min with 1 mM Na2HPO4, 1.7 mM NaH2PO4, and 150 mM NaCl (PBS); and immunoblotted for 2 h at room temperature with rabbit polyclonal antiplasma PAF-AH antibodies or with rabbit anti-PAF-AH (II) antibodies. After being washed once with PBS-T and washed three times with PBS, the blots were incubated for 1 h at room temperature with HRP-labeled anti-rabbit IgG, washed with PBS, developed for 2 min with enhanced chemiluminescence reagent (Pierce), and exposed to Biomax film (Kodak).
Immunoprecipitation
Adherent Mo (107 cells/ml) were stimulated with LPS (20 ng/ml) or GM-CSF (10 ng/ml) in the presence of HBSS-10 mM HEPES. At the end of the incubations, cells were rinsed twice with 5 ml of ice-cold HBSS-10 mM HEPES and lysed in 1 ml of lysis buffer containing 100 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 µM leupeptin, 1 mM sodium orthovanadate, 40 mM PMSF, and 2 mM dithiothreitol. Cells were scraped from the plates and passed several times through a 26-gauge needle to disperse large aggregates. Cell lysates were then transferred to microfuge tubes and gently rocked on an orbital shaker at 4°C for 30 min. Tubes were centrifuged for 15 min at 14,000 g in a precooled centrifuge, and the supernatant fraction was immediately transferred to fresh centrifuge tube. Polyclonal antibodies (5 µg) against AH (II) or plasma AH were added to 0.5 mg of cell lysates (1 mg/ml) to a final volume of 500 µl with lysis buffer and gently rocked on a orbital shaker for 2 h at 4°C. Immunocomplexes were captured by adding 50 µl of washed protein A-agarose (10% suspension) and by gently rocking on orbital shaker for 1 h at 4°C. After centrifugation in a precooled microcentrifuge at 14,000 g for 2 min, immunocomplexes were separated from supernatant and washed twice with 500 µl of lysis buffer before TAL assay. When used for the Western blot analysis, immunocomplexes were resuspended in 60 µl 2x sample buffer and boiled for 5 min. After centrifugation at 14,000 g for 10 min, the supernatants were transferred to a fresh microcentrifuge tube and used for the analysis.
PAF accumulation assay
To measure PAF accumulation, a metabolic labeling was performed as described previously (23)
. Briefly, adherent Mo or Mf were cultured in 100 mm culture dish at cell density of 5 x 106 cells/dish and incubated at 37°C with [3H]acetate (25 µCi) in the presence of 10 ml of HBSS-10 mM HEPES with or without agonists for various times. At the end of the incubations, the cells were washed twice with 5 ml of HBSS-10 mM HEPES to remove excess of radiolabel before scraping into 3 ml of methanol. The cellular lipids were extracted by the method of Bligh and Dyer (34)
, and the 1-radyl-[3H]acetyl-GPC fraction was isolated by thin layer chromatography (TLC) with a solvent system of CHCl3/CH3OH/CH3COOH/H2O (50:25:8:4 v/v/v/v/). Radioactivity of the lipid fractions was determined by area scraping of the silica gel into vials for liquid scintillation counting. The relative amounts of 1-acyl-[3H]acetyl–GPC and 1-alkyl-[3H]acetyl-GPC produced by stimulated Mo were determined after PLC hydrolysis, benzoylation and TLC analysis (23)
.
Determination of the rate of transfer of the [3H]acetyl group from [3H]acetyl-PAF to 1-acyl-2-lysoPAF
Mo were incubated at 37°C with agonists as described before in presence of [3H]acetyl-PAF (2.5 µCi) in 0.1% BSA. At the end of the incubations, the media were removed and the cells were rinsed twice with 5 ml of HBSS-10 mM HEPES before being scraped into 3 ml of methanol. The cellular lipids were extracted, and the amounts of [3H]acetyl groups transferred from [3H]acetyl PAF to 1-acyl-2-lysoPAF were measured after PLC hydrolysis, benzoylation and TLC analysis as described previously (23)
.
