Leptin is a hormone secreted by adipocytes in correlation with total body fat mass. In addition to regulating energy homeostasis, leptin modulates immune functions such as macrophage phagocytosis and cytokine synthesis. Previously, we reported defective leukotriene synthesis in macrophages from leptin-deficient mice that could be restored with exogenous leptin. In the present study, we utilized macrophages from normal rodents to explore the mechanism by which leptin could enhance cellular leukotriene synthesis. Leptin pretreatment of either rat alveolar or murine peritoneal macrophages for 16 h dose dependently increased the synthesis of leukotriene B4 and cysteinyl leukotrienes in response to calcium ionophore or the particulate zymosan. Leptin also enhanced calcium ionophore-stimulated release of free arachidonic acid. Calcium-dependent and -independent arachidonoyl-selective phospholipase activities in macrophage lysates were likewise increased following leptin treatment. Immunoblot analysis of leptin-treated cells revealed that group IVC iPLA2 (cPLA2γ) protein expression increased ∼80%. These data demonstrate for the first time that phospholipase A2 activity and cPLA2γ protein levels in alveolar macrophages represent targets for upregulation by leptin and provide previously unrecognized mechanisms by which this hormone can promote inflammatory responses.
- lipid mediators
- cytosolic phospholipase A2
the leukotrienes (LTs) are potent lipid mediators of inflammation that play important roles in a variety of inflammatory disease states as well as in host defense against infection (3, 34). The initial step in the generation of eicosanoids is the liberation of arachidonic acid (AA) and other 20-carbon unsaturated fatty acids from tissue phospholipid pools, which involves the actions of one or more phospholipase A2 (PLA2) enzymes. The enzyme 5-lipoxygenase (5-LO), in conjunction with 5-LO-activating protein (FLAP), can oxygenate AA to form LTA4. This intermediate can be hydrolyzed to form the potent neutrophil chemoattractant LTB4 or conjugated with glutathione to form the cysteinyl LTs (LTC4, LTD4, and LTE4) that elicit smooth muscle contraction and microvascular permeability (30). Alternative metabolism of AA via the cyclooxygenase pathway results in the formation of prostaglandins or thromboxane.
Leptin, the 16-kDa protein product of the ob gene, is a pleotropic hormone that is produced primarily by adipocytes. In general, circulating leptin levels are linearly correlated with total body fat mass (23, 41). However, leptin levels drop rapidly during fasting and increase in response to infection and inflammatory stimuli (1, 33). In addition to regulating appetite and energy expenditure, leptin is an important immunoregulatory hormone since it enhances a number of immune responses, including macrophage effector functions (16, 21), cytokine synthesis, and T helper (Th) cell polarization to a Th1 phenotype (22). In a previous report, we observed a link between leptin and LT synthesis because we found that LT synthesis was reduced in macrophages from leptin-deficient mice. The defect in LT synthesis could be reversed by pretreating macrophages overnight with exogenous leptin (25). In addition to this observation, leptin has been reported to enhance cytosolic PLA2 (cPLA2) activity in bone marrow stromal cells and increase hypothalamic prostaglandin synthesis in rats (9, 19). In this report, we utilized normal rat alveolar macrophages (AMs) and murine peritoneal macrophages (PMs) to examine the mechanisms by which leptin increases LT synthetic capacity.
MATERIALS AND METHODS
Isolation and culture of murine PMs and rat AMs.