PAF-AH (II) translocation
Cell homogenates were prepared from Mo (107 cells) treated with 20 ng/ml LPS (1 h) or 10 ng/ml GM-CSF (15 min). Postnuclear fraction (500 g supernatant) was centrifuged at 100,000 g for 1 h at 4°C to separate supernatant as the cytosol fraction and the pellet as membrane fraction. Both the cytosol and membrane fractions (50 µg) were used for the enzyme assay and for Western blot analysis with anti-PAF-AH (II) polyclonal antibody, as described above.
AT assay
AT activity was assayed as described previously (23)
. Briefly, incubation mixtures contained 500 µM [3H]acetyl-coenzyme A (0.2 µCi), 50 µM lyso-PAF suspended in 3.3% BSA-saline, 100 mM Tris-HCl (pH 7.2), and 100 µg of cell homogenate protein in a final volume of 0.5 ml. The samples were incubated at 37°C for 30 min and the lipids were extracted by the method of Bligh and Dyer (34)
except that 2% of acetic acid was included in the methanol. The extracted lipids were separated by TLC using a solvent system of CHCl3/CH3OH/NH4OH/H2O (60:35:8:2.3, v/v/v/v). The radioactivity of the areas corresponding to PAF was determined by liquid scintillation counting.
TA assays
TAL activity was determined by incubating at 37°C for 30 min [3H]acetyl-PAF (50 µM, 0.5 µCi in 0.1% BSA-saline) as acetyl donor and LPE (300 µM in 0.1% BSA-saline) as substrate acceptor in a mixture containing 100 µg of cell homogenate, 100 mM Tris-HCl (pH 7.4), 5 mM EDTA, and 1 mM sodium acetate in a final volume of 0.25 ml. The extracted lipids (34)
were separated by TLC using a solvent system of CHCl3/CH3OH/CH3COOH/H2O (50/25/8/4, v/v/v/v), and the radioactivity of the areas corresponding to [3H]acetyl-PAF and [3H]acetyl-PE were determined by liquid scintillation counting.
The TAS activity was determined as described by Karasawa et al. (21)
. Briefly, the TAS assay system contained [3H]acetyl-PAF(15 µM, 0.5 µCi), sphingosine (50 µM suspended in equal molar ratio of BSA), 100 mM Tris-HCl (pH 7.4), 10 mM EDTA, and 2 mM sodium acetate in a final volume of 0.25 ml. Incubations were carried out at 37°C for 30 min and the extracted lipids were separated by TLC using a solvent system of CHCl3/CH3OH (90/10, v/v). The radioactivity of the areas corresponding to [3H]acetyl-PAF and [3H]acetyl-sphingosine was determined by liquid scintillation counting.
AH assay
The AH activity assay system contained 20 µM [3H]acetyl-PAF (0.1 µCi), 1 mM EDTA, 100 mM potassium phosphate (pH 8.0), and 100 µg of cell homogenate in a final volume of 0.5 ml. The reaction was stopped after 15 min of incubation 37°C by adding 1 ml CHCl3, 1 ml CH3OH, and 0.5% sodium bicarbonate. The upper phase was washed three times with 1 ml CHCl3 and used for the radioactivity counting.
Statistical analysis
Data are expressed as ± SE from at least three independent experiments in duplicate. Statistical analysis was performed by Student’s t test. Probability values were considered significant at P < 0.05.
RESULTS
LPS and GM-CSF induce TAL activity in human Mo
To determine the possible TA involvement in the regulation of the PAF levels in the mononuclear phagocyte system cells, we first examined whether the TA activities were induced in Mo stimulated with LPS or GM-CSF, two agents known to activate inflammatory cells by stimulating the PAF metabolism (35
,36)
. As shown in the Figs. 1
and 2
, these two agonists stimulate not only the PAF biosynthesis (AT activity), as already known but, interestingly, also its degradation catalyzed by the TAL. In particular, only the TAL activity is induced by stimulation with LPS or GM-CSF, whereas the AH activity remains to near basal concentration, as well as the TAS activity (not shown). Specifically, as shown in the LPS dose-dependence curve (Fig. 1A
), the maximal enzyme activation occurs at concentration of 20 ng/ml and is approximately twofold for the TAL activity (specific activity of the control was 0.96 nmol/min/mg protein) and approximately fourfold for the AT activity (specific activity of the control was 13.7 nmol/min/mg protein). The time-course curve of the LPS-stimulated Mo (Fig. 1B
) shows an induction of the AT and TAL activities characterized by an early peak at 15 min and a delayed peak at 1 h, a pattern similar to the biphasic PAF accumulation previously reported (35)
. As illustrated in the insert of Fig. 1B
, pretreatment of Mo with actinomycin D (3 µg/ml) abrogated only the delayed but not the early peak of AT and TAL activations induced by LPS, suggesting that the second peak was dependent on the protein synthesis. The highest induction of the TAL activity (200% of the control) in Mo stimulated with GM-CSF occurs at concentration of 10 ng/ml for 15 min (Fig. 2A and B
) and is comparable to the AT induction (235% of the control). The specific induction of the TAL activity in activated Mo suggests that PAF is preferentially used for the synthesis of PAF analogues, rather than for the synthesis of N-acetyl-sphingosine (TAS activity) or for its hydrolysis (AH activity). To test this, we next examined the production of PAF and its acyl analogs by determining the subclasses of 1-radyl-2-[3H]acetyl-GPC produced by intact Mo on stimulation with LPS or GM-CSF. As shown in Fig. 3
A, PAF is the predominant specie produced by stimulated Mo, as indicated by the proportion of the 1-radyl-2-[3H]acetyl-GPC subclasses (70% for the 1-O-alkyl-linked and 30% for the 1-acyl-linked-2-[3H]acetyl-GPC). Although the quantities of acyl-PAF are low compared to PAF, a 4.1-fold increase in production of this acyl analog was observed in activated Mo, as indicated when the transfer of the [3H]acetyl group from [3H]acetyl-PAF to 1-acyl-2-lysoPAF was measured in intact cells (Fig. 3B
; from 390 cpm/dish in control cells to 1600 cpm/dish in cells treated with LPS or GM-CSF).
|
|
|
Among the several types of PAF-AH in mammals, namely the intracellular type I and II and the plasma type, only the intracellular PAF-AH (II) and the plasma PAF-AH possess TA activity (22)
. The PAF-AH (II) is a 40 kDa monomer with an amino acid sequence that exhibits 41% identity with that of plasma PAF-AH, a 43 kDa monomeric enzyme (12
,13)
.
To characterize the PAF-AH isoforms involved in the two peaks of enhanced TAL activity detected during activation of Mo by LPS (Fig. 1B
), we performed Western blot analysis of cell lysates with polyclonal antibodies against PAF-AH (II) and plasma PAF-AH. As shown in Fig. 4
A, Western blot analysis of the lysates from Mo activated with LPS for 15 min (lane 3 of gel 1 in A) and 1 h (lane 4 of gel 1 in A) shows an immunoreactive band comigrating on the gel with standard PAF-AH (II) protein. In contrast, no immunoreactive band comigrating with recombinant plasma PAF-AH can be visualized with polyclonal antibodies against plasma PAF-AH (A; gel 2). These data indicate that both the early and late peak of LPS-induced TA activity is due to PAF-AH (II). To confirm this observation, we performed measurement of the TAL activity after immunoprecipitation with specific antibodies against PAF-AH (II) or plasma PAF-AH. As shown in Fig. 4B
, when the cell lysates from Mo activated by GM-CSF for 15 min (10 ng/ml) or by LPS for 1 h (20 ng/ml) were incubated with polyclonal antibodies against PAF-AH (II) and the TAL activity was measured on the immunoprecipitates and supernatant fractions, 76% of the cell lysates activity was recovered in the immunoprecipitate fraction and 8% in the supernatant fraction. In contrast, when the cell lysates were immunoprecipitated with antibodies against plasma PAF-AH, the highest TAL activity was still detected in the supernatant fraction (80% of the cell lysate activity). Moreover, only the fractions immunoprecipitated with the anti-PAF-AH (II) and not with the antiplasma PAF-AH showed an immunoreactive band comigrating with the PAF-AH (II) standard protein, thus confirming that PAF-AH (II) was immunoprecipitated with the anti-PAF-AH (II) and not with the antiplasma PAF-AH (data not shown). These data indicate that the PAF-AH isoform involved in both early and late peak of increased TAL activity during activation of Mo by LPS and GM-CSF is PAF-AH (II).