Resident murine PMs and rat AMs were obtained from specific pathogen-free female C57BL/6j mice (Jackson Laboratory, Bar Harbor, ME) and female Wistar rats (Charles River, Portage, MI), respectively, by lavage as previously described (31). All experiments were conducted in compliance with the Animal Care and Use Committee of the University of Michigan. Greater than 95% of cells obtained from lavage of the lungs (rat) or the peritoneal cavity (mouse) were identified as macrophages following staining with Diff-Quik (American Scientific Products, McGaw Park, IL). After lavage fluid centrifugation at 4°C at 200 g for 5 min, the cell pellet was resuspended in Hanks' balanced salt solution (HBSS; GIBCO, Grand Island, NY), and the cells were enumerated using a hemocytometer. The cells were centrifuged a second time and resuspended in RPMI 1640 (GIBCO) to a concentration of 5 × 105 cells/ml. Murine PMs and rat AMs were adhered to cell culture wells (5 × 105/well) in RPMI 1640 for 1 h. After 1 h, the cell culture medium was replaced with complete medium (RPMI containing 10% fetal bovine serum and 1% penicillin/streptomycin, GIBCO), and the macrophages were cultured for 16 h.
Macrophage stimulation and eicosanoid analysis.
Macrophages were cultured with medium alone or with increasing doses of either murine leptin (Calbiochem, San Diego, CA; for murine PMs), rat leptin peptide (rLep; Diagnostic Systems Laboratories, Webster, TX), or rat leptin (Sigma, St. Louis, MO; for rat AMs). The following day, the cells were washed with warm PBS, and the medium was replaced with RPMI alone or with RPMI containing leptin. The cells were incubated with or without the agonist calcium ionophore A-23187 (Calbiochem) at 10 μM for 30 min. In some experiments, the cells were stimulated with the particulate stimulus zymosan (Sigma) at 0.1 mg/ml for 1 h. Medium was collected and stored at −70°C and was later assayed for LTB4, cysteinyl LTs, and PGE2 using commercially available enzyme immunoassay kits (Cayman Chemical, Ann Arbor, MI).
Rat AM phospholipids were isotopically labeled by culturing overnight in complete medium including 0.5 μCi of [3H]AA (specific activity 95–100 Ci/mmol; DuPont NEN, Boston, MA) per well at 37°C in a humidified atmosphere of 5% CO2. We have previously verified that >90% of the [3H]AA is incorporated into cell membrane phospholipids (37). Before stimulation, unincorporated label was removed by washing cells with HBSS containing 0.1% fatty acid-free BSA. [3H]AA incorporation was determined by scraping AM monolayers in methanol and scintillation counting and averaged ∼95,000 disintegrations per minute (dpm) per well. After overnight incubation, the culture medium was replaced with RPMI containing 0.1% BSA to trap the deacylated fatty acid, and the cells were stimulated with A-23187 (10 μM) for 30 min. Released AA was evaluated as the total counts recovered in culture medium and expressed as a percentage of incorporated radioactivity.
PLA2 activity assay.
PLA2 activity was determined in lysates of rat AMs (10 × 106 cells/ml) as previously described (39). Briefly, cells cultured overnight with and without leptin (5 ng/ml) were washed twice with warm PBS containing 4 mM EDTA and twice with PBS. The cells were then scraped in homogenization buffer (50 mM HEPES and 1 mM EDTA, pH 7.4) and disrupted by sonication (20% duty for 10–20 s). The liposomes of phospholipid substrates were prepared by sonicating (90% duty for 5 min) 1-palmitoyl-2-arachidonoyl-phosphatidylcholine (50 μM) and ∼100,000 dpm 1-palmitoyl-2-[14C]arachidonoyl-phosphatidylcholine (14C-AA-PC) or 1-palmitoyl-2-linoleoyl-phosphatidylcholine (50 μM) containing ∼100,000 dpm 1-palmitoyl-2-[14C]linoleoyl-phosphatidylcholine (14C-LA-PC; Perkin-Elmer Life Sciences, Boston, MA) in buffer containing 120 mM NaCl and 40 mM Tris·HCl (pH 9). Fifty micrograms of cell lysate protein were incubated with freshly sonicated phospholipid substrates in buffer containing 50 mM Tris·HCl (pH 8.0) and 0.5 mg/ml of BSA for 30 min at 37°C in a shaking water bath. The assays were performed in the presence of either 5 mM CaCl2 or 5 mM EGTA. The reactions were terminated by the addition of 1.2 ml of chloroform-methanol, 2:1 (vol/vol). The chloroform layer was extracted, dried under a stream of nitrogen, resuspended in 60 μl of chloroform, and spotted on silica gel thin layer chromatography plates (Whatman, Marlborough, MA). Free fatty acid was separated from phospholipid by thin-layer chromatography in hexane:diethyl ether:glacial acetic acid (70:30:2 by volume). The spots corresponding to the free fatty acids were visualized by I2 vapor and scraped into scintillation vials containing methanol. Fifteen minutes later, scintillation fluid was added to each vial, the radioactivity of the fatty acids was determined by scintillation counting, and the PLA2 activity was determined as picomoles of liberated fatty acid per milligram of cell lysate protein per 30 min. The data were normalized to the appropriate control and expressed as a percentage of the control.