|
PAF production is activated in human Mo-derived Mf
Before exploring the TA activities, we first determined the activation of the PAF remodeling route in intact Mf generated by in vitro differentiation of Mo in presence of animal serum (36)
. To confirm that under our experimental conditions the blood Mo acquire the morphological, biochemical, and phenotypical characteristics of mature Mf, we monitored the differentiation by assaying the ß-D-glucuronidase activity on cell lysates (37)
and by performing flow cytometry analysis (FACS). The FACS analysis performed on Mo incubated with 20% FCS for 10 days indicated that the forward- and the side-scatter signals of the cells were homogeneously increased in >95% of the cells and that the expression of CD11c, CD86, and HLA-DR, three major surface markers of Mf, was up-regulated after 10 days of incubation with 20% FCS compared to fresh Mo (day 0).
On the basis of these data, Mf cultured for 10 days in presence of 20% FCS were used in the next set of experiments designed to compare the PAF accumulation in Mo and Mf by quantifying the [3H]acetate incorporation into 1-radyl-2-lyso-GPC during cell activation. As shown in Fig. 5
, treatment for various times with GM-CSF 10 ng/ml, a concentration used to investigate PAF metabolism (36)
, had no or little effect on the 1-radyl-2-[3H]acetyl-GPC production compared to untreated Mf. In Mf starved for 12 h before stimulation (Fig. 5)
serum removal inactivated the cells (887±12 vs.1245±23 cpm/dish for the cells continually incubated with FCS). Stimulation of starved cells with GM-CSF for various times had no significant effects on the 1-radyl-2-[3H]acetyl-GPC production, whereas re-addition of FCS determined a time-dependent increase of the 1-radyl-2-[3H]acetyl-GPC production beginning at 10 min (1305±41 cpm/dish at 10 min vs. 868±10.5 cpm/dish at time zero). The results indicate that PAF synthesis had been already activated by differentiation and that stimulation with GM-CSF produced little or no 1-radyl-2-[3H]acetyl-GPC. These data also indicate that in activated Mf the PAF concentration is approximately four times less than that in GM-CSF-activated Mo.
|
TAL and AH activities increase during Mf differentiation
To determine the contribution of the TA in the mechanism regulating the PAF levels in Mf, we assayed the TA activities on the cell homogenates from Mf cultured for 10 days in presence of 20% FCS. As shown in Fig. 6
, we found that both TAL and AH activities progressively increased during in vitro Mf differentiation. The increase of the TAL activity during the differentiation process ranged from 180% (day 2) up to 334% of the control (day 7) and 410% of the control (day 10) and was more consistent than the increase of the AH activity (136, 252, and 320% of the control at days 2, 7, and 10, respectively). The TAL and AH specific activities in control cells, i.e., Mf at day 0, were 0.96 ± 0.05 and 3.6 ± 0.5 nmol/min/mg protein, respectively.
|
Moreover, the AT activity increased to a lesser extent compared to the TAL and AH activities. Particularly, the AT specific activity at day zero was 13.7 ± 0.3 nmol/min/mg protein and ranged from 110% (day 2) up to 140% of the control (day 10). In addition, according to the results shown in Fig. 5
, when the AT and TA activities were determined in Mf stimulated for 15 min with GM-CSF (10 ng/ml), the enzyme specific activities were comparable to untreated cells (data not shown). Finally, the TAS activity determined during Mf differentiation (data not shown) was comparable to Mo (specific activity of control was 50±0.5 pmol/min/mg protein).