Immunoprecipitation of cPLA2α.
Rat AM monolayers were lysed in buffer containing 1% Triton X-100 with 50 mM Tris (pH 8.0), 100 mM NaCl, 1 mM Na3VO4, 1 mM PMSF, 50 mM NaF, and 1 μg/ml leupeptin. Lysates were incubated overnight at 4°C with anti-cPLA2α (1:50) (Cell Signaling Technology, Beverly, MA). Protein A-Sepharose (Amersham Pharmacia Biotech, Arlington Heights, IL) was added to each sample and incubated for 3 h with agitation at 4°C. The beads were washed three times with lysis buffer without Triton X-100 and separated on 8% SDS-PAGE gels. The entire volume recovered after boiling the beads was loaded onto the gels. Because protein concentration could not be accurately assessed due to interference with the Coomassie protein assay by SDS contained in the loading buffer, lysates were derived from equal numbers of cells, but total cPLA2α per lane was subject to variation. Proteins were transferred to nitrocellulose membranes (Schleicher & Schuell, Keene, NH) overnight at 100 Amps and for 3 h at 200 Amps. Membranes were then probed with the antibodies to total phosphorylated cPLA2α as described below.
Immunoblot analysis of 5-LO, FLAP, cPLA2α, cPLA2γ, and phospho-cPLA2α.
Immunoblot analysis of macrophage 5-LO, FLAP, 85-kDa cPLA2α, and group IVC iPLA2 (cPLA2γ) was performed as previously described (8). Briefly, rat AMs (2 × 106/well) were cultured overnight with complete medium with and without 5 ng/ml of rat leptin. The following day, the macrophages were washed with HBSS and scraped with ice-cold homogenization buffer (50 mM Tris·HCl, 25 mM KCl, 5 mM MgCl2, 1 mM dithiothreitol, 0.5 mM PMSF, and 1 mM leupeptin, pH 7.4), and the cells were disrupted by sonication (10 bursts at 20% duty cycle). Twenty micrograms of protein from each sample, determined by a modified Coomassie dye binding assay (Pierce Chemical, Rockford, IL), were separated by SDS-PAGE under reducing conditions and transferred to nitrocellulose membranes. Membranes were probed with the rabbit polyclonal antibodies against human leukocyte 5-LO (provided by Dr. Jilly Evans, Merck Frosst Center for Therapeutic Research, Point-Claire-Dorval, Quebec, Canada; titer 1:5,000) (40), amino acid residues 41–52 of the human FLAP sequence (Dr. Jilly Evans; titer 1:5,000) (24), recombinant human 85-kDa cPLA2α (provided by Dr. James Clark, Wyeth Research, Cambridge, MA; titer 1:1,000), COOH-terminal residues of human cPLA2α (Cell Signaling Technologies; titer 1:1,000), and an antibody against 61-kDa cPLA2γ (provided by Dr. Christina C. Leslie, National Jewish Medical and Research Center, Denver, CO; titer 1:500) (38). In addition, membranes on which immunoprecipitated cPLA2α were separated were also probed with both anti-cPLA2α and anti-phospho (serine505)-cPLA2α (Cell Signaling Technologies; titer 1:1,000). Primary antibodies were detected using horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (titer 1:5,000) and visualized with the ECL detection system (Amersham Pharmacia Biotech, Piscataway, NJ). The densities of the luminescent bands were quantitated in appropriately exposed autoradiographs by video densitometry using NIH Image software (Scion, Frederick, MD). The density value of the anti-phospho-cPLA2α blot was divided by the density value of the anti-cPLA2α blot to normalize the relative band densities of immunoprecipitated phospho-cPLA2α.