These data suggest that the difference in PAF levels between FCS-activated Mf (Fig. 5)
and LPS or GM-CSF-activated Mo (Fig. 3)
is not solely due to a decreased synthesis (AT activity) but to a noticeable increase of its degradation (TAL and AH activities). In particular, in addition to the AH activity previously shown (15)
, the modulation of the PAF levels during Mf differentiation can be at least partially attributed to the TAL activity, thus highlighting a new role of this activity in the PAF metabolism.
Analysis of the PAF-AH isoform in Mf differentiated from Mo
The increase of both TAL and AH activities during Mf differentiation led us to analyze the PAF-AH isoform in Mf during in vitro differentiation. For this purpose, we prepared cell lysates from Mf grown for different days in presence of 20% FCS, as described in Fig. 6
legend.
The results from Western blot analysis with polyclonal antibodies against plasma PAF-AH and PAF-AH (II) indicated that Mf accumulate plasma PAF-AH starting from day 5 of differentiation. These cells, concomitantly, express PAF-AH (II), which seems to be the predominant acetylhydrolase isoform. Interestingly, the Western blot pattern indicates additional bands recognized by the antibody anti-PAF-AH (II) slightly shifted at lower molecular mass. Treatment of Mf lysates with alkaline phosphatase leads to changes of the immunoreactive pattern, which showed only bands comigrating with standard PAF-AH (II). Together these results suggest that the increase of TAL and AH activities in Mf during differentiation can be ascribed to both PAF-AH (II), which resulted to be regulated by phosphorylation, and to accumulation of plasma PAF-AH, an acetylhydrolase known to be regulated at transcriptional level (17)
.
ERKs and p38 MAP kinase pathways mediate TAL and AT induction in Mo
Our next goal was to analyze the signaling pathway(s) responsible for the PAF metabolizing enzyme activation in Mo and Mf. Previous studies have shown that GM-CSF and LPS activate multiple pathways, including those involving ERKs and p38 MAP kinase participation (26
27
28
29
30
31
, 38)
.
When selective inhibitors against p38 MAP kinase (SB 203580) and ERKs (PD 98059) were tested on Mo and Mf, we found that PD 98059 and SB 203580, at a concentration of 50 µM and 25 µM, respectively, inhibited >80% of their respective ERKs or p38 kinase responses to LPS or GM-CSF (data not shown). Results indicated that AT (Fig. 7
A) and TAL (Fig. 7B
) specific activities induced by LPS (1 h) or GM-CSF (15 min) returned near to control values when Mo were treated with p38 MAP kinase or ERKs inhibitors. Moreover, when the PAF biosynthesis was determined in intact Mo preincubated with PD98059 or SB 203580 (Fig. 7C
), both compounds reduced the LPS-stimulated 1-radyl-[3H]acetyl-GPC production from 4920 ± 334 to 2150 ± 225 cpm/dish and 1894 ± 172 cpm/dish, respectively (unstimulated cells: 1529±93 cpm/dish). Similarly, these two inhibitors reduced the 1-radyl-2-[3H]acetyl-GPC production induced by stimulation with GM-CSF to near control, specifically from 4529 ± 275 cpm/dish to 2210 ± 188 (50 µM PD 98059) or to 1987 ± 255 cpm/dish (25 µM SB 203580).
|
On the whole, this result shows that the AT and TAL activities in Mo are directly/indirectly regulated by p38/ERKs pathways and that inhibition of the AT activity by these MAP kinase inhibitors is reflected in the inhibition of the PAF production in response to both LPS and GM-CSF.
ERKs and p38 MAP kinase pathways regulate AT but not TAL and AH activation in Mf
We next asked whether in Mf the AT, TAL, and AH activities were regulated by p38 and ERKs-mediated phosphorylation by performing experiments on starved cells to induce a higher PAF synthesis (Fig. 5)
. Treatment of starved Mf with 20% FCS for 1 h determined an increase of the AT activity from 8.2 ± 1.5 to 17.8 ± 2.3 nmol/min/mg protein.
As shown in Fig. 8
, pretreatment with 50 µM PD 98059 or 25 µM SB 203580 before activation in presence of 20% FCS completely inhibited the AT activation whereas no significant effect was observed on the TAL and AH activities. Specifically, TAL and AH activities were slightly inhibited by pretreatment with 50 µM PD 98059 (15 and 12% of inhibition, respectively), while 25 µM SB 203580 inhibited the TAL activity by 31% and the AH activity by 28.4%. Moreover, even the simultaneous pretreatment of Mf with both kinase inhibitors had no effect on the TAL and AH activities (data not shown). These results suggest that in Mf ERKs and p38 kinase pathways are involved in the regulation of the AT activity and not in the regulation of the TAL and AH activities.
|
In vitro activation of AT, TAL, and AH by MAP kinase
To further explore the different regulation of the TAL and AH activities by MAP kinase in Mo and Mf and to complement the findings on intact cells, we evaluated the effect of recombinant, constitutively activated p38, ERK-1, or ERK-2 kinases on the AT, TAL, and AH activities in a cell-free system. As shown in Fig. 9
A and C) the addition of recombinant, constitutively activated p38, ERK-1, or ERK-2 to the untreated Mo homogenates increased both the TAL and AT activities reaching levels observed in homogenates from cells activated with LPS for 1 h. The AH activity, which is not induced by stimulation with LPS (Fig. 1B
), is not up-regulated by in vitro treatment with MAP kinase (Fig. 9B
). In Mf, only the AT activity was increased by incubation with recombinant p38, ERK-1, or ERK-2 as compared to FCS-activated Mf, whereas both the TAL and AH activity values remained close to those of untreated homogenates (Fig. 9, A-B
).
|
These data together with those from intact cells suggest that in Mo stimulus-induced activation of the AT and TAL is mediated by p38 and ERK-dependent phosphorylation and that the phosphorylation is primarily involved in the regulation of the TAL activity rather than AH activity of PAF-AH (II). In contrast, since in Mf only the AT activities appear to be influenced by p38 or ERK kinases it is likely that TAL and AH activities of PAF-AH (II) are regulated by kinases other than p38 and ERKs.
Translocation of the PAF-AH(II) from cytosol to membranes
PAF-AH (II) is known to be localized in both membrane and cytosol fractions and to translocate to the membrane in response to changes in the cell redox state and to direct/indirect phosphorylation (24
,39)
. To investigate whether agonist stimulation caused translocation of the PAF-AH (II), we analyzed the subcellular distribution of PAF-AH (II) in membrane fraction and in cytosol fraction on treatment of Mo with LPS (20 ng/ml for 1 h) and GM-CSF (10 ng/ml for 15 min). As shown in Fig. 10
A, PAF-AH (II), evenly distributed in the cytosol and membrane fractions of control Mo, translocated to the membranes on LPS and GM-CSF stimulation. When the TAL and AH activities were measured in the membrane fraction (Fig. 10B
), translocation of the PAF-AH (II) was specifically reflected in an increase of the membrane-associated TAL activity from 35% in control cells (% of the total units in postnuclear fraction) to 63% in LPS- or GM-CSF-treated cells. In contrast, the membrane-associated AH activity as well as the protein content of the membrane fraction remained the same in both control and stimulated cells. These data indicate that stimulation of Mo induces translocation of PAF-AH (II) to the cell membrane and that this process preferentially enhances the TAL but not the AH activity of this enzyme. The induced translocation of PAF-AH (II) with concomitant increase in the TAL activity can therefore be considered an event involved in the differential regulation of the PAF-AH (II) activities, i.e., TAL and AH, since the cellular location of the AH activity and the protein content were not affected by LPS and GM-CSF treatment of the cells.