Where appropriate, mean values were compared using an analysis of variance, a paired t-test, or a Kruskal-Wallis test on ranks from nonparametric data. The Dunnett's test or the Bonferroni test were used for mean separation. In all cases, a P value of <0.05 was considered significant.
Leptin enhances macrophage LT and PGE2 synthesis.
To determine whether leptin modulates LT synthesis in macrophages, we cultured rat AMs and resident murine PMs overnight with medium alone or with increasing doses of rLep [at the time these studies were initiated, the full-length rat leptin protein (16-kDa) was not commercially available] or murine leptin. On the following day, the cells were stimulated to produce LTs with calcium ionophore. In the absence of calcium ionophore, we could not detect LTs in medium from cells treated with or without leptin (data not shown). As shown in Fig. 1A, rLep pretreatment increased LTB4 synthesis in calcium ionophore-stimulated rat AMs. The peak LTB4 levels (3-fold greater than the control) were produced with a 5-ng/ml dose of rLep. At higher concentrations of rLep (50 and 500 ng/ml), LTB4 synthesis declined but was still ∼2.5-fold greater than the control. Similarly, when the full-length leptin protein was commercially available, we repeated the dose response experiment and found that LTB4 levels peaked with a 5-ng/ml dose of rat leptin, which was in the physiological range of 1–10 ng/ml and twofold greater than the control (data not shown). Likewise, with the particulate zymosan, a physiological stimulus for LT synthesis, rat leptin pretreatment augmented LTB4 synthesis at the dose of 5 ng/ml (Fig. 1B). Heat inactivation of leptin by boiling for 30 min completely inhibited the ability of leptin to augment LT synthesis, indicating that leptin, rather than potential endotoxin contamination, was responsible for the enhancement of LT synthesis (data not shown). We also observed that murine leptin pretreatment enhanced resident PM LTB4 and cysteinyl LT synthesis in a dose-dependent manner, indicating that leptin can enhance LT synthesis in other types of macrophages (Fig. 2).
To determine whether the effects of leptin were confined to the 5-LO pathway, we also measured the levels of the cyclooxygenase product, PGE2, in leptin-pretreated rat AMs stimulated with zymosan. As shown in Fig. 3, leptin augmented the synthesis of PGE2 in a dose-dependent manner. The synthesis of PGE2 in response to leptin pretreatment increased by 45% at 5 ng/ml, peaked at 10 ng/ml (although not a significant increase), and declined (25% increase) with the highest dose of 50 ng/ml. Although the augmentation of PGE2 synthesis was less than that observed for LT synthesis, this result suggests that leptin influences a proximal step in the eicosanoid synthetic pathway that is common to both oxygenate pathways.
Leptin enhances macrophage AA release.
Accordingly, we next assessed the ability of rLep to modulate substrate AA release in rat AMs. rLep did not significantly affect the uptake of [3H]AA (data not shown). As shown in Fig. 4, overnight rLep pretreatment augmented calcium ionophore-stimulated rat AM AA release in a dose-dependent manner. The maximal effect, a doubling of the control level of release, was observed at physiological levels of rLep (5 ng/ml) and declined at higher levels (50 and 500 ng/ml).
Leptin enhances PLA2 activity in rat AM lysates.
To determine whether the increase in AA release with leptin pretreatment reflected increased activity of the enzyme(s) responsible for deacylation, we performed cell-free PLA2 assays in rat AMs. At baseline, PLA2 activity against both of the phospholipid substrates, 14C-AA-PC (Fig. 5A) and 14C-LA-PC (Fig. 5B), was observed in both the presence and absence of calcium. However, PLA2 activity increased significantly with rLep (5 ng/ml) pretreatment only when 14C-AA-PC was used as the substrate. Both calcium-dependent and calcium-independent PLA2 activities were enhanced by rLep, by ∼50 and 90%, respectively.