|
DISCUSSION
PAF is one of the bioactive molecules produced by Mo and Mf. Although stimulated Mo release a large fraction of newly synthesized PAF, Mf are a major source of plasma PAF-AH (4
, 15)
. The balance of the PAF production and degradation during the differentiation of Mo into Mf is of pathophysiological importance; thus, it is carefully controlled during this process. It has been suggested that the PAF-AH released by Mf may influence local inflammatory events by degrading oxidized phospholipids and PAF and that PAF-AH may be transported to the blood, where it serves to limit the PAF half-life in plasma (15)
.
The mechanisms regulating PAF concentration, essential in the control of pathological inflammation, are still matter of studies since alterations of the enzymes that control PAF levels result in a number of pathologies (16)
. The role of both biosynthetic and catabolic enzymes in the mechanism responsible for the control of PAF levels in Mo/Mf has been extensively studied, and many data support the concept that the PAF degradation by AH represents the main pathway (15)
. However, the hypothesis that the PAF degradation could be associated with the production of other lipid mediators led us to investigate a possible involvement of the TA in the PAF metabolism during in vitro maturation of Mo into Mf. In fact, one of the physiological roles of TA is to diversify the biological function of PAF by producing different lipid mediators, such as ethanolamine plamalogen, acyl analogs of PAF, and N-acetylsphingosine (25)
.
The novelty of the present study is the demonstration that the TAL activity is consistently induced in Mo stimulated with LPS or GM-CSF (Figs. 1
2)
and in differentiated Mf (Fig. 6)
. Among the two types of PAF-AH that possess TA activity (22)
, namely the intracellular PAF-AH (II) and the plasma PAF-AH, both Western blot and enzyme assay of the immunoprecipitates revealed that the increased TAL activity in Mo activated by LPS for 15 min and 1 h or by GM-CSF for 15 min was due to PAF-AH (II) (Fig. 4, A and B
).
PAF-AH (II) has been shown to be a phospholipase that protects cells against cell death induced by oxidative stress through the hydrolysis of oxidized phospholipids. Both cellular function of PAF-AH (II) during oxidative stress and its substrate specificity have been deeply studied (13
, 40)
. The broad substrate specificity of PAF-AH (II) is compatible with the scavenger role of this enzyme, and it is suggested that oxidized phospholipids rather than PAF are the main substrates of the enzyme. More in detail, PAF-AH (II) shows little selectivity toward ester or ether bond at the position 1 of PAF molecule; hydrolyzes phospholipids with short to medium length sn-2 acyl chains, including truncated chains derived from oxidative cleavage; and possesses phospholipase A1 activity as well as significant activity toward short-chain diacyl- and triacyl-glycerols.
Although both TAL and AH activities are catalyzed by PAF-AH (II), Lee and et al. (24
, 25)
demonstrated that these two activities can be differently regulated by thiol modifying agents, by subcellular translocation, and by posttranslational modification, such as phosphorylation.
Moreover, analysis of site-directed mutants of PAF-AH (II) showed that the TAL activity was decreased in mutants at Gly2, a site for myristoylation, and at Cys120, whereas AH activity was not affected. This suggests that Cys120 and miristic acid attached to the N-terminal have a function for the TAL but not for the AH activity (22)
.
In the present study, we demonstrated that in Mo treated with LPS or GM-CSF phosphorylation is an important event for the different regulation of the TAL and AH activities of PAF-AH (II). In fact, in Mo stimulated with LPS or GM-CSF the TAL activity is inhibited by p38/ERKs inhibitors (SB 203580 and PD 98059) and activated by in vitro treatment with recombinant p38 or ERKs (Figs. 7
and 9A)
. At the same time, the AH activity, which is not induced in stimulated Mo (Figs. 1
and 2)
, is not up-regulated in vitro by treatment with recombinant MAP kinases (Fig. 9B
). Moreover, the stimulus-induced translocation of PAF-AH (II) from cytosol to membrane is specifically associated with the increase of the TAL but not AH activity (Fig. 10)
. This raises the possibility that the binding of the enzyme to the membrane somehow limits water availability necessary for the AH activity. These data suggest that, along with the p38/ERKs-mediated phosphorylation, the subcellular translocation of PAF-AH (II) participates in the mechanism(s) responsible for the dual regulation of the TAL and AH activities.