Effect of leptin on LT biosynthetic proteins.
We explored the possibility that some of the effects of leptin could be mediated by increases in the levels of the key regulatory enzymes in the LT synthetic pathway. We evaluated 5-LO, FLAP, cPLA2α, and cPLA2γ protein expression by immunoblot analysis in rat AMs following overnight pretreatment with rat leptin (5 ng/ml). Densitometric analysis of immunoblots indicated that leptin did not enhance 5-LO, FLAP, or cPLA2α protein expression, but it enhanced cPLA2γ protein expression by ∼80% (Fig. 6A). Finally, immunoblot analysis of immunoprecipitated cPLA2α revealed that leptin pretreatment did not enhance cPLA2α phosphorylation on serine505, an indicator of enzyme activation (Fig. 6B). Together, these results demonstrate that leptin enhances both the ability of PLA2 to liberate AA and the expression of cPLA2γ.
Previously, we observed that cysteinyl LT synthesis in macrophages from leptin-deficient mice was impaired and that in vitro addition of exogenous leptin was able to restore cysteinyl LT synthesis in PMs from these mice (25). In this study, we explored the mechanism by which leptin modulates LT synthesis in macrophages from normal mice and rats. We found that leptin pretreatment enhanced both LTB4 (rat AMs and murine PMs) and cysteinyl LT (murine PMs) synthesis in a dose-dependent manner. Together, these experiments demonstrate that the ability of leptin to increase LT synthesis extends to a variety of macrophage populations, AMs and PMs, rat and mouse, and stimulated with both soluble (A-23187) and particulate (zymosan) stimuli. These findings are pathophysiologically relevant since we have already shown that murine leptin levels rise to ∼10 ng/ml in the lungs and 7 ng/ml in blood in the setting of gram-negative pneumonia (25). Others have also observed increases in leptin in response to inflammatory stimuli in both rodents and humans (5–7, 17, 27, 28). Because LTs have been shown to play an important role in host defense against Klebsiella pneumoniae by augmenting phagocytosis and microbial killing (3, 26), the ability of leptin to increase LT synthesis by macrophages would be expected to contribute to its homeostatic role in antimicrobial host defense.
We noted that LTB4 synthesis in murine PMs increased nearly 5-fold at the highest dose of leptin, whereas the maximal increase of cysteinyl LT synthesis was only 2.5-fold. This increase was greater than what would have been predicted if the effect of leptin was limited to enhanced AA release. Although we did not observe changes in 5-LO or FLAP expression with a 5-ng/ml dose of leptin in rat AMs, we cannot rule out the possibility that higher concentrations of leptin might enhance the expression of these and other proteins that regulate the LT biosynthetic pathway. It also raises the possibility that leptin may also be capable of augmenting the activity and or expression of LTA4 hydrolase, the enzyme that converts LTA4 to LTB4 in macrophages (30). Future studies will be necessary to resolve these possibilities.
To determine whether the effect of leptin on eicosanoid synthesis was specific for LTs, we measured PGE2, a cyclooxygenase product produced by AMs in response to zymosan. This response increased by 45% when the PMs were pretreated with leptin (5 ng/ml). Interestingly, Brunetti and colleagues (9) found that exogenous leptin enhanced the synthesis of the cyclooxygenase products PGE2 and PGF2α in rat hypothalamic cells. Although the effect of leptin on PGE2 was much less than that observed for LTs, it suggests that leptin can augment a proximal step in eicosanoid biosynthesis common to both cyclooxygenase and 5-LO pathways.