The intriguing observation of the present study is that when Mo were uniformly differentiated in vitro into Mf in the presence of FCS, significant increases of both the TAL and AH activities were observed during the differentiation (Fig. 6)
. According to previous studies (15)
, the AH activity, which was not significantly induced in activated Mo, increased during Mf differentiation. The analysis of the PAF-AH isoforms during Mf differentiation indicated that PAF-AH (II) is the major isoform also in Mf, but since accumulation of plasma PAF-AH also occurs during differentiation, the increase of TAL and AH activities in Mf during differentiation can be ascribed to both enzymes.
The observation that the reduction of the PAF levels in mature Mf is linked to increases of both TAL and AH activities provides the first evidence of the TAL contribution to the regulation of the PAF levels and to the production of other lipid mediators, such as PAF analogs and ethanolamine plamalogens known to be produced by these cells (5
, 41
42
43
44)
.
Since the definition of the regulatory mechanism(s) of PAF levels can significantly contribute to effective therapeutic strategies to control inflammatory diseases, we next explored the signal transduction pathways involved in the regulation of PAF biosynthesis (AT activity) and degradation (TAL and AH activities) in Mo and Mf. ERKs and p38 MAP kinases have emerged as important elements in the regulation of the Mo and Mf functions, such as Mf proliferation or activation (26
, 28
, 38)
, GM-CSF generation from LPS-stimulated human Mo (27)
, suppression of IL-12 production by soluble CD40 (30)
, expression and release of GM-CSF in alveolar Mf (45)
, and LPS-induced activation of AT in mouse peritoneal macrophages (46)
. However, the role of these MAP kinases in the regulation of the TAL in Mo and Mf is currently an area of investigation.
We found that when Mf are differentiated in vitro from Mo, p38 and ERK kinase inhibitors blocked only the AT activation while, contrary to what was observed in Mo, TAL and AH activation was minimally influenced (Fig. 8)
. Data from cell-free systems, similarly, showed that a phosphorylating system containing constitutively activated p38, ERK-1, or ERK-2 kinase failed to activate both TAL and AH in homogenates from untreated Mf (Fig. 9)
. These data indicate that regulation of TAL is not the same in Mo and differentiated Mf. When we explored the regulation of TAL activity in Mf, at least two differences with Mo emerged. The first is the accumulation of plasma PAF-AH, an enzyme that possesses TAL activity but has different regulation from PAF-AH (II) (17)
. The second is the occurrence of a different pattern of phosphorylation of PAF-AH (II) in Mf, which is likely to depend on kinases other than p38 and ERKs. Additional studies are required to elucidate which kinase(s) is responsible for the PAF-AH (II) phosphorylation in Mf.
In conclusion, the identification of lysophospholipid transacetylase in human Mo and in vitro derived Mf highlights a new way for the modulation of PAF levels in inflammatory cells.
ACKNOWLEDGMENTS
This work was supported by a grant from the "Ministero dell’Università e della Ricerca Scientifica" (MIUR, Rome) to L. Servillo and in part by a grant from the "Ministero dell’ Università e della Ricerca Scientifica" (MIUR) to M. Triggiani.
The authors express gratitude to Dr. Ten-ching Lee (Oak Ridge Associated Universitie) for the kind gift of rabbit polyclonal antibodies against AH (II) and Dr. Ken Karasawa (Teikyo University, Japan) for helpful discussion and suggestions.
Received for publication September 7, 2005. Accepted for publication December 13, 2005.
REFERENCES