Because AA release is necessary for the synthesis of both PGE2 and LTs, we examined this step in both intact cell and cell-free assays. We observed that leptin enhanced AA release from rat AMs in a dose-dependent manner. The effect of leptin on AA release in intact cells was confirmed by the observation that leptin enhanced cell-free PLA2 activity in macrophage lysates. In particular, leptin enhanced PLA2 activity assayed under both calcium-dependent and calcium-independent conditions, but only when 14C-AA-PC, and not 14C-LA-PC, was the substrate. This finding suggests that leptin can enhance the enzymatic activity of an arachidonoyl-preferring PLA2.
It is well established that multiple isoforms of PLA2 enzymes can be expressed simultaneously in a given cell type. For example, the P338D1 macrophage cell line expresses the group V secretory calcium-dependent PLA2, the group IV cytosolic calcium-dependent isoform of PLA2 (cPLA2α), and the group VI calcium-independent PLA2 (35). Whereas the PLA2 isoforms expressed in AMs have not been comprehensively evaluated, rat AMs are known to express the calcium-dependent, AA-preferring group IV cPLA2α-isoform, which is an important source of AA release in response to calcium-dependent agonists (8). The AA-preferring, calcium-dependent activity identified in AM lysates is consistent with a group IV cPLA2α. Hοwever, we did not observe an increase in cPLA2α protein expression by immunoblot analysis of leptin-treated cells. Kim and coworkers (19) have previously shown that leptin enhances cPLA2 activity in human bone marrow stromal cells via the ERK1/2 mitogen-activated protein (MAP) kinase pathway. Although increases in cPLA2α activity have been correlated with phosphorylation of cPLA2α via the MAP kinase pathway (20), we did not observe an increase in cPLA2α phosphorylation on serine505 following overnight incubation with leptin. We cannot exclude the possibility that leptin might increase cPLA2α activity by phosphorylation of other sites such as serine727 (13).
Calcium is an essential cofactor for the liberation of AA by cPLA2α from cell membrane phospholipids and the subsequent conversion of AA to LTs by the enzyme 5-LO. We observed an enhancement of PLA2 activity in both the presence and absence of calcium. This suggests that leptin enhances the activity of a PLA2 enzyme that can function with and without calcium. We demonstrate for the first time that rat AMs express cPLA2γ, an arachidonoyl-preferring, calcium-independent enzyme, and this may have contributed to the AA-specific calcium-independent activity. This enzyme is capable of liberating AA in both the presence and absence of calcium (32). Interestingly, leptin increased the expression of this enzyme by ∼80%, and this may have contributed to the increase in AA released in intact macrophages following leptin pretreatment. This is also the first report, to our knowledge, of a mediator that is capable of increasing the expression of cPLA2γ. Although the activation of calcium-independent PLA2 isoforms has been implicated in constitutive phospholipid remodeling (2), agonist-stimulated AA release by calcium-independent PLA2 activity has also been linked to eicosanoid synthesis (4, 29, 32, 38).
Our data suggest previously unrecognized mechanisms by which leptin can promote inflammatory responses. These findings may be relevant to human diseases in which eicosanoids play a pathophysiological role. For example, recent epidemiological studies have shown an association between asthma and obesity in adults (10, 12). In addition, leptin has been implicated as an important mediator of osteoarthritis (14). It is possible that the increased incidence of asthma and osteoarthritis associated with obesity, a circumstance characterized by the overproduction of leptin, is related to the ability of leptin to augment LT and PGE2 synthesis, respectively. However, obese individuals exhibit leptin resistance, and our results may be an overestimate of the ability of leptin to augment LT synthesis in the obese (15). In contrast to obesity, malnutrition is a common primary problem and also a secondary consequence of many disease states, such as human immunodeficiency virus infection, emphysema, and cancer. By virtue of reduced LT synthetic capacity observed in many of these conditions, leptin deficiency may contribute to impaired host defense against infection (11, 18, 36). Our present results provide a rationale for future studies exploring leptin's effects on AA release and cPLA2γ in such pathophysiological states.
Support for this research was provided by grants from the American Lung Association (RG-056-N) and the University of Michigan Office of the Vice President of Research.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society